Psychological Distress and Multimorbidity in Primary Care

Fortin and colleagues have shown—quite elegantly—that, compared to a count of diseases, the CIRS accounted for more variance in psychological distress among patients with multimorbidity.(1) The study also showed the results to be uncertain for the count of diseases. The values in the confidence interval for the count of diseases go from a 3% reduction to a 29% increase in the odds associated with an increase of one disease. We cannot, with these results, reject the null hypothesis of no relationship between a count of diseases and psychological distress. But we don’t accept it either. The study does not discourage further studies on the potential usefulness of a simple count of diseases as a correlate or predictor of psychological distress. As Altman and Bland would have it: the absence of evidence is not evidence of absence.(2)

What I find most interesting is that the study compares two kinds of tools; we may call them two ways of knowing about the presence or absence of psychological distress among patients. Each of these tools comes with a mini-theory attached to it. The first, a simple count, is intuitive, a bit metaphorical in its theory: more is up and more of a bad thing can only lead to more bad things; the second, more sophisticated, does not give equal importance to all diseases, provides finer distinctions among several domains, and helps assess severity. Its mini-theory would be something like: the finer the distinctions among domains of illness, the finer the distinctions among corresponding outcomes. With a closer look at the 14 domains covered by the CIRS, we could guess that the CIRS would probably account for a higher variance in say cardiac outcomes than in social or quality of life outcomes of multiple illnesses. It seems likely that by adding a rating of the psychological domain to the calculation of the CIRS, the proportion of the variance in psychological outcomes would increase proportionately.

And that is why I would probably use the CIRS to adjust for patient complexity or study the relationship between psychological distress and multimorbidity.

Questions remain though: What do patients do in all of this? They are more likely to use their intuitive sense than the CIRS to assess their illness status. What do doctors prefer to use? Their cultivated clinical sense—the art of clinical practice—or more sophisticated technology? Can the two ways of knowing ever converge?

1. Fortin M, Bravo G, Hudon C, Lapointe L, Dubois MF, Almirall J. Psychological distress and multimorbidity in primary care. Ann Fam Med, 2006;4:417-422.

2. Altman DG, Bland JM. Absence of evidence is not evidence of absence. BMJ, 1995;311:485

Competing interests:   None declared

Fortin and colleagues have presented a very relevant and well performed study. It shows the importance of multimorbidity in terms of psychological distress. Furthermore, it once more shows that a simple count of diseases as a measure of multimorbidity is not specific enough to point out certain relations. Accounting for disease severity is one possible way to overcome this problem.

We have recently studied psycho-social characteristics in relation to multimorbidity, distinguishing between general susceptibility and disease related susceptibility(1). We hypothesized that taking into account known (pathophysiological) relations (i.e. disease related susceptibility) between diseases would result in stronger and more specific relations compared to combinations of disease that are not (yet) known to be related (i.e. general susceptibility). Despite some limitations, our study showed this distinction to be both feasible and promising.

In my opinion one of the other major challenges in multimorbidity research is to apply longitudinal databases, containing data on morbidity and mortality as well as data on psychological, sociological and demographic patient characteristics. Psychological factors could act both as a cause and a consequence of multimorbidity, even reinforcing each others effects. It would therefore be worthwhile to establish a longitudinal study with a sufficiently long follow-up period, monitoring both (multi-)morbidity and psychological characteristics.

1. Van den Akker M, Vos R, Knottnerus J. In an exploratory prospective study on multimorbidity general and disease related susceptibility could be distinguished. J Clin Epidemiol 2006;59(9):934-9.

Competing interests:   None declared