Skip to main content
NCBI home page
NIHPA Author Manuscripts logoLink to NIHPA Author Manuscripts
. Author manuscript; available in PMC: 2026 Jun 1.
Published in final edited form as: Am J Prev Med. 2024 Nov 28;68(6):1080–1090. doi: 10.1016/j.amepre.2024.11.005

Mortality Risks of US Healthcare Workers

Mark Olfson 1, Candace M Cosgrove 2, Melanie M Wall 1, Carlos Blanco 3
PMCID: PMC12092202  NIHMSID: NIHMS2039163  PMID: 39612966

Abstract

Introduction:

Physicians and nurses have lower annual mortality rates than the general population. One explanation for the low mortality rates of these healthcare workers emphasizes their specialized medical knowledge and greater access to healthcare while a second emphasizes their generally higher level of education. We evaluated the extent to which general educational level accounts for the lower all-cause mortality rates of US healthcare than non-healthcare workers. We also compared cause-specific mortality risks of healthcare and non-healthcare workers.

Methods:

A nationally representative sample of healthcare workers (n=176,000) and non-healthcare workers (n=1,662,000) from the 2008 American Community Survey (n=3,310,000) was followed through 2019 for mortality. Cox models estimated hazard ratios of all-cause and cause-specific mortality for six healthcare worker groups. Analyses were performed in 2024.

Results:

Age- and sex-standardized all-cause death rates per 100,000 were lower for healthcare (370.7, 95%CI=361.5–379.9) than non-healthcare (442.2, 95%CI=439.2–445.2) workers (p<0.001). However, this difference was not evident after additionally adjusting for level of education (aHR=1.00, 0.97–1.02). In fully adjusted models, hazards of mortality were lower for healthcare than non-healthcare workers for cardiovascular disease (aHR=0.94, 0.89–0.99) and lung cancer (aHR=0.89, 95%=0.81–0.98), but higher for pancreatic cancer (aHR=1.21, 1.05–1.39) and external causes of death (aHR=1.20, 1.10–1.30).

Conclusions:

Higher educational attainment accounted for the longer life expectancy of US healthcare workers than other workers. Nevertheless, healthcare work relative to non-healthcare work was associated with lower hazards of deaths due to cardiovascular diseases and lung cancer, but higher hazards of deaths due to external causes.

INTRODUCTION

The health of healthcare workers is vital to the optimal functioning of the healthcare system.1 Yet health effects of healthcare work are complex. Some relevant occupational hazards include infectious diseases,2 ionizing radiation,3 various chemicals,4 controlled substances,5 ergonomic strain,6 circadian rhythm disruptions,7 and psychological distress.8 The last four hazards could explain the increased risks of some healthcare workers to suicide9–10 and drug overdose.11

Despite these risks, physicians12–14 and nurses15 enjoy lower all-cause mortality rates than the general population. One explanation is that these healthcare workers have specialized medical knowledge, which motivates adoption of healthier behaviors.16 Compared to other workers, individuals involved in direct patient care are more likely to exercise, have a personal physician, and receive medical check-ups, and less likely to engage in heavy alcohol use.17 Healthcare professionals are also less likely than other workers to have obesity, diabetes, hypertension, and coronary artery disease.18

A second explanation posits that healthcare workers may have lower mortality rates than the general population because of the robust association between level of education and life expectancy.20,21 This association may be mediated by economic factors that affect access to medical care;22 sociopsychological resources that enable problem-solving and stress management;23 and health-related adaptive behaviors.24–25

In evaluating the contribution of educational level to mortality rates of healthcare workers, it is important to consider the great heterogeneity of their roles, responsibilities, and training. To practice as a physician, for example, individuals must graduate from medical school, complete 1 year of postgraduate training, and pass a national medical licensing examination. By contrast, healthcare support work typically involves on-the-job training and no more than a high school diploma.26 Yet healthcare workers in all roles would be expected to have greater knowledge of health and greater access to healthcare than other workers with a similar general level of education. If specialized medical knowledge or greater access to healthcare contributed to the lower mortality risk of healthcare workers, it would be expected that this lower risk would survive controlling for education level.

Beyond uncertainty over the causes of low mortality rates among healthcare workers, little is known about their relative risks of dying from specific causes of death. For example, physicians and nurses have lower tobacco use rates than other adults.27 However, a US study did not find a significant difference in lung cancer mortality of health assessment/treating occupations compared to other occupations (HR=0.71, 95%CI 0.36–1.43).28

The following analysis has two aims: 1) to evaluate the extent to which general educational level accounts for differences in mortality rates of healthcare and non-healthcare workers and 2) to compare cause-specific mortality rates of healthcare and non-healthcare workers. All-cause mortality rates of healthcare and non-healthcare workers were compared and the effects of adjusting for educational achievement and other sociodemographic characteristics were examined. Cause-specific mortality rates of 6 healthcare worker groups were compared to non-healthcare workers. Increasing understanding of the mortality profiles of healthcare workers, such as those related to cancer or infectious disease, might help target preventive health initiatives. It was anticipated that healthcare workers would have lower mortality rates than non-healthcare workers and that these differences would be attenuated, but persist following adjustment for educational level and other sociodemographic characteristics.

METHODS

Study Sample

The study cohort was identified within the Mortality Disparities in American Communities (MDAC)29 sample that links 2008 American Community Survey (ACS) data to National Death Index underlying cause of death certificate records from 2008 through 2019.30 The 2008 ACS, a cross-sectional nationally representative household survey of 1 the US population administered by the US Census Bureau, had a response rate of 97.9%. Its high response rate and sampling design person weights31 help to assure its national representativeness.

To increase the likelihood of completion of occupational training, we analyzed participants who were aged ≥26 years at ACS administration and who were employed, excluding those not employed (N=1,842,000). The US Census Disclosure Review Board approved the data releases (CBDRB-FY23-CES004–015, CBDRB-FY24-CES026–07, CBDRB-FY24-CES026–010, CBDRB-FY24-CES026–011, and CBDRB-FY24-CES0207–005).

Measures

Responses to occupational questions were coded into 509 categories.32 Healthcare workers included: 1) physicians and surgeons (physicians), 2) registered nurses, 3) other diagnosing or treating healthcare workers, 4) health technologists and technicians (health technicians), 5) healthcare support workers, and 6) social and behavioral health workers (Appendix Table 1). All other workers were non-healthcare workers.

