Hand Burn Injuries and Occupational Impairment: A Study on the Impact of Burn Injuries on Return-to-Work Outcomes from the Burn Model System Research Program

Abstract

Return to work (RTW) after burn injury is dependent on many variables, including type and location of burn injury, access to care, and preinjury mental and physical health. Noting that prior studies were limited by small sample sizes, we aimed to use a large database to explore the associations between hand burn severity, functional hand outcomes, and RTW postinjury. Data from a multicenter longitudinal study were analyzed. Adults with burn injuries were classified into 6 groups ranking in the severity of hand injury: (0) no hand burns, (1) single hand burn no grafting, (2) bilateral hand burn no grafting, (3) single hand burn requiring grafting, (4) bilateral hand burn requiring unilateral graft, and (5) bilateral hand burn requiring bilateral grafts. Grafting was used as a proxy for burn severity. Self-reported employment status, Patient-Reported Outcomes Measurement Information System (PROMIS) upper extremity (UE) scores, and reported requests for work accommodations were collected at discharge, 6-, 12-, and 24-month postinjury. Descriptive statistics and analysis of variance (ANOVA) with post-hoc Tukey test were completed to examine differences in outcomes by hand injury severity. A total of 4621 participants met the inclusion criteria. Group 5, those with the most severe burns, had significantly longer RTW times than groups 0-3 (P < 0.005). Group 5’s (bilateral burn/unilateral graft) average RTW was greater, however, not significantly, compared to group 4. At 6 months, the mean PROMIS UE scores for grafted groups (group 3, 40.6; group 5, 35.4) were significantly worse than non-grafted groups (group 1, 46.8; group 2, 45.0; P < 0.0001). At 12 and 24 months, mean PROMIS UE scores were worse for grafted groups, though differences were no longer significant compared to non-grafted groups. At every time point, the majority of respondents did not request accommodations for their injuries from their employers, regardless of severity. Burn severity plays a significant role in both RTW and hand function for participants with hand burns. In addition, the lack of correlation between burn severity and request for work accommodations hints at the baseline vulnerability of these populations. These findings suggest a need for systematic improvements in the way these patients are cared for and re-integrated into the workforce.

INTRODUCTION

Despite advances in the acute management and rehabilitation of people with major burns, these injuries continue to have long-term adverse consequences related to hypertrophic scarring, contractures, and psychosocial morbidity.^1,2 There is a decrease in health-related quality of life (HRQL) not only during early recovery but often long after the acute injury.³ In addition, burn injuries place people at risk for financial toxicity, difficulty with social reintegration, and challenging return to work (RTW). Return to work, specifically, is an important milestone for people with major burn injuries as it is often associated with lower financial stress, improved self-worth, and generally improved physical and mental health function.^4–7

Seminal works by Helm and Walker examined the causes of delayed RTW in people with burn injuries. Their reports suggest that burn size and anatomic distribution, percent of full-thickness burn injury, length of stay (LOS) for the index hospitalization, and the need for surgical wound closure were associated with longer time to RTW.^8,9 Functional use of the upper extremities, and more specifically the hands, is integral to most professions and trade work (a trade job is generally any job whose duties require advanced training and skills gained through means other than a bachelor’s degree). Therefore, it is valuable to understand the effect that significant burns of the hands have on one’s ability to RTW and how lower HRQL might be mitigated. Despite this need, few studies have examined the impact of burn location and severity on RTW for those working prior to injury.

To address this gap, we sought to determine the impact of hand burns on HRQL and timing of RTW. We hypothesized that people with more severe hand burns would have lower HRQL and longer time to RTW. We believe these findings will help clinicians counsel and provide rehabilitation for those with hand burns and improved RTW outcomes. Given limited vocational rehabilitation services that are offered at most burn centers across the country, understanding which patients may benefit most could improve the allocation of this important resource.^10,11

