Background/Aims: A potential risk of lymphoma associated with the use of topical calcineurin inhibitors is debated. We assessed the risk of lymphoma among patients treated with topical pimecrolimus, tacrolimus or corticosteroids. Methods: We conducted a cohort study using health insurance claims data. Cohorts of initiators of topical pimecrolimus, tacrolimus and corticosteroids, along with cohorts of persons with untreated dermatitis and randomly sampled enrollees were identified from January 2002 to June 2006. Lymphomas were identified using insurance claims and adjudicated by medical records review. We adjusted for confounders by propensity score matching. Results: Among 92,585 pimecrolimus initiators contributing 121,289 person-years of follow-up, we identified 26 lymphomas yielding an incidence of 21/100,000 person-years. This incidence of lymphoma was similar to that among tacrolimus users (rate ratio, RR = 1.16; 95% confidence interval, CI = 0.74–1.82) as well as corticosteroid users (RR = 1.15; 95% CI = 0.49–2.72). All three topical treatments were associated with an increased risk of lymphoma compared with the general population (RR_Pim = 2.89; RR_Tac = 2.82; RR_Cort = 2.10) suggesting increased detection of preexisting lymphomas. Conclusion: This study did not find an increased risk of lymphoma among initiators of topical pimecrolimus relative to other topical agents during an average follow-up of 1.3 years. Longer-term studies may be needed.

This paper explains the surrogate Henderson filters that are used in the X-11 variant of the Census Method II seasonal adjustment program to obtain trends at the ends of time series. It describes a prediction interpretation for these surrogate filters, justifies an approximation to the filters, proposed by Kenny & Durbin (1982), and proposes a further interpretation of the results. The starting point for the paper is unpublished work by Musgrave (1964aMusgrave ( , 1964b. His work has continuing relevance to current seasonal adjustment practice. This paper makes that work generally available for the first time, and reviews and extends it.

No abstract

No abstract

No abstract