Your Body Keeps the Score… of Bad Relationships<\/p>\n
Most of us know intuitively that difficult people are stressful<\/a>. The coworker who constantly stirs up drama, the parent who turns every phone call into a guilt<\/a> trip, the sibling<\/a> who makes every family gathering tense\u2014we feel the weight of these relationships. But a new study published in PNAS in February 2026 suggests the toll goes far deeper than feelings. It reaches all the way down to your DNA<\/a>.<\/p>\n A team of researchers led by Byungkyu Lee (NYU) and Brea L. Perry (Indiana University) studied more than 2,300 adults in Indiana, ranging in age from 18 to 103. They collected saliva samples and used cutting<\/a>-edge “epigenetic clocks”\u2014tools that measure biological aging by reading chemical modifications on DNA\u2014to assess how fast each person was aging at the molecular level. They also mapped out each participant’s social network<\/a> in unusual detail, asking not just who supports them but also who hassles them.<\/p>\n The results were striking. Each additional person in a subject’s network who created stress in their life (what the researchers called a “hassler”) was associated with 1.5 percent faster biological aging each year. Each hassler was also associated with being about 9 months older, biologically speaking, than peers of the same chronological age. To put that in context, the researchers compared the hassler effect to smoking<\/a> (one of the best-known accelerants of aging) and found it corresponded to about 13\u201317 percent of smoking’s impact on the same biological measures. That’s not trivial, especially because these effects compound over decades.<\/p>\n The study also found that the effects of difficult people extend far beyond epigenetics<\/a>. People with more hasslers reported higher rates of depression<\/a> and anxiety<\/a>, worse self-rated health, higher BMI, greater waist-to-hip ratios, and elevated inflammation scores. Notably, the number of hasslers had no association with adult height\u2014a stable trait that shouldn’t change due to social stress\u2014which gave the researchers added confidence<\/a> that they were observing real biological effects rather than statistical noise.<\/p>\n Not All Hasslers Are Created Equal<\/p>\n Perhaps the study’s most provocative finding concerns which hasslers matter most. The researchers broke down their results by relationship type\u2014spouse, family (kin), and non-family (nonkin)\u2014and found a clear hierarchy.<\/p>\n Family hasslers were the strongest and most consistent predictors of accelerated aging. Having a parent, sibling, or child who regularly makes your life difficult was robustly linked to both faster aging rates and greater cumulative biological age. The researchers explain this with an elegant insight about network structure: Kin hasslers tend to be embedded in your life in ways that make them nearly impossible to avoid. You can’t easily quit your family the way you might distance yourself from a difficult coworker. These relationships carry deep obligations and emotional weight, and when they turn sour, there’s often no escape valve.<\/p>\n Nonkin hasslers (coworkers, roommates, acquaintances) also contributed to cumulative aging, though their effects were weaker. What’s remarkable is that even these comparatively peripheral, weak-tie relationships left a detectable biological mark.<\/p>\n The biggest surprise was spousal hasslers. Despite decades of research highlighting marital conflict as a driver of health decline, spouses who were identified as hasslers showed no significant association with accelerated aging. The authors suggest this reflects the deeply ambivalent nature of marriage<\/a>: A hassling spouse is also, in most cases, a source of companionship, support, and shared daily life. That mix of positive and negative may create a buffer that simply doesn’t exist in other family relationships, where the strain is less likely to be offset by the same intensity of support.<\/p>\n What This Means for You and for How We Think About Loneliness<\/p>\n In recent years, loneliness and social isolation<\/a> have dominated the conversation about relationships and health. The U.S. Surgeon General declared loneliness an epidemic in 2023, and for good reason: The data on isolation and mortality are alarming. But this new study complicates that narrative in an important way. It suggests that it’s not just the absence of relationships that harms us\u2014the presence of bad ones is an independent risk factor. Having a large social network doesn’t necessarily protect you if that network includes people who chronically stress you out.<\/p>\n The study also reveals a troubling pattern of inequality. Hassler exposure isn’t random. Women, daily smokers, people with worse health, and those who experienced more adversity in childhood<\/a> are all significantly more likely to report hasslers in their lives. This creates a kind of double burden: The people already most vulnerable are also the ones most likely to be surrounded by difficult relationships, which in turn may accelerate their decline. The researchers describe this as a form of “relational inequality,” a mechanism through which social disadvantage reproduces itself.<\/p>\n So what can you do with this information? A few practical thoughts emerge. First, it’s worth taking an honest inventory of the relationships in your life and recognizing which ones are sources of chronic stress rather than occasional friction. The study deliberately excluded occasional hassling from its definition: Everyone argues sometimes, and that’s normal. What matters is the persistent, frequent pattern.<\/p>\n Second, the finding about kin hasslers underscores the importance of setting boundaries<\/a> with family, even when those relationships feel obligatory. You may not be able to cut ties, but you can limit exposure and seek support in managing those interactions. Third, the spouse finding is actually encouraging: It suggests that relationships where conflict coexists with genuine closeness and support may not carry the same biological cost.<\/p>\n