Abstract
This cohort study examines the risk of drug overdose death in individuals who live alone.
The increase in US drug overdose deaths has spurred interest in identifying high-risk populations. Being alone while using drugs poses added risks because it decreases opportunities for others to intervene. One study1 reported that 69% of overdose deaths occurred among people using drugs while alone. Living alone might be associated with increased fatal drug overdose risk. In this nationally representative cohort study, we compared overdose death risks among adults who were living alone or with others.
Methods
American Community Survey (ACS) data were linked to National Death Index records from 2008 to 2019 (N = 3 309 000 adults).2 Drug overdose deaths included underlying-cause-of-death codes for unintentional or undetermined intent poisonings with opioid, cocaine, cannabis, hallucinogen, anxiolytic-hypnotic, or psychostimulant codes. Data analysis was performed from May 11 to September 1, 2022. Reporting followed the US Census Bureau Disclosure Review Board guidelines and applied ACS weights. This study follows the STROBE reporting guideline for cohort studies. The New York State Psychiatric Institute Institutional Review Board deemed this study exempt from review as a secondary analysis of deidentified data.
Death rates per 100 000 person-years with 95% CIs were determined overall and by previously defined sociodemographic characteristics.2 Cox proportional hazards models, adjusted for age, sex, and race and ethnicity (included as demographic characteristics), estimated adjusted hazard ratios (aHRs) of overdose death. Event time was from the ACS until overdose death, death from other causes, or December 31, 2019, whichever came first. In separate models, interaction terms tested whether associations between living alone and overdose death were moderated by sociodemographic variables. Two-sided unpaired tests were used, with significance set at α = .05. Analyses were conducted in SAS, version 9.4 (SAS Institute).
Results
The overall drug overdose death rate from 2008 to 2019 was 10.56 per 100 000 person-years, including 11.84 for adults who lived alone and 10.37 for adults who lived with others (Table). After controlling for age, sex, and race and ethnicity, living alone was associated with a 1.42 (95% CI, 1.30-1.56) aHR for drug overdose death. A post hoc analysis limited to 1-year follow-up yielded similar findings for those living alone (aHR, 1.90; 95% CI, 1.37-2.64).
Table. Risk of Drug Overdose Death Among Adults Who Live Alone or With Others Stratified by Sociodemographic Characteristica.
| Characteristic | No. of persons | Drug overdose deaths per 100 000 person-years | aHR for living alone vs with othersb (95% CI) | Interaction termc P value | ||
|---|---|---|---|---|---|---|
| Total adults (95% CI) | Adults living alone (95% CI) | Adults living with others (95% CI) | ||||
| Total | 3 309 000 | 10.56 (10.23-10.90) | 11.84 (10.88-12.86) | 10.37 (10.02-10.73) | 1.42 (1.30-1.56) | NA |
| Age range, y | ||||||
| 18-29 | 574 000 | 16.37 (15.50-17.27) | 13.58 (10.77-16.89) | 16.59 (15.68-17.54) | 0.73 (0.58-0.91) | Reference |
| 30-49 | 1 158 000 | 12.61 (12.03-13.21) | 19.52 (17.29-21.96) | 11.84 (11.25-12.45) | 1.49 (1.31-1.70) | <.001 |
| ≥50 | 1 576 000 | 5.08 (4.71-5.47) | 7.48 (6.49-8.58) | 4.48 (4.10-4.90) | 1.77 (1.49-2.09) | <.001 |
| Sex | ||||||
| Male | 1 566 000 | 14.36 (13.80-14.93) | 17.73 (15.99-19.61) | 13.88 (13.30-14.49) | 1.45 (1.30-1.61) | .82 |
