# Widowhood mortality among married and cohabiting partners: a nationwide study in Finland

## TL;DR

This study finds that losing a cohabiting partner can have a bigger impact on mortality than losing a married partner, especially when the death is unnatural.

## Contribution

The study is novel in comparing widowhood mortality effects from marriage versus cohabitation and examining repartnering outcomes.

## Key findings

- Unnatural deaths of partners lead to higher mortality in both married and cohabiting individuals.
- Widowed cohabiters experience greater mortality risk than those from marriages.
- Repartnering after widowhood is linked to lower mortality compared to remaining unpartnered.

## Abstract

Widowhood mortality is typically studied among those widowed from marriage, with limited knowledge about its effects after the loss of a cohabiting partner. This study examines differences in widowhood mortality by: (1) natural vs. unnatural deaths, (2) marital vs. cohabiting partnerships, (3) repartnering status, and (4) sex.

Data were gathered from the registers of Statistics Finland for all adults aged 40–65 years in 1995 (N = 318,351), encompassing details on family relations and mortality between 1995 and 2020. Using propensity score matching, each widow(er) was matched with four control individuals based on their characteristics. Widowhood mortality was studied with Cox regression models across three time intervals: 0–3, 3–10, and 10–20 years post-widowing.

In the study sample, 53,852 were widowed from a marriage, 214,402 were their controls, 10,069 were widowed from a cohabiting partnership, and 40,028 were their controls. For both women and men, unnatural causes of widowhood led to a higher mortality than natural causes. The association was greater among the widowed from a cohabiting partnership than among the widowed from a marriage. Repartnering led to lower mortality than staying unpartnered after widowhood. The patterns persisted up to twenty years after partner bereavement, although the effects diminished over time.

Our findings imply that interventions for widowed individuals should be tailored depending on the cause of death (natural or unnatural) and the nature of the partnership (married or cohabitating). Future research should examine possible mechanisms to explain these effects, such as potential biological mechanisms and reduced socioeconomic security.