No Association Between Coffee and Caffeine Intake and Risk of Psoriasis in US Women FREE

Psoriasis is an immune-mediated disorder, but the involved genetic and environmental factors remain to be elucidated. The positive and negative effects of coffee and caffeine on psoriasis have been reported previously.^1- 6 Among the positive effects, coffee has anti-oxidative properties that may help quell inflammation¹; topical caffeine has been used for the psoriasis treatment²; and coffee intake may improve the efficacy of methotrexate and sulfasalazine for psoriasis treatment.³ On the other hand, diterpenes present in unfiltered coffee and caffeine may increase serum cholesterol levels and blood pressure¹; exceptionally high caffeine plasma levels were shown to induce an adverse effect of photochemotherapy on psoriasis⁴; and coffee and caffeine have been implicated as contributing to psoriasis and flaring psoriasis phenotypes, although this last association has not been scientifically proven.⁵

It would be of public health significance to elucidate the long-term relationship between coffee and caffeine intake and the risk of psoriasis. Currently, there is a paucity of research on this topic, and the association remains unclear.⁶ Herein, we evaluated the association between consumption of coffee, decaffeinated coffee, and caffeine and the risk of incident psoriasis in women in the United States.

Participants free of psoriasis in 1991 were included from the Nurses' Health Study (NHS) II⁷ and observed until 2005. In 2005, incidence of psoriasis was ascertained by self-report on questionnaires that inquired about the clinician-diagnosed psoriasis and the date of diagnosis. We confirmed the diagnosis by using the Psoriasis Screening Tool questionnaire,⁸ which has 99% sensitivity and 94% specificity.

Participants were asked about their daily intake of foods and beverages during the previous year for specified serving sizes in 1991, 1995, 1999, and 2003. Total caffeine intake was calculated according to the method of the US Department of Agriculture food composition data supplemented with other sources. The caffeine content was assumed as 137 mg per cup of coffee, 47 mg per cup of tea, 46 mg per can or 12-ounce bottle of caffeine-containing soft drink, and 7 mg per 1-ounce serving of chocolate candy. Data were available on coffee and caffeine consumption at baseline and during the follow-up as well as updated cumulative average consumption. To best ascertain long-term effect and reduce within-subject variation, we used the updated cumulative average intake from all available questionnaire cycles instead of using 1-time measurement. In addition, we examined coffee and caffeine intake only at baseline as well as updated intake at individual cycles in secondary analyses.

Cox proportional hazards regression models were used to estimate relative risks (RRs) and 95% CIs. Analyses were updated because the main exposure, outcome, and covariates were all time varying. We had multivariate models with or without smoking. Analyses were conducted using SAS software, version 9.2 (SAS Institute Inc). The study was approved by the institutional review board of Brigham and Women's Hospital. Our receipt of each completed questionnaire implied participant's informed consent of the present study.

A total of 82  539 participants were included. The baseline characteristics of participants by intake of caffeine (in quintiles) are listed in Table 1. Participants with higher consumption of caffeine were more likely to be current smokers and had a higher quantity of alcohol intake.

During 1  140  758 person-years of follow-up, we identified 986 incident cases of psoriasis. Risk of psoriasis was moderately elevated with increasing coffee consumption in the age-adjusted model. However, this trend became nonsignificant after adjustment for smoking. We also evaluated the association between decaffeinated coffee and risk of psoriasis, which was not significant. A trend toward increased risk of psoriasis was observed with higher caffeine intake in the age-adjusted model. The association became null after adjustment for smoking (Table 2).

Stratified analyses did not show significant findings among nonsmokers. Secondary analysis by only using different measures of coffee and caffeine consumption did not reveal material change of the effect estimation (eTable 1 and eTable 2).

In this prospective cohort study, we did not observe a material change of psoriasis incidence associated with coffee or caffeine intake, after adjusting for known confounders. Smoking appears to be the major confounder underlying the observed significant association between coffee and caffeine intake and risk of psoriasis in age-adjusted models. Consistent with published case-control studies, present data did not lend support to the effect of coffee or caffeine intake on risk of psoriasis.⁶ Our study had retrospective characteristics, given that information on psoriasis was collected in 2005, and misclassification is possible. Further studies are warranted to confirm our findings.