Abstract
BACKGROUND:
Needlestick injuries pose significant health hazards; however, the nationwide frequency of needlesticks and reporting practices among surgical residents are unknown. The objectives of this study were to examine the rate and circumstances of self-reported needlestick events in US surgery residents, assess factors associated with needlestick injuries, evaluate reporting practices, and identify reporting barriers.
STUDY DESIGN:
A survey administered after the American Board of Surgery In-Training Examination (January 2017) asked surgical residents how many times they experienced a needlestick during the last 6 months, circumstances of the most recent event, and reporting practices and barriers. Factors associated with needlestick events were examined using multivariable hierarchical regression models.
RESULTS:
Among 7,395 resident survey respondents from all 260 US general surgery residency programs (99.3% response rate), 27.7% (n = 2,051) noted experiencing a needlestick event in the last 6 months. Most events occurred in the operating room (77.5%) and involved residents sticking themselves (76.2%), mostly with solid needles (84.7%). Self-reported factors underlying needlestick events included residents’ own carelessness (48.8%) and feeling rushed (31.3%). Resident-level factors associated with self-reported needlestick events included senior residents (PGY5 29.9% vs PGY1 22.4%; odds ratio 1.66; 95% CI 1.41 to 1.96), female sex (31.9% vs male 25.2%; odds ratio 1.31; 95% CI 1.18 to 1.46), or frequently working more than 80 hours per week (odds ratio 1.42; 95% CI 1.20 to 1.68). More than one-fourth (28.7%) of residents did not report the needlestick event to employee health.
CONCLUSIONS:
In this comprehensive national survey of surgical residents, needlesticks occurred frequently. Many needlestick events were not reported and numerous reporting barriers exist. These findings offer guidance in identifying opportunities to reduce needlesticks and encourage reporting of these potentially preventable injuries among trainees.
Needlestick injuries have potentially significant consequences in healthcare workers due to exposure to bloodborne pathogens.1–3 Surgeons and surgical residents, in particular, are known to be at high risk for needlestick injuries because they perform invasive procedures frequently.4–12 Beyond potentially exposing healthcare workers to infectious diseases (eg HIV, hepatitis B, and hepatitis C),1–3,6,13,14 the event itself can cause substantial distress and can contribute to attrition.15,16
Needlestick events among trainees in healthcare are common; a majority of medical students and surgical residents have experienced a needlestick during their training.1,3,7,8,11,17,18 However, comprehensive data on needlestick events in surgical trainees are lacking. Earlier studies were mostly limited to a small number of institutions and/or had low survey response rates, with the largest study (699 residents from 19 centers) performed more than a decade ago.8 Low rates of reporting needlestick injuries to employee health complicate the issue.1,8,17,19–23 Early reporting can offer the opportunity to rapidly initiate prophylaxis or treatment, and ensures that healthcare needs and expenses are covered by the employer.24,25
Substantial knowledge gaps remain at a national level in the US concerning the frequency and circumstances of needlestick events in surgical residents, factors associated with the events, reporting rates, and barriers to reporting. In addition, since the last large study on needlesticks was published there have been changes in duty hour policies and a focus on resident wellness; it is unknown whether these contemporary issues have any associations with needlestick events and/or reporting of needlesticks. Using a national survey of all residents at the 260 ACGME-accredited general surgery programs in the US, our objectives were to examine the rate of self-reported needlestick events in surgery residents, assess resident- and program-level factors associated with needlestick injuries, understand the circumstances of the needlestick events, evaluate reporting practices of needlestick injuries among surgery residents, and identify barriers to reporting.
METHODS
Data source and study population
The American Board of Surgery In-Training Examination (ABSITE) is administered annually in January to all residents in ACGME-accredited general surgery programs in the US (n = 260 at the time of this study). In partnership with the American Board of Surgery (ABS), all residents taking the January 2017 ABSITE were offered an optional closed-ended survey at the end of the examination as a part of the FIRST (Flexibility in Duty Hour Requirements for Surgical Trainees) trial.26,27 Residents were included in the study population if they were clinically active and completed at least 1 question pertaining to needlesticks. Those residents who were not clinically active and those who did not answer any needlestick questions were excluded. Resident responses were collected by the ABS and de-identified before analysis. Resident- and program-level characteristics were provided by the ABS. The Northwestern University IRB office determined that this study constituted nonhuman subjects research as a part of the FIRST trial.26,27
Survey instrument
The survey included questions on duty hour violations and flexibility, resident well-being, and safety-related events, such as needlesticks. Residents were specifically asked how many times they experienced a needlestick in the last 6 months (0, 1, 2, 3 to 5, or more than 5 times), which represented the current academic year at the time of the survey, to minimize recall bias. The responses were dichotomized for the analyses as 0 times vs 1 or more times. Those residents who reported experiencing a needlestick were then asked yes/no questions about the circumstances of their most recent needlestick and to select from a list of potential factors that played a role in the needlestick event. All respondents were asked about perceived barriers to reporting needlesticks to employee health. Residents were allowed to select multiple options. Additional survey items covered violations of duty hour regulations, time for rest, and fatigue. The General Health Questionnaire-12 (GHQ-12), an instrument used to evaluate psychiatric well-being and potential for psychiatric disorders, including depression; anxiety; and social dysfunction, was included to understand the association between residents’ psychiatric well-being and their personal safety.28–32 Survey items were adapted from previously published surveys and review of the literature, pretested with residents using cognitive interviews, and iteratively revised and retested before inclusion.26,27
Statistical analyses
Cluster-weighted chi-square tests with clustering by residency program were used to assess differences in the rates of needlestick events, circumstances of the events, and reporting the event to employee health by resident and program characteristics. Separate multivariable hierarchical logistic regression models with random intercepts for residency program plus fixed effects for resident and program characteristics examined the association between these characteristics and (a) experiencing a needlestick in the last 6 months, and (b) reporting it to employee health.
Resident characteristics examined included PGY, sex, separate variables for duty hour violations of ACGME duty hour policies (ie 3 or more times in the most recent month of a general surgery rotation of working more than 28 hours continuously, having fewer than 8 hours off between regular shifts 3 or more times, or working more than 80 hours in a week), and quartile of the GHQ-12 score. In accordance with earlier studies of healthcare professionals, poor psychiatric well-being was defined as a GHQ-12 score ≥4.28–30 Program characteristics examined included program type, geographic region, and program size (dichotomized to ≤5 vs ≥6 categorical residents per PGY). Differences in the rates of reporting the most recent needlestick to employee health were examined based on resident self-reported factors that played a role in the needlestick and perceived reporting barriers to determine whether these variables were associated with reporting. All descriptive analyses included cluster-adjusted chi-square tests.
