Assessment of Pain in Rheumatic Diseases T

Home , Pain scale, Tenderness (medicine)

Assessment of pain in rheumatic diseases T. Sokka

Tuulikki Sokka, MD, PhD, Research ABSTRACT ture or myocardial infarction can pro- Assistant Professor, Vanderbilt University, Pain is the most prominent symptom in vide clinical information concerning Nashville, and Consultant Rheumatologist, people with musculoskeletal disorders, pain, and changes can be observed over Jyväskylä Central Hospital, Jyväskylä, and the most common motivation for the next few hours and days without the Finland. patients seeking medical help. Howev - apparent need for quantitative data. Reprinted and modified from Best Pract er, pain generally is not recorded quan - Pain and other symptoms are regarded Res Clin Rheumatol, vol. 17, Sokka T: "Assessment of pain in patients with titatively in routine medical care. Over as “subjective,” based on data obtained rheumatic diseases", pp. 17: 427-49, the last three decades, self-report ques - from the patient, and are viewed by the copyright 2003, with permission from t i o n n a i res have been developed in clinician largely as preliminary to criti- Elsevier. which a patient may record quantita - cal “objective” data obtained from the Please address correspondence and tively a pain score at baseline and over physical examination, laboratory tests, reprint requests to: Tuulikki Sokka, MD, time to determine whether their condi - or imaging procedures. This view is PhD, Vanderbilt University / Rheumato- tion has improved, remains unchanged, consistent with what has been termed logy, 203 Oxford House, Nashville, TN or has worsened. The most robust quan - the traditional “biomedical model” 37232-4500, USA. titative pain measure appears to be a (12), which has been applied so suc- E-mail: [email protected] simple 10 cm visual analog scale (VAS ) , cessfully in acute medical care during Clin Exp Rheumatol 2005; 23 (Suppl. 39): which can be completed by the patient the 20 th century that it is often applied S77-S84. and scored by a health professional in to chronic diseases as well. © Copyright CLINICALAND EXPERIMENTAL less than 10 seconds. Quantitative data At this time, chronic diseases are the RHEUMATOLOGY 2005. concerning pain cannot be obtained most important problem in medical Key words: Pain, assessment of pain, from any source other than the patient. care. In the management of chronic dis- rheumatoid arthritis, osteoarthritis, Quantitative assessment of pain at ea c h eases, the “biomedical model” is not as fibromyalgia, MDHAQ. visit in routine rheumatology care , useful as in acute diseases in guiding along with the assessment of functional diagnosis and management and has d i s a b i l i t y, global status, and other substantial limitations. For example, patient variables, using a patient self- the most effective predictors of mortal- re p o rt questionnaire might lead to ity in patients with rheumatoid arthritis improved patient care. (RA) include data from a patient questionnaire concerning the patient’s phys- Introduction ical function and level of formal educa- Pain is the most prominent symptom in tion, rather than data from a physical the majority of people with arthritis (2- examination, laboratory tests or radio- 5), a common reason for primary care graph (13-16). In the management of consultation (6-8), and a major source chronic rheumatic diseases, it is virtu- of health care costs (9). Musculoskele- ally impossible to assess pain over long tal pain appears to be much more com- periods without quantitative data to mon now than 40 years ago (10). None- estimate whether or not a patient’s con- theless, quantitative information con- dition is improved, unchanged or worse cerning pain, which is required to over months to years. assess and document possible improve- A clinical science of pain assessment ment, stabilization, or worsening of using patient self-report questionnaires pain over time, is generally not record- has been developed over the last few ed in routine medical care. In a survey decades to facilitate qualitative and of U.S. emergency department visits in quantitative assessment of pain status 1999, 52% included no recorded infor- at any given time (17-27). Despite lim- mation concerning the presenting level itations which are intrinsic in any sci- of pain (11). entific measurement, pain question- In acute medical situations, the primary naires have proven valuable in the stu- setting for most medical education and dy of the mechanisms underlying the training, the quantitative assessment causes and control of pain. and recording of pain levels may ap- In this essay, we review patient self- pear unnecessary.A patient with a frac- report questionnaires as quantitative