METHODS

Participants and procedures

Data from participants with and without hand burns enrolled in the Burn Model System-National Longitudinal Database (BMS-NLD) were included in this study. Institutional Review Board approval has been obtained for creation, participation, and use of data from the BMS-NLD database (STUDY00002340). The BMS-NLD is a federally funded research program that collects data prospectively from individuals with moderate to severe burn-related injuries.¹² Four BMS sites provided data for this study: Boston-Harvard Burn Injury Model System (https://www.bhbims.org); North Texas Burn Rehabilitation Model System (https://www.utsouthwestern.edu/education/medical-school/departments/physical-medicine/model-systems/ntbrms/); Northwest Regional Burn Model System (https://nwrbms.uw.edu); and Southern California Burn Model System (https://www.scbms.usc.edu). Data are collected using abstraction from the participants’ electronic medical records as well as patient-reported measures collected at hospital discharge (baseline data), and 6-, 12-, and 24-month postburn (follow-up assessments). Surveys are administered in English and Spanish. Current and past inclusion criteria for the BMS-NLD can be accessed online (https://burndata.washington.edu/standard-operating-procedures). For this study, BMS participants had to meet the current BMS enrollment criteria as outlined below and be of working age (≥ 18 years at time of injury).

• Age 18 years—64 years with 20% total body surface area (TBSA) burn size OR

• Age ≥ 65 years with ≥ 10%TBSA burn size OR

• Electrical high voltage/lightning injury OR

• Hand burn and/or face burn and/or feet burn AND

• Burn surgery for wound closure (surgery for closure of the burn wound must occur within 30 days of the initial burn injury) AND

• Received acute care treatment at one of the Burn Model System Centers for primary burn wound closure.

Though we recognize that electrical and lightning burns can potentially be more devastating than they initially appear due to subcutaneous propagation of the injury (thus larger true TBSA than what is externally seen) they were included as a part of this study as the decision to pursue surgery, ultimate surgical outcomes, and postoperative rehabilitation are not different than burns of other etiologies. For example, a deep partial-thickness or full-thickness thermal contact burn to the hand can result in the same devastating nerve/tendon/muscle injury as an electrical injury. All burn patients are debrided until healthy and bleeding tissue is encountered, the subsequent reconstructive options also remain the same. As discussed below, we use the need for/extent of grafting as a proxy for burn severity rather than TBSA or burn etiology. Thus, we felt it was reasonable to include electrical/lightning injuries in our study population.

Data from participants were divided into 6 groups based on injury severity (Table 1). Autografting to the participant’s hand(s) was used as the proxy for burn severity.

Table 1.

Injury Severity Groups

Group no.	0	1	2	3	4	5
Group description	No hand burn	Unilateral hand burn	Bilateral hand burns	Unilateral hand burn	Bilateral hand burns	Bilateral hand burns
Need for hand autografting	Not applicable	No	No	Yes; unilateral hand autografting	Yes; unilateral hand autografting	Yes; bilateral hand autografting

Participants in group 0 did not have hand burns, those in group 1 had unilateral burns not requiring autografting, participants in group 2 had bilateral hand burns not requiring autografting, participants in group 3 had a unilateral hand burn that required autografting, participants in group 4 had bilateral hand burns but only required unilateral hand autografting, and participants in group 5 had bilateral hand burns that required bilateral hand autografting.

Study instruments

Participant characteristic information

As part of the larger Burn Model System program of research, all participants provided written, informed consent for study participation. Study consent allowed for the review of the individual’s medical record and completion of patient-reported outcome measures. Demographic characteristics are reported by participant sex (male or female), race (black, white, or mixed/other), ethnicity (non-Hispanic, Hispanic), injury etiology (fire/flame, scald, grease, electrical, or other), and burn size (TBSA <10%, 10-19%, 20-39%, and ≥40%).

Study group participant characteristics were based on responses to RTW questions, PROMIS upper extremity surveys, and employment accommodation responses at any study follow-up timepoint. Sample size and mean and SD data are reported.

PROMIS upper extremity measure

The BMS-NLD study asks adult participants to complete the PROMIS-29 measure (healthmeasures.org). Within this survey is a set of 4 questions that address common actions one performs using the upper extremity. This 29-item measure has been validated with adult burn survivors.¹³ As with all PROMIS measures, a T-score of ≥ 50 correlates with better function. Responses to this measure were obtained at 6, 12, and 24 months postburn and addressed the ability to (1) cut food using utensils, (2) open a can with a hand can opener, (3) button a shirt, and (4) pick up coins from a tabletop.

RESULTS

Participant characteristics

Figure 1 outlines the study data acquisition. Data were obtained from BMS participants injured between January 1993 and August 2023. Table 2 outlines respondent demographics and characteristics.

Figure 1.

Data Acquisition Diagram

Table 2.