Finally, program-level rates of needlestick events among residents were calculated and plotted to assess variation across training programs. The level of significance was set at 0.05. Analyses were performed using Stata Software, version 14 (Stata Corp).
RESULTS
The total number of residents taking the 2017 ABSITE was 8,227. There were 786 residents who were not clinically active during the administration of the ABSITE, 46 who did not answer the clinically active question on the survey, and 4 residents who failed to answer any needlestick questions. Our study sample thus included 7,391 clinically active residents representing all 260 ACGME-accredited general surgery residency programs in the US. The response rate was 99.3%. Additional characteristics of the study cohort are described in Table 1.
Table 1.
Resident and Program Characteristics
| Characteristic | Total (n = 7,391) | |
|---|---|---|
| n | % | |
| Experienced needlestick event in the last 6 mo | 2,051 | 27.7 |
| PGY | ||
| 1 | 2,127 | 28.8 |
| 2 | 1,563 | 21.2 |
| 3 | 1,303 | 17.6 |
| 4 | 1,224 | 16.6 |
| 5 | 1,174 | 15.9 |
| Resident sex | ||
| Male | 4,528 | 61.3 |
| Female | 2,829 | 38.3 |
| Duty hour experience in most recent month of general surgery rotation | ||
| Worked >28 h continuously ≥3 times | 1,022 | 13.8 |
| Had <8 h off between regular shifts ≥3 times | 801 | 10.8 |
| Exceeded 80 h in a week ≥3 times | 1,120 | 15.2 |
| General Health Questionnaire-12 score quartile | ||
| First quartile (score range 0–1) | 2,158 | 29.2 |
| Second quartile of GHQ-12 scale (score range 2–3) | 1,959 | 26.5 |
| Third quartile of GHQ-12 scale (score range 4–5) | 1,431 | 19.4 |
| Fourth quartile of GHQ-12 scale (score range 6–12) | 1,839 | 24.9 |
| Program type | ||
| Academic | 4,448 | 60.2 |
| Community | 2,727 | 36.9 |
| Military | 216 | 2.9 |
| Geographic region | ||
| Northeast | 2,473 | 33.5 |
| Southeast | 1,425 | 19.3 |
| Midwest | 1,597 | 21.6 |
| Southwest | 852 | 11.5 |
| West | 1,044 | 14.1 |
| No. of PGY5 residents in program | ||
| 0–5 | 4,010 | 54.3 |
| 6 or more | 3,381 | 45.7 |
Duty hour experiences listed are not mutually exclusive categories. Each is asked as a separate question in the resident survey. Percentages might not sum to exactly 100% due to rounding or missing data (missing resident sex from 34 respondents (n = 7,357) and General Health Questionnaire-12 score is missing for 4 respondents (n =7,387).
GHQ, General Health Questionnaire.
Self-reported needlestick event rate and factors associated with needlestick injuries
Of the study sample, 27.7% of surgical residents (n = 2,051) self-reported experiencing a needlestick within the 6 months before the survey (ie current academic year). When we examined resident- and program-level factors associated with needlestick events (Table 2), we found that senior residents were more likely to experience a needlestick compared with PGY1 residents (PGY1 22.4% vs PGY5 29.9%; odds ratio [OR] 1.66; 95% CI 1.41 to 1.96). Women were more likely to experience a needlestick compared with men (women 31.9% vs men 25.2%; OR 1.31; 95% CI 1.18 to 1.46), and this pattern was similar at all PGY levels. Repeated violation of ACGME duty hour restrictions (working more than 28 hours continuously, having fewer than 8 hours off between shifts, or working more than 80 hours in a week 3 or more times during the most recent month on a general surgery rotation) was associated with greater likelihood of self-reporting a needlestick for all 3 types of violations (all, p < 0.05) (Table 2). Overall, 44.3% of residents were found to have poor psychiatric well-being (GHQ-12 scores ≥4). Residents with GHQ-12 scores of 6 to 12 (highest quartile) were more likely to experience a needlestick injury (31.2% vs GHQ-12 score 0 to 1 [lowest quartile] 23.0%; OR 1.33; 95% CI 1.15 to 1.54). At the program level, resident-reported needlestick rates ranged from 0% to 62.5%; only 2 residency programs had no resident-reported events during the past 6 months, and 8 programs had ≥50% of their residents self-reporting experiencing a needlestick (Fig. 1).
Table 2.