S-77 Assessment of pain in rheumatic diseases / T. Sokka measures of pain. We summarize the log scale, but that generally self-report- derivations of the HAQ have been dev- results generated using these question- ing is adequate thereafter. eloped, including a modified HAQ (M- naires in RA, osteoarthritis (OA), and The standard visual analog scale is a 10 HAQ) (36), a multidimensional HAQ fibromyalgia (FM). We also summarize cm scale with a border on each side. To (MDHAQ) (37), and the HAQII (38). data which reveal significant associa- the left of the “0” mark appears the The HAQ and its derivative versions tions between pain scores and the phys- indication “No pain at all”, and to the are discussed in greater detail in other ical examination, and radiographic and right of the “10” mark “Pain as bad as it chapters in this supplement. laboratory data. Such associations ap- could be”. There are occasional distor- pear weaker, however, than the associations through photocopying and print- The Arthritis Impact Measurement tions of pain scores with measures of ing, but adjustments can be made so Scales (AIMS) functional and psychological status. that the total score is 10. The Arthritis Impact Measurement Huskisson and colleagues also pointed Scales (AIMS) was developed by Assessment of pain using patient out that an alternative descriptive pain Meenan and colleagues (22) to assess questionnaires relief scale – based on the indications the physical, emotional and social well- The Minnesota Multiphasic Person- “complete”, “moderate”, “slight” and being of individuals with arthritis, with ality Inventory and the McGill Pain “no pain” relief – was possible, but scores for 9 categories: mobility, physi- Questionnaire much less sensitive than the visual ana- cal activity, social activity, social role, The Minnesota Multiphasic Personality log scale. A number of studies have activities of daily living, pain, dexteri- Inventory (MMPI) (28) is an early established that data from self-report ty, anxiety, and depression. Each score patient self-report questionnaire. A l- visual analog scales are reproducible is based on 4 to 6 items with response though not strictly a pain questionnaire, (34, 35). In one study (35) an absolute alternatives on Likert-format scales. the MMPI represents one of the first visual analog scale was found to be The AIMS pain category includes 4 widely used patient questionnaires more reproducible than a comparative questions: “During the past month, which gained acceptance over the last visual analog scale. how often have you had severe pain half-century. With the development of optical scan- from your arthritis ?”; “During the past The McGill Pain Questionnaire (17, ning technology for the automated month, how would you describe the 18) constituted a major advance in clin- computer entry of scores, visual analog arthritis pain you usually have ? ” ; ical research on pain. The question- scales have been presented in a format “During the past month, how long has naire is complex and completion re- of 21 small boxes or circles for patients your morning stiffness usually lasted quires 15-20 minutes, even in its short to assess their pain from 0-10 (or 100). from the time you wake up ?” “During form (29). Therefore, it is not easily Although formal direct comparative the past month, how often have you administered in a non-research clinical studies have not been performed to had pain in two or more joints at the setting, and simpler measures – such as analyze the results of automated optical same time ?” a visual analog pain scale – have scanning, they appear to have criterion The AIMS index has excellent psycho- become more widely accepted for use validity. The visual analog pain scale metric validity and greater reliability in clinical research, clinical trials, and has proven a great advance in the as- than the HAQ and its derivatives, and clinical care. sessment of pain. has been used in clinical trials to document the sensitivity of patient question- Visual analog pain scales The Health Assessment Questionnaire naires to changes in clinical status. A visual analog pain scale was initially (HAQ) and its derivatives: The Modi- However, the HAQ and its derivatives used in psychology by Freyd and oth- fied Health Assessment Questionnaire are more easily completed by patients ers since the early 1900s. Huskisson (MHAQ), Multi-Dimensional Health and more easily scored by health pro- and colleagues developed the use of a Assessment Questionnaire (MDHAQ), fessionals in clinical trials and in rou- pain VAS in rheumatology through a and Health Assessment Questionnaire tine care, and used considerably more series of investigations in the late (HAQ-II) widely than the AIMS. 1970s (19, 20, 30-33), pointing out that The HAQ was developed in the 1970s “only the patient can measure [pain] by Fries and associates and published The Western Ontario McMaster severity” (30). These investigators des- in Arthritis and Rheumatism in 1980 (WOMAC) cribed a variety of visual analog scales, (21). This questionnaire provided a The Western Ontario McMaster (WO- including vertical and horizontal scales, milestone in the development of a meth- MAC) questionnaire was developed, and scales with equally spaced lines odology based on patient self-reporting based on a survey of 100 patients with with the indications of mild, moderate to obtain information concerning func- primary OAof the hip or knee, initially and severe pain. They concluded that tional disability, pain and global status. for use in OA clinical trials (23, 24). numbers should not be included. They The HAQ includes visual analog scales The WOMAC consists of 24 items: 5 to also suggested that assistance from a for pain, as well as global status, al- assess pain, 2 to assess stiffness and 17 health professional is helpful the first though it was primarily designed to to assess physical function. The ques- time a patient completes a visual ana- measure functional disability. Several tions concerning pain include “walking

