Treatment pathways and economic analysis of treatment for severe psoriasis
Psoriasis is a chronic skin disease that affects up to 2% of the UK population. The clinical
presentation ranges from mild disease to extensive, severe disease that causes considerable
discomfort and distress. Severe disease usually requires photochemotherapy or systemic
treatment. Information about the effectiveness, safety and costs of the different treatments
is required to enable dermatologists to formulate evidence-based treatment guidelines.
Systematic reviews of the four main treatment modalities for moderate-severe psoriasis
(cyclosporin, methotrexate, systemic retinoids and photochemotherapy) were performed.
Randomised controlled trials were located systematically by electronic searching, hand
searching and personal communications. Data on trial characteristics and outcomes were
extracted and tabulated. Where possible data were pooled to give summary effect sizes as
odds ratios, rate differences or numbers needed to treat (NNTs). Firm RCf evidence of
efficacy was found for cyclosporin, oral ret.inoids, particularly in combination with PUV A,
phototherapy, photochemotherapy and for combinations of topical calcipotriol or steroids
with phototherapy. The corresponding NNTs were low, indicating high levels of efficacy.
RCI' evidence of efficacy is lacking for methotrexate. Two observational studies of
patients attending the Psoriasis Specialty Clinic were performed. The first was a crosssectional
study that used data in existing disease assessment docwnentation to identify the
characteristics of a group of 256 patients. The second was a longitudinal study that
followed the treatment pathways of 166 patients in the first group. These studies confirmed
that this group of patients and their treatments were comparable with those described in the
literature. An economic analysis was performed, using a previously published decisionanalytic
model, to compare four treatment strategies for severe psoriasis from the health
service perspective. The results (cost-effectiveness ratios) showed that methotrexate was
the most cost-effective primary treatment followed by cyc1osporin, acitretin and PUV A.
The rank order was not sensitive to changes in response rates. Modifications to the decision
analytic model are proposed including a wider array of pathways and an allowance for
adverse effects of treatment. Future analyses should include narrowband UVB alone as a