Participant characteristics were collected at the ACS survey. Socio-demographic characteristics included age (26–35,36–45,46–55,>55), sex, race/ethnicity, marital status, education, household annual income, and residence (urban vs. rural).33

NDI data indicated whether each MDAC participant died over the 11-years from their ACS survey date. Nineteen specific causes of deaths were considered based International Classification of Diseases, 10th Revision (ICD–10) mortality codes (Appendix Table 2). Detailed external cause of death analyses are presented elsewhere.10,11,34

Statistical Analysis

Background characteristics of healthcare and non-healthcare workers were first compared. Annual unadjusted all-cause and cause specific mortality rates per 100,000 person-years were determined for healthcare workers and non-healthcare workers. With the combined groups as the reference group, age- and sex-standardized all-cause and cause specific mortality rates per 100,000 person-years were derived. Adjusted Cox proportional hazards regressions estimated hazards of all-cause and cause specific mortality for healthcare workers with non-healthcare workers (reference group) and socio-demographic characteristics as covariates. Event time was measured continuously from ACS administration until death date or December 31, 2019.

The six healthcare worker and non-healthcare worker groups were compared with respect to baseline socio-demographic characteristics, age- and sex-standardized all-cause and cause specific mortality rates, and adjusted hazards ratios. Cox proportional hazards models of all-cause mortality were fit stepping in covariates for age and sex; age, sex, and education; age, sex, education, and income; and all characteristics. We stepped in education before income to evaluate to what extent adjusting for the general education level of healthcare and non-healthcare workers changed the hazards of healthcare worker mortality.

Analyses were conducted in SAS version 9.4 (Cary, NC). To increase generalizability, individual-level observations were weighted to account for non-equal probability of ACS selection. Reporting followed Census Bureau’s Disclosure Review Board disclosure guidelines. The New York State Psychiatric Institute IRB deemed this research exempt.

RESULTS

There were several sociodemographic differences across occupational groups (Appendix Table 3). Healthcare support workers and registered nurses included the largest percentages of women, while physicians included the largest percentage of men. Non-healthcare workers had the largest percentage of Hispanic people and healthcare support workers had the largest percentage of non-Hispanic Black people. Except for healthcare support workers, the other healthcare groups achieved higher educational attainment than non-healthcare workers.

Applying Census rounding rules, the study cohorts included 176,000 healthcare workers of whom 6,900 died and 1,662,000 non-healthcare workers of whom 90,500 died during follow-up. The annual age- and sex-standardized rate of all-cause deaths per 100,000 healthcare workers (370.7, 95%CI=361.5–379.9) was significantly lower than that of non-healthcare workers (442.2, 95%CI=439.2–445.2) (Table 1). The hazards of all-cause mortality were significantly lower for healthcare than non-healthcare workers in age- and sex-adjusted analyses (HR=0.86, 95%CI=0.84–0.88), but not in age-, sex-, and education-adjusted analyses (HR=1.00, 95%CI=0.97–1.02) (Table 2).

Table 1.

Mortality of healthcare workers and non-healthcare workers, United States

Age- and Sex- Standardized Mortality Rate per 100,000 person-years (95% CI) Full Adjusted Hazards Ratioc 95% CI (Non-Healthcare Workers, reference)
Cause of Death Healthcare Workers (n=176,000)a Non-Healthcare Workers (N=1,662,000)a Adjusted Hazards Ratiob 95% CI (Non-Healthcare Workers, reference)
All causes 370.7 (361.5-379.9) 442.2 (439.2-445.2) 0.86 (0.84-0.88) 0.99 (0.97-1.02)
Natural deaths 321.4 (312-9-329.9) 393.9 (391.1-396.7) 0.84 (0.82-0.86) 0.97 (0.94-1.00)
 Cardiovascular disease 92.2 (87.4-97.0) 119.4 (117.8-121.0) 0.79 (0.75-0.83) 0.94 (0.89-0.99)
  Ischemic heart disease 45.5 (42.0-49.0) 59.2 (58.1-60.3) 0.79 (0.73-0.86) 0.99 (0.92-1.08)
  Non-ischemic heart disease 27.3 (24.7-29.9) 34.7 (33.9-35.5) 0.79 (0.71-0.87) 0.90 (0.82-1.00)
  Cerebrovascular disease 11.2 (9.6-12.8) 15.0 (14.4-15.6) 0.73 (0.63-0.85) 0.82 (0.70-0.95)
  Other circulatory disease 8.2(6.9-9.5) 10.6 (10.1-11.1) 0.87 (0.73-1.02) 0.97 (0.82-1.15)
 Cancer 126.6 (121.5-131.7) 150.9 (149.1-152.7) 0.88 (0.84-0.91) 0.98 (0.94-1.03)
  Lung 23.1 (21.0-25.2) 38.0 (37.1-38.9) 0.68 (0.61-0.75) 0.89 (0.81-0.98)
  Breastd 22.0 (19.6-24.4) 23.4 (22.4-24.4) 0.93 (0.83-1.05) 0.93 (0.82-1.04)
  Pancreatic 12.9 (11.2-14.6) 11.3 (10.8-11.8) 1.16 (1.01-1.33) 1.21 (1.05-1.39)
  Colorectal 12.9 (11.2-14.6) 13.7 (13.2-14.2) 0.94 (0.82-1.08) 0.99 (0.86-1.15)
  Other cancer 67.5 (63.7-71.4) 77.0 (75.7-78.3) 0.91 (0.86-0.97) 1.00 (0.94-1.06)
 Diabetes mellitus 11.6 (10.0-13.2) 12.7 (12.2-13.2) 0.95 (0.81-1.10) 1.10 (0.94-1.29)
 Influenza and pneumonia 5.1 (4.0-6.2) 6.0 (5.6-6.4) 0.83 (0.66-1.05) 1.00 (0.79-1.26)
 Sepsis 5.3 (4.2-6.4) 5.4 (5.1-5.7) 0.99 (0.80-1.23) 1.20 (0.96-1.50)
 COPD 9.8 (8.4-11.2) 16.3 (15.7-16.9) 0.63 (0.55-0.74) 0.92 (0.79-1.07)
 Liver disease 11.6 (10.0-13.2) 15.2 (14.6-15.8) 0.82 (0.70-0.94) 1.04 (0.90-1.21)
 Other natural causes 59.3 (55.6-63.0) 68.0 (66.8-69.2) 0.85 (0.80-0.91) 0.96 (0.90-1.03)
External causes of death 44.3 (40.9-47.7) 44.7 (43.8-45.6) 1.03 (0.95-1.12) 1.20 (1.10-1.30)
Open in a new tab

Data from MDAC. Limited to adults aged ≥26 years who were employed at time of ACS. Analysis excludes deaths of unknown cause.

a

Numbers in column headings are based on unweighted n’s and rounded to 500s following Census rules.

b

Adjusted for age, sex, and age by sex interaction

c

Adjusted for age, sex, race and ethnicity, marital status, education, income, and residence.

d

Analysis limited to females.