S-78 Assessment of pain in rheumatic diseases / T. Sokka on flat surfaces”, “going up and down much bodily pain have you had during well as a drug to treat the patient. It is stairs”, “at night while in bed”, “sitting the past four weeks?” with response less perfectly expressed, but regarded or lying”, and “standing upright”. options 1 =none; 2= very mild; 3= as valid, in tests such as the cardio- The WOMAC has been administered mild; 4 = moderate; 5 = severe; 6 = gram, or assays for serum glucose and as a Likert Scale with 5 or 7 response very severe, and “During the past four rheumatoid factor, in which there is a options, and as a series of 10 cm visual weeks, how much did pain interfere strong probability of a diagnosis based analog scales. It has been extensively with your normal work (including both on a positive finding in a test, but not used in OA clinical trials throughout work outside the home and house- an absolute correlation. the world, and is regarded as the “gold work) ?” with response options 1 = not Analyses of scores for pain in RAindi- standard” for the assessment of OA of at all; 2 = a little bit; 3 = moderately; 4 cate that there does exist a statistically the lower extremities. = quite a bit; 5 = extremely. significant correlation between pain The SF-36 has documented validity in scores and findings on radiographs and Nottingham Health Profile (NHP) normal healthy populations and diverse other objective measures. The assess- The Nottingham Health Profile (NHP) patient groups and is widely used. It is ment of pain and its correlation with was introduced in the early 1980s (25) sensitive to changes in clinical status, traditional measures is discussed in as a generic health status questionnaire. and occupies a well-earned place in greater detail below for the three most Generic health status questionnaires clinical trials. The scoring procedure is common rheumatic diseases – rheuma- were developed for use in many types complex, with recoding of the respons- toid arthritis, osteoarthritis and fibro- of diseases, in contrast to the HAQ and es according to instructions on a scale myalgia. AIMS, which were developed for use of 0-100, where 100 indicates “the in patients with rheumatic diseases. best” and 0 “the worst” health situa- Pain in rheumatoid arthritis The NHP is based on patient perception, and calculation of the mean value Pain is the major reason for patients tions of health, and was designed to for the recoded responses. The compli- with RAto seek medical care (2-5, 40), help people express how they feel when cated scoring system makes the SF-36 although these patients experience experiencing various states of ill health. unfeasible for use in standard clinical many other symptoms such as joint The pain section of the NHP includes 8 care. swelling, tenderness, deformities, and questions concerning pain with the morning stiffness. Furthermore, pain is response alternatives “yes” and “no”. Rheumatoid Arthritis Pain Scale the area of health in which most of The scoring includes a weighting of all (RAPS) patients with RA would like to see “yes” responses with a certain popula- The Rheumatoid Arthritis Pain Scale improvement (5, 40). Fries et al. (41, tion specific value, and adding the (RAPS) (27) was developed to measure 42) showed that DMARDs are the best scores of individual questions together. pain in adult patients with RA. The do- drugs in the long-term for relieving The final score for each concept ranges mains of RAPS include physiologic, pain in RA. More frequent visits to from 0, indicating good health, to 100, a ffective, sensory-discriminative, and rheumatologists were associated with which indicates poor health. cognitive components, and consists of greater improvements in pain and func- The NHP has been used in clinical re- 24 items that are scored using a 7-point tional capacity over one year (43). Sev- search, although it also has a floor ef- Likert scale ranging from “0 = never” eral recent clinical trials of DMARDs fect, i.e. it is poorly sensitive to small to “6 = always”, which is considered to showed statistically significant im- degrees of change in health (39). Fur- represent a greater severity of pain. The provement in pain over 6 to 24 months thermore, the questionnaire is long and RAPS, like the McGill pain question- in treatment groups compared to incorporates a complicated scoring sys- naire, clearly provides more informa- groups that received control medica- tem, and hence is not practical for use tion than a visual analog pain scale and tions or placebo (1). Data concerning in most clinical trials and routine clini- is useful in clinical research, but less so pain independent of other measures are cal care. outside specialized research settings. not often reported in clinical trial results, as data are presented in the form Short-Form 36 (SF-36) Associations of pain and other of pooled indices such as the A C R A 36-item questionnaire called the measures of clinical status response criteria. Short-Form 36 (SF-36) was developed The dominant paradigm of 20th century Pain follows the same pattern of devel- by Ware and associates (26), initially medicine is the “biomedical model”, in opment as other parameters of disease for use in health policy surveys. The which symptoms are regarded as being activity in groups of patients with RA SF-36 assesses 8 health concepts: 1) explainable by “objective” information (44). After initial improvement, pain physical activities; 2) social activities; from a physical examination, radio- scores deteriorate over the years. Borg 3) role activities; 4) bodily pain; 5) graphs, imaging studies, laboratory and Dawes (45) found that pain at the general mental health; 6) role activities tests, and other high technology proce- onset of the disease did not predict the because of emotional problems; 7) vi- dures. This paradigm is expressed opti- pain level at 3 years, while in another tality; and 8) general health. The pain mally in acute infectious diseases, in study baseline pain in early disease was section includes two questions: “How which a test identifies a pathogen, as the only significant predictor of cumu-

S-79 Assessment of pain in rheumatic diseases / T. Sokka

Table I. Pain in selected longitudinal observational studies over 5 years or more in patients with rheumatoid arthritis.

Mean disease duration at Duration of follow-up, Pain measure Pain at baseline Pain at P-value for baseline, years years evaluation paired data

Egsmose et al., 1995 (48) 0.8 (all < 2 years) 5 VAS, Early: 44 17 < 0.001 scale 0-100 Delayed: 51 40 < 0.02

Eberhardt and Fex, 1995 (49) 0.9 (all < 2 years) 5 VAS, 1.4 1.0 £ 0.01 scale 0-3

Lindqvist et al., 2002 (50) 10 (included VAS, 1.2 1.2 NS 168 patients) Scale 0-3

Muhlerin et al., 1996 (51) 2.4 (range 0.2-12.0) 6 VAS, 47 32 < 0.005 scale 0-100

Callahan et al., 1997 (52) 9.7 5 VAS, 52 47 NS scale 0-100

Munro et al., 1998 (53) Range 0-2 years 5 VAS, 1.7 1.1 < 0.01 Range 2-5 years scale 0-3 1.8 1.5 0.313 Range > 5 yrars 1.9 1.5 0.039

Leirisalo-Repo et al., 1999 (44) 0.7 (all < 2 years) 8-9 VAS, 43 17 scale 0-100

Uhlig et al., 2000 (54) 2.2 (all < 4 years) 5 AIMS, pain 4.6 4.7 0.12 scale 0-10

*Heiberg et al., 2005 (55) Disease duration was 13 years in patients with data VAS, 46 36 Significant; at baseline, and 14 years in patients who were scale 0-100 p-values not evaluated 7 years later. provided

**Pincus et al., 2005 (47) Median disease duration was 7 years in patients VAS, 52 49 0.38 included in the first evaluation (1985) and 9 years scale 0-100 in the second (2000)

