Table 3.

Description of studies identified in our systematic review of evidence

Author, Year	Region	Study objectives	Study design	Sample size and brief description	Length of observation	Racial/ethnic groups (n [%])*	Injury or illness type	Outcome	Quality
Asher AL, et al., 2017	United States of America (US)	To create a predictive model of patients’ ability to return to work (RTW) following lumbar spine surgery	Prospective cohort study	• 4,694 • Patients undergoing elective spine surgery for degenerative lumbar disease	Three months post initial surgery	• White (4,246 [91%]) • African American (302 [6.0%]) • Other race (146 [3.0%])	Non-occupational	• RTW	Medium
Ben-Shalom Y, Mamun AA., 2015	US	To study factors associated with achieving RTW milestones	Retrospective cohort study	• 417,238 • A representative proportion of disability insurance beneficiaries from 1996 to 2004	Five years after receipt of first disability insurance benefit	• White non-Hispanic (294,153 [71%]) • Black non-Hispanic (76,355[18%]) • Hispanic (28,372 [6.8%]) • Other or unknown race (17,941 [4.3%])	Non-occupational	• Enrollment in employment services • Start of trial work period • Trial work period completion • Suspension or termination of benefits due to RTW	High
Blinder V, et al., 2013 (a) Blinder V, et al., 2012 (b)	US	To identify early correlates of not returning to work (a) or not being employed (b) for low-income women treated for breast cancer	Prospective cohort study	• 274 (a) / 290 (b) • Women employed at time of a breast cancer diagnosis and were uninsured or underinsured and have a family income ≤200% of federal poverty level	Six, 18-, and 36- and 60-months following breast cancer diagnosis	• Latina (a:145 [53%]) (b:179 [62%]) • Non-Latina white (a:90 [33%] (b:111 [38%])	Non-occupational	• Not returning to work (a)/ not reporting employment (b)	Medium
Bradley CJ, Wilk A., 2014	US	To examine differences between African American and non-Hispanic white women in employment and change in work hours in women newly diagnosed with breast cancer who were initially employed and insured	Prospective cohort study	• 548 • Employed women diagnosed with breast cancer within two months of initiating treatment with intent to cure	Two- and nine-months following baseline interview	• Non-Hispanic white (429 [78%]) • African American (119 [22%])	Non-occupational	• Employment status • Hours worked/week	Medium
Busch MA, et al., 2009	United Kingdom (UK)	To investigate the frequency and determinants of return to paid work after stroke in a multi-ethnic urban population	Prospective cohort study	• 400 • Patients in a stroke registry with first ever stroke between January 1995 and December 2004 and working immediately before stroke	12 months following stroke	• White (231 [59%]) • Black (128 [33%]) • Other (31 [8.0%])	Non-occupational	• Employment status	Medium
Carrougher GJ, et al., 2020	US	To investigate the effect of patient and injury characteristics on employment for working-aged adult survivors of burn injury	Prospective cohort study	• 967 • Survivors of burn injury, ≥18 years of age or 18 years at follow-up, inpatient hospitalization of ≥ 3 days, known pre-burn employment status	12 months following burn injury	Race • White (654 [78%]) • Black (127 [15%]) • American Indian/Alaska Native (24 [2.9%]) • Asian (15 [1.8%]) • Other (9 [1.0%]) • Multiracial (6 [0.70%]) • Native Hawaiian/Pacific Islander (3 [0.40%]) Ethnicity • Hispanic/Latino (157 [17%]) • Non-Hispanic/non-Latino (761 [83%])	Non-occupational	• Employment status	Medium
Chibnall JT, Tait RC., 2009	US	To investigate sociodemographic, claim process and short-term adjustment predictors of long-term clinical adjustment among workers’ compensation claimants with low back pain	Prospective cohort study	• 374 • Workers’ compensation claimants with low back pain whose claims were settled in Missouri between January 1, 2001 and June 2, 2002	Six years post workers’ compensation settlement	• Caucasian (203 [61%]) • African American (171 [52%])	Occupational	• Employment status • Receipt of social security disability insurance	High
Chibnall JT, et al., 2005 (a) Tait RC, et al., 2004 (b)	US	To examine relationships among race, socioeconomic status, and post-settlement outcomes (a) and case management (b) of occupational back injuries	Cross-sectional study	• 1,472 • First-incident workers’ compensation claimants with low back injuries whose claims were settled between January 1, 2001 and June 1, 2002	Not reported	• Caucasian (892 [61%]) • African American (580 [39%])	Occupational	• Financial struggle (a) • Claim duration (injury to settlement) (b) • Receipt of temporary disability payment (b)	High
Friedman LS, Ruestow P, Frost L., 2012	US	To assess ethnic disparities in monetary compensation among construction workers injured on the job	Retrospective cohort study	• 1,039 • Construction workers who filed a workers’ compensation claim in Illinois between 2000 and 2005	Not reported	• White (724 [70%]) • Hispanic (168 [16%]) • Other (79 [7.6%] • Black (68 [6.5%])	Occupational	• Total workers’ compensation claim ($USD) • Mean temporary work disability (weeks)	High
