Grief leaves a mark on the body. Loss of loved ones and biological age

Among people who had lost more than one loved one, several epigenetic clocks indicated faster biological aging before they had even reached middle age.

Grief leaves a mark on the body. Loss of loved ones and biological age

Table of contents

    Grief may reach deeper than the mind

    The death of a parent, partner, child or sibling is usually considered primarily a psychological and social experience. New findings suggest, however, that losing a loved one may also be associated with changes that can be detected at the biological level.

    In a study involving nearly 4,000 American adults, people who had experienced the death of someone close to them had, on average, an older biological profile according to several epigenetic clocks. The association was stronger among participants who had experienced at least two such losses.

    This does not mean that grief automatically ages every person by a specific number of years. The study identified a statistical association at the population level rather than an inevitable process affecting everyone in the same way. It does, however, suggest that severe social experiences may remain embedded in the body long after the most acute emotional crisis has passed.

    Study details

    The study combined information about major life experiences collected over more than two decades with DNA methylation measurements taken from participants in adulthood.

    • Publication title: Familial Loss of a Loved One and Biological Aging: NIMHD Social Epigenomics Program.
    • Authors: Allison E. Aiello, Aura Ankita Mishra, Chantel L. Martin and colleagues from Columbia University, North Carolina State University, the University of North Carolina at Chapel Hill and the University of Texas at Austin.
    • Publication date: July 29, 2024.
    • Journal: JAMA Network Open, a peer-reviewed open-access scientific journal.
    • DOI: 10.1001/jamanetworkopen.2024.21869.
    • Full text: JAMA Network Open.
    • Study type: An observational cohort study using data from the US National Longitudinal Study of Adolescent to Adult Health, commonly known as Add Health.
    • Follow-up period: Participants entered the project in 1994–1995, when they were attending grades 7 through 12, and were followed until the fifth study wave conducted between 2016 and 2018.
    • Population and sample: The analysis included 3,963 people with complete data on bereavement experiences, covariates included in the statistical models and blood samples suitable for DNA methylation analysis. The median age during the fifth study wave was approximately 38 years.
    • Exposure: The death of a biological parent or parental figure, sibling, spouse or partner, or child. Losses were also classified according to when they occurred: before the age of 18 or during adulthood.
    • Primary outcome: Biological aging assessed using four epigenetic measures: Horvath, PhenoAge, GrimAge and DunedinPACE.
    • Funding: The project was funded primarily by institutes within the US National Institutes of Health. The authors reported that the funders had no role in the study design, data analysis or decision to publish.
    • Conflicts of interest: One author, Daniel Belsky, reported holding a patent related to DunedinPACE and receiving royalties and consulting fees. Other reported disclosures mainly concerned research grants.

    The study did not directly measure the intensity, duration or clinical course of grief. The analyzed exposure was the death of a loved one, so the findings should not be interpreted as measuring the biological effects of every form of sadness or emotional stress.

    More losses were linked to an older biological profile

    By the time of the final assessment, almost 40% of participants had lost at least one loved one. Approximately 29% had experienced one loss, while nearly 10% had experienced two or more.

    The most consistent pattern concerned the number of bereavement experiences. As the number of losses increased, results from three of the four aging measures became less favorable:

    • Participants who had experienced one loss showed an older profile according to GrimAge and a faster pace of aging according to DunedinPACE than those who had experienced no losses. No significant difference was observed for PhenoAge or the Horvath clock.
    • Among participants who had experienced two or more losses, an older biological profile was detected using PhenoAge and GrimAge, together with a faster pace of aging according to DunedinPACE.
    • Parental loss was associated with older PhenoAge and GrimAge results and faster DunedinPACE, regardless of whether the loss occurred during childhood or adulthood.
    • Losses that occurred during adulthood showed stronger associations with the analyzed biomarkers than losses experienced during childhood or adolescence.

    The coefficients presented in the publication were expressed as standardized differences rather than a specific number of additional “biological years.” It is therefore not possible to conclude that one bereavement ages the body by one, two or any other fixed number of years.

    Instead, the findings suggest a cumulative effect: the greater the burden of loss, the less favorable the average biological aging profile.

