Measuring patient-relevant benefits in the treatment of psoriasis with the Patient Benefit Index: development and preliminary validation of a 10-item short form

Dear Editor, Patient-relevant benefits from treatment of inflammatory skin diseases can be measured with the validated 25-item questionnaire Patient Benefit Index Standard version (PBI-S; www.patient-benefit-index.com).^{1, 2} It consists of two parts: the Patient Needs Questionnaire, in which patients rate the importance of 25 different treatment goals, and the Patient Benefit Questionnaire, in which patients evaluate the achievement of these goals by therapy. From all items, an importance-weighted total benefit score is determined. However, shorter instruments are often preferable both in practice and in clinical trials. We therefore developed and tested a 10-item short form of the PBI-S named the ‘PBI-S-10’, which should maintain both content validity and good psychometric properties. The short form should allow for the calculation of a total score that indicates overall patient-relevant benefit and demonstrate high agreement with the long form.

We used data from the German Psoriasis Registry PsoBest at baseline and first follow-up 1–5 months later (n = 4164)³ and PsoHealth4, a cross-sectional study on psoriasis care in Germany (n = 1185).⁴ Ten items for inclusion in the short form were selected by a consensus group (methodologists, dermatologists and statisticians) based on PBI-S item characteristics (PsoBest) as well as clinical criteria. These included average importance rating, correlation with the PBI-S total score, percentage of missing values, inter-item correlations, floor/ceiling effects and clinical relevance. Criterion validity of the short form was evaluated based on its association with the long form [intraclass correlation (ICC) ≥ 0·8; Pearson correlation]. In addition, we determined its internal consistency (Cronbach’s alpha of Patient Needs Questionnaire items), floor and ceiling effects (percentage of lowest/highest possible score), missing values, convergent validity (Pearson correlation of the PBI-S-10 with convergent criteria) and known-groups validity (testing hypotheses on differences in PBI-S-10 in patient subgroups assumed to have different treatment benefit).

The mean age of patients was 48·6 and 50·9 years, respectively, in the two databases; the majority were male (58·6% and 53·3%, respectively). Patients had been diagnosed with psoriasis an average of 18·4 and 20·8 years ago. Both disease severity and quality of life impairment were higher in PsoBest at baseline than in PsoHealth4 [Psoriasis Area and Severity Index (PASI):⁵ 14·9 vs. 6·9; Dermatology Life Quality Index (DLQI):⁶ 11·5 vs. 6·0].

The total score generated from the 10 selected items showed excellent⁷ agreement with the corresponding score of the long form in both databases (ICC = 0·958 and 0·940; r = 0·973 and 0·945, in the PsoBest and PsoHealth4, respectively). Internal consistency was good in both databases (0·794 and 0·816) and floor effects were low (2·86% and 2·19%). Ceiling effects were 8·4% in PsoBest and 13·8% in PsoHealth4, indicating that these patients rated all of the 10 goals that were applicable to them as ‘very’ much achieved.

In PsoBest, the PBI-S-10 correlated in the expected direction with changes in DLQI, PASI, Patient Body Surface Area (PatBSA, a grid scheme in which patients marked the affected areas)⁸ and Patient’s Global Assessment of disease severity (ranging from 0, ‘no skin alterations’ to 10, ‘strongest skin alterations’ on a visual analogue scale), supporting convergent validity (Table 1). Known-groups validity was indicated by strong effects for the PBI-S-10 comparison between patients who achieved vs. did not achieve a DLQI of 0, and between patients who achieved vs. did not achieve PASI 75 (a percentage PASI improvement of 75 or higher). Differences between patients receiving biological vs. conventional systemic treatment were smaller. In PsoHealth4, correlations of PBI-S-10 with DLQI, PatBSA, PASI and Physician’s Global Assessment of disease severity (visual analogue scale with five subdivisions labelled ‘none’, ‘mild’, ‘moderate’, ‘severe’ and ‘very severe’) also supported convergent validity. Patients receiving more potent treatment achieved higher PBI-S-10 total scores.

As a limitation, it should be noted that this study included only patients from Germany. In addition, responses to single items of a questionnaire can be affected by the content and number of other items. This is why future validation studies of the PBI-S-10 that administer the questionnaire in its final form are needed to confirm its psychometric properties. As a further limitation, patient partners were not included in this study; instead, patient preferences for different treatment goals were considered by using their importance ratings for item selection and by asking 10 patients for feedback on the draft questionnaire.

We recommend use of the PBI-S-10 in clinical studies that focus on overall patient-relevant benefit from psoriasis treatment, and for shared decision making if time does not allow for use of the long form. However, in clinical practice, a more differentiated assessment of patient needs including all 25 items of the PBI-S may be advisable in order to align the treatment to all those goals relevant to the person.

Author Contributions

Christine Blome: Conceptualization (lead); data curation (supporting); formal analysis (lead); methodology (lead); project administration (lead); writing – original draft (lead); writing – review and editing (lead). Catharina C. von Stülpnagel: Formal analysis (supporting); methodology (supporting); writing – original draft (supporting); writing – review and editing (supporting). Matthias Augustin: Conceptualization (supporting); data curation (supporting); methodology (supporting); resources (lead); supervision (lead); writing – original draft (supporting); writing – review and editing (supporting). Ulrich Mrowietz: Data curation (supporting); writing – review and editing (supporting). Kristian Reich: Data curation (supporting); writing – review and editing (supporting). Holger Muehlan: Methodology (supporting); writing – original draft (supporting); writing – review and editing (supporting). Natalia Kirsten: Writing – review and editing (supporting). Anna K. Langenbruch: Data curation (equal); writing – review and editing (supporting). Christina Sorbe: Data curation (equal); methodology (supporting); writing – review and editing (supporting).Toni M. Klein: Methodology (supporting); writing – review and editing (supporting).