Factors Associated with Experiencing a Needlestick Event Within the Last 6 Months
| Variable | Experienced nee- dlestick event (n = 7,391) |
p Value* | Adjusted odds ratio† for experiencing needlestick event (95% CI) |
|
|---|---|---|---|---|
| n | % | |||
| All residents | 2,051 | 27.7 | ||
| PGY | <0.001 | |||
| 1 | 477 | 22.4 | 1.00 (ref) | |
| 2 | 448 | 28.7 | 1.41 (1.21 — 1.64)‡ | |
| 3 | 389 | 29.9 | 1.52 (1.29—1.79)‡ | |
| 4 | 386 | 31.5 | 1.71 (1.45—2.01)‡ | |
| 5 | 351 | 29.9 | 1.66 (1.41—1.96)‡ | |
| Resident sex | <0.001 | |||
| Male | 1,141 | 25.2 | 1.00 (ref) | |
| Female | 901 | 31.9 | 1.31 (1.18—1.46)‡ | |
| Duty hour experience in most recent month of general surgery rotation | ||||
| Worked >28 h 0—2 times | 1,675 | 26.3 | <0.001 | 1.00 (ref) |
| Worked >28 h 3 or more times | 376 | 36.8 | 1.24 (1.05—1.47)§ | |
| Had <8 h off between shifts 0—2 times | 1,739 | 26.4 | <0.001 | 1.00 (ref) |
| Had <8 h off between shifts ≥3 times | 312 | 39.0 | 1.34 (1.12—1.61)∥ | |
| Worked >80 h in a week 0—2 times | 1,624 | 25.9 | <0.001 | 1.00 (ref) |
| Worked >80 h in a week ≥3 times | 427 | 38.1 | 1.42 (1.20—1.68)‡ | |
| General Health Questionnaire-12 score quartile | <0.001 | |||
| First quartile (score range 0—1) | 496 | 23.0 | 1.00 (ref) | |
| Second quartile of GHQ-12 scale (score range 2—3) | 525 | 26.8 | 1.17 (1.01 —1.35)§ | |
| Third quartile of GHQ-12 scale (score range 4—5) | 455 | 31.8 | 1.40 (1.20—1.64)‡ | |
| Fourth quartile of GHQ-12 scale (score range 6—12) | 574 | 31.2 | 1.33 (1.15—1.54)‡ | |
| Program type | 0.005 | |||
| Academic | 1,294 | 29.1 | 1.00 (ref) | |
| Community | 715 | 26.2 | 0.87 (0.75—1.00) | |
| Military | 42 | 19.4 | 0.59 (0.40—0.89)§ | |
| Geographic region | 0.89 | |||
| Northeast | 700 | 28.3 | 1.00 (ref) | |
| Southeast | 380 | 26.7 | 0.95 (0.80—1.13) | |
| Midwest | 436 | 27.3 | 0.97 (0.82—1.15) | |
| Southwest | 242 | 28.4 | 1.04 (0.84—1.28) | |
| West | 293 | 28.1 | 1.01 (0.84—1.23) | |
| No. of PGY5 residents in program | 0.16 | |||
| 0—5 | 1,080 | 26.9 | 1.00 (ref) | |
| ≥6 | 971 | 28.7 | 1.00 (0.87—1.14) | |
Bivariate p values are from cluster-corrected chi-square tests of differences in needlestick rates across categories of each variable with responses clustered by residency program.
Adjusted odds ratios are from a hierarchical logistic regression model containing all variables displayed and with random intercepts for residency program.
p < 0.001.
p < 0.5.
p < 0.01.
GHQ, General Health Questionnaire.
Figure 1.
Variation in program-level rates of surgical resident needlestick events during the most recent 6-month period (representing the current academic year).
Circumstances and self-reported factors playing a role in needlestick events
When asked about the circumstances of their most recent needlestick injury (Table 3), residents reported that most occurred in the operating room (77.5%; n = 1,590), commonly involved sticking themselves (76.2%; n = 1,563), and frequently involved solid needles (84.7%; n = 1,738). PGY1 residents were more likely to answer that they stuck themselves than PGY5 residents (p = 0.01). There were no sex differences in self-reported circumstances of the needlestick event.
Table 3.
Resident-Reported Circumstances and Factors in Most Recent Needlestick Event and Perceived Barriers to Reporting by PGY and Sex
| PGY | Resident sex | |||||||||||||||||
|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|
| Total (n = 2,051) |
1 (n = 477) |
2 (n = 448) |
3 (n = 389) |
4 (n = 386) |
5 (n = 351) |
Male (n = 1141) |
Female (n = 901) |
|||||||||||
| Variable | n | % | n | % | n | % | n | % | n | % | n | % | p Value* | n | % | n | % | p Value* |
| Circumstance of most recent needlestick incident | ||||||||||||||||||
| Stick occurred in operating room | 1,590 | 77.5 | 326 | 68.3 | 303 | 67.6 | 311 | 80.0 | 332 | 86.0 | 318 | 90.6 | <0.001 | 880 | 77.1 | 703 | 78.0 | 0.67 |
| Stuck oneself | 1,563 | 76.2 | 380 | 79.7 | 353 | 78.8 | 299 | 76.9 | 285 | 73.8 | 246 | 70.1 | 0.01 | 852 | 74.7 | 705 | 78.3 | 0.06 |
| Stick occurred during a call shift | 752 | 36.7 | 141 | 29.6 | 180 | 40.2 | 166 | 42.7 | 153 | 39.6 | 112 | 31.9 | 0.001 | 401 | 35.1 | 346 | 38.4 | 0.19 |
| Stuck with a hollow-bore needle | 313 | 15.3 | 84 | 17.6 | 76 | 17.0 | 64 | 16.5 | 48 | 12.4 | 41 | 11.7 | 0.13 | 169 | 14.8 | 142 | 15.8 | 0.60 |
| Factor that played a role | ||||||||||||||||||
| Personal inexperience played a role | 431 | 21.0 | 229 | 48.0 | 106 | 23.7 | 53 | 13.6 | 29 | 7.5 | 14 | 4.0 | <0.001 | 262 | 23.0 | 168 | 18.7 | 0.02 |
| My fatigue played a role | 439 | 21.4 | 97 | 20.3 | 97 | 21.7 | 117 | 30.1 | 71 | 18.4 | 57 | 16.2 | <0.001 | 220 | 19.3 | 215 | 23.9 | 0.02 |
| I felt rushed | 641 | 31.3 | 187 | 39.2 | 163 | 36.4 | 123 | 31.6 | 100 | 25.9 | 68 | 19.4 | <0.001 | 337 | 29.5 | 300 | 33.3 | 0.11 |
| My carelessness played a role | 1,001 | 48.8 | 222 | 46.5 | 234 | 52.2 | 200 | 51.4 | 187 | 48.5 | 158 | 45.0 | 0.19 | 572 | 50.1 | 424 | 47.1 | 0.17 |
| Carelessness of another played a role | 448 | 21.8 | 89 | 18.7 | 84 | 18.8 | 80 | 20.6 | 95 | 24.6 | 100 | 28.5 | 0.004 | 242 | 21.2 | 204 | 22.6 | 0.45 |
| Could not have been prevented | 306 | 14.9 | 58 | 12.2 | 58 | 13.0 | 61 | 15.7 | 66 | 17.1 | 63 | 18.0 | 0.11 | 166 | 14.6 | 140 | 15.5 | 0.56 |
| Perceived barrier to needlestick reporting | ||||||||||||||||||
| Takes too much time | 1,647 | 80.3 | 368 | 77.2 | 351 | 78.4 | 320 | 82.3 | 316 | 81.9 | 292 | 83.2 | 0.17 | 891 | 78.1 | 750 | 83.2 | 0.007 |
| Too much of a disruption to operative case | 1,073 | 52.3 | 231 | 48.4 | 200 | 44.6 | 222 | 57.1 | 232 | 60.1 | 188 | 53.6 | <0.001 | 562 | 49.3 | 508 | 56.4 | 0.005 |
| Do not know how to report event | 229 | 11.2 | 81 | 17.0 | 49 | 10.9 | 37 | 9.5 | 31 | 8.0 | 31 | 8.8 | <0.001 | 127 | 11.1 | 100 | 11.1 | 0.98 |
| No need to report because infection risk is low | 443 | 21.6 | 97 | 20.3 | 101 | 22.5 | 92 | 23.7 | 88 | 22.8 | 65 | 18.5 | 0.41 | 281 | 24.6 | 161 | 17.9 | <0.001 |
| There is pressure not to report | 153 | 7.5 | 43 | 9.0 | 29 | 6.5 | 34 | 8.7 | 28 | 7.3 | 19 | 5.4 | 0.35 | 75 | 6.6 | 77 | 8.6 | 0.13 |
| Negative stigma associated with needlestick event | 523 | 25.5 | 153 | 32.1 | 128 | 28.6 | 98 | 25.2 | 83 | 21.5 | 61 | 17.4 | <0.001 | 295 | 25.9 | 227 | 25.2 | 0.73 |
| Do not want to know test result | 140 | 6.8 | 32 | 6.7 | 32 | 7.1 | 31 | 8.0 | 27 | 7.0 | 18 | 5.1 | 0.65 | 97 | 8.5 | 42 | 4.7 | <0.001 |
p Values are from cluster-corrected chi-square tests of differences in needlestick rates across level and sex with responses clustered by residency program.
