Effectiveness of Patient-Centered Interventions on Falls in the Acute Care Setting: a Quantitative Systematic Review Protocol

SYSTEMATIC REVIEW PROTOCOL

Effectiveness of patient-centered interventions on falls in the acute care setting: a quantitative systematic review protocol

1 1 1 1 1,2 Donna Avanecean Dawn Calliste Teresita Contreras Yeogyeong Lim Aileen Fitzpatrick

1College of Health Professions, Pace University, New York, New York, USA, 2The Northeast Institute for Evidence Synthesis and Translation (NEST): a Joanna Briggs Institute Center of Excellence, Newark, New Jersey, USA

Review objective: The quantitative objective of this review is to identify the effectiveness of patient-centered interventions on fall rates in adult patients in the acute care setting. Keywords Acute care; fall prevention; falls; hospital; patient-centered care JBI Database System Rev Implement Rep 2017; 15(1): 55–65.

Background related to the natural course of the patient’s ill- 3,5 nintentional falls during hospitalization remain ness. The most common fall-related injuries an ongoing concern for healthcare institutions include soft tissue injuries, bone fractures, head U injuries, decreased mobility and independence, on a global level, despite implementation of various 4 improvement strategies. The World Health Organ- anxiety and death. The WHO estimates that ization (WHO) defines falls as an event that results in 424,000 individuals will die from unintentional or accidental falls globally and approximately 37.3 a person coming to rest inadvertently on the ground, 1 floor or other lower level.1 Although the incidence of million individuals will require medical attention. These estimates are probably under-representative of falls has been of heightened focus for many years and 5 numerous studies have been done evaluating differ- actual fall rates because not all falls are reported. ent approaches for falls prevention, fall rates con- Falls and fall-related injuries have a great impact 2 on patients, physically, mentally, socially and tinue to remain high in acute care settings. Recent 5 estimates of the incidence of falls in the United States emotionally. Unintentional falls may also have an during an acute care admission range from an aver- economic impact on healthcare institutions as a result age rate of 3.3 to 11.5 per 1000 patient-days.3 A of an increase in medical costs from additional treat- patient-day is a unit in a system of accounting by ment of fall injuries and an increase in the length of a healthcare facilities and heathcare planners and is hospital stay. Patients who fall have on average a 12.3 day longer length of stay and injuries from falls have defined as the number of days that patients are 5 hospitalized.3 The day of admission, but not the led to a 61 per cent increase in patient care costs. In day of discharge, is counted as a patient-day.3 2010, costs of unintentional falls per individual In the United States, up to 50 per cent of hospi- patient, not resulting in injury, were between talized patients are at risk for falls.4 Between 30 and US$1586 and 3500; those resulting in minor injuries were US$9996-13,316 and those resulting in serious 50 per cent of all falls that occur during a hospital- 3,6 7 ization will result in injury and falls are the leading injury were US$24,249-27,000. Wong et al. cause of sentinel events: an unanticipated event in a examined costs of falls in three Midwest hospitals healthcare setting resulting in death or serious in the United States and reported an increased length physical or psychological injury to a patient, not of stay of 6.3 days per patient resulting in an average cost of US$14,000. Assessment of individuals at risk for falls is para- Correspondence: Donna Avanecean, [email protected] mount for prevention. Various fall risk scales have There is no conflict of interest in this project. been developed to assist registered nurses in the DOI: 10.11124/JBISRIR-2016-002981 assessment of patients at risk for falls including

JBI Database of Systematic Reviews and Implementation Reports ß 2017 THE JOANNA BRIGGS INSTITUTE 55