VAS: Visual Analog Scale for pain; AIMS: Arthritis Impact Measurement Scales; NS: not significant. *Study was cross-sectional in part; **study was entirely cross-sectional. lative pain over a year (46). A cross- ESR and radiographs in patients with a pain VAS < 20 mm is a useful cut-off sectional quantitative assessment of RA(Table II) (49, 56-64), r levels < 0.3 point for pain (67). In a study that consecutive RA patients seen in 1985 indicate that less than 10% of the varia- included patients with inflammatory versus 2000 in the same clinic indicat- tion in pain scores can be explained by and degenerative rheumatic conditions, ed significantly better status in the joint the variation in radiographic scores or 75% of patients considered their status count, radiographic, laboratory, patient laboratory tests. In some studies, a as “acceptable” when their pain level questionnaire, and physical function stronger association has been seen was < 25mm (68). The estimated aver- measures in 2000 compared to 1985, between pain scores and joint swelling age level of pain was 20 on a 100 mm but not in the pain scores (47). On the and tenderness, up to r = 0.4-0.6, indi- VAS in the elderly general population general population level, musculoske- cating an explanation of 15%-35% of (69). Therefore, it appears that a score letal pain is much more common now the variation in these measures by the of “0” for pain is not a realistic goal, than 40 years ago (10). variation in pain scores (57, 59). How- and a score of 10-25 on a 100 mm VAS Changes in pain levels over 5 years or e v e r, the strongest associations be- may indicate normal status. more have been reported in long-term tween pain scores and other variables follow-up studies during the last de- in RA are seen in the scores for func- Pain in osteoarthritis cade (44, 47-55) (TableI). Overall, im- tional status, and in the scores for the Pain is the most important determinant provement in pain over 5 years is more psychological constructs of anxiety, of disability in patients with OA (70- significant in patients with early dis- depression, helplessness and lack of 73). OAis not inevitably a progressive, ease, while improvement in pain is less self-efficacy (Table II). degenerative disease, but rather a col- pronounced in patients with a longer The absence of joint pain is included in lection of heterogenous conditions with disease duration at the outset of obser- the ACR remission criteria for RA(65). a dynamic course that may also include vation. On a 100 mm VAS scale, < 10 has been repair and periods of structural stability Although significant correlations can interpreted as “no pain” (66). The 6th (74-76). be seen between pain scores and the OMERACT conference suggested that Structural changes such as cartilage

S-80 Assessment of pain in rheumatic diseases / T. Sokka

Table II. Correlation coefficients between pain and other measures in rheumatoid arthritis.

Pain No. pts. Correlation coefficients between pain and other measures measure Age Disease Education HAQ/ Patient Disease ESR SJC TJC AM x-rays duration MHAQ global activity, pooled

Callahan et al., 1987 (56) VAS 385 -0.06 0.17 -0.14 0.55 0.53 - 0.24 0.33 0.34 0.24 -

Hagglund et al., 1989 (57) AIMS 53 ------0.14 0.45 - - - pain scale

Serbo and Jajic 1991 (58) VAS 61 - - - 0.52 ------

Stenström et al.,1992 (59) VAS, 69 0.08 0.12 - 0.55 - - 0.06 0.04 0.66 - 0.21 activity induced pain

Hakala et al., 1994 (60) VAS 103 - - - 0.36 ------

Smedstad et al., 1995 (61) VAS 238 ------0.06 - -

Eberhardt and Fex,1995 (49) VAS 67 - - - 0.44 ------(baseline) 67 - - - 0.59 ------(at 5 years)

Rojkovich and Gibson, VAS Daytime 0.14 0.16 - - - - 0.17 0.03 0.25 - 0.09 1998 (62) movement n=251 Daytime rest 0.19 0.02 - - - - 0.18 0.03 0.25 - 0.16 n=232 Nocturnal 0.02 -0.13 - - - - 0.25 0.21 0.36 - 0.08 n=181

Sokka et al., 2000 (63) VAS 141 - - - 0.65 ------0.01

Sarzi-Puttini et al., 2002 (64) VAS 105 0.04 0.07 - 0.42 0.57 - 0.38 0.05 0.04 0.43 0.06

–: not available. VAS: visual analog scale; AIMS: Arthritis Impact Measurement Scales; HAQ: Health Assessment Questionnaire; MHAQ: Modified Health Assessment Questionnaire; ESR: erythrocyte sedimentation rate; SJC: swollen joint count; TJC: tender joint count; AM: morning stiffness.

Table III. Rationale to assess pain on a self-report questionnaire.

• Pain is the most prominent symptom in people with musculoskeletal disorders, and the most common reason for seeking medical help, but is rarely quantitatively measured or recorded in routine medical care.

• Since pain is a personal experience, estimated changes in levels of pain over long periods in patients with rheumatic diseases cannot be obtained from any source other than the patient.

• Several self-report questionnaires have been developed to measure pain and disability, and are well-documented tools for clinical trials.

• Short questionnaires that do not involve complicated scoring systems are the most feasible instruments for implementation as a part of routine care.

• The MDHAQ is a one-page, two-sided questionnaire which includes a visual analog pain scale. It can be completed by the patient as a self-report questionnaire in less than 10 minutes in the waiting room and can be scored by a health professional in less than 30 seconds. loss, periarticular bone growth, osteo- NES1) (80) indicated that 53% of indi- proportion of patients who seek clinical phyte formation and sclerosis, are only viduals who had radiographic findings care because of pain have radiographic weakly correlated with the severity of of stage 3-4 OA according to the Kell- abnormalities, but the correlation is far symptoms, including joint pain, use- gren-Lawrence scale did not report any from perfect. related stiffness, and disability (77-79). knee pain. Conversely about 85% of Furthermore, De Bock et al. (81) found A population-based study of OA de- people who reported significant knee a low association between a patient’s rived from the US National Health and pain did not have significant radio- perception of pain and the physician’s Nutrition Examination Survey (NHA- graphic abnormalities. Clearly a higher assessment of pain in patients with OA.

