VU Research Portal

Pain in people with dementia van Kooten, J.

2019

document version Publisher's PDF, also known as Version of record

Link to publication in VU Research Portal

citation for published version (APA) van Kooten, J. (2019). Pain in people with dementia: Determining the prevalence of pain and its associated factors in nursing home residents with dementia.

General rights Copyright and moral rights for the publications made accessible in the public portal are retained by the authors and/or other copyright owners and it is a condition of accessing publications that users recognise and abide by the legal requirements associated with these rights.

• Users may download and print one copy of any publication from the public portal for the purpose of private study or research. • You may not further distribute the material or use it for any profit-making activity or commercial gain • You may freely distribute the URL identifying the publication in the public portal ?

Take down policy If you believe that this document breaches copyright please contact us providing details, and we will remove access to the work immediately and investigate your claim.

E-mail address: [email protected]

Download date: 10. Oct. 2021 Pain in people with dementia

Determining the prevalence of pain and its associated factors in nursing home residents with dementia

Janine van Kooten VRIJE UNIVERSITEIT Pain in people with dementia

Determining the prevalence of pain and its associated Colofon factors in nursing home residents with dementia

This thesis was prepared at Amsterdam UMC, Vrije Universiteit Amsterdam, at the department of General Practice and Elderly Care Medicine, within the Amsterdam Public Health research institute.

The study was funded by grants from Alzheimer Nederland [WE.09-2012-02], Fund NutsOhra care ACADEMISCH PROEFSCHRIFT subsidies [1130-046], Amstelring, Roomsch Catholijk Ouderen Armen Kantoor [ U-9128-2012 HM/lw], Stichting Beroepsopleiding Huisartsen (SBOH), and Stichting Henriëtte Hofje. ter verkrijging van de graad Doctor of Philosophy aan de Vrije Universiteit Amsterdam, The following organizations participated in the study: Amstelring (www.amstelring.nl), Cordaan (www. op gezag van de rector magnificus cordaan.nl), Zonnehuisgroep Amstelland (ww.zhga.nl). prof.dr. V. Subramaniam, in het openbaar te verdedigen Financial support for the printing of this thesis was kindly provided by SBOH and Vivium Zorggroep. ten overstaan van de promotiecommissie van de Faculteit der Geneeskunde op woensdag 29 mei 2019 om 11.45 uur in de aula van de universiteit, De Boelelaan 1105

ISBN/EAN: 978-94-6323-609-6 Cover design and lay-out: Lisette Poelhekken & Ilse Modder (www.ilsemodder.nl) Cover: Jacqueline de Vries – JAC (www.jacquelinedevries.nl) Printed by: Gildeprint Eschede, www.gildeprint.nl

door Copyright © Janine van Kooten All rights reserved. No part of this book may be reproduced, stored in retrieval system, or transmitted, Janine van Kooten in any form of by any means, electronic, mechanical, photocopying, recording or otherwise, without prior written permission of the author. geboren te Naarden promotoren: prof.dr. C.M.P.M. Hertogh prof.dr. M.L. Stek copromotoren: dr. M. Smalbrugge dr. J.C. van der Wouden

“I don’t mind pain, so long as it doesn’t hurt” Oscar Wilde Contents

Chapter 1 General introduction 09 Chapter 6 Evaluation of a pain assessment procedure in long-term care 125 residents with pain and dementia Chapter 2 Pain experience in Dementia Subtypes: A Systematic Review 23 Published as: van Kooten J, Smalbrugge M, van der Wouden JC, Stek ML, Published as: Binnekade TT, Van Kooten J, Lobbezoo F, Rhebergen D, Van Hertogh CMPM. der Wouden JC, Smalbrugge M, Scherder EJA. Evaluation of a Pain Assessment Procedure in Long-Term Care Residents Pain Experience in Dementia Subtypes: A Systematic Review. Current With Pain and Dementia. Journal of Pain and Symptom Management, 2017 Alzheimer Research, 2017;14(5):471-485. Nov;54(5):727-731.

Chapter 3 A review of pain prevalence in Alzheimer’s, vascular, 59 Chapter 7 Pain, neuropsychiatric symptoms, and quality of life of 137 frontotemporal and Lewy body dementias nursing home residents with advanced dementia in the Published as: van Kooten J, Binnekade TT, van der Wouden JC, Stek ML, Netherlands: a cross-sectional study Scherder EJ, Husebø BS, Smalbrugge M, Hertogh CM. Published as: van Kooten J, van der Wouden JC, Sikkes SAM, Smalbrugge M, A Review of Pain Prevalence in Alzheimer’s, Vascular, Frontotemporal Hertogh CMPM, Stek ML. and Lewy Body Dementias. Dementia and Geriatric Cognitive Disorders, Pain, Neuropsychiatric Symptoms, and Quality of Life of Nursing Home 2016;41(3-4):220-32. Residents With Advanced Dementia in The Netherlands: A Cross- sectional Study. Alzheimer Disease & Associated Disorders, 2017 Oct- Chapter 4 Pain in dementia: prevalence and associated factors: protocol 81 Dec;31(4):315-321. of a multidisciplinary study Published as: van Kooten J, Delwel S, Binnekade TT, Smalbrugge M, van Chapter 8 General discussion 157 der Wouden JC, Perez RS, Rhebergen D, Zuurmond WW, Stek ML, Lobbezoo F, Hertogh CM, Scherder EJ. Pain in dementia: prevalence and associated Addendum Nederlandse samenvatting 175 factors: protocol of a multidisciplinary study. BMC Geriatrics, 2015 Mar 21;15:29. Curriculum Vitae 179

Chapter 5 Prevalence of pain in nursing home residents: the role of 105 Dankwoord 180 dementia stage and dementia subtypes. Published as: van Kooten J, Smalbrugge M, van der Wouden JC, Stek ML, Hertogh CMPM. Prevalence of Pain in Nursing Home Residents: The Role of Dementia Stage and Dementia Subtypes. Journal of the American Medical Directors Association, 2017 Jun 1;18(6):522-527.

6 7 1 General introduction 1 General introduction progressive neurological/neurodegenerative disorder. The specific symptoms that someone with dementia experiences depend on the parts of the brain that are damaged and the disease that is causing dementia. There are over 50 diseases that may cause dementia, but the most common dementia subtypes This thesis is about pain in people with dementia with a special focus on those are: Alzheimer’s dementia, vascular dementia, frontal temporal dementia and who live in a nursing home. Determining the prevalence of pain and its associated dementia with Lewy bodies.5 Regardless of which type of dementia is diagnosed, factors in nursing home residents with dementia will help optimizing pain each person will experience his dementia in his own unique way. management and might serve as a building block to improve palliative care for people with dementia. This general introduction provides background information Although the disease trajectories are variable, dementia should be viewed as a on nursing home care, dementia, and specifically pain in people with dementia. life-limiting disease, and as there is no curative treatment available, palliative Finally, the objectives and outline of this thesis are presented. care is regarded appropriate.6,7

Palliative care in dementia Palliative care is defined by the World Health Organization as “an approach that Nursing home care for people with dementia in the Netherlands improves the quality of life of patients and their families facing the problems associated with life-threatening illness, through the prevention and relief of There are several forms of institutional care in the Netherlands. The most suffering by means of early identification and impeccable assessment and intensive form of institutional care is provided in nursing homes. For several treatment of pain and other problems, physical, psychosocial and spiritual”.8 years, the government policy has been to stimulate the use of formal care at In addition to this definition, the European Association for Palliative Care has home instead of institutional care. As a result of this policy, the percentage of presented an 11 domain-framework for palliative care in dementia providing people with dementia that live in a nursing home has not changed in recent guidance for clinical practice, policy and research.9 One of the 11 domains is years. In contrast to this, the volume of care (i.e. the number of people with ‘Optimal treatment of symptoms and providing comfort’ and includes the heavy care needs) has increased by 2.2% per year on average.1 Nowadays, assessment and treatment of frequently reported symptoms such as pain in only people with severe physical disabilities or severe psychogeriatric problems people with dementia. This domain was found important by experts and had the (e.g. dementia) are admitted to nursing homes.1 highest research priority rating. More insight into the assessment and treatment of pain in dementia might serve as a building block to improve palliative care for In 2013, approximately 50,000 people with dementia and dementia-like disorders people with dementia. were living in a nursing home.2 Most of them lived in small-scale special care units, as a result of improvements in dementia care with more attention for residents’ needs and their security and safety.3 The care in dementia special care units is characterized by a multidisciplinary approach, involving nursing staff, Pain in people with dementia elderly care physician, psychologists and paramedical staff.4 Pain in people with dementia is a frequent occurrence and has been cited as one of the main reasons for a decrease in quality of life, especially in residents with dementia who live in a nursing home.10 However, pain in people with dementia is Dementia also known to be difficult to manage adequately.11 Particularly people with dementia who live in nursing homes have been identified to be at increased risk Dementia is a term used to describe a combination of symptoms (i.e. syndrome) for experiencing pain and receiving inadequate treatment.12,13 which have a considerable impact on an individual’s daily life: problems with memory, having trouble communicating and behavioural changes (i.e. neuro­ psychiatric symptoms). It is caused by brain cell damage, mainly due to a

10 | GENERAL INTRODUCTION CHAPTER 1 | 11 Pain awareness of pain.36 In dementia with Lewy bodies, pain perception might be Pain is described by the International Association for the Study of Pain as altered due to damage caused by Lewy bodies, as well as, cortical atrophy.37 “an unpleasant sensory and emotional experience associated with actual or potential tissue damage, or described in terms of such damage”.14 Pain is the However, evidence from neuropathological, neuroimaging, experimental, and experience we associate with actual or potential tissue damage and it is always clinical research is still conflicting for all dementia subtypes and there is no subjective.14 This is also true for people with dementia, and when they report pain, conclusive evidence of the impact of dementia related neuropathological changes specifically those with mild to moderate dementia, it should be regarded as pain.15 on pain perception.26 Knowledge of the prevalence of pain in people with dementia of different subtypes will be helpful in optimizing the treatment of pain in people Pain prevalence in people with dementia with dementia; it is therefore relevant to investigate the prevalence of pain in a The prevalence of pain increases with advancing age, often exceeding 50% in population with various dementia subtypes. The implications of assessment on community dwelling older people without dementia.16 In recent years, similar pain in different dementia subtypes are discussed further on in this chapter. prevalence rates have been found for pain in people with dementia.11,17 Pain management in people with dementia Two decades ago studies on pain in dementia were scarce, especially in those Management of pain in dementia includes both recognition and treatment of who live in a nursing home.18 Ferrell et al.(1995) were the first to report on the pain. Pain recognition is an important and challenging task because pain prevalence of pain in nursing home residents with dementia and they found a expression in people with dementia might be different from those without prevalence of 62%.19 Their findings were followed by some evidence that pain dementia. The clinical manifestations of pain are frequently multifactorial, was not always recognized,20 and that pain was structurally undertreated in and people with dementia are more vulnerable to side effects of analgesic drugs. people with dementia.21-23 Along with these findings a discussion was started about the interaction between pain and dementia.24 It has been suggested that Pain recognition is especially difficult in nursing home residents with dementia due to the specific neuropathological changes pain might be perceived differently who are unable to articulate their pain verbally. The non-verbal manifestations of per dementia subtype.25 Since, the number of studies on neuropathological pain must be recognized and interpreted correctly, so that the distress caused to changes in pain and dementia has grown gradually.26 First, animal studies these residents can be treated adequately. Few guidelines exist to aid in the found an elevated pain tolerance in mice suffering from Alzheimer’s dementia, 27 recognition and treatment of pain. In the Netherlands, the Dutch Association of subsequently experimental studies in people with Alzheimer’s dementia Elderly Care Physicians (Verenso) has developed the multidisciplinary guideline replicated this finding.28 With atrophy of gray matter as main neuropathological ‘Recognition and treatment of chronic pain in vulnerable elderly people’ (2011).38 feature in Alzheimer’s dementia, neuropathological changes in Alzheimer’s This guideline provides a summary of the relevant literature on pain in elderly dementia have been suggested to cause an increase in pain tolerance. 27-29 people, and contains recommendations for recognition as well as for treatment This increase in pain tolerance is expected to result in a diminished pain of pain in vulnerable elderly people, including people with dementia. experience in people with Alzheimer’s dementia,28,29 however, more recent Internationally, it is one of the first guidelines specifically concerning pain studies have shown conflicting results.30-32 These studies show that pain management in nursing home residents with dementia, including detection of perception, and therefore pain experience, was not diminished in people with pain and its treatment. Adherence to the guideline in clinical practice has not Alzheimer’s dementia. Moreover, Jensen-Dahm et al.(2014) showed a lower been evaluated yet. tolerance to cold pressor pain in people with AD. 30 In the case of vascular dementia, with atrophy of white matter as main neuropathological feature, Pain assessment in people with dementia neuropathological changes have been suggested to decrease the pain Assessing pain and quantifying the experience of pain in older people with tolerance.25,33 And this decrease in pain tolerance might lead to both an increased dementia is both important and challenging. It becomes even more challenging pain experience and an increased pain prevalence.34 In frontotemporal dementia, in the presence of severe cognitive impairment, and communication difficulties. atrophy of grey matter is seen in the frontal, lateral temporal and parietal brain Older people with severe cognitive impairment, in particular those who live in areas.35 These neuropathological changes have been suggested to influence the nursing homes, may find it difficult to articulate their pain verbally and their emotional-affective aspects of pain, and therefore, might lead to a loss of ability to self-report can become impaired or even be absent. Behavioral changes

12 | GENERAL INTRODUCTION CHAPTER 1 | 13 may be the only available externally observable sign of pain, but the interpretation nursing home residents with a combination of Alzheimer’s dementia and of these behaviors might be difficult. It is particularly important that pain is vascular dementia tend to have more pain than those without dementia.45 As considered as a possible underlying cause for behavioral change and that steps both studies were not primarily designed to compare pain prevalence in people are taken to identify the possible cause in order to provide comfort, and therefore with dementia of different subtypes, it is relevant to study the prevalence of pain improve the quality of life. In the presence of cognitive impairment, behavioral in a population of nursing home residents with a known dementia subtype measures of pain can be used with moderate to reasonable interrater reliability, diagnosis, and to use several dementia-specific assessment tools concurrently.47 but even in cognitively intact older people they are not always a good measure of intensity.15,39 Pain treatment in people with dementia Pain treatment in people with dementia in itself is not so different from that of For people with mild to moderate dementia who can still communicate, self-report people without dementia. After recognizing that someone is in pain, classifying scales have been proven useful and have been shown to be valid.40,41 For people pain in pathophysiologic terms can be a guidance in achieving adequate pain with severe dementia who are unable to communicate whether they are in pain, relief. There are two main pain types:14,48 there are dementia-specific observational pain scales. However, the robustness of their psychometric properties has been object of discussion.42 Observational 1. Nociceptive pain: this type of pain is caused by tissue damage and is most pain scales are based on behaviour that is interpreted as signaling pain. often derived from the stimulation of pain receptors. Nociceptive pain is Pain observations may be based on behaviours that typically indicate to others usually described as “sharp” and “aching”. This type of pain responds well that an individual is experiencing pain such as facial expressions, and physical to common analgesic medications and non-pharmacological strategies. signs, but also on less specific behaviours such as avoiding functional activities or recreational activities. Some people become aggressive and start verbalising 2. Neuropathic pain: this type of pain is caused by nerve damage and is and vocalising their pain, whereas other people may only show subtle changes usually described as “tingling” or “shooting”. Neuropathic pain does not such as becoming withdrawn. It should be remembered that such behaviour respond as predictably to analgesic therapy as nociceptive pain. However, may point to pain, but may have causes other than pain. Observational pain neuropathic pain has been noted to respond to so called ‘anti-neuropathic tools might pick up other causes of distress, for example anxiety.43 Therefore, drugs’ such as tricyclic antidepressants and anticonvulsants.38 the relationship between signs and symptoms pointing to pain and the usual signs and symptoms of the dementia syndrome needs to be weighed carefully The most commonly used strategy to manage pain is to start with analgesic before concluding that the observed behaviour is indicative of pain.44 drugs. However, the number of analgesic drugs that can be used in older people is limited. Older people are more likely to experience side-effects related to Nowadays a number of promising pain assessment tools are available and changes in their pharmacokinetics (i.e. metabolism and elimination of analgesic research into the psychometric characteristics of these tools has shown that drugs); therefore, several medications should be avoided in older people (with several of these assessment tools enable the effective identification of pain and dementia), such as nonsteroidal anti-inflammatory agents (NSAIDs), paracetamol pain-related behaviour in people with dementia that have difficulties with codeine, or, weak opioids such as tramadol.37,38 This limits the analgesic communicating pain.11, 42, 45 Recent studies that used these specific observational treatment possibilities and might result in a generally restricted analgesic policy scales have shown that nursing home residents with severe dementia are at in frail older people with dementia, which in turn may result in undertreatment highest risk to be in pain that is neither detected nor adequately treated.45,46 of pain.38 These studies also looked at the role of dementia subtype, but did not find a significant difference between dementia subtypes. In the study by Zwakhalen et Indeed, studies in nursing home residents with dementia have shown al.(2009) the dementia subtype diagnosis was missing in more than half of the undertreatment and/or too low dosages of analgesics.13,49-51 Apart from difficulties participants due to the fact that most of the nursing home residents had only a in assessing pain and restricted analgesic policy, undertreatment has been diagnosis of ‘dementia not otherwise specified’ (NOS) at that 46time. Husebo et attributed to differences in pain perception in connection with dementia subtype, al.(2010) in their study on the psychometric properties of the Mobilization- clinical difficulties in assessing neuropathic pain in people with dementia, Observation-Behaviour-Intensity-Dementia (MOBID-2) Pain Scale found that as well as poor knowledge and training of nursing home staff.26,51,52

14 | GENERAL INTRODUCTION CHAPTER 1 | 15 In contrast to these findings, a few recent studies have shown a tendency to an and the role of dementia subtype. Chapter 6 reports on the longitudinal study increased use of analgesics in people with dementia.47 For example, Haasum et that explored the course of pain and applied pain management strategies after a al.(2011) found that people with dementia had a higher probability to use guideline-based assessment procedure. Chapter 7 reports on the cross-sectional paracetamol than those without dementia, but they found no difference in use of study that investigated the association between pain, neuropsychiatric symptoms opioids or specific drugs to treat neuropathic pain.53 The observed increased use and quality of life in nursing home residents. of analgesic drugs can be perceived as both promising and alarming. It may be considered promising because it might reflect an increased alertness when it Finally, chapter 8 provides a general discussion, which reflects on the results comes to pain in people with dementia. On the other hand, a high frequency of described in the previous chapters. Methodological issues of the study are prescription of analgesic drugs does not necessarily mean that appropriate discussed as well as implications for practice and recommendations for future treatment is being given to the right individual at the right time. To enhance research. appropriate pain treatment in nursing home residents with dementia, more insight in current pain treatment strategies is needed.

Objectives of this thesis

The aim of this dissertation is threefold. First, it aims at gaining insight into the prevalence of pain in people with dementia and factors associated with pain. Second, it aims at investigating the relationship of pain with neuropsychiatric symptoms (e.g. depression and agitation) and quality of life. Third, it aims to investigate the current situation in Dutch nursing homes considering management and the effect of feedback about presence of pain combined with a treatment advice on the course of pain.

Outline of this thesis

To address the objectives of this thesis, we performed both literature research and clinical research. As a recent overview of pain experience data and pain prevalence data in dementia subtypes was lacking, we performed two systematic reviews with the aim to summarize existing knowledge on pain experience (Chapter 2) and pain prevalence (Chapter 3) in the four most common dementia subtypes.

As part of the clinical research, we collected data on the prevalence of pain in a population with different dementia subtypes and in various stages of dementia. Chapter 4 describes the rationale and methods of this observational, cross- sectional, partly longitudinal, cohort study, the “PAINdemiA” study. The results of the “PAINdemiA” study are presented in chapter 5, 6 and 7. Chapter 5 describes the pain prevalence in nursing home residents with dementia

16 | GENERAL INTRODUCTION CHAPTER 1 | 17 References Study of Pain, 2017. Available at: http://www.iasp-pain.org/Taxonomy. Accessed April 2018. 1. Mot E. The Dutch system of long-term care. The Hague: Centraal Planbureau, 2010. 15. Hadjistavropoulos T, Herr K, Turk DC, et al. An interdisciplinary expert 2. RIVM. Aantal mensen met dementie. Bilthoven: Rijksinstituut voor consensus statement on assessment of pain in older persons. Clin J Pain. Volksgezondheid en Milieu, 2016. 2007;23:S1-43. Available at: https://www.volksgezondheidenzorg.info/onderwerp/dementie/ 16. Gibson SJ. IASP global year against pain in older persons: highlighting the cijfers-context/huidige-situatie#node-aantal-mensen-met-dementie. current status and future perspectives in geriatric pain. Expert Rev Accessed April 2018. Neurother. 2007;7:627-35. 3. Verbeek H, van Rossum E, Zwakhalen SM, et al. Small, homelike care 17. Shega JW, Ersek M, Herr K, et al. The multidimensional experience of environments for older people with dementia: a literature review. Int noncancer pain: does cognitive status matter? Pain Med. 2010;11:1680-7. Psychogeriatr. 2009;21:252-64. 18. Takai Y, Yamamoto-Mitani N, Okamoto Y, et al. Literature review of pain 4. Koopmans RT, Lavrijsen JC, Hoek JF, et al. Dutch elderly care physician: a prevalence among older residents of nursing homes. Pain Manag Nurs. new generation of nursing home physician specialists. J Am Geriatr Soc. 2010;11:209-223. 2010;58:1807-9. 19. Ferrell BA, Ferrell BR, Rivera L. Pain in cognitively impaired nursing home 5. Alzheimer’s Disease International. World Alzheimer report 2014. Dementia patients. J Pain Symptom Manage. 1995;10:591-8. and risk reduction, an analysis of protective and modifiable factors. London, 20. Cook AK, Niven CA, Downs MG. Assessing the pain of people with cognitive 2014. impairment. Int J Geriatr Psychiatry. 1999;14:421-5. 6. Hughes JC, Jolley D, Jordan A, Sampson EL. Palliative care in dementia: 21. Feldt KS, Ryden MB, Miles S. Treatment of pain in cognitively impaired issues and evidence. Advances in Psychiatric Treatment. 2007;13:251-260. compared with cognitively intact older patients with hip-fracture. J Am 7. Todd S, Barr S, Roberts M, Passmore AP. Survival in dementia and Geriatr Soc. 1998;46:1079-85. predictors of mortality: a review. Int J Geriatr Psychiatry. 2013;28:1109-24. 22. Morrison RS, Siu AL. A comparison of pain and its treatment in advanced 8. WHO. National cancer control programmes: policies and managerial dementia and cognitively intact patients with hip fracture. J Pain Symptom guidelines. Geneva, World Health Organization, 2002. Manage. 2000;19:240-8. Available at: http://www.who.int/cancer/palliative/definition/en/. 23. Closs SJ, Barr B, Briggs M. Cognitive status and analgesic provision in Accessed April 2018. nursing home residents. Br J Gen Pract. 2004;54:919-21. 9. van der Steen JT, Radbruch L, Hertogh CM, et al. White paper defining 24. Farrell MJ, Katz B, Helme RD. The impact of dementia on the pain optimal palliative care in older people with dementia: a Delphi study and experience. Pain 1996;67:7-15. recommendations from the European Association for Palliative Care. Palliat 25. Scherder EJ, Sergeant JA, Swaab DF. Pain processing in dementia and its Med. 2014;28:197-209. relation to neuropathology. Lancet Neurol. 2003;2:677-686. 10. Beerens HC, Zwakhalen SM, Verbeek H, et al. Factors associated with 26. Achterberg WP, Pieper MJ, van Dalen-Kok AH, et al. Pain management in quality of life of people with dementia in long-term care facilities: patients with dementia. Clin Interv Aging. 2013;8:1471-1482. a systematic review. Int J Nurs Stud. 2013;50:1259-70. 27. Wirths O, Weis J, Kayed R, et al. Age-dependent axonal degeneration in an 11. Corbett A, Husebo B, Malcangio M, et al. Assessment and treatment of pain Alzheimer mouse model. Neurobiol Aging. 2007;28:1689-99. in people with dementia. Nat. Rev. Neurol. 2012;8:264-274. 28. Benedetti F, Vighetti S, Ricco C, et al. Pain threshold and tolerance in 12. Achterberg WP, Gambassi G, Finne-Soveri H, et al. Pain in European long- Alzheimer’s disease. Pain. 1999;80:377-382. term care facilities: cross-national study in Finland, Italy and The Netherlands. 29. Scherder E, Bouma A, Borkent M, Rahman O. Alzheimer patients report Pain. 2010;148:70-4. less pain intensity and pain affect than non-demented elderly. Psychiatry. 13. Plooij B, van der Spek K, Scherder EJ. Pain medication and global cognitive 1999;62:265-272. functioning in dementia patients with painful conditions. Drugs Aging. 30. Jensen-Dahm C, Werner MU, Dahl JB, et al. Quantitative sensory testing 2012;29:377-384. and pain tolerance in patients with mild to moderate Alzheimer disease 14. IASP. IASP Taxonomy. Washington DC: International Association for the compared to healthy control subjects. Pain. 2014;155:1439-1445.

18 | GENERAL INTRODUCTION CHAPTER 1 | 19 31. Jensen-Dahm C, Werner MU, Jensen TS, et al. Discrepancy between 46. Zwakhalen SM, Koopmans RT, Geels PJ, et al. The prevalence of pain in stimulus response and tolerance of pain in Alzheimer disease. Neurology. nursing home residents with dementia measured using an observational 2015;84:1575-81. pain scale. Eur J Pain. 2009;13:89-93. 32. Cole LJ, Farrell MJ, Duff EP, et al. Pain sensitivity and fMRI pain-related 47. Corbett A, Husebo B, Malcangio M, et al. Assessment and treatment of pain brain activity in Alzheimer’s disease. Brain. 2006;129:2957-65. in people with dementia. Nat Rev Neurol. 2012;8:264-274. 33. Oosterman JM, van Harten B, Weinstein HC, et al. Pain intensity and pain 48. AGS panel on persistent pain in older persons. The management of affect in relation to white matter changes. Pain. 2006;125:74-81. persistent pain in older persons. J Am Geriatr Soc. 2002;50:S205-24. 34. Scherder EJ, Plooij B, Achterberg WP, et al. Chronic pain in “probable” 49. Nygaard HA, Jarland M. Are nursing home patients with dementia diagnosis vascular dementia: preliminary findings. Pain Med. 2015;16:442-450. at increased risk for inadequate pain treatment? Int J Geriatr Psychiatry. 35. Hajjar ER, Hanlon JT, Artz MB, et al. Adverse drug reaction risk factors in 2005;20:730-7. older outpatients. Am J Geriatr Pharmacother. 2003;1:82-9. 50. Achterberg WP, Scherder E, Pot AM, Ribbe MW. Cardiovascular risk factors 35. Bastos Leite AJ, Scheltens P, Barkhof F. Pathological aging of the brain: an in cognitively impaired nursing home patients: a relationship with pain? overview. Top Magn Reson Imaging. 2004; 15(6): 369-89. Eur J Pain. 2007;11:707-710. 36. Bathgate D, Snowden JS, Varma A, Blackshaw A, Neary D. Behaviour in 51. Husebo BS, Strand LI, Moe-Nilssen R, et al. Who suffers most? Dementia frontotemporal dementia, Alzheimer’s disease and vascular dementia. Acta and pain in nursing home patients: a cross-sectional study. J Am Med Dir Neurol Scand. 2001;103(6):367-78. Assoc. 2008;9:427-433. 37. Burton EJ, McKeith IG, Burn DJ, Firbank MJ, O’Brien JT. Progression of 52. Scherder EJ, Plooij B. Assessment and management of pain, with particular white matter hyperintensities in Alzheimer disease, dementia with Lewy emphasis on central neuropathic pain, in moderate to severe dementia. bodies, and Parkinson disease dementia: a comparison with normal aging. Drugs Aging. 2012;29:701-706. Am J Geriatr Psychiatry. 2006; 14: 842–849. 53. Haasum Y, Fastbom J, Fratiglioni L, et al. Pain treatment in elderly persons 38. Achterberg WP, de Ruiter CM, de Weerd-Spaetgens CM, et al. with and without dementia: a population-based study of institutionalized [Multidisciplinary guideline ‘Recognition and treatment of chronic pain in and home-dwelling elderly. Drugs Aging. 2011;28:283-293. vulnerable elderly people’]. Ned Tijdschr Geneeskd. 2012;155:A4606. 39. Zwakhalen SM, Hamers JP, Abu-Saad HH, Berger MP. Pain in elderly people with severe dementia: a systematic review of behavioural pain assessment tools. BMC Geriatr. 2006;6:3. 40. Closs SJ, Briggs M. Patients’ verbal descriptions of pain and discomfort following orthopaedic surgery. Int J Nurs Stud. 2002;39:563-572. 41. Pautex S, Michon A, Guedira M, et al. Pain in severe dementia: self- assessment or observational scales? J Am Geriatr Soc. 2006;54:1040-1045. 42. Zwakhalen SM, Hamers JP, Berger MP. The psychometric quality and clinical usefulness of three pain assessment tools for elderly people with dementia. Pain. 2006;126:210-220. 43. Regnard C. Pain is not the only cause of distress in dementia. BMJ 2011;343:d5355; author reply d5356. 44. Jordan A, Regnard C, O’Brien JT, Hughes JC. Pain and distress in advanced dementia: choosing the right tools for the job. Palliat Med. 2012;26:873-8. 45. Husebo BS, Strand LI, Moe-Nilssen R, et al. Pain in older persons with severe dementia. Psychometric properties of the Mobilization-Observation- Behaviour-Intensity-Dementia (MOBID-2) Pain Scale in a clinical setting. Scand J Caring Sci. 2010;24:380-391.

20 | GENERAL INTRODUCTION CHAPTER 1 | 21 Pain experience in Dementia Subtypes: 2 A Systematic Review 2 Pain experience in Introduction Dementia Subtypes: Aging is associated with an increase in the number of painful conditions, e.g., osteoarthritis.1,2 Age is also an important risk factor for dementia, in particular in A Systematic Review dementia due to Alzheimer’s disease (AD).3,4 The prevalence of pain in people suffering from dementia exceeds 50%.5-7

During the last two decades, the number of studies on pain in dementia has Tarik T. Binnekade, Janine van Kooten, Frank Lobbezoo, Didi Rhebergen, Johannes C. increased considerably. However, most studies focus solely on a diagnosis of van der Wouden, Martin Smalbrugge, Erik J.A. Scherder “dementia”, ignoring the fact that dementia concerns a heterogeneous disorder consisting of a number of subtypes. The most prevalent subtypes are AD, vascular dementia (VaD), frontotemporal dementia (FTD), and dementia with Lewy Bodies Abstract (DLB)].8,9 The neuropathology partly overlaps, but also differs between the various subtypes. Particularly the neuropathology, characterizing each subtype, may have Recently, the number of studies focusing on pain in dementia has increased its own influence on how pain is processed.10 considerably. Still, little attention has been paid to the influence of the neuropathology of different dementia subtypes on pain experience. AD is characterized by grey matter atrophy, e.g., in the hippocampus.11,12 The In 2003, a review identified several studies that indicated a relation between hippocampus is involved in the processing of motivational-affective aspects of dementia subtype and pain experience. Now, ten years later, an update is pain.10 Atrophy of the hippocampus may therefore lead to an increase in pain warranted. We conducted a systematic review to identify studies that assessed tolerance,13-16 implying a decrease in pain experience.10 On the other hand, pain experience and dementia subtypes by searching PubMed, Embase, PsycINFO, increased activity in brain regions involved in the processing of affective CINAHL, and Cochrane Library. Inclusion criteria were: (1) major dementia components of pain has been observed in a mild stage of AD,17 which implies that subtype diagnosis, i.e., Alzheimer’s dementia (AD), Vascular dementia (VaD), pain experience may be unaltered or even increased in this stage of AD. The frontotemporal dementia (FTD), Dementia with Lewy Bodies (DLB); (2) age ≥ 60 lateral pain system, which processes the sensory-discriminative aspects of pain years; and (3) pain experience. We identified twelve studies that addressed AD, experience, remains largely intact in AD, resulting in an intact sense of pain three studies VaD, one study FTD, and no studies DLB. location and intensity.10 In AD, studies on clinical pain indicate a reduced pain experience compared to controls, whereas experimental studies show inconsistent findings. In VaD, clinical Next to grey matter atrophy, AD is also characterized by white matter lesions.18 studies found that primary caregivers rated pain equal to cognitively intact White matter lesions may disrupt connections between cortical and subcortical controls, although more painful locations were reported. During self-report, brain areas (de-afferentiation),19 resulting in an increase in both pain experience elderly with VaD reported higher pain levels than cognitively intact controls. and pain prevalence. More specifically, white matter lesions may disrupt In FTD, a significantly lower pain sensitivity to experimental pain was found. connections between the intralaminar thalamic nuclei and somatosensory areas Considering the limited number of studies, these findings should be considered (SI, SII), resulting in abnormal sensations (dysesthesia).20 In VaD, white matter with caution. Existing literature provides some evidence that dementia subtype lesions are the most prominent neuropathology.21 Indeed, there are some affects pain experience. Further research is needed to clarify the relation between indications that patients with VaD may experience an increase in pain.22,23 In FTD, dementia subtype and pain experience as it could serve as basis for improving the neuronal loss and spongiform changes are seen in the frontal, lateral temporal, assessment and management of pain in people with dementia. and parietal brain areas.24,25 The areas affected in FTD are strongly related to emotional states and motivation.26 This may explain why a loss of awareness of Key messages: Alzheimer, dementia, Frontotemporal dementia, Lewy Bodies, pain was reported significantly more often by caregivers of elderly with FTD than pain experience, pain, Vascular dementia. those of elderly with AD or VaD.27 In DLB, the amount of white matter lesions is similar to that in AD.28 However, grey matter such as the amygdala and medial

24 | PAIN EXPERIENCE IN DEMENTIA SUBTYPES: A SYSTEMATIC REVIEW CHAPTER 2 | 25 thalamic nuclei are also affected by DLB.29 Like the hippocampus, the amygdala Methods and the medial thalamic nuclei are involved in the affective aspects of pain, implying that the affective pain experience, or, suffering from pain, might be Search decreased in this type of dementia.10 A systematic search of major electronic databases was conducted from inception until September 2014. Databases searched were PubMed, EMBASE, CINAHL, The possible alterations in pain experience in the various subtypes of dementia PsycINFO, and the Cochrane Library. Search terms included were subtypes of emphasize the clinical relevance of assessing pain in dementia; undiagnosed dementia (Alzheimer disease OR Vascular dementia OR Multi-infarct dementia changes in pain experience may have serious consequences. An increase in pain OR Frontotemporal lobar degeneration OR frontotemporal dementia OR Pick’s experience that remains unnoticed may lead to undertreatment of pain, whereas disease OR Lewy Body disease OR Lewy Body Dementia) and pain in the title or a decrease in pain experience may lead to serious bodily harm as patients may abstract, or medical subject headings. Medical subheadings were adapted to not react appropriately to potential damaging situations, e.g., hot water.27 each electronic database. There was no restriction with respect to study design, as we expected to find few published papers determining dementia subtypes. Particularly in view of the difference in neuropathology between the various One of the two reviewers (TB) searched the databases as detailed above and both subtypes of dementia and its putative consequences for pain experience, a review reviewers (TB/JvK) searched for additional studies. Additional studies were in 2003 included those studies that examined pain experience in these distinct identified by hand searching of reference lists of included papers andby groups of patients.10 The review included both clinical and experimental pain contacting experts. Duplicates were removed and subsequently, the two studies. The clinical pain studies compared the pain experience of elderly with reviewers independently conducted a two-stage inclusion process: (1) eligibility dementia and cognitively intact controls who were faced with similar chronic screening of title and abstracts, after which consensus between both reviewers painful conditions. The experimental pain studies applied standardized pain was reached on the list of studies proceeding to the following stage; and (2) stimuli to both elderly with dementia and cognitively intact controls. The main screening based on full text papers. If in doubt at stage 1 and/or if the abstract conclusions were that patients with AD, and especially those with FTD, may was absent, the full-text paper was obtained for further selection. experience a decrease in pain experience, whereas VaD patients may report an increase in pain experience. Clinical studies on DLB were not available. The reviewers’ selections of studies were compared and in case of disagreement, Limitations of that review were, that it was not conducted systematically, and the a third reviewer (ES) was asked for advice to reach consensus. Studies were number of studies per subtype was low (AD, VaD and FTD) or no studies were included if they: (1) involved elderly (60+) patients diagnosed with one or more available (DLB). Now, more than 10 years later, seems a timely opportunity to subtypes of dementia (i.e., Alzheimer’s Disease, Vascular Dementia, update the 2003 review by using up-to-date methodology. Therefore we Frontotemporal Dementia, Dementia with Lewy Bodies); (2) included clinically systematically searched databases for studies, which focused on pain experience and/or experimentally derived measurements on pain experience. All studies in the various dementia subtypes. were included irrespective of the presence of a control group. Studies were excluded if they: (1) involved patients without a known dementia subtype The goals of the present review were to: (1) systematically identify and appraise diagnosis or patients with a diagnosis of dementia other than the four subtypes both clinical and experimental studies that focus on pain experience in one of the named earlier; (2) involved chronic malignant pain; or (3) were not based on four major dementia subtypes, i.e., AD, VaD, FTD, and DLB; and (2) to give an primary empirical data, or were only published as an abstract, case report, update of the conclusions drawn in the review published in 2003. editorial, or letter. No language restriction was utilized. Studies on malignant pain were excluded, because pain due to cancer is often considered separately from non-malignant pain. The nature of the disease and the treatment of the disease reduce comparability with non-malignant pain studies.

Primary authors were contacted when studies reported one or more dementia subtypes, but did not display the results of the subtypes separately. Papers were subsequently excluded if the authors were not able to provide the required data.