the Hendrich II Fall Risk Model, Morse Fall Scale targeted to specific sub-groups of patients including and New York Presbyterian Hospital fall risk instru- patients with cognitive impairments for active man- ment.8 The available fall risk assessment tools have agement of post-operative complications, not just varying reliability and validity, and the majority are for a generalized adult population across multiple geared toward the assessment of the geriatric acute care settings.20 The authors concluded that fall patient.9 Only the Morse Fall Scale shows promise prevention programs that are more patient specific for its international applicability.10-18 and can be used in providing care for adults in In the hospitalized patient, risk factors leading to multiple acute care settings are needed to reduce fall unintentional falls can be divided into two rates.20 categories: intrinsic and extrinsic factors. Intrinsic Falls prevention evidence has changed and factors include physiological conditions such as low evolved over time with positive and negative studies vision, dizziness, incontinence, age, cognitive diffi- revealing that a one-size-fits-all approach is not the culties, balance and gait difficulties and polyphar- solution.24 Patient education has been identified in macy. Extrinsic factors refer to an individual’s the literature as having a potential benefit in the immediate environment including cluttered hospital reduction of falls. In an RCT, Haines et al.25 found rooms, loose electrical cords and spills.12 Patient that patient education programs that provided writ- falls risk scales tend to address only particular intrin- ten information as a part of a targeted multi-factorial sic and extrinsic factors, but do not adequately assess program prevented falls. In addition, other recom- the patients’ current fall risk status, subsequently mended strategies included the use of web-based highlighting a need for more patient-centered risk education programs and videos, utilization of vari- assessments and interventions. ous equipments such as gait belts or chair/bed alarms Environmental hazards in the acute care setting and interventions geared toward frequent toileting have also been identified with increased fall risk. have also been well documented in the literature.25 Several studies examined the usefulness of specially Conventional methods, such as the use of fall risk designed patient care rooms, low beds, flooring and assessment tools, have not been consistently effective safety alarms (bed and chair alarms), skid proof in reducing fall rates. Since in-patient falls can be socks and hourly checks.19-22 Various combinations caused by many intrinsic and extrinsic factors and of these factors have been incorporated in the fall patients who fall often have numerous fall risk assessment tools developed so far, but no single tool factors, it may be more beneficial to adopt a has been adopted universally.23 Institutions tend to multi-modal patient-centered approach to prevent- develop their own assessment tools, which are inves- ing falls, which is tailored to a patient’s needs and tigated in these institutions only, and thus have not incorporates multiple fall prevention interventions. been independently evaluated for validity and Patient-centered care replaces our current pro- reliability.23 vider-centered system with one that revolves around A systematic review of seven randomized con- the patient.26 Effective care is generally defined by or trolled trials (RCTs) conducted in three countries in consultation with patients rather than by pro- found that certain interventions may be effective in vider-dependent tools or standards.26 Although the reducing the incidence of falls in older adults in the phrase patient-centered care was coined several dec- acute care setting.20 Interventions identified ades ago, it entered the health policy lexicon in 2001, included the use of fall risk cards, short-term admin- when it was featured as one of the six aims for high- istration of vitamin D supplementation, one-on-one quality health care in the Institute of Medicine’s patient-centered education focusing on the individ- landmark report, Crossing the Quality Chasm.27 ual patient’s risk factors and preventative strategies, The Institute of Medicine defined patient-centered and targeted fall risk factor reduction intervention care as ‘‘providing care that is respectful of and that includes fall risk factor screening.20 The findings responsive to individual patient preferences, needs, of this review concluded that there may be an added and values and ensuring that patient values guide all benefit in multi-disciplinary, multi-factorial inter- clinical decisions.’’28(p.6) ventions focused on the systematic assessment and Patient-centered care includes: knowledge of treatment of identified risk factors.20 The findings shared responsibility between patients and care- further identified the need for better designed RCTs givers; communication approaches that allow

JBI Database of Systematic Reviews and Implementation Reports ß 2017 THE JOANNA BRIGGS INSTITUTE 56

