The association between psoriasis, diabetes mellitus, and atherosclerosis in Israel: a case-control study

Previous reports demonstrated an association between psoriasis and other diseases including heart failure and diabetes mellitus.

Our aim was to describe the association between psoriasis, diabetes mellitus, and atherosclerosis in Israel.

A cross-sectional study was performed utilizing the database of Maccabi Healthcare Services (MHS), a large health provider organization in Israel. Case patients were defined as subjects who were diagnosed with psoriasis. Patients with diabetes and atherosclerosis were identified by using the MHS diabetes and cardiovascular registries, respectively. The control group included MHS enrollees without psoriasis. The proportion of diabetes and atherosclerosis among case and control groups was compared. Chi-square tests were used to compare categorical parameters. Logistic regression models were used for multivariate analyses.

The study included 46,095 patients with psoriasis (case patients) and 1,579,037 subjects without psoriasis (control patients). The age-adjusted proportion of diabetes was significantly higher in psoriasis patients as compared with the control group (odds ratio [OR] 1.27, 95% confidence interval [CI] 1.1-1.48). The age-adjusted proportion of atherosclerosis was significantly higher in psoriasis patients as compared with the control group (OR 1.28, 95% CI 1.04-1.59). In patients with psoriasis, a multivariate logistic regression model demonstrated an association between diabetes and the multiple use of very potent topical steroids (P < .05) or use of systemic medication for psoriasis (methotrexate, cyclosporine or acitretin) (P < .001). A similar model demonstrated an association between atherosclerosis and the use of phototherapy (P < .001).

Our study was based on a computerized database. The diagnosis of psoriasis was based on digitally transmitted data. Therefore overestimation (false-positive cases) and underestimation (false-negative cases) of psoriasis patients may exist, thereby being a source for information bias. A second limitation is selection bias that may occur due to the possibility that reporting of both psoriasis and associated illnesses is higher in individuals who are seeking medical care. A third limitation concerns the causal effect between occurrence of psoriasis and atherosclerosis or diabetes. The dataset of MHS records diagnoses only from 1997 and does not record the date of disease onset.

Our study supports previous reports for an association between psoriasis and atherosclerosis and psoriasis and diabetes. Further study is needed to support this observation.