DRB approval #: CBDRB-FY24-CES026-011

Note: Bold face indicates statistical significance (p<.05).

Table 2.

All-cause mortality hazards of US healthcare workers and non-healthcare workers

Characteristics Age and Sex Adjusted HR (95%CI) Age, Sex, and Education Adjusted HR (95%CI) Age, Sex, Education, and Income Adjusted HR (95% CI) Fully Adjusted HR (95% CI)
Worker status
 All healthcare workers 0.86 (0.83-0.88) 1.00 (0.97-1.02) 1.01 (0.98-1.03) 0.99 (0.97-1.02)
 Non-healthcare workers 1.00 1.00 1.00 1.00
Age, years
 26-35 years 0.24 (0.23-0.25) 0.24 (0.23-0.25) 0.22 (0.22-0.23) 0.23 (0.22-0.23)
 36-45 years 0.46 (0.45-0.47) 0.46 (0.45-0.47) 0.45 (0.44-0.46) 0.46 (0.45-0.47)
 46-55 years 1.00 1.00 1.00 1.00
 >55 years 2.88 (2.83-2.92) 2.92 (2.88-2.97) 2.86 (2.82-2.90) 2.77 (2.72-2.81)
Sex
 Female 0.64 (0.63-0.65) 0.62 (0.61-0.63) 0.60 (0.59-0.60) 0.55 (0.54-0.56)
 Male 1.00 1.00 1.00 1.00
Education
 Less than high school 2.10 (2.02-2.18) 1.52 (1.46-1.59) 1.84 (1.76-1.92)
 High school/GED 1.88 (1.81-1.95) 1.47 (1.41-1.53) 1.53 (1.47-1.59)
 Some college/Assoc deg. 1.54 (1.48-1.59) 1.27 (0.22-1.32) 1.30 (1.25-1.35)
 Bachelor’s degree 1.04 (1.00-1.08) 0.94 (0.91-0.98) 0.94 (0.91-0.98)
 Master’s degree 0.90 (0.86-0.94) 0.85 (0.82-0.89) 0.85 (0.81-0.89)
 Doctorate degreea 1.00 1.00 1.00
Income, $
 0 to $40,000 \ (Loss)b 1.98 (1.93-2.02) 1.72 (1.68-1.76)
 $40,001 to $75,000 1.49 (1.46-1.52) 1.41 (1.37-1.44)
 $75,001 to $125,000 1.18 (1.15-121) 1.16 (1.14-1.19)
 >$125,000 1.00 1.00
Race/Ethnicity
 White, non-Hispanic 1.05 (1.02-1.07)
 Black, non-Hispanic 1.00
 Hispanic 0.53 (0.52-0.55)
 Other 1.10 (1.06-1.13)
Marital status
 Married 1.00
 Separated/Divorced 1.35 (1.32-1.37)
 Widowed 2.04 (1.99-2.10)
 Never Married 1.41 (1.38-1.44)
Residence
 Urban 1.08 (1.07-1.10)
 Rural 1.00
Open in a new tab

Data from MDAC. Limited to adults aged ≥26 years who were employed at time of ACS.

a

Includes doctorate degree and professional degree beyond a BA. Associates degree denoted as Assoc degree.

b

Loss refers to negative or a loss in income dollars for the past 12 months.

DRB approval #: CBDRB-FY24-CES026-010.

Note: Bold face indicates statistical significance (p<.05).

Age- and sex- standardized mortality rates were significantly lower for healthcare than non-healthcare workers for natural deaths including deaths due to cardiovascular disease (including all subgroups except “other circulatory diseases”), cancer (including lung cancer and “other” cancer), and liver disease (Table 1). In the fully adjusted models, the hazards of mortality were significantly lower for healthcare than non-healthcare workers for cardiovascular disease including cerebrovascular disease and for lung cancer. In contrast, the hazards of mortality were higher for healthcare workers than non-healthcare workers for pancreatic cancer and external causes of death. Unadjusted mortality rates appear in Appendix Table 4.

Physicians had lower standardized rates than non-healthcare workers of deaths due to cardiovascular disease including ischemic and non-ischemic heart disease, cancer including lung and breast cancer, COPD, and liver disease (Figure 1, Appendix Table 5). Conversely, standardized rates of pancreatic cancer were higher for physicians than non-healthcare workers. After controlling for the sociodemographic variables, physicians had higher hazards of pancreatic cancer mortality (Table 3).

Figure 1. Age and sex standardized mortality rates per 100,000 person years of healthcare occupational groups and non-healthcare workers.

Figure 1.

Open in a new tab

Data from MDAC. Limited to adults aged ≥26 years who were employed at time of ACS. DRB approval #: CBDRB-FY24-CES026–010. Analysis excludes deaths of unknown cause.

Table 3.

Adjusted mortality hazards ratios for healthcare occupational groups and non-healthcare workers

