European Journal of Health Economics最新文献

Background: The EQ-5D-Y-3L is a generic, preference-accompanied health measure intended for pediatric populations. EQ-5D-Y-3L health states are valued using the perspective of a hypothetical 10-year-old child ('child perspective') rather than adults valuing for themselves ('adult perspective'). The perspective used has been shown to influence valuation outcomes, affecting comparability of health utilities. This study explored within-respondent differences in values between adult and child perspectives using data from Hungary.

Methods: A secondary analysis was conducted using composite time trade-off (cTTO) data from the Hungarian EQ-5D-Y-3L valuation study. Two hundred adults valued 10 health states from the child perspective and four from the adult perspective. The cTTO values for the matched health states (valued from both perspectives) were compared, with differences analyzed using t-tests and random-intercept regression. Associations with respondent characteristics were also explored.

Results: Differences in cTTO values were observed between perspectives, particularly for more severe health states. Compared to the adult perspective, the child perspective yielded significantly lower values for worse-than-dead observations, but higher values for better-than-dead observations. After adjusting for within-subject variation and respondent characteristics, perspective was not a significant predictor of cTTO values. Instead, differences were partly explained by education, region of residence, parental status, and the view that a child's life is more valuable than an adult's.

Conclusions: This is the first study to explore perspective differences in EQ-5D-Y-3L health state valuation within respondents using nationally representative data from outside Western Europe. The findings highlight the importance of considering individual-level attributes in pediatric health valuation.

Background: Smoking is one of the main causes of preventable disease and premature death. While existing evidence syntheses the cost-effectiveness of smoking cessation interventions for hospitalized patients and high-risk groups such as pregnant women or children, the evidence on the cost-effectiveness of non-pharmacological interventions for the general population remains relatively underdeveloped.

Methods: A systematic literature review was performed using MEDLINE, EMBASE, Cochrane Library, and NHS Health Economic Evaluation Database along with grey literature, pre-prints, and HTA reports.

Results: A total of 9,541 abstracts were screened, with 23 studies meeting the eligibility criteria. These studies focused on four main intervention types (i) face-to-face counseling, (ii) digital/telephone counseling, (iii) reimbursement, and (iv) awareness building. The ICERs ranged from - 332,320 EUR/QALY to 156,310 EUR/QALY. Of the 39 ICERs reported, 30 (76.9%) demonstrated superior cost-effectiveness for smoking cessation interventions. Nine studies reported strong dominance, where the intervention not only gained QALYs but also saved costs. On average, 0.02 QALYs (SD = 0.02) were gained per person. When Life-Years Saved (LYS) were used as the effectiveness measure with a range from EUR 192/LYS to EUR 17,908/LYS. All ICERs werebelow the EUR 25,000/LYS willingness to pay threshold.

Conclusion: The evidence suggests that smoking cessation interventions are general cost-effective in Europe. Personal counseling appears crucial for digital interventions to demonstrate cost-effectiveness. Also, awareness building could serve as a cost-effective means of supporting existing cessation programs.

Background: Since the introduction of the EQ-5D-Y-3L valuation protocol, a considerable number of EQ-5D-Y-3L value sets have been published. This provides an opportunity to explore the differences and similarities between EQ-5D-Y-3L value sets across countries, and their similarity to their EQ-5D-5L counterparts.

Methods: EQ-5D-Y-3L value set publications for 11 countries identified key methodological, sampling and value set characteristics. Similarity between value sets was assessed using kernel density plots and other key characteristics. Preference patterns between groups of value sets were explored. EQ-5D-Y-3L value set properties were compared with those of EQ-5D-5L value sets from the same country.

Results: All EQ-5D-Y-3L valuation studies used the same DCE design. Six studies used expanded health state designs in the composite Time Trade Off. Analytical strategies differed between studies. Values for state 33333 ranged from - 0.691 (Slovenia) to 0.289 (Japan); the number of negative values ranged from 0 to 21%. Pain/discomfort level 3 received the largest weight in all EQ-5D-Y-3L studies, while self-care level 3 received the smallest weight in 8 out of 11 studies. Similarities were found between European value sets, and between Asian value sets. Value sets for Australia and Brazil had similar scale lengths as the Asian value sets, but differed in other ways.

Discussion: Although substantial differences were observed between EQ-5D-Y-3L value sets (e.g. regarding the length of the value scale), striking similarities between them existed (e.g. pain/discomfort consistently received the largest weight). Comparing EQ-5D-Y-3L value sets to EQ-5D-5L values generally suggests less willingness to trade life years for life quality for children.

Background: International studies have demonstrated hospital-related variations in the outcomes of total hip replacement (THR) based on clinical outcome measures and Patient-Reported Outcomes (PROs). Therefore, this study explored hospital-related variations for THR based on PROs in Germany.

