Patient Preferences for Topical Psoriasis Treatments are Diverse and Difficult to Predict

Participants aged ≥18 years with a clinical diagnosis of psoriasis vulgaris affecting the scalp plus at least one other skin area, who used prescription psoriasis topical treatments during the previous year, were recruited. We wanted to maximize the number of participants with scalp psoriasis. The scalp is a commonly affected and inconvenient application site that patients dislike treating, so it is important to gain patient insight into treating this area. Participants were excluded if they were allergic to any fragrance-free topical products. To enable comparison across demographic categories, participants were selectively recruited with the aim of having equal proportions in the following categories: male/female; <40/≥40 years of age; working/not working.

Literature on patient preferences for topical psoriasis treatment formulations and previous market research (LEO Pharma) on burden of psoriasis treatment was used to identify 13 topical treatment attributes that, in addition to treatment efficacy, safety and cost, are important to patients [3, 14, 16‐19]. Based on these attributes plus the target attributes for the new gel Applicator [15], a user survey comprising 16 statements was developed against which participants evaluated the three treatment presentations. Some statements were deliberately presented in their negative form to enhance careful reading and consideration of responses. One topical treatment attribute, spreadability, was not reflected in the survey due to the fact that the dedicated spreading surface of the Applicator could confuse responders as to what they were asked to evaluate—the usefulness of the spreading surface or the inherent spreadability of the gel (Table 1).

Participants attended two sessions conducted 5 days apart at a non-clinic location (Fig. 1). During the first instruction session, participants ranked topical treatment attributes from the provided 13-attribute list during a card-sorting exercise (scale 1–13, where 1 is of the highest importance and 13 is of the least importance). Participants were provided with three inactive topical treatments and instructions for use: Cal/BD ointment vehicle in an aluminum tube; Cal/BD gel vehicle in a plastic bottle; and Cal/BD gel vehicle in the Applicator (pre-assembled and primed). They were also provided with a user survey for each treatment presentation and survey completion instructions. On each day of testing, participants applied one treatment presentation to at least one area affected by psoriasis; all participants tested all presentations in an order of their own choice. Participants graded the treatment presentation against the 16 user survey statements on a seven-point scale: totally disagree (1), largely disagree (2), slightly disagree (3), neither disagree nor agree (4), slightly agree (5), largely agree (6) and totally agree (7). During the second feedback session, participants indicated their preference for each of the three topical treatment presentations (1 is first choice and 3 is last choice, or not relevant) in relation to their plaques in the following locations of relevance to them: scalp (hairy/bald), face, neck, chest/stomach (hairy/non-hairy), back/buttocks (hairy/non-hairy), arms (hairy/non-hairy), legs (hairy/non-hairy), hands and ankles/feet.

The study objectives were to determine patient preferences for different topical treatment attributes and to identify patient characteristics that could help guide topical treatment prescription, thereby maximizing patient acceptance.

Data collection complied with the 2014 version of the Legal and Ethical Guidelines for Healthcare Market Research developed by the Ethics Team of the British Healthcare Business Intelligence Association, the UK Data Protection Act 1998 and Directive 95/46/EC on the protection of individuals with regard to the processing of personal data and on the free movement of such data. Independent Ethics Committee approval is not required for market research in the UK. Informed consent was obtained from all patients for being included in the study.

Standard descriptive statistical methods were used as the primary method of data analysis (SAS 9.4, SAS Institute, Inc., Cary, NC, USA). Multiple logistic regression analysis was used to screen for associations between participant characteristics and topical treatment preference by plaque location. Initially, it was planned that participants would be classified as working/not working, but prior to analysis, participants were re-classified as working full time or not working/not working full time. This was considered more relevant from a ‘burden of treatment’ perspective, since part-time work would allow more flexibility and time for topical treatment application.

One hundred and twelve participants were recruited and all completed the study (Table 2). Briefly, the study was performed in two parts: The first group participated from 27 November to 2 December 2014 and the second group 22 to 27 January 2015. Recruitment started in October 2014; participants were recruited by Medicys Ltd through health-care professionals (68%), the Medicys database (21%) and external recruiters with their own contacts (11%). The median age was between 40 and 49 years for both males and females and 49% of the participants had psoriasis for 15 years or more (Table 3). Most participants (94%) experienced face-to-face contact all or a lot of the time in their daily life, and 67% had such contact at leisure time.

Participants had psoriasis on up to 11 skin areas and 80% reported having at least one affected area where they found application of topical treatment difficult (Fig. 2a, b). Upon recruitment, participants were required to have both scalp and at least one other skin area affected by psoriasis. However, at the time of the study, 12% of participants reported not having scalp psoriasis; these patients were not excluded from the analysis, so as not to waste their contributions with regard to their general priorities and evaluation of the products on other skin areas.