26 | PAIN EXPERIENCE IN DEMENTIA SUBTYPES: A SYSTEMATIC REVIEW CHAPTER 2 | 27 Quality assessment Finally, sixteen studies met the inclusion criteria, seven of these were also included in the review from 2003. No additional studies from before 2003 were All studies were assessed for methodological quality using the Newcastle- identified that were not included in the earlier review. One study thatwas Ottawa Scale (NOS).30 The NOS assesses quality of non-randomized studies by included in 2003 did not meet criteria for the current review, because it did not judging a study on three main concepts: the selection of the study groups; the include clinically or experimentally derived pain measures. comparability of the groups; and the ascertainment of the outcome. The NOS has predefined scoring criteria, but requires further specification depending on the topic of study. In the current study, the NOS was adapted as follows: (1) the case Figure 1. PRISMA 2009 Flow diagram of search strategy.34 definition should include CT/MRI imaging and follow international criteria for dementia subtype diagnosis and (2) groups should be comparable for age and Records identified through gender or one other relevant factors associated with pain, like the use of pain database searching (n = 2408) medication, comorbid anxiety and depression or other factors depending on the study design. Identification ≥ Records after duplicates removed Records excluded The maximum NOS score is 9 points. Studies that score 5 points are generally (n = 1606) 31-33 (n = 802) considered to be of satisfactory methodological quality. Studies with scores Reasons for exclusion: less than 5 should be interpreted with caution, for there is an elevated risk of - No subtyping bias. The quality assessment in the current study was performed independently - No main diagnosis of by two authors (TB/JvK). Consensus was reached through discussion. dementia - No outcome measurement on pain Data extraction and analysis - Not published as primary empirical data Two reviewers (TB/JvK) independently extracted data on: (1) the characteristics Screening Records screened on titles - No full text available and no of the study sample (e.g., sample size, gender, dementia subtype); (2) pain and/or abstract (n = 1606) response of the author intensity and pain affect in clinical studies; (3) pain threshold and pain tolerance in experimental studies; and (4) conclusions of the included studies. Because we expected considerable heterogeneity in study design, case mix and measurement instruments, no attempts were made to perform a meta-analysis. Additional records identified Full-text articles excluded through other sources (n = 1488) (n = 2) Reasons for exclusion: - Results not specified per

Eligibility dementia subtype Results - No main diagnosis of dementia Full-text articles assessed - Malignant cancer pain for eligibility Study selection - No clinical or experimental (n = 120) The initial electronic search yielded 2408 hits: 475 from PubMed, 1151 from study Embase, 437 from PsycINFO, 311 from CINAHL, and 34 from the Cochrane - Author e-mailed additional information but does not Library. After removing duplicates, there were 1606 unique hits, see figure 1. Two meet eligibility criteria additional publications were identified by reference checking, so in total 1608 Studies included in - Author did not have access

articles were checked for eligibility based on title and abstract. A total of 120 Included qualitative synthesis to required data (n = 16) - Not published as primary articles proceeded to the full text screening. At this phase, eleven authors were empirical data contacted for additional information, and nine studies were subsequently excluded. Following the full text review, a total of 104 studies were excluded.

28 | PAIN EXPERIENCE IN DEMENTIA SUBTYPES: A SYSTEMATIC REVIEW CHAPTER 2 | 29 Study characteristics

In total, 16 studies were identified that focused on pain in dementia subtypes. the other 15 studies had a case-control design. Sample sizes of the case-control Characteristics of the included studies are presented in tables 1 and 2. One study studies were generally small ranging from 27 to 63,36,37 while the cross-sectional was reported with extra not published data.35 Eight of these were clinical studies study with 173 participants was considerably larger.35 All studies included both and 8 were experimental studies. One study was a cross-sectional study,35 elderly with a subtype of dementia and a cognitively intact control group.

Table 1. Clinical studies reporting on the intensity of pain in patients with dementia.

Dementia subtype Author, year, country Design NOS score Sample Mean Cognition Outcome Pain intensity Pain affect Findings (SD) Mean (SD) measures Mean (SD) Mean (SD) AD, VaD, AD+VaD Husebo et al. (2008), Norway Crosssectional 7/9 N= 173 AD N = 47 MOBID-2 AD 2.4 (2.2) Observed pain intensity did not differ study 84.7 (6.7) years VaD N = 72 VaD 2.5 (1.7) between subtypes. Patients with VaD, 25% male AD+VaD N = 32 AD+VaD 2.9 and VaD+AD had more pain locations Nursing home ND N = 30 (1.9) than AD. MMSE 12.3 (9.7) ND 2.2 (1.5) Diagnosis based on clinical findings and CT. Although CT was present in 71% of cases. AD Scherder et al. (1999), Case-control 5/9 N = 37 AD N = 19 VAS VAS NWC-A Elderly with AD reported experiencing The Netherlands study AD 84.2 (n.a.) years CST 9.4 NWC-A AD 5.5* AD .21 lower pain intensities and less suffering ND 86.6 (n.a.) (range: 8.5-13) MPI-DLV ND 22.4 ND .94 from pain than controls when matched years NHP-Pain for the presence of chronic painful AD 10.5% male ND N = 18 NHP-Pain* MPI-DLV* conditions. ND 5.6% male CST 17.5 AD 7.9 AD .32 No difference in age, gender or Nursing home (range: 14-20) ND 38.6 education. Diagnosis of “probable” AD. AD Scherder & Case-control 4/9 N = 60 Early AD CAS CAS* Significance levels: Early-Mid*, Bouma. study Early AD 86.8 years N = 20 Early AD 2.9 Control-Early* control Mid*. Patients (2000), (range: 75-95) MMSE-12 Midstage AD with AD reported significantly lower pain The Netherlands Midstage AD 82.3 8.0 (range: 7-10) 1.5 intensities when matched for chronic years (range: ND 8.4 painful conditions. The ND group had 76-92) Midstage AD significantly more comorbid conditions ND 87.1 years N = 20 including hypertension. Results for FAS (range: 76-96) MMSE-12 and FPS were presented although test 3.7 (range: 1-6) comprehension was <80% in dementia Early AD 17.5% groups. Therefore excluded from this male ND N = 20 review. Midstage AD 20% MMSE-12 male 11.4 ND 15% male (range: 11-12) Nursing home AD Scherder et Case-control 4/9 N = 31 AD N =14 CAS CAS-intensity * CASaffect* Pain intensity and affect were measured al. (2001), study AD 87.9 years MMSE-12 FPS at 3 intervals, multiple times per day. AD The Netherlands (range: 78-99) 7.4 (range: 6-9) NWC-A FPS* NWC-A* reported significantly lower pain intensity ND 87.4 years and pain affect at all intervals. Results (range: 78-97) ND N = 17 CNPI CNPI not reported due to high number of AD 14% male MMSE-12 measurements. No significant difference ND 18 % male 10.2 (range: 7-12) in observed pain behavior. Nursing home

30 | PAIN EXPERIENCE IN DEMENTIA SUBTYPES: A SYSTEMATIC REVIEW CHAPTER 2 | 31 Dementia subtype Author, year, country Design NOS score Sample Mean Cognition Outcome Pain intensity Pain affect Findings (SD) Mean (SD) measures Mean (SD) Mean (SD)

AD Scherder & van Manen Case-control 4/9 N = 37 AD N = 20 CAS CAS intensity CAS affect Self-reported pain both in rest and after (2005), The Netherlands study AD 84.5 years FPS At rest At rest walking a short distance. CNPI only (range: 74-98) MMSE CNPI AD 16.1 after walking. ND 85.8 years 18.7 (24.9) AD 14.4 (range: 72-93) (range: 12-24) ND 31.3 (23.2) Only significant difference between AD Nursing home (32.4) ND 28.8 and ND on pain intensity after walking. ND N = 17 (31.7) MMSE 26.3 After walking Same number of chronic painful (range: 25-30) AD 12.8 After conditions. (22.3)* walking ND 30.3 AD 13.3 (33.8) (23.5) ND 25.2 FPS (28.9) At rest AD 0.9 (1.7) CNPI ND 1.6 (1.7) After walking After walking AD 0.9 AD 1.5 (1.7) (1.3) ND 0.8 (1.6) ND 0.8 (1.2) AD Scherder et al. (2008), The Case-control 7/9 N= 39 AD N = 19 CAS CAS intensity CAS-affect Patients with AD reported significantly Netherlands study AD 86.4 (5.3) years MMSE 19.6 (1.8) FPS T(0)* T(0) lower pain intensity and pain affect at ND 85.7 (6.4) years NWC-A AD 21.6 AD 15.8 baseline (T0) and 1 month later (T1) AD 11% male ND N = 20 (22.7) ND (21.3) ND ND 20% male MMSE 27.3 (2.0) 50.8 (23.2) 41.2 (29.0) Inclusion criterion was the presence of Nursing home arthrosis/arthritis of lower extremity T(1)* T(1) * joints or lumbar spine. AD 27.0 AD 20.9 (29.7) ND (27.2) ND Clinical diagnosis of “probable” AD. 45.9 (23.4) 40.8 (27.6)

FPS NWC-A T(0)* T(0)* AD 0.6 (0.7) AD 2.0 ND 2.4 (1.7) (2.7) ND 4.4 T(1)* (2.7) AD 0.8 (1.1) ND 2.3 (1.5) T(1) AD 2.4 (3.6) ND 3.8 (3.0)

32 | PAIN EXPERIENCE IN DEMENTIA SUBTYPES: A SYSTEMATIC REVIEW CHAPTER 2 | 33 Dementia subtype Author, year, country Design NOS score Sample Mean Cognition Outcome Pain intensity Pain affect Findings (SD) Mean (SD) measures Mean (SD) Mean (SD)

VaD Scherder et al. (2003), Case-control 5/9 N = 40 VaD N = 20 CAS CAS-intensity CAS-Affect* Self-reported pain intensity and pain The Netherlands study VaD 82.5 (9.2) years MMSE-12 VAS VaD 30.0 VAD 30.2 affect were similar for VaD and ND. ND 84.7 (4.6) years 7.3 (1.3) FPS (34.8) (35.3) Observed pain behavior did also not VaD 25% male NWC-A ND 17.5 ND 13.2 differ between VaD and ND. Still, VaD ND 16% male ND N = 20 CNPI (22.3) (20.2) group took significantly more Nursing home MMSE-12 acetaminophen than controls. 10.7 (1.3) FPS* NWC-A VaD 2.0 (1.4) VaD 2.8 Patients had diagnosis of “possible” VaD. ND 1.3 (1.5) (3.9) Matched for chronic painful conditions. ND 1.7 AD and ND not the same type of chronic (2.6) painful conditions. In addition, VaD had more comorbid conditions. CNPI VaD .55 (1.05) (N = 10) ND .20 (.52) VaD Scherder et al. (2014), Case-control 6/9 N = 42 VaD CAS CAS intensity CAS affect* The pain scores were summed scores The Netherlands study VaD 80.9 (4.9) years N = 20 FPS VaD 103.8 VaD 102.3 taken at baseline and 2 months later. ND 85.6 (5.6) years MMSE (59.4) (53.4) The probable VaD patients in a mild to VaD 30% male 19.5 ND 74.1 ND 59.2 moderate stage of dementia reported a ND 18.2% male (range: 11-24) (43.2) (38.8) higher pain affect on the modified CAS. Nursing home No significant difference in analgesic use ND FPS was found. A diagnosis of “probable” VaD N= 22 VaD 4.0 (2.3) including CT (85% of cases). Inclusion MMSE ND 3.1 (2.7) criterion for both groups was the 27.0 presence of arthritis/arthrosis or (range: 25-29) osteoporosis.

AD, Alzheimer’s Disease; VaD, Vascular Dementia; ND, No Dementia; CAS, Coloured Analogue Scale; CST, Cognitive Screening Test; CNPI, Checklist of Non-verbal Pain Indicators; FPS, Faces Pain Scale; NHP, Nottingham Health Profile; NWC-A, Number of Words Chosen- Affective; MPI-DLV, Multidimensional pain inventory - Dutch language version; MPQ-DLV, McGill Pain questionnaire- Dutch language version; VAS, Visual Analogue Scale; NOS, Newcastle-Ottawa Scale; *p<.05.

34 | PAIN EXPERIENCE IN DEMENTIA SUBTYPES: A SYSTEMATIC REVIEW CHAPTER 2 | 35 Table 2. Experimental studies reporting on the intensity of pain in patients with dementia.

Dementia subtype Author, year, country Pain stimuli NOS score Sample Mean Cognition MMSE Outcome Pain threshold Pain tolerance Findings (SD) mean (SD) measures Mean (SD) or (CI) Mean (SD) or (CI) AD Benedetti et al. (1999), Italy Electrical and 4/9 N = 48 AD N = 24 mA AD 15.9 (2.2) AD 41.3 (5.8)* In AD pain thresholds were ischemic AD 64.5 (4.7) years Average (electrical) ND 15.8 (2.7) ND 32.7 (4.9) similar to controls. Pain ND 62.9 (5.8) years n.a. (range: 10-19) tolerance was increased in AD 35.7% male ND N = 24 minutes AD 5.5 (2.7) AD 22.7 (7.8)* AD. As MMSE decreased ND 45.8% male MMSE n.a. (ischemic) ND 5.4 (2.9) ND 14.3 (6.2) the pain tolerance Outpatient clinic increased for both electric and ischemic pain. AD Rainero et al. (2000), Electrical 4/9 N = 40 AD N = 20 mA AD 14.1 (2.2) Pain intensity 2x The autonomic responses Italy AD 68.6 (3.8) years 13 (2.8) (electrical) ND 15.1 (2.7) pain threshold to painful stimuli depend ND 68.8 (3.2) years ND N = 20 on the intensity of the AD 40% male MMSE NRS NRS NRS stimulus and not on the ND 50% male n.a. AD=ND AD 5.3 (1.3) * pain experience. Outpatient clinic ND 8.5 (1.0) Selfreported pain intensity is similar for AD and control at baseline. At 2x pain threshold the selfreported pain intensity is significantly blunted in AD. Pain intensity rating at 2x threshold is closely related to MMSE score. AD Gibson et al. (2001), Heat laser 5/9 N = 30 AD N = 15 watts AD = ND Pain sensitivity and Australia stimulation AD 82.4 (1.4) years 12.7 (CO2 laser) processing of acute ND 81.6 (1.7) years (range: 2-19) painful stimuli remain AD 20% male ND N = 15 intact despite cognitive ND 33% male 29.73 impairment in AD, as Setting unclear (range: 29-30) measured with cerebral event related potentials (CERP). AD Cole et al. (2006), Australia Mechanical 7/9 N = 29 AD N = 14 kg s/m2 Weak pain AD required higher AD 79.0 (4.8) years 19.4 (5.7) (mechanical) AD 3.9 (1.1)* mechanical pressure to ND 78.6 (4.8) years ND 3.0 (1.0) elicit weak pain and AD 50% male ND N = 15 moderate pain. The self ND 46.7% male 29.3 (0.1) Moderate pain reported suffering from Outpatient clinic AD 5.9 (1.3) the pain was similar for ND 5.4 (1.4) both AD and ND. Functional connectivity of the lateral and medial pain network not reduced in mild to moderate AD. AD Lints-Martindale et al. Electrical 6/9 N = 63 AD N = 27 Weak pain Moderate pain No significant differences (2007), Australia and mechanical Electrical ND N = 26 were found between AD stimulation group and ND with respect to the AD 78.4 (4,0) years ES-group mA Electrical Electrical selfreported unpleasant­ ND 78,6 (4,3) years AD N = 13 (electrical) AD 4.2 (3.6) AD 5.3 (4.7) ness of the electrical and 21.2 (5.4) ND 4.0 (3.2) ND 5.2 (3.9) mechanical pain stimuli. Mechanical ND N = 22 stimulation group 28.8 (2.2) kg s/m2 Mechanical Mechanical AD 80.0 (6.3) years (mechanical) AD 3.3 (1.2) AD 5.2 (2.1) ND 78.3 (3.8) years MS-group ND 2.7 (1.0) ND 4.6 (1.2) AD N = 14 Gender n.a. 22.8 (3.9) Outpatient clinic ND N = 14 29.4 (0.9)

36 | PAIN EXPERIENCE IN DEMENTIA SUBTYPES: A SYSTEMATIC REVIEW CHAPTER 2 | 37 Dementia subtype Author, year, country Pain stimuli NOS score Sample Mean Cognition MMSE Outcome Pain threshold Pain tolerance Findings (SD) mean (SD) measures Mean (SD) or (CI) Mean (SD) or (CI)

AD Limongi et al. (2013), Italy Mechanical, 4/9 N = 27 AD N = 10 Mechanical n.s. Thermal tolerance Mechanical pressure, thermal, and 80.9 (8.0) years ND N = 17 n.s. average of measurements electrical 37% male Electrical n.s. Electrical +30% on three fingers of both Nursing home above threshold hands. n.s. Exact numbers not presented, no significant difference was found between AD and ND. AD Jensen-Dahm et al. (2014), Thermal, 6/9 N= 58 AD N = 29 °C (thermal) Thermal Pain threshold and Denmark mechanical, AD 70.9 (6.9) years 22.4 (1.2) Median AD 41.2°C tolerance measurements and cold ND 70.5 (5.5) years ND N= 29 (IQR) (40.0°C - 42.4°C) were taken in 3 sessions pressor AD 48.3% male 29.7 (0.5) ND 42.3°C spread over 2 days. Mean ND 48.3% male (41.1°C - 43.5°C) Mechanical * (thermal) or median Outpatient clinic AD 213 kPa (mechanical, CPT) values kPa Mechanical (188 - 306) are displayed. (mechanical) AD 120 kPa ND 289 kPa Median (IQR) (100 - 142) (262 - 360) Reduced mechanical ND 131 kPa pressure tolerance in mild (113 - 192) CPT to moderate AD patients. AD 31.3 seconds CPT (21.2 - 50.7) CAS scores for +0.5 and Seconds AD 14.9 seconds ND 28.1 seconds +2.5 °C heat pain (cold (11.7 - 20.5) (20.2 - 45.6) threshold were calculated. pressor) ND 11.8 seconds Scores did not differ Median (IQR) (9.4 - 20.0) between groups. No difference in selfreported suffering from pain. FTD Carlino et al. (2010), Italy Electrical 6/9 N = 41 FTD mA Right hand * Right hand Pain threshold is FTD 66.1 (8.2) N = 23 (electrical) FT 28.0 (16.9) FT 41.1 (14.5)* increased in both FT and years 21.7 (6.1) FAB 25.9 FAB 33.9 (16.9) FAB compared to ND. Pain ND 63.3 (6.4) years FT group (17.4) ND 28.9 (10.3) tolerance is only increased Outpatient clinic ND N = 18 based on ND 14.9 (4.6) in FT group compared to MMSE SPECT control. n.a. frontal and/ or temporal hypoper­ fusion

FAB group Left hand * Left hand Based on FT 26.3 (17.3) FT 39.8 (14.9)* Frontal FAB 25.2 FAB 35.7 (16.3) Assessment (16.9) ND 27.3 (11.4) Battery ND 14.8 (6.1) scores

AD, Alzheimer’s Disease; FTD, Frontotemporal dementia; ND, No Dementia; mA, milliampere; NRS, Numeric Rating Scale; ES, Electrical Stimulation; MS, Mechanical Stimulation; NOS, Newcastle-Ottawa Scale; IQR, Inter Quartile Range; CI, confidence interval; *p<.05.

38 | PAIN EXPERIENCE IN DEMENTIA SUBTYPES: A SYSTEMATIC REVIEW CHAPTER 2 | 39 Methodological quality Alzheimer’s disease

Quality assessment scores were calculated for the 16 included studies in this Findings from clinical studies review using the NOS. The NOS-scores were as follows: six studies had a score In the review from 2003, three clinical studies were identified.15,16,41 These studies of 4/9, three had a score of 5/9, four of 6/9, and three 7/9. The results of the compared patients with AD living in nursing homes with cognitively intact elderly quality assessment are displayed in Appendix 1 and 2. on measures of self-reported pain. Both groups were similar with regard to age, gender, and education, while they also had a comparable number of chronic painful conditions and comorbid conditions. All three studies found that elderly with AD reported to experience less intense pain than cognitively intact elderly. Study outcome measures In addition, elderly with AD reported to suffer less from their pain than cognitively intact elderly. All sixteen studies compared elderly with dementia with cognitively intact controls. Sensory-discriminative aspects of pain are reported in the form of pain Since the review of 2003, three more clinical studies were identified.35,42,43 intensity in clinical studies and pain intensity and/or pain threshold in Two studies assessed self-reported pain on two occasions.42,43 The study by experimental studies. Emotional-affective aspects of pain are reported as pain Scherder & van Manen (2005)43 assessed self-reported pain in rest and after a affect in clinical studies and pain affect and/or pain tolerance in experimental short walk, in AD and controls with a comparable number of chronic painful studies. conditions. The AD group reported to experience less intense pain after a short walk than controls. However, they did not report suffering less from their pain at The clinical studies did not administer painful stimuli but instead included either baseline or after a short walk than cognitively intact elderly. Interestingly, participants with similar painful conditions. A difference in pain experience is the personal nursing assistant evaluated the pain of the patients much higher hence attributed to the presence of a certain dementia subtype. The outcome than the patients themselves. This is not consistent with the second study,42 measures are self-reported pain intensity and pain affect, and in three studies which assessed self-reported pain in elderly suffering from arthrosis/arthritis of the observed pain intensity was included.15,23,35 lower extremity joints or lumbar spine, at baseline and after one month, without any intervention. Elderly with AD reported experiencing less intense pain and In the eight experimental studies participants were administered experimental less suffering from their pain at both baseline and after one month. Still, one pain stimuli, including mechanical, electrical, thermal, and ischemic pain clinical study did not support the notion that elderly with AD experience less pain stimuli. In all studies, the pain threshold was determined and in three studies than cognitively intact elderly.35 Nursing home patients in multiple stages of the pain tolerance level.13,36,38 According to the International Association for the dementia and with different dementia subtypes (i.e., AD, VaD, AD+VaD) were Study of Pain (IASP), the pain threshold is “the minimum intensity of a stimulus compared with cognitively intact controls; no difference in observed pain intensity that is perceived as painful”.39 The IASP specifies the pain tolerance as “The was found between elderly with AD and the other dementias or control group.35 maximum intensity of a pain-producing stimulus that a subject is willing to accept in a given situation”. Two studies included weak and moderate pain Findings from experimental studies stimuli using the self-report Gracely Box Scale.17,37,40 In the 2003 review, three studies were identified that administered pain stimuli to elderly with AD and controls.13,14,44 These studies applied different pain stimuli Not all studies specified the definition of the pain threshold or pain tolerance. including: heat laser stimulation,44 ischemic pain,13 and electrical pain stimuli.13,14 The absence of a clearly stated definition may lead to misinterpretation as is the All three studies found that the pain threshold is similar for both AD and elderly case in one study,36 which reports to study the thermal pain threshold, but gives without dementia. Only two studies applied stronger pain stimuli, either at pain the definition of the thermal pain tolerance level. tolerance level,13 or two times pain threshold.14 Both studies found that the self- rated pain intensity is lower for AD than for controls for these stronger pain stimuli, whilst the electrical stimulation needed to reach pain tolerance is higher. Interestingly, both studies found a positive correlation between MMSE score and

40 | PAIN EXPERIENCE IN DEMENTIA SUBTYPES: A SYSTEMATIC REVIEW CHAPTER 2 | 41 self-reported pain intensity for the strongest pain stimuli, indicating that elderly However, they do report a significantly higher pain affect, or suffering from pain. in a more severe stage of AD may be less sensitive to strong pain stimuli. In the study of Husebo and collegues, pain in VaD patients was evaluated through observation with the MOBID-2 pain scale by the primary caregivers. VaD patients For this update, four more experimental studies were identified.17,36-38 Pain did not differ significantly from those without dementia. However, the number of stimuli included: thermal stimuli,36,38 mechanical pressure stimuli,17,36-38 electrical pain locations (e.g., shoulder, hip, knee, back) was significantly higher in the VaD stimuli, and a cold pressor test.38 A study applying mechanical pressure to the group and in the AD+VaD group compared to the AD group.35 thumbnail of AD patients in a mild to moderate stage and cognitively intact elderly found that AD patients do not report a significantly lower pain affect as a Findings from experimental studies reaction to the mechanical stimuli than controls.17 In addition, fMRI scans No experimental studies were identified. indicated that the neuronal activity in the medial and lateral pain system, responsible for the suffering and perception of pain, respectively,10 was even higher in AD patients than in controls. The authors conclude that the emotional Frontotemporal dementia experience of pain, or pain affect, is not attenuated in AD patients. In another experimental study,37 mechanical and electrical painful stimuli were applied to Findings from clinical studies AD patients and older persons without dementia. In reaction to painful stimuli, No clinical studies were identified. no significant differences were found between both groups, neither in the rating of unpleasantness, nor in facial expressions. In a third experimental pain study, Findings from experimental studies pain was provoked in patients with mild to moderate AD and controls by More than ten years ago, the assumption that FTD patients may experience less mechanical pressure, thermal stimulation, and a cold pressor test 38. pain, emerged from only one study,27 which did not meet inclusion criteria for the AD patients showed a lower pain tolerance for mechanical pressure, compared current review. In that study, almost half of the carers of FTD patients reported a to controls; the responses to the cold pressure test and the thermal stimulation loss of awareness of pain. Since then, only one additional study could be found were similar for both groups. A fourth experimental pain study provoked pain by that examined pain in FTD. This study was an experimental study applying thermal, electrical, and mechanical stimuli.36 In contrast to the above-mentioned electrical pain stimuli to FTD patients and controls.46 Two FTD subgroups were studies, in reaction to painful stimuli, a decrease in the affective processing of made based on clinical and neuroimaging characteristics. One group was pain, measured by sensory evoked potentials, was observed in patients with mild composed of patients with executive dysfunctions, e.g., abstract reasoning and to moderate AD, compared to controls. mental flexibility, assessed by neuropsychological tests. The other group consisted of patients with frontal and/or temporal hypoperfusion, determined by single photon emission computed tomography (SPECT). The results show that the Vascular dementia pain threshold was significantly higher in the group with executive dysfunctions, compared to the controls. In addition, a significantly higher pain threshold and pain Findings from clinical studies tolerance was observed in the group of patients with frontal and/or temporal In 2003, one study was identified that focused on VaD patients.23 Patients with hypoperfusion, compared to those without dementia. The results of both studies “possible” VaD were compared with cognitively intact controls who were matched point to a possible decrease in pain experience in FTD. for chronic painful conditions. The elderly with VaD reported experiencing pain at a higher intensity and suffering more from their pain than controls. An important limitation of this study was that the diagnosis of VaD was not confirmed by CT/ Dementia with Lewy Bodies MRI or autopsy. No studies were identified. Since 2003, two pain studies included VaD patients.35,45 The results of the most recent study show that “probable” VaD patients in a mild to moderate state of dementia, report a similar pain intensity as cognitively intact controls.45

42 | PAIN EXPERIENCE IN DEMENTIA SUBTYPES: A SYSTEMATIC REVIEW CHAPTER 2 | 43 Discussion This lack of studies may be explained by the fact that VaD is a highly heterogeneous disorder, with a wide range of neuropathology and associated The present systematic review provides an overview of currently available studies clinical implications.49 comparing pain experience in different dementia subtypes. We set out to systematically identify and appraise both clinical and experimental studies that Frontotemporal Dementia focus on pain experience in the major dementia subtypes, i.e., AD, VaD, FTD and In 2003, one study that interviewed primary caregivers of elderly with FTD DLB. In addition, we give an update of the conclusions drawn in the review by indicated that elderly with FTD experience less pain than elderly suffering from Scherder and colleagues from 2003.10 AD or VaD.27 Since then, only one experimental study was found that focused on pain in FTD.46 The view that patients with FTD may experience less pain than Alzheimer’s disease controls still stands. An important nuance has come to light. Carlino and In 2003, it was suggested by both clinical15,41 and experimental studies 13,14 that colleagues found that FTD patients with strictly localized SPECT frontal and/or elderly with AD experience less pain than cognitively intact controls. Only one temporal hypoperfusion showed a significant increase in both pain threshold and experimental study did not report a difference in pain experience between these pain tolerance, whereas the FTD group based on executive dysfunctions groups.44 Recent clinical studies are less consistent in their findings reporting measured by neuropsychological tests showed only an increase in pain either a decrease in self-reported pain experience or a pain experience threshold.46 comparable to cognitively intact controls.35,42,43 Four experimental studies including elderly with AD have been published since 2003.17,36-38 One study finds Dementia with Lewy Bodies reduced affective processing of pain, measured by sensory evoked potentials, in In 2003, no studies were identified that focus on pain experience in DLB. Now, elderly with mild to moderate AD compared to controls.36 Still, three experimental after more than ten years still no studies were found. One explanation for the studies do not support the notion of a reduced pain experience in AD.17,37,38 Elderly lack of studies in this field is that these patients may be studied within trials with AD and cognitively intact controls had similar pain threshold and tolerance assessing pain in Parkinson’s disease (PD), of which there are many. 50 In fact, values and did not report suffering more from their pain than cognitively intact Lewy Body neuropathology is a key feature of PD. 50,51 Pain is an important non- elderly. These findings are in congruence with the functional cerebral imaging motor symptom associated with PD.52 Prevalence estimations indicate that 40 studies that did not find reduced functionality of the lateral and medial pain to 60% of PD patients may suffer from one or multiple painful conditions. 53 system in AD.17,44 Other possible explanations are that DLB is often misdiagnosed as AD and that AD and DLB often co-occur.54,55 Still, estimates indicate that DLB may be the Vascular dementia second most prevalent dementia subtype when diagnosed correctly,56,57 therefore So far, pain experience in VaD has hardly been studied. In 2003, only one clinical it is imperative to first improve DLB subtype assessment after which more study was identified. It was reported that elderly with VaD experience more pain studies should focus on pain in this disorder. than cognitively intact elderly, when faced with similar painful conditions.23 Since then, two more clinical studies were identified.35, 45 The three clinical studies on Theoretical considerations pain in VaD reported inconsistent results, two found higher self-reported pain There are inconsistent findings concerning the pain experience of elderly with a intensity and/or affect in elderly with VaD,23,45 whereas one study did not find an similar dementia subtype. One possible explanation lies in the heterogeneity of increased pain experience in VaD using an observational instrument.35 However, even the dementia subtypes themselves. AD can be divided in three distinct in this study the number of pain locations (e.g., shoulder, hip, knee, back) was neuropathological subtypes based on the amount of neurofibrillary tangles in the significantly higher in the VaD group and in the AD+VaD group compared to the cortex and hippocampus, i.e., typical AD, hippocampal-sparing AD, and limbic- cognitively intact elderly. Compared to controls, the higher experience of pain in predominant AD,58,59 all of which may have a different influence on pain VaD may be a result of damage to the spino-thalamo-cortical tract, known as post- experience. In FTD, the influence of neuropathological heterogeneity on pain stroke or deafferentiation pain.47,48 White matter lesions are characteristic for experience has already been shown.46 Moreover, VaD is characterized by multiple VaD,21 but may also present in AD,18 making it possible that deafferentiation pain is subtypes, and a large heterogeneity in neuropathology.49 a factor in both dementia subtypes. No experimental pain studies were identified.

44 | PAIN EXPERIENCE IN DEMENTIA SUBTYPES: A SYSTEMATIC REVIEW CHAPTER 2 | 45 Strengths and limitations groups of 47-72 patients.35 The studies that rely on self-report show a high There are a number of limitations to the studies included in this review. Most variability in pain scores. This variability applies to patients, but also cognitively clinical studies only supplied a clinical diagnosis of either AD or VaD, which is not intact controls. The combination of small sample sizes and large variability in the substantiated by CT or MRI. Although this is common in clinical practice, at least in outcome measures may lead to underpowered studies, which in turn may result the Netherlands, this poses a problem for studying this population. Neuroimaging in both Type I and Type II errors.63 In addition, according to our assessment using reduces the risk of including patients with combined pathology, e.g., AD+VaD, or the NOS, at least six studies had an insufficient methodological quality. Removing other comorbid neuropathology that influences pain experience. Most of the these studies would however not have changed the present findings. experimental studies did include neuroimaging in the dementia subtype diagnosis. Clinical implications Another limitation of current studies on pain in dementia subtypes is the The struggle to find out whether pain experience in patients with dementia ascertainment of pain. Self-report is considered the gold standard of pain increases, remains constant, or even decreases, might explain the variety in the assessment, however self-report instruments become increasingly unreliable as extent in which analgesics are prescribed to patients with dementia. A number of the dementia progresses.41,60 This may have had the most influence on the clinical studies indicate that elderly with dementia receive less analgesics than studies in which patients in a more severe stage of dementia were studied. For this cognitively intact controls.2,64-68 Still, others find no difference in the use of reason, patients’ comprehension of the self-report instruments was first evaluated, analgesics35,69,70 or even a higher use in elderly with dementia.71 Although the which resulted in a number of participants’ scores being excluded in clinical variety in analgesic prescription might parallel the possible alterations in pain studies.15,23 A number of studies utilized pain observation tools, although the experience, the overall clinical judgment is that patients with dementia who question remains how well these observations correspond with the patients’ own are no longer able to indicate that they are in pain, have a higher risk for under­ pain experience.15,43 Currently, an international group of researchers is developing treatment.72 The risk for undertreatment of pain is particularly high in patients a new instrument for the assessment of pain in cognitively impaired persons called with vascular problems, i.e., patients with AD+VaD35 and VaD.10,27 The increase in the: “Pain assessment in Impaired Cognition (PAIC) meta-tool”.61 This instrument pain in these patients might reflect central neuropathic pain. Due to the which will be based on a combination of currently available pain observation progressive degeneration of white matter, patients may experience an increase instruments, expert opinion and American Geriatric Society guidelines. The aim is in pain but at the same time, are increasingly less able to communicate about it. to develop a methodologically sound instrument that can be apply internationally In those cases, the medical staff should be aware that (an increase in) pain may in both clinical practice and research. express behaviorally, e.g., physical inactivity.73

The strength of the included experimental pain studies is the ability to apply a Recommendations for future research standardized pain stimulus and directly link it to relatively natural self-report, i.e., Future research should focus on a stricter group definition, based on both clinical pain present and pain intolerable, pain observation and functional brain imaging. presentation and neuropathology. A diagnosis of, for example, Alzheimer’s Therefore, these findings may provide a more reliable insight into possible disease, or FTD alone, may not be sufficient to make firm predictions about pain alterations in pain experience in patients with dementia than clinical pain studies, experience. In clinical studies, the influence of mixed dementia pathology and where the participants require cognitive functions, suchs as, memory and abstract other comorbid neuropathology should be adressed by adding CT or MRI cerebral reasoning, to translate their pain to a numerical score or drawn image of a face. imaging. In addition, larger sample sizes and the use of more comparable pain On the other hand, the question remains whether clinical chronic pain, assessed assessment measures are required in order to report more robust findings. by self-report, can be easily compared to pain evoked by experimental pain stimuli. The challenge for future studies lies in studying the relationship between clinical A recent review from studies on pain in patients with schizophrenia found that pain and neuropathology in patients with dementia. Especially patients suffering experimental pain studies indicate that schizophrenic patients may be insensitive from VaD, FTD and DLB neuropathology deserve more attention when it comes to pain, whereas the few available studies on clinical pain indicate the opposite.62 to pain. Alternative pain measures such as facial responses, autonomic responses, and functional brain imaging should be further explored in order to Most studies included in the present review involve only small experimental groups reduce the influence of cognitive functioning and subjective interpretation on ranging from 14-29 patients in all but one study which includes experimental pain assessment.

46 | PAIN EXPERIENCE IN DEMENTIA SUBTYPES: A SYSTEMATIC REVIEW CHAPTER 2 | 47 Conclusion

In conclusion, this systematic review identified 16 studies which assess pain experience in dementia subtypes. In 2003, it was suggested that AD may reduce pain experience. More recent studies are less consistent, finding both altered and unchanged pain experience in AD. In VaD, the notion of an increase in pain experience is not yet falsified. In FTD, the view of a decrease in pain experience still stands. For patients with DLB, even after 10 more years no studies were identified.

The present findings should be interpreted with caution, given that a number of studies have limited methodological quality or are underpowered. In addition, given the subjective nature of current pain assessment instruments, and the influence of cognitive decline on pain report, a degree of variability inpain experience may be attributed to measurement error. This review further underlines the importance of focusing on clear clinical and neuropathological profiles influencing pain experience in dementia and on the search for more objective instruments to assess pain in dementia.