Adjusted hazards ratios (95% Confidence Intervals)
Cause of death Physiciansa (N=13,000) Registered Nursesa (N=42,000) Other Diagnosing or Treating Workersa (N=22,500) Health Techniciansa (N=32,500) Healthcare Support Workersa (N=39,000) Social and Behavioral Health Workersa (N=27,000) Non-Healthcare Workersa (N=1,666,000) Reference Group
All causes 1.04 (0.95-1.15) 0.96 (0.90-1.01) 1.01 (0.93-1.09) 0.99 (0.94-1.05) 0.98 (0.93-1.03) 1.04 (0.98-1.11) 1.00
Natural deaths 1.05 (0.95-1.16) 0.93 (0.87-0.98) 0.99 (0.91-1.08) 0.98 (0.92-1.04) 0.95 (0.90-1.00) 1.02 (0.96-1.10) 1.00
 Cardiovascular disease 0.97 (0.80-1.19) 0.83 (0.73-0.95) 0.94 (0.80-1.10) 0.89 0.79-1.01) 0.96 (0.87-1.05) 1.07 (0.94-1.21) 1.00
  Ischemic heart disease 0.98 (0.74-1.30) 0.95 (0.79-1.15) 0.97 (0.77-1.23) 0.90 (0.75-1.09) 1.04 (0.90-1.20) 1.10 (0.91-1.33) 1.00
  Non-ischemic heart dis. 0.94 (0.65-1.38) 0.80 (0.63-1.02) 1.01 (0.76-1.35) 0.96 (0.77-1.18) 0.84 (0.70-1.01) 0.95 (0.75-1.21) 1.00
  Cerebrovascular disease 1.04 (0.64-1.68) 0.58 (0.39-0.85) 0.82 (0.53-1.27) 0.89 (0.65-1.23) 0.76 (0.58-1.00) 1.09 (0.78-1.53) 1.00
  Other circulatory disease 0.94 (0.49-1.81) 0.82 (0.54-1.23) 0.74 (0.41-1.33) 0.60 (0.38-0.97) 1.26 (0.97-1.65) 1.24 (0.86-1.80) 1.00
 Cancer 1.11 (0.95-1.29) 0.97 (0.89-1.06) 0.98 (0.86-1.11) 1.01 (0.92-1.11) 0.96 (0.88-1.04) 0.96 (0.86-1.07) 1.00
  Lung 0.95 (0.61-1.47) 0.76 (0.61-0.95) 0.98 (0.72-1.34) 0.94 (0.76-1.15) 0.91 (0.77-1.08) 0.89 (0.69-1.16) 1.00
  Breastb 0.37 (0.12-1.16) 1.08 (0.88-1.33) 0.79 (0.51-1.23) 0.93 (0.71-1.20) 0.83 (0.67-1.04) 0.99 (0.74-1.33) 1.00
  Pancreatic 1.86 (1.25-2.77) 1.48 (1.14-1.94) 0.87 (0.55-1.39) 1.20 (0.87-1.65) 0.95 (0.70-1.30) 1.16 (0.83-1.63) 1.00
  Colorectal 1.22 (0.78-1.93) 0.95 (0.70-1.30) 0.96 (0.63-1.46) 0.81 (0.57-1.16) 1.15 (0.90-1.47) 0.89 (0.61-1.29) 1.00
  Other cancer 1.05 (0.86-1.28) 0.93 (0.81-1.06) 1.01 (0.85-1.20) 1.08 (0.95-1.24) 0.99 (0.88-1.21) 0.94 (0.81-1.09) 1.00
 Diabetes mellitus 1.37 (0.74-2.53) 0.76 (0.51-1.14) 0.80 (0.45-1.39) 1.30 (0.96-1.76) 1.11 (0.85-1.44) 1.34 (0.95-1.87) 1.00
 Influenza and pneumonia 1.34 (0.65-2.74) 1.24 (0.77-2.01) 0.30 (0.09-1.01) 1.29 (0.81-2.03) 0.92 (0.60-1.42) 0.76 (0.40-1.43) 1.00
 Sepsis 0.94 (0.36-2.43) 1.04 (0.62-1.75) 1.25 (0.62-2.55) 1.59 (1.06-2.38) 1.11 (0.76-1.62) 1.14 (0.65-1.98) 1.00
 COPD 0.92 (0.79-1.07) 0.72 (0.50-1.00) 1.12 (0.69-1.82) 0.87 (0.62-1.22) 1.02 (0.80-1.31) 0.92 (0.79-1.07) 1.00
 Liver disease 0.70 (0.31-1.58) 1.34 (1.01-1.79) 1.62 (1.11-2.37) 1.04 (0.76-1.41) 0.73 (0.54-0.99) 1.05 (0.72-1.54) 1.00
 Other natural causes 1.04(0.82-1.30) 0.86 (0.74-1.00) 1.04 (0.86-1.25) 0.90 (0.78-1.05) 0.94 (0.84-1.07) 1.06 (0.91-1.24) 1.00
External causes of death 0.99 (0.74-1.34) 1.29 (1.08-1.55) 1.05 (0.82-1.35) 1.12 (0.94-1.34) 1.27 (1.10-1.48) 1.27 (1.03-1.55) 1.00
Open in a new tab

Data from MDAC. Limited to adults aged ≥26 years who were employed at time of ACS. Analysis excludes deaths of unknown cause.

a

Numbers in column headings are based on unweighted n’s and rounded to 500s following Census rules.

b

Analysis limited to females.

Results adjusted for age, sex, race/ethnicity, marital status, education, income, and residence.

Note: Bold face indicates statistical significance (p<.05).

DRB approval #: CBDRB-FY24-CES026-010

Compared to non-healthcare workers, physicians had significantly decreased age- and sex-standardized rates of all-cause death including from natural and external causes. The hazards of all-cause mortality were significantly lower for physicians than non-healthcare workers in age- and sex-adjusted analyses (aHR=0.65, 0.59–0.71), but not in age-, sex-, and education-adjusted analyses (aHR=0.96, 0.87–1.05) (Appendix Table 6).

Registered nurses had a significantly lower standardized all-cause and natural cause mortality rates than non-healthcare workers. Among natural causes, nurses had lower standardized mortality rates of each cardiovascular disease group, cancer deaths including lung and “other cancer”, diabetes, and COPD (Figure 1, Appendix Table 5). In fully adjusted models, the hazards of mortality were lower for nurses than non-healthcare workers for natural causes (aHR=0.93,0.87–0.98) including cardiovascular disease (aHR=0.83, 0.73–0.95) and lung cancer (aHR=0.76, 0.61–0.95). By contrast, the adjusted hazards of mortality for registered nurses compared to non-healthcare workers were higher for pancreatic cancer (aHR=1.48,1.14–1.94), liver diseases (aHR=1.34,1.01–1.79), and external causes of death (aHR=1.29, 1.08–1.55) (Table 3).

Relative to non-healthcare workers, other diagnosing or treating workers had lower standardized all-cause and natural cause deaths. Among natural causes, their standardized mortality rate was lower than that of non-healthcare workers for the cardiovascular diseases, cancer including lung and “other” cancers, diabetes, influenza and pneumonia, COPD, and other natural causes. Compared to non-healthcare workers, other diagnosing or treating workers also had lower standardized rates of external causes of death (Figure 1, Appendix Table 5). In the fully adjusted models, the hazards of mortality for this group compared to non-healthcare workers was higher for liver diseases (aHR=1.62, 1.11–2.37) (Table 3).

