Skip to content
What Causes What
SMOKE+ INCREASES

Widowhood → All-cause mortality

Higher Widowhood increases All-cause mortality.

In 373,189 elderly US couples followed from 1993 to 2002, a wife's death sat with 18 percent higher all-cause death hazard in the husband (HR 1.18, 1.16 to 1.19) and a husband's death sat with 16 percent in the wife (HR 1.16, 1.14 to 1.17). Nobody assigned widowhood.

Observational · Low · N=373189

The association disappears when you account for other factors, like socioeconomic status or healthy-user bias.

Mechanism

Widowhood is the death of a spouse. Authors name the loss of care and of a person who noticed illness. They also name who married whom and a shared house: the same smoke, the same street, the same money. Elwert and Christakis (2008) followed 373,189 elderly married couples in Medicare from 1993 to 2002 (746,378 people, 317,300 deaths). After age, race, poverty, both spouses' baseline Charlson scores and hospital days, and zip and county context, a wife's death sat with all-cause hazard 1.18 in men (1.16 to 1.19). A husband's death sat with 1.16 in women (1.14 to 1.17). The increment held after almost every cause of the first death. The exception was Alzheimer's or Parkinson's. Cause-specific hazards in the survivor rose for most cancers, infections, and vascular deaths. Causes came from two years of Medicare diagnoses, not the certificate. Stahl, Arnold, Chen, Anderson and Schulz (2017) matched 593 people widowed in the Cardiovascular Health Study between 1989 and 1999 to 593 still-married people of the same age and sex. Over three years, 199 of 1,186 died. Bereavement overall sat at hazard 0.98 (0.74 to 1.30). Men sat at 1.77 (1.14 to 2.75). Women sat at 0.67 (0.46 to 0.97). The rise in men showed without cardiovascular disease. The fall in women showed with it. A CES-D of 8 or more sat with death (1.74, 1.13 to 2.68) and pulled down the men-without-CVD increment.

Caveats

The stamp is smoke because nobody randomized a death. These are cohorts. Observation alone cannot close the backdoor, so the stamp stays smoke. More papers of the same design do not settle the claim. Elwert and Christakis treat the increment as close to causal. The design still does not identify. Couples share a house. Residual smoking, accidents that hit both, and who married whom remain open. Elwert inferred causes from claims, not certificates. Same-day deaths were treated as not widowed. Alzheimer's or Parkinson's in the first spouse did not raise the survivor's all-cause hazard. Stahl is 1,186 people and three years. The overall hazard crosses 1. The sex split is the opposite of Elwert's two increments above 1. Depression scores were missing for 276 people, often because they died before the next visit. Omitted Abell and Steptoe 2021: living alone, with divorce and bereavement lumped in one transition. The exposure is the household, not widowhood. Omitted heat and clozapine files: they do not identify this page. This is a journal entry, not medical advice and not a reason to treat a number.

Effect

odds_ratio = 1.18. Elwert 2008: 373,189 couples. Men HR 1.18 (1.16-1.19) after wife's death; women HR 1.16 (1.14-1.17) after husband's. Observational. Design does not identify.

Nodes

Socioeconomic statusWidowhoodBiology & endocrinologyAll-cause mortality

Named confounders

Cite this page

What Causes What. “Widowhood → All-cause mortality.” https://whatcauseswhat.org/edges/e-widowhood-mortality (atlas updated 2026-09-02).

https://whatcauseswhat.org/edges/e-widowhood-mortality

Papers