Participant characteristics

	All study respondents (n = 3,212)	RTW date respondents (n = 1,176)	PROMIS upper extremity respondents (n = 412)	Employment accommodation respondents (n = 280)
Age (M ± SD)	43.4 ± 15.8	39.5 ± 13.3	45.6 ± 16.4	42.4 ± 14.1
Sex (n, %) Male Female	2,346 (73%) 866 (27%)	916 (78%) 260 (22%)	287 (69%) 126 (31%)	195 (70%) 85 (30%)
Race (n, %) Black White Mixed/Other	376 (13%) 2,350 (81%) 190 (7%)	143 (14%) 850 (81%) 53 (5%)	26 (6%) 350 (87%) 25 (6%)	17 (6%) 232 (86%) 22 (8%)
Ethnicity (n, %) Hispanic Non-Hispanic	2,635 (86%) 430 (14%)	956 (84%) 182 (16%)	310 (78%) 89 (22%)	207 (78%) 59 (22%)
Injury etiology (n, %) Fire/flame Scald Grease Electrical Other	1,905 (60%) 337 (11%) 301 (9%) 220 (7%) 419 (13%)	661 (57%) 108 (9%) 140 (12%) 105 (9%) 147 (13%)	279 (68%) 19 (5%) 45 (11%) 28 (7%) 41 (10%)	149 (53%) 42 (15%) 45 (16%) 16 (6%) 27 (10%)
Burn size groups (n, %) <10% TBSA 10%-19% 20%-39% ≥40%	1,238 (39%) 764 (24%) 826 (26%) 383 (12%)	447 (38%) 280 (24%) 326 (28%) 122 (10%)	169 (41%) 82 (20%) 105 (26%) 55 (13%)	154 (55%) 50 (18%) 53 (19%) 22 (8%)
Burn size (M ± SD)	18.3 ± 16.8	18.0 ± 16.0	19.0 ± 17.8	13.6 ± 14.6
LOS (M ± SD)	26.6 ± 25.8	21.4 ± 19.8	29.3 ± 35.0	19.8 ± 27.0

There were 3212 participants (all study respondent group; Table 2) with an average age of 43.4 years (SD 15.8) with most identifying as male (73%). By race 81% identified as White, 13% as Black, and 7% as Mixed/Other. Ethnicity was primarily Hispanic (86%). Breaking down by etiology, most injuries were caused by flame (60%), followed by other (13%), scald (11%), grease (9%), and finally electrical (7%). The greatest proportion of respondents had burn size TBSA < 10% (39%) followed by 20%-39% (26%), then 10%-19% (24%), and finally ≥40% (12%). The number of participants who responded to RTW questions was 1176, compared to 2937 who responded to employment questions at 6 months postburn. There were 412 participants who responded to the PROMIS questions at any follow-up timepoint and 280 who responded to workplace accommodation questions.

Return to work

Of the 3212 patients with hand burns, 1176 submitted responses regarding time to RTW after sustaining the injury. ANOVA comparison of the 5 groups demonstrated a significant difference in the reported days to RTW (P < 0.0001, F = 6.21). Further post-hoc pairwise testing demonstrated significant differences in RTW between group 5 (mean 242.6 days) and the following groups: group 0 (no hand burn; mean 129.6 days, P < 0.005), group 1 (unilateral burn, no graft; mean 150.6 days, P < 0.005), group 2 (bilateral burn, no graft; mean 134.8 days, P < 0.005), and group 3 (unilateral burn, unilateral graft; mean 155.6 days, P < 0.005).

PROMIS upper extremity scores

266 responses were recorded at 6 months, 254 at 12 months, and 207 at 24 months. Surveys were administered in person at clinic follow-up or requested via email. Participants were contacted for 2 years postinjury asking for responses to the study surveys. At all time points, ANOVA comparison demonstrated a significant difference in scores amongst the groups (6 months P < 0.0001, F = 8.6; 12 months, P = 0.0016, F = 4.48; 24 months, P = 0.048, F = 2.44) (Table 3). Subsequent post-hoc pairwise analysis demonstrated that at 6 months, group 5 (bilateral burn, bilateral graft; mean score 35.4) had significantly worse scores than group 1 (single hand, no graft; mean score 46.8, P < 0.05), group 2 (bilateral burn, no graft; mean score 45.0, P < 0.05), and group 4 (bilateral burn, unilateral graft; mean score 41.6, P < 0.05). Post-hoc pairwise analysis at 12 months demonstrated that a significant difference in scores was only seen between group 5 (bilateral burn, bilateral graft; mean score 39.1) and group 1 (one hand, no graft; mean score 48.7, P < 0.05). Finally, post-hoc pairwise analysis at 24 months also demonstrated that there was only a significant difference between group 5 (bilateral burn, bilateral graft; mean score 49) and group 1 (unilateral burn, no graft; mean score 49.0, P < 0.05).