Based on responses from 2,051 residents who reported experiencing a needlestick in the 6 months before survey administration. Each potential circumstance was asked as an independent yes/no question. Factors and perceived barriers were presented as lists where respondents could select all that applied.
Residents were asked about the conditions of their most recent needlestick event (Table 3). They most commonly described that their own carelessness played a role (48.8%; n = 1,001). Fatigue (21.4%; n = 439) and feeling rushed (31.3%; n = 641) were not commonly reported as contributors to the needlestick events. PGY1 residents more frequently attributed their needlestick event to inexperience (48.0% vs PGY5 4.0%; p < 0.001) or feeling rushed (39.2% vs PGY5 19.4%; p < 0.001). Higher PGY levels were associated with residents citing carelessness of a colleague as contributing to their needlestick compared with PGY1 residents (PGY5 28.5% vs PGY1 18.7%; p = 0.004). Women more frequently stated that their fatigue played a role in the needlestick event (23.9% vs men 19.3%; p = 0.02), and men more frequently stated that their inexperience played a role (23.0% vs women 18.7%; p = 0.02). Only 14.9% (n = 306) of the cohort thought that the needlestick event could not have been prevented.
Needlestick reporting practices, barriers to reporting, and factors associated with reporting
Overall, 28.7% (n = 589) of the residents who noted experiencing a needlestick event in the past 6 months did not report that event to employee health (Table 4). PGY level was not significantly associated with reporting. Men were less likely to report their most recent needlestick event (men 69.5% vs women 73.8%; OR 0.77; 95% CI 0.63 to 0.94). When duty hour policy violations were examined, residents who reported frequently having fewer than 8 hours off between shifts (fewer than 8 hours off 3 or more times 62.5% vs fewer than 8 hours off 0 to 2 times 72.9%; OR 0.73; 95% CI 0.54 to 0.99) or frequently working more than 80 hours in a week (more than 80 hours in a week 3 or more times 63.0% vs more than 80 hours in a week 0 to 2 times 73.5%; OR 0.70; 95% CI 0.52 to 0.93) were less likely to report their most recent needlestick. Residents with the poorest psychiatric well-being (top quartile of GHQ-12 scores) were also less likely to report needlesticks (score 6 to 12 [top quartile] 65.0% vs score 0 to 1 [bottom quartile] 76.8%; OR 0.61; 95% CI 0.46 to 0.81).
Table 4.
Factors Associated with Reporting Most Recent Needlestick to Employee Health
| Variable | Rate of reporting to employee health (n = 2,051) |
p Value* | Adjusted odds ratio† for reporting needlestick event to employee health (95% CI) |
|
|---|---|---|---|---|
| n | % | |||
| Needlestick event reported to employee health | 1,462 | 71.3 | ||
| PGY | 0.58 | |||
| 1 | 344 | 72.1 | 1.00 (ref) | |
| 2 | 319 | 71.2 | 0.84 (0.62—1.13) | |
| 3 | 269 | 69.2 | 0.76 (0.56—1.04) | |
| 4 | 268 | 69.4 | 0.75 (0.55—1.03) | |
| 5 | 262 | 74.6 | 0.96 (0.69—1.33) | |
| Resident sex | 0.06 | |||
| Female | 665 | 73.8 | 1.00 (ref) | |
| Male | 793 | 69.5 | 0.77 (0.63—0.94)‡ | |
| Duty hour experience in most recent month of general surgery rotation | ||||
| Worked >28 h 0—2 times | 1,197 | 71.5 | 0.74 | 1.00 (ref) |
| Worked >28 h 3 or more times | 265 | 70.5 | 1.33 (0.99—1.78) | |
| Had <8 h off between shifts 0—2 times | 1,267 | 72.9 | 0.001 | 1.00 (ref) |
| Had <8 h off between shifts ≥3 times | 195 | 62.5 | 0.73 (0.54—0.99)‡ | |
| Worked >80 h in a week 0—2 times | 1,193 | 73.5 | <0.001 | 1.00 (ref) |
| Worked >80 h in a week ≥3 times | 269 | 63.0 | 0.70 (0.52—0.93)‡ | |
| General Health Questionnaire-12 score quartile | 0.002 | |||
| First quartile (score range 0—1) | 381 | 76.8 | 1.00 (ref) | |
| Second quartile of GHQ-12 scale (score range 2—3) | 382 | 72.8 | 0.84 (0.62—1.12) | |
| Third quartile of GHQ-12 scale (score range 4—5) | 325 | 71.4 | 0.79 (0.58—1.07) | |
| Fourth quartile of GHQ-12 scale (score range 6—12) | 373 | 65.0 | 0.61 (0.46—0.81)§ | |
| Program type | 0.15 | |||
| Academic | 901 | 69.6 | 1.00 (ref) | |
| Community | 530 | 74.1 | 1.25 (0.97—1.62) | |
| Military | 31 | 73.8 | 1.34 (0.62—2.88) | |
| Geographic region | 0.12 | |||
| Southwest | 183 | 75.6 | 1.00 (ref) | |
| Northeast | 482 | 68.9 | 0.68 (0.47—0.98)‡ | |
| Southeast | 287 | 75.5 | 0.95 (0.64—1.43) | |
| Midwest | 311 | 71.3 | 0.71 (0.48—1.05) | |
| West | 199 | 67.9 | 0.62 (0.41—0.94)‡ | |
| No. of PGY5 residents in program | 0.02 | |||
| 0—5 | 796 | 73.7 | 1.00 (ref) | |
| ≥6 | 666 | 68.6 | 0.77 (0.61—0.98)‡ | |
Based on responses from 2,051 residents who indicated they experienced a needlestick in the 6 months before survey administration.
