Demonstration of a link between spouse depression and disability and disease activity of persons with rheumatoid arthritis


  • James C. Coyne

    Corresponding author
    1. University of Pennsylvania School of Medicine, Philadelphia, and Northern Centre for Healthcare Research, University Medical Center Groningen, University of Groningen, Groningen, The Netherlands
    • Department of Psychiatry, University of Pennsylvania School of Medicine, 3535 Market Street, Room 676, Philadelphia, PA 19106
    Search for more papers by this author

In this issue of Arthritis Care & Research, Lam and colleagues report a longitudinal study in which they demonstrate an association between spouse depressive symptoms and disability and disease course in persons with rheumatoid arthritis (PWRA) (1). Their study represents an advance over past studies that have suggested such a link. This study differs by being longitudinal, not cross-sectional. Lam et al linked spouses' self-report of their own mood to PWRA reports of disability and disease activity, circumventing the obvious methodologic problems of an exclusive reliance on the report of the PWRA, and they introduced statistical controls for a full range of confounds. Overall, Lam and colleagues provide a rather stringent test: any accumulative effects of spouse mood on PWRA disability and disease activity apparent at the start of observation is controlled as a potential confound. Given the stability of mood, the results of Lam et al provide what is likely a conservative estimate of effects of spousal mood on the PWRA.

The findings by Lam et al suggest a number of possible mechanisms, and for some they can provide relevant data (1). First, they considered the possibility of a simple mood contagion. Correlations between partners' moods in the face of acute and chronic illness are commonly found (2), and the temptation is to assume that such associations simply indicate some sort of contagion such that one partner's negative distress and demoralization affects the other's mood, perhaps in a reverberating fashion. Lam and colleagues argue that the persistence of their effect after their statistical control procedures suggests that something more than contagion is involved. They can seemingly dismiss initial similarity between partners' moods because such assortative mating effects are modest, and most likely these couples became involved before the onset of significant PWRA disability and disease activity, which directly affects PWRA mood.

Lam et al recommend focusing attention on care giving and support processes, and leave for future research the explication of any underlying biobehavioral mechanisms. Importantly, intimate couples are to a considerable degree interdependent in their social and physical role functioning, and strains can arise when restrictions on one partner impact what the other one can do. This becomes a readily testable hypothesis with richer data concerning daily role functioning of these couples. Also, in the face of an intermittent chronic condition, the quality of the provision of spousal social support can deteriorate (3), and the emergence of spousal hostile criticism can have deleterious effects on PWRA social and emotional functioning, as well as physical effects. Kiecolt-Glaser and colleagues (4) have produced some provocative demonstrations of the effects of hostile marital interaction on immune functioning and inflammation, and this can provide a readily testable hypothesis concerning biobehavioral mechanisms underlying spouse effects on the PWRA health and functioning. I would give investigation of related phenomena in couples with PWRA a priority.

The results from Lam et al offer encouragement for the commitment of further resources to study the means by which PWRA and spouses affect each other. We need to move beyond provocative, but ultimately ambiguous and unsatisfying demonstrations of cross-sectional associations, particularly those limited to the reports of PWRA. Studies need to incorporate daily diaries, momentary assessment methodologies, and observations of laboratory interactions, complete with appropriate stress hormone and immune assessments.

What, if any, are the clinical implications at this point? First, I should dispel any confusion about “depression” as it is measured by Lam and colleagues. Their assessment involves a well-validated self-report measure of depressive symptoms, but elevations on it are not equivalent to a clinical diagnosis of depression. Given the range of scores and the modest specificity of this measure as an indicator of clinical disorder, it is unlikely that many of these spouses were clinically depressed or warranted intervention with medication. At this point, we do not have much to recommend clinically except perhaps a referral for couples' counseling if clinicians observe or hear reports of hostile and unsupportive marital interactions with PWRA. Such observations or reports are sufficient to justify a referral without any expectation of physical health benefits, and who knows, resolution of these couples' problems might just affect disability and disease activity of PWRA. Whether such effects can be convincingly demonstrated is left for future research.