S-81 Assessment of pain in rheumatic diseases / T. Sokka

Although pain and structural damage in likely to have chronic widespread pain doctor in treating patients with rheu- OA have been significantly correlated 7 years later; 77% of the subjects who matic conditions. A rationale to assess in groups of patients, many patients reported chronic widespread pain ini- pain on a self-report questionnaire is with minimal pain may have consider- tially also reported chronic widespread presented in Table III. able structural damage while others pain after 7 years. Although not clini- References may report extensive pain with little cally examined to establish the diagno- 1. S O K K A T: Assessment of pain in patients damage. Therefore, it appears reason- sis, it appears likely that most of these with rheumatic diseases. Best Pract Res Clin able to use patient questionnaires to subjects had FM. Rheumatol 2003; 17: 427-49. document the amount of pain and dis- 2. K A Z I S LE, MEENAN R F, A N D E R S O N J J: Pain in the rheumatic diseases: investigation ability in these patients, not only in The use of patient self-rep o r t ques- of a key health status component. Arthritis clinical trials but also in standard clini- ti o n n a i r es in routine clinical care Rheum 1983; 26: 1017-22. cal care. In a clinical trial comparing D i fferences exist between research 3. GIBSON T, CLARK B: Use of simple anal- two study drugs in OA(82), differences questionnaires, which may be very gesics in rheumatoid arthritis. Ann Rheum Dis 1985; 44: 27-9. in the results of treatment based on a lengthy and require complex scoring 4. MCKENNA F, WRIGHT V: Pain and rheuma- VAS pain scale were as substantial as systems, and simple questionnaires de- toid arthritis (Letter). Ann Rheum Dis 1985; those based on the WOMAC scale, signed for use in standard clinical care. 44: 805. suggesting that a simple pain VAS is In routine clinical care and for most 5. ANDERSON KO, BRADLEYLA, TURNER RA, AGUDELO CA, PISKO EJ: Pain behavior of more than adequate in standard care clinical research, a visual analog scale rheumatoid arthritis patients enrolled in ex- (83). score appears to capture just as much perimental drug trials. A rthritis Care Res information as more elaborate ques- 1994; 7: 64-8. Pain in fibromyalgia tionnaires designed to measure pain, 6. R E K O L A K E , K E I N A N E N - K I U K A A N N I E M I K , TA K A L A J: Use of primary health ser- FM is a syndrome characterized by and is often more sensitive to changes vices in sparsely populated country districts widespread pain, which often involves in clinical trials and clinical care. The by patients with musculoskeletal symptoms: all four quadrants of the body as well as longer questionnaires are of value to consultations with a physician. J Epidemiol the axial skeleton, and diffuse tender- analyze the mechanisms and patho- Community Health 1993; 47: 153-7. 7. MANTYSELKA P, KUMPUSALO E, AHONEN ness, but without evidence of structural physiology of pain, but a VAS scale for R et al.: Pain as a reason to visit the doctor: a damage such as that seen in OA or in- pain appears to capture all of the infor- study in Finnish primary health care. Pain flammation as seen in RA. The etiolo- mation needed in a clinical study, in- 2001; 89: 175-80. 8. UHLIG T, HAGEN KB, KVIEN TK: Why do gy and pathogenesis of pain in FM are cluding clinical trials. patients with chronic musculoskeletal disor- unknown. Reports using positron emis- The evidence that scores for pain are ders consult their primary care physicians? sion tomography and functional mag- not directly correlated with objective Curr Opin Rheumatol 2002; 14: 104-8. netic resonance imaging (fMRI) sug- data indicates that data concerning pain 9. CROOK J, RIDEOUT E, BROWNE G: T h e prevalence of pain complaints in a general gest that a group of brain structures are should be derived from the patient population. Pain 1984; 18: 229-314. activated during painful conditions rather than from efforts to measure pain 10. HARKNESS EF, MACFARLANE GJ, SILMAN (84-89). through objective measures. The most AJ, MCBETH J: Is musculoskeletal pain more One approach to identifying patients reproducible data on pain are derived common now than 40 years ago ?Two population-based cross-sectional studies. Rheum - with FM is suggested by a study which from patients rather than from a health atology (Oxford) 2005; 44: 890-5. reported higher scores for pain relative professional. 11. MCLEAN SA, MAIO RF, DOMEIER RM: The to scores for functional disability in epidemiology of pain in the prehospital set- about 50% of patients with FM com- Conclusion ting. Prehosp Emerg Care 2002; 6: 402-5. 12. ENGELGL: The need for a new medical mod- pared to RA (90). Furthermore, the ra- Pain is a personal experience. There- el: a challenge for biomedicine. S c i e n c e tios of scores for fatigue on a VA S fore, information about pain can be ob- 1977; 196: 129-36. compared to MHAQ were also consid- tained best from the patient. Several 13. PINCUS T, CALLAHAN LF, SALE W G , erably higher in patients with FM com- self-report questionnaires – including BROOKS AL, PAYNE LE, VAUGHN W K: Severe functional declines, work disability, pared to RA (91). Indeed, receiver op- lengthy research questionnaires and a and increased mortality in seventy-five rheu- erator curve data indicated that the re- simple visual analog pain scale – have matoid arthritis patients studied over nine sults compared favorably with the use been developed over the last few de- years. Arthritis Rheum 1984; 27: 864-72. 14. P I N C U S T, CALLAHAN L F: Formal educa- of ESR to distinguish patients with FM cades, and all of them constitute well- tion as a marker for increased mortality and from patients with RA. documented tools for research purpos- morbidity in rheumatoid arthritis. J Chronic FM pain is persistent, and complete es. It is recommended that a quantita- Dis 1985; 38: 973-84. sustained remission of pain is rare (92). tive assessment of pain be carried out at 15. P I N C U S T, BROOKS RH, CALLAHAN LF: Prediction of long-term mortality in patients One study (93) which focused on chro- each visit in routine rheumatology care, with rheumatoid arthritis according to simple nic widespread pain in the population along with an assessment of functional questionnaire and joint count measures. Ann indicated that subjects with widespread disability, the global status, and other Intern Med 1994; 120: 26-34. pain who were more than 50 years old patient variables, using a patient self- 16. PINCUS T, CALLAHAN LF: Associations of low formal education level and poor health and reported daytime tiredness and report questionnaire, which provides status: Behavioral, in addition to demograph- somatic symptoms initially were most clinically useful information to the ic and medical, explanations ? J Clin Epi -