48 | PAIN EXPERIENCE IN DEMENTIA SUBTYPES: A SYSTEMATIC REVIEW CHAPTER 2 | 49 References 17. Cole LJ, Farrell MJ, Duff EP, Barber JB, Egan GF, Gibson SJ. Pain sensitivity and fMRI pain-related brain activity in Alzheimer’s disease. Brain. 1. Gibson SJ, Lussier D. Prevalence and relevance of pain in older persons. 2006;129(Pt 11): 2957–65. Pain Med. 2012;13 Suppl 2: S23–6. 18. O’Brien JT. Role of imaging techniques in the diagnosis of dementia. Br J 2. Horgas AL, Elliott AF. Pain assessment and management in persons with Radiol. 2007;80: S71–7. dementia. Nurs Clin North Am. 2004;39(3): 593–606. 19. Mori E. Impact of Subcortical Ischemic Lesions on Behavior and Cognition. 3. Mayeux R, Stern Epidemiology of Alzheimer disease. Cold Spring Harb Ann N Y Acad Sci. 2002;977(1): 141–8. Perspect Med. 2012;2(8). 20. Schmahmann JD. Parietal Pseudothalamic Pain Syndrome. Arch Neurol. 4. Skoog I. Psychiatric epidemiology of old age: the H70 study - the NAPE 1992;49(10): 1032–7. Lecture 2003. Acta Psychiatr Scand. 2003;109(1): 4–18. 21. Roman G, Benavente O. The neuropathology of vascular dementia. Semin 5. Black BS, Finucane T, Baker A, et al. Health problems and correlates of Cerebrovasc Dis Stroke. 2004;4(2): 87–96. pain in nursing home residents with advanced dementia. Alzheimer Dis 22. F arrell MJ, Katz B, Helme RD. The impact of dementia on the pain Assoc Disord. 2006;20(4): 283–90. experience. Pain. 1996;67(1): 7–15. 6. Leong IY-O, Nuo TH. Prevalence of pain in nursing home residents with 23. Scherder EJA, Slaets J, Deijen JB, et al. Pain Assessment in Patients with different cognitive and communicative abilities. Clin J Pain. 2007;23(2): 119–27. Possible Vascular Dementia. Psychiatry Interpers Biol Process. 2003;66(2): 7. Zwakhalen SMG, Koopmans RTCM, Geels PJEM, Berger MPF, Hamers JPH. 133–45. The prevalence of pain in nursing home residents with dementia measured 24. Varma AR, Adams W, Lloyd JJ, et al. Diagnostic patterns of regional atrophy using an observational pain scale. Eur J Pain. European Federation of Chapters on MRI and regional cerebral blood flow change on SPECT in young onset of the International Association for the Study of Pain. 2009;13(1): 89–93. patients with Alzheimer’s disease, frontotemporal dementia and vascular 8. Brunnström H, Gustafson L, Passant U, Englund E. Prevalence of dementia dementia. Acta Neurol Scand. 2002;105(4): 261–9. subtypes: a 30-year retrospective survey of neuropathological reports. Arch 25. Bastos Leite AJ, Scheltens P, Barkhof F. Pathological aging of the brain: an Gerontol Geriatr. 2009;49(1): 146–9. overview. Top Magn Reson Imaging. 2004;15(6): 369–89. 9. Aarsland D, Rongve A, Nore SP, et al. Frequency and case identification of 26. Varrone A, Pappatà S, Caracò C, et al. Voxel-based comparison of rCBF dementia with Lewy bodies using the revised consensus criteria. Dement SPET images in frontotemporal dementia and Alzheimer’s disease highlights Geriatr Cogn Disord. 2008; 26(5): 445–52. the involvement of different cortical networks. Eur J Nucl Med Mol Imaging. 10. Scherder EJA, Sergeant JA, Swaab DF. Pain processing in dementia and its 2002;29(11): 1447–54. relation to neuropathology. Lancet Neurol. 2003;2(11): 677–86. 27. Bathgate D, Snowden JS, Varma A, Blackshaw A, Neary D. Behaviour in 11. Braak H, Braak E. Staging of alzheimer’s disease-related neurofibrillary frontotemporal dementia, Alzheimer’s disease and vascular dementia. Acta changes. Neurobiol Aging. 2005; 16(3): 271–8. Neurol Scand. 2001;103(6): 367–78. 12. Reid AT, Evans AC. Structural networks in Alzheimer’s disease. 28. Barber R, Scheltens P, Gholkar A, et al. White matter lesions on magnetic Eur Neuropsychopharmacol. 1999;23(1): 63–77. resonance imaging in dementia with Lewy bodies, Alzheimer’s disease, 13. Benedetti F, Vighetti S, Ricco C, Bergamasco B, Pinessi L, Rainero I. Pain vascular dementia, and normal aging. J Neurol Neurosurg Psychiatry. threshold and tolerance in Alzheimer’s disease. Pain. 1999;80(1-2): 377–82. 1999;67(1): 66–72. 14. Rainero I, Vighetti S, Bergamasco B, Pinessi L, Benedetti F. Autonomic 29. Mirzaei S, Knoll P, Koehn H, Bruecke T. Assessment of diffuse Lewy body responses and pain perception in Alzheimer’s disease. Eur J disease by 2-18F.fluoro-2-deoxy-D-glucose positron emission tomography Pain.2000;4(3): 267–74. (FDG PET). BMC Nucl Med. 2003;3(1):1. 15. Scherder E, Bouma A, Slaets J, et al. Repeated Pain Assessment in 30. Wells GA, Shea B, O’Connell D, et al. The Newcastle-Ottawa Scale (NOS) Alzheimer’s Disease. Dement Geriatr Cogn Disord. 2001;12(6): 400–7. for assessing the quality of nonrandomised studies in meta-analyses. 16. Scherder E, Bouma A, Borkent M, Rahman O. Alzheimer patients report 1993. p. 198. Available at: less pain intensity and pain affect than non-demented elderly. Psychiatry. http://www.ohri.ca/programs/clinical_epidemiology/oxford.asp. Accessed 2014. 1999;62(3): 265–72. 31. Kwon BK, Roffey DM, Bishop PB, Dagenais S, Wai EK. Systematic review:

50 | PAIN EXPERIENCE IN DEMENTIA SUBTYPES: A SYSTEMATIC REVIEW CHAPTER 2 | 51 occupational physical activity and low back pain. Occup Med (Lond). dioxide laser stimulation in patients with Alzheimer’s disease and age- 2011;61(8): 541–8. matched control volunteers. Pain Res Manag. 2001;6(3): 126–32. 32. Lo W-K, Chan WW. Proton pump inhibitor use and the risk of small intestinal 45. Scherder EJA, Plooij B, Achterberg WP, et al. Chronic Pain in “Probable” bacterial overgrowth: a meta-analysis. Clin Gastroenterol Hepatol. Vascular Dementia: Preliminary Findings. Pain Med. 2015;16: 442-50. 2013;11(5): 483–90. 46. Carlino E, Benedetti F, Rainero I, et al. Pain perception and tolerance in 33. Stubbs B, Binnekade T, Eggermont L, Sepehry AA, Patchay S, Schofield patients with frontotemporal dementia. Pain. 2010;151(3): 783–9. P. Pain and the risk for falls in community-dwelling older adults: 47. Vestergaard K, Nielsen J, Andersen G, Ingeman-Nielsen M, Arendt- systematic review and meta-analysis. Arch Phys Med Rehabil. 2014;95(1): Nielsen L, Jensen TS. Sensory abnormalities in consecutive , unselected 175–87. patients with central post-stroke pain. Pain. 1995;61: 177–86. 34. Moher D, Liberati A, Tetzlaff J, Altman DG. Reprint--preferred reporting 48. Hong JH, Bai DS, Jeong JY, et al. Injury of the spino-thalamo-cortical items for systematic reviews and meta-analyses: the PRISMA statement. pathway is necessary for central post-stroke pain. Eur Neurol. 2010;64(3): Phys Ther. 2009;89: 873–80. 163–8. 35. Husebo BS, Strand LI, Moe-Nilssen R, Husebo B, Aarsland D, Ljunggren 49. Wallin A, Milos V, Sjögren M, Pantoni L, Erkinjuntti T. Classification and AE. Who suffers most? Dementia and pain in nursing home patients: a subtypes of vascular dementia. Int Psychogeriatr. 2003;15(S1): 27–37. cross-sectional study. J Am Med Dir Assoc. 2008;9(6): 427–33. 50. Wasner G, Deuschl G. Pains in Parkinson disease--many syndromes under 36. Limongi F, Radaelli S, Noale M, Maggi S, Crepaldi G. Somatosensory Evoked one umbrella. Nat Rev Neurol. 2012;8(5): 284–94. Potentials and pain assessment in Alzheimer’s disease. Eur Geriatr Med. 51. Braak H, Tredici K Del, Rüb U, de Vos RA, Jansen Steur EN, Braak E. 2013;4(6): 384–8. Staging of brain pathology related to sporadic Parkinson’s disease. 37. Lints-Martindale AC, Hadjistavropoulos T, Barber B, Gibson SJ. Neurobiol Aging. 2003;4(2): 197–211. A psychophysical investigation of the facial action coding system as an index 52. Barone P, Antonini A, Colosimo C, et al. The PRIAMO study: A multicenter of pain variability among older adults with and without Alzheimer’s disease. assessment of nonmotor symptoms and their impact on quality of life in Pain Med. 2007; 8(8): 678–89. Parkinson’s disease. Mov Disord. 2009;24(11): 1641–9. 38. Jensen-Dahm C, Werner MU, Dahl JB, et al. Quantitative sensory testing 53. Borsook D. Neurological diseases and pain. Brain. 2011;135(2): 320–44. and pain tolerance in patients with mild to moderate Alzheimer disease 54. Barker WW, Luis CA, Kashuba A, et al. Relative frequencies of Alzheimer compared to healthy control subjects. Pain. 2014;155(8): 1439–45. disease, Lewy body, vascular and frontotemporal dementia, and hippocampal 39. IASP. IASP Taxonomy. Washington DC, International Association for the sclerosis in the State of Florida Brain Bank. Alzheimer Dis Assoc Disord. Study of Pain, 2017. Available at: http://www.iasp-pain.org/Taxonomy. 2002;16(4): 203–12. Accessed 2014. 55. Hansen L, Salmon D, Galasko D, et al. The Lewy body variant of Alzheimer’s 40. Gracely RH, Kwilosz DM. The Descriptor Differential Scale: applying disease: a clinical and pathologic entity. Neurology. 1990;40(1): 1–8. psychophysical principles to clinical pain assessment. Pain. 1988;35(3): 56. Barker WW, Luis CA, Kashuba A, et al. Relative frequencies of Alzheimer 279–88. disease, Lewy body, vascular and frontotemporal dementia, and hippocampal 41. Scherder EJA, Bouma A. Visual Analogue Scales for Pain Assessment in sclerosis in the State of Florida Brain Bank. Alzheimer Dis Assoc Disord. Alzheimer’s Disease. Gerontology. 2000;46(1): 47–53. 2002;16(4): 203–12. 42. Scherder EJA, Eggermont L, Plooij B, et al. Relationship between chronic 57. Aarsland D, Rongve A, Nore SP, et al. Frequency and case identification of pain and cognition in cognitively intact older persons and in patients with dementia with Lewy bodies using the revised consensus criteria. Dement Alzheimer’s disease. The need to control for mood. Gerontology. 2008;54(1): Geriatr Cogn Disord. 2008;26(5):445–52. 50–8. 58. Janocko NJ, Brodersen KA, Soto-Ortolaza AI, et al. Neuropathologically 43. Scherder E, van Manen F. Pain in Alzheimer’s disease: nursing assistants’ defined subtypes of Alzheimer’s disease differ significantly from and patients’ evaluations. J Adv Nurs. 2005;52(2):151–8. neurofibrillary tangle-predominant dementia. Acta Neuropathol. 44. Gibson SJ, Voukelatos X, Ames D, Flicker L, Helme RD. An examination of 2012;124(5): 681–92. pain perception and cerebral event-related potentials following carbon 59. Whitwell JL, Dickson DW, Murray ME, et al. Neuroimaging correlates of

52 | PAIN EXPERIENCE IN DEMENTIA SUBTYPES: A SYSTEMATIC REVIEW CHAPTER 2 | 53 pathologically defined subtypes of Alzheimer’s disease: a case-control 73. Plooij B, Scherder EJA, Eggermont LHP. Physical inactivity in aging and study. Lancet Neurol. 2012;11(10): 868–77. dementia: a review of its relationship to pain. J Clin Nurs. 2012;21: 3002–8. 60. Kunz M, Scharmann S, Hemmeter U, Schepelmann K, Lautenbacher S. The facial expression of pain in patients with dementia. Pain. 2007;133(1-3): 221–8. 61. Corbett A, Achterberg W, Husebo B, et al. An international road map to improve pain assessment in people with impaired cognition: the development of the Pain Assessment in Impaired Cognition (PAIC) meta-tool. BMC Neurol. 2014;14(1): 229. 62. Engels G, Francke AL, van Meijel B, et al. Clinical pain in schizophrenia: a systematic review. J Pain. 2014;15(5): 457–67. 63. Christley RM. Power and Error: Increased Risk of False Positive Results in Underpowered Studies. Open Epidem J. 2010;3(1): 16–9. 64. Balfour JE, O’Rourke N. Older adults with Alzheimer disease, comorbid arthritis and prescription of psychotropic medications. Pain Res Manag. 2003;8(4):198–204. 65. De Souto Barreto P, Lapeyre-Mestre M, Vellas B, Rolland Y. Potential underuse of analgesics for recognized pain in nursing home residents with dementia: a cross-sectional study. Pain. 2013;154(11): 2427–31. 66. Hoffmann F, van den Bussche H, Wiese B, Glaeske G, Kaduszkiewicz H. Diagnoses indicating pain and analgesic drug prescription in patients with dementia: a comparison to age- and sex-matched controls. BMC Geriatr. 2014;14(1): 20. 67. Morrison RS, Siu AL. A Comparison of Pain and Its Treatment in Advanced Dementia and Cognitively Intact Patients with Hip Fracture. J Pain Symptom Manage. 2000;19(4): 240–8. 68. Reynolds KS, Hanson LC, DeVellis RF, Henderson M, Steinhauser KE. Disparities in pain management between cognitively intact and cognitively impaired nursing home residents. J Pain Symptom Manage. 2008;35(4): 388–96. 69. Lövheim H, Karlsson S, Gustafson Y. The use of central nervous system drugs and analgesics among very old people with and without dementia. Pharmacoepidemiol Drug Saf. 2008;17(April): 912–8. 70. Nygaard HA, Jarland M. Are nursing home patients with dementia diagnosis at increased risk for inadequate pain treatment? Int J Geriatr Psychiatry. 2005;20(8): 730–7. 71. Haasum Y, Fastbom J, Fratiglioni L, Kåreholt I, Johnell K. Pain treatment in elderly persons with and without dementia: a population-based study of institutionalized and home-dwelling elderly. Drugs Aging. 2011;28(4): 283–93. 72. Herr K, Coyne PJ, McCaffery M, Manworren R, Merkel S. Pain assessment in the patient unable to self-report: position statement with clinical practice recommendations. Pain Manag Nurs. 2011;12(4): 230–50.

54 | PAIN EXPERIENCE IN DEMENTIA SUBTYPES: A SYSTEMATIC REVIEW CHAPTER 2 | 55 Appendix

Appendix 1. Quality assessment Clinical Studies

Study Selection Comparability Exposure Score same method of patient representativeness of selection definition Gender or ascertainment non response Age ascertainment for Total definition patients of controls of controls other of pain rate cases and controls Husebo et al., 2008 MET MET MET MET UNMET UNMET MET MET MET 7/9 Scherder et al., 1999 UNMET UNMET MET MET UNMET MET MET MET UNMET 5/9 Scherder & Bouma, 2000 UNMET UNMET UNMET MET UNMET MET MET MET UNMET 4/9 Scherder et al., 2001 UNMET UNMET UNMET MET UNMET MET MET MET UNMET 4/9 Scherder et al., 2003 UNMET UNMET MET MET UNMET MET MET MET UNMET 5/9 Scherder & van Manen, 2005 UNMET UNMET MET MET UNCLEAR UNCLEAR MET MET UNCLEAR 4/9 Scherder et al., 2008 UNMET MET MET MET MET MET MET MET UNCLEAR 7/9 Scherder et al., 2014 MET UNMET MET MET UNMET MET MET MET UNMET 6/9

Appendix 2. Quality Assessment Experimental Studies

Study Selection Comparability Exposure Score same method of patient representativeness of selection definition Gender or ascertainment non response Age ascertainment for Total definition patients of controls of controls other of pain rate cases and controls Benedetti et al., 1999 MET UNMET UNMET MET UNMET UNMET MET MET UNCLEAR 4/9 Carlino et al., 2010 MET MET UNMET MET MET UNMET MET MET UNCLEAR 6/9 Cole et al., 2006 MET UNMET UNMET MET MET MET MET MET MET 7/9 Gibson et al., 2001 UNCLEAR UNMET UNMET MET MET MET MET MET UNCLEAR 5/9 Jensen-Dahm et al., 2014 MET UNMET UNMET MET MET MET MET MET UNMET 6/9 Limongi et al., 2013 UNMET UNMET MET MET UNMET UNMET MET MET UNMET 4/9 Lints-Martindale et al., MET UNMET UNMET MET MET UNMET MET MET MET 6/9 2007 Rainero et al., 2000 MET UNMET UNMET MET UNMET UNMET MET MET UNCLEAR 4/9

56 | PAIN EXPERIENCE IN DEMENTIA SUBTYPES: A SYSTEMATIC REVIEW CHAPTER 2 | 57 A review of pain prevalence in Alzheimer’s, vascular, frontotemporal and Lewy 3 body dementias 3 A review of pain prevalence Introduction in Alzheimer’s, vascular, Pain in people with dementia is a common problem, and in view of the demographic changes and aging world population more people can be expected to suffer from frontotemporal and Lewy both dementia and a painful condition in the near future.1-4 The prevalence of pain in people with dementia has been estimated to be approximately 50%,3 and it has body dementias been suggested that the prevalence of pain differs between dementia subtypes.5

Neuropathological changes differ between dementia subtypes, and these are held responsible for the decline in function as well as for alterations in pain Janine van Kooten, Tarik T. Binnekade, Johannes C. van der Wouden, Max L. Stek, perception.3,5 For example, pain perception in Alzheimer’s disease (AD) may be Erik J.A. Scherder, Bettina S. Husebø, Martin Smalbrugge, Cees M.P.M. Hertogh altered because of a loss of grey matter, which leads to an increased pain tolerance without a change in pain threshold.6,7 Abstract Pain experience in vascular dementia (VaD) may increase because of white matter Background: Numerous studies have reported on pain in dementia. It has been lesions, while pain experience in frontotemporal dementia (FTD) may decrease hypothesized that pain perception differs between dementia subtypes, and because atrophy predominates in the frontal cortex.8 This area is related to therefore, prevalence of pain differs between dementia subtypes. However, there emotional states and motivation, which may explain the frequently reported loss remains a paucity of evidence on differences in prevalence of pain in different of awareness of pain people with FTD.9 Pain perception in dementia with Lewy dementia subtypes. This review aimed to determine the prevalence of pain for bodies (DLB) may be altered due to brain damage caused by Lewy bodies and the major dementia subtypes: Alzheimer’s disease (AD), vascular dementia cortical atrophy.10 As our understanding of the underlying neuropathology has (VaD), frontotemporal dementia (FTD) and dementia with Lewy bodies (DLB). increased, it seems worthwhile to take into account dementia subtype in the management of pain in people with dementia. However, prevalence of pain has Summary: We found 10 studies that met our inclusion criteria. Most of these not been determined systematically for each dementia subtype.8,11 studies reported on AD; studies reporting the prevalence of pain in people with DLB were scarce and for FTD, we found no studies. The sample-weighted Our objectives were to determine the prevalence of pain per dementia subtype in prevalence of pain could only be calculated for AD, VaD and mixed dementia: AD published studies that focused specifically on one of the four major dementia 45.8% (95% confidence interval, CI: 33.4% - 58.5%), VaD 56.2% (95% CI: 47.7% subtypes (i.e. AD, VaD, FTD and DLB) in samples from both the community and - 64.4%) and mixed dementia 53.9% (95% CI: 37.4% - 70.1%). long term care facilities (LTCFs).12 Because of the differences in neuropathological changes, we hypothesized that the prevalence of pain differs between the four Key messages: Studies investigating the prevalence of pain in dementia dementia subtypes. subtypes were scarce; however, we found a high prevalence of pain in dementia without significant differences between the dementia subtypes. More studies are required to draw firm conclusions on differences in prevalence of pain between dementia subtypes. Methods

Search Strategy We searched MEDLINE, EMBASE, CINAHL and PsycINFO for articles about patients with a known dementia subtype and pain. Databases were searched from inception until August 2014. The search strategy consisted of a combination

60 | A REVIEW OF PAIN PREVALENCE IN ALZHEIMER’S, VASCULAR, FRONTOTEMPORAL AND LEWY BODY DEMENTIAS CHAPTER 3 | 61 of free terms and medical subject headings (e.g. MeSH) relating to pain and the Internal validity is defined as the possible amount of error in measuring the 4 commonest subtypes of dementia (i.e. AD, VaD, FTD and DLB). Articles were conditions , and was addressed by the following question: screened for dementia subtype and pain in the title or abstract text. Additionally, Was an acceptable pain assessment tool used for the assessment of pain? we identified studies by handsearching reference lists of published literature For each criterion, three options were possible: ‘+’ = low risk of bias, ‘-’ = possible reviews on pain in dementia and studies that met our inclusion criteria. risk of bias or ‘?’= risk of bias unclear (due to poor reporting). Each study was There were no language restrictions. rated independently by two reviewers. A total score was calculated as the sum of the individual criteria using a score of ‘1’ for low risk of bias and a score of ‘0’ for Selection of Studies possible or unclear risk of bias. Scores for the criterion ‘nonreponse’ were only For inclusion, studies had to meet the following criteria: presenting primary data reported in two studies; therefore we decided to exclude this criterion from the including prevalence of pain in patients with a dementia subtype diagnosis. total score. We qualitatively evaluated individual studies for similarities and Studies were independently screened for eligibility on titles and abstract by two differences in study design and results. reviewers (T.T.B. /J.v.K.) and after both reviewers reached consensus, they reviewed the full text of the thus selected papers. Studies that reported pain for Analysis both cognitively intact and cognitively impaired patients were included if they The sample weighted averages with 95% confidence intervals (CI) were calculated reported the results separately to enable data extraction of pain prevalence per for subgroups that consisted of ≥100 participants. We used a random-effects dementia subtype. We excluded studies that used pain as an inclusion or model if the I2 statistic was greater than 50%, and a fixed-effect model if the2 I exclusion criterion, studies on induced pain (e.g. Alcolea et al.13), case reports or statistic was under 50%.15,16 Statistical analyses were performed using MedCalc for case series. Authors were contacted when studies reported one or more Windows, version 5.5- 32 bit (MedCalc Software, Ostend, Belgium). In subgroups dementia subtypes, but did not display the results of the subtypes separately. consisting of <100 participants we confined ourselves to descriptive analyses. Papers were subsequently excluded if the authors were not able to provide the required data or did not respond to our request.

Data Extraction and Synthesis Results Two reviewers independently extracted data on study design, setting, population, dementia subtype, prevalence of pain and measurement instruments. For one The initial electronic search yielded 2,374 hits: 475 from MEDLINE, 1,151 from study, published in Spanish, data was extracted by only one reviewer, who had EMBASE, 311 from CINAHL and 437 from PsycINFO. After removing duplicates, a knowledge of the Spanish language. total of 1,709 articles were screened for eligibility based on title and abstract. Another 4 publications were identified by reference checking and a total of 85 We assessed the external and internal validity of study findings using the underwent full text review (fig. 1). We contacted 16 authors for further information Methodological Evaluation of Observational research (MORE) checklist, which and five of them provided additional information. The main reasons for exclusion was adapted to the specific research question.14 were no data on prevalence of pain and no data on dementia subtype. Eventually, 10 studies met all inclusion criteria. The characteristics and main findings of External validity is defined as the extent to which the results of a study can be included studies are described in table 1. generalized to the target population. For this study, the checklist comprised of the following questions: All studies were published in the last 10 years. Eight studies were conducted in • Was the sampling frame a close representation of the target population? Europe,17-25 and two studies in the USA.26,27 Five studies examined outcomes in • Was an appropriate case definition used for dementia (subtype)? nursing homes,23-27 whereas five studies examined outcomes in community • Was dementia stage measured in the target population? dwelling people or patients recruited from outpatient clinics.17,18,20-22 Five studies • Was some form of random selection used to select the sample? were cross-sectional,17,21,24,25,27 two longitudinal observational studies,19,23 one • Was the likelihood of non-response bias minimal? was a case-control study,18 one a retrospective cohort study,26 and one a randomized controlled trial.20 The mean subject age ranged from 70 to 86 years.

62 | A REVIEW OF PAIN PREVALENCE IN ALZHEIMER’S, VASCULAR, FRONTOTEMPORAL AND LEWY BODY DEMENTIAS CHAPTER 3 | 63 Figure 1. Study selection process. one study the Discomfort Behavioural Scale (DBS),26 one study a questionnaire that included pain domains,19 and in one study, the physician assessed if pain Records identified through was present.23 database searching (n = 2,374) Risk Assessment Most studies did not meet all criteria for validity for observational studies. The

Records after duplicates removed Records excluded nonreponse rate was only reported for two studies. With respect to the external (n = 1,709) (n = 1,628) validity, the risk of bias was considered possible in most studies, and the most Reasons for exclusion: frequent source of bias was found in the random selection of participants. - No subtyping Additionally, with respect to the internal validity, the risk of bias of the - No main diagnosis of dementia measurement of pain was considered low in most of the studies. Only two studies - No outcome measurement used a non-specific measurement tool, and in one study, a proxy was asked to fill Records screened on titles on pain in or help to fill in the EQ-5D (table 2). and/or abstract - Not published as primary (n = 1,709) empirical data - No full text available and no Overall Sample response of the author The 10 published reports provided information on 51,810 participants. Except for Additional records identified one study, which provided information on 49,627 patients, the sample size ranged through other sources (reference checking) Full-text articles excluded from 48 to 929. Most studies included more women than men, except for the (n = 4) (n = 75) studies of Colosimo et al19 and Boström et al.,18 which both included people with Reasons for exclusion: DLB. The mean age of the participants in LTCFs was above 80 years, and the - No subtyping mean age of the other participants (i.e. outpatient and community-dwelling Full-text articles assessed - No main diagnosis of for eligibility dementia people) was <80 years. Nine studies provided cross-sectional data on pain (n = 85) - No outcome measurement prevalence, one study measured pain at the end of life.26 on pain - Author e-mailed additional info but does not meet Prevalence of pain in dementia subtypes 17,18,20,21,23-27 Studies included in eligibility criteria Nine studies reported the proportion of pain in people with AD. Eight qualitative synthesis - Author did not respond or studies used cross sectional data and one study reported data from a case (n = 10) did not have access to required data control study. The sample of Alzheimer patients was 50,911, including one study - Not published as primary with 49,627 patients, and age ranged from 48 to 103 years. The mean prevalence empirical data of pain in AD in our study was 45.8% (95% CI: 33.4% - 58.5%) with a substantial heterogeneity (I²=98%; fig.2a).

Three studies focused solely on patients with AD,17,21,22,27 five studies included Three studies reported on pain prevalence solely in people with VaD.22,23,26 These patients with AD, VaD, mixed pathologies and/or DLB,20,23-26 one study was a case studies collected data on more than one dementia subtype all studies were control study of AD and DLB,18 and one study included dementia subtypes based carried out in LTCFs. The sample of patients with VaD was a homogeneous on motor neuron symptoms (i.e. parkinsonian syndromes).19 No study reported subgroup (I²=0%) that consisted of 142 participants and the mean prevalence of the prevalence of pain specific to patients with FTD. pain in VaD was 56.2% (95% CI: 47.7% - 64.4%; fig. 2b).

Different instruments for pain measurement were reported; four studies used We found mixed dementia (AD and VaD) to be reported frequently as a separate the EQ-5D,17,18,20,21 two studies the Minimum Data Set (MDS),25,27 one study the entity next to AD and VaD, and therefore, we presented pain prevalence for mixed Mobilization-Observation-Behaviour-Intensity-Dementia-2 (MOBID-2) Pain Scale,24 dementia separately. Four studies reported on mixed dementia; three studies

64 | A REVIEW OF PAIN PREVALENCE IN ALZHEIMER’S, VASCULAR, FRONTOTEMPORAL AND LEWY BODY DEMENTIAS CHAPTER 3 | 65 used cross-sectional data and one study provided data on prevalence using the Three studies reported on DLB, whereas one study combined Parkinson Disease baseline data of a randomized controlled trial.20,23-25 The sample of patients with and DLB into one group. After contact with the author, it appeared that they could mixed dementia was 649, and the mean prevalence of pain in mixed dementia was not be untangled, in their sample were 18 people and nine (50%) of them suffered 53.9% (95% CI: 37.4% - 70.1%) with a substantial heterogeneity (I²=91%; fig. 2c). from pain23. The study of Boström et al18 was a case-control study that included 34 outpatients with AD and 34 outpatients with DLB. The prevalence of pain (n=24) in DLB was 70% and in those with AD (n=32) 30%. The study by Colosimo et al19 collected data on non-motor symptoms in parkinsonian syndromes and Figure 2.  Forest plot for effect of the sample-weighted pain prevalence in AD ( a ), VaD ( b ) and mixed dementia ( c ). they included 14 people with DLB, of whom seven (50%) indicated that they were in pain. We did not find enough studies to calculate a sample-weighted prevalence Baquero et al. 200917 for DLB. Boström et al. 200718 Buchanan et al. 200527 Alzheimer’s disease We found no studies on FTD, but we retrieved one study concerning two Hendriks et al. 201523 conditions closely related to FTD: corticobasal degeneration (CBD) and 24 Husebo et al. 2008 progressive supranuclear palsy (PSP).19,28 Colosimo et al.19 collected data from 11 Jensen-Dahm et al. 201220 patients with CBD and from 30 patients with PSP. Among those with CBD, A Oliva-Moreno et al. 201021 Volicer et al. 200925 4 patients (37%) indicated that they had pain, and among those with PSP, Monroe et al. 201226 7 patients expressed pain (23%).

Total (random effects)

0 20 40 60 80 100 0 20 40 60 80 2 100 0Pain prevalence20 (%)40 60 80 I = 98%100 Discussion

In this systematic review, we reviewed the current literature on pain in dementia Vascular dementia Hendriks et al. 201523 to present the prevalence of pain per dementia subtype, although, we found not Husebo et al. 200824 enough studies to report a sample-weighted prevalence for each dementia Monroe et al. 201226 subtype. We identified 10 studies that reported on pain prevalence in AD, VaD,

B Total (random effects) DLB or mixed dementia (AD and VaD). We found no studies that reported on pain 0 20 40 60 80 100 prevalence in FTD. 0 20 40 60 80 100 0Pain prevalence20 (%)40 60 80 I2 = 98%100 We calculated the sample-weighted prevalence, which was 45.8% (95% CI: 33.4% - 58.5%) for AD, 56.2% (95% CI: 47.7% - 64.4%) for VaD, and 53.9% (95% CI: 37.4% - 70.1%) for mixed dementia. Our findings could not confirm the

Hendriks et al. 201523 hypothesis that pain prevalence differs significantly between the different Mixed dementia Volicer et al. 200925 dementia subtypes. Jensen-Dahm et al. 201220 Husebo et al. 200824 The sample-weighted prevalence for people with AD, VaD and mixed dementia C 29 Total (random effects) are comparable to the estimated pain prevalence of 50% that Zwakhalen et al 0 20 40 60 80 100 0 20 40 60 80 100 found in their review on pain in dementia (2005); however, in their review pain 0 20 40 60 80 100 was not reported per dementia subtype. As AD, VaD and mixed dementia together Pain prevalence (%) I2 = 98% account for two-third of all dementia cases, our findings seem to be in line with those of previous studies.29,30

66 | A REVIEW OF PAIN PREVALENCE IN ALZHEIMER’S, VASCULAR, FRONTOTEMPORAL AND LEWY BODY DEMENTIAS CHAPTER 3 | 67 The wide range of pain prevalence might be explained by the heterogeneity in and we were not able to control for confounding by dementia severity and study designs, settings, patient characteristics and definition of pain. For differences in pain caused by comorbidity in subtypes of dementia. example, in one study, pain was reported twice: one self-report and one proxy- rating. As self-report is the golden standard in pain-assessment,31 we decided to extract the self-reported data, but prevalence measured by self-report was lower than measured by proxy. This is in line with the findings of Scherder and van Conclusion Manen,32 who found that self-reported pain in patients with AD was lower than pain observed by a nursing assistant. This may indicate that people with AD do This is the first study that used a systematic approach to review the available not suffer from pain to the extent that their care givers expect them to do or that literature on pain prevalence in dementia subtypes and calculated the sample the prevalence of pain may depend on the methods and data sources used as weighted prevalence of pain in patients with AD, VaD, and mixed dementia. well as the time frame of pain detection.31,33 However, we found not enough studies to calculate a sample-weighted prevalence for DLB, and we found no studies that reported on pain prevalence in This review is not without limitations. To begin with, literature on this specific FTD. The pain prevalence in FTD and DLB has not been investigated extensively, topic seems to be scarce; only 10 studies on pain in dementia subtypes fulfilled which may be due to epidemiological issues, such as the higher prevalence of our inclusion criteria, and these studies showed substantial heterogeneity in AD, VaD, and mixed dementias as well as methodological issues. Whilst this study design. For example, prevalence of pain was usually not the main objective study could not show a difference in pain prevalence for the different dementia of studies included in our review. In addition, we found no studies on pain in FTD. subtypes, this review underlines that pain is reported frequent in people with This lack of studies could be explained by several epidemiological factors; for dementia and we recommend uniformity in pain assessment. More well- instance, FTD is relatively more prevalent in a younger population, whereas pain designed studies are required to draw firm conclusions on differences in is more prevalent in the older population.4 A specific reason for the lack of prevalence of pain between dementia subtypes as a consequence of the studies in DLB could be that these patients may be studied within trials assessing differences in neuropathological changes. These studies should take into account pain in PD, as Lewy body neuropathology is a key feature of PD and prevalence of comorbidity as potential confounder and focus especially on people with DLB pain in PD is estimated to be between 40% and 60%.34,35 Furthermore, we found and FTD, as both groups are underrepresented in earlier research. Research that the methods of pain assessment varied widely and many of the used should also focus on the causes, course and characteristics of pain. Eventually, instruments were not specifically designed for the population, for example, the pain prevalence and its possible impact on quality of life should be topic of future EQ5D, which is an instrument to measure quality of life and only incorporates research. one pain item. We found the use of nonspecific assessment tools to be more common than the use of dementia specific pain assessment tools. On the one hand this was surprising, as more than 30 specific measurement instruments have been designed to measure pain especially in people with dementia,29,36 but Acknowledgements on the other hand, this was not surprising, as the main objectives of the included studies did not include prevalence of pain. Another limitation that should be We would like to thank Bettina Husebø, Todd Monroe, Elizabeth Londos, Ladislav noted is that we could not correct for the influence of co-morbid pain conditions, Volicer, and Simone Hendriks for providing us with additional data for the present as most of the included studies did not report co-morbidity. review.

Our findings correspond to the previously reported prevalence of pain in people with dementia of 50%; also, our analyses showed that pain prevalence in dementia subtypes was not statistically significantly different. However, it cannot be completely ruled out that the prevalence of pain differs between the four dementia subtypes as a consequence of the differences in neuropathological changes subtypes, as we did not identify studies on pain in FTD and DLB,

68 | A REVIEW OF PAIN PREVALENCE IN ALZHEIMER’S, VASCULAR, FRONTOTEMPORAL AND LEWY BODY DEMENTIAS CHAPTER 3 | 69 Table.1 Studies reporting prevalence of pain in dementia subtypes

First Author Study design Setting Dementia Mean age in Female Dementia Mean MMSE Pain Pain prevalence subtype(s) years (SD), severity score (SD), range instrument range Baquero et al. (2009), Cross-sectional outpatients with and AD (n = 128) 76.2 (NA), 68% GDS 4 27.7% 12.3 (6.89), EQ-5D no pain 53% Spain17 without AD who donated 48-93 GDS 5 30.5% 0-26 moderate pain 41% material to biobank GDS 6 27.0% (n = 121) intense pain 6% (n = 141) GDS 7 5.7% Boström et al. (2007), Case-control outpatients AD (n = 34) 78.2 (NA), 44% NA 16.9 (NA), EQ-5D 31% (n = 9) Sweden18 with AD and DLB 63-92 0-30 intensity: (n = 68) no pain 64.8% mild-moderate 29.4% severe pain 0% missing 5% DLB (n = 34) 77.4 (NA), 44% 17.3 (NA), 70% (n = 17) 64-87 0-17) intensity: no pain 20.6% mild- moderate 29.5% severe 20% missing 29.4% Colosimo et al. (2010), Cross-sectional outpatients with CBD (n = 11) 70.1 (10.2), 56.5% NA MMSE Question-naire 36.4% Italy19 longitudinal parkinsonism range NA <23.5 with five pain observational (n = 1,302) 55% domains study DLB (n = 14) 74.1(6.5), 14.3% MMSE 50% range NA <23.5 93% PSP (n = 30) 69.7 (6.3), 46.7% MMSE 40% range NA <23.5 24% Jensen-Dahm et al. Baseline data RCT community dwelling AD (n = 239) 76.2 (7.1), 54.8% NA 24 (2.6), EQ-5D 33.9% (self-report) (2012), Denmark20 people with AD and other 74.5% range NA 20-30 self-report vs. 50.2% (proxy) dementia subtypes proxy Mix (n = 82) 31.7% (self-report) (n = 321) 22.5% 42.7%(proxy) “HR quality of life of Cross-sectional community dwelling AD (n = 237) 75.5 (8.5), 70.9% mild 20% NA EQ5D 68.6% Canary Island citizens” people with AD (n = 243) NA moderate 40% Oliva-Moreno, 2010 severe 40% and Lopez-Bastida21 et al. (2006),Spain22 Buchanan et al. (2005), Cross-sectional residents from LTCFs AD (SCU, 80.6 (7.9) * 62.1% CPS 2 6.6% NA MDS-pain no pain 78.8% USA27 who live in special care n = 11,311) CPS 3 44.1% mild pain(pain less than daily) unit (SCU) and who don’t CPS 4 16.3% 14.4% live in special care units CPS 5 26.7% moderate pain 6.2% (nSCU) CPS 6 5.0% excruciating pain 0.6% (n = 49,627) AD (nSCU, 82.5 (8.2) ** 69.7% CPS 2 10.5% no pain 71.3% n = 38,316) CPS 3 37.7% mild pain 17.8% CPS 4 12.7% moderate pain 9.5% CPS 5 18.0% excruciating pain 1.3% CPS 6 15.2%

70 | A REVIEW OF PAIN PREVALENCE IN ALZHEIMER’S, VASCULAR, FRONTOTEMPORAL AND LEWY BODY DEMENTIAS CHAPTER 3 | 71 First Author Study design Setting Dementia Mean age in Female Dementia Mean MMSE Pain Pain prevalence subtype(s) years (SD), severity score (SD), range instrument range Hendriks et al. (2015), Longitudinal baseline characteristics AD (n = 147) 84 (7) 70% advanced NA Frequency never 47.6% The Netherlands23 observational from residents of LTCFs 46% range NA dementia 9% assessed by rarely 19.7%, study with dementia physicians sometimes 15.6% (n = 372) often 9.5% almost daily 7.5% VaD (n = 72) never 43% 23% rarely 19.5%, sometimes 19.5% often 9.7% almost daily 8.3% Mix (n = 65) never 47.7% 18% rarely 18.5% sometimes 16.9% often 10.8% almost daily 6.1% DLB/PD never 50% (n = 18) rarely 27.8% 5% sometimes 11.2% often 5.5% almost daily 5.5% Husebo et al. (2008), Cross-sectional residents from LTCFs AD (n = 45) 84.7 (6.7) 75% CDR 0 17% 12.3 (9.7) MOBID-2 51.1% Norway24 with and without 25% 65-103 CDR 1 18% 0-30 Pain Scale mean pain score 2.4, dementia CDR 2 23% SD 2.2, range 0-7 (n = 181) CDR 3 43% VaD (n = 66) 56.1% 37% mean pain score 2.5, SD 1.7, range 0-8 Mix (n = 32) 68.8% 18% Mean pain score 2.9, SD 1.9, range 0-7 Monroe et al. (2012), Retrospective residents from LTCFs AD (n = 43) 86 (8), 54% CPS 2 14.6% NA DBS 46.5% USA26 between-groups with dementia (n = 48) 90% range NA CPS 3 22.9% Mean DBS score 6, cross-sectional CPS 4 20.8% SD 8, median 0, design at the end CPS 5 16.7% range 0-32 of life CPS 6 25.0% VaD (n = 4) 50% 8% mean DBS score 15, SD 22, median 6, range 0-48 DLB (n = 1) 100% 2% mean DBS score 6 Volicer et al. (2009), Cross-sectional residents from LTCFs AD (n = 406) 84.5 (7), 74.5% CPS 2 7.4% NA MDS-pain 50.2% The Netherlands25 with dementia 44.8% 65-102 CPS 3 34.6% Frequency Intensity: (n = 929) CPS 4 7.1% Intensity no pain 52.4% CPS 5 36.2% mild 22.7% CPS 6 12.2% moderate 21.4% severe 3.5% VaD +Mix 64.2% (n=470) Intensity: 51.1 % no pain 38.5% mild 28.5% moderate 28.7% severe 4.3%

AD: Alzheimer Disease; CBD: corticobasal degeneration; CDR: Clinical Dementia Rating; CPS: Cognitive PD: Parkinson Disease; PSP: progressive supranuclear palsy; QOL: quality of life; VaD: Vascular Dementia; Performance Scale; DBS: Discomfort Behaviour Scale; DLB: Dementia with Lewy Bodies; GDS: Global Deterioration VAS: Visual analog Scale; *<60 yr 1.5%; 61-70 yrs 6.9%; 71-80yrs 35%; 81-90 yrs 47.6%; over 91yrs 9.2%; **<60 yr Scale; MMSE: Minimal Mental State Examination; Mix: Alzheimer and Vascular Dementia; NA: not available; 1.3%; 61-70 yrs 4.8%; 71-80yrs 27.4%; 81-90 yrs 51.2%; over 91yrs 15.4%.