Table 3.

PROMIS Upper Extremity Mean Scores

Group # (M ± SD)	6 months (n = 266)	12 months (n = 254)	24 months (n = 207)
1	46.8 (±9.9)	48.7 (±8.6)	49.0 (±8.6)
2	45 (±12.0)	44.4 (±12.8)	43.5 (±12.4)
3	40.6 (±11.4)*	44.7 (±11.0)	45.1 (±10.5)
4	41.6 (±11.7)	44.0 (±11.1)	45.8 (±9.3)
5	35.4 (±12.8)*	39.1 (±13.6)*	41.5 (±12.1)*

* Significant difference with group 1 (P < 0.05).

Workplace accommodations

There were 144 participants with hand burns who provided responses to questions concerning requests for work accommodations (defined as modifications in job responsibilities to assist people with postinjury restrictions). When the 4 levels of accommodation were compared against the 6 hand burn severity groups (groups 1–5), there were no significant differences in levels of requested accommodation amongst the severity groups at any time point (P values: 6 months 0.26, 12 months 0.22, and 24 months 0.18).

DISCUSSION

This study was performed to determine the impact of hand burn severity on HRQL and RTW. In addition, we sought to determine whether accommodations were received in a manner commensurate with the severity of burn injury. We found that there were significant associations between hand burn severity, upper extremity physical function, and RTW. In general, people with more severe injuries (indicated by bilateral burns and/or burns requiring skin grafting) were significantly more likely to have lower PROMIS upper extremity scores and longer time to RTW than people with less severe burns. Importantly, people with more severe hand burn injuries were not necessarily given additional accommodations to facilitate RTW.

As one might expect, the severity of burn injury was negatively associated with PROMIS upper extremity scores. However, at 6 months, there were more group pairs with significant differences than at the 12- and 24-month time points. Specifically, there was a significant difference in PROMIS upper extremity scores between the bilateral burn/bilateral graft group (group 5) and the bilateral burn/no graft (group 2) at 6 months that did not persist at 12 and 24 months. This likely indicates that, in the short-term, more severe burns are correlated with worse functionality. With time, people’s function, accommodations, adaptations to their environment, and/or coping capabilities improve. This information and time scale can be used to provide assurance and motivation for people struggling to function successfully in their environment in the weeks and months after injury and acute reconstruction. Furthermore, this provides a time course for therapists and vocational rehabilitation counselors to use when developing and communicating work hardening and RTW accommodation schedules.

Existing literature on RTW after burn injury has generally demonstrated that burn size tends to be a better predictor of RTW rather the anatomic location(s) of injury. For example, a systematic review of RTW literature by Tanttula et al. demonstrated that RTW rates were not significantly different in patients with and without hand burns. However, they did find that time to RTW was significantly longer in patients with hand burns compared to those without hand burns.¹⁴ However, this may have been due to examining people with hand burn injuries as a monolith and not as a spectrum injury. Their review generally reflects what was seen in our data—patients with hand burns requiring grafting and/or bilateral burns were associated with longer RTW times. A limitation of our study is that we were unable to separate participants based on job type, physical intensity, and/or need for manual dexterity. Existing literature logically demonstrates that RTW is significantly longer for patients for physical labor-related jobs. Future directions may include a closer look at non-clinical variables that influence RTW, including type of job, patient’s preinjury workplace rank and experience, and relationship with employer. Each of these has been shown to be important for successful RTW after burn injury.¹⁵