patient access to information and achieve patient not specific to fall risk prevention, these studies understanding; consideration of patients’ individu- highlighted the importance of patient-centered care ality, values and needs; and focus on the use of on improving patient-specific outcomes. related population-based strategies to improve Challenges in evaluating patient-centered inter- appropriate use of health services.28 In accordance ventions include: lack of theoretical and conceptual with patient-centeredness as one of the six goals for a clarity, unexamined assumptions, lack of adequate 21st century healthcare system, the Patient Protec- control for patient characteristics and social con- tion and Affordable Care Act has mandated the use texts, modest correlations between survey and obser- of measures of the quality of care, public reporting vational measures, and overlap of patient-centered and performance payments that reflect this ambi- care with other constructs.32 Despite these chal- tious aim.29 The law repeatedly refers to patient- lenges, studies evaluating patient-centered target centeredness, patient satisfaction, patient experience interventions aimed at reducing falls have shown of care, patient engagement and shared decision positive outcomes.44,45 making in its provisions.30 Even when the law only An RCT by Ang et al. evaluated the effectiveness uses the more general term, ‘‘quality measures’’, of multiple targeted interventions in the reduction of patient-centered assessments will be required when fall rates in an acute care hospital in Singapore in these provisions are turned into regulations for patients identified as high risk for falls. Patients specific programs such as with Medicare’s Value- randomized to the intervention arm of the study Based Purchasing Program.30 In October 2007, the received targeted interventions based on their Hen- Centers for Medicare and Medicaid Services intro- drich II Fall Risk Model Score in addition to usual duced a regulation that was reinforced in 2008 care. Results of this study demonstrated a fall inci- stating that hospitals will no longer receive reim- dence rate of 1.5% (95% confidence interval [CI]: bursement for treating injuries related to falls of 0.9-2/6) in the control arm and 0.4% (95% CI: 0.2– patients admitted to hospitals.31 Therefore, health- 1.1) in the intervention arm. Furthermore, the inves- care institutions have a major financial motivation to tigators noted that there was an increased time to re-evaluate their fall prevention plans for hospital- first fall in the intervention group compared to the ized patients and to adopt and implement newer control group (hazard ratio ¼ 0.29).44 more effective methods that would reduce the Haines et al., in a RCT conducted in a sub-acute amount of falls from occurring. hospital setting in Sydney, Australia, examined the Multiple studies have shown that patient-centered effectiveness of targeted falls interventions on all care improves patient satisfaction, quality of care patients admitted. Targeted interventions were and health outcomes while reducing healthcare costs selected by nursing staff based on patients fall risk and disparities in healthcare.27,32-43 Providers prac- assessment. Potential interventions included fall risk ticing patient-centered care may improve their alert cards with information brochures, exercise patients’ clinical outcomes and satisfaction rates programs and hip protectors. Results of this study by improving the quality of the provider-patient demonstrated a 30% (P ¼ 0.045) reduction of first- relationship, while at the same time decreasing the time falls. This study further demonstrated a utilization of diagnostic testing, prescriptions, hos- reduction in recurrent fall events (relative risk pitalizations and referrals.26 Stewart et al.32 showed 0.78; 95% CI 0.56–1.06) in those individuals that expert opinion could not be correlated with randomized to the intervention arm.45 positive results, but patient-perceived patient-cen- An initial search of the JBI Database of System- tered care correlated with ‘‘better recovery from atic Reviews and Implementation Reports, the their discomfort and concern, better emotional Cochrane Database of Systematic Reviews, the health 2 months later, and fewer diagnostic tests Cumulative Index to Nursing and Allied Health and referrals.’’27(p.2) Bertakis and Azari34 concluded Literature (CINAHL) and PubMed was conducted that patient-centered care was associated with for existing reviews on fall risk prevention. A num- decreased utilization of healthcare services and ber of systematic reviews20,43 that evaluated the lower total annual charges.34 Reduced annual effectiveness of various fall intervention strategies, medical care charges may be an important outcome including educating nursing staff, implementation of medical visits that are patient-centered.32 While of fall risk assessment scales or falls risks and

JBI Database of Systematic Reviews and Implementation Reports ß 2017 THE JOANNA BRIGGS INSTITUTE 57

prevention were identified. However, none of these EBSCOhost, followed by analysis of the text words reviews looked specifically at patient-centered inter- contained in the title and abstract, and of the index ventions to reduce the risk of falls. Therefore, this terms used to describe the article. A second search review seeks to determine the effectiveness of using all identified keywords and index terms will patient-centered interventions on falls rates of adult then be undertaken across all included databases. patients hospitalized in acute care settings. Third, the reference list of all identified reports and articles will be searched for additional studies. Stud- Inclusion criteria ies published in English will be considered for Types of participants inclusion in this review. Studies published from The current review will consider studies that include the inception of databases to the present date will adult patients, aged 18 years and older, of any race, be considered for inclusion in this review. ethnicity or gender who are admitted to medical or The databases to be searched include: surgical acute care unit for any condition or illness. PubMed, CINAHL, Embase and Nursing/Academic edition. Types of intervention The search for unpublished studies will include: The current review will consider all studies that ProQuest Dissertations and Thesis, the New York include fall prevention interventions that are ident- Academy of Medicine and the Virginia Henderson ified as patient-centered. For the purpose of this Global Nursing e-Respository. review, patient-centered intervention is defined as Initial keywords to be used will be: any intervention geared toward a patient’s assessed acute care, hospital, patient-centered care, falls and individual needs, values and preferences. fall prevention.