Khan I, et al., 2019	US	To investigate the factors associated with RTW in patients who achieved otherwise favorable outcomes after lumbar spine surgery	Retrospective cohort study	• 12,435 • Patients undergoing lumbar surgery and were employed prior to surgery and completed 12-month follow-up	12 months post-surgery	• Caucasian (11,016 [90%]) • African American (886 [7.2%]) • Other race (353 [2.9%])	Non-occupational	• RTW	High
MacKenzie EJ, et al., 2006	US	To understand the factors that influence RTW after major limb trauma	Prospective cohort study	• 432 • Patients admitted to one of eight level 1 trauma centres for treatment of a severe lower extremity injury • English or Spanish speaking and aged 18–69 years	Three, six, 12-, 24- and 84-months post-injury	• White† • Non-white†	Non-occupational	• RTW	Medium
Marom BS, et al., 2020 (a) Marom BS, et al., 2018 (b)	Israel	To determine the time to RTW and to examine the effect of ethnicity and other prognostic variables on RTW among male manual workers who had experienced a hand injury	Prospective cohort study	• 178 • Male participants, aged 22–65 years who sustained acute hand injury below the elbow, either during or outside of working hours • Manual workers prior to hand injury and were referred by a physician to one of seven occupational therapy clinics • Read or write in Arabic or Hebrew language	12 months post-injury	• Jewish (88 [50%]) • Arab (90 [51%])	Non-occupational	• Time to RTW (a) • Employment status (b)	Medium
Meade MA, et al., 2004	US	To examine issues of employment and race for persons with spinal cord injury	Retrospective cohort study	• 5,925 • Participants who experienced a spinal cord injury between 1972–2002 and part of a larger database • Classified as African American or white • Ages 18–65 years at time of injury	One-year post-injury (multivariable model) Five, ten, 15-, and 20-years post-injury (descriptive analysis)	• White (4,210 [71%]) • African American (1,715 [29%])	Non-occupational	• Employment status	High
Pham TN, et al., 2020	US	To examine the association of extremity contractures with employment after burn injury	Retrospective cohort study	• 1,203 • Participants aged 18–64 years identified in a burn database (1994–2003) who were working prior to injury	Six-, 12- and 24-month post-injury	• Caucasian† • Non-Caucasian†	Non-occupational	• RTW	Medium
Sanchez KM, Richardson JL, Mason HRC, 2005	US	To assess the role of person, disease, and work-related factors on a timely return to the workplace among employed colorectal cancer survivors who sustained employment five years after initial cancer diagnosis	Cross-sectional study	• 142 • Participants who survived a colorectal cancer diagnosis between 1994 and 1995 were randomly selected from two population-based cancer registries • Speak and read English or Spanish • Ages 30–59 years at the survey		• White (104 [73%]) • Black (22 [16%]) • Hispanic (16 [14%])	Non-occupational	• Delayed RTW (> 2 months following diagnosis)	High
Savitsky B, et al., 2020 (a) Savitsky B, et al., 2020 (b)	Israel	To examine socioeconomic and other predictors of RTW (a) and duration to RTW (b)	Retrospective cohort study	• 44,740 (a) / 45,291(a) • Resident to Israel and aged 21–67 years • Injured and hospitalized between January 1, 2008 and December 31st 2013 • Employed 2 months prior to injury as a salaried worker•	One month, one year and 2 years post-injury	• Other Israelis (26,858 [59.3%]) • Israeli Arabs (11,232 [24.8%]) • Immigrants from former Soviet Union (6,341 [14%]) • Ethiopian Immigrants (679 [1.5%]) • Missing (181[0.4%])	Occupational and non-occupational	• Non-RTW	High
Skolarus LE, et al., 2016	US	To compare the proportion of stroke survivors who returned to work within 90 days following their stroke by ethnicity and investigate the role of sociodemographic factors and stroke severity	Prospective cohort study	• 125 • Stroke patients from a larger population-based surveillance study stroke survivors (August 2011-December 2013) who were employed at the time of their stroke	90 days post-stroke	• Mexican Americans (71 [57%]) • Non-Hispanic white (54 [43%])	Non-occupational	• RTW	Medium
Strong LL, Zimmerman FJ, 2005	US	To examine the relationship between race/ethnicity and the number of workdays missed owing to an injury or illness that caused a participant to miss work according to gender	Prospective cohort study	• 35,710 • Cohort drawn from ten waves of data (years 1988–2000) from a nationally representative population-level dataset	Not reported	• Non-Hispanic white (20,377 [57%]) • Hispanic (5,387 [15%]) • African American (9,946 [28%])	Non-occupational	• Workdays missed due to injury or illness	High
Tait RC, Chibnall JT, 2000	US	To examine medical and psychosocial factors associated with work injury management decisions for patients with low back pain	Retrospective cohort study	• 132 • Workers’ compensation claimants with refractory low back injuries whose impairment was physician assigned	Not reported	• Caucasian (84 [64%]) • African American (47 [36%])	Occupational	• Temporary total disability costs related to excused time off work	High

Notes: * = Racial groups presented based on the study author’s classification; † = author did not report a breakdown of racial groups included in their study