|
Abstract
|
[
Introduction
Despite an increase in alcohol-related deaths in the US over the last 2 decades,1,2 little is known about alcohol-related mortality risks among physicians3 and other health care workers. Compared with non–health care workers, health care workers have increased drug overdose death risks.4 An occupational liability to substance use or to risks of painful injuries and access to controlled substances may explain these findings. Only the first hypothesis would contribute to an increased risk of alcohol-related mortality. We estimated risks for alcohol-related deaths among US health care workers compared with non–health care workers.
Methods
This cross-sectional study was deemed exempt from review and the need for informed consent by the institutional review board of the New York State Psychiatric Institute owing to the use of only deidentified data. We followed the STROBE reporting guideline.
The 2008 American Community Survey (ACS) was a cross-sectional, nationally representative survey of approximately 2.9 million addresses with a 97.9% response rate. Data were linked to National Death Index records from 2008 to 2019.5 After excluding unemployed individuals and those younger than 26 years, the cohort included 1 838 000 individuals. Alcohol-related underlying or contributing causes of death were identified (eTable 1 in Supplement 1). Health care workers included (1) registered nurses, (2) support workers, (3) technicians, 4) social and behavioral workers, (5) other diagnosing or treating clinicians (eg, dentists), and (6) physicians (eTable 2 in Supplement 1). Event time was from ACS administration to alcohol-related death, death from other causes, or December 31, 2019, whichever came first, with follow-up between 11 and 12 years for participants still living.
Analyses were performed in SAS, version 9.4 (SAS Institute Inc). Unadjusted and age- and sex-standardized alcohol-related death rates per 100 000 person-years with 95% CIs were calculated (2-sided P < .05 indicated statistical significance). Cox proportional hazards regression models estimated alcohol-related death hazard ratios for the 6 health care groups adjusted for age, sex, race and ethnicity, marital status, educational level, and income. Race and ethnicity were included as covariates given known disparities in alcohol-related deaths among US racial and ethnic groups. We applied ACS weights.6
Results
The median age of the overall cohort was 44 (IQR, 35-53) years; 47.7% were female and 52.3% were male. Group sociodemographic characteristics are presented in Table 1. Unadjusted alcohol-related death rates per 100 000 person-years were significantly lower for each health care worker group than for non–health care workers. Following age and sex standardization, only physicians (4.2) and other diagnosing or treating clinicians (11.6) had significantly lower alcohol-related death rates than non–health care workers (18.1). Controlling for the other sociodemographic characteristics, alcohol-related mortality hazards did not significantly differ between each health care worker group and non–health care workers ...
Language: en |