Compared to non-healthcare workers, health technicians had lower standardized rates of other circulatory disease deaths, but higher rates of death due to sepsis (Figure 1, Appendix Table 5). In fully adjusted Cox models, health technicians compared to non-healthcare workers had lower hazards of other circulatory disease deaths (aHR=0.60, 0.38–0.97), but higher hazards of deaths due to sepsis (aHR=1.59, 1.06–2.38) (Table 3).

Healthcare support workers compared to non-healthcare workers had significantly increased standardized all-cause, natural cause, and external cause mortality (Figure 1, Appendix Table 5). The increase in natural cause deaths included cardiovascular disease including ischemic heart disease and other circulatory disease, diabetes, other natural causes, and external causes of death. In fully adjusted models, healthcare support workers had decreased adjusted hazards of mortality due to liver disease (aHR=0.73, 0.54–0.99), but increased hazards of external causes of death (aHR=1.27, 1.10–1.48) (Table 3).

Social and behavioral health workers compared to non-healthcare workers had lower standardized all-cause and natural mortality rates (Figure 1, Appendix Table 5). They also had lower standardized rates of cardiovascular disease deaths including ischemic and non-ischemic heart disease, cancer deaths including lung and “other” cancer, and COPD. In the fully adjusted models, social and behavioral health workers had increased adjusted hazards of external causes of death (aHR=1.27, 1.03–1.55) (Table 3).

DISCUSSION

US healthcare workers tend to have lower mortality rates than other workers. Contrary to our expectations, the level of education of healthcare workers fully accounted for their lower all-cause mortality rate. Among healthcare worker groups, only healthcare support workers, the one healthcare group with lower education attainment than non-healthcare workers, had a higher standardized mortality rate than non-healthcare workers.

In evaluating these findings, it important to acknowledge that the longer time spent in school by healthcare than non-healthcare workers may not necessarily have caused them to have lower mortality rates. Factors that precede completion of education, such as parental income or education, could have accounted for the education and mortality association. It is also not possible to determine which downstream factors associated with education contributed to decreasing the group differences in mortality. A large literature, spanning randomized controlled trials, twin studies, and quasi-experimental designs, has examined associations between education and mortality.20,35 Beneficial education effects on mortality risk have been observed in some but not all populations and contexts.20 Yet educational disparities in life expectancy within the US have widened over the last few decades.36 Because nearly one in ten US adults aged ≥25 years has less than a high school education,37 programs and practices that promote high school graduation38 and beyond might offer opportunities to reduce mortality rates.

The cause-specific mortality findings provide some clues to potential occupational health benefits and risks of healthcare work. Accounting for education and other socio-demographic differences, registered nurses compared to non-healthcare workers had lower risks of cardiovascular and lung cancer mortality. Factors that protect registered nurses from these causes of mortality cannot be determined from this study. However, nurses have lower rates of tobacco use27 and obesity39–40 than the general population. According to a recent meta-analysis, 23.2% of nurses in the Americas (84% US studies) were obese39 compared to 37.0% of US women and 35.5% of US men.40

Registered nurses along with other diagnosing or treating healthcare workers were at increased risk of dying from liver diseases. While sources of this elevated risk can also not be determined from the study, US nurses41 and some other health professionals42 are less likely than controls to have problems with alcohol use. An earlier MDAC analysis further reported that these two groups compared to non-healthcare workers were not at increased risk of alcohol-related deaths.34 However, nurses and other frontline healthcare workers have increased exposure to hepatitis C virus (HCV).43 HCV infections of healthcare workers, which are often unreported,44 can be transmitted by contact of infected blood with injuries to the skin or mucosal membranes. While the present findings do not specifically implicate HCV in elevated liver disease deaths among these healthcare workers, the results are suggestive and should motivate further clinical research, particularly given availability of curative direct acting HCV antivirals.45

Healthcare technicians were at increased risk of death due to sepsis that can result from a wide range of severe infections. Increased occupational exposure to blood borne pathogens through needle stick injury are one known significant hazard to health technicians and other healthcare workers.46 These injuries can lead to serious or even fatal infections from hepatitis B, HCV, HIV, and other blood borne pathogens,47 though reported cases of sepsis attributed to health-related work are rare.48

Compared to non-healthcare workers, physicians, and to a lesser extent registered nurses, were unexpectedly at increased risk of pancreatic cancer mortality. Because neither physicians or registered nurses have significantly elevated common pancreatic cancer risk factors, such as smoking,27 obesity,37 or excessive alcohol use,41,47 the source of this unexpected risk is unknown. If physicians and nurses receive more extensive diagnostic evaluations than non-healthcare workers, surveillance bias might contribute to this finding. Among the major cancers, pancreatic cancer has the lowest rate of histological verification leading to its underestimation,49,50 while cancer of unknown primary site is associated with limited diagnostic evaluations51 and low health literacy.52 In a post hoc analysis, it was found that the percentage of cancer deaths that were unknown primary was 5.4% for non-healthcare workers, 3.9% for registered nurses, and below the reporting threshold for physicians.

As has been previously reported, healthcare workers were vulnerable to external causes of death.11,53 Registered nurses, healthcare support workers, and social and behavioral health workers compared to non-healthcare workers were at elevated risk of external causes of death, such as suicide and unintentional overdose, after controlling for several potentially confounding characteristics. These risks, some of which have been described elsewhere,10–11 underscore the importance of ensuring that healthcare workers have timely access to confidential and effective mental health and substance use treatment services. Risks of external causes of death could vary by occupational roles. For example, healthcare support workers often lift patients or engage in other physically strenuous tasks that put them at high risk of musculoskeletal injuries54,55 and opioid prescriptions, which may in turn pose risks of unintentional overdose56 and suicide.57

The contribution of educational level to the differential mortality rate of all healthcare workers extended to physicians. While physicians had lower hazards of mortality than non-health care workers after controlling for age and sex, a significant difference was not observed after further controlling for educational level. This observation is broadly consistent with findings from Spain (2001–2011),58 Norway (1960–2000),11 and Finland (1971–1980)59 that physicians do not have lower mortality rates compared to other highly educated professionals. One of many factors that may contribute to this finding is that the quality of healthcare physicians receive is only of slightly higher quality than that received by non-physicians.60 Other health advantages upstream to the health care system, which are shared by highly educated individuals, such as higher incomes, safer neighborhoods, and healthier lifestyles, may also make contributions to their health.23