Comparator Assessment of methodological quality The current review will consider studies that com- Papers selected for retrieval will be assessed by two pare patient-centered fall prevention interventions independent reviewers for methodological validity with usual care or general, non-patient-centered fall prior to inclusion in the review, using standardized prevention interventions. critical appraisal instruments from the Joanna Briggs Outcomes Institute Meta Analysis of Statistics Assessment and The current review will consider studies that Review Instrument (JBI-MAStARI) (Appendix I). include reported incidence of falls and fall rates Any disagreements that arise between the reviewers as measured by the number of falls during an acute will be resolved through discussion or with a care admission. third reviewer.

Types of studies Data extraction The current review will consider both experimental Data will be extracted from papers included in the and epidemiological study designs including RCTs, review using the standardized data extraction tool quasi-experimental studies, before and after studies, from JBI-MAStARI (Appendix II). The data prospective and retrospective cohort studies, case- extracted will include specific details about the control studies and analytical cross-sectional studies interventions, populations, study methods and out- for inclusion. In the absence of these, this review will comes of significance to the review question and also consider descriptive epidemiological study specific objectives. Reviewers will attempt to contact designs including case series, individual case reports study authors to seek missing data or provide clarity and descriptive cross-sectional studies for inclusion. in reported results.

Search strategy Data synthesis The search strategy aims to find both published and Quantitative data will, where possible, be pooled unpublished studies. A three-step search strategy will in statistical meta-analysis using JBI-MAStARI. All be utilized in this review. An initial limited search of results will be subject to double data entry. Effect MEDLINE and CINAHL will be undertaken using sizes are expressed as odds ratio (for categorical

JBI Database of Systematic Reviews and Implementation Reports ß 2017 THE JOANNA BRIGGS INSTITUTE 58