In the initial card-sorting exercise, all 13 topical treatment attributes received similar mean ranking scores (range 6.0–8.1; scale 1–13), indicating that participants had very different priorities. In most cases, the proportion of participants who indicated an attribute was of high importance (i.e., ranked 1–3 of 13) was similar to the proportion who indicated the same attribute was of low importance (i.e., ranked 11–13 of 13; Fig. 3). However, there was a numerical difference between participant rankings for some attributes. For example, the fact that participants did not need to touch the product (mean ranking score 6.0) was considered of high importance by 29% of participants, while 14% considered this attribute to be of low importance (Fig. 3). Attributes that relatively few patients considered of neither high nor low importance, such as ‘I can use the same product for all affected areas’, were concluded to be of medium priority overall.

When analyzed by gender, age or employment status, little difference was observed between the proportion of participants rating the same attribute as of low and high importance. However, numerical differences were observed for some attributes. Of note, the product being easy to spread onto the skin was of higher importance (i.e., ranked 1–3) to females (31%) than males (11%), and the fact that the product was not greasy on the skin was of higher importance to participants who worked full time (39%) than those who did not (18%). Precise application (‘only where I want it’) was considered of higher importance to participants under 40 years of age (33%) than older participants (18%), and being able to get the product out of the container easily was considered of higher importance to those who worked full time (28%) than those who did not (13%). More males (28% vs 14% of females) and more participants aged over 40 years (26% vs 14% of participants aged less than 40 years) indicated it was important that the product was absorbed quickly into the skin. Additionally, an acceptable time taken to apply the product was of higher importance to participants who worked full time (22% vs 11% of participants not working or working part time).

Participants’ mean score ratings of the three presentations indicated that the gel presentations matched the topical treatment wish list attributes to a higher degree than the ointment and, in particular, more participants scored the gel Applicator the highest (Fig. 4a, b). For most topical treatment attributes, the participant ratings increased stepwise from the ointment in a tube, to gel in a bottle, to gel in the Applicator (Fig. 4a). A larger proportion of participants largely or totally agreed (score of 6 or 7 on a scale of 1–7) that they can avoid wasting the product when using the gel Applicator compared with the gel in a bottle and the ointment in a tube (Fig. 4a, b). More participants largely or totally agreed it is easy to get the ointment out of the tube compared with getting the gel out of the bottle or the gel Applicator (Fig. 4a, b). Additionally, more participants largely or totally agreed the gel Applicator allows them to apply the product without having to touch it with their fingers compared with the gel in a bottle and the ointment (Fig. 4a, b). Out of the 29% of participants for whom the ‘no touch’ attribute was of high importance, 30% largely or totally agreed that the Applicator provided this option.

Additional analysis was performed to evaluate the extent to which each treatment presentation delivered on each attribute (on a scale of 1–7, where 1 is totally disagree and 7 is totally agree) using data only from those participants who classified the attribute as most important to them (ranking score 1–3). There were no clear differences between this subgroup and the overall study population.

Of the 112 participants, 27 (24%) reported pain, reduced strength or reduced mobility in their hands. For the attributes: ‘I can easily get the product out of the container’; ‘I feel in control of the applied amount’; ‘I can reach to apply the product to all skin areas affected’; ‘It is easy to apply the product only where I want it’; and ‘I can use this product for all my affected skin areas’, analysis of mean rating scores showed no differences between participants with pain or reduced strength or mobility in their hands and the rest of the population.

Participants indicated which topical treatment they preferred for each plaque location of relevance to them. Overall, the gel Applicator was preferred for 8/14 plaque locations, while the ointment was preferred for 5/14 locations (Fig. 5). The two gel presentations combined were preferred for all locations except the neck (ointment preferred by 52%) and hairy back/buttocks (50%). Using the gel Applicator as reference and adjusting for the other participant characteristics, none of the characteristics analyzed (gender, age, work status and hand pain/impairment) showed a strong independent association with patient preference. For treatment of non-hairy back/buttocks, female gender and working full time were independently associated with lower preference for ointment in tube versus gel in Applicator [females: odds ratio (OR) 0.14; 95% confidence interval (CI) 0.02–0.95; working full time: OR 0.06; 95% CI 0.01–0.59]. For treatment of hands, female gender and working full time were independently associated with lower preference for both ointment in tube (females: OR 0.07; 95% CI 0.008–0.50; working full time: OR 0.1; 95% CI 0.01–0.95) and gel in bottle (females: OR 0.02; 95% CI 0.001–0.23; working full time: OR 0.043; 95% CI 0.003–0.66) compared with gel in Applicator. Of note, no presentation was consistently preferred for treatment of hairy versus non-hairy areas. For treatment of psoriasis on a hairy scalp, there was almost equal preference for each of the two gel presentations (bottle 47%; gel Applicator 44%), and very low preference for ointment (9%). Among participants who had applied the gel presentations to hairy scalp, there were no participant characteristics (gender, age, working status, hand pain or hand impairment) independently associated with preference for either treatment presentation.