Limitations

This analysis has several limitations. First, the mortality data, which terminated in 2019, do not reflect effects of the COVID-19 pandemic.61 Second, because no adjustments were made for multiple comparisons, the confidence intervals should be cautiously interpreted. Third, occupational status was based on a one-time assessment that does not capture subsequent employment transitions and some individuals start healthcare work before or after age 26. Fourth, the ACS does not measure health risk factors that might precede initiating healthcare employment and influence occupation selection. Fifth, the results do not provide insights into which aspects of healthcare work confer mortality risks or advantages. Sixth, the results are based on non-healthcare workers as the reference group and might have differed if another reference group had been selected. Finally, sample size limitations constrained the precision of the mortality estimates, particularly for physicians, and prevented meaningful comparisons among healthcare worker groups.

CONCLUSIONS

Comparatively low death rates of US healthcare workers appear to be related to their high level of general education. Relative to non-healthcare workers, there were also some cause-specific increased mortality risks for individual health worker groups that could reflect occupation-specific hazards. There might also be general lessons from overall healthcare worker mortality patterns concerning the protective role of education through adoption of health behaviors, such as avoiding tobacco use. Programs and practices that help expand educational opportunities, a social determinant of health,62 might reduce mortality rates in other populations.

Supplementary Material

1

ACKNOWLEDGEMENTS

Funding:

Funding was provided by interagency agreements between the National Heart, Lung, and Blood Institute and the National Institute on Aging interagency with Census Bureau.

Footnotes

Declaration of conflict of interest: No financial disclosures have been reported by the authors of this paper.

Disclaimer: Any opinions and conclusions expressed herein are those of the authors and do not represent the views of the U.S. Census Bureau, National Heart, Lung, and Blood Institute or the National Institute on Drug Abuse. The Census Bureau has ensured appropriate access and use of confidential data and has reviewed these results for disclosure avoidance protection (Project MDACMS#056: DRB Approval Numbers: CBDRB-FY23-CES004-015, CBDRB-FY24-CES026-07, CBDRB-FY24-CES026-010, CBDRB-FY24-CES026-011, and CBDRG-FY24-CES0207-005).

CREDIT AUTHOR STATEMENT

Author Statement: Mark Olfson: Conceptualization, Methodology, Writing-Original Draft, Writing-Review & Editing. Candace M. Cosgrove: Methodology, Formal Analysis, Resources, Data Curation. Melanie M. Wall: Conceptualization, Methodology, Visualization, Writing-Review & Editing. Carlos Blanco: Conceptualization, Methodology, Writing-Review & Editing.

Publisher's Disclaimer: This is a PDF file of an unedited manuscript that has been accepted for publication. As a service to our customers we are providing this early version of the manuscript. The manuscript will undergo copyediting, typesetting, and review of the resulting proof before it is published in its final form. Please note that during the production process errors may be discovered which could affect the content, and all legal disclaimers that apply to the journal pertain.

