Value in Health最新文献

Objectives: Value-based healthcare has recently gained recognition. Part of this framework uses the outcome information from daily care. This study evaluated the effects of patients' perceived use of outcome information on shared decision making, patient experiences with healthcare, treatment credibility, and outcome expectations.

Methods: Data were collected from 25 hand surgery and therapy clinics. We created 2 groups based on whether patients indicated that outcome information was used (Outcome Information group) or not (control group) during the clinician consultation. Patients' experiences with healthcare were assessed after the first consultation using a digitally distributed patient-reported experience measure and a questionnaire to measure treatment credibility and expectations. Confounders were controlled for using propensity score matching in a 3:1 ratio. We calculated Cliff's delta as an effect size measure (0.11-0.27 small, 0.28-0.42 medium, and >0.43 large).

Results: After propensity score matching, we included 636 patients in the Outcome Information group and 212 in the control groups, respectively. The Outcome Information group experienced more shared decision making (Cliff's delta 0.33 [0.24-0.40], P < .001) and scored better on all patient-reported experience measure items. Patients in the Outcome Information group had more positive expectations of the treatment outcome (Cliff's delta: 0.21 [0.12-0.29], P < .001) and found their treatment more credible (Cliff's delta: 0.26 [0.18-0.34], P < .001) than those in the control group.

Conclusions: The perceived use of outcome information by patients leads to more shared decision making, better experiences with healthcare, and more positive outcome expectations and treatment credibility. Therefore, we recommend the use of outcome information in daily care to fulfill the promise of value-based healthcare.

Objectives: Our objective was to develop and assess the psychometric properties of relevant bolt-on items for the EQ-5D-5L in patients with rare diseases (RDs).

Methods: Nineteen new EQ-5D-5L bolt-ons were developed based on literature review, expert input, and qualitative interviews and focus groups with patients, caregivers, and representatives of patient associations. A nationwide, cross-sectional, web-based survey in China included patients or caregivers of patients with 31 RDs in China (n = 9190). In each RDs, participants completed the EQ-5D-5L and 3 of 20 (1 existing and 19 newly developed) bolt-ons. Ceiling, explanatory power, convergent, divergent, and known-group validity were examined.

Results: Among the bolt-ons, itching had the lowest ceiling (6.5%), whereas social relationships had the highest (42.2%). The absolute reduction in the ceiling of the EQ-5D-5L with the addition of any bolt-ons was limited, ranging from 0 (respiratory problems) to 8.3% points (isolation). Dignity and vitality resulted in the largest increase in explained variance in EQ VAS. The isolation, fertility, and visual acuity bolt-ons showed good divergent validity from the EQ-5D-5L items. There was strong convergent validity between SF-12 and conceptually related bolt-ons (eg, physical health composite and muscle problems bolt-on). Various bolt-ons improved the known-groups validity in specific patient groups, eg, Huntington's disease (oral expressions), scleroderma (dexterity), myasthenia gravis (muscle problems), neuromyelitis optica and multiple sclerosis (fatigue), Marfan syndrome (self-image), and Pompe disease (safety).

Conclusion: The EQ-5D-5L shows sufficient validity in most RDs, but incorporating relevant, specific bolt-ons could enhance its ability to more comprehensively assess health-related quality of life in these patients.

Objectives: Several trial-level surrogate methods have been proposed in the literature. However, often only 1 method is presented in practice. By plotting trial-level associations between surrogate and final outcomes with prediction intervals and by presenting results from cross-validation procedures, this research demonstrates the value of comparing a range of model predictions.

Methods: Two oncology data sets were used as examples. One contained 34 trials and had an overall moderate surrogate association; the other contained 14 trials and had an overall strong association. The models fitted included weighted linear regression, meta-regression, and Bayesian bivariate random-effects meta-analysis (BRMA).

Results: Predictions from the models showed a high degree of variation when there was a moderate association (surrogate threshold effect of 0.413-0.906) and less variation when there was a strong association (surrogate threshold effect of 0.696-0.887). For both data sets, BRMA provided the most robust results, although informative priors for the heterogeneity distribution were needed for the smaller data set. Weighted linear regression models provided reasonable predictions in cases of moderate association. However, in the case of strong association, Bayesian BRMA demonstrated greater uncertainty in predictions.

Conclusions: Weighted linear regression provides a useful reference because prediction intervals represent 95% of variance in the data. However, the weights used in such a model must include information on follow-up time. In cases with small data sets, and in cases in which there appeared to be a strong association, Bayesian BRMA provided predictions that were more robust than those provided by weighted linear regression.

