Single cases are complex: Illustrated by Flink et al. ‘Happy despite pain: A pilot study of a positive psychology intervention for patients with chronic pain’

Abstract

In this issue of the Scandinavian Journal of Pain Ida Katrina link and her colleagues report a short series (n = 4) of single case xperiments testing the potential impact of a positive psychology ntervention for people with chronic pain on self-report measures f affect and catastrophizing [1]. The study is notable for several easons. First, it is among the first to apply positive psychology echniques to chronic pain. Most current psychological methods re guided by the ubiquitous cognitive-behavioural strategy that ocuses on ‘negative’ thinking and appraisal processes that are preumed to be causally related to poor adjustment. The primary aim f CBT is thus to reducing distress and improving function. By way of ontrast positive psychology aims to increase the ratio of positive to egative emotions by strengthening positive affect and well-being. Flink et al. used a set of positive psychology exercises that have een shown to produce beneficial effects with other groups (sumarised in Table 2 of their article). Second, the authors elected o use experimental single case methodology to test the interention. The fundamental features of single case methods are any repeated observations across different conditions, e.g., noreatment and treatment, so that the participant acts as their own ontrol. The test of the impact of the treatment is made by comarison of the measures across control and treatment conditions. hird, Flink et al. attempted a replicated case series rather than a ingle opportunistic case report. They provide an account of the ampling frame and selection of participants; this is not always ound in reports of single case series. Fourth, they made multiple easurements at different ‘levels’ of data. This feature is elaborated ater. Notwithstanding these features this is a difficult dataset to interret unambiguously. Interpretation of the graphical data displays s not easy because there is marked variability within the baseline nd treatment phases and the data plots are complicated by the preentation of multiple measures within the same plot (Figures 5–7). ut there is some evidence of effects, especially for participant 2, ut none of the effects is marked. The merit of this article is the