Importantly, our data demonstrated that there was no difference in workplace accommodations provided to participants across the 6 severity groups. We defined an accommodation as a patient-driven request to modify job tasks to comply with postinjury restrictions. BMS-NLD data, however, does not specifically look at changes in job tasks. A change in job task is an employer-driven change that may not necessarily benefit the patient (eg, fewer hours, less complicated, more poorly compensated tasks). Tanttula et al. demonstrated that patients with hand burns frequently required changes in job tasks despite being able to return to their original employer.¹⁴ Thus, simply looking at RTW and requesting workplace accommodations may not be reflective of a participant’s postinjury vocational experience. The complexity of this experience calls for vocational rehabilitation resources within burn clinics to help injured workers appropriately navigate the postinjury labor market. Ultimately, we recognize that although it is a burn provider’s goal to return patients as close to their preinjury function as possible that in practice functional “success” can look different posinjury. This may mean that patients are better off finding different jobs postinjury. The incorporation of vocational counselors into burn clinics is important as they, in conjunction with the clinical and therapy team, can help patients reasonably fulfill their postinjury expectations.

The findings should be examined in light of several additional limitations. First, it is difficult to collect and categorize data on the severity of hand burn injury and method of reconstruction (eg, use of dermal template, local flaps for joint or tendon coverage). For example, a unilateral burn is considered less severe than a bilateral burn and burns requiring grafting are considered more severe than burns that did not require grafting. Although this is generally accurate, it may be an oversimplification of these injuries and incompletely describes the complexity of acute reconstruction. The BMS-NLD does not collect granular surgical data to improve the classification of these injuries. However, the categorization scheme used for this study is clinically logical and was associated with increasingly worse hand function. A standardized scoring system to grade hand burn injuries could reduce this potential measurement bias. Another limitation includes the fact that participants included in this study did not necessarily have isolated hand burns. Patients were included if they had hand burns (and met other inclusion criteria) though could have also had burns to other parts of their body. This potentially confounds data as it is unclear the degree of contribution the hand burn had compared to burns to other parts of the body. BMS-NLD data also lack objective measures of hand function (eg, goniometry, 2-point discrimination) against which to compare subjective patient-reported outcomes. Finally, the BMS-NLD is not granular enough to include information on how or why patients are cleared to RTW, thus it is not possible to develop a consistent RTW “criteria.” This, however, we feel is reflective of the fact that decision to RTW often cannot be consistent among patients with complex burns involving the hands. Decision to RTW is often dependent on the patient’s job tasks and whether or not the patient and therapist feel they are able to complete those tasks. It also depends on whether or not employers are willing to excuse these patients for necessary postdischarge clinical and therapy appointments.

CONCLUSION

Patient-reported HRQL and function measures and RTW are important benchmarks in the care and recovery of people with burn injuries. Allocating therapy and vocational rehabilitation resources in accordance with the severity of the injury and functional limitations may improve RTW and/or injured workers’ experiences during the RTW process. Furthermore, this study unveiled the disconnect between burn severity and request for work accommodations. Our data point to the need for further study into current strategies and resources for postburn vocational rehabilitation.

Author Contributions:

Nikhitha Thrikutam (Writing—original draft, Writing—review & editing [lead]), Charles Kopp (Writing—original draft, Writing—review & editing [equal]), Caitlin Orton (Conceptualization, Data curation, Investigation, Methodology, Project administration, Supervision [equal], Writing—review & editing [supporting]), Alyssa Bamer (Data curation, Formal analysis [lead], Methodology, Writing—original draft [equal]), Jeffrey Schneider (Conceptualization, Project administration, Supervision [equal]), Kyra Solis-Beach (Conceptualization, Data curation, Project administration, Supervision [equal]), Lewis Kazis (Conceptualization, Supervision [equal]), Haig Yenikomshian (Conceptualization, Project administration, Supervision [equal]), Karen Kowalske (Conceptualization, Project administration, Supervision [equal]), and Barclay Stewart (Conceptualization, Investigation, Methodology, Project administration, Supervision, Writing—review & editing [lead])

Funding:

The contents of this manuscript were developed under grants from the National Institute on Disability, Independent Living, and Rehabilitation Research (NIDILRR grants 90DPBU0005, 90DPBU0006, 90DPBU0007, 90DPBU0008, 90DPBU0009). NIDILRR is a Center within the Administration for Community Living (ACL), Department of Health and Human Services (HHS). The contents of this manuscript do not necessarily represent the policy of NIDILRR, ACL, HHS, and do not assume endorsement by the Federal Government.

Conflict Of Interest Statement:

There are no conflicts of interest for any authors.