REFERENCES

  • 1.Wallace JE, Lemaire JB, Ghali WA. Physician wellness: a missing quality indicator. Lancet. 2009;374(9702):1714–21. doi-org.ezproxy.cul.columbia.edu/10.1016/S0140-6736(09)61424-0 [DOI] [PubMed] [Google Scholar]
  • 2.Sepkowitz KA, Eisenberg L. Occupational deaths among healthcare workers. Emerg Infect Dis 2005;11(7):1003–1008. doi: 10.3201/eid1107.041038. [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 3.Kitahara CM, Linet MS, Drozdovitch V, Alexander BH, et al. , Cancer and circulatory disease risks in US radiologic technologists associated with performing procedures involving radionuclides. Occup Environ Med 2015;72(11):770–776. doi: 10.1136/oemed-2015-102834 [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 4.Arif AA, Delclose GL. Association between cleaning-related chemicals and work-related asthma and asthma symptoms among healthcare professionals. Occup Environ Med 2012;69(1):35–40. doi: 10.1136/oem.2011.064865 [DOI] [PubMed] [Google Scholar]
  • 5.Trinkhoff AM, Selby VL, Baek H, Storr CL, et al. , Workplace exposures and prescription drug misuse among nurses. JONA 2022;52(11):591–597. 10.1097/nna.0000000000001215 [DOI] [PubMed] [Google Scholar]
  • 6.Gomaa AE, Tapp LC, Luckhaupt SE, Vanoli K, et al. , Occupational traumatic injuries among workers in health care facilities – United States, 2012–2013. MMRW Morb Mortal Wkly Rep 2015;64(15):405–410. [PMC free article] [PubMed] [Google Scholar]
  • 7.Jorgensen JT, Karlsen S, Stayner L, Hansen J, et al. , Shift work and overall and cause-specific mortality in the Danish nurse cohort. Scand J Work Environ Health 2017;43(2):117–126. doi: 10.5271/sjweh.3612 [DOI] [PubMed] [Google Scholar]
  • 8.Feskanich D, Hastrup JL, Marshall JR, Colditz GA, et al. , Stress and suicide in the Nurses’ Health Study. J Epidemiol Commun Health 2002;56:95–98. 10.1136/jech.56.2.95 [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 9.Dutheil F, Aubert C, Pereira B, Dambrun M, et al. Suicide among physicians and health-care workers: A systematic review and meta-analysis. PLoS ONE 2019;14(12):e0226361. 10.1371/journal.pone.0226361 [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 10.Olfson M, Cosgrove CM, Wall MM, Blanco C. Suicide risks of healthcare workers in the United States. JAMA 2023;330(12):1161–1166. 10.1001/jama.2023.15787 [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 11.Olfson M, Cosgrove CM, Wall MM, Blanco C. Fatal drug overdose risks of healthcare workers in the United States: a population-based cohort study. Ann Int Med 2023;176(8):1081–1088. doi: 10.7326/M23-0902. [DOI] [PubMed] [Google Scholar]
  • 12.Aasland OG, Hem E, Haldorsen T, Ekeberg O. Mortality among Norwegian doctors 1960–2000. BMC Pub Health 2011;11:173. doi: 10.1186/1471-2458-11-173. [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 13.Innos K, Rahu K, Baburin A, Rahi M. Cancer incidence and cause-specific mortality in male and female physicians: a cohort study in Estonia. Scand J Public Health 2002;30:133–140. doi: 10.1080/14034940210133735 [DOI] [PubMed] [Google Scholar]
  • 14.Juel K, Mosbech J, Hansen ES. Mortality and causes of death among Danish medical doctors 1973–1992. Int J Epidemiol 1999;28:456–460. doi: 10.1093/ije/28.3.456. [DOI] [PubMed] [Google Scholar]
  • 15.Gunnarsdittor H, Rafnsson V. Mortality among Icelandic nurses. Scan J Work Environ Health 1995;21:24–29. 10.5271/sjweh.4 [DOI] [PubMed] [Google Scholar]
  • 16.Frank E, Biola H, Burnett CA. Mortality rates and causes among US physicians. Am J Prev Med 2000;19(3):155–159. doi: 10.1016/s0749-3797(00)00201-4. [DOI] [PubMed] [Google Scholar]
  • 17.Helfand BK, Mukamal KJ. Healthcare and lifestyle practices of healthcare workers: do healthcare workers practice what they preach? JAMA Intern Med 2013;173(3):242–244. doi: 10.1001/2013.jamainternmed.1039 [DOI] [PubMed] [Google Scholar]
  • 18.Dayob E, Jena AB. Chronic disease prevalence and healthy lifestyle behaviors among US health care professionals. Mayo Clin Proc 2015;90 (12):1659–1662. . 10.1016/j.mayocp.2015.08.002 [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 19.Ko DT, Chu A, Austin PC, Johnston S, et al. , Comparison of cardiovascular risk factors and outcomes among practicing physicians vs the general population in Ontario, Canada. JAMA Network Open 2019;2(11):e1915983. doi: 10.1001/jamanetworkopen.2019.15983 [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 20.Hummer RA, Hernandez EM. The effect of educational attainment on adult mortality in the United States. Popul Bull 2013;68(1):1–16. [PMC free article] [PubMed] [Google Scholar]
  • 21.Raghupathi V, Raphupathi W. The influence of education on health: an empirical assessment of OECD countries for the period 1995–2015. Arch Pub Health 2020;78:20. 10.1186/s13690-020-00402-5 [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 22.Fox JB, Shaw FE. Relationship of income and health care coverage to receipt of recommended clinical preventive services by adults – United States, 2011–2012. MMWR Morb Mortal Wkly Rep 2014;63(31):666–670. [PMC free article] [PubMed] [Google Scholar]
  • 23.Zajacova Lawrence EM. The relationship between education and health: reducing disparities through a contextual approach. Ann Rev Pub Health 2018;39:273–289. doi: 10.1146/annurev-publhealth-031816-044628 [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 24.Bennie JA, De Cocker K, Teychenne MJ, Brown WJ, et al. , The epidemiology of aerobic physical activity and muscle-strengthening activity guideline adherence among 383,928 US adults. Int J Behav Nutr Physical Activity 2019;16:34. 10.1186/s12966-019-0797-2 [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 25.Cornelius ME, Loretan CG, Wang TW, Jamal A, et al. Tobacco use among adults – United States, 2020. MMRW Morb Mortal Wkly Rep 2022. 71(11) :397–405. [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 26.Bureau of Labor Statistics, Healthcare support workers. https://www.bls.gov/ors/factsheet/pdf/healthcare-support-occupations.pdf
  • 27.Nilan K, McKeever T, McNeill A, Raw M, et al. , Prevalence of tobacco use in healthcare workers: A systematic review and meta-analysis. PLoS One 2019;14(7):e0220168. doi: 10.1371/journal.pone.0220168. [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 28.Lee DJ, Fleming LE, Le Blanc WG, Arheart KL, et al. , Occupation and lung cancer mortality in a nationally representative US cohort: the National Health Interview Survey (NHIS). JOEM 2006;48(8):823–832. doi: 10.1097/01.jom.0000225137.19863.4e. [DOI] [PubMed] [Google Scholar]
  • 29.United States Census Bureau, Mortality Disparities in American Communities (MDAC) https://www.census.gov/topics/research/mdac.html Accessed September 12, 2024
  • 30.National Center for Health Statistics, The National Death Index. https://www.cdc.gov/nchs/ndi/index.html Accessed September 12, 2024.
  • 31.United States Census Bureau, American Community Survey (ACS). https://www.census.gov/programs-surveys/acs Accessed September 12, 2024.
  • 32.US Bureau of Labor Statistics, Injuries, Illness, and Fatalities: Survey of Occupational Injuries and Illnesses Data. https://www.bls.gov/iif/nonfatal-injuries-and-illnesses-tables.htm#summary Accessed September 12, 2024.
