Does caregiving shorten life? What studies since 1999 show about the health of family caregivers

For a quarter of a century, people looking after a sick husband, wife or parent have been told that the work could kill them. The warning traces to one study of 819 older Americans, published in 1999. It has been repeated in policy reports, news stories and clinics ever since.

A paper published in September in the journal Biopsychosocial Science and Medicine argues that the warning is aimed at the wrong target. Its authors, led by the neuroscientist Naomi Eisenberger of the University of California, Los Angeles, point to larger studies in which caregivers lived as long as other people, or longer.

The record is more mixed than either the old warning or the new correction suggests. Caregivers do report more stress and depression than other people. The survival advantage seen in several studies weakens under scrutiny. And no study has been able to separate the effect of caregiving from the kind of person who takes it on.

Where the idea that caregiving shortens life came from

The source is a paper in the Journal of the American Medical Association by Richard Schulz and Scott Beach of the University of Pittsburgh. They followed a total of 392 caregivers and 427 noncaregivers aged 66 to 96 years who were living with their spouses, in four US communities.

Participants were sorted into four groups. In the first, the spouse was not disabled. In the second, the spouse was disabled but the participant was not helping. In the third, the participant was helping and reported no strain. In the fourth, the participant was helping and reported mental or emotional strain.

After 4 years of follow-up, 103 participants (12.6%) died. Once age, existing illness and other differences were accounted for, the strained caregivers had mortality risks that were 63% higher than noncaregiving controls.

Three details of that result tend to be lost. The estimate was imprecise: its plausible range started at no extra risk at all and ran to more than two and a half times the risk, so it only just counted as a difference. Caregivers who reported no strain had an estimated risk 8% higher than the comparison group, a gap that could not be told apart from zero. And people whose spouse was disabled but who were not providing care had an estimate 37% higher, which also could not be told apart from zero.

The authors’ conclusion was specific. Being a caregiver under strain, they wrote, is an independent risk factor for mortality among elderly spousal caregivers, meaning a risk factor in its own right. The statement concerned strained, elderly spouses. It traveled as a claim about caregiving.

It traveled far. The paper has since been cited almost 5,300 times, according to Science News, which reported on the new challenge to it this week. The new paper says the claim has shaped public policy, media narratives, and clinical practice for over two decades.

What later studies of caregivers and mortality found

Larger studies followed, and they did not repeat the result.

The biggest came in 2013. Researchers led by David Roth drew on a national US study of stroke risk. They identified 3,503 family caregivers and paired each with a noncaregiver who resembled them on 15 characteristics covering demographics, health history and health habits. Over an average of six years, 264 (7.5%) of the caregivers died, against 315 of the matched noncaregivers, or 9.0%.

The authors’ analysis put the death rate 18% lower for caregivers. Analyses by race, sex, relationship and reported strain failed to identify any subgroups with increased rates of death compared with matched noncaregivers. The authors concluded that providing care is not linked to a higher risk of death in most cases, but may instead be associated with modest survival benefits for the caregivers.

An earlier study, published in 2009, had reached a similar place, and Schulz was one of its authors. Using a national survey of 3,376 older married people, it found that spending at least 14 hr per week providing care to a spouse predicted decreased mortality for the caregiver. The result held regardless of how impaired the spouse was.

By 2015, Roth and two colleagues could count the studies. Since the 1999 paper, they wrote, 5 subsequent studies drawn from the general population have found reduced mortality and extended longevity for caregivers as a whole. They argued that policy reports and media coverage commonly present an overly dire picture of the health risks associated with caregiving.

A pooled analysis published in 2019 combined 12 such studies, seven from the United States and five from the United Kingdom, Japan and Australia. Across them, the death rate was 16% lower in favor of caregivers.

Why the caregiving studies disagree

Three explanations have been offered, and they are not exclusive.

The comparison group. The 1999 study compared strained caregivers with people whose spouses were healthy. Those two groups differ in more than caregiving. One has a seriously ill husband or wife and the other does not. Eisenberger and her colleagues, as Science News reports their argument, point to the group in the original data who had a disabled spouse but gave no care. Those people, the Science News article says, still died at the same rate as caregivers experiencing distress. In the adjusted figures published in 1999, the two estimates were 37% and 63% above the comparison group, with wide and overlapping margins of error. Either way, the pattern would suggest that the hazard lies in having a sick spouse.

The 2009 study was designed around that problem. Traditional investigations, its authors wrote, do not disentangle the effects of providing care from those of being continuously exposed to an ailing loved one. They wrote that it may be premature to conclude that health risks for caregivers are due to providing active help.

Who becomes a caregiver. People in poor health are less able to take on the care of someone else. If healthier people are the ones who become and remain caregivers, caregivers will outlive other people whether or not caregiving does them any good. The 2013 study tried to deal with this by matching on 15 characteristics, but matching can only cover what was measured.