    Not all epigenetic clocks showed the same result

    One of the most important aspects of the study is that the results differed depending on the measure used. Associations with loss were observed for GrimAge, PhenoAge and DunedinPACE, but not for the Horvath clock.

    This does not necessarily represent a contradiction. Individual epigenetic clocks were developed to measure slightly different aspects of aging:

    • The Horvath clock was designed primarily to predict chronological age based on DNA methylation. It may be less sensitive to environmental stressors and processes directly related to disease risk.
    • PhenoAge was developed to reflect a broader biological aging phenotype and incorporates patterns associated with health and disease risk.
    • GrimAge is linked to outcomes including disease and mortality risk, which may make it more sensitive to the biological consequences of long-term stress.
    • DunedinPACE does not estimate biological age in years. Instead, it estimates the pace at which the body is aging over a given period.

    The absence of an association with the Horvath clock limits the ability to claim that bereavement affects every dimension of epigenetic aging. At the same time, the agreement between three newer measures suggests that the observed signal did not depend on a single algorithm.

    How grief could be linked to biological aging

    The study cannot identify a specific mechanism, but the authors describe several pathways through which losing a loved one could affect health.

    The first involves chronic psychological stress. Prolonged activation of the stress-response system may influence hormonal regulation, inflammation, immune function and cellular repair processes.

    A second pathway involves behavioral changes. After the death of a loved one, sleep, diet, physical activity and engagement with preventive healthcare may deteriorate. Alcohol or other substance use may also increase in some people.

    A third mechanism is social and economic. The death of a partner, parent or another family member can also mean the loss of emotional support, household income, childcare or practical help with daily life.

    In practice, these factors may overlap:

    • Psychological distress may disrupt sleep and recovery, making it more difficult for the nervous and hormonal systems to return to equilibrium.
    • Reduced social contact may intensify stress, especially when the deceased person had been the main source of emotional or practical support.
    • Financial difficulties may restrict access to healthcare, rest and nutritious food, potentially prolonging the biological consequences of the crisis.
    • Additional losses may occur before the body has recovered from earlier ones, which could help explain the stronger association observed among people who experienced at least two losses.

    These are plausible mechanisms, however, rather than processes directly demonstrated by this analysis.

    Key limitations

    The findings are important, but they do not prove that losing a loved one directly causes accelerated aging. The study was observational, which means that the influence of other factors cannot be completely excluded.

    The main limitations include:

    • DNA methylation was measured at only one point in time. It is therefore unknown what participants’ epigenetic profiles looked like before the loss or how quickly they changed afterward.
    • The course of grief was not assessed. The researchers did not have detailed information about its intensity, duration, coping strategies, social support or use of psychological care.
    • Different types of loss were combined into broader categories. The death of a child, partner, sibling or parent may have different consequences that could not be estimated precisely because some subgroups were small.
    • Familial factors may have influenced the results. The premature death of a parent may be associated with inherited or environmental health risks that also affect biological aging markers in their children.
    • The findings relate to a specific US cohort. Although the sample was weighted to reflect the population represented by the Add Health study, the results should not automatically be generalized to every country or social group.
    • Not all epigenetic clocks confirmed the association. The lack of an effect in the Horvath clock shows that the result depends partly on which dimension of biological aging is being measured.

    The statistical models accounted for factors including age, sex, race and ethnicity, parental education, neighborhood poverty, household size and caregiver smoking. These adjustments reduce the risk of confounding but cannot eliminate it entirely.

    What the study means

    The central conclusion is not that grief inevitably accelerates aging. Instead, the results suggest that the loss of loved ones may be one of the life experiences associated with long-term biological burden, particularly when several losses accumulate.

    Epigenetic clocks are also not a verdict or a clinical diagnosis for an individual person. Their results can reflect many factors, including metabolic health, smoking, stress, social conditions, physical activity and other exposures that accumulate over many years.

    The study does, however, strengthen the argument that support after bereavement should not be viewed solely as a matter of emotional comfort. Access to close relationships, psychological care, financial stability and the ability to return to healthy routines may also matter for long-term physical health.

    Further research is needed in which biological measurements are collected both before a loss and repeatedly afterward. Only this type of study can determine whether the observed changes truly emerge as a consequence of bereavement, how long they persist and whether appropriate support can reduce or reverse them.

    Sources