Economic Evaluations of Educational, Physical, and Psychological Treatments for Fibromyalgia

A Systematic Review With Meta-Analysis

Cristina Maria Nunes Cabral; Gisela Cristiane Miyamoto; Katherinne Ferro Moura Franco; Judith Ekkina Bosmans

Disclosures

Pain. 2021;162(9):2331-2345. 

In This Article

Results

Our search strategy identified 2899 studies, of which 703 were duplicates. After screening of titles and abstracts, the full-text of 77 studies were checked, and 11 studies were included. No study was found in protocol registration databases and through hand searching. The main reasons of exclusion were not full article but, for example, conference abstracts, and not economic evaluations (Appendix 2, available as supplemental digital content at http://links.lww.com/PAIN/B305). A flow diagram of the study selection is presented in Figure 1.

Figure 1.

Flow diagram of the study selection. HTA, Health Technology Assessment; NHS EED, National Health Service Economic Evaluation Database.

Overview of the Included Studies

The main characteristics of the included studies are presented in Table 1. The most recent study was published in 2020,[8] and the oldest study was published in 1996.[10] Most studies were conducted in Europe;[8,10,11,14,25,26,28,35,49,55] only 1 study was conducted in the United States.[34] The studies included 2028 participants in total, varying from 33[11] to 600 participants[34] per study. All studies were randomized controlled trials.[8,10,11,14,25,26,28,34,35,49,55] Most studies were performed from both the healthcare and societal perspectives;[11,14,25,26,28,35,49] 2 were performed only from the healthcare perspective[8,34] and 2 only from the societal perspective.[10,55] The time horizon was shorter than 9 months in 5 studies,[8,11,14,26,28] 12 months in 5 studies,[10,25,34,35,55] and 24 months in 1 study.[49] Two studies conducted only a cost-effectiveness analysis,[34,49] 6 studies conducted only a cost-utility analysis,[10,11,25,26,35,55] and 3 studies conducted both analyses.[8,14,28] Nine studies measured utilities to calculate QALYs: 7 studies using the EuroQoL-5 Dimension (EQ-5D),[8,11,14,25,26,28,35] 1 study using the Short-Form 6 Dimension (SF-6D),[55] and 1 study using the Maastricht Utility Measurement Questionnaire.[10] The other clinical outcome measured was the impact of fibromyalgia, assessed in 5 studies using the Fibromyalgia Impact Questionnaire.[8,14,28,34,49] Healthcare costs were assessed in all studies,[8,10,11,14,25,26,28,34,35,49,55] patient and family costs were assessed in 6 studies,[10,11,14,25,49,55] and lost productivity costs were also assessed in 6 studies.[10,14,26,28,35,55]

The included studies compared a psychological intervention with an educational discussion group,[10] pharmacological treatment,[26,28] mindfulness-based stress reduction,[35] another psychological intervention,[8] waiting list,[10,14] and usual care.[25,26,28,35] Two studies compared aquatic program[11] or spa treatment[55] to usual care, one study compared aerobic exercise with multidisciplinary intervention and usual care,[49] and the last study compared social support to social support plus education and the control group (CG).[34] A detailed description of the interventions evaluated in each study can be found in Appendix 3 (available as supplemental digital content at http://links.lww.com/PAIN/B305).

Quality Assessment

Based on the CHEERS statement, most of the economic evaluations[8,11,14,25,26,28,35,49,55] was considered to be of sufficient quality (Table 2). Items referring to title, target population and subgroups, setting and location, comparators, time horizon, choice of health outcomes, measurement of effectiveness, estimating resources and costs, and source of funding were scored as yes in all included studies.[8,10,11,14,25,26,28,34,35,49,55] The item referring to abstract was scored as yes in only 2 studies,[8,26] and the item referring to study findings, limitations, generalizability, and current knowledge was scored as yes in 6 included studies.[8,11,14,35,49,55] Finally, the item referring to the discount rate was scored as not applicable in 4 studies[10,14,34,55] because the time horizon was 12 months or less.

All included studies were considered to have a high overall risk of bias (Table 3). The domain regarding bias in measurement of the outcome was classified as having high risk of bias in all studies because QALYs and impact of fibromyalgia are participant-reported outcomes, and blinding is not possible, while the domain regarding bias in selection of the reported result was classified as having a low risk of bias in all studies.[8,10,11,14,25,26,28,34,35,49,55] Bias arising from the randomization process mostly arose because of the absence of concealed allocation.[10,11,25,34,35] Bias due to deviations from the intended interventions mostly arose because participants and carers were aware of participants' assigned intervention.[8,10,11,14,25,26,28,34,35,49,55] Finally, bias due to missing outcome data mostly arose because data were not available for nearly all participants randomized.[10,34,35,49,55]

Cost-effectiveness of the Educational, Physical, and Psychological Treatments

Table 4 presents the costs and effects outcomes, and the ICERs for the included studies.