How caregiving is defined. The 2019 pooled analysis found the advantage was fragile. It was not present in the US studies taken on their own. It was also not evident among those studies in which informal caregiving was operationalized precisely, meaning defined as hands-on help with daily tasks instead of a broad yes-or-no question. After weighing the other characteristics of participants, the authors wrote, the results “provided more support for stress theory than the healthy caregiver hypothesis”. Stress theory is the older view that the strain of caregiving wears down health. The healthy caregiver hypothesis is the idea that caregivers do well because healthier people take on the role and keep it up. In other words, once those characteristics were allowed for, the pooled advantage no longer looked solid.

So the larger studies undercut the claim that caregiving kills. They do not establish that it protects.

What studies find about caregivers’ mental and physical health

On psychological health the evidence is consistent, and it is not reassuring. A pooled analysis of 84 papers compared caregivers with other people on several measures. The largest differences were found with regard to depression and stress, where the gaps were moderate in size. The gap in physical health was small. And larger differences were found between dementia caregivers and noncaregivers than for caregivers in general.

The body’s response has been harder to pin down. Long-term stress is widely thought to drive inflammation, the immune system’s alarm state, which is tied to many diseases. Caregivers ought to show it.

A 2020 study in the Proceedings of the National Academy of Sciences looked for it with an unusually careful design. It followed 480 people for nine years, starting when none of them was a caregiver. Along the way 239 took on extensive family care, and they were compared with 241 similar people who did not. The researchers tracked six biomarkers of inflammation, meaning substances in the blood that rise with it.

Only one of the six rose more in the new caregivers, and by a small amount. And no biomarker effects were found for a subset of strained spouse caregivers of persons with dementia, a group of 45 people. The authors said the results suggest minimal systemic inflammation in response to chronic caregiving stress, systemic meaning body-wide.

Why that would be is not known. The gerontologist and clinical psychologist William Haley of the University of South Florida, a co-author of that study, told Science News as much.

The new argument about caregiving: helping versus witnessing

Eisenberger and her two co-authors offer an answer. They note that the mortality claim originated from a single study, and they argue that it has survived because the word caregiving bundles together two different experiences.

One is helping. The authors call it prosocial behavior, meaning action taken for someone else’s benefit. They argue that it engages what they term the mammalian caregiving system, a set of brain and body responses, described so far mostly in rodents, that evolved for looking after helpless young and that can quiet the stress response.

The other is watching someone loved decline and die, which can trigger empathic distress and grief.

In their account, the field has tied the negative effects of caregiving to the act of helping others, when the harm belongs to the witnessing. One consequence, they say, is that helping has gone unstudied as something that might be good for health. The paper ends with a call for systematic investigation of prosocial behavior as a behavioral determinant of health.

That is a proposal, and the paper reports no new study of caregivers. The biology behind it is early. Studies primarily in rodents hint at how that caregiving system works, Science News reports Eisenberger as saying, with human evidence only beginning to emerge.

Eisenberger came to the question as a caregiver. Her father had dementia and died in January. Looking after him was stressful and also meaningful, she told Science News. “It gave me a lot of joy to bring him some happiness,” she said.

She is not prescribing it. “I’m not suggesting everyone go out and engage in informal caregiving,” she said.

What family caregivers report about stress and purpose

The scale is large. In 2025, some 63 million people in the United States, almost a quarter of all adults, provided care to adults or children with a medical condition or disability, according to a survey by AARP, the organization for older Americans, and the National Alliance for Caregiving.

Their answers point both ways. Almost 65 percent of caregivers surveyed reported experiencing moderate or high emotional stress and almost half reported financial difficulties. More than half also said that caring gave them a sense of purpose. Roth and colleagues made the same observation a decade ago: most caregivers also report benefits from caregiving.

Not everyone who studies caregivers welcomes the change of emphasis. Rebecca Utz, a sociologist and gerontologist at the University of Utah who was not involved in the new paper, told Science News that stressing the positives could put scarce support at risk. “I don’t like to take our foot off the pedal and say that caregiving is so positive because it’s hard, and people need support,” Utz said.

The 2015 review made a compatible point from the other direction. A more accurate picture, its authors argued, would allow services to be aimed at the subgroup of caregivers who are highly strained or otherwise at risk.

Where the caregiving evidence runs out

Cause. Nobody can be assigned at random to care for a dying spouse. Every study here observed people who became caregivers for their own reasons, and any of those reasons could also affect how long they lived.

Wellbeing. Living as long as other people is a low bar. The same body of research finds more depression and more stress among caregivers, and the survival studies do not measure what the years were like.

The hardest cases. Averages across thousands of caregivers include many whose duties are light. The people at the center of the original warning, older spouses under heavy strain, are few in any one study. The inflammation study had 45 of them.

Whether helping is protective. The new paper’s central idea rests on laboratory experiments and animal research. It has not been tested by following caregivers.

What became of the original result. It was never as broad as its reputation, and one of its two authors later co-wrote a study pointing the other way. Neither Schulz nor Beach could be reached for comment by Science News.

Larger studies since 1999 have not found that family caregivers die sooner than other people, and several found they lived longer, though none could rule out that healthier people are the ones who become caregivers.