S-82 Assessment of pain in rheumatic diseases / T. Sokka

demiol 1994; 47: 355-61. 1983; 26: 1346-53. BROOKS RH, NANCE EP, JR, KAYE JJ: Mea- 17. MELZACK R: The McGill Pain Question- 37. PINCUS T, SWEARINGEN C, WOLFE F: To- sures of activity and damage in rheumatoid naire: major properties and scoring methods. ward a multidimensional health assessment arthritis: Depiction of changes and prediction Pain 1975; 1: 277-99. questionnaire (MDHAQ): Assessment of of mortality over five years. Arthritis Care 18. M E L Z A C K R: The McGill pain question- advanced activities of daily living and psy- Res 1997; 10: 381-94. naire. In MELZACK R (Ed.): Pain Measure - chological status in the patient friendly health 53. MUNRO R, HAMPSON R, MCE N T E G A RT A , ment and A s s e s s m e n t. New York, Raven assessment questionnaire format. A rt h r i t i s THOMPSON EA, MADHOK R, CAPELL H: Press, 1983: 41-7. Rheum 1999; 42: 2220-30. Improved functional outcome in patients 19. HUSKISSON EC: Measurement of pain. 38. W O L F E F, MICHAUD K, PINCUS T: Devel- with early rheumatoid arthritis treated with Lancet 1974; 2: 1127-31. opment and validation of the health assess- intramuscular gold: Results of a five year 20. HUSKISSON EC : Assessment for clinical tri- ment questionnaire II: A revised version of prospective study. Ann Rheum Dis 1998; 57: als. Clin Rheum Dis 1976; 2: 37-49. the health assessment questionnaire. Arthritis 88-93. 21. FRIES JF, SPITZ P, KRAINES RG, HOLMAN Rheum 2004; 50: 3296-305. 54. U H L I G T, SMEDSTAD LM, VAGLUM P, H R: Measurement of patient outcome in 39. C A R R AJ, THOMPSON PW, KIRWAN JR: M O U M T, GERARD N, KVIEN T K: T h e arthritis. Arthritis Rheum 1980; 23: 137-45. Quality of life measures. Br J Rheumatol course of rheumatoid arthritis and predictors 22. MEENAN RF, GERTMAN PM, MASON JH: 1996; 35: 275-81. of psychological, physical and radiographic Measuring health status in arthritis: the 40. HEIBERG T, KVIEN TK: Preferences for im- outcome after 5 years of follow-up. Rheuma - arthritis impact measurement scales. Arthritis proved health examined in 1,024 patients tology 2000; 39: 732-41. Rheum 1980; 23: 146-52. with rheumatoid arthritis: pain has highest 55. HEIBERG T, FINSET A, UHLIG T, KVIEN TK: 23. BELLAMYN, BUCHANAN WW, GOLDSMITH priority. Arthritis Rheum 2002; 47: 391-7. Seven year changes in health status and pri- CH, CAMPBELLJ, STITT LW:Validation study 41. FRIES JF, SPITZ PW, MITCHELL DM, ROTH orities for improvement of health in patients of WOMAC: A health status instrument for SH, WOLFE F, BLOCH DA: Impact of specif- with rheumatoid arthritis. Ann Rheum Dis measuring clinically important patient rele- ic therapy upon rheumatoid arthritis. Arthri - 2005; 64: 191-5. vant outcomes to antirheumatic drug therapy tis Rheum 1986; 29: 620-7. 56. CALLAHAN LF, BROOKS RH, SUMMEY JA, in patients with osteoarthritis of the hip or 42. FRIES JF: Re-evaluating the therapeutic ap- PINCUS T: Quantitative pain assessment for knee. J Rheumatol 1988; 15: 1833-40. proach to rheumatoid arthritis: the “saw- routine care of rheumatoid arthritis patients, 24. BELLAMY N: Pain assessment in osteoarthri- tooth” strategy. J Rheumatol 1990; 17 (Sup- using a pain scale based on activities of daily tis: experience with the WOMAC osteo- pl. 22): 12-5. living and a visual analogue pain scale. arthritis index. Semin Arthritis Rheum 1989; 43. WARD MM : Rheumatology visit frequency Arthritis Rheum 1987; 30: 630-6. 18 (Suppl. 2): 14-7. and changes in functional disability and pain 57. HAGGLUND KJ, HALEY WE, REVEILLE JD, 25. HUNT SM, MCEWEN J, MCKENNA SP: Mea- in patients with rheumatoid arthritis. J ALARCON GS: Predicting individual differ- suring health status: A new tool for clinicians Rheumatol 1997; 24: 35-42. ences in pain and functional impairment and epidemiologists. J R Coll Gen Pract 44. L E I R I S A L O - R E P O M, PA I M E L A L, PELTO- among patients with rheumatoid arthritis. 1985; 35: 185-8. M A A R et al.: Functional and radiological Arthritis Rheum 1989; 32: 851-8. 26. WARE JE JR, SHERBOURNE CD: The MOS outcome in patients with early RA -A longi- 58. SERBO B, JAJIC I: Relationship of the func- 36-item short-form health survey (SF-36): I. tudinal observational study. Arthritis Rheum tional status, duration of the disease and pain Conceptual framework and item selection. 