| Female | 1 743 000 | 7.04 (6.66-7.43) | 7.03 (6.05-8.12) | 7.04 (6.63-7.46) | 1.34 (1.14-1.57) | Reference |
| Race and ethnicityd | ||||||
| Non-Hispanic | ||||||
| Black | 279 000 | 9.08 (8.17-10.05) | 10.11 (7.85-12.83) | 8.87 (7.90-9.94) | 1.19 (0.91-1.55) | .26 |
| White | 2 499 500 | 12.19 (11.76-12.64) | 12.00 (10.88-13.20) | 12.22 (11.76-12.71) | 1.33 (1.20-1.47) | Reference |
| Hispanic | 198 000 | 5.81 (5.17-6.52) | 10.56 (7.46-14.52) | 5.45 (4.80-6.17) | 2.21 (1.57-3.11) | <.001 |
| Other non-Hispanic | 332 000 | 6.29 (5.32-7.39) | 16.55 (11.63-22.85) | 5.23 (4.31-6.30) | 3.44 (2.37-4.99) | <.001 |
| Marital status | ||||||
| Married | 1 965 500 | 5.14 (4.83-5.47) | 9.01 (5.37-14.20) | 5.10 (4.79-5.43) | 2.05 (1.29-3.25) | <.001 |
| Separated/divorced | 413 000 | 18.60 (17.40-19.86) | 14.88 (13.11-16.83) | 20.65 (19.08-22.31) | 0.80 (0.69-0.93) | .10 |
| Widowed | 231 000 | 4.85 (3.88-5.99) | 3.44 (2.40-4.79) | 6.71 (5.00-8.80) | 1.28 (0.77-2.12) | .49 |
| Never married | 699 000 | 18.56 (17.72-19.44) | 13.96 (12.30-15.79) | 19.62 (18.66-20.62) | 0.61 (0.53-0.70) | Reference |
| Employment | ||||||
| Employed | 2 091 000 | 7.97 (7.62-8.33) | 9.48 (8.40-10.67) | 7.77 (7.40-8.15) | 1.35 (1.19-1.53) | Reference |
| Unemployed | ||||||
| <65 y | 658 000 | 23.79 (22.71-24.90) | 36.58 (32.32-41.25) | 22.45 (21.35-23.59) | 1.59 (1.39-1.82) | <.001 |
| ≥65 y | 559 000 | 0.86 (0.60-1.20) | 1.83 (1.12-2.81) | 0.49 (0.27-0.82) | 5.83 (2.78-12.24) | <.001 |
| Educational level | ||||||
| Less than high school | 445 000 | 15.20 (14.15-16.31) | 15.85 (12.86-19.34) | 15.11 (14.00-16.29) | 1.59 (1.27-2.00) | .11 |
| High school/GED | 966 000 | 14.39 (13.66-15.14) | 14.59 (12.55-16.85) | 14.36 (13.58-15.17) | 1.45 (1.24-1.70) | .05 |
| Some college/associate degree | 1 003 000 | 11.30 (10.68-11.94) | 14.75 (12.83-16.88) | 10.79 (10.15-11.46) | 1.65 (1.42-1.91) | .41 |
| Bachelor’s degree or higher | 895 000 | 3.17 (2.82-3.54) | 4.85 (3.79-6.13) | 2.87 (2.51-3.26) | 1.74 (1.33-2.27) | Reference |
| Annual income, $ | ||||||
| 0-40 000 (Loss) | 1 028 000 | 15.83 (15.09-16.59) | 14.52 (13.2-15.93) | 16.34 (15.46-17.25) | 1.01 (0.91-1.13) | .04 |
| 40 001-75 000 | 968 000 | 9.99 (9.39-10.61) | 6.63 (5.2-8.33) | 10.39 (9.74-11.06) | 0.69 (0.54-0.88) | <.001 |
| 75 001-125 000 | 791 000 | 7.41 (6.85-8.00) | 6.04 (3.87-8.99) | 7.47 (6.9-8.09) | 1.05 (0.7-1.58) | .03 |
| >125 000 | 522 000 | 6.29 (5.66-6.97) | 10.46 (6.07-16.79) | 6.17 (5.54-6.86) | 2.32 (1.43-3.77) | Reference |
| Geographic residence | ||||||
| Urban | 2 258 000 | 11.18 (10.79-11.58) | 12.58 (12.86-13.74) | 10.95 (10.53-11.38) | 1.37 (1.25-1.51) | .52 |
| Rural | 1 050 000 | 8.55 (7.94-9.20) | 8.37 (12.55-10.58) | 8.57 (7.92-9.25) | 1.38 (1.08-1.77) | Reference |
| Region | ||||||
| Northeast | 630 000 | 11.56 (10.76-12.41) | 13.22 (10.95-15.82) | 11.30 (10.45-12.20) | 1.76 (1.44-2.16) | .02 |
| Midwest | 898 000 | 11.04 (10.32-11.80) | 11.16 (9.29-13.30) | 11.02 (10.24-11.84) | 1.28 (1.06-1.55) | <.001 |
| South | 1 095 000 | 10.64 (10.09-11.22) | 9.76 (8.35-11.34) | 10.78 (10.18-11.40) | 1.14 (0.97-1.33) | <.001 |
| West | 686 000 | 9.21 (8.58-9.88) | 14.98 (12.67-17.58) | 8.44 (7.79-9.13) | 1.89 (1.57-2.26) | Reference |
| Functional disabilitye | ||||||
| Present | 512 000 | 25.34 (23.88-26.86) | 25.29 (22.25-28.62) | 25.35 (23.70-27.08) | 1.58 (1.37-1.83) | Reference |
| Absent | 2 797 000 | 8.48 (8.16-8.80) | 8.29 (7.39-9.26) | 8.50 (8.16-8.85) | 1.18 (1.05-1.32) | .23 |
Abbreviations: aHR, adjusted hazard ratio; GED, general educational development; NA, not applicable.