data), and their 95% CIs will be calculated for supplement-to-american-nurse-today-best-practices-for- analysis. Heterogeneity will be assessed statisti- falls-reduction-a-practical-guide/. [Cited March 2016]. cally using the standard chi-square and also 10. Hendrich AL, Bender PS, Nyhuis A. Validation of the Hen- explored using sub-group analyses based on the drich II fall risk model: a large concurrent case/control study different study designs included in this review. of hospitalized patients. Appl Nurs Res 2003;16(1):9–21. 11. Heinze C, Halfeins RJG, Roll S, Dassen T. Psychometric Where statistical pooling is not possible, the find- evaluation of the Hendrich II Fall Risk Model. J Adv Nurs ings will be presented in narrative form including 2006;53(3):327–32. tablesandfigurestoaidindatapresentationwhere 12. Currie LM, Mello LV, Cimino JJ, Bakken S. Development of appropriate. and representation of a fall-injury risk assessment in a clinical information system. Stud Health Technol Inform Acknowledgements 2004;107(Pt 1):721–5. The current review will partially fulfill degree 13. Morse JM, Morse RM, Tylko SJ. Development of a scale to identify the fall-prone patient. Can J Aging 1989;8(4):366–77. requirements for successful completion of the 14. Wagner LM, Scott V, Silver M. Current approaches to fall risk Doctor of Nursing Practice Program at Pace Univer- assessment in nursing homes. Geriatr Nurs 2011;32(4):238–44. sity, College of Health Professions, New York, New 15. Chow SK, Lai CK, Wong TK, Suen LK, Kong SKF, Chan CK, et al. York, for Donna Avanecean, Dawn Calliste, Tere- Evaluation of the Morse Fall Scale: applicability in Chinese sita Contreras, Yeogyeong Lim. hospital populations. Int J Nurs Stud 2007;44(4):556–65. 16. Baek S, Piao J, Jin Y, Lee SM. Validity of the Morse Fall Scale References implemented in an electronic medical record system. J Clin 1. World Health Organization. Media centre falls fact sheet no. Nurs 2014;23(17-18):2434–41. 344.2012; Available from: http://www.who.int/mediacentre/ 17. Tang WS, Chow YL, Koh SS. The inter-rater reliability of the factsheets/fs344/en. [Updated October 2012; cited Febru- modified Morse Fall Scale among patients 55 years old in ary 23, 2016]; [Internet]. an acute care hospital in Singapore. Int J Nurs Pract 2. Magaziner J, Hawkes W, Hebel R, Zimmerman S, Fox K, Dolan 2014;20(1):32–8. M, et al. Recovery from hip fracture in eight areas of function. J 18. Morse JM. The Modified Morse Fall Scale. Int J Nurs Pract Gerontol A Biol Sci Med Sci 2000;55A(9):498–507. 2006;12(3):174–5. 3. Bouldin ED, Andresen EM, Dunton NE, Simon M, Waters TM, 19. Spiva LA, Roberston B, Delk ML, Patrick S, Kimrey MM, Green Liu M, et al. Falls among adult patients hospitalized in the B, et al. Effectiveness of team training on fall prevention. United States: prevalence and trends. J Patient Saf Stud Health Technol Inform 2014;29(2):164–73. 2013;9(1):13–7. 20. Stern C, Jayasekara R. Interventions to reduce the incidence 4. Donaldson N, Brown DS, Aydin CE, Bolton MLB, Rutledge of falls in older adult patients in acute-care hospitals: a DN. Leveraging nurse-related dashboard benchmarks to systematic review. Int J Evid Based Healthc 2009;7:243–9. expedite performance improvement and document excel- 21. Haines TP, Hill AM, Hill KD, McPhail BS, Oliver D, Brauer S, lence. J Nurs Adm 2005;35(4):163–72. et al. Patient education to prevent falls among older hos- 5. Quigley P, White S. Hospital-based fall program measure- pitalized inpatients: a randomized control trial. Arch Intern ment and improvement in high reliability organizations. Med 2011;171(6):516–24. Online J Issues Nurs 2012;18(2):5. 22. Quigley P,Goff L. Current and emerging innovations to keep 6. Wu S, Keeler EB, Rubenstein LZ, Maglione MA, Shekelle PG. patients safe. Best practices for falls reduction: a practical A cost effectiveness analysis of a proposed national falls guide 2011. Available from: https://library.hill-rom.com/ prevention program. Clin Geriatr Med 2010;26(4):751–66. Supporting-Evidence/Fall-Prevention/Evidence-Based- 7. Wong CA, Recktenwald AJ, Jones ML, Waterman BM, Bollini Solutions/Current-and-emerging-innovations-to-keep- ML, Dunagan WC. The cost of serious fall related injuries in patients-safe-Technological-innovations-play-a-leading- three Midwestern hospitals. Jt Comm J Qual Patient Saf role-in-falls-prevention-programs/. [Cited February 23, 2011;37(2):81–7. 2016]; [Internet]. 8. National Database of Nursing Quality Indicators. Patient 23. Kelly A, Dowling M. Reducing the likelihood of falls in older fall module: data entry2012; Available from: https:// people. Nurs Stand 2004;18(49):33–40. www.2.kumc.edu/chalk/training/NDNQI/module_patient_ 24. Haines TP, Waldron NG. Translation of falls prevention falls_data_entry_01.asp. [Cited February 23, 2016]; [Internet]. knowledge into action in hospitals: what should be trans- 9. Kulik C. Components of a comprehensive fall-risk assess- lated and how should it be done? J Safety Res ment. Special supplement to American Nurse Today – best 2011;42(6):431–2. practices for falls reduction: a practical guide 2011:6–7; 25. Haines TP, Bell RAR, Varghese PN. Pragmatic, cluster Available from: http://americannursetoday.com/special- randomized trial of a policy to introduce low-low beds to

JBI Database of Systematic Reviews and Implementation Reports ß 2017 THE JOANNA BRIGGS INSTITUTE 59