72 | A REVIEW OF PAIN PREVALENCE IN ALZHEIMER’S, VASCULAR, FRONTOTEMPORAL AND LEWY BODY DEMENTIAS CHAPTER 3 | 73 Table 2. Risk of bias assessment

Author Year Country Dementia subtype(s) Sample Risk of bias size Sampling Case Dementia Selection of Response Case Total score frame definition for stage patients Definition for dementia pain subtype Baquero et al.17 2009 Spain AD 128 + + + - ? - 3 Boström et al.18 2007 Sweden AD, DLB 68 + + + ? ? - 3

Buchanan et al.27 2005 USA AD 49627 + - + + ? + 4

Colosimo et al.19 2010 Italy DLB, PS, CBD 55 + + ? - ? ? 2

Hendriks et al.23 2014 The Netherlands AD, VaD, mix, DLB 330 - + - ? ? - 1 Husebo et al.24 2008 Norway AD, VaD, mix, no dementia 181 + + + - ? + 4 Jensen-Dahm et al.20 2012 Denmark AD, mix 321 + + - - ? - 2 Monroe et al.26 2012 USA AD, VaD, DLB 48 - - + - ? ? 1

Oliva-Moreno21 & Lopez-Bastida et 2006 Spain AD 237 + ? + - - - 2 al.22 Volicer et al.25 2009 The Netherlands AD, VaD, mix, other 929 + - + - ? - 2

AD: Alzheimer Disease; CBD: corticobasal degeneration; DLB: Dementia with Lewy Bodies; Mix: Alzheimer Disease and Vascular Dementia; PSP: progressive supranuclear palsy; VaD: Vascular Dementia; + = low risk of bias; - = high risk of bias; ? = risk of bias unclear.

74 | A REVIEW OF PAIN PREVALENCE IN ALZHEIMER’S, VASCULAR, FRONTOTEMPORAL AND LEWY BODY DEMENTIAS CHAPTER 3 | 75 References in meta-analyses. BMJ 2003; 327: 557-560. 17. Baquero TM, Peset V, Burguera JA, et al. Quality of life in Alzheimer’s 1. Woolf AD, Vos T, March L. How to measure the impact of musculoskeletal disease. Rev Neurol 2009; 49: 337-342. conditions. Best Pract Res Clin Rheumatol 2010; 24: 723-732. 18. Bostrom F, Jonsson L, Minthon L, Londos E. Patients with dementia with 2. Cook AK, Niven CA, Downs MG. Assessing the pain of people with cognitive Lewy bodies have more impaired quality of life than patients with Alzheimer impairment. Int J Geriatr Psychiatry 1999; 14: 421-425. disease. Alzheimer Dis Assoc Disord 2007; 21: 150-154. 3. Achterberg WP, Pieper MJ, van Dalen-Kok AH, et al. Pain management in 19. Colosimo C, Morgante L, Antonini A, et al. Non-motor symptoms in atypical patients with dementia. Clin Interv Aging 2013; 8: 1471-1482. and secondary parkinsonism: The PRIAMO study. J Neurol 2010; 257: 5-14. 4. Duncan R, Francis RM, Collerton J, et al. Prevalence of arthritis and joint 20. Jensen-Dahm C, Vogel A, Waldorff FB, Waldemar G. Discrepancy between pain in the oldest old: findings from the Newcastle 85+ study. Age Ageing self- and proxy-rated pain in Alzheimer’s disease: Results from the Danish 2011; 40: 752-755. Alzheimer Intervention Study. J Am Geriatr Soc 2012; 60: 1274-1278. 5. Scherder EJ, Sergeant JA, Swaab DF. Pain processing in dementia and its 21. Oliva-Moreno J, Lopez-Bastida J, Worbes-Cerezo M, Serrano-Aguilar P. relation to neuropathology. Lancet Neurol 2003; 2: 677-686. Health related quality of life of Canary Island citizens. BMC Public Health 6. Jensen-Dahm C, Werner MU, Dahl JB, et al. Quantitative sensory testing 2010; 10: 675. and pain tolerance in patients with mild to moderate Alzheimer disease 22. Lopez-Bastida J, Serrano-Aguilar P, Perestelo-Perez L, Oliva-Moreno J. compared to healthy control subjects. Pain 2014; 155: 1439-1445. Social-economic costs and quality of life of Alzheimer disease in the Canary 7. Scherder E, Oosterman J, Swaab D, et al. Recent developments in pain in Islands, Spain. Neurology 2006; 67: 2186-2191. dementia. BMJ 2005; 330: 461-464. 23. Hendriks SA, Smalbrugge M, Galindo-Garre F, Hertogh CM, Van Der Steen 8. Scherder EJ, Slaets J, Deijen JB, et al. Pain assessment in patients with JT. From admission to death: prevalence and course of pain, agitation, and possible vascular dementia. Psychiatry 2003; 66: 133-145. shortness of breath, and treatment of these symptoms in nursing home 9. Bathgate D, Snowden JS, Varma A, Blackshaw A, Neary D. Behaviour in residents with dementia. J Am Med Dir Assoc 2015; 16: 475-481. frontotemporal dementia, Alzheimer’s disease and vascular dementia. Acta 24. Husebo BS, Strand LI, Moe-Nilssen R, Husebo SB, Aarsland D, Ljunggren Neurol Scand 2001; 103: 367-378. AE. Who Suffers Most? Dementia and Pain in Nursing Home Patients: A 10. Burton EJ, McKeith IG, Burn DJ, Firbank MJ, O’Brien JT. Progression of Cross-sectional Study. J Am Med Dir Assoc 2008; 9: 427-433. white matter hyperintensities in Alzheimer disease, dementia with lewy 25. Volicer L, Van Der Steen JT, Frijters DHM. Modifiable Factors Related to bodies, and Parkinson disease dementia: a comparison with normal aging. Abusive Behaviors in Nursing Home Residents With Dementia. J Am Med Am J Geriatr Psychiatry 2006; 14: 842-849. Dir Assoc 2009; 10: 617-622. 11. Corbett A, Husebo B, Malcangio M, et al. Assessment and treatment of pain 26. Monroe T, Carter M, Feldt K, Tolley B, Cowan RL. Assessing advanced in people with dementia. Nat Rev Neurol 2012; 8: 264-274. cancer pain in older adults with dementia at the end-of-life. J Adv Nurs 12. van der Flier WM, Scheltens P. Epidemiology and risk factors of dementia. J 2012; 68: 2070-2078. Neurol Neurosurg Psychiatry 2005; 76 Suppl 5: v2-v7. 27. Buchanan RJ, Choi M, Wang S, Ju H, Graber D. Nursing home residents with 13. Alcolea D, Martinez-Lage P, Izagirre A, et al. Feasibility of lumbar puncture Alzheimer’s disease in special care units compared to other residents with in the study of cerebrospinal fluid biomarkers for Alzheimer’s disease: a Alzheimer’s disease. Dementia: The International Journal of Social multicenter study in Spain. J Alzheimers Dis 2014; 39: 719-726. Research and Practice 2005; 4: 249-267. 14. Shamliyan TA, Kane RL, Ansari MT, et al. Development quality criteria to 28. Sieben A, Van Langenhove T, Engelborghs S, et al. The genetics and evaluate nontherapeutic studies of incidence, prevalence, or risk factors of neuropathology of frontotemporal lobar degeneration. Acta Neuropathol chronic diseases: pilot study of new checklists. J Clin Epidemiol 2011; 64: 2012; 124: 353-372. 637-657. 29. Zwakhalen SM, Hamers JP, Berger MP. The psychometric quality and 15. Higgins JP, Thompson SG. Quantifying heterogeneity in a meta-analysis. clinical usefulness of three pain assessment tools for elderly people with Stat Med 2002; 21: 1539-1558. dementia. Pain 2006; 126: 210-220. 16. Higgins JP, Thompson SG, Deeks JJ, Altman DG. Measuring inconsistency 30. Kalaria RN, Maestre GE, Arizaga R, et al. Alzheimer’s disease and vascular

76 | A REVIEW OF PAIN PREVALENCE IN ALZHEIMER’S, VASCULAR, FRONTOTEMPORAL AND LEWY BODY DEMENTIAS CHAPTER 3 | 77 dementia in developing countries: prevalence, management, and risk factors. Lancet Neurol 2008; 7: 812-826. 31. Pautex S, Michon A, Guedira M, et al. Pain in severe dementia: self- assessment or observational scales? J Am Geriatr Soc 2006; 54: 1040-1045. 32. Scherder E, van Manen F. Pain in Alzheimer’s disease: nursing assistants’ and patients’ evaluations. J Adv Nurs 2005; 52: 151-158. 33. Scherder E, Bouma AF, Slaets JF, et al. Repeated pain assessment in Alzheimer’s disease. Dement Geratric Cogn Disord 2001;12:400-407. 34. Wasner G, Deuschl G. Pains in Parkinson disease--many syndromes under one umbrella. Nat Rev Neurol 2012; 8: 284-294. 35. Borsook D. Neurological diseases and pain. Brain 2012; 135: 320-344. 36. Zwakhalen SM, Hamers JP, Abu-Saad HH, Berger MP. Pain in elderly people with severe dementia: a systematic review of behavioural pain assessment tools. BMC Geriatr 2006; 6: 3.

78 | A REVIEW OF PAIN PREVALENCE IN ALZHEIMER’S, VASCULAR, FRONTOTEMPORAL AND LEWY BODY DEMENTIAS CHAPTER 3 | 79 Pain in dementia: prevalence and associated factors: protocol of a 4 multidisciplinary study 4 Pain in dementia: prevalence Discussion: This study will help to enhance our knowledge regarding the prevalence of different types of pain in different dementia subtypes, i.e. AD, VaD, and associated factors: protocol FTD and DLB. This study also aims to contribute to a better understanding of oral health status in people with dementia, the use of autonomic responses in of a multidisciplinary study the assessment of pain in people with dementia and the relationships between pain and cognitive symptoms, neuropsychiatric symptoms and quality of life in people with various dementia subtypes and in different stages of the disease.

Janine van Kooten, Suzanne Delwel, Tarik T. Binnekade, Martin Smalbrugge, Johannes C. van der Wouden, Roberto S. G. M. Perez, Didi Rhebergen, Wouter W. A. Zuurmond, Max L. Stek, Frank Lobbezoo, Cees M. P. M. Hertogh, Erik J. A. Scherder

Background Abstract Pain is a common problem in people with dementia. The prevalence of pain in Background: Pain is a common problem in people with dementia, however people with dementia is high;1-4 there is good agreement in both large and small the exact prevalence of pain in dementia subtypes, e.g. Alzheimer’s Disease studies that about 50% of the people with dementia regularly experience pain.5 (AD), vascular Dementia (VaD), frontotemporal dementia (FTD) and dementia This is not surprising, considering that advanced age is an important risk factor with Lewy bodies (DLB), is unknown, as is the relation between pain and for developing dementia, as well as for pain.6,7 the different subtypes of dementia. In this study, the prevalence of pain in people with dementia will be investigated per dementia subtype and the Pain in dementia is associated with occurrence of neuropsychiatric symptoms,8,9 relationship between the various subtypes of dementia and the presence of and also with a decline in cognitive functioning, as well as in performance of specific types of pain (i.e. musculoskeletal pain, neuropathic pain and orofacial Activities of Daily Living (ADL).10 Next to neuropsychiatric symptoms, pain is the pain) will be examined. Secondly, associations between various types of pain, most cited reason for a decrease in quality of life in dementia.11 Therefore, cognitive functioning, neuropsychiatric symptoms and quality of life in people recognition and adequate treatment of pain in people with dementia should have with dementia will be examined. A third purpose is to study the value of the high priority. assessment of autonomic responses in assessing pain in people with dementia. Finally, the effect of feedback to the attending physician on the presence of pain, Although knowledge about the differences in pain prevalence between dementia based on examination by investigators with backgrounds in neuropsychology, subtypes is scarce, it is suggested that differences in neuropathology between geriatric dentistry and elderly care medicine, will be evaluated. dementia subtypes could be of vital importance in the treatment of pain.12 Results from previous small studies showed that specific neuropathological changes Methods/Design: A cross-sectional, partially longitudinal observational study may affect pain experience and pain perception differently in the four commonest in 400 participants with dementia, aged 60 years and older. Participants will dementia subtypes,12 i.e., Alzheimer’s Disease (AD), vascular dementia (VaD), be recruited from an outpatient memory clinic and dementia special care units. dementia with Lewy bodies (DLB), and frontotemporal dementia (FTD). It is All participants will be examined by an elderly care medicine trainee, a dentist suggested that with a preponderance of white matter lesions, pain experience in with experience in geriatric dentistry, and a neuropsychologist. The primary VaD may increase compared to other dementia subtypes,12 while in AD and FTD it outcome is presence of pain. Secondary outcomes will include oral health, is hypothesized that neuropathological changes like gray matter atrophy lead to autonomic responses to pain stimulus, vital sensibility and gnostic sensibility, an increase in pain tolerance.13-15 musculoskeletal examination, cognitive functioning, neuropsychiatric symptoms, and quality of life. Current knowledge about the prevalence of specific types of pain, e.g., musculoskeletal pain or neuropathic pain, and causes of pain in people with

82 | PAIN IN DEMENTIA: PREVALENCE AND ASSOCIATED FACTORS: PROTOCOL OF A MULTIDISCIPLINARY STUDY CHAPTER 4 | 83 dementia is limited. Nociceptive pain of musculoskeletal origin is considered the Methods/Design most prevalent pain type, but orofacial pain and neuropathic pain are also likely to have considerable prevalence rates.16,17 For example, orofacial pain is one of Study design the most prevalent types of pain in older people, and has hardly been studied in This protocol describes an observational, cross-sectional and partly longitudinal people with dementia.17,18 Assessment of the specific type of pain and classifying cohort study. Measurements take place at baseline, and for the people with pain its probable causes would be useful as a complement for the assessment of pain at baseline, a follow-up pain measurement takes place after 3 months. in people with dementia, and would provide a sound basis for appropriate pain treatment when combined with dementia subtyping in people with dementia. Recruitment To ensure this study population covers the broad range of the elderly population An important problem encountered in the assessment of pain in patients with with dementia, participants will be recruited in different stages of dementia. dementia, especially in more advanced stages, is that assessment often is based Participants with mild to moderate dementia, will be recruited at the outpatient on observation and interpretation of behavior, for which the validity is unequivocally memory clinic of the Center of Geriatric Medicine Amsterdam (COGA), VU established.5,19,20 More objective methods for assessing pain in a more indirect way University medical center. Participants with moderate to severe dementia will be in this population are therefore warranted. Assessing autonomic responses to pain recruited from Amstelring, a health organization delivering specialized care to might be such a method.16,21 residents with dementia in 17 nursing homes covering a large area in the northwestern part of the Netherlands, including Amsterdam. Considering the limited number of studies on pain in each subtype of dementia, and the absence of studies on type of pain in dementia subtypes and in different Inclusion criteria stages of dementia, we developed a unique multidisciplinary approach. Participants who meet the following criteria will be included: aged 60 or older, Investigators with backgrounds in neuropsychology, geriatric dentistry, elderly diagnosis of dementia, i.e. AD, VaD, FTD, or DLB, and a signed informed consent care medicine, and pain medicine will study in close cooperation the presence by the participant or legal representative. and type of pain, in various stages of the four main subtypes of dementia (AD, VaD, FTD and DLB). In a subset of patients we will explore the feasibility of Exclusion criteria measuring autonomic responses as a tool to assess the presence of pain. Primarily mentally disabled persons who subsequently develop dementia, e.g. people with Down’s syndrome and AD, will be excluded. Persons who indicate Research aims either verbally or non-verbally, that they do not wish to participate will be The primary aim of this study is to investigate the prevalence and intensity of excluded as well, despite of given consent by a legal representative. pain (subtypes) in the four main dementia subtypes and to examine the relationship between the different types of pain and the different dementia Ethical approval and informed consent subtypes, i.e., AD, VaD, FTD and DLB. The study protocol has been approved by an accredited Medical Ethics Review Secondary aims are: Committee (VU University Medical Center; NL.43861.029.13). • to examine the associations between neuropsychiatric symptoms and pain, quality of life and pain, and cognitive function and pain for all different Informed consent will be obtained of all participants. For participants unable to give dementia subtypes; consent, the legal representative will be asked to sign an informed consent form. • to explore whether autonomic responses can be used for assessing pain in patients with more advanced stages of dementia; Measurement instruments • to test the psychometric properties of the newly developed Orofacial Pain Variables measured in this study (primary outcome measures, secondary Scale for Non Verbal Individuals (OPS-NVI); outcome measures, demographic, and control variables) are summarized in • to examine the oral health of older adults with dementia; table 1. These will be assessed by using a combination of self-report instruments • evaluate the effect of providing feedback on the presence of pain (including and observational instruments, which are completed by caretaker or by a contact treatment advice), to the attending physician and/or dentist. nurse.

84 | PAIN IN DEMENTIA: PREVALENCE AND ASSOCIATED FACTORS: PROTOCOL OF A MULTIDISCIPLINARY STUDY CHAPTER 4 | 85 Table 1. Measurement instruments Primary outcomes

Instrument Source Time T0 memory T0 T1 The primary outcome will be: the presence and intensity of pain in dementia (in min) clinic nursing nursing Home home subtypes. Pain will be assessed, using a combination of methods concurrently. Rey Auditory Verbal Learning Test 1 15 * (15 Word Test) In participants who are able to communicate about the presence of pain, the Amsterdam Dementia Screening Brief Pain Inventory (BPI) will be applied.22 The BPI is a questionnaire, assessing 1 5 * Test Meander (ADS-6) the severity of pain by four numeric rating scale items and, explores the degree Behavioural Assessment of the to which pain interferes with daily activities by seven interference items. The BPI Dysexecutive Syndrome 1 5-10 * (BADS) – Rule Shift Cards is a valid and reliable assessment tool for pain assessment in epidemiological 22 Brief Pain Inventory (BPI) 1 5 * * * studies. Concurrently with the BPI, the participant will be asked to score the Dementia Quality of Life (DQoL) 1,2 5 * intensity of pain, when he or she is in pain, on a self-report scale. The Numeric 23-25 Faces Pain Scale – Revised (FPS-R) 1 3 * * * Rating Scale (NRS), Verbal Descriptor Scale (VDS), and Faces Pain Scale 26,27 Global Deterioration Scale (GDS) 3 1 * Revised (FPS-R) can be used for this purpose, dependent on understanding Katz-ADL by the participant of each of the scales. Mini Mental State Exam (MMSE) 1 10 * * Mobilization Observation Behaviour In participants with difficulties to communicate their pain, a contact nurse will be Intensity Dementia-2 (MOBID-2) 3 5 * * asked to complete the Pain Assessment IN Advanced Dementia Scale (PAINAD).28 Pain Scale The PAINAD is a brief and easy to administer observational scale, especially Neuropsychiatric Inventory- 3 15 * used in people with advanced dementia. The reliability and validity of the PAINAD Nursing Home (NPI-NH) have been studied extensively and a cut-off score of 2 has been recommended to Neuropsychiatric Inventory- 2 5 * 28-32 Questionnaire (NPI-Q) indicate the probable presence of pain. Numeric Rating Scale (NRS) 1 2 * * * Orofacial Pain Scale for Apart from these pain assessment tools, the Mobilization Observation Behaviour 3 10 * * * Non Verbal Individuals (OPS-NVI) Intensity Dementia-2 (MOBID-2) Pain Scale will be administered in all Oral Examination 1 20 * * participants by a contact nurse to measure pain, using standardized movements.33 Pain Assessment IN Advanced 3 The MOBID-2 is a reliable, valid and time effective instrument to measure pain in Dementia (PAINAD) people with advanced dementia.33 For this study, the MOBID-2 was translated into Quantitative Sensory Testing (QST) 1 15 * * Dutch, based on the guidelines for cross-cultural adaptation.34 The translated Qualidem 3 10 * version showed good feasibility in a pilot study. An overall score of NRS ≥ 3 in the Revised Index for Social 3 5 * 33 Engagement (RISE) MOBID-2 will be regarded as clinically relevant pain. Rey Complex Figure Test (RCFT) 1 5-10 * For the measurement of orofacial pain, Delwel and Lobbezoo developed the Stroop Color-Word Test 1 5 * Orofacial Pain Scale for Non Verbal Individuals (OPS-NVI), which demonstrated Structured physical examination 1,3 15 * * good face validity and usability in a pilot study.17,35 For further validation of the Trail Making Test (TMT) 1 5 * OPS-NVI, a number of steps will be performed. First, the verbal and nonverbal Verbal Descriptor Scale (VDS) 1 2 * * * participants will be observed using the OPS-NVI, during rest, drinking, eating, Verbal Fluency Test (VFT) - Animals 1 3-10 * and mouth care. These activities will be observed for indicators of pain behavior Visual Association Test (VAT) 1 * and the pain intensity will be assessed by two observers, whenever possible. VU University- Ambulatory Monitoring System (VU-AMS) 1 10 * * Second, the group of verbal participants will be asked about the presence of pain and, if pain is present, the pain intensity with the NRS, VDS, and FPS-R. 1 = Patient, 2 = Caretaker, 3 = Health professionals,T0 = baseline, T1 = after 3 months.

86 | PAIN IN DEMENTIA: PREVALENCE AND ASSOCIATED FACTORS: PROTOCOL OF A MULTIDISCIPLINARY STUDY CHAPTER 4 | 87 Third, a standardized dental examination will be performed in both verbal and testing of vital and Quantitative Sensory Testing (QST) will be done (see below). nonverbal participants, to evaluate their oral health, which will be described The oral examination will be evaluated and given as feedback to the participant, hereafter. The results of observation, self-assessment and the dental caretakers and/or medical staff. examination will be compared in order to validate the OPS-NVI. Autonomic responses to pain stimulus Autonomic responses to pain will be measured with the VU University Ambulatory Monitoring System (VU-AMS), a certified device that records the Secondary outcomes electrocardiogram (ECG) and changes in thorax impedance (dZ) to determine Inter-Beat-Interval (IBI), Pre Ejection Period (PEP), and Respiratory Sinus Secondary outcomes include oral health, autonomic responses to a pain Arrhythmia (RSA). In addition, Skin Conductance Level (SCL) will be recorded stimulus, vital sensibility, musculoskeletal examination, cognitive functioning, using the VU-AMS by two electrodes (Biopac TSD203) combined with their neuropsychiatric symptoms, and quality of life. isotonic electrode gel (GEL101) on the index and middle finger.46-48

Oral health At the outpatient memory clinic autonomic responses to a pain stimulus in the Oral health will be evaluated by a dentist with experience in geriatric dentistry, form of a venipuncture will be measured in all patients. First the patients will be performing a structured oral examination, including history taking, extraoral, asked to rate their current pain on the NRS followed by a two minute rest and a functional, intraoral, and (whenever indicated) additional examination. baseline measurement. Next, a venipuncture will be performed by a geriatric nurse and directly after this procedure the patient will be asked to rate the pain The function of the temporomandibular joint will be assessed with an Active intensity of the venipuncture on the NRS. Subsequently, a second baseline Mandibular Examination (AME). The participant will be asked to make different measurement will be performed during a three-minute rest period. The patients jaw movements and the dentist will measure the excursions.36 The presence or will be asked to rate their current pain after the second baseline measurement absence of pain will be observed; if the procedure is too painful, it will be to ensure that their pain level is not increased by sitting still for a prolonged stopped.37 Whenever applicable, removable prosthetics will be examined for period of time. hygiene, retention, and occlusion/articulation.38 The status of the oral tissues will be determined by examining the lips, cheeks, palate, tongue, floor of the In nursing home residents, autonomic responses to pain stimuli are only mouth, alveolar ridges, saliva, and mouth odor. These elements will be classified measured in patients who undergo changes of wound dressings as part of usual as healthy or abnormal by the dentist.38 The number of opposing molars, the care in the treatment of pressure ulcers. First the patients who are able to occlusal units (OU), will be counted as an indication of the chewing ability.39 communicate, will be asked to rate their current pain level on the NRS followed The status of the participant’s natural teeth will be scored with the Decayed by a five minute baseline measurement in rest. After the baseline measurement, Missing and Filled Teeth (DMFT) Index.40 For this index, the number of teeth with the wound treatment procedure will be explained. The renewal of the wound caries, missing teeth, and teeth with fillings will be recorded. Also the amount of dressing will be utilized as a painful episode. The current wound dressing will be dental wear will be quantified and itemized.41 The periodontal status will be removed, the wound cleaned and new dressing will be applied. Afterwards, the determined by measuring the periodontal pockets according to the Dutch pain intensity of the treatment episode will be assessed in verbal patients by Periodontal Screening Index (DPSI)42 and the physiological mobility of the teeth.43 using the NRS. Finally, another 5 minute baseline measurement in rest will be The oral hygiene will be measured, according to the Plaque Index by Silness and performed. Loe.44 Somatosensory changes of the face will be examined with brief quantitative sensory tests for orofacial pain. Three simple tests will be used: a cotton swab, Vital sensibility pinprick, and brush test. The three innervation areas of the N. Trigeminus will be Vital sensibility will be assessed by Quantitative Sensory Testing (QST), including tested: above the eyebrow, under the eye and at the lower lip. The results will be touch perception, sharp/dull distinction and temperature discrimination. used to determine if a participant has signs of neuropathic pain in the orofacial All sensory procedures will be administered to the inside of both lower arms. region.45 For determining of signs of neuropathic pain elsewhere in the body, This site is often used in somatosensory studies because of the high density of

88 | PAIN IN DEMENTIA: PREVALENCE AND ASSOCIATED FACTORS: PROTOCOL OF A MULTIDISCIPLINARY STUDY CHAPTER 4 | 89 intra-epidermal nerve fibers.49-51 QST will only be performed in patients with a Syndrome,56 (4) the Rey Complex Figure Test,57 (5) the Stroop Color-Word Test,58 Mini Mental State Examination (MMSE)-score ≥.14 (6) the Trail Making Test,59 (7) the Verbal Fluency Test Animals,60 (8) the Visual Association Test,61 and (9) the Mini Mental State Examination (MMSE),62 a brief The touch perception threshold will be assessed by nylon monofilaments based instrument used to screen for cognitive impairment and well validated in both on the Von Frey hairs (Somedic AB, Sweden). A maximum number of 19 filaments research and clinical practice.63,64 In nursing home patients only the MMSE will will be administered to the inside of both lower arms, until the participant be used for measuring cognitive functioning. indicates the sensation of touch. False positive responses will be controlled by administering a null stimulus (no filament administered). The monofilaments Neuropsychiatric symptoms increase in diameter, resulting in an increase in force that is applied to the skin. Neuropsychiatric symptoms will be measured with a brief questionnaire form of The monofilaments are numbered in ascending order. The number of the first the Neuropsychiatric Inventory (NPI-Q)65 in participants at the outpatient memory monofilament resulting in a response will be used as an outcome. clinic, and with the Neuropsychiatric Inventory Nursing Home Version (NPI-NH)66 in nursing home residents. The NPI-Q is a revised version of the original NPI and The ability to sense the difference between sharp and dull stimuli will be has been validated against the original NPI.67 The NPI-NH is developed for rating assessed using the Neuropen (Owen Mumford, U.S.A.). This instrument contains by professional caregivers within institutions and proved to be valid and reliable both a monofilament for touch and a tip for sharpness. The monofilament will be for trained nursing staff.66, 68 The NPI-NH is the only nursing home instrument for pressed to the skin surface at a 90° angle until it bows (10 g) and subsequently a broad range of neuropsychiatric symptoms that has been translated into hold in this position for 2 seconds. The sharp tip will be pressed to the skin Dutch.69 surface at a 90° angle, until the marker is within the 40 g marker zone and subsequently held in this position for 2 seconds. Number of errors (maximum of Quality of life 6) will be used as the outcome measure. Quality of life will be assessed by the Dementia Quality of Life instrument (DQol)70 or the Qualidem.71,72 The DQol is a valid instrument to assess the quality of life in For temperature discrimination the Rolltemp (Somedic AB, Sweden) will be persons with a mild to moderate stage of dementia.70 In people with moderate to applied in random order to the left forearm (total of 3 times) and to the right severe dementia, the Qualidem will be used, which is a reliable and valid forearm (total of 3 times). The subject will have to indicate whether the stimulus instrument that provides a quality of life profile of persons living in nursing was cold or warm. The rollers will be placed in their docking stations between homes.71,72 trials so their temperature is either 25° C or 40° C. Compared with the normal human skin temperature of 30° C, the rollers will be perceived as either cold or In addition to quality of life, social participation will be measured using the warm, by humans with normal temperature sensitivity. The number of errors Revised Index for Social Engagement (RISE).73 The RISE is a valid instrument for (maximum six) will be used as an outcome measure. assessing social engagement in nursing home residents and contains six questions about the social interaction of the participant. A contact nurse who is Musculoskeletal examination familiar with the participant completes this questionnaire.73 A structured musculoskeletal examination will be performed by a physician in patients with clinically pain to determine whether the pain originates from the Dementia subtype and stage musculoskeletal system. Clinically relevant pain is either defined as BPI score Participants recruited at the outpatient memory clinic will be screened for ≥ 3, or self-report score ≥ 3, or MOBID-2 pain score ≥ 3, or PAINAD score ≥ 2. dementia, including Computed Tomography (CT) or Magnetic Resonance Imaging (MRI). The clinical diagnosis will be established by consensus within a Cognitive functioning multidisciplinary consultation. The National Institute of Neurological and Cognitive functioning will be measured using the following memory measures: Communicative Disorders and Stroke (NINCDS) and the Alzheimer’s Disease and (1) the 15 Word test adapted from the Rey Auditory Verbal Learning Test,52-54 (2) Related Disorders Association (ADRDA) criteria for Alzheimer dementia,74 the the Meander from the Amsterdam Dementia Screening Test,55 (3) the Rule Shift National Institute of Neurological Disorders and Stroke Association (NINDS) Cards Tests subtest of the Behavioural Assessment of the Dysexecutive Internationale pour la Recherche et l’Enseignement en Neurosciences (AIREN)

90 | PAIN IN DEMENTIA: PREVALENCE AND ASSOCIATED FACTORS: PROTOCOL OF A MULTIDISCIPLINARY STUDY CHAPTER 4 | 91 criteria for vascular dementia,75 the revised criteria for FTD,76 and the revised At the start of the memory-screening day, the neuropsychologist will install the criteria for DLB are used.77 Participants recruited from dementia special care VU-AMS device after which the VU-AMS protocol will be executed. units will have a diagnosis of dementia usually based on DSM-IV criteria (2000), and the medical record will be reviewed for a probable subtype diagnosis Participants will undergo a physical examination by a physician from the determined by a geriatrician, neurologist, elderly care physician, or old age outpatient clinic; for this study, the physical examination will comprise a psychiatrist, and preferably completed by CT or MRI. structured examination of the musculoskeletal system. Besides, patients will undergo QST by a trained neuropsychologist or research assistant. A dental Dementia stage will be determined using the Global Deterioration Scale (GDS), a examination will be performed by a dentist with experience in geriatric dentistry. well-validated seven point scale.78 The GDS defines seven stages of dementia, If pain is observed in participants, recruited in the outpatient memory clinic, ranging from ‘no global impairment’ (1) to ‘very severe global impairment’ (7). feedback about the pain and the most probable cause(s) will be given to the attending physician. Demographics, baseline, and control variables Patient characteristics (i.e. sex and age) will be derived from medical records. Study measures after consent in nursing homes The Charlson co-morbidity Index will be used to classify co-morbidities,79 while To reduce participant burden, measurements in nursing home residents will be ADL performance will be measured by the Katz-ADL.80 The Katz-ADL assesses split into three parts, which will preferably be performed on 3 different days; i.e. functional status and is used extensively in the field of elderly care medicine and (1) structured pain assessment including musculoskeletal examination when geriatrics. clinically relevant pain is present, (2) oral examination, and (3) cognitive assessment, QST and VU-AMS, if applicable. Structured pain assessment Medication includes the BPI, self-report-scales, or PAINAD. Besides, all participants will be Information on medication use will be derived from the medical records at the observed for presence of pain during morning care using the MOBID-2. Trained outpatient memory clinic. To ensure a complete overview of pain medication, research assistants will collect data on neuropsychiatric symptoms, quality of participants will explicitly be asked about their use of over-the-counter pain life, participation, and ADL by interviewing the contact nurse. Dental examination medication. In the nursing homes information on medication use will be derived will be performed by a dentist. Cognitive assessment will be done by a from the pharmacists’ electronic patient records. neuropsychologist. QST will only be assessed in patients with a MMSE-score ≥ 14, and the VU-AMS will only be used in nursing home residents who undergo changes of wound dressings as part of usual decubitus care.

Study procedure If pain is observed in nursing home residents during baseline measurement, feedback about the type(s) of pain and a differential diagnosis of the most Study measures after consent at the outpatient memory clinic Measurements probable cause(s), including information about treatment options, will be given will be performed during the memory-screening day in the outpatient clinic. to the attending physician. Reassessment of these participants will take place Since several measurements are already part of the standard routine, for after three months with the same pain measurement instruments as applied at research purposes the following measurements will be added, i.e. a structured baseline. During this reassessment, information about pain treatments in the pain assessment, an oral examination, the measurements of autonomic last three months will be collected to evaluate the effect of feedback to the responses and sensibility (touch perception, sharp/dull distinction, temperature attending physician. discrimination), the presence of neuropsychiatric symptoms, and the measurement of quality of life. Demographics, dementia subtype diagnosis, Sample size and power analysis cognitive performance (i.e. neuropsychological assessment), ADL (i.e. Katz- Our first objective is to assess the prevalence of pain in dementia subtypes. To ADL), co-morbidity (i.e. Charlson co-morbidity Index), and prescribed medication date, no clear estimates of the prevalence of pain in dementia subtypes are will be obtained from medical notes. known. Therefore, we used an assumed prevalence of 50% for our sample size calculation. Aiming at a precision of 5 percentage points (i.e., estimate +/− 2.5%)

92 | PAIN IN DEMENTIA: PREVALENCE AND ASSOCIATED FACTORS: PROTOCOL OF A MULTIDISCIPLINARY STUDY CHAPTER 4 | 93 with a significance level (alpha) of 0.05 and a power (beta) of 0.95, we will need to In this study, a unique multidisciplinary approach is used, in which a recruit 384 participants. Taking into consideration that prevalence measurements neuropsychologist, dentist, and elderly care medicine trainee all examine the do not require follow-up, we did not correct for attrition. same group of participants. All participants will have a cognitive assessment, an oral examination and physical assessment. Results will be gathered and a To analyze the assumed association between neuropsychiatric symptoms and differential diagnosis, including a treatment advice, if desired, will be provided to pain, quality of life and pain, and cognitive functioning and pain in all different the attending physician. dementia subtypes, multivariable linear regression analyses will be used. Power depends on the amount of explanatory variables and therefore we decided to We assume that the prevalence of pain will be different for the four most common include a maximum of 10 explanatory variables in our analyses. Based on the dementia subtypes, and will be the highest in VaD and we expect to see more sample size calculations, using the software-tool ‘G*power version 3.1.2’,81 118 neuropathic pain in VaD, in which it is more likely to see neuropathic pain due to participants per major dementia subtype (i.e. AD and VaD) need to be recruited white matter lesions.12, 82 to ensure a 95% confidence interval with 80% power. This study will provide data about the prevalence and intensity of orofacial pain Data analysis and oral health problems in persons with dementia. Although there is an increase We will use descriptive statistics for the demographic features of the cohort. in attention for oral health in people with dementia, studies on oral health in The prevalence and intensity of pain and the causes of pain will be determined in people with dementia are scarce. Furthermore, available data indicate that oral the overall group of people with dementia. In addition, the frequency of pain health in people with dementia is poorer compared to cognitive intact people.83 subtypes per dementia subtype and dementia stage will be analyzed for group The chosen design seems suitable for our purposes. We will include patients in differences using Chi square tests. all different stages of the disease by recruiting at the outpatient memory clinic and in nursing homes, and we will also consider all possible causes of pain that For the secondary outcome analyses, descriptive statistics and multivariable are most likely to be diagnosed due to our multidisciplinary approach. The linear regression analyses will be used. Multivariable linear regression will be chosen measurement instruments are all commonly used in the field of elderly used to evaluate the relationships between neuropsychiatric symptoms and pain, care medicine and geriatrics, except for the MOBID-2,OPS-NVI and the VU-AMS. quality of life and pain, and cognitive functioning and pain, in dementia subtypes. Observational pain assessment tools have limited validity and could lead to Differences in autonomic responses per dementia subtype, measured by the misinterpretation of pain behavior. For example, a grimace might be interpreted VU-AMS, as well as the effect of feedback on the presence of pain will be analyzed as a smile, or a positive score on an observational pain tool might be attributed using paired t-tests. to pain instead of some other cause of distress.20 However, there is currently no better method of assessing pain in a population with severe dementia beyond the proposed approach in this study. We acknowledge that this may lead to an overdetection of pain in this study. As we use several observational instruments Discussion this enables to determine correlations between the instruments. This will give insight in the probability of overdetection and also enhance knowledge about the The aim of the study is to investigate the prevalence and intensity of pain in the concurrent validity of the instruments. major dementia subtypes and for different dementia stages, as well as the relationship between the different types of pain and the major dementia For that reason we planned to investigate the contribution of autonomic subtypes. Primary outcome is the presence and intensity of pain in dementia responses to pain stimuli and QST as more objective measurement methods in subtypes. Secondary outcomes are oral health, autonomic responses to pain the assessment of pain in people with dementia. The reliability of QST in people stimuli, vital sensitivity, cognitive functioning, neuropsychiatric symptoms, and with dementia has already been demonstrated,21 although the use of QST in quality of life. Additionally, an evaluation of effectiveness of feedback on the epidemiologic studies on pain in dementia is scarce. presence of pain will be carried out, and the possibility of using autonomic responses for assessing pain in patients with dementia will be explored.