Limited to adults aged 18 years or older; excludes respondents living in group quarters.
Adjusted for age, sex, and race and ethnicity.
Group × living alone.
Race and ethnicity were classified based on self-identification from options predetermined by the American Community Survey.
Functional disability defined as self-reported serious difficulties with 1 of the following: hearing, vision, concentrating, remembering, making decisions, walking or climbing stairs, dressing or bathing, or independent living.
The aHRs between living alone and overdose death varied with the largest aHRs observed among unemployed adults aged 65 years or older, the other non-Hispanic racial and ethnic group, adults with incomes above $125 000, and Hispanic adults. For adults aged 18 to 29 years, separated or divorced, never married, and with incomes of $40 001 to $75 000, living alone was associated with a lower aHR of overdose death. In keeping with prior research,3 functional disabilities (25.34), unemployment (23.79), and separated or divorced marital status (18.60) were also associated with increased overdose death rates per 100 000 person-years.
Discussion
Living alone was associated with a 42% increased risk of overdose death after adjusting for age, sex, and race and ethnicity. The analysis has limitations. Because the ACS does not measure key overdose risk factors, including substance use disorders,4 nonfatal drug overdoses,5 chronic pain, or mental health conditions, these factors could not be analyzed. There is potential for misclassification of overdose deaths, although our International Statistical Classification of Diseases and Related Health Problems, 10th Revision, Clinical Modification, classification scheme aligns with Centers for Disease Control and Prevention recommendations.6 Although living arrangements can change, a 1-year post hoc analysis yielded similar results. Lack of follow-up assessments prevented examination of shifts in living arrangements. The analysis also did not account for shifting ecologic risks in drug overdose. The results of this study suggest that people who live alone are a potential at-risk population for overdose death.
Data Sharing Statement
References
- 1.British Columbia Coroners Service . Illicit drug overdose deaths in BC: findings of coroners’ investigations. Ministry of Public Safety and Solicitor General. September 27, 2018. Accessed November 10, 2022. https://www2.gov.bc.ca/assets/gov/birth-adoption-death-marriage-and-divorce/deaths/coroners-service/statistical/illicitdrugoverdosedeathsinbc-findingsofcoronersinvestigations-final.pdf
- 2.Olfson M, Cosgrove CM, Altekruse SF, Wall MM, Blanco C. Living alone and suicide risk in the United States, 2008-2019. Am J Public Health. 2022;112(12):1774-1782. doi: 10.2105/AJPH.2022.307080 [DOI] [PMC free article] [PubMed] [Google Scholar]
- 3.Olfson M, Cosgrove C, Altekruse SF, Wall MM, Blanco C. Deaths of despair: adults at high risk for death by suicide, poisoning, or chronic liver disease in the US. Health Aff (Millwood). 2021;40(3):505-512. doi: 10.1377/hlthaff.2020.01573 [DOI] [PubMed] [Google Scholar]
- 4.Bohnert ASB, Ilgen MA, Ignacio RV, McCarthy JF, Valenstein M, Blow FC. Risk of death from accidental overdose associated with psychiatric and substance use disorders. Am J Psychiatry. 2012;169(1):64-70. doi: 10.1176/appi.ajp.2011.10101476 [DOI] [PubMed] [Google Scholar]
- 5.Krawczyk N, Eisenberg M, Schneider KE, et al. Predictors of overdose death among high-risk emergency department patients with substance-related encounters: a data linkage cohort study. Ann Emerg Med. 2020;75(1):1-12. doi: 10.1016/j.annemergmed.2019.07.014 [DOI] [PMC free article] [PubMed] [Google Scholar]
- 6.Rudd RA, Seth P, David F, Scholl L. Increases in drug and opioid-involved overdose deaths—United States, 2010–2015. MMWR Morb Mortal Wkly Rep. 2016;65(50-51):1445-1452. doi: 10.15585/mmwr.mm655051e1 [DOI] [PubMed] [Google Scholar]
Associated Data
This section collects any data citations, data availability statements, or supplementary materials included in this article.
Supplementary Materials
Data Sharing Statement