1999; 42: S130. intensity and some psychological variables in Med Care 1992; 30: 473-81. 45. B O R G A A , DAW E S P T: Measures of pain patients with rheumatoid arthritis. C l i n 27. ANDERSON DL: Development of an instru- and disease activity in rheumatoid arthritis. Rheumatol 1991; 10: 19-22. ment to measure pain in rheumatoid arthritis: Br J Rheumatol 1993; 32: 1028-9. 59. STENSTRÖM CH, LINDELLB, SWANBERG P, rheumatoid arthritis pain scale (raps). Arthri - 46. VAN DER HEIDE A, JACOBS JW, HAANEN HC, N O R D E M A R R, HARMS-RINGDAHL K: A c- tis Care Res 2001; 45: 317-23. BIJLSMAJW: Is it possible to predict the first tivity-induced pain in rheumatoid arthritis 28. MCKINLEY JC, HATHAWAY SR: The identifi- year extent of pain and disability for patients functional class II and its relations with dem- cation and measurement of the psychoneu- with rheumatoid arthritis? J Rheumatol 1995; ographic, medical, functional, psychosocial, roses in medical practice: the Minnesota mul- 22: 1466-70. and work variables. Arthritis Care Res 1992; tiphasic personality inventory. JAMA 1943; 47. PINCUS T, SOKKA T, KAUTIAINEN H: 5: 42-8. 122: 161-7. Patients seen for standard rheumatoid arthri- 60. H A K A L A M, NIEMINEN P, MANELIUS J: 29. MELZACK R: The short-form McGill Pain tis care have significantly better articular, Joint impairment is strongly correlated with Questionnaire. Pain 1987; 30: 191-7. radiographic, laboratory, and functional sta- disability measured by self-report question- 30. SCOTT J, HUSKISSON EC: Graphic represen- tus in 2000 than in 1985. Arthritis Rheum naires. Functional status assessment of indi- tation of pain. Pain 1976; 2: 175-84. 2005; 52: 1009-19. viduals with rheumatoid arthritis in a popula- 31. SCOTT J, HUSKISSON EC: Vertical or hori- 48. EGSMOSE C, LUND B, BORG G et al.: Pa- tion based series. J Rheumatol 1994; 21: 64- zontal visual analogue scales. Ann Rheum tients with rheumatoid arthritis benefit from 9. Dis 1979; 38: 560. early 2nd line therapy: 5 year followup of a 61. S M E D S TAD LM, VAGLUM P, KVIEN T K , 32. HUSKISSON EC: Measurement of pain. J prospective double blind placebo controlled MOUM T: The relationship between self- Rheumatol 1982; 9: 768-9. study. J Rheumatol 1995; 22: 2208-13. reported pain and sociodemographic varia- 33. HUSKISSON EC: Visual analogue scales. In 49. EBERHARDT KB, FEX E: Functional impair- bles, anxiety, and depressive symptoms in MELZACK R (Ed.). Pain Measurement and ment and disability in early rheumatoid arth- rheumatoid arthritis. J Rheumatol 1995; 22: Assessment. New York, Raven Press, 1983: ritis – development over 5 years. J Rheum - 514-20. 33-7. atol 1995; 22: 1037-42. 62. ROJKOVICH B, GIBSON T: Day and night 34. DIXON JS, BIRD HA: Reproducibility along a 50. LINDQVIST E, SAXNE T, GEBOREK P, EBER - pain measurement in rheumatoid arthritis. 10 cm vertical visual analogue scale. Ann HARDT K: Ten year outcome in a cohort of Ann Rheum Dis 1998; 57: 434-6. Rheum Dis 1981; 40: 87-9. patients with early rheumatoid arthritis: 63. S O K K A T, KANKAINEN A, HANNONEN P: 35. CARLSSON AM: Assessment of chronic pain. health status, disease process, and damage. Scores for functional disability in patients I. Aspects of the reliability and validity of the Ann Rheum Dis 2002; 61: 1055-9. with rheumatoid arthritis are correlated at visual analogue scale. P a i n 1983; 16: 87- 51. MULHERIN D, FITZGERALD O, BRESNIHAN higher levels with pain scores than with 101. B: Clinical improvement and radiological radiographic scores. Arthritis Rheum 2000; 36. PINCUS T, SUMMEY JA, SORACI SA JR, deterioration in rheumatoid arthritis: Evi- 43: 386-9. WALLSTON KA, HUMMON NP: Assessment dence that pathogenesis of synovial inflam- 64. SARZI-PUTTINI P, FIORINI T, PANNI B, of patient satisfaction in activities of daily mation and articular erosion may differ. Br J TURIEL M, CAZZOLA M, ATZENI F: Corre- living using a modified Stanford health as- Rheumatol 1996; 35: 1263-8. lation of the score for subjective pain with sessment questionnaire. A rthritis Rheum 52. CALLAHAN LF, PINCUS T, HUSTON JW, III, physical disability, clinical and radiographic