94 | PAIN IN DEMENTIA: PREVALENCE AND ASSOCIATED FACTORS: PROTOCOL OF A MULTIDISCIPLINARY STUDY CHAPTER 4 | 95 The success of this study in nursing home residents depends highly upon support References of the nursing staff. Participating in this research may be perceived as time- consuming by the nursing staff. Therefore, we decided to support the nurses by 1. Zwakhalen SM, Koopmans RT, Geels PJ, Berger MP, Hamers JP. The deploying research assistants. The research assistants will collect data on prevalence of pain in nursing home residents with dementia measured using behavior, quality of life, participation, and ADL by interviewing the contact nurse an observational pain scale. Eur J Pain. 2009;13:89–93. to ensure that participation is as little time-consuming as possible. These 2. Shega JW, Hougham GW, Stocking CB, Cox-Hayley D, Sachs GA. Pain in research assistants will be trained in administering the instruments correctly community-dwelling persons with dementia: frequency, intensity, and and consistently. congruence between patient and caregiver report. J Pain Symptom Manage. 2004;28:585–92. In conclusion, this study will help to enhance our knowledge regarding the 3. Van ‘t Hof CE, Zwakhalen SM, Hamers JP. Interventions after diagnosing prevalence of different types of pain in different dementia subtypes i.e. AD, VaD, pain in nursing home residents with dementia: the pilot implementation of FTD and DLB. This study also aims to contribute to a better understanding of an observational pain scale (PACSLAC-D). Tijdschr Gerontol Geriatr. oral health problems in people with dementia and the relationships between 2011;42:67–78. pain and cognitive symptoms, neuropsychiatric symptoms, and quality of life in 4. Patel KV, Guralnik JM, Dansie EJ, Turk DC. Prevalence and impact of pain people with various dementia subtypes and in different stages of the disease. among older adults in the United States: findings from the 2011 National Health and Aging Trends Study. Pain. 2013;154:2649–57. 5. Corbett A, Husebo B, Malcangio M, et al. Assessment and treatment of pain in people with dementia. Nat Rev Neurol. 2012;8:264–74. 6. Woolf AD, Vos T, March L. How to measure the impact of musculoskeletal conditions. Best Pract Res Clin Rheumatol. 2010;24:723–32. 7. Duncan R, Francis RM, Collerton J, et al. Prevalence of arthritis and joint pain in the oldest old: findings from the Newcastle 85+ study. Age Ageing. 2011;40:752–5. 8. Ahn H, Horgas A. The relationship between pain and disruptive behaviors in nursing home residents with dementia. BMC Geriatr. 2013;13:14. 9. Volicer L, Frijters DH, Van der Steen JT. Relationship between symptoms of depression and agitation in nursing home residents with dementia. Int J Geriatr Psychiatry. 2012;27:749–54. 10. Scherder EJ, Eggermont L, Plooij B, et al. Relationship between chronic pain and cognition in cognitively intact older persons and in patients with Alzheimer’s disease. The need to control for mood. Gerontology. 2008;54:50– 8. 11. Beerens HC, Sutcliffe C, Renom-Guiteras A, et al. Quality of Life of and Quality of Care for People With Dementia Receiving Long Term Institutional Care or Professional Home Care: The European RightTimePlaceCare Study. J Am Med Dir Assoc. 2013;15:54-61. 12. Scherder EJ, Sergeant JA, Swaab DF. Pain processing in dementia and its relation to neuropathology. Lancet Neurol. 2003;2:677–86. 13. Benedetti F, Arduino C, Costa S, et al. Loss of expectation-related mechanisms in Alzheimer’s disease makes analgesic therapies less effective. Pain. 2006;121:133–44.

96 | PAIN IN DEMENTIA: PREVALENCE AND ASSOCIATED FACTORS: PROTOCOL OF A MULTIDISCIPLINARY STUDY CHAPTER 4 | 97 14. Rainero I, Vighetti S, Bergamasco B, Pinessi L, Benedetti F. Autonomic dementia. Pain. 2006;126:210–20. responses and pain perception in Alzheimer’s disease. Eur J Pain. 30. Ersek M, Herr K, Neradilek MB, Buck HG, Black B. Comparing the 2000;4:267–74. psychometric properties of the Checklist of Nonverbal Pain Behaviors (CNPI) 15. Carlino E, Benedetti F, Rainero I, et al. Pain perception and tolerance in and the Pain Assessment in Advanced Dementia (PAIN-AD) instruments. patients with frontotemporal dementia. Pain. 2010;151:783–9. Pain Med. 2010;11:395–404. 16. Scherder EJ, Plooij B. Assessment and management of pain, with particular 31. Zwakhalen SM, Hamers JP, Abu-Saad HH, Berger MP. Pain in elderly people emphasis on central neuropathic pain, in moderate to severe dementia. with severe dementia: a systematic review of behavioural pain assessment Drugs Aging. 2012;29:701–6. tools. BMC Geriatr. 2006;6:3. 17. Lobbezoo F, Weijenberg RA, Scherder EJ. Topical review: orofacial pain in 32. Zwakhalen SM, van der Steen JT, Najim MD. Which score most likely dementia patients. A diagnostic challenge. J Orofac Pain. 2011;25:6–14. represents pain on the observational PAINAD pain scale for patients with 18. Cox MO. The issues and challenges of orofacial pain in the elderly. Spec Care dementia? J Am Med Dir Assoc. 2012;13:384–9. Dentist. 2000;20:245–9. 33. Husebo BS, Strand LI, Moe-Nilssen R, Husebo SB, Ljunggren AE. Pain in 19. Hadjistavropoulos T, Herr K, Turk DC, et al. An interdisciplinary expert older persons with severe dementia. Psychometric properties of the consensus statement on assessment of pain in older persons. Clin J Pain. Mobilization-Observation-Behaviour-Intensity-Dementia (MOBID-2) Pain 2007;23:S1–43. Scale in a clinical setting. Scand J Caring Sci. 2010;24:380–91. 20. Jordan A, Regnard C, O’Brien JT, Hughes JC. Pain and distress in advanced 34. Beaton DE, Bombardier C, Guillemin F, Ferraz MB. Guidelines for the dementia: choosing the right tools for the job. Palliat Med. 2012;26:873–8. process of cross-cultural adaptation of self-report measures. Spine. 21. Jensen-Dahm C, Werner MU, Dahl JB, et al. Quantitative sensory testing 2000;25:3186–91. Phila Pa 1976. and pain tolerance in patients with mild to moderate Alzheimer disease 35. Corbett A, Achterberg W, Husebo B, et al. An international road map to compared to healthy control subjects. Pain. 2014;155:1439-1445. improve pain assessment in people with impaired cognition. The development 22. Cleeland CS, Ryan KM. Pain assessment: global use of the Brief Pain of the Pain Assessment in Impaired Cognition (PAIC) tool. BMC Neurology, in Inventory. Ann Acad Med Singapore. 1994;23:129–38. press. 23. Gracely RH, McGrath P, Dubner R. Ratio scales of sensory and affective 36. Bonn CE, van Selms MK, te Poel F, Visscher CM, Lobbezoo F, Naeije M. verbal pain descriptors. Pain. 1978;5:5–18. Onderscheid tussen temporomandibulaire pijn en andere vormen van 24. Gracely RH, McGrath P, Dubner R. Validity and sensitivity of ratio scales of orofaciale pijn op basis van een vragenlijst. Ned Tijdschr Tandheelkd. sensory and affective verbal pain descriptors: manipulation of affect by 2013;120:28–33. diazepam. Pain. 1978;5:19–29. 37. Naeije M, Lobbezoo F, van Loon LA, et al. Behandelingsprotocol voor 25. Heft MW, Gracely RH, Dubner R, McGrath PA. A validation model for verbal craniomandibulaire dysfunctie 2. Behandeling Ned Tijdschr Tandheelkd. description scaling of human clinical pain. Pain. 1980;9:363–73. 2000;107:406–12. 26. Bieri D, Reeve RA, Champion GD, Addicoat L, Ziegler JB. The Faces Pain 38. van der Putten GJ, de Visschere L, Vanobbergen J, Schols JM, de Baat C. De Scale for the self-assessment of the severity of pain experienced by children: richtlijn mondzorg voor zorgafhankelijke cliënten in verpleeghuizen 2007: development, initial validation, and preliminary investigation for ratio scale Bittere noodzaak! Tijdschr Gerontol Geriatr. 2008;39:202–7. properties. Pain. 1990;41:139–50. 39. Kayser AF. Shortened dental arches and oral function. J Oral Rehabil. 27. Hicks CL, von Baeyer CL, Spafford PA, van Korlaar I, Goodenough B. The 1981;8:457–62. Faces Pain Scale-Revised: toward a common metric in pediatric pain 40. WHO. Oral Health Surveys - Basic methods. Geneva. World Health measurement. Pain. 2001;93:173–83. Organization, 1987 28. Warden V, Hurley AC, Volicer L. Development and psychometric evaluation 41. Wetselaar P, van der Zaag J, Lobbezoo F. Een beoordelingssysteem voor of the Pain Assessment in Advanced Dementia (PAINAD) scale. J Am Med Dir gebitsslijtage. Ned Tijdschr Tandheelkd. 2011;118:324–8. Assoc. 2003;4:9–15. 42. van der Velden V. The Dutch periodontal screening index validation and its 29. Zwakhalen SM, Hamers JP, Berger MP. The psychometric quality and application in The Netherlands. J Clin Periodontol. 2009;36:1018–24. clinical usefulness of three pain assessment tools for elderly people with 43. NVvP. Protocol Parodontale Diagnostiek en Behandeling. 2011.

98 | PAIN IN DEMENTIA: PREVALENCE AND ASSOCIATED FACTORS: PROTOCOL OF A MULTIDISCIPLINARY STUDY CHAPTER 4 | 99 44. Silness J, Loe H. Periodontal disease in pregnancy. II. Correlation between Br J Clin Psychol. 2000;39:181–91. oral hygiene and periodontal condition. Acta Odontol Scand. 1964;22:121–35. 61. Lindeboom J, Schmand B, Tulner L, Walstra G, Jonker C. Visual association 45. Svensson P, Drangsholt M, Pfau DB, List T. Neurosensory testing of orofacial test to detect early dementia of the Alzheimer type. J Neurol Neurosurg pain in the dental clinic. J Am Dent Assoc. 2012;143:e37–9. Psychiatry. 2002;73:126–33. 46. de Geus EJ, Willemsen GH, Klaver CH, van Doornen LJ. Ambulatory 62. Folstein MF, Folstein FE, McHugh PR. Mini-Mental State. A practical method measurement of respiratory sinus arrhythmia and respiration rate. Biol for grading the cognitive state of patients for the clinician. J Psychiatric Res Psychol. 1995;41:205–27. 1975;3:189–198. 47. Licht CM, Penninx BW, de Geus EJ. Effects of antidepressants, but not 63. Lopez MN, Charter RA, Mostafavi B, Nibut LP, Smith WE. Psychometric psychopathology, on cardiac sympathetic control: a longitudinal study. properties of the Folstein Mini-Mental State Examination. Assessment. Neuropsychopharmacology. 2012;37:2487–95. 2005;12:137–44. 48. Willemsen GH, de Geus EJ, Klaver CH, van Doornen LJ, Carroll D. 64. Wind AW, Schellevis FG, van Staveren G, Scholten RP, Jonker C, van Eijk Ambulatory monitoring of the impedance cardiogram. Psychophysiology. JT. Limitations of the Mini-Mental State Examination in diagnosing dementia 1996;33:184–93. in general practice. Int J Geriatr Psychiatry. 1997;12:101–8. 49. Olausson H, Wessberg J, Kakuda N. Tactile directional sensibility: peripheral 65. Kaufer DI, Cummings JL, Ketchel P, et al. Validation of the NPI-Q, a brief neural mechanisms in man. Brain Res. 2000;866:178–87. clinical form of the Neuropsychiatric Inventory. J Neuropsychiatry Clin 50. Scherder EJ, Rommelse NN, Broring T, Faraone SV, Sergeant JA. Neurosci. 2000;12:233–9. Somatosensory functioning and experienced pain in ADHD-families: a pilot 66. Wood S, Cummings JL, Hsu MA, et al. The use of the neuropsychiatric study. Eur J Paediatr Neurol. 2008;12:461–9. inventory in nursing home residents. Characterization and measurement. 51. Kawakami T, Ishihara M, Mihara M. Distribution density of intraepidermal Am J Geriatr Psychiatry. 2000;8:75–83. nerve fibers in normal human skin. J Dermatol. 2001;28:63–70. 67. Cummings JL, Mega M, Gray K, Rosenberg-Thompson S, Carusi DA, 52. van der Elst W, van Boxtel MPJ, van Breukelen GJP, Jolles J. The Stroop. Gornbein J. The Neuropsychiatric Inventory: comprehensive assessment of Color-Word Test: Influence of Age, Sex, and Education; and Normative Data psychopathology in dementia. Neurology. 1994;44:2308–14. for a Large Sample Across the Adult Age Range. Assessment. 2006;13:62– 68. Lange RT, Hopp GA, Kang N. Psychometric properties and factor structure 79. of the Neuropsychiatric Inventory Nursing Home version in an elderly 53. Saan RJ, Deelman BG. De 15-Woorden Test A en B (Een voorlopige neuropsychiatric population. Int J Geriatr Psychiatry. 2004;19:440–8. handleiding). AZG afd. Neuropsychologie, Groningen, 1986. 69. Kat MG, de Jonghe JF, Aalten P, Kalisvaart CJ, Droes RM. Verhey FR. 54. Rey A. L’examen clinique en psychologie. Paris: Presse de Universitaire de [Neuropsychiatric symptoms of dementia: psychometric aspects of the France; 1964. Dutch Neuropsychiatric Inventory (NPI)]. Tijdschr Gerontol Geriatr. 55. Lindeboom J, Jonker C. A. brief test for dementia screening: the ADS3. 2002;33:150–5. Tijdschr Gerontol Geriatr. 1988;19:97–102. 70. Brod M, Stewart AL, Sands L, Walton P. Conceptualization and measurement 56. Wilson BA, Evans JJ, Emslie H, Alderman N, Burgess P. The development of of quality of life in dementia: the dementia quality of life instrument (DQoL). an ecologically valid test for assessing patients with a dysexecutive Gerontologist. 1999;39:25–35. syndrome. Neuropsychol Rehabil. 1998;8:213–28. 71. Ettema TP, Droes RM, de Lange J, Mellenbergh GJ, Ribbe MW. QUALIDEM: 57. Osterrieth PA. Le test de copie d’une figure complexe: Contribution à l’étude development and evaluation of a dementia specific quality of life instrument– de la perception et de la mémoire. Arch Psychol. 1944;30:286–356. validation. Int J Geriatr Psychiatry. 2007;22:424–30. 58. Hammes JGW. De Stroop Kleur-Woord Test: Handleiding. Amsterdam: Swets 72. Ettema TP, Droes RM, de Lange J, Mellenbergh GJ, Ribbe MW. QUALIDEM: & Zeitlinger; 1973. development and evaluation of a dementia specific quality of life instrument. 59. Reitan RM. Validity of the Trail Making Test as an indicator of organic brain Scalability, reliability and internal structure. Int J Geriatr Psychiatry. damage. Percept Mot Skills. 1958;8:271–6. 2007;22:549–56. 60. Harrison JE, Buxton P, Husain M, Wise R. Short test of semantic and 73. Gerritsen DL, Steverink N, Frijters DH, Hirdes JP, Ooms ME, Ribbe MW. A phonological fluency: Normal performance, validity and test-retest reliability. revised Index for Social Engagement for long-term care. J Gerontol Nurs.

100 | PAIN IN DEMENTIA: PREVALENCE AND ASSOCIATED FACTORS: PROTOCOL OF A MULTIDISCIPLINARY STUDY CHAPTER 4 | 101 2008;34:40–8. 74. McKhann GM, Knopman DS, Chertkow H, et al. The diagnosis of dementia due to Alzheimer’s disease: recommendations from the National Institute on Aging-Alzheimer’s Association workgroups on diagnostic guidelines for Alzheimer’s disease. Alzheimers Dement. 2011;7:263–9. 75. Roman GC, Tatemichi TK, Erkinjuntti T, et al. Vascular dementia: diagnostic criteria for research studies. Report NINDS-AIREN Int Workshop Neurol. 1993;43:250–60. 76. Rascovsky K, Hodges JR, Knopman D, et al. Sensitivity of revised diagnostic criteria for the behavioural variant of frontotemporal dementia. Brain. 2011;134:2456–77. 77. McKeith IG, Dickson DW, Lowe J, et al. Diagnosis and management of dementia with Lewy bodies: third report of the DLB Consortium. Neurology. 2005;65:1863–72. 78. Reisberg B, Ferris SH, de Leon MJ, Crook T. The Global Deterioration Scale for assessment of primary degenerative dementia. Am J Psychiatry. 1982;139:1136–9. 79. Charlson ME, Pompei P, Ales KL, MacKenzie CR. A new method of classifying prognostic comorbidity in longitudinal studies: development and validation. J Chronic Dis. 1987;40:373–83. 80. Katz S, Ford AB, Moskowitz RW, Jackson BA, Jaffe MW. Studies of illness in the aged. The index of ADL: A standardized measure of biological and psychosocial function. JAMA. 1963;185:914–9. 81. Faul F, Erdfelder E, Lang AG, Buchner A. G*Power 3: a flexible statistical power analysis program for the social, behavioral, and biomedical sciences. Behav Res Methods. 2007;39:175–91. 82. Achterberg WP, Pot AM, Scherder EJ, Ribbe MW. Pain in the nursing home: assessment and treatment on different types of care wards. J Pain Symptom Manage. 2007;34:480–7. 83. Rejnefelt I, Andersson P, Renvert S. Oral health status in individuals with dementia living in special facilities. Int J Dent Hyg. 2006;4:67–71.

102 | PAIN IN DEMENTIA: PREVALENCE AND ASSOCIATED FACTORS: PROTOCOL OF A MULTIDISCIPLINARY STUDY CHAPTER 4 | 103 Prevalence of pain in nursing home residents: the role of dementia stage and dementia 5 subtypes 5 Prevalence of pain in nursing Acetaminophen (paracetamol) was the most prescribed analgesic (79.6%). home residents: the role of Conclusion: Most of the participating nursing home residents had no pain; however, pain was observed more often in residents with severe dementia, dementia stage and dementia whereas residents in the early stages of VaD self-reported pain more often that subtypes those with other dementia subtypes. As one-third of the residents with clinically relevant pain had moderate to Janine van Kooten, Martin Smalbrugge, Johannes C van der Wouden, Max L Stek, severe pain regardless of using pain medication, more focus should be on how Cees M P M Hertogh pain management could use more tailored approaches and be regularly adjusted to individual needs.

Abstract Introduction Objectives: To study pain prevalence, pain type, and its pharmacological treatment in Dutch nursing home residents in relation to dementia subtype and Pain is a common problem in people with dementia. Previous studies have re- dementia severity. ported that the prevalence of pain in people with dementia is approximately 50% and that pain is mainly caused by age-related musculoskeletal disorders like os- Design: Data were collected as part of the PAINdemiA study, an observational teoporosis and arthritis.1,2 cross-sectional study conducted between May 2014 and December 2015. Pain can be divided into two types: nociceptive pain and neuropathic pain. Noci- Setting: Ten nursing homes in the Netherlands ceptive pain is often associated with osteoarthritis, and it is considered the most prevalent pain type, while neuropathic pain, which can result from stroke and Participants: 199 nursing home residents in various stages of dementia. diabetes, is also a common phenomenon.3 However, neuropathic pain has been barely explored in nursing home residents with dementia.3 This is not surprising, Measurements: We collected data on pain (by observation: MOBID-2 Pain considering that the diagnosis of neuropathic pain lies in the expression of specif- Scale and by self-report scales), pain type, pain medication, dementia subtype, ic complaints and this ability might be limited in people with dementia.3,4 As pre- dementia severity (GDS) and demographic features. vious studies in the general population have reported higher pain intensities for people suffering from pain with neuropathic characteristics (e.g. tingling, burning Results: In the whole sample the prevalence of pain was 43% (95% CI = 36- or shooting) than for people suffering from pain without neuropathic characteris- 50%) using the MOBID-2 Pain Scale. Regardless of regularly scheduled tics,5 recognizing pain with neuropathic characteristics is a pivotal point in provid- analgesics, approximately one-third of the residents with pain suffered from ing optimal pain treatment in nursing home residents with dementia. moderate to severe pain. Pain assessment with the MOBID-2 Pain Scale showed no difference in pain between dementia subtypes, but residents with Pain assessment in nursing home residents with dementia is a challenging task.6 more severe dementia experienced pain more often than those with less severe In the earlier stages of dementia pain assessment is usually possible via self-re- dementia (27% vs 15%). The prevalence of self-reported pain was significantly port of pain, while in the later stages self-report becomes impossible. In these higher in residents with vascular dementia (VaD) (53%) compared to those with stages pain assessment depends on observation of pain and detection of pain Alzheimer’s disease (18%) and other dementia subtypes (14%). Nociceptive related behavior.7 Dementia specific assessment tools are helpful in detecting pain was the predominant type of pain (70%) followed by mixed pain (25%). these pain related behaviors; although, these tools are not able to distinguish pain from other causes of discomfort, and they cannot make a clear distinction

106 | PREVALENCE OF PAIN IN NURSING HOME RESIDENTS: THE ROLE OF DEMENTIA STAGE AND DEMENTIA SUBTYPES. CHAPTER 5 | 107 between the different pain types.6,8 Yet, the challenge of accurately identifying From the source population for this study (679 nursing home residents that were pain remains necessary for effective pain treatment.9 living at a dementia special care unit in one of the participating nursing homes) we could finally include a sample of 199 residents. We received a signed informed Some studies indicate that residents with severe dementia are particularly at risk consent for 219 residents (32.3%), of these residents 20 (2.9%) were subsequently to suffer from pain,10,11 whereas other studies found that pain prevalence did not excluded because they did not meet the inclusion criteria or moved or died before differ among residents in different stages of dementia.12 Dementia-specific neu- the start of the assessment. For 252 residents we did not receive any response, ropathological changes may be responsible for alterations in pain perception.1,13 and 208 residents refused to participate (Figure 1). Disturbances in the pain system by white matter lesions may increase pain ex- perience in people with vascular dementia (VaD) compared to other dementia Figure 1. Flowchart of inclusion. subtypes, while gray matter atrophy may lead to a decrease in pain experience in people with Alzheimer’s Disease (AD) and frontotemporal dementia (FTD).13-15 Nevertheless, the current body of research is ambiguous as the findings on pain source population experience in dementia subtypes are inconsistent, and the direction of the impact (n = 679) of neuropathology may differ in dementia subtypes.16 refusal (n = 208) & no response (n = 252) The aim of this study was to determine and compare pain prevalence, pain type, signed informed consent and its pharmacological treatment in Dutch nursing home residents in relation to (n = 219) dementia subtype and dementia severity. residents excluded (n = 20) Reasons for exclusion: participants - no diagnosis of dementia Methods (n = 199) - died or moved before start of the assessment

Design Study design A cross-sectional observational study of nursing home residents was carried out as part of the PAINdemiA-study. Data was collected from May 2014 to December 2015, and all measurements of a resident were administered within a time frame of 1 week. The full protocol has been published elsewhere.17 Measurement and residents’ characteristics Study population Pain was assessed by different methods that were partly adjusted to the level We invited all residents (n = 679) that were living at a dementia special care unit in of communication of the resident. In all residents the Mobilization Observation one of the participating Dutch nursing homes (n = 10) to participate in the study. Behaviour Intensity Dementia (MOBID-2) Pain Scale was used to assess pain. The Once a nursing home had agreed to participate, legal representatives of all res- MOBID-2 is a two part, nursing staff-administered observational instrument: it idents were informed and asked to give consent. Inclusion criteria were aged 60 assesses immediate pain behaviors in response to five standardized active, guid- years or older, being diagnosed with dementia, and having a signed informed ed movements of different body parts during morning care, and it assesses pain consent form by the legal representative. Residents were excluded when they behaviors that might originate from internal organs, head, and skin over time.18,19 had a diagnosis of Huntington’s disease, Parkinson’s disease dementia, alcohol Using a numeric rating scale, the overall pain intensity of the MOBID-2 ranges related dementia, or when they suffered from cognitive deficits due to psychi- from 0 to 10, and an overall score of ≥ 3 is regarded as clinically relevant pain.19 atric disorders, or when they were primary mentally disabled. This study was Studies on the psychometric properties of the MOBID-2 have shown good reli- approved by the Medical Ethics Review Committee of the VU University Medical ability, validity, and feasibility in clinical practice.18,19 As the frequency distribution Center (Amsterdam, The Netherlands). of this scale was skewed, MOBID-2 scores were categorized into three groups: no pain (0-2), mild pain (3-4), and moderate-severe pain (5 and above).

108 | PREVALENCE OF PAIN IN NURSING HOME RESIDENTS: THE ROLE OF DEMENTIA STAGE AND DEMENTIA SUBTYPES. CHAPTER 5 | 109 In residents who were able to communicate about the presence of pain the re- pain, there were no specific diagnosis or references to potential neuro- searcher also asked the resident about pain through self-report. Depending on pathic pain in the medical record. the residents’ understanding of self-report scales, as judged by the researcher, 3. Mixed pain : during physical examination there were complaints and/or the Numeric Rating Scale (NRS), Verbal Descriptor Scale (VDS), and/or Faces signs that could be related to both nociceptive and neuropathic pain. In Pain Scale Revised (FPS-R) were used to measure pain.20-22 The NRS ranges from the medical record were references to potential neuropathic pain and/or 0 to 10, and it has a cut-off score of 4 and more for clinically relevant pain. The to conditions that are known to cause mainly nociceptive pain. VDS has five ordinal categories (no pain- mild pain- moderate pain- severe pain- worst pain possible), and pain categorized as ‘moderate pain’ or more is regarded For all residents data on age and gender was collected using electronical medical as clinically relevant pain. The FPS-R consists of 6 drawings of faces ranging records. from a neutral face to a grimacing face. From the third face onward pain is re- garded as clinically relevant.23 The dementia subtype diagnosis was taken from the medical record, but in case the dementia subtype was not specified (n = 2), the attending physician was con- In residents with difficulties to communicate their pain, as judged by the re- tacted to obtain the dementia subtype diagnosis. We categorized these subtypes searcher, the Pain Assessment IN Advanced Dementia Scale (PAINAD) was used into four groups: Alzheimer’s Disease (AD), vascular dementia (VaD), mixed de- to assess pain.24 The PAINAD is a brief and easy to administer observational scale mentia, and other dementias. Different from our protocol,17 we added the cate- especially used in people with advanced dementia. It assesses five behaviors re- gory mixed dementia, and we categorized the people with DLB and FTD in the lated to pain, and each of these is rated on a three point scale (0-2). The total subgroup of other dementias.26 score ranges from 0-10 an a cut-off score of 2 has been recommended as indica- tion for the presence of pain.25 The reliability and validity of the PAINAD have been Dementia severity was assessed using the Global Deterioration Scale (GDS).27 studied extensively, and they are well established.24 This is a seven point scale that describes seven stages of cognitive impairment from ‘no global impairment’ (1) to ‘very severe global impairment’ (7). Stages In all residents with an indication of clinically relevant pain, defined as NRS ≥ 4, 4 and higher are considered to represent consecutive dementia stages. The in- or VDS ‘moderate pain’ or more, or FPS ≥ ‘third face’, or PAINAD ≥ 2, or MOBID-2 formation to assign global ratings of cognitive impairment was derived through ≥ 3, physical examination was performed with emphasis on the musculoskeletal interviews with the nursing staff. system. The type of pain was determined by combining the findings of this exam- ination and the medical (history) data in the medical record.3,4 We distinguished Activities of daily living (ADL) were measured using the modified version of Katz three groups3,7: index of independence in ADL (Katz-ADL), an observational scale asking if the 1. Neuropathic pain: resident needs help with six basic activities of daily living (bathing, dressing, toi- a. Current Neuropathic Pain (CNP): during physical examination com- leting, continence, getting out of a chair, and eating), and it has good psycho- plaints, signs, and/or symptoms associated with neuropathic pain metric properties.28 Information about comorbidity was recorded from residents’ were observed, and a diagnosis of neuropathic pain was reported in medical records, and the disease groups of Charlson Comorbidity Index were the medical record. used to categorize comorbidity.29 The sum of the disease groups was dichoto- b. Possible Neuropathic Pain (PNP): during physical examination com- mized on the median. plaints, signs, and/or symptoms associated with neuropathic pain were observed, and in the medical record we found references relat- Data on analgesic drug use was derived from pharmacists’ files and classified ed to neuropathic pain: for instance, descriptions of signs, or symp- according to the Anatomical Therapeutic Chemical classification (ATC).30 Five toms; or a history suggestive for the pain to be of neuropathic origin groups of analgesics were distinguished: Acetaminophen (paracetamol) (ATC (e.g. stroke, diabetes mellitus, surgery). code N02 BE01), non-steroidal anti-inflammatory drugs (NSAIDs) (ATC code 2. Nociceptive pain: during physical examination there were complaints M01), weak opioids (ATC code N02 AA59), strong opioids (ATC code N02 AA01, and/or signs that referred to pain and most likely arise from actual or -AA03, -AA05, -AE01, -AB03, -AX02), and specific antineuropathic drugs (ATC threatened damage to non-neural tissue, and in contrast to neuropathic code N06 AA09 (amitriptyline), N06 AA10 (nortriptyline), N06 AX21 (duloxetine),

110 | PREVALENCE OF PAIN IN NURSING HOME RESIDENTS: THE ROLE OF DEMENTIA STAGE AND DEMENTIA SUBTYPES. CHAPTER 5 | 111 Table 1. N02 AX02 (tramadol), N03 AF01 (carbamazepine), N03AX12 (gabapentin), and Measurement instruments N03 AX16 (pregabalin)). We only classified drugs as antineuropathic when neu- ropathic pain was mentioned as indication in the medical record. Furthermore, Characteristics n=199 a distinction was made between regularly scheduled analgesics and analgesics prescribed as needed. Age in years, mean (SD) 84.9 (6.5) Female 154 (77.4%) Analyses Male 45 (22.6%) We used descriptive statistics (means and standard deviations (SDs) for dis- Dementia severity tributed continuous level variables, medians and interquartile ranges (IQRs) GDS unknown 4 (2%) for non-normally distributed continuous level variables, and percentages for GDS 4 12 (6%) categorical variables), and calculated the prevalence of pain including the 95% GDS 5 46 (23.1%) confidence interval (CI). Differences in pain between dementia subtypes were GDS 6 99 (49.7%) determined with Chi Square analyses and post-hoc analyses using Bonferroni GDS 7 38 (19.1%) correction. Differences in pain between dementia stages were determined with Dementia subtype Chi Square analyses and post-hoc analyses using logistic multinomial regression AD 106 (53.3%) analyses. SPSS 22.0 (IBM Corp., Armonk, NY) was used to analyze the data. VaD 20 (10.1%) Mixed dementia 31 (15.6%) Results not otherwise specified 42 (21.1%) Katz-ADL (SD) 3.77 (1.77)

Sample Number of comorbidities (IQR) 1 (2) Table 1 shows the residents’ characteristics. In our study we included 199 nurs- Myocardial infarction 22 (11.1%) ing home residents who were mostly female (77.4%). Age ranged from 66 to 102 years with a mean age of 84.9 (SD 6.5) years. The predominant dementia type was Congestive heart failure 21 (10.6%) Alzheimer’s Disease (53.3%), followed by mixed dementia, and vascular demen- Peripheral vascular disease 15 (7.5%) tia. In 17.6% of the cases dementia was classified as ‘other’. Most residents were Cerebrovascular disease 60 (30.2%) in a moderately severe stage of dementia (GDS 6) and there was no difference in dementia severity between the dementia subtypes. The number of comorbidities Chronic pulmonary disease 28 (14.1%) was rather low, especially the number of musculoskeletal and connective tissue Musculoskeletal and connective tissue 4 (2%) disorders, that did not include osteoarthritis. The number of comorbidities (di- disorders(osteoarthritis not included)s chotomized at the median) differed significantly between the dementia subtypes Stomach ulcer 9 (4.5%) (Pearson Chi Square 8.998; df 3; p = 0.029), with more comorbidities in those with Liver disease 1 (0.5%) VaD compared to those with AD . Renal disease 35 (17.6) Diabetes mellitus 43 (21.6%) Cancer 7 (3.5%) Prevalence of pain 1 = Patient, 2 = Caretaker, 3 = Health professionals,T0 = baseline, T1 = after 3 months. MOBID-2 The MOBID-2 was assessed in almost all participants (n = 195; 98%), and the prevalence of pain measured as MOBID-2 score ≥3 was 43.1% (95% CI 36.1- 50.1%). The median pain score of the 84 residents with MOBID-2 score ≥3 was 4

112 | PREVALENCE OF PAIN IN NURSING HOME RESIDENTS: THE ROLE OF DEMENTIA STAGE AND DEMENTIA SUBTYPES. CHAPTER 5 | 113 (IQR 3; ranges 3-9), and 41 residents (21%) had moderate-severe pain (MOBID-2 Figure 2. Self-reported pain prevalence related to dementia subtypes. ≥ 5). Of the residents with pain (i.e. MOBID-2 ≥ 3) 26.8% had a GDS score of 7 as compared to 14.7% in those without pain (Pearson Chi Square 4.33; df 1; p = 0.037). The prevalence of pain did not differ between dementia subtypes (Pearson Chi Square 2.69; df 3; p = 0.441). Additional analyses showed that moderate-se- vere pain (MOBID-2 ≥ 5) was significantly associated with the GDS score of 7 (OR 2.58; p = 0.031) (Table 2).

PAINAD Of the 67 residents that were not able to self-report their pain 18 residents had a PAINAD score of ≥ 2, and the prevalence of pain was 26.9% (95% CI 16-38%). The median score was 4 (IQR 4), and 15 of the residents with pain had a GDS score of 7 (83%, 95% CI 64-100%), whereas 21 of the 49 residents without pain had a GDS *pain measured on at least one of the three self-report scales: NRS, VDS or FPS. score of 7 (43%, 95% CI 28-57%). The prevalence of pain measured by the PAINAD AD: Alzheimer’s Disease; VaD: Vascular Dementia. did not differ between dementia subtypes.

Self-reported pain There were 122 residents that were able to self-report their pain on at least one of Pain types the three self-report scales. All of them had a GDS score < 7, and the prevalence We determined the type of pain (i.e. neuropathic pain, nociceptive pain, or mixed of pain was 22.1% (95% CI 14.7-29.6%). The median score of the eight residents pain) in residents with presence of clinically relevant pain (i.e. pain measured on with a NRS score > 3 was 5 (IQR 3), the median score of the residents with pain on one of the five measurement scales) and a complete physical examination; two the VDS (n = 23) was moderate pain, and for those reported pain using the FPS (n residents were categorized to have current neuropathic pain, 59 residents were = 7) the median inferred score was 4 (IQR 4). categorized in the group with nociceptive pain, and 21 residents in the group with mixed pain (i.e. nociceptive pain combined with possible neuropathic pain). In this Post-hoc analyses showed significant differences in self-reported pain preva- sample all cases with possible neuropathic pain had features of nociceptive pain, lence between residents with VaD and both residents with AD (p = 0.025) and therefore they were categorized as having mixed pain (see Figure 3). The causes residents with other dementias (p = 0.026) (Figure 2). of neuropathic pain in the current neuropathic pain group were polyneuropathy and neuropathic pain of multicausal origin, and in the mixed pain group the differ- ential diagnoses commonly included old fractures, contractures, and pain origi- Table 2. nating from the musculoskeletal system. There was no difference in type of pain prevalence of pain per measurement tool and per dementia subtype (current neuropathic pain and mixed pain vs nociceptive pain) between dementia subtypes (Pearson Chi Square 1.95; df 3; p = 0.582) nor a difference in type of pain Total AD Mixed VaD (n=20) Other (n=199) (n=106) (n=31) (n=42) between residents with GDS score of 7 and residents with GDS score <7. MOBID-2(n=195) 84/195 (43%) 43 (41.7%) 12 (38.7%) 12 (60%) 17 (41.5%)

Self-report scale*(n=122) 27/122(22.1%) 11 (17.7%) 5 (26.3%) 7 (53.8%) 4 (14.3%) Pharmacological treatment We found that 43% of the whole study population and 58% of the residents with PAINAD (n=67) 18/67(26.9%) 11(28.9%) 1 (10%) 1 (14.3%) 5 (41.7%) MOBID-2 score ≥ 3 used regularly scheduled analgesics. Residents with identified Data are number of residents . AD: Alzheimer’s Disease; VaD: Vascular Dementia; MOBID-2: Mobilization pain were more likely to have regularly scheduled analgesic drug use than those Observation Behaviour Intensity Dementia (MOBID-2) Pain Scale; *pain measured on at least one of the three self- without pain (Pearson Chi Square 23.39; df 1; p < 0.001). We found no difference report scales used in this study; PAINAD: Pain Assessment IN Advanced Dementia Scale (PAINAD). in the use of regularly scheduled analgesic drugs between dementia subtypes

114 | PREVALENCE OF PAIN IN NURSING HOME RESIDENTS: THE ROLE OF DEMENTIA STAGE AND DEMENTIA SUBTYPES. CHAPTER 5 | 115 Table 3. Figure 3. Pain type related to dementia subtypes. dementia subtype and prescription of analgesics (n=199)

Total AD (n=106) Mix (n=31) VaD (n=20) Other (n=42)

Analgesic monotherapy 68/199 68/199 12/31 5/20 16/42 Acetaminophen 11 3 15 NSAIDs 0 0 0

Opioids 0 2 1 antineuropathic drugs 1 0 0

Analgesic combination therapy 19/199 19/199 2/31 3/20 3/42 No prescribed analgesics 76/199 76/199 10/31 7/20 15/42 AD: Alzheimer’s Disease; VaD: Vascular Dementia; mixed pain: nociceptive pain combined Prescribed as needed with possible neuropathic pain. 36/199 36/199 7/31 5/20 8/42

Data are number of residents.; NSAIDs: non-steroidal anti-inflammatory drugs; antineuropathic drugs: e.g. amitriptyline, gabapentin, carbamazepine.

(Pearson Chi Square 1.152; df 3; p = 0.764) nor between residents with a GDS score of 7 and GDS score < 7 (Fisher’s exact test p = 0.146) (Table 3).

Acetaminophen (paracetamol) was the most prescribed analgesic (79.6%), fol- Discussion lowed by oxycodone (6.6%), and fentanyl patches (6.6%). We found five prescrip- tions for anti-neuropathic drugs in four residents, who were still in pain. There This study determined and compared pain prevalence, pain type, and its phar- were no significant differences in types of prescribed analgesics between demen- macological treatment in Dutch nursing home residents in relation to dementia tia subtypes nor dementia severity. subtype and dementia severity. The prevalence of pain in the whole sample was 43% using the MOBID-2; that is, most of the participating nursing home residents Additional analyses showed that, regardless of regularly scheduled analgesics, did not show pain behaviour that could be regarded as clinically relevant pain. In about one third of the residents with clinically relevant pain suffered from moder- addition, pain assessment with the MOBID-2 showed that residents with more se- ate to severe pain (MOBID-2 ≥ 5). vere dementia experienced pain more often and with higher pain intensities than those with less severe dementia. Prevalence of pain as measured by the MOBID-2 did not differ between the dementia subtypes.

