Patient preference for biologic treatments of psoriasis in Japan

Abstract

Psoriasis is a chronic autoimmune disease affecting skin which may also manifest in nails and joints. Several biologic treatments have been approved in Japan for psoriasis. Each biologic has a different profile for efficacy and safety, including different dosing regimens and co-payment considerations which may complicate treatment decisions made by patients and physicians during short consultations. Elucidating patient preference is expected to contribute to shared decision-making between patients and physicians to optimize treatment satisfaction and outcomes. However, the number of studies investigating this in Japan is very limited. The study used a discrete choice experiment methodology to elicit patient preferences for hypothetical options in an experimental framework. Participants were asked to choose their preferred treatment option from two hypothetical choices, defined by different levels of six attributes (i.e. early onset of efficacy, long-term efficacy, sustained efficacy after drug withdrawal, dosing convenience, co-payment and risk of serious infection). The survey included 16 treatment choice scenarios and was completed by 395 participants. Across all participants, the attribute regarded as most important was sustained efficacy after drug withdrawal, followed by dosing convenience, co-payment, long-term efficacy, early onset of efficacy and risk of serious infection. The study found that patients prefer treatments which have durable efficacy and lower treatment burden characterized as fewer injection frequency and lower co-payment. These results may be helpful to understand patient preference for biologic treatments used for psoriasis in Japan and contribute to shared decision-making between patients and physicians to improve patient satisfaction and treatment outcomes.

Keywords: Japan; biologics; discrete choice experiment; patient preference; psoriasis.

Figure 1

Example of treatment choice scenario for discrete choice experiment.

Figure 2

Relative importance of attributes (overall results, n = 395). Relative importance is relatively described values calculated by the distance between the highest and the lowest attribute levels. For example, the distance of sustained efficacy after drug withdrawal is 0.66, addition of the highest level of 0.316 and the lowest level of 0.344. This value is converted to percentage of total which is the sum of all attributes' distance. The higher percentages indicate the more preferred attribute.

Figure 3

Preference weights for attribute levels (overall results, n = 395). Preference weights are showed on the vertical scale, describing how much each level was selected within one attribute. Regarding risk of serious infection, one patient in 100 was the highest and therefore the most preferred level in this attribute. Non‐overlapping error bars indicate statistically significant differences across levels within attributes.

Figure 4

Relative importance of attributes (<60 years, n = 256). Relative importance is relatively described values calculated by the distance between the highest and the lowest attribute levels. The higher percentages indicate the more preferred attribute.

Figure 5

Preference weights for attribute levels (<60 years, n = 256). Preference weights are showed on the vertical scale, describing how much each level was selected within one attribute. The highest value indicates the most preferred level and the lowest value indicates the least preferred level in each attribute. Non‐overlapping error bars indicate statistically significant differences across levels within attributes.

Figure 6

Relative importance for attributes (≥60 years, n = 139). Relative importance is relatively described values calculated by the distance between the highest and the lowest attribute levels. The higher percentages indicate the more preferred attribute.

Figure 7

Preference weights for attribute levels (≥60 years, n = 139). Preference weights are showed on the vertical scale, describing how much each level was selected within one attribute. The highest value indicates the most preferred level and the lowest value indicates the least preferred level in each attribute. Non‐overlapping error bars indicate statistically significant differences across levels within attributes.

Figure 8

Relative importance for attributes (males, n = 297). Relative importance is relatively described values calculated by the distance between the highest and the lowest attribute levels. The higher percentages indicate the more preferred attribute.

Figure 9

Preference weights for attribute levels (males, n = 297). Preference weights are showed on the vertical scale, describing how much each level was selected within one attribute. The highest value indicates the most preferred level and the lowest value indicates the least preferred level in each attribute. Non‐overlapping error bars indicate statistically significant differences across levels within attributes.

Figure 10

Relative importance for attributes (females, n = 98). Relative importance is relatively described values calculated by the distance between the highest and the lowest levels in each attribute. The larger values indicate the more preferred options.

Figure 11

Preference weights for attribute levels (females, n = 98). Preference weights are showed on the vertical scale, describing how much each level was selected within one attribute. The highest value indicates the most preferred level and the lowest value indicates the least preferred level in each attribute. Non‐overlapping error bars indicate statistically significant differences across levels within attributes.