Psychological Interventions. Among studies that evaluated a psychological intervention compared with another intervention or usual care/CG,[8,10,14,25,26,28,35] mean healthcare costs in the psychological intervention group ranged between $961 (€698)[8] and $3542 (€3013).[10] From the societal perspective, mean costs in the psychological intervention group varied from $2696 (€2092)[26] to $17,238 (€14,663).[10] Effects also showed high variability, with the incremental QALYs ranging from 0.01[26,28] to 0.12[25] and improvement on the impact of fibromyalgia score ranging from 15.62[28] to 21.07.[14] Incremental cost-effectiveness ratios for QALYs showed that the psychological intervention was dominant (ie, less expensive and more effective) over the comparator treatment from both the healthcare and societal perspectives in 4 studies[25,26,28,35] and from the societal perspective in one study.[14] In one study,[10] the psychological intervention was not cost-effective compared with the control intervention from both perspectives using the NICE willingness to pay threshold. In another study,[14] the psychological intervention was cost-effective compared with control from the healthcare perspective using the NICE willingness to pay threshold. Incremental cost-effectiveness ratios for impact of fibromyalgia showed that the psychological intervention was also dominant compared with the comparator treatment from the societal perspective in one study[14] and from both perspectives in another study.[28] Finally, the most recent study[8] compared 2 psychological interventions, and ICERs for QALYs and impact of fibromyalgia indicated that attachment-based compassion therapy was dominant over the active CG (relaxation) from the healthcare perspective.

Meta-analyses were conducted for incremental QALYs and incremental costs. The first meta-analysis included 2 studies with a total of 217 participants over a 6-month time horizon[26,28] (Figure 2) and showed that the difference in QALYs between psychological treatment and usual care was not statistically significant different, whereas healthcare and societal costs were statistically significantly lower in the psychological intervention group compared with usual care (mean difference: $−2087, 95% CI: −3061 to −1112; mean difference: $−2411, 95% CI: −3582 to −1,240, respectively). Heterogeneity for QALYs was high in this meta-analysis. The second meta-analysis included 2 studies with a total of 216 participants and compared psychological treatment with recommended pharmacological intervention over a 6-month time horizon[26,28] (Figure 3). In this analysis, there was only a statistically significant difference in healthcare costs, which were lower in the psychological intervention group than in the recommended pharmacological intervention group (mean difference: $−1443, 95% CI: −2165 to −721). The same was not found for QALYs and societal costs. For societal costs, heterogeneity was high in this meta-analysis. The third meta-analysis included 2 studies and a total of 352 participants over a 12-month time horizon[25,35] (Figure 4), and showed that healthcare costs in the psychological intervention group were significantly lower than in the usual care (mean difference: $−538, 95% CI: −917 to −158). The same was not found for societal costs.

Figure 2.

Forest plots of the comparison between psychological treatment vs usual care from a 6-month time horizon. CI, confidence interval; QALY, quality-adjusted life-years.

Figure 3.

Forest plots of the comparison between psychological treatment vs recommended pharmacological treatment from a 6-month time horizon. CI, confidence interval; QALY, quality-adjusted life-years.

Figure 4.

Forest plots of the comparison between psychological treatment vs usual care from a 12-month time horizon. CI, confidence interval.

Physical Interventions. Two studies compared exercise therapy with usual care. In the comparison between aquatic exercise with usual care,[11] healthcare and societal costs were higher in the aquatic exercise group (EG) compared with usual care, but no measure of uncertainty was presented. The aquatic EG gained 0.131 QALYs compared with the usual care group, which was a statistically significant difference.[11] Both from the healthcare and societal perspectives, the ICERs were lower than the NICE willingness to pay threshold, indicating that aquatic exercise was cost-effective compared with usual care. In the comparison between aerobic exercise, multidisciplinary intervention, and usual care,[49] costs varied depending on the perspective. The multidisciplinary intervention was associated with the lowest costs from the healthcare perspective and the highest costs from the societal perspective, although not statistically significantly so. In addition, the multidisciplinary intervention resulted in statistically significantly larger effects than aerobic exercise and usual care. Aerobic exercise was dominated by the multidisciplinary intervention and was dominant compared with usual care from the healthcare perspective for the impact of fibromyalgia. From the societal perspective, the ICER for the comparison between aerobic exercise and multidisciplinary intervention showed that to lose 1 point of improvement in impact of fibromyalgia, $564 (€442) is saved in the aerobic EG compared with the multidisciplinary group. For the comparison between aerobic exercise and usual care from the societal perspective, the ICER showed that to gain 1 point of improvement in impact of fibromyalgia, $24 (€19) should be invested in the aerobic EG compared with the usual care group.

Other Interventions. In 2 studies[34,55] evaluating other interventions than psychological interventions, it was not possible to recalculate the ICERs. In the first study,[34] incremental healthcare costs and effects were in favour of the comparison intervention (social support group), although differences were not statistically significant. In the second study,[55] healthcare and societal costs in the spa treatment group were nonsignificantly higher than in the CG, and no significant differences in QALYs were observed.

processing....