  • 33.Ratcliffe M, Burd C, Holder K, Fields A. Defining rural at the US Census Bureau. ACSGEO-1, U.S. Census Bureau, Washington, DC, 2016. [Google Scholar]
  • 34.Olfson M, Cosgrove CM, Wall MM, Blanco C: Alcohol-related deaths of US healthcare workers. JAMA Network Open 2024;7(5):e2410248. doi: 10.1001/jamanetworkopen.2024.10248. [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 35.Galama TJ, Lleras-Muney A, van Kippersluis H. The effect of education on health and mortality; a review of experimental and quasi-experimental evidence. NBER Working Paper 24225, July 2018. [Google Scholar]
  • 36.Bosworth B Increasing disparities in mortality by socioeconomic status. Ann Rev Pub Health 2018;39:237–251. doi: 10.1146/annurev-publhealth-040617-014615. [DOI] [PubMed] [Google Scholar]
  • 37.United States Census Bureau, Educational attainment in the United States: 2022. https://www.census.gov/data/tables/2022/demo/educational-attainment/cps-detailed-tables.html Accessed September 12, 2024
  • 38.Zaff JF, Donian A, Gunning A, Anderson SE, et al. , Factors that promote high school graduation: a review of the literature. Educ Psychol Rev 2017;29:447–476. [Google Scholar]
  • 39.Sadali UB, Kamal KKBN, Park J, Chew HSJ, et al. , The global prevalence of overweight and obesity among nurses: a systematic review and meta-analysis. J Clin Nurs 2023;32:7934–7955. doi: 10.1111/jocn.16861. [DOI] [PubMed] [Google Scholar]
  • 40.World Health Organization, Global Health Observatory data repository, prevalence of obesity among adults. https://apps.who.int/gho/data/view.main.CTRY2450A Accessed September 12, 2024
  • 41.Trinkoff AM, Eaton WW, Anthony JC. The prevalence of substance abuse among registered nurses. Nurs Res. 1991;40(3):172–5. [PubMed] [Google Scholar]
  • 42.Kenna GA, Wood MD. The prevalence of alcohol, cigarette and illicit drug use and problems among dentists. J Am Dent Assoc 2005;136(7):1023–32. doi: 10.14219/jada.archive.2005.0297. [DOI] [PubMed] [Google Scholar]
  • 43.Westerman C, Peters C, Lisiak B, Lamberti M, Nienhaus A. The prevalence of hepatitis C among healthcare workers: a systematic review and meta-analysis. Occup Environ Med 2015;72:880–888. 10.1136/oemed-2015-102879 [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 44.Elseviers MM, Arias-Guillen M, Gorke A, et al. Sharps injuries amongst healthcare workers: review of incidence, transmissions and costs. J Ren Care 2014;40:150–156. 10.1111/jorc.12050 [DOI] [PubMed] [Google Scholar]
  • 45.Fleurence RL, Collins FS. A national hepatitis C elimination program in the United States: A historic opportunity. JAMA 2023; 329 (15): 1251–1252. 10.1001/jama.2023.3692 [DOI] [PubMed] [Google Scholar]
  • 46.Cooke CE, Stephens JM. Clinical, economic, and humanistic burden of needlestick injuries in healthcare workers. Med Dev: Evid Res 2017; 10: 225–235. 10.2147/mder.s140846 [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 47.Panlilio AL, Orelien JG, Srivastava PU, Jagger J, et al. Estimate of the annual number of percutaneous injuries among hospital-based healthcare workers in the United States, 1997–1998. Infect Control Hosp Epidemiol 2004;25:556–562. doi: 10.1086/502439. [DOI] [PubMed] [Google Scholar]
  • 48.Epidemiological notes and reports laboratory-acquired meningococcemia – California and Massachusetts. MMWR Weekly 1991;40(3):46–47, 55. [PubMed] [Google Scholar]
  • 49.Fest J, Ruiter R, van Rooij FJA, van de Geest FJA, et al. , Underestimation of pancreatic cancer in the national cancer registry – Reconsidering the incidence and survival rates. Eur J Cancer 2017;72:186–191. 10.1016/j.ejca.2016.11.026 [DOI] [PubMed] [Google Scholar]
  • 50.Margulis AV, Fortuny J, Kaye JA, et al. Validation of cancer cases using primary care, cancer registry, and hospitalization data in the United Kingdom. Epidemiology. 2018;29(2):308–313. doi: 10.1097/EDE.0000000000000786. [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 51.Mileshkin L, Bochtler T, Beringer A, Mueller-Ohldach M, et al. , Patterns of care and outcomes in carcinoma of unknown primary: a SEER-Medicare study. Ann Oncol 2020;31:S4:S459. doi: 10.3390/cancers14122905. [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 52.Vadjdic CM, Perez-Concha O, Dobbins T, Ward RL, et al. , Demographic, social, and lifestyle risk factors for cancer registry-notified cancer of unknown primary site (CUP). Cancer Epidemiology 2019;60:156–161. 10.1016/j.canep.2019.04.004 [DOI] [PubMed] [Google Scholar]
  • 53.Morano LH, Steege AL, Luckhaupt SE. Occupational patterns of unintentional and undetermined drug involved and opioid-involved overdose deaths -United States, 2007–2012. MMWR Morb Mortal Wkly Rep 2018;67(33):925–930. doi.org/ 10.15585/mmwr.mm6733a3 [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 54.US Bureau of Labor Statistics, Injuries, Illnesses, and fatalities, Table R98. https://www.bls.gov/web/osh/cd_r98.htm Accessed September 12, 2024
  • 55.Waters TR. Introduction to ergonomics for healthcare workers. Rehab Nursing 2010;35(10):185–191. doi: 10.1002/j.2048-7940.2010.tb00046.x. [DOI] [PubMed] [Google Scholar]
  • 56.Bohnert ASB, Valenstein M, Blair MJ, Ganoczy D, et al. , Association between opioid-prescribing patterns and opioid overdose-related deaths. JAMA 2011; 305 (13):1315–1321. 10.1001/jama.2011.370 [DOI] [PubMed] [Google Scholar]
  • 57.Ilgen MA, Bohnert ASB, Ganoczy D, Bair MJ, et al. , Opioid dose and risk of suicide. Pain 2016;157(5):1079–1084. 10.1097/j.pain.0000000000000484 [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 58.Huertas-Herrero L, Anderica E, Belz MJ, Ronda E, et al. , Avoidable mortality for causes amenable to medical care and suicide in physicians in Spain. Int Arch Occup Envion Health 2022;951147–1155. 10.1007/s00420-021-01813-9 [DOI] [PubMed] [Google Scholar]
  • 59.Rimpelä AH, Nurminen MM, Pulkkinen PO, Rimpelä MK, Valkonen T. Mortality of doctors: do doctors benefit from their medical knowledge? Lancet 1987;1(8524):84–86. doi: 10.1016/s0140-6736(87)91919-2. [DOI] [PubMed] [Google Scholar]
  • 60.Frakes MD, Gruber J, Jena A. Is great information good enough? Evidence from physicians as patients. NBER Working Paper 26038. http://www.nber.org/papers/w26038 [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 61.Lin S, Deng X, Ryan I, Zhang K,et al. , COVID-19 symptoms and deaths amonghealthcare workers, United States. Emerg Infect Dis 2022;28(8):1624–1632. doi: 10.3201/eid2808.212200. [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 62.Kubota Y, Heiss G, MacLehose RF, Roetker NS, Folsom AR. Association ofeducational attainment with lifetime risk of cardiovascular disease: the atherosclerosis risk in communities study. JAMA Intern Med 2017;177:1165–1172. doi: 10.1001/jamainternmed.2017.1877. [DOI] [PMC free article] [PubMed] [Google Scholar]

Associated Data

This section collects any data citations, data availability statements, or supplementary materials included in this article.

Supplementary Materials

1
NIHMS2039163-supplement-1.pdf (221.4KB, pdf)

RESOURCES