Objectives: To examine recent economic evaluations and understand whether any type 2 diabetes mellitus (T2DM) screening designs may represent better value for money and to rate their methodological qualities.

Methods: We systematically searched 3 concepts (economic evaluations [EEs], T2DM, screening) in 5 databases (Medline, Embase, EconLit, Web of Science, and Cochrane) for EEs published between 2010 and 2023. Two independent reviewers screened for and rated their methodological quality (using the Consensus on Health Economics Criteria Checklist-Extended).

Results: Of 32 EEs, a majority were from high-income countries (69%). Half used single biomarkers (50%) to screen adults ≥30 to <60 years old (60%) but did not report locations (69%), treatments for those diagnosed (66%), diagnostic methods (57%), or screening intervals (54%). Compared with no screening, T2DM screening using single biomarkers was found to be not cost-effective (23/54 comparisons), inconclusive (16/54), dominant (11/54), or cost-effective (4/54). Compared with no screening, screening with a risk score and single biomarkers was found to be cost-effective (21/40) or dominant (19/40). The risk score alone was mostly dominant (6/10). Compared with universal screening, targeted screening among obese, overweight, or older people may be cost-effective or dominant. Compared with fasting plasma glucose or fasting capillary glucose, screening using risk scores was found to be mostly dominant or cost-effective. Expanding screening locations or lowering HbA1c or fasting plasma glucose thresholds was found to be dominant or cost-effective. Each EE had 4 to 17 items (median 13/20) on Consensus on Health Economics Criteria Checklist-Extended rated "Yes/Rather Yes."

Conclusions: EE findings varied based on screening tools, intervals, locations, minimum screening age, diagnostic methods, and treatment. Future EEs should more comprehensively report screening designs and evaluate T2DM screening in low-income countries.

Objectives: The Weight-Specific Adolescent Instrument for Economic Evaluation (WAItE) is a weight-specific patient-reported outcome measure for use in adolescence, consisting of 7 domains, each with 5 response levels. The objective of this study was to generate a UK value set for the WAItE, enabling the calculation of utility values.

Methods: An online discrete choice experiment (DCE) completed by an adult sample representative of the working population of the United Kingdom was used to estimate the preferences for the 5 levels of the 7 domains. DCE data were analyzed using multinomial and mixed logit models. The latent values were then anchored onto the 0-1 death-full health quality-adjusted life year scale using 2 different anchoring techniques, the time trade-off method and the DCE-visual analog scale method.

Results: A total of 1004 adults from the United Kingdom were included in the final estimation sample for the DCE. From the latent estimates, the majority of the levels of the dimensions followed the monotonic nature of the WAItE; however, some levels of the Tiredness-, Walking-, and Sports-related dimensions were not monotonic and combined to generate the final value set. The results from the time trade-off and DCE-visual analog scale anchoring methods were similar, with values for the PITS state (the worst health state possible from the WAItE) of 0.289 and 0.230.

Conclusions: This study has developed a value set for the WAItE based on the preferences of the UK population, enabling the use of the WAItE in cost-utility analyses of interventions targeting obesity in adolescents.

Objectives: This study aimed to develop a value set for the EQ-5D-5L based on preferences of the general adult population of the United Arab Emirates (UAE).

Methods: The study followed the EuroQol EQ-5D-5L valuation protocol and involved conducting interviewer-administered face-to-face or online interviews in Arabic or English, using the EuroQol Valuation Technology with a sample of 1005 adults representing the UAE general population. Sample recruitment involved a 2-stage quota sampling strategy across the 7 emirates of the UAE, ensuring representation of nationals and expatriates. Various models using composite time trade-off data only, discrete choice experiment data only, and hybrid using both composite time trade-off and discrete choice experiment data were examined, along with various sensitivity analyses to examine the robustness of the models.

Results: The average age of respondents was 39 years (SD 10.8), 44.5% were female, and 11% were UAE nationals. The best-performing model to generate the value set for the EQ-5D-5L was the hybrid tobit model censored at -1.0, corrected for heteroskedasticity. Values ranged from -0.654 for the worst health state (55555) to 1 for full health (11111) and 0.962 for 11211, with 15.3% of predicted values worse than dead. Mobility problems had the largest impact on health state preference values relative to other dimensions.

Conclusion: This value set will facilitate the application and use of the EQ-5D-5L instrument in the UAE population in generating local evidence on the cost-effectiveness of healthcare interventions, as well as to enhance other applications of EQ-5D in population health assessment and health systems.