Bivariate p values are from cluster-corrected chi-square tests of differences in reporting rates across categories of each variable with responses clustered by residency program.
Adjusted odds ratios are from a hierarchical logistic regression model containing all variables displayed and with random intercepts for residency program.
p < 0.05.
p < 0.01.
GHQ, General Health Questionnaire.
Residents were asked about the barriers to reporting their most recent needlestick event (Table 3). The most commonly cited reasons were that reporting takes too much time (80.3%; n=1,647) and is too much of a disruption to operative cases (52.3%; n = 1,073). Women more frequently said that needlestick reporting takes too much time (women 83.2% vs men 78.1%; p = 0.007) and that reporting is too much of a disruption to operative cases (women 56.4% vs men 49.3%; p = 0.005).
When asked about reporting procedures, 11.2% (n = 229) of residents stated that they did not know how to report the event. PGY1 residents were more likely to respond that they did not know how to report a needlestick event compared with PGY2 to PGY5 residents (PGY1 17.0% vs PGY5 8.8%; p < 0.001). Additionally, 21.6% (n = 443) of residents thought that they did not need to report the event because they thought the infection risk was low, and 6.8% of residents (n = 140) did not report the event because they did not want to know the result of testing for communicable diseases. Men more frequently responded that there is no need to report needlesticks because the infection risk is low (men 24.6% vs women 17.9%; p < 0.001) or because they did not want to know the result (men 8.5% vs women 4.7%; p < 0.001). Only 7.5% of residents (n = 153) responded that they felt pressure not to report, but 25.5% of residents (n = 523) responded that there was a negative stigma associated with experiencing a needlestick event, and this negative stigma was most frequently perceived in PGY1 residents (PGY1 32.1% vs PGY5 17.4%; p < 0.001).
When we examined the rates of reporting their most recent needlestick to employee health according to the residents’ self-reported circumstances of the event and barriers to reporting (Table 5), needlestick reporting rates were lowest when residents responded that there was no need to report because the infection risk was low (47.6% vs residents who disagreed that infection risk was low 77.8%; p < 0.001). Needlestick reporting rates were lower when residents thought certain factors contributed to their most recent needlestick, including inexperience (66.6% vs disagreed inexperience played a role 72.5%; p = 0.03), fatigue (65.4% vs disagreed fatigue played a role 72.9%; p = 0.006), feeling rushed (67.9% vs disagreed feeling rushed played a role 72.8%; p = 0.04), or their own carelessness (66.7% vs disagreed own carelessness played a role 75.6%; p < 0.001). Residents were also less likely to report the event if they responded that reporting was too much of a disruption to operative cases (66.4% vs disagreed reporting was too much of a disruption 76.7%; p < 0.001). Residents were less likely to report the event to employee health if they disagreed that the carelessness of others played a role in their most recent needlestick (68.5% vs agreed carelessness of others played a role 81.3%; p < 0.001) or if they thought that the needlestick could have been prevented (69.7% vs agreed needlestick could not have been prevented 80.1%; p = 0.001).
Table 5.
Resident Rates of Reporting Most Recent Needlestick Events to Employee Health by Perceptions of Contributing Factors and Reporting Barriers
| Factor/barrier, agreement | Reported to employee health |
p Value | |
|---|---|---|---|
| n | % | ||
| Most recent needlestick event factor | |||
| Personal inexperience played a role | 0.03 | ||
| Agreed | 287 | 66.6 | |
| Disagreed | 1,175 | 72.5 | |
| My fatigue played a role | 0.006 | ||
| Agreed | 287 | 65.4 | |
| Disagreed | 1,175 | 72.9 | |
| I felt rushed when needlestick occurred | 0.04 | ||
| Agreed | 435 | 67.9 | |
| Disagreed | 1,027 | 72.8 | |
| My carelessness played a role | <0.001 | ||
| Agreed | 668 | 66.7 | |
| Disagreed | 794 | 75.6 | |
| Carelessness of another person played a role | <0.001 | ||
| Agreed | 364 | 81.3 | |
| Disagreed | 1,098 | 68.5 | |
| The needlestick could not have been prevented | 0.001 | ||
| Agreed | 245 | 80.1 | |
| Disagreed | 1,217 | 69.7 | |
| Perceived barrier to needlestick reporting, agreement with factor/barrier | |||
| Reporting the event takes too much time | 0.33 | ||
| Agreed | 1,183 | 71.8 | |
| Disagreed | 279 | 69.1 | |
| Too much disruption to operative case | <0.001 | ||
| Agreed | 712 | 66.4 | |
| Disagreed | 750 | 76.7 | |
| Do not know how to report the event | 0.92 | ||
| Agreed | 164 | 71.6 | |
| Disagreed | 1,298 | 71.2 | |
| No need to report the event because the risk of infection is low | <0.001 | ||
| Agreed | 211 | 47.6 | |
| Disagreed | 1,251 | 77.8 | |
| There is pressure not to report | 0.24 | ||
| Agreed | 102 | 66.7 | |
| Disagreed | 1,360 | 71.7 | |
| Negative stigma associated with a needlestick | 0.94 | ||
| Agreed | 372 | 71.1 | |
| Disagreed | 1,090 | 71.3 | |
| Do not want to know the test result | 0.24 | ||
| Agreed | 93 | 66.4 | |
| Disagreed | 1,369 | 71.6 | |
Based on responses from 2,051 residents who reported experiencing a needlestick in the 6 months before survey administration, of whom 1,462 (71.3%) reported their most recent needlestick to employee health. p Values are from cluster-corrected chi-square tests of differences in needlestick rates across categories of each variable with responses clustered by residency program.
