The UFS-QOL demonstrated responsiveness to change among women treated with MRgFUS, suggesting that the instrument is sensitive to change in patient-perceived Symptom Severity and HRQL. The UFS-QOL was able to detect patient improvement from baseline to 3 months and baseline to 6 months, with mean change scores greater than or equal to 20 points for all subscales. The large change scores are consistent with the large effect sizes observed for the Symptom Severity and HRQL subscales, all of which were 0.80 or higher at six months with the exception of Sexual Function, which was 0.70. Using Cohen's guide for social phenomena , these values can be interpreted as large for Symptom Severity, Concern, Activities, Energy/Mood, Control, Self-consciousness, and total HRQL, and medium for Sexual Function. Our findings are consistent with those recently reported by the Fibroid Registry Outcomes for Outcomes Data (FIBROID), a registry study of over 2,000 patients undergoing uterine embolization for leiomyomata . Six-month follow-up data from this study also reported large change scores of more than 35 points from baseline for both Symptom Severity and Total HRQL.
When using ancillary measures as benchmarks for patient improvement, the UFS-QOL was able to discriminate among levels of change in patient-reported overall treatment effect and satisfaction. With respect to overall treatment effect, the UFS-QOL was able to demonstrate significantly greater improvements in Symptom Severity and HRQL among those who reported that their fibroid symptoms improved compared to those who reported no improvement. The UFS-QOL was also able to discriminate between patients who were satisfied with treatment (compared to those who were not) and between patients who reported that their treatment was effective (compared to those who did not) in all subscales of HRQL. The UFS-QOL was not able to discriminate between patients who indicated that they would recommend the treatment to a friend compared to patients who would not, but this may be an indication that this particular item is not a good measure of patient satisfaction.
In prior research, the SF-36 subscales Social Function, Bodily Pain, and Vitality have been shown to be responsive to hysterectomy [28, 29]. In this study, effect sizes for these subscales ranged were 0.7, 0.9, and 0.9 for Social Functioning, Bodily Pain, and Vitality respectively, indicating moderate to large effect sizes. In addition, all SF-36 subscales with the exception of General Health were able to detect changes in patient perceived health status over time. While the SF-36 was able to demonstrate adequate responsiveness to change in the health status of patients treated for uterine fibroids, our findings suggest that the disease-specific UFS-QOL with its focus solely on UF symptoms was clearly more responsive with change score differences and effect sizes of a much greater magnitude compared to the SF-36.
The approach we used to assess the responsiveness of the UFS-QOL is consistent with that outlined by Juniper et al. [24
] by addressing the following questions:
In patients who truly change their health status, can we measure this change by comparing baseline and follow-up scores?
Is the instrument able to distinguish between patients whose health status changes and those who remain stable?
What is the magnitude of the instrument's responsiveness?
A major strength of this study is the demonstrated responsiveness of the UFS-QOL in several different analyses. Greater confidence can be placed in the interpretation of clinically significant effects with consistent evidence from multiple sources. In the current study, the UFS-QOL Symptom Severity and HRQL subscale scores between consecutive assessments demonstrated improved symptoms and HRQL for within-patient comparisons. Discriminant ability with patient perceptions was also observed in addition to the large effect sizes. The consistency of our findings strongly suggests that the observed score changes are valid and clinically meaningful.
Due to ethical considerations, a placebo group (or sham treatment) could not be included in this study. As such, a limitation of this study is that change in the UFS-QOL due to placebo effect cannot entirely be ruled out, although follow-up of this patient cohort shows that the reductions in symptom severity appear to be maintained over time. Future research examining the responsiveness of the UFS-QOL with placebo-controlled alternative therapies for which the uterus remains in tact is needed to address this limitation.
Patient-reported outcomes of symptom severity and HRQL are becoming increasingly important for evaluating treatment of uterine fibroids within the clinical setting. With the increasing availability of noninvasive therapies to hysterectomy, it will be especially important to assess symptom reduction of uterine fibroids and evaluate the clinical success of patients who choose these treatment options. Psychometrically sound patient-reported outcome measures, with demonstrated reliability, validity, and responsiveness, are needed to evaluate interventions among this patient population. The current study indicates that the UFS-QOL is responsive to change in patient health status and able to discriminate between those who improve with treatment for uterine fibroids and those who do not. Consequently, the UFS-QOL appears to be a useful evaluative tool for assessing symptoms and HRQL in studies among patients with uterine fibroids.