In residents that could self-report their pain, there was a higher prevalence of pain in residents with VaD compared to those with AD and to those with other de- mentia subtypes. This can be regarded as support for the hypothesis that people with VaD experience pain more frequently.13,31 As the residents in the self-report group were in less severe stages of dementia, it is possible that neuropathologi- cal differences play a more obvious role in the early stages of dementia, and that this role becomes ambiguous as dementia progresses.

116 | PREVALENCE OF PAIN IN NURSING HOME RESIDENTS: THE ROLE OF DEMENTIA STAGE AND DEMENTIA SUBTYPES. CHAPTER 5 | 117 In addition to the finding that the majority of the participating nursing home res- A strength of this study was the use of the MOBID-2 that scores pain based on idents did not suffer from clinical relevant pain, it should be also mentioned that standardized movements during morning care as well as pain behavior related most of the residents with clinical relevant pain, based on the MOBID-2 score, to other parts of the body over time. Observed pain during day-by-day activities suffered from mild pain. This is in line with earlier studies that used the MO- and guided movement during morning care are close to the day-to-day reality; BID-2; for example, the study by Husebo et al.(2010)19 that found a mean score therefore, they reflect clinical practice. Further, we consider the use of a combi- of 4.1 in residents with MOBID-2 score ≥ 3, or the more recent study by Sandvik nation of pain measurements as a strength, because the MOBID-2 mostly iden- et al.(2014)32 that found a mean score of 3.7 in residents with MOBID-2 score ≥ 3. tified nociceptive pain related to the musculoskeletal system, whether the self- However, we found that 21% of the residents had moderate-severe pain. Perhaps report-scales were of additional value in identifying pain originated outside the this is because of co-morbid depressive symptoms which are associated with musculoskeletal system. more pain complaints and greater pain intensities.33,34 These depressive symp- toms are also known to complicate pain management,34 and as we found clinical This study has also some limitations. First, the response rate was 32.2%. This significant signs of depression in 25% of the residents,35 attention should be paid possibly limits the external validity of the study (i.e. generalizability); however, to the coexistence of depressive symptoms when assessing and treating pain. the characteristics of the participants in our study are congruent with that of other studies including Dutch nursing home residents with dementia.40,41 Second, Nociceptive pain was the predominant type of pain followed by mixed pain, and we included only a few people with FTD and DLB, for this reason we could not there were only two cases of current neuropathic pain. As expected from the find- report data on pain for these dementia subtypes separately. Third, determining ings of Kollenburg et al.(2012),3 the number of cases with current neuropathic the type of pain, especially neuropathic pain, is challenging in nursing residents pain was actually low, but mixed pain seemed rather common in our sample. with dementia because it is generally based on history taking and neurophysical This is in line with earlier reports on the prevalence of neuropathic like pain in measurements; therefore, not all cases of neuropathic pain may have been iden- older people with osteoarthritis.36 As we found a low number of prescriptions of tified correctly. To facilitate this in the future, Quantative Sensory Testing (QST) antineuropathic drugs this might be a starting point to improve pain treatment; could be a promising tool.42 for example, by prescribing antineuropathic drugs as adjuvant analgesics in res- In conclusion, most of the participating nursing home residents did not show idents with mixed pain.1,6 pain behaviour that could be regarded as clinically relevant pain. Residents with severe dementia had more often pain and also higher pain intensities than resi- The pharmacological treatment of pain did not differ between between dementia dents with mild to moderate dementia (27% vs. 15%) and residents with VaD who subtypes-or for dementia severity. The rate of regularly scheduled analgesics could report pain themselves reported pain more often than residents with AD was in line with previously published data, but we found a considerable number or other dementia subtypes. As one-third of the residents with clinically relevant of residents that still suffered from moderate to severe pain (MOBID ≥ 5), despite pain had moderate to severe pain regardless of using pain medication, more fo- the use of regularly scheduled analgesics.37,38 This association has been shown cus should be on how pain management could use more tailored approaches and before, and it suggests that regularly scheduled analgesics not necessarily re- be regularly adjusted to individual needs. flects optimal pain treatment.10,36 Moreover, we found lower rates of opioid use than Sandvik et al.(2014), who also found that increasing the dosages of analgesic drugs significantly improved the overall pain intensity in nursing home residents with dementia.32 Painmanagement in the way of complete relief is obviously very difficult, but systematic medication reviews might contribute to a more tailored treatment approach and eventually reduce pain intensity.

Strengths and limitations This was the first attempt to determine pain types in nursing home residents with dementia.

118 | PREVALENCE OF PAIN IN NURSING HOME RESIDENTS: THE ROLE OF DEMENTIA STAGE AND DEMENTIA SUBTYPES. CHAPTER 5 | 119 References subtypes: A systematic review. Curr Alzheimer Res. In press. 17. van Kooten J, Delwel S, Binnekade TT et al. Pain in dementia: prevalence 1. Achterberg WP, Pieper MJ, van Dalen-Kok AH, et al. Pain management in and associated factors: protocol of a multidisciplinary study. BMC Geriatr patients with dementia. Clin Interv Aging 2013;8:1471-1482. 2015;15:29. 2. Fejer R, Ruhe A. What is the prevalence of musculoskeletal problems in the 18. Husebo BS, Strand LI, Moe-Nilssen R, Husebo SB, Snow AL, Ljunggren AE. elderly population in developed countries? A systematic critical literature re- Mobilization-Observation-Behavior-Intensity-Dementia Pain Scale (MOBID): view. Chiropr Man Therap 2012;20:31. development and validation of a nurse-administered pain assessment tool 3. van Kollenburg EG, Lavrijsen JC, Verhagen SC, Zuidema SU, Schalkwijk for use in dementia. J Pain Symptom Manage 2007;34:67-80. A, Vissers KC. Prevalence, causes, and treatment of neuropathic pain in 19. Husebo BS, Strand LI, Moe-Nilssen R, Husebo SB, Ljunggren AE. Pain in Dutch nursing home residents: a retrospective chart review. J Am Geriatr older persons with severe dementia. Psychometric properties of the Mobili- Soc 2012;60:1418-1425. zation-Observation-Behaviour-Intensity-Dementia (MOBID-2) Pain Scale in a 4. Treede RD, Jensen TS, Campbell JN, et al. Neuropathic pain: redefini- clinical setting. Scand J Caring Sci 2010;24:380-391. tion and a grading system for clinical and research purposes. Neurology 20. Heft MW, Gracely RH, Dubner R, McGrath PA. A validation model for verbal 2008;70:1630-1635. description scaling of human clinical pain. Pain 1980;9:363-373. 5. Bouhassira D, Lanteri-Minet M, Attal N, Laurent B, Touboul C. Prevalence of 21. Hicks CL, von Baeyer CL, Spafford PA, van Korlaar I, Goodenough B. The chronic pain with neuropathic characteristics in the general population. Pain Faces Pain Scale-Revised: toward a common metric in pediatric pain mea- 2008;136:380-387. surement. Pain 2001;93:173-183. 6. Corbett A, Husebo B, Malcangio M, et al. Assessment and treatment of pain 22. Price DD, McGrath PA, Rafii A, Buckingham B. The validation of visual ana- in people with dementia. Nat Rev Neurol 2012;8:264-274. logue scales as ratio scale measures for chronic and experimental pain. Pain 7. AGS Panel on Persistent Pain in Older Persons. The management of per- 1983;17:45-56. sistent pain in older persons. J Am Geriatr Soc 2002;50:S205-S224. 23. Gerbershagen HJ, Rothaug J, Kalkman CJ, Meissner W. Determination of 8. Jordan A, Regnard C, O’Brien JT, Hughes JC. Pain and distress in advanced moderate-to-severe postoperative pain on the numeric rating scale: a cut-off dementia: choosing the right tools for the job. Palliat Med 2012;26:873-878. point analysis applying four different methods. Br J Anaesth 2011;107:619- 9. Zwakhalen SM, Hamers JP, Abu-Saad HH, Berger MP. Pain in elderly people 626. with severe dementia: a systematic review of behavioural pain assessment 24. Warden V, Hurley AC, Volicer L. Development and psychometric evaluation tools. BMC Geriatr 2006;6:3. of the Pain Assessment in Advanced Dementia (PAINAD) scale. J Am Med Dir 10. Zwakhalen SM, Koopmans RT, Geels PJ, Berger MP, Hamers JP. The prev- Assoc 2003;4:9-15. alence of pain in nursing home residents with dementia measured using an 25. Zwakhalen SM, van der Steen JT, Najim MD. Which score most likely rep- observational pain scale. Eur J Pain 2009;13:89-93. resents pain on the observational PAINAD pain scale for patients with de- 11. Herr K, Zwakhalen S, Swafford K. Observation of Pain in Dementia. Curr mentia? J Am Med Dir Assoc 2012;13:384-389. Alzheimer Res 2016. 26. Lo CD, Lopez G, Corrao S. Cognitive impairment and stroke in elderly pa- 12. Takai Y, Yamamoto-Mitani N, Okamoto Y, Koyama K, Honda A. Literature re- tients. Vasc Health Risk Manag 2016;12:105-116. view of pain prevalence among older residents of nursing homes. Pain Manag 27. Reisberg B, Ferris SH, de Leon MJ, Crook T. The Global Deterioration Nurs 2010;11:209-223. Scale for assessment of primary degenerative dementia. Am J Psychiatry 13. Scherder EJ, Sergeant JA, Swaab DF. Pain processing in dementia and its 1982;139:1136-1139. relation to neuropathology. Lancet Neurol 2003;2:677-686. 28. Katz S. Assessing self-maintenance: activities of daily living, mobility, and 14. Benedetti F, Vighetti S, Ricco C, et al. Pain threshold and tolerance in Alzhei- instrumental activities of daily living. J Am Geriatr Soc 1983;31:721-727. mer’s disease. Pain 1999;80:377-382. 29. Charlson ME, Pompei P, Ales KL, MacKenzie CR. A new method of classify- 15. Carlino E, Benedetti F, Rainero I, et al. Pain perception and tolerance in pa- ing prognostic comorbidity in longitudinal studies: development and valida- tients with frontotemporal dementia. Pain 2010;151:783-789. tion. J Chronic Dis 1987;40:373-383. 16. Binnekade TT, van Kooten J, Lobbezoo F, et al. Pain experience in dementia 30. World Health Organization (WHO). Anatomical Therapeutical Chemical (ATC)

120 | PREVALENCE OF PAIN IN NURSING HOME RESIDENTS: THE ROLE OF DEMENTIA STAGE AND DEMENTIA SUBTYPES. CHAPTER 5 | 121 classification index including defined daily doses (DDDs) for plain substanc- es. Oslo,World Health Organization Coolaborative Centre for Drugs Statistics Methodology, 2011. 31. Scherder EJ, Plooij B, Achterberg WP, et al. Chronic pain in “probable” vas- cular dementia: preliminary findings. Pain Med 2015;16:442-450. 32. Sandvik RK, Selbaek G, Seifert R, et al. Impact of a stepwise protocol for treating pain on pain intensity in nursing home patients with dementia: a cluster randomized trial. Eur J Pain 2014;18:1490-1500. 33. Smalbrugge M, Jongenelis LK, Pot AM, Beekman AT, Eefsting JA. Pain among nursing home patients in the Netherlands: prevalence, course, clini- cal correlates, recognition and analgesic treatment – an observational co- hort study. BMC Geriatrics 2007;7:3. 34. Bair MJ, Robinson RL, Katon W, Kroenke K. Depression and pain comor- bidity: A literature review. Archives of Internal Medicine 2003;163:2433-2445. 35. Van KootenJ, Van der Wouden JC, Sikkes SAM, Smalbrugge M, Hertogh CMPM, Stek ML. Pain, neuropsychiatric symptoms, and quality of life of nurs- ing home residents with advanced dementia in the Netherlands: A cross-sec- tional study. Alzheimer Dis Assoc Disord 2017;31:315-321. 36. de Luca KE, Parkinson L, Byles JE, Lo TK, Pollard HP, Blyth FM. The preva- lence and cross-sectional associations of neuropathic-like pain among older, community-dwelling women with arthritis. Pain Med 2016. 37. de Souto BP, Lapeyre-Mestre M, Vellas B, Rolland Y. Potential underuse of analgesics for recognized pain in nursing home residents with dementia: a cross-sectional study. Pain 2013;154:2427-2431. 38. Sandvik R, Selbaek G, Kirkevold O, Aarsland D, Husebo BS. Analgesic pre- scribing patterns in Norwegian nursing homes from 2000 to 2011: trend anal- yses of four data samples. Age Ageing 2016;45:54-60. 39. Husebo BS, Strand LI, Moe-Nilssen R, Husebo SB, Aarsland D, Ljunggren AE. Who suffers most? Dementia and pain in nursing home patients: a cross-sectional study. J Am Med Dir Assoc 2008;9:427-433. 40. Pieper MJ, Francke AL, van der Steen JT, et al. Effects of a stepwise mul- tidisciplinary intervention for challenging behavior in advanced dementia: a cluster randomized controlled trial. J Am Geriatr Soc 2016;64:261-269. 41. Zwijsen SA, Smalbrugge M, Eefsting JA, et al. Coming to grips with chal- lenging behavior: a cluster randomized controlled trial on the effects of a multidisciplinary care program for challenging behavior in dementia. J Am Med Dir Assoc 2014;15:531-10. 42. Krumova EK, Geber C, Westermann A, Maier C. Neuropathic pain: is quanti- tative sensory testing helpful? Curr Diab Rep 2012;12:393-402.

122 | PREVALENCE OF PAIN IN NURSING HOME RESIDENTS: THE ROLE OF DEMENTIA STAGE AND DEMENTIA SUBTYPES. CHAPTER 5 | 123 Evaluation of a pain assessment procedure in long-term care residents with 6 pain and dementia 6 Evaluation of a pain assessment Background procedure in long-term Pain is a disturbing, and frequently seen, symptom in long-term care (LTC) res- idents with dementia. Pain is mainly caused by age-related chronic musculo- care residents with pain and skeletal conditions and the prevalence of pain has been estimated to be 50% in LTC residents with dementia.1,2 Although pain is well known to be present in this dementia population, it is often ineffectively managed.2 Despite best-practice recommen- dations for pain assessment and treatment in LTC residents with dementia,3,4 it Janine van Kooten, Martin Smalbrugge, Johannes C. van der Wouden, Max L. seems that physicians frequently do not adhere to these recommendations in Stek, Cees M. P. M. Hertogh everyday work. This may result in underdetection of pain and eventually inappro- priate pain treatment. Earlier work to improve pain treatment in LTC residents with dementia focused on implementation of pain assessment, promotion of re- assessment, and education of health care staff to dispel misconceptions about Abstract pain in dementia.2 Recently, a review on improvement of treatment strategies in palliative care reported positive effects of audit and feedback, a strategy that Background: The management of pain in long-term care (LTC) residents with aims to improve professional practice by showing the professional inconsisten- dementia is complex. A prospective exploratory study was conducted to describe cies in their clinical practice.5 the course of pain and pain management strategies following a guideline-based pain assessment procedure in LTC residents with pain and dementia. One of the challenges to achieve adequate pain management is to ensure that, in addition to gathering information, health care professionals actually use pain Measures: Pain observation with the Mobilization Observation Behaviour assessment information.2 This prospective study explored the course of pain and Intensity Dementia (MOBID-2) pain Scale, a review of the electronic patient file pain management strategies following a guideline-based pain assessment pro- and pharmacy files and physical examination of LTC residents with pain and cedure and included audit and feedback through communication of the assess- dementia. ment results together with a written treatment recommendation to the attending physician in LTC residents with pain and dementia. We report on course of pain, Intervention: Communication of the assessment results to the attending applied pain management strategies, including pharmacological treatment, and physician including a guideline-based treatment recommendations. agreement between the treatment recommendations and the applied pain man- agement strategies. Outcomes: After three months, complete follow-up data was obtained for 64 residents. Pain intensity was significantly reduced (p < 0.001). The proportion of residents with persistent pain was 58% and the total number of analgesic prescriptions did not change significantly. Measures

Conclusions: There is room for improvement regarding pain management in Data Collection and Study Population long-term care residents with pain and dementia, and performance feedback Data were collected as part of the PAINdemiA study, with the overall aim to inves- seems a promising strategy to explore further. tigate the prevalence of pain per dementia subtype. The included 199 LTC res- idents were from 10 LTC-facilities (n = 679) in The Netherlands, recruited from Key words: dementia, pain, treatment, feedback, long-term care May 2014 to December 2015. The study protocol was approved by the Medical Ethics Review Committee of the VU University Medical Center (Amsterdam, The Netherlands). The full protocol has been published elsewhere.6

126 | EVALUATION OF A PAIN ASSESSMENT PROCEDURE IN LONG-TERM CARE RESIDENTS WITH PAIN AND DEMENTIA CHAPTER 6 | 127 Study Procedure the dementia subtype diagnosis. We categorized these subtypes into four groups: The pain assessment procedure followed the Dutch guideline ‘Recogniton and Alzheimer’s Disease , vascular dementia , mixed dementia, and other dementias treatment of chronic pain in vulnerable elderly people’,4 and included a pain (i.e. dementia not otherwise specified, dementia with Lewy Bodies and fronto- measurement by a nurse who was familiar with the resident, chart review, and temporal dementia). Dementia severity was assessed using the Global Deterio- physical examination to identify possible causes and the type of pain: nociceptive ration Scale (GDS).9 This is a 7-point scale describing stages of cognitive impair- pain, neuropathic pain, or mixed pain (i.e. combination of nociceptive pain and ment from ‘no global impairment’ (1) to ‘very severe global impairment’ (7). neuropathic pain). In the week before the assessment procedure was scheduled, nursing staff were instructed in the use of the pain assessment scale during an 1 Statistical analyses hour-meeting with the researcher. The researcher, at the time of data collection Descriptive statistics (means and standard deviations for continuous level an elderly care medicine trainee (i.e., the equivalent of a geriatric medicine train- variables, and percentages for categorical variables) were used for residents’ ee), performed the chart reviews and the physical examinations. Approximately, ­demographic characteristics, treatment recommendations, and applied treat- three months after the baseline assessment, the nursing staff was asked to re- ment strategies. To estimate differences in mean and standard error (SE) of the assess pain in participants with pain during the baseline assessment, and the MOBID-2 between baseline and follow-up measurement, we used the Wilcoxon ­ researcher performed another chart review to obtain information about applied signed rank test; and for differences in pharmacological treatment between pain management strategies. baseline and follow-up, we used the McNemar test. We report the proportion of residents that had persistent pain at follow-up. All statistical analyses were Outcome measures ­performed using SPSS v 22.0 (IBM Corporation, New York, NY). Pain was assessed with the Mobilization Observation Behaviour Intensity Demen- tia (MOBID-2) Pain Scale, a two part, staff-administered observational instrument Intervention that assesses immediate pain behaviors in response to five standardized active, For each resident, all findings of the guideline-based pain assessment procedure guided movements of different body parts during morning care, and assesses were summarized into a report by the elderly care medicine trainee in coopera- pain behaviors that might originate from internal organs, head, and skin over tion with the research team that included two elderly care physicians and a spe- time.7 Using a numeric rating scale, the overall pain intensity of the MOBID-2 cialist in palliative medicine. The reports included treatment recommendations ranges from 0 to 10, and an overall score of ≥ 3 is regarded as clinically relevant for the attending elderly care physician and was send by e-mail within one week pain. Psychometric properties of the MOBID-2 in clinical practice were found to after the assessment. be appropriate and a change score of ≥ 3 is considered clinically relevant.7,8 Outcomes Information about the applied pain management strategies was obtained from We obtained complete follow-up data for 64 residents, for 20 residents there were the electronical patient file (EPF). We searched for documentation of pain man- no observations obtained at follow-up: 8 residents had died, 3 moved and for 9 agement strategies from the date of the first assessment until date of the second residents there were no observations available due to other causes (e.g., work assessment and assessed the types of pain management strategies applied to load nursing staff, residents on holiday). There were no differences in demo- alleviate pain. graphic characteristics or pain intensity at baseline between the residents with and without follow-up data. Pharmacological treatment provided for pain was derived from pharmacy files -cdata on type and dosage of analgesics. Furthermore, a distinction was made The mean age was 84.2 (SD 6.4) and the majority was female (75%). Most resi- between regularly scheduled analgesics and analgesics prescribed as needed dents had severe or very severe dementia (GDS 6 or GDS 7), and the most com- (PRN). mon dementia subtype was Alzheimer’s disease (45.3%).

Other Measures In 63 of the 64 residents, a report on the presence of pain and possible causes of The dementia subtype diagnosis was taken from the EPF; and in case the demen- pain had been provided following the guideline-based pain assessment; and in 58 tia subtype was not specified, the attending physician was contacted to obtain residents, a treatment recommendation was part of the report. For two residents,

128 | EVALUATION OF A PAIN ASSESSMENT PROCEDURE IN LONG-TERM CARE RESIDENTS WITH PAIN AND DEMENTIA CHAPTER 6 | 129 Table. 1 pain management was recently changed and now seemed adequate; two resi- Treatment recommendations, implementation and subsequent change in pain dents did not accept physical examination and/or treatment, and one report did not include a treatment recommendation, because the resident was erroneously Type of recommendation No. Documentation Implementation Change in pain scored as not suffering from pain. Recommendations could be both nonpharma- send to the attending in medical re- intensity physician cords cological as well as pharmacological, and followed the Dutch guideline ‘Recog- 4 1. Non-pharmacological, for 8 4 Implemented - - niton and treatment of chronic pain in vulnerable elderly people’. Table 1 shows example, consultation with the recommendations and which were used. an occupational therapist, Partly implemented 2 2 Pain ↓ physical therapist and/or Not implemented 2 1 Pain ↓ psychologist; to use washing 1 Pain ↑ Course of Pain without water; to consider The mean MOBID-2 score at baseline was 4.80 (SD 1.5) and after three months adapted clothing or orthope- dic shoes; to promote more 3.23 (SD 2.5). The results show a significant improvement over time with a mean exercise; to consider bed rest change of 1.57 (SE 0.3; Z -4.329; p < 0.001) and we found that 57.8% (n = 37) was during the day.

still in pain after three months (MOBID-2 score ≥ 3). The proportion of residents with severe pain, MOBID-2 score 5, did not significantly change over time (50% 2. To start/increase analge- 16 9 Implemented 7 4 Pain ↓ ≥ sic drugs 2 Pain ↑ vs 54%). 1 Pain = Partly implemented 1 1 Pain ↓ Pain management strategies Not implemented 1 1 Pain ↓ We found documentation of applied pain management strategies in 36 of the 64 3. To stop/decrease analge- 1 1 Implemented 1 1 Pain ↓ EPF (56.3%). Table 1 shows the used recommendations and how these recom- sic drugs Partly implemented - - mendations affected the pain intensity. We found that 18 residents, of those with applied pain management strategies documented in their medical records, had a Not implemented - - MOBID-2 score of ≥ 3, 10 of them even had an MOBID-2 score of ≥ 5. Of the resi- 4. Combination of 1 + 2 19 10 Implemented 6 2 Pain ↓ 3 Pain ↑ dents without applied pain management strategies documented in their EPF, 19 1 Pain = residents had an MOBID-2 score of ≥ 3, 10 of them had an MOBID-2 score of ≥ 5. Partly implemented 1 1 Pain ↓ There was no difference in pain persistence or pain intensity between residents Not implemented 3 1 Pain ↓ with and without documentation of applied pain management strategies in their 2 Pain ↑ medical records. 5. Additional examination by 4 2 Implemented 2 1 Pain ↓ the attending elderly care Pharmacological treatment physician and/or multidisci- 1 Pain = plinary team Analyses of pharmacological treatment showed that there were 77 prescriptions Partly implemented - - of analgesic drugs in 49 residents at baseline, prescription of medication as Not implemented - - needed included (32.5%). After three months, there were 64 prescriptions of an- 6. To continue current treat- 10 10 Implemented 10 10 8 Pain ↓ algesic drugs in 44 residents and prescription of medication as needed included ment regime 2 Pain = (31.3%). There was no significant difference in the total number of prescriptions Partly implemented - - of analgesic drug between baseline and follow-up Not implemented - - (table 2). Total 58 36 36

130 | EVALUATION OF A PAIN ASSESSMENT PROCEDURE IN LONG-TERM CARE RESIDENTS WITH PAIN AND DEMENTIA CHAPTER 6 | 131 Table. 2 Number of analgesic drug prescriptions per group of analgesic drug mented at reassessment, our guideline-based pain assessment procedure might have improved pain management or at least boosted the awareness of pain. The Baseline Three months † p-value awareness of pain might have increased by participating in this study and the Acetaminophen 53 45 0.687 presence of external researchers. NSAIDs 3 2 1.00 Of note, we stratified residents on the threshold level for pain intensity (i.e. MO- Weak opioids 0 1 1.00 BID-2 ≥ 3) at baseline. With this selection we introduced potential for pain inten- Strong opioids 17 12 1.00 sity to show regression toward the mean, which was conformed with a scatter- Anti-neuropathic drugs 4 4 1.00 plot of change against baseline measurements. This is in line with the study of Sandvik et al.,10 who used the MOBID-2 Pain Scale and found a mean change of NSAIDs = nonsteroidal anti-inflammatory drugs. 1.0 (SD 2.2) over time in their control group.10 But as our mean change of 1.57 † Mc Nemar test. (SE 0.3; Z -4.329; p 0.001) is slightly higher, the change in pain may not be fully Data are numbers of prescriptions. < explained by regression to the mean. Another explanation might be the course of pain; Hendriks et al.11 found an intermittent course of pain in some residents and reported resolution of pain proportions of 10% to 13%. It should be noted that the In more detail, the pharmacological treatment data revealed 47 changes in pre- design of our study did not include reassessment of residents without pain, nor scription of analgesics. In 19 residents, there was no change in analgesic treat- corrected for dementia stage or exclude residents who were near the end of life, ment; in 10 residents, analgesic treatment was ended or the dosage was lowered, while burdensome symptoms as pain are known to increase at the end of life.11 for example, several weeks after a fall; in 4 residents, analgesic treatment was There was no significant difference in prescriptions of analgesic drugs three started or the dosage was increase; and in 14 residents, who used multiple an- months after the guideline-based pain assessment procedure. It could be specu- algesic drugs, one of the analgesic drug prescriptions was changed (seven times lated that communication of assessment results together with written treatment dosage increase and seven times dosage decrease). recommendations may have encouraged some elderly care physicians to review Most changes in analgesic prescriptions (33 of 40) were documented in the EPF their treatment decisions and reflect on them. also as applied pain management strategy. Of the residents with persistent pain (MOBID-2 score of ≥ 3) at three months (n = 37), 29 residents (78%) had a pre- Strengths and limitations scription for analgesic of whom 62% used regularly scheduled pain medication. We performed a structural assessment of pain at two different moments in time by observers who were familiar with the resident. Furthermore, we used a vali- Conclusions dated and reliable observational instrument.7 The findings of this study are valu- The aims of this exploratory study were to describe the course of pain and pain able for daily practice, because of their ecologic validity. That is, this study took management strategies following a guideline-based pain assessment procedure place in a field setting with data collection by people who could be using the pain that included communication of the assessment results and written treatment assessment in real life, and therefore, the results are probable more generaliz- recommendations to the attending physician of LTC residents with pain and de- able. Lastly, another strength of our study is presenting the results of the pain mentia. We designed this guideline- based assessment procedure to improve assessment including treatment recommendations to the attending physician, pain management in LTC facilities, and we found a significant change in the mean avoiding possible communication problems between nursing staff and the at- pain intensity between baseline and three months later. However, pain persisted tending physician. in more than half of the participating residents, and there was no change in the proportion of residents with severe pain nor in the use of analgesic drugs. There are also some limitations that should be mentioned. First, our study did not include a control group and it will be necessary to repeat this observation in Pain intensity was significantly lower three months after the guideline-based future studies. Second, the lack of a comparison group limits the interpretation of pain assessment procedure. There are several possible explanations for this change in pain intensity over time. Third, the selection of residents was based on result. Although only half of our treatment recommendations had been imple- one single measurement at baseline which might have increased the measure-

132 | EVALUATION OF A PAIN ASSESSMENT PROCEDURE IN LONG-TERM CARE RESIDENTS WITH PAIN AND DEMENTIA CHAPTER 6 | 133 ment variability. Lastly, data on treatment strategies were collected by retrospec- References tive chart review. As medical records are not designed for research purposes, this study is prone to information bias. To minimize this bias, data on treatment 1. Corbett A, Husebo B, Malcangio M, et al. Assessment and treatment of pain strategies were complemented with data from pharmacy files. in people with dementia. Nat. Rev. Neurol 2012;8:264-274. 2. Herr K. Pain assessment strategies in older patients. J Pain 2011;12:S3-S13. Lessons Learned 3. Hadjistavropoulos T, Herr K, Turk DC, et al. An interdisciplinary expert This study showed that pain treatment is a complex task, and that solely commu- consensus statement on assessment of pain in older persons. Clin. J Pain nicating guideline-based pain assessment results including treatment recom- 2007;23:S1-43. mendations did not result in a clinical relevant change in pain intensity. However, 4. Achterberg WP, de Ruiter CM, de Weerd-Spaetgens CM, Geels P, Horikx A, the intervention presented in this paper can be easily applied in clinical practice, Verduij M. [Multidisciplinary guideline ‘Recognition and treatment of chronic and future studies should combine performance feedback with implementation pain in vulnerable elderly people’]. Ned Tijdschr Geneeskd. 2012;155: A4606.. methods that take the local situation into account. Furthermore, we recommend 5. van Riet PJ, Vernooij-Dassen M, Sommerbakk R, et al. Implementation of to include a control group, and to use case report forms to record data in a pro- improvement strategies in palliative care: an integrative review. Implement. spective and systematic way instead of retrospective chart review. Sci 2015;10:103. 6. van Kooten J, Delwel S, Binnekade TT, et al. Pain in dementia: prevalence and associated factors: protocol of a multidisciplinary study. BMC Geriatr 2015;15:29. 7. Husebo BS, Strand LI, Moe-Nilssen R, et al. Pain in older persons with se- vere dementia. Psychometric properties of the Mobilization-Observation-Be- haviour-Intensity-Dementia (MOBID-2) Pain Scale in a clinical setting. Scand. J. Caring. Sci 2010;24:380-391. 8. Husebo BS, Ostelo R, Strand LI. The MOBID-2 pain scale: reliability and re- sponsiveness to pain in patients with dementia. Eur. J. Pain 2014;18:1419- 1430. 9. Reisberg B, Ferris SH, de Leon MJ, Crook T. The Global Deterioration Scale for assessment of primary degenerative dementia. Am. J Psychiatry 1982;139:1136-1139. 10. Sandvik RK, Selbaek G, Seifert R, et al. Impact of a stepwise protocol for treating pain on pain intensity in nursing home patients with dementia: a cluster randomized trial. Eur. J. Pain 2014;18:1490-1500. 11. Hendriks SA, Smalbrugge M, Galindo-Garre F, et al. From admission to death: prevalence and course of pain, agitation, and shortness of breath, and treatment of these symptoms in nursing home residents with dementia. J. Am. Med. Dir. Assoc 2015;16:475-481.

134 | EVALUATION OF A PAIN ASSESSMENT PROCEDURE IN LONG-TERM CARE RESIDENTS WITH PAIN AND DEMENTIA CHAPTER 6 | 135 Pain, neuropsychiatric symptoms, and quality of life of nursing home residents with advanced dementia in the Netherlands: a 7 cross-sectional study 7 Pain, neuropsychiatric relationship between pain and QoL subdomains, but NPS, in particular agitation symptoms, and quality of and depressive symptoms, were significantly associated with lower QoL subdomain scores. Agitation was related to lower scores on the subdomains life of nursing home residents “relationship”(95% confidence interval, CI, -0.083 to -0.059), “positive affect”(95% CI -0.037 to -0.013), “restless tense behaviour”(95% CI -0.003 to with advanced dementia in -0.004), and “social relations”(95% CI -0.033 to -0.009), whereas depression was related to lower scores on the subdomains “positive affect”(95% CI the Netherlands: -0.054 to -0.014), “negative affect”(95% CI -0.114 to -0.074), “restless tense behaviour”(95% CI -0.075 to -0.025), and “social relations”(95% CI -0.046 to a cross-sectional study -0.002).

Janine van Kooten, Johannes C. van der Wouden, Sietske A.M. Sikkes, Martin Conclusion: Only NPS were significantly associated with QoL in nursing home Smalbrugge, Cees M.P.M. Hertogh, Max L. Stek residents with dementia. Further longitudinal research is needed to estimate the nature of the relationship between pain, NPS and QoL.

Abstract

Background: Many studies have investigated factors associated with Quality Introduction of life (QoL) in nursing home residents with dementia. Both pain and neuropsychiatric symptoms (NPS) are clinically relevant and individually Dementia is one of the most burdensome health conditions worldwide, and the associated with a lower QoL; however, there are no studies that investigated number of persons with dementia is still growing. The estimated number of peo- pain and NPS together in relation to QoL. ple affected will amount to 65.7 million in 2030.1 As dementia is not curable yet, maintaining and maximizing of quality of life (QoL) is one of the treatment goals Purpose: In this study we explored the relationship of pain and NPS tot QoL in in caring for people with dementia,2,3 especially for those residing in a nursing nursing home residents with dementia by investigating the association between home.4,5 pain concurrently with NPS, and QoL. QoL is generally described as a complex, multidimensional construct and is de- Methods and patients: Secondary data analyses of cross-sectional data from fined by the World Health Organization as: “individuals‘ perception of their posi- 199 residents collected by observations at dementia special care units of 10 tion in life in the context of the culture and value systems in which they live and nursing homes. QoL was measured with Qualidem, pain with the Mobilization in relation to their goals, expectations, standards and concerns.”6 QoL consists Observation Behaviour Intensity Dementia (MOBID-2) Pain Scale and NPS with of different domains, which might be interrelated and all contribute to an overall the Neuropsychiatric Symptoms Inventory. The relation of pain and NPS to QoL QoL.7 This overall QoL could differ in structure between subpopulations; howev- was studied using multiple linear regression analyses. Analyses were adjusted er, two domains are important to all populations: self-esteem and good social for age, sex, activities of daily living, comorbidity, medication use and dementia contacts.8 According to Ettema et al.,9 the concept of QoL in the subpopulation severity. of people dementia could be defined as: “the multidimensional evaluation of the person-environment system of the individual, in terms of adaptation to the per- Results: Regression models with pain and NPS, showed no independent ceived consequences of the dementia.” Domains that are often used to determine QoL in dementia are: self-esteem, social relations or interactions, mood or af-

138 | PAIN, NEUROPSYCHIATRIC SYMPTOMS, AND QUALITY OF LIFE OF NURSING HOME RESIDENTS WITH ADVANCED CHAPTER 7 | 139 DEMENTIA IN THE NETHERLANDS: A CROSS-SECTIONAL STUDY fect, physical functioning and (social) environment. living at a dementia special care unit in one of the participating nursing homes to participate in the study. Nursing home residents were eligible to participate when In recent years, many studies have investigated factors influencing QoL in peo- they had a diagnosis of dementia, were aged 60 years or older and had a signed ple with dementia;10-13 however, few studies have investigated the association be- informed consent by the legal representative. tween pain and QoL. In case an association was found, pain was often associated with a lower QoL in people with dementia.12,14-16 For example, Torvik et al.14 found Nursing home residents were excluded when they had a diagnosis of Hunting- that self-reported pain was related to lower scores on the negative affect domain, ton’s disease, Parkinson’s disease dementia, alcohol related dementia, or when but they did not control for the presence of neuropsychiatric symptoms (NPS), people suffered from cognitive deficits due to psychiatric disorders or when peo- such as agitation and depressive symptoms, which, like pain, are highly prevalent ple were primary mentally disabled. and also related to lower QoL in people with dementia.17,18 Of the source population of 679 residents, we received 219 (32.3%) signed in- The etiology of NPS is multifactorial and among these factors is pain. For this formed consent forms; of these 219 residents, 20 (2.9%) were subsequently ex- reason NPS might be an expression of pain, however, the presence of NPS in cluded because they did not meet the inclusion criteria or moved or died before itself might also attribute to a higher pain intensity.19,20 In literature, a higher pain the start of the data collection. Data were collected by trained care staff, research intensity has been associated with, for example, agitation and depressive symp- assistants and an elderly care medicine trainee (J.v.K.).This study was approved toms.20,21 Moreover, depressive symptoms have been related to a lower QoL and by the Medical Ethics Review Committee of the VU University Medical Center it has even been the most consistent predictor of a reduced QoL in people with (Amsterdam, The Netherlands). dementia.12 Yet, there are no studies that investigated the relation between QoL and NPS together with pain. Quality of life QoL was assessed by a trained research assistant using the Qualidem question- As QoL has increasingly become an outcome measure of research studies, gain- naire, which is a proxy rated dementia-specific instrument that provides a QoL ing more insight into the contribution of pain and NPS to QoL might help indicate profile of nursing home residents with dementia.23,24 For people with mild to se- directions towards new interventions aiming at improving QoL in people with de- vere dementia this questionnaire consist of 37 items with 4 possible responses mentia, especially in nursing homes. In this study we explored the relationship of (never, rarely, sometimes and frequently) and the responses to these items rep- pain and NPS to QoL by investigating the association between pain, NPS and QoL resent 9 subdomains: (1) care relationship ( 7 items, Cronbach α = 0.82 in this in data from the “Pain in dementia Amsterdam(PAINdemiA)” study.22 sample), (2) positive affect (6 items, Cronbach α = 0.86 ), (3) negative affect (3 items, Cronbach α = 0.73), (4) restless tense behaviour (3 items, Cronbach α = 0.61), (5) positive self-image (3 items, Cronbach α = 0.44), (6) social relations (6 items, Cronbach α = 0.70), (7) social isolation (3 items, Cronbach α = Methods 0.26), (8) feeling at home (4 items, Cronbach α = 0.65), and (9) having something to do (2 items, Cronbach α = 0.52). We only included subscales with a Cronbach Design and study population α of ≥ 0.55.25 For people with very severe dementia the questionnaire consist of Secondary data analyses of data from a cross-sectional study of nursing home 18 items representing 6 subdomains (care relationship, positive affect, negative residents that was carried out as part of the “PAINdemiA” study. We collected affect, restless tense behaviour, social relations, social isolation).24 For all resi- data between May 2014 and December 2015 from 199 nursing home residents dents we subsequently calculated combined subscale scores (range, 0 to 3) by in The Netherlands. Ten nursing homes in and near Amsterdam participated the adding the item scores for each subdomain and dividing the sum by the number “PAINdemiA” study.22 Care organizations were approached by the University Net- of items used, with a high score indicating better QoL.26 work of Organisation for Care for the Elderly of the VU University Medical Cen- ter (UNO-VUmc) to allow their nursing homes to take part in the study. Once a Pain nursing home agreed to participate, legal representatives of all residents were Pain was measured with the Mobilization Observation Behaviour Intensity De- informed and asked to give consent. We invited all residents (n = 679) that were mentia (MOBID-2) Pain Scale, a staff-administered observational instrument, on