DISCUSSION
Surgeons and, in particular, surgical residents are at high risk for needlestick injuries, given the procedural nature of the specialty. By surveying 7,391 surgery residents representing all 260 ACGME-accredited general surgery residents in the US with a 99.3% response rate, we were able to perform a detailed national evaluation of self-reported needlestick events in surgical residents and examine specifics about reporting practices, making this study the most comprehensive analysis of needlestick injuries to date. We found that 27.7% of surgical residents surveyed reported experiencing a needlestick within the last 6 months, and more than one-fourth of the residents did not report the event to employee health.
Needlestick event rate and factors associated with needlestick injuries
More than one-quarter (27.7%) of the residents surveyed self-reported experiencing a needlestick injury within the 6 months before being surveyed. Previous studies have shown that trainees have a high incidence of needlestick injuries during training, ranging from 12% to 83%, and that the rate tends to increase with time.3,7,8,11,18,33,34 However, most studies were performed in smaller cohorts, often limited to a single-institution or smaller multi-institutional studies, and were usually not focused solely on surgical residents; the largest study to date was from 19 centers,8 and our study included all 260 ACGME-accredited general surgery residency programs in the US.
At the program level, there was considerable variation in needlestick injury rates. Although 2 programs had no needlesticks among their residents during the 6-month period included in this study, 8 programs had ≥50% of their residents reporting a needlestick injury. This variation suggests that some programs have fairly low needlestick event rates, and there might be differences in program-level teaching and culture about handling needles that contributes to the low rate and can serve as examples of best practices.
We identified several factors that were associated with surgical residents experiencing needlestick injuries. Women were more likely to self-report experiencing a needlestick injury compared with men. This might be due to higher actual event rates or could be related to gender differences in health awareness and help-seeking behaviors.35–39 Residents who reported repeated violations of ACGME duty hour policies, specifically the limitations on shift length, time off between shifts, and limitations on total work hours, also had an increased likelihood of experiencing a recent needlestick injury. The easiest explanation for this association is resident fatigue.40–42 However, although fatigue was not a commonly cited factor underlying the needlestick injury in this study, most needlestick injuries involved residents sticking themselves and almost half of residents blamed their own carelessness. This might indicate residents are not aware that their own fatigue can play a role in needlestick events. It is also important to acknowledge that it is uncertain whether more restrictive duty hour policies than those that are currently in place lead to improved resident or patient safety.43
Although most residents in this study responded that their most recent needlestick event could have been prevented, 14.9% of residents thought that their most recent needlestick could not have been prevented. Because it is well established that needlestick events can be considered a preventable workplace injury,2 the findings from this study highlight the need to educate not only residents, but also attending surgeons and other healthcare workers, in safe needle and sharp instrument handling practices, especially considering that 21.8% of residents reported that the carelessness of another played a role in their needlestick event.
Circumstances and self-reported factors playing a role in needlestick events
As with previous studies,3,5,8 most needlesticks in surgical residents occurred in the operating room with solid needles, presumably when suturing. Although sharp-tip needles cannot be eliminated, some have advocated for the use of blunt-tip suture needles whenever possible, which have been shown to reduce the risk of needlestick injury.1,2,44,45 Another way to reduce the risk of transmission of blood-borne pathogens is through encouraging the practice of wearing 2 layers of gloves during bedside procedures, in addition to the safe needle handling education.1,46
Needlestick reporting practices, barriers to reporting, and factors associated with reporting
Reporting of needlestick injuries by residents to their institution’s employee health service is essential for many reasons. Reporting the event allows for testing of both the involved resident and patient for the presence and transmission of blood-borne pathogens, rapid initiation of prophylaxis in appropriate circumstances, early treatment of actual pathogen transmissions, and ensures that any healthcare needs and expenses related to this occupational safety event will be covered by the employer.24,25 More than one-fourth (28.7%) of the residents in this cohort who noted experiencing a needlestick did not report that event to employee health and, distressingly, a considerable proportion (21.6%) of residents reported feeling no need to report their needlestick injury because they thought the infection risk was low. Although the reporting rate in this study is higher than reported historical rates (usually around 50%),8,11,18–21,23,47 additional efforts are needed to educate all healthcare providers about the benefits of reporting needlestick injuries and the risks of not reporting and eliminate fears that there are negative consequences to reporting (eg negative perceptions of the resident among attending surgeons).
Overt external pressure on residents to not report needlestick events was rare, which is similar to a previous single-institution report.11 However, 52.3% of residents responded that reporting is too much of a disruption to operative cases, and 25.5% of residents thought there is a negative stigma associated with needlestick events. Therefore, attending surgeons and other hospital staff need to be aware of the subtle, implicit pressures that residents might incorrectly or correctly perceive to prioritize their work over their health. Of additional concern is that 17.0% of PGY1 residents reported not knowing how to report a needlestick event and the majority of surgery residents (80.3%) thought that reporting takes too much time. More work might be needed to reinforce workplace injury-reporting procedures in junior residents and make injury-reporting procedures easier and more accessible for all residents. This is likely to help all healthcare workers.
Similar to the association observed with needlestick events, violations of ACGME duty hour policies, specifically the limitations on shift length, time off between shifts, and limitations on total work hours, were associated with a lower likelihood of reporting that injury to employee health. Of additional concern is that residents with the poorest psychiatric well-being were more likely to self-report experiencing a needlestick and were less likely to report their needlestick injury. When these data are considered together, it might be that residents with poor psychiatric well-being, possibly from either being fatigued or burned out by excessive work demands, are at higher risk for preventable workplace injuries, such as needlesticks, and then not reporting that injury. However, there is a notable limitation to this hypothesis in that we were only able to identify associations, and the converse hypothesis could be true: a needlestick event could be a major contributor to poor psychiatric well-being. Another possibility is that the frequent duty hour policy violations and poor resident psychiatric well-being reflect negative training environments in which residents at these programs are tacitly or actively discouraged from reporting workplace injuries. Easier and more accessible workplace injury reporting processes might be beneficial in these programs. Although it is tempting to examine residents’ responses to specific questions from the GHQ-12 to gain additional insight into this association, there is research to support that the GHQ-12 assessment is unidimensional; therefore, use and interpretation of subscale scores of the GHQ-12 is not recommended.48
Limitations
This study has several limitations. First, this study was based on a survey of surgery residents and might not be generalizable to all specialties. Second, though all survey-based studies are susceptible to recall and reporting bias that might lead to either underestimation or overestimation of needlestick events, we believe the shorter time frame of recall (6 months, representing the current academic year at the time of the survey vs 1 year or longer in most studies8,11,17,18,34), a broad study population that included residents in every ACGME-accredited general surgery residency program, and the exceptionally high response rate mitigate these biases. It should also be emphasized that the associations identified in this study are correlations, and conclusions about causation cannot be made. Finally, although we found an association between frequent violations of duty hour policies and having experienced a recent needlestick injury, we are unable to determine whether current duty hour policies have a stronger association with needlestick injuries compared with more restrictive policies.
