Differences in efficacy between intention-to-treat and per-protocol analyses for patients with psoriasis vulgaris and atopic dermatitis: clinical andpharmacoeconomic implications

Background Pharmacoeconomic outcome research is based on three criteria: (i ) evaluation of objective therapeutic effects; (ii) quality of life; and (i ii) treatment costs. Evaluation of therapeutic effect is mainly based on th e results of clinical trials using objective clinical measures, e.g.: Psori asis Area and Severity Index (PASI) (score for psoriasis vulgaris) and the Severity Scoring of Atopic Dermatitis (SCORAD) (score for atopic dermatitis ). In most studies, only results for a treatment-optimized subpopulation (p atients treated according to the protocol) are presented in publications. T he relevance of such data for daily routine therapy is doubtful. Objectives Our purpose was to investigate the expected loss of effectivenes s of switching from a clinical trial to daily routine therapy for the synch ronous application of narrow-band ultraviolet (UV) B phototherapy (311 nm) and bathing in 10% Dead Sea salt solution (synchronous balneophototherapy) for patients with psoriasis vulgaris and atopic dermatitis. Methods We conducted a multicentre, uncontrolled observational study of out patients. To achieve data for 'clinical trial' and 'daily routine' situatio ns, two populations were compared: (i) all patients strictly treated accord ing to the protocol (ATP) with no protocol deviations (data published in cl inical trials), and (ii) all patients participating in the study who receiv ed active treatment at least once, despite treatment irregularities, non-co mpliance, early withdrawal or other protocol violations [intention-to-treat -population (ITT), model for 'daily routine']. Results A total of 2526 patients were included in the ITT analysis for psor iasis vulgaris (n = 487 for atopic dermatitis), of which 818 patients could be analysed according to protocol (n = 104 for atopic dermatitis). Strikin g differences in the therapeutic effect between both groups (ITT and ATP) w ere found using relative PASI and SCORAD score improvement: 11% (57% 'daily routine' vs. 68% in 'clinical trial') for psoriasis vulgaris and 16% (39% 'daily routine' vs. 55% 'clinical trial') for atopic dermatitis. The main r easons for excluding patients from the 'clinical trial' group were early st udy withdrawal in 29% (atopic dermatitis, 47%) of patients and fewer treatm ents per week than planned in the protocol in 24% (atopic dermatitis, 52%). Conclusions Our data clearly indicate that for the prediction of the therap eutic effect for daily routine therapy the ITT data appear to be more relev ant than the ATP results (i.e. those presented in clinical trials). Althoug h these data are only a first step for evaluating the 'real' therapeutic ef fect of a treatment modality in daily routine, they seem to support the req uirements for ITT analyses in efficacy studies and demonstrate the necessit y of ITT data for pharmacoeconomic evaluation.