140 | PAIN, NEUROPSYCHIATRIC SYMPTOMS, AND QUALITY OF LIFE OF NURSING HOME RESIDENTS WITH ADVANCED CHAPTER 7 | 141 DEMENTIA IN THE NETHERLANDS: A CROSS-SECTIONAL STUDY the basis of pain behaviours in response to 5 standardized active, guided move- to severe dementia) and GDS = 7 (i.e., very severe dementia). ments of different body parts and pain behaviours related to internal organs, head, and skin.27 The overall pain intensity ranges from 0 to 10 and a score of ≥ 3 Statistical Analyses is regarded as clinically relevant pain.27,28 As the frequency distribution of the MO- Descriptive statistics were calculated (means and SDs for normal distributed BID-2 pain scale was skewed, data were categorized into three groups: no pain (0 continuous level variables, median and interquartile ranges for non-normal dis- to 2), mild pain (3 to 4) and moderate-severe pain (≥ 5). tributed continuous level variables, and percentages for categorical variables) and differences in QoL subdomain scores between residents in GDS < 7 and GDS NPS = 7 were compared using the independent sample t test for normal distributed Residents’ NPS were assessed by a trained research assistant using the Dutch continuous variables. Version of the Neuropsychiatric Inventory (NPI), which contains 12 items: delu- sions, hallucinations, agitation, anxiety, depression, elation, apathy, disinhibition, We used multiple linear regression analyses models to analyze the association irritability, aberrant motor behaviour, sleep problems and eating disorders.29,30 between pain (no pain, mild pain, severe pain), NPS (total score or five factor The care staff member was asked if a symptom was present over the past 2 scores) and all QoL subdomains with a Cronbach α of ≥ 0.55.25 Initially, a random weeks and if a symptom was present they were ask to rate both the frequency intercept was added to adjust for potential dependency of the residents within the (1 to 4) and severity (1 to 3) of each symptom. Total score (range, 0 to 144) and nursing homes. As there was no dependency of the residents within the nursing factor scores were used in the analyses. Factor scores were calculated as a sum homes, we present the results without random intercept. The calculated intra- of the component scores (F x S): (1) agitation consisting of agitation/aggression, class correlation coefficients were between 0.000 and 0.036.34 euphoria, disinhibition and irritability (range, 0 to 48); (2) depression including depression and anxiety (range, 0 to 24); (3) psychosis comprising delusions and We adjusted all analyses for the a priori selected variables: sex (female/male), hallucinations (range, 0 to 24); (4) psychomotor agitation encompassing aberrant age (in years), ADL (total score), comorbidity (1, 2, 3, ≥ 4), psychoactive drug use motor behaviour and sleep problems (range, 0 to 24); and (5) apathy including (yes/no), and severity of dementia (GDS < 7, GDS = 7). Because of the number apathy and eating disorders (range, 0 to 24).11,17 of participating residents (n = 199), interactions between independent variables were not modelled. We used SPSS 22.0 (IBM Corp., Armonk, NY) for the descrip- Other measurements tive analyses and for all other analyses we used MLwin version 2.28 (University of Data on age and sex was collected using electronic medical records for all res- Bristol, Bristol, UK). idents. Activities of daily living (ADL) were measured with the modified version of Katz index of independence in ADL (Katz-ADL) (range, 1 to 6) and comorbidity was assessed with the Charlson Index of Comorbidity (range, 0 to 26).31,32 As the frequency distribution was skewed, we categorized the data into four percentile Results groups based on index score: 1, 2, 3 and 4 or above. Data on psychoactive drug use were derived from pharmacists’ files and classified according to the Anatom- Characteristics ical Therapeutic Chemical classification. We dichotomized the data in two cate- We included 199 residents, but for 8 residents (6%) we did not collect all the data. gories: daily psychoactive drug use “yes” or “no”. In 5 residents data on QoL and NPS were missing, in 4 residents data on pain were missing, and in 4 residents data on dementia severity was missing. Table 1 The dementia subtype diagnosis was taken from the medical record. We cate- shows the residents’ characteristics. The total study sample (n = 199) had a mean gorized them into 4 groups: Alzheimer’s Disease, Vascular Dementia, mixed de- age of 84.6 (SD, 6.5) and 77.4% (n = 154) of the residents were female. The ma- mentia and ‘miscellaneous’. Dementia severity was assessed using the Global jority of the residents were in a moderately severe stage of dementia (GDS = 6). Deterioration Scale (GDS).33 This is a 7 point scale that describes seven stages Pain was present in 43.2% (n = 84) of the residents and 83% (n = 162) of the of cognitive impairment from ‘no global impairment’ (1) to ‘very severe global residents had shown at least 1 NPS in the previous 2 weeks. The most common impairment’ (7). Stages 4 and higher are considered to represent consecutive symptom was agitation/aggression (49%) followed by irritability (41%) and apa- dementia stages.33 We dichotomized the data in two categories: GDS < 7(i.e., mild thy (30%). Hallucinations, euphoria, and appetite showed the lowest percentage

142 | PAIN, NEUROPSYCHIATRIC SYMPTOMS, AND QUALITY OF LIFE OF NURSING HOME RESIDENTS WITH ADVANCED CHAPTER 7 | 143 DEMENTIA IN THE NETHERLANDS: A CROSS-SECTIONAL STUDY rates (7 to 15%). About 48% (n = 96) of the residents used at least 1 psychotropic Quality of life drug a day; antidepressants (28%), antipsychotics (18.2%), anxiolytics (16.8%), As can been seen in Table 2, the Qol measured with Qualidem was generally on hypnotics (14.7%), anti-dementia drugs (14.7%) and anti-epileptics (7.7%). a moderate to high level; with the highest scores on the subdomains “positive affect” and “feeling at home”. Subdomains “positive affect” and “social relations” had significantly higher scores for those with mild to severe dementia (GDS< 7).

Table 1. Demographic and clinical characteristics of the study sample (n=199)

Table 2. Characteristics Mean/Median/N* % Quality of Life- Qualidem Subscales Mean age (SD), years 84.9 (6.5) % Female 77.4% Mean (SD) Dementia severity Total sample GDS 4-6 (n=154) GDS 7 P % GDS unknown 4 2% (n=191) (n=37) % GDS 4 12 6% Care relationship (0-3) 2.16 (0.76) 2.20 (0.69) 2.01 (0.99) 0.278 % GDS 5 46 23.1% Positive affect (0-3) 2.42 (0.63) 2.49 (0.57) 2.13 (0.79) 0.013 * % GDS 6 99 49.7% Negative affect (0-3) 2.21 (0.77) 2.25 (0.75) 2.05 (0.86) 0.204 % GDS 7 38 19.1% Restless tense behaviour (0-3) 1.92 (0.87) 1.98 (0.84) 1.67 (0.93) 0.064 Charlson comorbidity (range 0-26), median (IQR) 2 (2) % Psychotropic drug use 96 48.2% Positive self-image (0-3)† 2.56 (0.59) 2.56 (0.59) † Mean ADL (SD) 3.77 (1.77) Social relations (0-3) 1.89 (0.69) 1.95 (0.64) 1.61 (0.82) 0.023 * Dementia subtype Social isolation (0-3) 2.19 (0.71) 2.21 (0.66) 2.13 (0.89) 0.654 % AD 106 53.3% Feeling at home (0-3)† 2.44 (0.67) 2.44 (0.67) † % VaD 20 10.1% Having something to do (0-3)† 1.07 (1.03) 1.07 (1.03) † % Mixed dementia 31 15.6% % other 42 21.1% † values only calculated for residents with GDS score < 7. GDS indicates Global Deterioration Scale. MOBID-2 (range 0-10), median (IQR) 2 (4) *Significant different, P< 0.05. % missing 4 2% % no pain (≤2) 111 55.8% % mild pain (3-4) 43 21.6% % moderate-severe pain (≥5) 41 20.6% Relationship of pain and neuropsychiatric symptoms to QoL NPI (range 0-144), mean (SD) 16.2 (15.3) Table 3 shows the associations of pain and NPI total score with the QoL sub-

NPI factor agitation (0-48) 6.00 (7.94) domains “care relationship”, “positive affect”, “negative affect”, “restless tense behavior” and “social relations”. Differences in Qualidem scores appeared to be NPI factor depression (0-24) 2.69 (4.40) mostly explained by the presence of NPS. Both unadjusted and adjusted analyses NPI factor psychosis (0-24) 1.82 (3.88) showed that NPS, when analysed together with pain, were related to lower scores NPI factor psychomotor agitation (0-24) 3.02 (4.72) on all QoL subdomains in nursing home residents with dementia (p < 0.001). In NPI factor apathy (0-24) 1.69 (4.10) the unadjusted analyses, a higher pain score was significantly associated with a

*Mean, median or number of residents. Figures in parentheses are standard deviations or interquartile ranges lower score on the subdomain “negative affect” ( 95% confidence interval -0.525 GDS: Global Deterioration Scale; IQR: interquartile range; ADL: Activities of Daily Living; AD: Alzheimer’s Disease; to -0.032). No other association between pain and QoL subdomains approached VaD: Vascular Dementia; MOBID-2: Mobilization Observation Behaviour Intensity Dementia (MOBID-2) Pain Scale; statistical significance. NPI: Neuropsychiatric Inventory.

144 | PAIN, NEUROPSYCHIATRIC SYMPTOMS, AND QUALITY OF LIFE OF NURSING HOME RESIDENTS WITH ADVANCED CHAPTER 7 | 145 DEMENTIA IN THE NETHERLANDS: A CROSS-SECTIONAL STUDY Table 3. Table 4 shows the results from both unadjusted and adjusted analyses with pain Relationship between pain and NPI total score on different Qualidem Subscales and NPS groups. Agitation and depression were both significantly related to almost all QoL subdomains. Agitation was related to lower scores on the sub- Unadjusted model Adjusted model domains “relationship”, “positive affect”, “restless tense behavior” and “social B (95% CI) P B (95% CI) P relations”, whereas depression was related to lower scores on the subdomains Care relationship “positive affect”, “negative affect”, “restless tense behavior” and “social rela- Mild vs. no pain 0.084 (-0.139 to 0.307) 0.461 0.107 (-0.118 to 0.332) 0.352 tions”. After agitation and depression, we found that psychomotor agitation and Moderate-severe -0.120 (-0.34 to 0.107) 0.301 -0.106 (-0.341 to 0.129) 0.377 apathy were both associated with 2 QoL subdomains. Psychomotor agitation was vs. no pain associated with lower scores on the subdomain “restless tense behavior” and NPI total score -0.028 (-0.034 to -0.022)*** < 0.001 -0.029 (-0.035 to -0.023)*** < 0.001 higher scores on the subdomain “social relations”, whereas apathy was associ- Positive affect ated with lower scores on the subdomains “positive affect” and “social relations”. There seems to be more, instead of less, social contact when residents showed Mild vs. no pain 0.119 (-0.090 to 0.329) 0.267 0.200 (-0.005 to 0.406) 0.057 psychomotor agitation. Moderate-severe -0.108 (-0.322 to 0.106) 0.322 -0.099 (-0.315 to 0.012) 0.368 vs. no pain NPI total score -0.015 (-0.021 to -0.009)*** < 0.001 -0.016 (-0.022 to -0.010)*** < 0.001 Negative affect Mild vs. no pain 0.001 (-0.242 to 0.244) N/A 0.060 (-0.177 to 0.297) 0.620 Moderate-severe -0.279 (-0.525 to -0.032)* -0.240 (-0.487 to 0.007) 0.057 vs. no pain 0.027 NPI total score -0.023 (-0.029 to -0.017)*** < 0.001 -0.021 (-0.027 to -0.015)*** < 0.001 Restless tense behaviour Mild vs. no pain 0.119 (-0.147 to 0.386) 0.382 0.175 (-0.091 to 0.441) 0.198 Moderate-severe -0.103 (-0.375 to 0.169) 0.459 -0.105 (-0.383 to 0.173) 0.460 vs. no pain NPI total score -0.029 (-0.037 to -0.021)*** < 0.001 -0.029 (-0.037 to -0.021)*** < 0.001 Social relations Mild vs. no pain 0.024 (-0.213 to 0.261) 0.133 (-0.094 to 0.360) 0.843 0.252 Moderate-severe 0.066 (-0.177 to 0.309) 0.139 (-0.090 to 0.376) vs. no pain 0.595 0.251 NPI total score -0.012 (-0.018 to -0.006)*** < 0.001 -0.01 (-0.016 to -0.004) < 0.001

Unadjusted analyses with pain [no pain (reference), mild pain, moderate-severe pain] and NPI total score as independent variables and the Qualidem subscales as dependent variables; adjusted analyses with potential confounders: age, sex, activities of daily living dependency, comorbidity, psychotropic drug use, dementia subtype, and dementia severity. CI indicates confidence interval; N/A: not available; NPI, Neuropsychiatric Inventory. *P < 0.05,** P < 0.01, ***P < 0.001.

146 | PAIN, NEUROPSYCHIATRIC SYMPTOMS, AND QUALITY OF LIFE OF NURSING HOME RESIDENTS WITH ADVANCED CHAPTER 7 | 147 DEMENTIA IN THE NETHERLANDS: A CROSS-SECTIONAL STUDY Table 4. Table 4 CONTINUED. Relationship between QoL, pain and NPI Factor Scores Relationship between QoL, pain and NPI Factor Scores

Unadjusted model Adjusted model Unadjusted model Adjusted model B (95% CI) P B (95% CI) P B (95% CI) P B (95% CI) P Care relationship Social relations Mild vs. no pain -0.052 (-0.281 to 0.177) 0.663 0.048 (-0.172 to 0.268) 0.668 Mild vs. no pain 0.043 (-0.151 to 0.237) 0.664 0.06 (-0.130 to 0.254) 0.527 Moderate-severe 0.124 (-0.109 to 0.357) 0.170 0.166 (-0.061 to 0.393) 0.152 Moderate-severe vs. no pain -0.173 (-0.369 to 0.023) 0.086 -0.164 (-0.362 to 0.034) 0.104 vs. no pain Agitation -0.021 (-0.035 to -0.007)** 0.003 -0.021 (-0.033 to -0.009)*** < 0.001 Agitation -0.070 (-0.082 to -0.058)*** < 0.001 -0.071 (-0.083 to -0.059)*** < 0.001 Depression -0.024 (-0.046 to -0.002)* 0.029 -0.024 (-0.046 to -0.002)* 0.029 Depression 0.000 (-0.018 to 0.018) 1.000 0.003 (-0.017 to 0.023) 0.794 Psychosis 0.008 (-0.018 to 0.033) 0.538 0.003 (-0.021 to 0.026) 0.803 Psychosis -0.005 (-0.027 to 0.017) 0.649 -0.006 (-0.028 to 0.016) 0.585 Psychomotor 0.024 (0.002 to 0.046)* 0.029 0.031 (0.009 to 0.053)** 0.005 Psychomotor 0.017 (-0.001 to 0.035) 0.059 0.021 (0.003 to 0.039)* 0.020 Apathy -0.045 (-0.067 to -0.023)*** < 0.001 -0.034 (-0.055 to -0.012)** 0.002 Apathy -0.008 (-0.028 to 0.012) 0.424 -0.014 (-0.034 to 0.006) 0.162 Positive affect Unadjusted analyses with pain [no pain (reference), mild pain, moderate-severe pain] and NPI total score as independent variables and the Qualidem subscales as dependent variables; adjusted analyses with potential Mild vs. no pain 0.048 (-0.158 to 0.254) 0.648 0.122 (-0.080 to 0.324) 0.236 confounders: age, sex, activities of daily living dependency, comorbidity, psychotropic drug use, dementia subtype, Moderate-severe and dementia severity. -0.076 (-0.287 to 0.129) 0.456 -0.082 (-0.290 to 0.126) 0.439 vs. no pain CI indicates confidence interval; NPI, Neuropsychiatric Inventory; QoL, quality of life. Agitation -0.025 (-0.038 to -0.013)*** < 0.001 -0.025 (-0.037 to -0.013)*** < 0.001 * P < 0.05, ** P < 0.01, ***P < 0.001. Depression -0.036 (-0.056 to -0.016)*** < 0.001 -0.034 (-0.054 to -0.014)*** < 0.001 Psychosis 0.011 (-0.011 to 0.033) 0.317 0.002 (-0.020 to 0.024) 0.856 Psychomotor 0.014 (-0.006 to 0.034) 0.162 0.018 (-0.002 to 0.038) 0.072 Apathy -0.027 (-0.047 to -0.007)** 0.007 -0.029 (-0.051 to -0.007)** 0.008 Discussion Negative affect Mild vs. no pain -0.114 (-0.322 to 0.094) 0.282 -0.055 (-0.261 to 0.151) 0.600 This study investigated the relationship of pain and neuropsychiatric symptoms Moderate-severe -0.229 (-0.439 to -0.019)* 0.032 -0.185 (-0.397 to 0.027) 0.087 (NPS) to the quality of life (QoL) in nursing home residents with dementia. The re- vs. no pain sults of this study show that the relationship between pain, NPS and QoL as mea- Agitation 0.005 (-0.007 to 0.017) 0.405 0.006 (-0.006 to 0.018) 0.317 sured with the Qualidem was not straightforward in this cohort. The presence Depression -0.102 (-0.122 to -0.082)*** < 0.001 -0.094 (-0.114 to -0.074)*** < 0.001 of NPS, mainly agitation and depressive symptoms, was significantly associated Psychosis -0.039 (-0.063 to -0.015)** 0.001 -0.045 (-0.069 to -0.021)*** < 0.001 with lower scores on all QoL subdomains, whereas pain was not associated with Psychomotor -0.004 (-0.024 to 0.016) 0.639 0.000 (-0.020 to 0.020) 1.000 any of the QoL subdomains. Apathy -0.011 (-0.031 to 0.009) 0.271 -0.006 (-0.028 to 0.016) 0.585 Restless Tens Behaviour The association between the presence of NPS and lower score on all QoL sub- Mild vs. no pain 0.163 (-0.092 to 0.418) 0.210 0.212 (-0.043 to 0.467) 0.103 domains has been shown in previous studies.4,11,35 These studies showed that Moderate-severe -0.143 (-0.400 to 0.114) 0.275 -0.135 (-0.398 to 0.128) 0.314 vs. no pain almost 50% of QoL was related to the presence of NPS and that this was the Agitation -0.018 (-0.032 to -0.004)* 0.010 -0.018 (-0.003 to -0.004) * 0.010 most consistent factor associated with QoL in people with dementia, although these researchers had not controlled for pain. In addition to previous studies, we Depression -0.054 (-0.078 to -0.030)*** < 0.001 -0.050 (-0.075 to -0.025)*** < 0.001 Psychosis -0.005 (-0.032 to 0.022) 0.721 -0.011 (-0.038 to 0.016) 0.432 performed analyses that included pain and NPI factor scores and we found that agitation, depressive symptoms and, to a more limited extent, apathy and psy- Psychomotor -0.067 (-0.091 to -0.043)*** < 0.001 -0.066 (-0.089 to -0.042)*** < 0.001 chomotor agitation were related to alterations in QoL subdomain scores. These Apathy 0.002 (-0.023 to 0.027) 0.878 0.011 (-0.014 to 0.036) 0.397 alterations seem in line with earlier research and the gained knowledge from us-

148 | PAIN, NEUROPSYCHIATRIC SYMPTOMS, AND QUALITY OF LIFE OF NURSING HOME RESIDENTS WITH ADVANCED CHAPTER 7 | 149 DEMENTIA IN THE NETHERLANDS: A CROSS-SECTIONAL STUDY ing NPI factor scores might benefit approaches aiming at improvement of QoL.36 between QoL and NPS. The presence of NPS might have negatively influenced Conversely, we did not find an independent association between pain and QoL QoL or the absence of NPS might have positively influenced QoL. Moreover, it re- subdomains using multiple linear regression, which suggests that pain is not veals the difficulties with the underlying concept of QoL and the measurement of linearly related to QoL in nursing home residents with dementia. This finding the construct of QoL. Some factors are part of the QoL construct, while they also is consistent with that of Beer et al.,4 who found that self-rated pain and NPI to- have been identified as predictor of QoL. tal score remained independently associated to self-rated QoL in their multiple Besides, we cannot exclude the possibility of selection bias. Two third of the in- regression model, but that observed pain was not independently associated to vited residents did not agree to participate or did not respond to the invitation. staff measured QoL. Differences in assessment instruments might account for Despite this limitation, the residents’ characteristics of our sample are compa- the reported variation in contributing factors associated with QoL. In fact, severe rable to other studies that included Dutch nursing home residents.37,38 Finally, as dementia requires other assessment instruments. As the majority of our sample testing of multiple hypotheses, without adjusting the individual significance level had severe to very severe dementia, we used dementia specific measurement in- of 0.05, may increase the probability that 1 or more of the findings are spurious, struments; therefore, our findings might strengthen that of Beer and colleagues, we presented the absolute P-values. indicating no linearly relation between pain and subdomains of Qualidem in nurs- ing home residents with dementia. This study showed that pain and NPS did not contribute to QoL in the same way, and, therefore, suggests a relationship of different nature. For example, pain may This study makes several noteworthy contributions to the field of dementia spe- exacerbate NPS, moreover, the relation may be better explained by a mediation cific research. Firstly, it investigated the relationships of pain and NPS to various model with pain as a mediator variable between NPS and QoL. This might also be subdomains of QoL, while adjusting for other known confounding variables. Sec- a plausible explanation for the results of the study by Husebo et al.,39 who found ondly, it showed that the relationship between pain, NPS and QoL, as measured that systematic pain treatment could be effective in treating agitation in nursing with Qualidem, is probably not straightforward. The strengths of this study are the home residents with dementia. Gathering these results with our results implies use of validated dementia specific measurement instruments to measure QoL, that pain treatment has beneficial effect on QoL via NPS. However, further longi- pain and NPS, the use of multiple regression analyses including pain categories tudinal research is needed to investigate whether a mediation model is a better instead of dichotomized pain reports, and the additional analyses including NPI approach to describe the relation between pain, NPS and QoL in nursing home factor scores instead of NPI total score. residents with dementia.

This study also has some limitations. As we used cross-sectional data, we could In conclusion, the relationship between pain, NPS and QoL, as measured with not draw conclusions on causality. It is therefore theoretically possible that a low- the Qualidem, seems not straightforward. Regardless of pain, differences in QoL er QoL may result in more NPS instead of the opposite. Although self-report is subdomains appear to be mostly explained by the NPS. In particular, agitation the preferred method of assessing QoL, as well as pain, this is often not possi- and depressive symptoms were independently related to lower QoL subdomain ble in nursing home residents with advanced dementia. Different observational scores. Our results should be considered hypothesis generating and further lon- methods are available to assess pain and QoL in people with advanced demen- gitudinal research is needed to investigate the exact nature of the relationship tia; however, observational methods always have the possibility of observer bias. between pain, NPS and QoL. Qualidem is a recommended questionnaire to evaluate important domains of QoL in residents with severe dementia, but for almost 20% of our sample the use of Qualidem was restricted by the need to omit items for those with very se- vere dementia. Further, some subscales were expected to have a low reliability and therefore were left out for analyses. As Qualidem is based on recall by the proxy-rater, there is a possibility of recall bias. Notably, there might be overlap between the Qualidem and NPI questionnaires, for example, the NPI question- naire asks “Does the resident seem depressed?” and the Qualidem consists of items like “is sad” or “cries”. This overlap might have influenced the relationship

150 | PAIN, NEUROPSYCHIATRIC SYMPTOMS, AND QUALITY OF LIFE OF NURSING HOME RESIDENTS WITH ADVANCED CHAPTER 7 | 151 DEMENTIA IN THE NETHERLANDS: A CROSS-SECTIONAL STUDY References of Norwegian nursing homes. Pain Manag Nurs. 2010;11:35-44. 15. Cipher DJ, Clifford PA. Dementia, pain, depression, behavioral disturbanc- 1. Prince M, Bryce R, Albanese E, et al. The global prevalence of dementia: A es, and ADLs: toward a comprehensive conceptualization of quality of life in systematic review and metaanalysis. Alzheimers Dement. 2013;9:63-75. long-term care. Int J Geriatr Psychiatry. 2004;19:741-748. 2. Torke AM. Building the evidence base for palliative care and dementia. Palli- 16. Hodgson N, Gitlin LN, Huang J. The influence of sleep disruption and pain at Med. 2014;28:195-196. perception on indicators of quality of life in individuals living with dementia at 3. Haberstroh J, Hampel H, Pantel J. Optimal management of Alzheimer’s home. Geriatr Nurs. 2014;35:394-398. disease patients: Clinical guidelines and family advice. Neuropsychiatr Dis 17. Zuidema S, Koopmans R, Verhey F. Prevalence and predictors of neuropsy- Treat. 2010;6:243-253. chiatric symptoms in cognitively impaired nursing home patients. J Geriatr 4. Beer C, Flicker L, Horner B et al. Factors associated with self and informant Psychiatry Neurol. 2007;20:41-49. ratings of the quality of life of people with dementia living in care facilities: a 18. Corbett A, Husebo B, Malcangio M, et al. Assessment and treatment of pain cross sectional study. PLoS One. 2010;5:e15621. in people with dementia. Nat Rev Neurol. 2012;8:264-274. 5. Sloane PD, Zimmerman S, Williams CS, et al. Evaluating the quality of life 19. Lyketsos CG, Colenda CC, Beck C, et al. Position statement of the American of long-term care residents with dementia. Gerontologist. 2005;45 Spec No Association for Geriatric Psychiatry regarding principles of care for patients 1:37-49. with dementia resulting from Alzheimer disease. Am J Geriatr Psychiatry. 6. The World Health Organization Quality of Life assessment (WHOQOL). po- 2006;14:561-572. sition paper from the World Health Organization. Soc Sci Med. 1995;41:1403- 20. Cohen-Mansfield J, Werner P.Longitudinal predictors of non-aggressive ag- 1409. itated behaviors in the elderly. Int J Geriatr Psychiatry. 1999;14:831-844. 7. Gerritsen DL, Steverink N, Ooms ME, et al. Finding a useful conceptual basis 21. Bair MJ, Robinson RL, Katon W, et al. Depression and pain comorbidity: a for enhancing the quality of life of nursing home residents. Qual Life Res. literature review. Arch Intern Med. 2003;163:2433-2445. 2004;13:611-624. 22. van Kooten J, Delwel S, Binnekade TT, et al. Pain in dementia: prevalence 8. Pietersma S, de Vries M, van den Akker-van Marle ME. Domains of quality and associated factors: protocol of a multidisciplinary study. BMC Geriatr. of life: results of a three-stage Delphi consensus procedure among patients, 2015;15:29. family of patients, clinicians, scientists and the general public. Qual Life Res. 23. Ettema TP, Droes RM, de Lange J, et al. QUALIDEM: development and evalu- 2014;23:1543-1556. ation of a dementia specific quality of life instrument--validation. Int J Geriatr 9. Ettema TP, Droes RM, de Lange J, et al. The concept of quality of life in de- Psychiatry. 2007;22:424-430. mentia in the different stages of the disease. Int Psychogeriatr. 2005;17:353- 24. Ettema TP, Droes RM, de Lange J, et al. QUALIDEM: development and eval- 370. uation of a dementia specific quality of life instrument. Scalability, reliability 10. Barca ML, Engedal K, Laks J, et al. Quality of life among elderly patients with and internal structure. Int J Geriatr Psychiatry. 2007;22:549-556. dementia in institutions. Dement Geriatr Cogn Disord. 2011;31:435-442. 25. Bouman AI, Ettema TP, Wetzels RB, et al. Evaluation of Qualidem: a demen- 11. Wetzels RB, Zuidema SU, de Jonghe JF, et al. Determinants of quality of tia-specific quality of life instrument for persons with dementia in residential life in nursing home residents with dementia. Dement Geriatr Cogn Disord. settings; scalability and reliability of subscales in four Dutch field surveys. Int 2010;29:189-197. J Geriatr Psychiatry. 2011;26:711-722. 12. Beerens HC, Zwakhalen SM, Verbeek H, et al. Factors associated with qual- 26. Smit D, de Lange J, Willemse B, et al. Activity involvement and quality of life ity of life of people with dementia in long-term care facilities: a systematic of people at different Stages of dementia in long term care facilities. Aging review. Int J Nurs Stud. 2013;50:1259-1270. Ment Health. 2016;20:100-109. 13. Banerjee S, Samsi K, Petrie CD, et al. What do we know about quality of life 27. Husebo BS, Strand LI, Moe-Nilssen R, et al. Pain in older persons with se- in dementia? A review of the emerging evidence on the predictive and explan- vere dementia. Psychometric properties of the Mobilization-Observation-Be- atory value of disease specific measures of health related quality of life in haviour-Intensity-Dementia (MOBID-2) Pain Scale in a clinical setting. Scand people with dementia. Int J Geriatr Psychiatry. 2009;24:15-24. J Caring Sci. 2010;24:380-391. 14. Torvik K, Kaasa S, Kirkevold O, et al. Pain and quality of life among residents 28. Husebo BS, Strand LI, Moe-Nilssen R, et al. Mobilization-Observation-Be-

152 | PAIN, NEUROPSYCHIATRIC SYMPTOMS, AND QUALITY OF LIFE OF NURSING HOME RESIDENTS WITH ADVANCED CHAPTER 7 | 153 DEMENTIA IN THE NETHERLANDS: A CROSS-SECTIONAL STUDY havior-Intensity-Dementia Pain Scale (MOBID): development and validation of a nurse-administered pain assessment tool for use in dementia. J Pain Symptom Manage. 2007;34:67-80. 29. Kat MG, de Jonghe JF, Aalten P, et al. [Neuropsychiatric symptoms of de- mentia: psychometric aspects of the Dutch Neuropsychiatric Inventory (NPI)]. Tijdschr Gerontol Geriatr. 2002;33:150-155. 30. Wood S, Cummings JL, Hsu MA, et al. The use of the neuropsychiatric inven- tory in nursing home residents. Characterization and measurement. Am J Geriatr Psychiatry. 2000;8:75-83. 31. Katz S. Assessing self-maintenance: activities of daily living, mobility, and instrumental activities of daily living. J Am Geriatr Soc. 1983;31:721-727. 32. Charlson ME, Pompei P, Ales KL, et al. A new method of classifying prognos- tic comorbidity in longitudinal studies: development and validation. J Chronic Dis. 1987;40:373-383. 33. Reisberg B, Ferris SH, de Leon MJ, et al. The Global Deterioration Scale for assessment of primary degenerative dementia. Am J Psychiatry. 1982;139:1136-1139. 34. Twisk JWR. Applied multilevel analysis. Cambridge, England: Cambridge University Press; 2006. 35. van de Ven-Vakhteeva J, Bor HF, Wetzels RB, et al. The impact of antipsy- chotics and neuropsychiatric symptoms on the quality of life of people with dementia living in nursing homes. Int J Geriatr Psychiatry 2013;28:530-8.. 36. Samus QM, Rosenblatt A, Steele C, et al. The association of neuropsychiatric symptoms and environment with quality of life in assisted living residents with dementia. Gerontologist. 2005;45 Spec No 1:19-26. 37. Pieper MJ, Francke AL, van der Steen JT et al. Effects of a stepwise multi- disciplinary intervention for challenging behavior in advanced dementia: a cluster randomized controlled trial. J Am Geriatr Soc. 2016;64:261-269. 38. Zwijsen SA, Smalbrugge M, Eefsting JA et al. Coming to grips with chal- lenging behavior: a cluster randomized controlled trial on the effects of a multidisciplinary care program for challenging behavior in dementia. J Am Med Dir Assoc. 2014;15:531-10. 39. Husebo BS, Ballard C, Sandvik R, et al. Efficacy of treating pain to reduce be- havioural disturbances in residents of nursing homes with dementia: cluster randomised clinical trial. BMJ. 2011;343:d4065.

154 | PAIN, NEUROPSYCHIATRIC SYMPTOMS, AND QUALITY OF LIFE OF NURSING HOME RESIDENTS WITH ADVANCED CHAPTER 7 | 155 DEMENTIA IN THE NETHERLANDS: A CROSS-SECTIONAL STUDY 8 General discussion 8 General discussion scales), pain type, pain medication, dementia subtypes, dementia severity and demographic features. We found that the prevalence of observed pain was 43% (95% CI = 36-50%) using the MOBID-2 Pain Scale. Half of the nursing home residents with observed pain had mild pain. Pain assessment with the MOBID-2 This chapter summarizes and discusses the main findings of this thesis. It also Pain Scale showed no difference in pain between dementia subtypes, but addresses the most important methodological issues and implications for residents with more severe dementia experienced pain more often than those clinical practice. Finally, the implications for further research are discussed. with less severe dementia (27% vs. 15%). About one third of the nursing home residents with observed pain suffered from moderate to severe pain, despite treatment with regularly scheduled analgesics. The prevalence of self-reported Summary of the main findings pain was significantly higher in nursing home residents with VaD (53%) compared to those with AD (18%) and other dementia subtypes (14%). Nociceptive pain was Review studies the predominant type of pain (70%), followed by mixed pain (neuropathic and We performed two systematic reviews (chapters 2 and 3). The first systematic nociceptive pain) (25%). Acetaminophen (paracetamol) was the most commonly review (chapter 2) reports on pain experience in dementia subtypes. We identified prescribed analgesic (80%). twelve studies that addressed pain experience in dementia subtypes. For Alzheimer’s Dementia (AD), studies on clinical pain indicated a reduced pain The nursing home residents with observed pain, as assessed with the MOBID-2 experience compared to controls without dementia, whereas the findings of Pain Scale, were included in a small prospective exploratory study (chapter 6). experimental studies were inconsistent. For vascular dementia (VaD), clinical Following the pain assessment, we communicated who of participating nursing studies found that primary caregivers rated pain is equal to rating of pain by home residents were identified as being in pain to the attending physician. primary caregivers in controls without dementia, although primary caregivers We also included a non-binding treatment recommendation based on the Dutch reported more painful locations in people with VaD. During self-report, guideline “Recognition and management of chronic pain in vulnerable elderly” people with VaD reported higher pain levels than controls without dementia. (Verenso)1. After three months, reassessment of pain took place and we found For frontotemporal dementia (FTD), experimental studies found a reduced pain that the pain intensity was significantly reduced (p < 0.001). The proportion of experience compared to controls without dementia. We found no studies on pain nursing residents with persistent pain was 58%. The use of analgesic drugs did experience in dementia with Lewy bodies (DLB). not change.