CONCLUSIONS
In this most comprehensive national survey of surgical residents to date, needlesticks occurred frequently. Many needlestick events continue to not be reported, and numerous barriers to reporting exist. These findings can help identify opportunities to reduce needlestick injuries among trainees and encourage reporting of these preventable injuries to receive timely care.
Acknowledgments
Support for this study: This study was funded by the American Board of Surgery, American College of Surgeons, and the Accreditation Council for Graduate Medical Education as a part of the FIRST trial. Dr Yang’s research is supported by the National Heart Lung and Blood Institute of NIH (K08HL145139). Dr Hewitt’s research is supported by a postdoctoral research fellowship from the Agency for Healthcare Research and Quality (5T32HS000078).
Abbreviations and Acronyms
- ABS
American Board of Surgery
- ABSITE
American Board of Surgery In-Training Examination
- GHQ-12
General Health Questionnaire-12
- OR
odds ratio
Footnotes
Disclosure Information: Nothing to disclose.
REFERENCES
- 1.Beltrami EM, Williams IT, Shapiro CN, Chamberland ME. Risk and management of blood-borne infections in health care workers. Clin Microbiol Rev 2000;13:385–407. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 2.NIOSH Fast Facts. How to Prevent Needlestick and Sharps Injuries. Available at: https://www.cdc.gov/niosh/docs/2012123/pdfs/2012-123.pdf?id=10.26616/NIOSHPUB2012123. Accessed April 18, 2018.
- 3.O’Neill TM, Abbott AV, Radecki SE. Risk of needlesticks and occupational exposures among residents and medical students. Arch Intern Med 1992;152:1451–1456. [PubMed] [Google Scholar]
- 4.Hussain SA, Latif AB, Choudhary AA. Risk to surgeons: a survey of accidental injuries during operations. Br J Surg 1988;75: 314–316. [DOI] [PubMed] [Google Scholar]
- 5.McGeer A, Simor AE, Low DE. Epidemiology of needlestick injuries in house officers. J Infect Dis 1990;162:961–964. [DOI] [PubMed] [Google Scholar]
- 6.Weiss ES, Makary MA, Wang T, et al. Prevalence of bloodborne pathogens in an urban, university-based general surgical practice. Ann Surg 2005;241:803–807; discussion 807–809. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 7.Deisenhammer S, Radon K, Nowak D, Reichert J. Needlestick injuries during medical training. J Hosp Infect 2006;63: 263–267. [DOI] [PubMed] [Google Scholar]
- 8.Makary MA, Al-Attar A, Holzmueller CG, et al. Needlestick injuries among surgeons in training. N Engl J Med 2007; 356[26]:2693–2699. [DOI] [PubMed] [Google Scholar]
- 9.Lakbala P, Sobhani G, Lakbala M, et al. Sharps injuries in the operating room. Environ Health Prev Med 2014;19: 348–353. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 10.Myers DJ, Lipscomb HJ, Epling C, et al. Surgical procedure characteristics and risk of sharps-related blood and body fluid exposure. Infect Control Hosp Epidemiol 2016;37:80–87. [DOI] [PubMed] [Google Scholar]
- 11.Ouyang B, Li LD, Mount J, et al. Incidence and characteristics of needlestick injuries among medical trainees at a community teaching hospital: a cross-sectional study. J Occup Health 2017;59:63–73. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 12.Jagger J, Bentley M, Tereskerz P. A study of patterns and prevention of blood exposures in OR personnel. AORN J 1998; 67:979–981. 983–974, 986–977 passim. [DOI] [PubMed] [Google Scholar]
- 13.Tarantola A, Abiteboul D, Rachline A. Infection risks following accidental exposure to blood or body fluids in health care workers: a review of pathogens transmitted in published cases. Am J Infect Control 2006;34:367–375. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 14.Weiss ES, Cornwell EE 3rd, Wang T, et al. Human immunodeficiency virus and hepatitis testing and prevalence among surgical patients in an urban university hospital. Am J Surg 2007;193:55–60. [DOI] [PubMed] [Google Scholar]
- 15.Henry K, Campbell S, Jackson B, et al. Long-term followup of health care workers with work-site exposure to human immunodeficiency virus. JAMA 1990;263[13]: 1765–1766. [PubMed] [Google Scholar]
- 16.Wicker S, Stirn AV, Rabenau HF, et al. Needlestick injuries: causes, preventability and psychological impact. Infection 2014;42:549–552. [DOI] [PubMed] [Google Scholar]
- 17.Cervini P, Bell C. Brief report: needlestick injury and inadequate post-exposure practice in medical students. J Gen Intern Med 2005;20:419–421. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 18.Sharma GK, Gilson MM, Nathan H, Makary MA. Needlestick injuries among medical students: incidence and implications. Acad Med 2009;84:1815–1821. [DOI] [PubMed] [Google Scholar]
- 19.Hamory BH. Underreporting of needlestick injuries in a university hospital. Am J Infect Control 1983;11:174–177. [DOI] [PubMed] [Google Scholar]
- 20.Hamory BH. Error: percent in “underreporting of needlestick injuries” was “underreported”. Am J Infect Control 1984;12:68. [DOI] [PubMed] [Google Scholar]
- 21.Salzer HJ, Hoenigl M, Kessler HH, et al. Lack of risk-awareness and reporting behavior towards HIV infection through needlestick injury among European medical students. Int J Hyg Environ Health 2011;214:407–410. [DOI] [PubMed] [Google Scholar]
- 22.Talbot TR, Wang D, Swift M, et al. Implementation of an enhanced safety-engineered sharp device oversight and bloodborne pathogen protection program at a large academic medical center. Infect Control Hosp Epidemiol 2014;35: 1383–1390. [DOI] [PubMed] [Google Scholar]
