Kiecolt-Glaser · Psychosomatic medicine 1997 · Observational laboratory interaction study · n=62

Marital conflict in older adults: endocrinological and immunological correlates.

Level 4 - case-series / case-control

Cross-sectional observational laboratory study examining physiological correlates of marital interaction

PubMed 9251151 · doi:10.1097/00006842-199707000-00001 · record verified 2026-08-26

What was done

Thirty-one older couples (mean age 67 years, married an average of 42 years) were admitted to a clinical research center to evaluate endocrine and immune correlates of marital conflict and satisfaction. Baseline blood samples were drawn via venous catheter for immunological assays (blastogenic response to two T-cell mitogens and latent Epstein-Barr virus antibody titers). Five sequential blood samples were collected across a 30-minute videotaped conflict discussion and a 15-minute recovery period to measure cortisol, adrenocorticotropic hormone (ACTH), norepinephrine, and epinephrine. Videotapes were coded for problem-solving behaviors using the Marital Interaction Coding System (MICS).

What was found

Among wives, marital satisfaction and escalation of negative behavior during conflict accounted for 16% to 21% of the variance in the rates of change of cortisol, ACTH, and norepinephrine, while no significant relationship was observed for epinephrine. In husbands, endocrine measures did not show significant relationships with negative behavior or marital quality. Across both husbands and wives, individuals with poorer immunological responses across the three functional assays displayed more negative behavior during the conflict discussion and reported their typical marital disagreements as more negative compared to those with stronger immune responses.

Why it matters

The study indicates that abrasive behavioral patterns in long-term relationships remain linked to biological stress markers, including neuroendocrine reactivity in women and down-regulated immune function in both sexes, even among older adults married for decades.

Limits

The sample size was small (31 couples). The cross-sectional, laboratory-based design prevents establishing whether negative marital dynamics cause endocrine and immune alterations or reflect broader underlying health or behavioral traits. Absolute baseline and peak hormone or immune assay values, statistical significance levels (p-values), and effect sizes for immunological comparisons were not reported in the abstract.

Cited by