S-83 Assessment of pain in rheumatic diseases / T. Sokka

scores in recent onset rheumatoid arthritis. pain measures. J Rheumatol 1999; 26: 1785- the knee or hip: face validity of self-report BMC Musculoskeletal Disorders 2002; 3. 92. questionnaire ratings. J Rheumatol 2005; 32: 65. P I N A L S RS, MASI AT, LARSEN RA et al. : 74. S P E C TOR TD, DACRE JE, HARRIS PA , 533-9. Preliminary criteria for clinical remission in HUSKISSON EC: Radiological progression of 84. JONES AK, BROWN WD, FRISTON KJ, QI LY, rheumatoid arthritis. Arthritis Rheum 1981; osteoarthritis: an 11 year follow up study of FRACKOWIAK RSJ: Cortical and subcortical 24: 1308-15. the knee. Ann Rheum Dis 1992; 51: 1107-10. localization of response to pain in man using 66. SOKKA T, PINCUS T: Most patients receiv- 75. FELSON DT, ZHANG Y, HANNAN MT et al.: positron emission tomography. Proc R Soc ing routine care for rheumatoid arthritis in The incidence and natural history of knee Lond B 1991; 244: 39-44. 2001 did not meet inclusion criteria for most osteoarthritis in the elderly. The Framingham 85. MOUNTZ JD, ZHOU T, CHEN J: Use of sen- recent clinical trials or American College of Osteoarthritis Study. Arthritis Rheum 1995; sitive assays to detect soluble Fas in patients Rheumatology criteria for remission. J Rheu - 38: 1500-5. with systemic lupus erythematosus: Com- matol 2003; 30: 1138-46. 76. PEAT G, CROFT P, HAY E: Clinical assessment on the article by Knipping et al. and the 67. WELLS G, ANDERSON J, BOERS M et al. : ment of the osteoarthritis patient. Best Pract article by Goel et al. (letter). Arthritis Rheum MCID/Low Disease Activity State Wo r k- Res Clin Rheumatol 2001; 15: 527-44. 1996; 39: 1611-2. shop: summary, recommendations, and re- 77. ODDING E, VALKENBURG HA, A L G R A D , 86. CASEY KL: Match and mismatch: identify- search agenda. J Rheumatol 2003; 30: 1115- VANDENOUWELAND FA, GROBBEE DE, ing the neuronal determinants of pain. Ann 8. HOFMAN A: Associations of radiological os- Intern Med 1996; 124: 995-8. 68. FA L G A R O N E G, ZERKAK D, BAUER C , teoarthritis of the hip and knee with locomo- 87. DERBYSHIRE SWG: Imaging the brain in MESSOW M, DOUGADOS M: How to define tor disability in the Rotterdam Study. Ann pain. American Pain Society 1999; 9: 7-8. a Minimal Clinically Individual State Rheum Dis 1998; 57: 203-8. 88. KW I AT E K R, BARNDEN L, T E D M A N R e t (MCIS) with pain VAS in daily practice for 78. D I E P P E PA, CUSHNAGHAN J, SHEPSTO N E al.: Regional cerebral blood flow in fibro- patients suffering from musculoskeletal dis- L: The Bristol ‘OA500’study: progression of myalgia: Single-photon-emission computed orders. Clin Exp Rheumatol 2005; 23: 235-8. osteoarthritis (OA) over 3 years and the rela- tomography evidence of reduction in the 69. K R I S H N A N E, HAKKINIEN A, SOKKA T, tionship between clinical and radiographic pontine tegmentum and thalami. A rt h r i t i s HANNONEN P: Impact of age and comorbidi- changes at the knee joint. Osteoarthritis Car - Rheum 2000; 43: 2823-33. ties on the criteria for remission and response tilage 1997; 5: 87-97. 89. GRACELY RH, PETZKE F, WOLF JM, CLAUW in rheumatoid arthritis. Ann Rheum Dis 2005. 79. C R E A M E R P, LETHBRIDGE-CEJKU M, DJ: Functional magnetic resonance imaging 70. S A L A F F I F, CAVA L I E R I F, NOLLI M, FER- HOCHBERG MC: Factors associated with evidence of augmented pain processing in RACCIOLI G: Analysis of disability in knee functional impairment in symptomatic knee fibromyalgia. Arthritis Rheum (in press). osteoarthritis. Relationship with age and psy- osteoarthritis. Rheumatology 2000; 39: 490- 90. CALLAHAN LF, PINCUS T: The P-VA S / D - chological variables but not with radiograph- 6. ADL ratio: A clue from a self-report ques- ic score. J Rheumatol 1991; 18: 1581-6. 80. HANNAN MT, FELSON DT, PINCUS T: Ana- tionnaire to distinguish rheumatoid arthritis 71. JORDAN J, LUTA G, RENNER J, DRAGOMIR lysis of the discordance between radiograph- from noninflammatory diffuse musculoske- A, HOCHBERG M, FRYER J: Knee pain and ic changes and knee pain in osteoarthritis of letal pain. Arthritis Rheum 1990; 33: 1317- knee osteoarthritis severity in self-reported the knee. J Rheumatol 2000; 27: 1513-7. 22. task specific disability: the Johnston County 81. DE BOCK GH, VAN MARWIJK HWJ, KAPTEIN 91. DEWALT DA, REED GW, PINCUS T: Further Osteoarthritis Project. J Rheumatol 1997; 24: AA, MULDER JD: Osteoarthritis pain assess- clues to recognition of patients with fibro- 1344-9. ment in family practice. Arthritis Care Res myalgia from a simple 2-page patient multi- 72. VAN BAAR ME, DEKKER J, LEMMENS JAM, 1994; 7: 40-5. dimensional health assessment questionnaire O O S T E N D O R P R A B , B I J L S M A J W J: Pain 82. P I N C U S T, KOCH GG, SOKKA T et al.: A (MDHAQ). Clin Exp Rheumatol 2004; 22: and disability in patients with osteoarthritis randomized, double-blind, crossover clinical 453-61. of hip or knee: the relationship with articular, trial of diclofenac plus misoprostol versus 92. FORSETH KKO, GRAN JT: Management of kinesiological, and psychological character- acetaminophen in patients with osteoarthritis fibromyalgia: what are the best treatment istics. J Rheumatol 1998; 25: 125-33. of the hip or knee. Arthritis Rheum 2001; 44: choices ? Drugs 2002; 62: 577-92. 73. CREAMER P, LETHBRIDGE-CEJKU M, 1587-98. 93. PAPAGEORGIOU AC, SILMAN AJ, MACFAR- HOCHBERG MC: Determinants of pain sever- 83. PINCUS T, WANG X, CHUNG C, SOKKA T, LANE GJ: Chronic widespread pain in the ity in knee osteoarthritis: effect of demo- KOCH GG: Patient preference in a crossover population: a seven year follow up study. graphic and psychosocial variables using 3 clinical trial of patients with osteoarthritis of Ann Rheum Dis 2002; 61: 1071-4.

S-84