- 23.Fritzsche C, Heine M, Loebermann M, et al. Reducing the underreporting of percutaneous exposure incidents: a single-center experience. Am J Infect Control 2016;44: 941–943. [DOI] [PubMed] [Google Scholar]
- 24.US Department of Labor, Occupational Safety and Health Administration. Occupational exposure to bloodborne pathogens; needlestick and other sharps injuries; final rule. Fed Regist 2001;66:5318–5325. Available at: https://www.osha.gov/pls/oshaweb/owadisp.show_document?p_id=16265&p_table=FEDERAL_REGISTER. Accessed 2 October 2019. [PubMed] [Google Scholar]
- 25.Needlestick Safety and Prevention Act of 2000. H.R.5178, 106th Cong. 2000. [Google Scholar]
- 26.Bilimoria KY, Chung JW, Hedges LV, et al. Development of the Flexibility in Duty Hour Requirements for Surgical Trainees (FIRST) trial protocol: a national cluster-randomized trial of resident duty hour policies. JAMA Surg 2016;151:273–281. [DOI] [PubMed] [Google Scholar]
- 27.Bilimoria KY, Chung JW, Hedges LV, et al. National cluster-randomized trial of duty-hour flexibility in surgical training. N Engl J Med 2016;374:713–727. [DOI] [PubMed] [Google Scholar]
- 28.Goldberg DP, Williams P. A User’s Guide To the General Health Questionnaire. Windsor, UK: NFER-Nelson; 1988. [Google Scholar]
- 29.Goodwin L, Ben-Zion I, Fear NT, et al. Are reports of psychological stress higher in occupational studies? A systematic review across occupational and population based studies. PLoS One 2013;8:e78693. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 30.Goldberg DP, Oldehinkel T, Ormel J. Why GHQ threshold varies from one place to another. Psychol Med 1998;28: 915–921. [DOI] [PubMed] [Google Scholar]
- 31.McManus IC, Keeling A, Paice E. Stress, burnout and doctors’ attitudes to work are determined by personality and learning style: a twelve year longitudinal study of UK medical graduates. BMC Med 2004;2:29. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 32.McManus IC, Winder BC, Gordon D. The causal links between stress and burnout in a longitudinal study of UK doctors. Lancet 2002;359[9323]:2089–2090. [DOI] [PubMed] [Google Scholar]
- 33.Lopez RA, Rayan GM, Monlux R. Hand injuries during hand surgery: a survey of intraoperative sharp injuries of the hand among hand surgeons. J Hand Surgery 2008;33:661–666. [DOI] [PubMed] [Google Scholar]
- 34.Wicker S, Nurnberger F, Schulze JB, Rabenau HF. Needlestick injuries among German medical students: time to take a different approach? Med Educ 2008;42:742–745. [DOI] [PubMed] [Google Scholar]
- 35.Lorence D, Park H. Gender and online health information: a partitioned technology assessment. Health Info Libr J 2007; 24:204–209. [DOI] [PubMed] [Google Scholar]
- 36.Mansfield AK, Addis ME, Mahalik JR. “Why won’t he go to the doctor?”: the psychology of men’s help seeking. Int J Mens Health 2003;2:93–109. [Google Scholar]
- 37.Rakowski W, Assaf AR, Lefebvre RC, et al. Information-seeking about health in a community sample of adults: correlates and associations with other health-related practices. Health Educ Q 1990;17:379–393. [DOI] [PubMed] [Google Scholar]
- 38.Rice RE. Influences, usage, and outcomes of Internet health information searching: multivariate results from the Pew surveys. Int J Med Inform 2006;75:8–28. [DOI] [PubMed] [Google Scholar]
- 39.Yousaf O, Grunfeld EA, Hunter MS. A systematic review of the factors associated with delays in medical and psychological help-seeking among men. Health Psychol Rev 2015;9: 264–276. [DOI] [PubMed] [Google Scholar]
- 40.Ayas NT, Barger LK, Cade BE, et al. Extended work duration and the risk of self-reported percutaneous injuries in interns. JAMA 2006;296:1055–1062. [DOI] [PubMed] [Google Scholar]
- 41.Lockley SW, Barger LK, Ayas NT, et al. Effects of health care provider work hours and sleep deprivation on safety and performance. Jt Comm J Qual Patient Saf 2007;33 [Suppl]:7–18. [DOI] [PubMed] [Google Scholar]
- 42.Anderson C, Sullivan JP, Flynn-Evans EE, et al. Deterioration of neurobehavioral performance in resident physicians during repeated exposure to extended duration work shifts. Sleep 2012;35:1137–1146. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 43.Levine AC, Adusumilli J, Landrigan CP. Effects of reducing or eliminating resident work shifts over 16 hours: a systematic review. Sleep 2010;33:1043–1053. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 44.CDC Evaluation. of blunt suture needles in preventing percutaneous injuries among health-care workers during gynecologic surgical proceduresd—New York City, March 1993-June 1994. MMWR Morb Mortal Wkly Rep 1997;46:25–29. [PubMed] [Google Scholar]
- 45.Mingoli A, Sapienza P, Sgarzini G, et al. Influence of blunt needles on surgical glove perforation and safety for the surgeon. Am J Surg 1996;172:512–516; discussion 516–517. [DOI] [PubMed] [Google Scholar]
- 46.Al Maqbali MA. Using double gloves in surgical procedures: a literature review. Br J Nurs 2014;23:1116–1122. [DOI] [PubMed] [Google Scholar]
- 47.Cutter J, Jordan S. Uptake of guidelines to avoid and report exposure to blood and body fluids. J Adv Nurs 2004;46: 441–452. [DOI] [PubMed] [Google Scholar]
- 48.Gnambs T, Staufenbiel T. The structure of the General Health Questionnaire (GHQ-12): two meta-analytic factor analyses. Health Psychol Rev 2018;12:179–194. [DOI] [PubMed] [Google Scholar]