ObjectiveThe objective of this study was to assess the predictive value of a discrete choice experiment (DCE) in public health by comparing stated preferences to actual behavior.Methods780 Type 2 diabetes mellitus (T2DM) patients received a questionnaire, containing a DCE with five attributes related to T2DM patients’ willingness to participate in a combined lifestyle intervention. Panel mixed-multinomial-logit models were used to estimate the stated preferences based on 206 completed DCE questionnaires. Actual participation status was retrieved for 54 respondents based on patients’ medical records and a second questionnaire. Predicted and actual behavior data were compared at population level and at individual level.ResultsBased on the estimated utility function, 81.8 % of all answers that individual respondents provided on the choice tasks were predicted correctly. The actual participation rate at the aggregated population level was minimally underestimated (70.1 vs. 75.9 %). Of all individual choices, 74.1 % were predicted correctly with a positive predictive value of 0.80 and a negative predictive value of 0.44.ConclusionStated preferences derived from a DCE can adequately predict actual behavior in a public health setting.

BackgroundCost-sharing programs are often too complex to be easily understood by the average insured individual. Consequently, it is often difficult to determine the amount of expenses in advance. This may preclude well-informed decisions of insured individuals to adhere to medical treatment advised by the treating physician. Preliminary research has showed that the uncertainty in these cost-sharing payments are affected by four design characteristics, i.e. 1) type of payments (copayments, coinsurances or deductibles), 2) rate of payments, 3) annual caps on cost-sharing and 4) moment that these payments must be made (directly at point of care or billed afterwards by the insurer).MethodsAn online discrete choice experiment was used to assess the extent to which design characteristics of cost-sharing programs affect the decision of individuals to adhere to recommended care (prescribed medications, ordered diagnostic tests and referrals to medical specialist care). Analyses were performed using mixed multinomial logits.ResultsThe questionnaire was completed by 7921 members of a patient organization. Analyses showed that 1) cost-sharing programs that offer clear information in advance on actual expenses that are billed afterwards, stimulate adherence to care recommended by the treating physician; 2) the relative importance of the design characteristics differed between respondents who reported to have forgone health care due to cost-sharing and those who did not; 3) price-awareness among respondents was limited; 4) the utility derived from attributes and respondents’ characteristics were positively correlated; 5) an optimized cost-sharing program revealed an adherence of more than 72.9% among those who reported to have forgone health care.ConclusionsThe analyses revealed that less complex cost-sharing programs stimulate adherence to recommended care. If these programs are redesigned accordingly, individuals who had reported to have forgone a health service recommended by their treating physician due to cost-sharing, would be more likely to use this service. Such redesigned programs provide a policy option to reduce adverse health effects of cost-sharing in these groups. Considering the upcoming shift from volume-based to value-based health care provision, insights into the characteristics of a cost-sharing program that stimulates the use of recommended care may help to design value-based insurance plans.Electronic supplementary materialThe online version of this article (10.1186/s12913-018-3598-4) contains supplementary material, which is available to authorized users.

BackgroundThere is an increasing number of quality indicators being reported publicly with aim to improve the transparency on hospital care quality. However, they are little used by patients. Knowledge on patients’ preferences regarding quality may help to optimise the information presented to them.ObjectiveTo measure the preferences of patients with breast and colon cancers regarding publicly reported quality indicators of Dutch hospital care.MethodsFrom the existing set of clinical quality indicators, participants of patient group discussions first assessed an indicator’s suitability as choice information and then identified the most relevant ones. We used the final selection as attributes in two discrete choice experiments (DCEs). Questionnaires included choice vignettes as well as a direct ranking exercise, and were distributed among patient communities. Data were analysed using mixed logit models.ResultsBased on the patient group discussions, 6 of 52 indicators (breast cancer) and 5 of 21 indicators (colon cancer) were selected as attributes. The questionnaire was completed by 84 (breast cancer) and 145 respondents (colon cancer). In the patient group discussions and in the DCEs, respondents valued outcome indicators as most important: those reflecting tumour residual (breast cancer) and failure to rescue (colon cancer). Probability analyses revealed a larger range in percentage change of choice probabilities for breast cancer (10.9%–69.9%) relative to colon cancer (7.9%–20.9%). Subgroup analyses showed few differences in preferences across ages and educational levels. DCE findings partly matched with those of direct ranking.ConclusionStudy findings show that patients focused on a subset of indicators when making their choice of hospital and that they valued outcome indicators the most. In addition, patients with breast cancer were more responsive to quality information than patients with colon cancer.

Hospital quality indicators provide valuable insights for quality improvement, empower patients to choose providers, and have become a cornerstone of value-based payment. As outcome indicators are cumbersome and expensive to measure, many health systems have relied on proxy indicators, such as structure and process indicators. In this paper, we assess the extent to which publicly reported structure and process indicators are correlated with outcome indicators, to determine if these provide useful signals to inform the public about the outcomes. Quality indicators for three conditions (breast and colorectal cancer, and hip replacement surgery) for Dutch hospitals (2011–2018) were collected. Structure and process indicators were compared to condition-specific outcome indicators and in-hospital mortality ratios in a between-hospital comparison (cross-sectional and between-effects models) and in within-hospital comparison (fixed-effects models). Systematic association could not be observed for any of the models. Both positive and negative signs were observed where negative associations were to be expected. Despite sufficient statistical power, the share of significant correlations was small [mean share: 13.2% (cross-sectional); 26.3% (between-effects); 13.2% (fixed-effects)]. These findings persisted in stratified analyses by type of hospital and in models using a multivariate approach. We conclude that, in the context of compulsory public reporting, structure and process indicators are not correlated with outcome indicators, neither in between-hospital comparisons nor in within-hospital comparisons. While structure and process indicators remain valuable for internal quality improvement, they are unsuitable as signals for informing the public about hospital differences in health outcomes.

Background In patient choice, patients are expected to select the provider that best fits their preferences. In this study, we assess to what extent the hospital choice of patients in practice corresponds with their preferred choice. Methods Dutch patients with breast cancer (n = 631) and cataract (n = 1109) were recruited. We employed a discrete choice experiment (DCE) per condition to measure stated preferences and predict the distribution of patients across four hospitals. Each DCE included five attributes: patient experiences, a clinical outcome indicator, waiting time, travel distance and whether the hospital had been recommended (e.g., by the General Practitioner (GP)). Revealed choices were derived from claims data. Results Hospital quality was valued as most important in the DCE; the largest marginal rates of substitution (willingness to wait) were observed for the clinical outcome indicator (breast cancer: 38.6 days (95% confidence interval (95%CI): 32.9–44.2); cataract: 210.5 days (95%CI: 140.8–280.2)). In practice, it was of lesser importance. In revealed choices, travel distance became the most important attribute; it accounted for 85.5% (breast cancer) and 95.5% (cataract) of the log-likelihood. The predicted distribution of patients differed from that observed in practice in terms of absolute value and, for breast cancer, also in relative order. Similar results were observed in population weighted analyses. Discussion Study findings show that patients highly valued quality information in the choice for a hospital. However, in practice these preferences did not prevail. Our findings suggest that GPs played a major role and that patients mostly ended up selecting the nearest hospital.