hospital wards for the prevention of falls and fall injuries. J 36. Mead N, Bower P. Patient-centred consultations and out- Am Geriatr Soc 2010;58(3):435–41. comes in primary care: a review of the literature. Patient 26. Rickert J. Patient-centered care: what it means and how to Educ Couns 2002;48(1):51–61. get thereHealth Affairs Blog; 2012; Retrieved from: http:// 37. Fiscella K, Meldrum S, Franks P, Shields C, Duberstein P, healthaffairs.org/blog/2012/01/24/patient-centered-care- McDaniel S, et al. Patient trust: is it related to patient- what-it-means-and-how-to-get-there/. [Cited May 6, 2016]; centered behavior of primary care physicians? Med Care [Internet]. 2004;42(11):1049–55. 27. Epstein RM, Fiscella K, Lesser CS, Stange KC. Why the nation 38. Epstein RM, Franks P, Shields CG, Meldrum SC, Miller KN, needs a policy push on patient-centered health care. Health Campbell TL, et al. Patient-centered communication Aff 2010;29(8):1489–95. and diagnostic testing. Ann Fam Med 2005;3(5): 28. Institute of Medicine. Crossing the quality chasm: a new 415–21. health system for the 21st century Washington, DC: 39. Mallinger JB, Griggs JJ, Shields CG. Patient-centered care National Academy Press; 2001. and breast cancer survivors’ satisfaction with information. 29. U.S. Congress. Patient Protection and Affordable Care Act, Patient Educ Couns 2005;57(3):342–9. H.R. 3590. Public Law 111-148. 111th Cong. March 23, 2010. 40. Safran DG, Karp M, Coltin K, Chang H, Li A, Ogren J, et al. Retrieved from: https://www.congress.gov/111/plaws/ Measuring patients’ experiences with individual primary publ148/PLAW-111publ148.pdf [Cited May 6, 2016] care physicians: results of a statewide demonstration proj- 30. Millenson ML, Macri J. Will the Affordable Care Act move ect. J Gen Intern Med 2006;21(1):13–21. patient-centeredness to center stage? PsychEXTRA Dataset 41. Epstein RM, Street RL. Patient-centered communication in 2012:1–10. cancer care: promoting healing and reducing suffering 31. Inouye SK, Brown CJ, Tinnetti ME. Medicare nonpayment, Bethesda, MD: National Cancer Institute; 2007. hospital falls, and unintended consequences. N Engl J Med 42. Epstein RM, Shields CG, Franks P, Meldrum SC, Feldman M, 2009;360(23):2390–3. Kravitz RL. Exploring and validating patient concerns: 32. Stewart M, Brown JB, Donner A, McWhinney IR, Oates J, relation to prescribing for depression. Ann Fam Med Weston WW, et al. The impact of patient-centered care on 2007;5(1):21–8. outcomes. J Fam Pract 2000;49(9):796–804. 43. Rao JK, Anderson LA, Inui TS, Frankel RM. Communication 33. Epstein RM, Franks P, Fiscella K, Shields CG, Meldrum SG, interventions make a difference in conversations between Kravitz RL, et al. Measuring patient-centered communi- physicians and patients: a systematic review of the evi- cation in patient-physician consultations: theoretical and dence. Med Care 2007;45(4):340–9. practical issues. Soc Sci Med 2005;61(17):1516–28. 44. Ang E, Mordiffi SZ, Wong HB. Evaluating the use of a 34. Bertakis KD, Azari R. Patient-centered care is associated with targeted multiple intervention strategy in reducing patient decreased health care utilization. J Am Board Fam Med falls in an acute care hospital: a randomized controlled trial. 2011;24(3):229–39. J Adv Nurs 2011;67(9):1984–92. 35. Little P, Everitt H, Williamson I, Warner G, Moore M, Gould C, 45. Haines TP, Bennell KL, Osborne RH, Hill KD. Effectiveness of et al. Observational study of effect of patient centeredness targeted falls prevention programme in subacute hospital and positive approach on outcomes of general practice setting: randomised controlled trial. BMJ 2004;328(7441): consultations. BMJ 2001;323(7318):908–11. 676.

JBI Database of Systematic Reviews and Implementation Reports ß 2017 THE JOANNA BRIGGS INSTITUTE 60

Appendix I: Appraisal instruments MAStARI appraisal instrument

JBI Database of Systematic Reviews and Implementation Reports ß 2017 THE JOANNA BRIGGS INSTITUTE 61

JBI Database of Systematic Reviews and Implementation Reports ß 2017 THE JOANNA BRIGGS INSTITUTE 62

JBI Database of Systematic Reviews and Implementation Reports ß 2017 THE JOANNA BRIGGS INSTITUTE 63

Appendix II: Data extraction instruments MAStARI data extraction instrument

JBI Database of Systematic Reviews and Implementation Reports ß 2017 THE JOANNA BRIGGS INSTITUTE 64

JBI Database of Systematic Reviews and Implementation Reports ß 2017 THE JOANNA BRIGGS INSTITUTE 65