The second systematic review (chapter 3) reports the prevalence of pain in dementia Furthermore, we collected data concerning quality of life and the presence of subtypes. The number of studies reporting the prevalence of pain per dementia neuropsychiatric symptoms in the participating nursing home residents (chapter 7). subtype was limited to ten studies, and these studies showed no significant Both pain and neuropsychiatric symptoms have been associated with quality of difference in prevalence of pain between the dementia subtypes. This review life in people with dementia, but there were no studies investigating the relation confirms that pain is frequently reported in people with dementia, and shows that between quality of life and neuropsychiatric symptoms together with pain. the majority of people with dementia do not suffer from high pain intensities. We analyzed the relationship between pain, neuropsychiatric symptoms and quality of life using regression models. We found that there was no independent relationship between pain and quality of life. However, neuropsychiatric Clinical studies symptoms, in particular agitation and depression, were significantly associated with a lower quality of life. In the clinical section of this thesis, chapter 5 reports the results from an observational cross-sectional cohort study in nursing home residents in the Netherlands. We included 199 nursing home residents with dementia living at dementia special care units of 10 nursing homes. We collected data on presence of pain (by observation: MOBID-2 Pain Scale and PAINAD, and by self-report

158 | GENERAL DISCUSSION CHAPTER 8 | 159 Methodological reflections on the clinical studies 700 people. We received informed consent for one third of the nursing home residents. To estimate the possible influence of sampling bias, we compared the This section considers the methodological strengths and limitations that should demographic features of our sample with other samples of nursing home be taken into account when interpreting the results described in the clinical residents with dementia in other Dutch health care organizations. We found that section of this thesis. The design of the study and the study sample, as well as our sample had the same age distribution as the samples of Pieper et al.(2017)3 the measurements and factors related to the data-collection are discussed. and Zwijsen et al.(2014),4 who both performed studies that included a representative sample of nursing home residents with dementia. Moreover, the Study design distribution of dementia subtypes was corresponding with the distribution found The results described in the clinical section of this thesis were based on data in the study of Zwijsen et al.,4 and the prevalence of pain was similar to the collected within the “PAINdemiA” study, a cross-sectional, and partly longitudinal prevalence of observed and self-reported pain found in the study of Pieper et al.3 cohort study that collected data on the prevalence of pain in people with different Therefore, we consider the influence of sampling bias to be low. dementia subtypes and in various dementia stages. In this multidisciplinary study a neuropsychologist, dentist and elderly care medicine trainee collected data in the same group of participants. In the clinical section of this thesis we Measurements used data collected from people with dementia that live in nursing homes. Pain measurements A strength of this study is that the data collection incorporated dementia Our study is unique in that we investigated the prevalence of both observed pain subtypes, which enables the investigation of pain prevalence in people with and self-reported pain in nursing home residents with dementia. dementia of different subtypes. Pain assessment is dependent upon situation, context and instrument, i.e. patient A cross-sectional design was chosen for the “PAINdemiA” study, because the care or clinical research, acute or chronic pain, neuropathic or non-neuropathic primary aim was to collect data on the prevalence of pain. However, the cross- pain, patient with or without cognitive impairment.5, 6 Many pain assessment sectional design inhibits inference concerning causality or direction of the instruments exist and for adequate pain assessment it is necessary to choose association between pain, neuropsychiatric symptoms and quality of life (chapter the right tool for the job. For research purposes it is fundamental to select 7). Next to the cross-sectional data, we also collected longitudinal data. measurement instruments that are reliable, valid, and sensitive to change.5 We collected data on the course of pain and pain management strategies three Self-report is the gold standard to measure pain intensity, although in severe months after the first assessment, hypothesizing that audit and feedback dementia this may not be possible. Therefore, we chose assessment instruments (i.e. communicating the assessment results to the attending physician including that have been developed or standardized in older adults with dementia, have a guideline-based treatment recommendation) would eventually lead to a lower been proven time efficient, and do not pose an excessive burden on the resident. pain intensity.2 We only collected longitudinal data of nursing home residents who were assessed as being in pain during the first pain assessment. With this As accurate measurement of pain intensity was necessary, we identified pain by selection we introduced potential for pain intensity to show regression toward using two independent instruments.7 We chose to use the MOBID-2 Pain Scale, the mean. a dementia-specific observational pain scale, and to combine this scale with another instrument that was adjusted to the communications skills of the Sample resident (i.e. a self-report scale or the PAINAD).5-7 Nursing home organizations were approached to participate in our study via UNO-VUmc, a network of nursing homes affiliated with VUmc for research, The MOBID-2 Pain Scale assesses pain intensity inferred by observations of pain developing and sharing of practice innovations and education/training of nursing behaviours during standardized, guided movement with focus on the home professionals. We included ten nursing homes from three health care musculoskeletal system, and previous observations of other pain behaviours organizations delivering specialized care to residents with dementia in the over the last days. It combines pain originating from musculoskeletal conditions northwestern part of the Netherlands. Our source population consisted of almost and pain caused by other conditions such as pressure ulcers or symptoms from

160 | GENERAL DISCUSSION CHAPTER 8 | 161 internal organs. Studies on the psychometric properties of the MOBID-2 have because the researchers involved in this study visited the nursing homes to see shown good reliability, validity, and feasibility in clinical practice.8-10 all participants and to collect data. When data was missing (e.g. pain observations) the researchers were able to deal with it immediately and most of We collapsed the MOBID-2 pain intensity score (0-10) into three categories: the time they were able to retrieve the missing data. no pain (0-2), mild pain (3-4), or moderate to severe pain (≥ 5). This may have resulted in a loss of information. Indeed, the total score distribution was right- Other measurements skewed. However, we consider the categorical response easier to interpret, Data concerning age, gender, dementia subtype diagnosis, medical history and and it increases the usefulness of our findings for clinical practice. medical treatment information was retracted from the medical records. Data in medical records are originally collected for other purposes than research. The PAINAD is a frequently used dementia-specific pain scale and measures pain Therefore, retrospective review of medical records might be limited by incomplete by direct observations. It assesses five behaviours related to pain, and each of documentation, difficulty interpreting information found in the documents, these is rated on a three point scale (0-2).11 The reliability and validity of the PAINAD difficulty establishing cause and effect, and variation in the quality of information have been studied extensively, and they are well established.12 However, we found recorded by medical professionals. Nevertheless, we found in more than 80% of the items of the PAINAD showing substantial overlap with items of (dis)comfort the participants a dementia subtype diagnosis and for most people we could tools. This might be one of the explanations for the difference in pain rates between determine the pain type by combining the finding of the physical examination the MOBID-2 and PAINAD. with the information found in the medical record. For example pain behaviour during examinations of the legs and an old fracture in the medical history. We defined being able to use a self-report pain scale on the basis of dementia severity and the performance of the resident at the day of the data collection. Data on dementia severity, Activities of daily living (ADL), neuropsychiatric All residents that were able to use a self-report scale had a GDS (Global symptoms and quality of life was collected using questionnaires. Dementia severity Deterioration Scale) score < 7. was assessed using the GDS,13 ADL were measured using the modified version of Katz index of independence in ADL (Katz-ADL),14 neuropsychiatric symptoms were We found that the pain prevalence rates were significantly different between the measured using the NPI,15 and quality of life was measured with the Qualidem.16, 17 different pain measurement instruments (range: 22%-43%). This could be explained by the difference in assessment techniques. The pain assessment with a These questionnaires were administered via face to face interviews with the self-report pain scale scores actual pain at the time of asking, while the MOBID-2 nursing staff. We chose for interviews to have the possibility to explain the Pain Scale combines provoked pain observations with more chronic pain conditions questionnaires as well as to decrease the possibility of missing data. As the (during the last 3-4 days), and the PAINAD items measures pain more indirectly. information obtained in the interviews was based on recall by the proxy-rater, there is a possibility of recall bias. However, the maximum recall period was two Another explanation for the differences found might be the time of assessment. weeks and a nurse was responsible for approximately seven residents, so the In our study, the PAINAD observations and assessment via self-report were risk of recall bias is limited. mostly performed at rest, which contrasts with half of the items of the MOBID-2 Pain Scale that were performed during morning care. Reflections on the findings Pain behaviour can only be assessed by people who know the residents and who are able to observe the resident for a prolonged period of time. Therefore, Prevalence of pain in dementia subtypes nursing staff were responsible for the pain observations. All nursing home In the field of dementia research different measurement instruments are used to organizations were visited to instruct the nursing staff in an approximately one assess pain in dementia.6 These instruments address different constructs; hour training session about administering the MOBID-2 Pain Scale and the observational pain scales are based on pain behaviours observed by a nurse or PAINAD, shortly before the start of the data collection. other caregiver, and self-report scales are based on pain experience reported by The amount of missing data (e.g. a missing pain observation) was relatively low, the persons themselves.18 However, in literature these constructs (i.e. observed

162 | GENERAL DISCUSSION CHAPTER 8 | 163 pain and self-reported pain) have been used interchangeably. For this reason, it Another explanation might be (cardiovascular) comorbidity and the presence of is difficult to compare and to interpret earlier findings concerning pain prevalence its consequences that are known to cause pain, for example, hemiplegia and in dementia. Therefore, we presented the prevalence of observed pain and self- shoulder dysfunction.23 In our study we actually found these conditions in people reported pain separately in the clinical section of this thesis. with VaD during physical examination.

We found that the prevalence of observed pain in all participating nursing home Another finding of this thesis is that observed pain according to the MOBID-2 residents, as measured with the MOBID-2 Pain Scale, was 43% and in half of Pain Scale was reported more frequently in residents with a GDS score of 7 (27%) these residents this was pain of low intensity. This finding can be compared to than in residents with a GDS score of < 7. Moreover, a GDS score of 7 was those of other recent studies performed in nursing home residents with significantly associated with moderate to severe pain. Further, all residents with dementia, for example, with the results of Erdal et al.(2017),19 who also used a GDS score of 7 suffered from a certain level of paratonia. Paratonia is a form of MOBID-2 Pain Scale and found observed pain to be present in 40% of the nursing hypertonia that is present in 90-100% of people in the advanced stages of home residents with dementia. dementia,24,25 and is accompanied by pain, a decline in mobility, and a reduced quality of life. We found that pain commonly originated from the musculoskeletal We concluded that the majority (57%) of the nursing home residents with system and during physical examination we found a considerable proportion of dementia had no pain. The prevalence of pain is slightly lower than we found in residents suffering from contractures or paratonia (n = 15). The presence of pain our systematic literature review. However, one third of the nursing home in people with a GDS score of 7 and moderate to severe pain (n = 12), as seen in residents with clinically relevant pain had moderate to severe pain regardless of our study, is an important signal and might be regarded as a trigger to evaluate using pain medication. This is in line with the findings of Hendriks et al.(2015)20 current pain management in order to improve palliative care in these persons who found that pain was persistent in nursing home residents with dementia and with dementia. symptom management only intensified at the end of life. We might conclude that pain is not relieved adequately in Dutch nursing home resident with dementia. This is surprising as Dutch nursing home care is characterized by the presence Current state of pain management in nursing homes of an elderly care physician. Dutch elderly care physicians follow a vocational training that includes training in palliative care, and therefore, they should be In our study, we found that 43% of all nursing home residents used analgesics on aware of the presence of burdensome symptoms as pain. On the other hand, a a daily base, irrespective of pain presence. However, 40% of the residents with a treatment recommendation of an elderly care physician also depends on the MOBID-2 pain score of ≥ 3 did not use analgesics on a daily base, whereas one- observation of pain by the nursing staff. More focus should be on the evaluation third of the residents with pain that used analgesic on a daily base still suffered of pain by both elderly care physician and nursing staff. from moderate to severe pain. This suggests that pain management is suboptimal, and that there may be room for improvement. As the number of people with dementia with Lewy bodies (DLB) and frontotemporal dementia (FTD) was small, firm conclusions on the prevalence of Overall, acetaminophen (paracetamol) was the most prescribed analgesic (80%) observed pain could not be drawn for people with DLB and FTD. However, the in both people with and without pain. This is in line with the finding of Hendriks et small number of people with DLB and FTD in our study reflects the number of al.(2015),20 who also found that acetaminophen was frequently provided in people with these dementia subtypes in clinical practice. In the subpopulation of nursing home residents. The use of stronger analgesics, such as opioids, people who were able to self-report pain we found that the prevalence of self- was only seen in nursing home residents with pain, with the highest percentage reported pain was significantly higher in people with vascular dementia (VaD). in those with moderate to severe pain (25%). Comparing our results with the This is of importance since it suggests that there is, at least in those who are current literature, we found a percentage comparable to Pieper et al.(2017)3 and able to self-report, a difference in experienced pain between people with Sandvik et al.(2016).26 Sandvik et al. suggests a shift from undertreatment of pain Alzheimer’s dementia (AD) and VaD.21 In literature this difference in pain in people with dementia to an equal prescription rate of analgesics, including experience has been contributed to white matter lesions in VaD, which leads to opioids, in both people with and without dementia. We found that 13 residents altered pain processing and, therefore, to an increased pain experience.21,22 (32%) suffered from moderate to severe pain despite using opioids. This raises

164 | GENERAL DISCUSSION CHAPTER 8 | 165 the question whether opioids were administered in an appropriate dosage, or the Qualidem, the relation between quality of life, pain and neuropsychiatric that treatment could be improved by the use of adjuvant treatment with so-called symptoms might be better described by a different model. An example is a anti-neuropathic drugs. mediation model with pain as mediator variable between neuropsychiatric symptoms and quality of life. However, in a recent systematic review, van Neuropathic pain is frequently resistant to standard pain treatment,18 but its Dalen-Kok et al.(2015)32 concluded that the available evidence does not support detection is essential to optimize its management, especially in patients with a strong association between pain and neuropsychiatric symptoms. On the other dementia.27 We determined the type of pain and found that nociceptive pain was hand, systematic pain treatment has shown to be effective in treating agitation in the predominant type of pain (70%) followed by mixed pain (25%), i.e. a combination nursing home residents with dementia.33,34 This suggests that, at least in nursing of nociceptive pain and neuropathic pain.18,27 home residents, pain should be considered as a possible and treatable cause of agitation.18,35 In addition, recent studies also found observed pain associated with Nevertheless, pharmacological treatment with specific anti-neuropathic drugs depressive symptoms in people with dementia.19,35 Altogether, these findings was rare in our study population (n = 4). Future research into alternative show that there is a need for more multidimensional evaluation of burdensome treatment options such as treatment with anti-neuropathic drug and non- symptoms (e.g. pain, agitation, depression), including adequate assessment, pharmacological treatment options is needed in order to establish the most treatment, and evaluation. appropriate pain treatment for people with dementia.28

In our prospective exploratory study we communicated the results of the pain Recommendation for practice assessment to the attending elderly care physician with a treatment recommendation (n = 58). We found that the intensity of observed pain was Although pain in dementia has been topic of interest of many publications in the significantly reduced after three months, but the percentage of analgesic drugs last years, it is still a relevant topic nowadays. The results of this thesis have did not change. However, persistent pain was observed in 58% of the residents. several implications for future practice. Although we found half of our treatment recommendations being implemented, there was no difference in pain intensity or pain persistence between residents Pain management in Dutch nursing homes with or without annotations of the treatment recommendation in their records. Pain management appears to be suboptimal in nursing home residents with This finding suggests that communicating the assessment results together with dementia. Since assessment is a prerequisite for appropriate pain management, a treatment recommendation is not sufficient to improve pain management in implementation of pain assessment tools in daily practice is needed. To provide nursing homes. However, pain assessment itself has shown to be an effective adequate pain treatment, (elderly care) physicians should be more aware of the intervention to improve pain management,29 and as we found a decrease in pain pain type and it specific treatment regime. Moreover, they should consider more intensity this might still be an explanation for the reduced pain intensity after often the presence of neuropathic pain. Considering the evaluation of treatment three months. Moreover, our findings also suggest that pain management in effects, special attention should be paid to nursing home residents with relatively Dutch nursing homes is suboptimal and that there may be room for improvement. high MOBID-2 pain scores and daily prescribed analgesics. Especially in those with The overarching aim to pay attention to burdensome symptoms as pain is to a GDS score of 7, timely and meticulous evaluation of the effect is needed. For improve quality of life. In previous studies pain was found to be associated with a example, relieving pain and preventing an increase in pain in people with paratonia reduced quality of life in people with dementia.30 Contrary to previous studies, we by consulting an occupational therapist, or adjuvant therapy with anti-neuropathic found no independent relationship between pain and quality of life in our drugs in case of neuropathic pain features. More tailored approaches that are regression models. In a cross-sectional study about characteristics associated regularly adjusted to individual needs may help improve pain management, and, with quality of life Klapwijk et al.(2016)31 found that the presence of pain was therefore, palliative care in nursing home residents with dementia. associated with two subdomains of the Qualidem (p = 0.05) in multivariate During the data-collection we met several residents who did not have a wish regression analyses, whereas neuropsychiatric symptoms were associated to all for additional analgesics, because they experienced their own pain as bearable. subdomains of the Qualidem ( p < 0.01 to p = 0.04). As we also found that To act upon this wish might lead to insufficient treatment of pain, but this shows neuropsychiatric symptoms were significantly associated with all subdomains of above all that we should not rely solely on pain scores in determining pain

166 | GENERAL DISCUSSION CHAPTER 8 | 167 treatment; we need to communicate with the resident about their pain, and their Another important aspect of pain management is the timely evaluation of need for analgesics in order to provide appropriate pain treatment tailored to the treatment effects. Research should focus on what is the optimal way of evaluating resident himself. treatment effects, and how to specify the appropriate treatment tailored to individual needs of people with pain and dementia. For example, is the most appropriate treatment either non-pharmacological or pharmacological or Recommendations for future research a combination of the two, which type of pharmacological treatment is adequate, or how could we specify the optimal dosage of medication in terms of effect and Prevalence of pain in dementia subtypes side-effects? Moreover, the appropriateness of chronic use of analgesics in As the number of participants with frontotemporal dementia (FTD) and dementia people without pain during assessment should be topic of further research. with Lewy bodies (DLB) in our study was small, it was not possible to draw firm These research recommendations can all contribute to the development of conclusions about differences in the prevalence of pain in dementia subtypes. practical guidelines for the treatment of pain adjusted to the individual needs of Hence, it would be worthwhile to investigate the prevalence of pain and type of people with pain and dementia. pain in nursing home residents with FTD and DLB.

This thesis acknowledges that pain in dementia is complex. In order to facilitate Concluding remarks the interpretation of future study findings, further research should use the different constructs of observed pain and self-reported pain. Also attention Based on the findings of the “PAINdemiA” study we concluded that the prevalence should be paid to the constructs of ‘provoked pain’(i.e. acute pain) and chronic of observed pain was 43% and did not differ between dementia subtypes. For self- pain conditions in people with dementia. We consider provoked pain and chronic reported pain we found a trend toward more frequent experienced pain by people pain to be different clinical entities, and the treatment of these pain types differs with vascular dementia. And overall, we found that there is room for improvement from each other. Differentiation between these two pain types will contribute to of pain management in nursing home residents with dementia. more appropriate treatment of pain in people with dementia. For example, musculoskeletal pain that is only present during morning care (i.e. provoked Our study is unique in that we investigated the prevalence of both observed pain and pain) may be better treated with non-pharmacological treatment or short-acting self-reported pain in nursing home residents with dementia. Moreover, under­ painkillers instead of long-acting painkillers. standing the current state of pain management in nursing home residents with dementia is a prerequisite for improving pain management in nursing home Pain management residents with dementia. Further research should focus on how to best perform Research interest in the management of pain in people with dementia has evaluation of treatment, and how to specify the appropriate treatment tailored to increased in recent years.20,36,37 Besides better pain management in itself, there individual needs of nursing home residents with dementia. And, as pain is one of the is an even greater need for more multidimensional evaluation of discomfort or factors associated with quality of life in dementia, future research should focus on a burdensome symptoms. Especially in the light of aiming at preservation of multidimensional approach that includes pain management in nursing home quality of life, the focus should be more multidimensional instead of handling residents with dementia that suffer from burdensome symptoms, such as pain. This one piece of the puzzle at the time. It might be worthwhile to examine the will help to preserve, or even improve of quality of life in nursing home residents with possibility of providing an approach to preserve (or even improve) quality of life in dementia, and therefore, contribute to better palliative care for people with dementia. people with dementia, which can be applied in the presence of various burdensome symptoms. The Social Production Functions (SPF) theory can be used as a conceptual basis for the development of such approach.38,39 The SPF theory assumes that QoL is the result of the realization of physical wellbeing and social wellbeing. A certain amount of each is necessary, but there is also the possibility of substitution. For example, the loss of physical comfort (e.g. pain) can be partly substituted by activities that improve the level of social wellbeing.

168 | GENERAL DISCUSSION CHAPTER 8 | 169 References 13. Reisberg B, Ferris SH, de Leon MJ, Crook T. The Global Deterioration Scale for assessment of primary degenerative dementia. Am J Psychiatry. 1. Achterberg WP, de Ruiter CM, de Weerd-Spaetgens CM, Geels P, Horikx A, 1982;139: 1136-1139. Verduijn M. [Multidisciplinary guideline ‘Recognition and treatment of 14. Katz S. Assessing self-maintenance: activities of daily living, mobility, and chronic pain in vulnerable elderly people’](In Dutch). Ned Tijdschr Geneeskd. instrumental activities of daily living. J Am Geriatr Soc. 1983;31: 721-727. 2012;155: A4606. 15. Kat MG, de Jonghe JF, Aalten P, Kalisvaart CJ, Droes RM, Verhey FR. 2. van Riet Paap J, Vernooij-Dassen M, Sommerbakk R, et al. Implementation [Neuropsychiatric symptoms of dementia: psychometric aspects of the of improvement strategies in palliative care: an integrative review. Dutch Neuropsychiatric Inventory (NPI)](In Dutch). Tijdschr Gerontol Geriatr. Implement Sci. 2015;10: 103. 2002;33: 150-155. 3. Pieper MJ, van der Steen JT, Francke AL, Scherder EJ, Twisk JW, 16. Ettema TP, Droes RM, de Lange J, Mellenbergh GJ, Ribbe MW. Achterberg WP. Effects on pain of a stepwise multidisciplinary intervention QUALIDEM: development and evaluation of a dementia specific quality of life (STA OP!) that targets pain and behavior in advanced dementia: A cluster instrument--validation. Int J Geriatr Psychiatry. 2007;22: 424-430. randomized controlled trial. Palliat Med. 2017: 269216316689237. 17. Ettema TP, Droes RM, de Lange J, Mellenbergh GJ, Ribbe MW. 4. Zwijsen SA, Smalbrugge M, Eefsting JA, et al. Coming to grips with QUALIDEM: development and evaluation of a dementia specific quality of life challenging behavior: a cluster randomized controlled trial on the effects of instrument. Scalability, reliability and internal structure. Int J Geriatr a multidisciplinary care program for challenging behavior in dementia. J Am Psychiatry. 2007;22: 549-556. Med Dir Assoc. 2014;15: 531 e531-510. 18. AGS panel on persistent pain in older persons. The management of 5. Hadjistavropoulos T, Herr K, Turk DC, et al. An interdisciplinary expert persistent pain in older persons. J Am Geriatr Soc 2002;50:S205-24. consensus statement on assessment of pain in older persons. Clin J Pain. 19. Erdal A, Flo E, Selbaek G, et al. Associations between pain and depression 2007;23: S1-43. in nursing home patients at different stages of dementia. J Affect Disord. 6. Colloca L, Ludman T, Bouhassira D, et al. Neuropathic pain. Nature reviews 2017;218: 8-14. disease primers. 2017;3: 17002. 20. Hendriks SA, Smalbrugge M, Galindo-Garre F, Hertogh CM, van der Steen JT. 7. Corbett A, Husebo B, Malcangio M, et al. Assessment and treatment of pain From admission to death: prevalence and course of pain, agitation, and in people with dementia. Nat Rev Neurol. 2012;8: 264-274. shortness of breath, and treatment of these symptoms in nursing home 8. Husebo BS, Strand LI, Moe-Nilssen R, Husebo SB, Snow AL, Ljunggren AE. residents with dementia. J Am Med Dir Assoc. 2015;16(6):475-81. Mobilization-Observation-Behavior-Intensity-Dementia Pain Scale (MOBID): 21. Achterberg WP, Pieper MJ, van Dalen-Kok AH, et al. Pain management in development and validation of a nurse-administered pain assessment tool patients with dementia. Clin Interv Aging. 2013;8: 1471-1482. for use in dementia. J Pain Symptom Manage. 2007;34: 67-80. 22. Scherder EJ, Sergeant JA, Swaab DF. Pain processing in dementia and its 9. Husebo BS, Strand LI, Moe-Nilssen R, Husebo SB, Ljunggren AE. Pain in relation to neuropathology. Lancet Neurol. 2003;2: 677-686. older persons with severe dementia. Psychometric properties of the 23. van Almenkerk S, Depla MF, Smalbrugge M, Eefsting JA, Hertogh CM. Mobilization-Observation-Behaviour-Intensity-Dementia (MOBID-2) Pain Pain among institutionalized stroke patients and its relation to emotional Scale in a clinical setting. Scand J Caring Sci. 2010;24: 380-391. distress and social engagement. Int J Geriatr Psychiatry 2015; 30(10):1023-31. 10. Husebo BS, Strand LI, Moe-Nilssen R, BorgeHusebo S, Aarsland D, 24. Hobbelen JS, Koopmans RT, Verhey FR, Van Peppen RP, de Bie RA. Ljunggren AE. Who suffers most? Dementia and pain in nursing home Paratonia: a Delphi procedure for consensus definition. J Geriatr Phys Ther. patients: a cross-sectional study. J Am Med Dir Assoc. 2008;9: 427-433. 2006;29: 50-56. 11. Warden V, Hurley AC, Volicer L. Development and psychometric evaluation 25. Souren LE, Franssen EH, Reisberg B. Neuromotor changes in Alzheimer’s of the Pain Assessment in Advanced Dementia (PAINAD) scale. J Am Med disease: implications for patient care. J Geriatr Psychiatry Neurol. 1997;10: Dir Assoc. 2003;4: 9-15. 93-98. 12. Zwakhalen SM, van der Steen JT, Najim MD. Which score most likely 26. Sandvik R, Selbaek G, Kirkevold O, Aarsland D, Husebo BS. Analgesic represents pain on the observational PAINAD pain scale for patients with prescribing patterns in Norwegian nursing homes from 2000 to 2011: trend dementia? J Am Med Dir Assoc. 2012;13: 384-389. analyses of four data samples. Age Ageing. 2016;45: 54-60.

170 | GENERAL DISCUSSION CHAPTER 8 | 171 27. van Kollenburg EG, Lavrijsen JC, Verhagen SC, Zuidema SU, Schalkwijk A, Reinhard Wippler [Explanatory Sociology: essays for Reinhard Wippler]. Vissers KC. Prevalence, causes, and treatment of neuropathic pain in Dutch Amsterdam: Thesis Publisher, 1996; 169-184. nursing home residents: a retrospective chart review. J Am Geriatr Soc. 2012;60: 1418-1425. 28. Ellis JM, Wells Y, Ong JSM. Non-pharmacological approaches to pain management in residential aged care: a pre-post-test study. Clinical gerontologist. 2017: 1-11. 29. Fuchs-Lacelle S, Hadjistavropoulos T, Lix L. Pain assessment as intervention: a study of older adults with severe dementia. Clin J Pain. 2008;24: 697-707. 30. Beerens HC, Zwakhalen SM, Verbeek H, Ruwaard D, Hamers JP. Factors associated with quality of life of people with dementia in long-term care facilities: a systematic review. Int J Nurs Stud. 2013;50: 1259-1270. 31. Klapwijk MS, Caljouw MA, Pieper MJ, van der Steen JT, Achterberg WP. Characteristics associated with quality of life in long-term care residents with dementia: a cross-sectional study. Dement Geriatr Cogn Disord. 2016;42: 186-197. 32. van Dalen-Kok AH, Pieper MJ, de Waal MW, Lukas A, Husebo BS, Achterberg WP. Association between pain, neuropsychiatric symptoms, and physical function in dementia: a systematic review and meta-analysis. BMC Geriatr. 2015;15: 49. 33. Husebo BS, Ballard C, Sandvik R, Nilsen OB, Aarsland D. Efficacy of treating pain to reduce behavioural disturbances in residents of nursing homes with dementia: cluster randomised clinical trial. BMJ. 2011;343: d4065. 34. Husebo BS, Ballard C, Fritze F, Sandvik RK, Aarsland D. Efficacy of pain treatment on mood syndrome in patients with dementia: a randomized clinical trial. Int J Geriatr Psychiatry. 2014;29: 828-836. 35. Malara A, De Biase GA, Bettarini F, et al. Pain assessment in elderly with behavioral and psychological symptoms of dementia. J Alzheimer Dis. 2016;50: 1217-1225. 36. Corbett A, Husebo BS, Achterberg WP, Aarsland D, Erdal A, Flo E. The importance of pain management in older people with dementia. Br Med Bull. 2014;111: 139-148. 37. Husebo BS, Achterberg W, Flo E. Identifying and managing pain in people with Alzheimer’s Disease and other types of dementia: a systematic review. CNS Drugs. 2016;30: 481-497. 38. Gerritsen DL, Steverink N, Ooms ME, Ribbe MW. Finding a useful conceptual basis for enhancing the quality of life of nursing home residents. Qual Life Res. 2004 Apr;13(3):611-24. 39. Lindenberg S. Continuities in the theory of social production functions. In: Ganzeboom H, Lindenberg S (eds)., Verklarende sociologie: opstellen voor

172 | GENERAL DISCUSSION CHAPTER 8 | 173 Nederlandse samenvatting Curriculum Vitae A Dankwoord A Nederlandse samenvatting er werd eveneens een beperkt aantal studies gevonden. Op basis van de gevon- den studies is er niet met zekerheid te zeggen dat de pijnbeleving anders is voor de verschillende soorten dementie.

Pijn bij mensen met dementie, In het praktijkonderzoek is de aanwezigheid en mate van pijn onderzocht door met speciale aandacht voor verpleeghuisbewoners middel van een pijnobservatie en bij wie dit nog kon, is er gevraagd naar de aan- wezigheid van pijn. Daarnaast zijn er gegevens verzameld over de soort demen- tie, dementie ernst, (pijn)medicatie, medische voorgeschiedenis en kwaliteit van Pijn bij mensen met dementie heeft de afgelopen jaren steeds meer aandacht leven. De bevindingen uit het praktijkonderzoek zijn: gekregen. Het is zelfs een groot probleem genoemd. De helft van de mensen met - De meeste verpleeghuisbewoners met dementie hadden geen pijn. dementie zou pijn hebben en de aanwezigheid van pijn is in verband gebracht - Als er toch sprake was van pijn (43%), dan was dit in de helft van de gevallen met een verminderde de kwaliteit van leven bij mensen met dementie. Pijn bij een pijn met een lage intensiteit. dementie wordt toegeschreven aan zowel de ouderdom als de dementie zelf. Dit - Het voorkomen van pijn verschilde niet significant tussen de verschillende betekent dat hoe pijn wordt ervaren en het voorkomen van pijn kan veranderen soorten dementie. als er sprake is van dementie. Er zijn bovendien aanwijzingen dat de soort de- - Pijn kwam vaker voor bij verpleeghuisbewoners met ernstige tot zeer ernsti- mentie ook van invloed is op dit proces; mensen met vasculaire dementie zouden ge dementie. anders en vaker pijn ervaren dan mensen met Alzheimer dementie. Er is echter - Een derde van de verpleeghuisbewoners met pijn, die met pijnstillers werd nog niet veel onderzoek naar gedaan. Inzicht in het voorkomen van pijn bij ver- behandeld, had alsnog matig-ernstige tot ernstige pijn. schillende vormen van dementie zou kunnen bijdragen aan het optimaliseren van - 25% van de verpleeghuisbewoners met pijn had gemengde pijn (pijn met ken- de behandeling van pijn bij mensen met dementie. merken van zenuwpijn). - Pijnstillende middelen tegen zenuwpijn werden nauwelijks voorgeschreven. Voor dit proefschrift is middels literatuuronderzoek en praktijkonderzoek ge- - Pijn is in deze studie niet direct geassocieerd met kwaliteit van leven. keken naar het voorkomen van pijn en de pijnbeleving bij de vier meest voorko- - Bij de groep verpleeghuisbewoners die bij de eerste meting pijn hadden mende soorten dementie (Alzheimer dementie, vasculaire dementie, dementie en waarbij een terugkoppeling was gegeven aan de eigen arts was er na 3 met Lewy Bodies en frontotemproale dementie). In het praktijkonderzoek is ook maanden een significante afname van de pijnintensiteit, de pijn persisteerde gekeken naar de huidige praktijksituatie in Nederlandse verpleeghuizen ten aan- bij 58%. zien van het voorkomen van pijn, de mate van pijn die wordt ervaren, de invloed - Het gebruik van pijnmedicatie nam niet significant toe na 3 maanden. van de ernst van dementie op het voorkomen van pijn, de soort pijn die wordt er- varen, hoe pijn wordt behandeld en de relatie van pijn met kwaliteit van leven. Het De studies beschreven in dit proefschrift nuanceren de grootte van het ‘probleem’ praktijkonderzoek is uitgevoerd bij 200 verpleeghuisbewoners in 10 verschillende pijn bij mensen met dementie. Bovendien lijkt het soort dementie niet van in- verpleeghuizen en bij de groep bewoners met pijn werd dit teruggekoppeld aan vloed te zijn op het voorkomen van pijn. Bovenstaande bevindingen laten echter de eigen arts met een behandeladvies gebaseerd op de richtlijn ‘Herkenning en zien dat er een verbetering mogelijk is in het herkennen en behandelen van pijn behandeling van chronische pijn bij kwetsbare ouderen’ (Verenso). Na 3 maanden bij mensen met dementie. Bijvoorbeeld wanneer er sprake is van zenuwpijn of werden bij deze groep bewoners nogmaals gekeken naar de aanwezigheid van bij pijn ondanks het gebruik van pijnstillende middelen. Gepersonaliseerd pijn- pijn en de ingezette behandeling. management kan er voor zorgen dat pijnklachten herkend worden en optimaal behandeld kunnen worden. Middels een literatuuronderzoek is gekeken naar wat er in de literatuur bekend was over het voorkomen van pijn bij verschillende soorten dementie. Er is een Implicaties voor de praktijk beperkt aantal studies gevonden en deze studies lieten geen significant verschil Aanbevelingen voor de praktijk 1) Gepersonaliseerd pijnmanagement aangepast zien tussen het voorkomen van pijn bij de verschillende soorten dementie. Er is aan de individuele behoeften van de verpleeghuisbewoner met dementie. 2) Pijn- ook gekeken of de pijnbeleving verschilt bij verschillende soort dementie, maar behandeling bij mensen met dementie (in het verpleeghuis) zou op regelmatige

176 177 basis geëvalueerd moeten worden en zo nodig aangepast om daarmee comfort A Curriculum Vitae te verhogen.

Implicaties voor toekomstig onderzoek In toekomstige studies met verpleeghuisbewoners met dementie en pijn zou Curriculum Vitae (Nederlands) gekeken kunnen worden naar wat de meest passende behandeling is en het Janine van Kooten werd op 26 april 1984 in Naarden geboren. Na het behalen van evalueren van de effecten van deze behandeling. Dit heeft als doel om meer ge- haar gymnasium diploma aan OSG de Meergronden in Almere, begon zij in 2001 personaliseerd pijnmanagement aangepast aan de individuele behoeften van de met de studie geneeskunde aan de Vrije Universiteit in Amsterdam. Ze studeerde verpleeghuisbewoner met dementie te kunnen adviseren. af in 2008, waarna zij als basisarts is gaan werken. In september 2010 startte zij met de opleiding tot specialist ouderengeneeskunde. Vanaf september 2012 combineerde zij haar opleiding met een promotieonderzoek, waarvan u de resul- taten in dit proefschrift kunt lezen. In september 2016 heeft zij de opleiding tot specialist ouderengeneeskunde afgerond. Daarnaast studeerde zij in december 2016 af als epidemioloog. Sinds november 2016 is zij werkzaam als specialist ouderengeneeskunde bij Vivium Zorggroep.

Curriciulum Vitae (English) Janine van Kooten was born April 26, 1984 in Naarden, the Netherlands. In 2001 she completed her secondary education at OSG de Meergronden in Almere, the Netherlands. Afterwards, she studied medicine at VU University in Amsterdam. She graduated in 2008 and started working as a physician. In September 2010 she started as elderly care medicine trainee. In 2012 she started the PhD-project­ described in this thesis, which she combined with the elderly care medicine training. She was registered as elderly care physician in September 2016. She also followed a post‐master program Epidemiology, and was registered as an epidemiologist in December 2016. Since November 2016 she works as an elderly care physician at Vivium Zorggroep.

178 179 A Dankwoord Verder verdienen de andere projectgroepleden van de PAINdemiA studie een plek in het dankwoord. Allereerst Tarik en Suzanne, we hebben het toch maar mooi gedaan met elkaar! En hoewel we niet op 1 dag zijn/gaan promoveren, zullen we het alle drie in 2019 afronden. Chapeau voor ons! Eindelijk kan ik zeggen: het is klaar, mijn proefschrift is af. De weg was niet zonder hobbels, maar met de hulp en steun van velen heb ik mijn doel gehaald. Erik Scherder, initiator van het PAINdemiA project, Ik heb bewondering voor jouw enthousiasme, onuitputbare energie en passie. Bedankt voor de kansen die je mij Mijn dank gaat allereerst uit naar de betrokken zorgorganisaties, en dan met hebt gegeven. name de verzorgenden van de psychogeriatrische afdelingen in verpleeghuis Bornholm, Gerrie Knetemannlaan, Groenelaan, Groenhof, Leo Polak, Otto Hel- En alle andere projectgroepleden: Frank Lobbezoo, Andrea Maier, Didi Rheber- dringstraat, Saskia van Uijlenburgkade, het Schouw, St Jacob, Vreugdehof, en het gen, Roberto Perez (in herinnering) en Wouter Zuurmond, bedankt voor jullie Zonnehuis. Zonder jullie inzet was dit project niet mogelijk geweest. inbreng en begeleiding. Om met collega’s uit zoveel verschillende vakgebieden samen te werken heb ik als inspirerend ervaren en het heeft mijn blik verruimd. De leden van de beoordelingscommissie: prof.dr. W.A. van Gool, prof.dr. R.T.C.M. Koopmans, prof.dr. T.H. Muller, prof.dr. N. van den Noorgate en prof.dr. S.M.G. Verder verdienen mijn collega’s van de afdeling huisartsgeneeskunde & ouderen- Zwakhalen, bedankt voor het kritische lezen en beoordelen van mijn proefschrift. geneeskunde een woord van dank; ik kon de afgelopen jaren altijd wel bij iemand van jullie terecht met vragen, om te klankborden of te spuien. En niet te vergeten Graag wil ik mijn promotoren, Cees Hertogh en Max Stek, en co-promoteren, onze vrijdagmiddagborrels. Martin Smalbrugge en Hans van der Wouden, bedanken voor hun inbreng en be- geleiding bij het volbrengen van mijn promotietraject. Mijn (oud) kamergenootjes: Ewout, Lisanne (H-463), Simone, Sandra, Kirstin, Nienke en Letty (B-543). Daarnaast waren er nog vele andere collega’s die bij- Cees, bedankt voor het vertrouwen in mij om dit promotietraject aan te gaan en droegen aan een fijne werkomgeving, waaronder Laura, Tessa, Marike, Lizette, tot een goed einde te brengen. Jouw visie op de ouderengeneeskunde en hoe Ruth en alle andere collega’s en betrokkenen die ik niet genoemd heb (het is niets deze in te richten heeft zeker zijn weg naar dit proefschrift gevonden en daar wil persoonlijk als je hier niet bij naam genoemd staat). Dank! ik je hartelijk voor bedanken. Ook mijn opleidingsgroep en de medewerkers van Gerion wil ik hier genoemd Max, hoewel jij meer op afstand betrokken was bij mijn proefschrift, kon ik altijd hebben. Anouk, Annemiek, Hanneke, Jeannine, Lakshmi, Mahtab, Maaike, Men- rekenen op waardevolle inbreng en een persoonlijke noot. Ik waardeer je toegan- no en Michelle bedankt voor jullie support als mede-AIOS. Door jullie had ik al- kelijkheid en betrokkenheid zeer. tijd een groep waar ik thuis kwam. Claudia, Cora, Judith, Anne Marije, Jos en alle andere collega’s die ik niet genoemd heb. Dank voor jullie steun en flexibiliteit bij Martin, naast co-promotor ben je ook collega en vertrouwenspersoon geweest het maken van afspraken, opleidingsschema’s en nog veel meer. tijdens dit traject. Met jouw wetenschappelijke kennis en kritische blik werd elke versie van de artikelen in dit proefschrift weer beter. Maar ook op persoonlijk vlak En dan de mensen in mijn privéomgeving, met jullie steun en afleiding kan ik nu heb jij mij doen groeien. Heel veel dank voor de afgelopen jaren, ik hoop dat we in eindelijk zeggen dat het klaar is. Een kopje koffie op de markt in Ankeveen, een de toekomst kunnen blijven samenwerken. rondje (hard)lopen met of zonder dolichoten, zwemmen in de spiegelplas of met de masters van Triton, het heeft allemaal bijgedragen aan de totstandkoming van Hans, als ‘vreemde eend’ kwam jij bij mijn project. Maar de verschillende tussen dit proefschrift. kinderen en ouderen bleken toch niet zo groot. Jouw kennis over methodiek, in het bijzonder over het doen van systematische reviews, is van grote meerwaarde De laatste loodjes waren het zwaarst, maar gelukkig kan ik altijd bij je terecht, geweest voor mijn proefschrift. Daarnaast was je altijd bereid om mee te denken lieve Lisette. Jouw luisterende oor en stimulans om het nu toch echt af te maken en dingen in perspectief te zetten. Dank! hebben geholpen. Bedankt voor het opmaken van mijn proefschrift, het is super-

180 181 mooi geworden.

Simone, wij begonnen nagenoeg samen aan dit traject en ik mag dat nu, met jou als paranimf aan mijn zijde, ook gezamenlijk afsluiten. Ik ben blij dat ik je heb leren kennen en hoop dat we nog vele jaren van vriendschap voor ons hebben.

Lieve Joke, waar je ook bent op de wereld, altijd ben je er voor mij. En nu mag jij een van mijn paranimfen zijn.

Mama en papa, bedankt voor jullie onvoorwaardelijke liefde en steun. Mama, jij liet mij als kind kennis maken met het verpleeghuis en hebt zodoende het zaadje geplant. Papa, jij hebt mij altijd gestimuleerd om verder te leren. Dit dankwoord was er zonder jullie niet geweest.

Jorg, je had gelijk ♥

“Het boek is vol”, mevr V. (102 jaar)

182