Methods: We performed a multicenter, prospective, longitudinal cohort study. Patients were recruited in ten high-volume hospitals in Lower Saxony, Germany (11/2019-2/2022). We measured the difference between the preoperative and 6-month postoperative absolute scores using the Oxford Hip Score (OHS). Therefore, we employed a multifactorial analysis of covariance model with OHS change from baseline as the dependent variable.

Results: In total, 583 patients (65.54 ± 9.91 years; 62.1% female) were included in our analysis. The unadjusted mean OHS score increased from 21.61 ± 7.63 to 40.75 ± 8.10 points, thus indicating a change of 19.14 ± 9.58 points. Overall, 503 patients (86.3%) experienced a minimal important change (MIC) of at least 9 OHS points from the preoperative period to 6-months postoperative. The adjusted change in OHS points for participating hospitals varied between 13.41 and 17.99 OHS points. We found differences between the top-performing hospital and two hospitals (p < 0.05 each); however, those differences were shown to be below the minimal important difference (MID) of at least 5.22 OHS points.

Conclusions: We identified differences in the quality of care for THR among high-volume hospitals in Germany; however, those differences were below the MID threshold. Furthermore, both higher preoperative OHS scores and lower rates of pain relief usage was associated with lower change scores. We recommend studies to explore hospital-related clinically relevant variation for THR which also include low-volume hospitals and an evaluation of both MIC and MID thresholds in Germany.

Loss of Exclusivity (LOE) marks a turning point for brand-name drugs, triggering rapid share erosion as generics enter the market. Yet existing forecasting methods-often based on simple parametric decay curves-struggle with data scarcity, suffer from low predictive accuracy, and fail to capture counter-trend share rebounds. In this study, we assemble a 20-year panel of chronic-disease LOE events in South Korea (Hypertension, Lipidemia, Diabetes) and evaluate a range of forecasting algorithms, from classical machine learning to cutting-edge neural networks. Our best-in-class N-BEATS model predicts absolute brand share with an RMSE of .034 and MAPE of .073, while a Random Forest achieves an RMSE of .014 and MAPE of .147 for quarter-to-quarter changes. Notably, our approach successfully captures rare post-LOE share recoveries that traditional parametric benchmarks miss. SHAP analysis reveals that the number of generics is the dominant driver of absolute share levels, whereas time since LOE most strongly influences short-term fluctuations; additionally, brand-holder partnerships with local distributors meaningfully affect both outcomes. These findings lay the groundwork for a data-driven, practical forecasting framework-enabling patent holders and generic manufacturers to align lifecycle and market-access strategies with the key levers of post-LOE dynamics-and point toward even more robust insights as larger LOE datasets become available in future research.

Introduction: In the working population, diminished productivity increases the economic burden of ischaemic stroke. This study aimed to estimate costs of paid productivity losses per modified Rankin Scale (mRS) score at different time points up to two-years post-stroke.

Patients and methods: Data from a multicentre cross-sectional study on post-stroke costs were utilised. Only patients with a pre-stroke professional activity were included in this study. Patients participated in a telephone mRS-interview and returned a questionnaire on paid productivity loss due to absenteeism and presenteeism in the preceding three months. Records were categorised into three-month, one-year, and two-year post-stroke time points based on the interview date. Costs of paid productivity loss were calculated using the human-capital approach (HCA) and friction cost method (FCM). Bootstrapping was combined with multiple imputation to derive mean costs per time point and mRS score (0, 1, 2, and 3-5).  RESULTS: Out of 1106 records, 280 respondents had a pre-stroke professional activity. Costs of paid productivity loss according to both the HCA and FCM ranged from €8980 (mRS 1) to €17,620 (mRS 3-5) in the first three months post-stroke. At one- and two years post-stroke, cost estimates according to the HCA were very similar to those at three months, while costs according to the FCM were significantly lower (0-775 euro at one year and 113-1884 euro at two years post-stroke).

Discussion: Costs of productivity loss following ischaemic stroke in the working population are substantial and rise with higher mRS scores.

Conclusion: These estimates may be used to inform health economic evaluations.

Objectives: EQ-5D has separate three-level versions for children/adolescents (EQ-5D-Y-3L) and adults (EQ-5D-3L), assessing the same five dimensions of health-related quality of life (HRQoL) using age-appropriate language. Little is known about how differences in wording affect self-reported HRQoL assessments. This study aimed to compare the measurement properties of the EQ-5D-Y-3L and EQ-5D-3L in an adult general population sample.

Methods: A cross-sectional study was conducted in a Hungarian adult general population sample representative by age and gender (n = 1,196). Measurement properties, including ceiling, floor, informativity, agreement (Kendall's tau) and known-groups validity (effect sizes) based on self-perceived health and chronic conditions were compared across instruments.

Results: EQ-5D-Y-3L and EQ-5D-3L yielded 85 and 47 unique health states, respectively. Identical health profiles were reported by 59.0%. Overall ceiling was lower using the EQ-5D-Y-3L (34.8%) than the EQ-5D-3L (46.8%), with the largest dimension-level difference for EQ-5D-Y-3L worried/sad/unhappy (56.8%) vs. EQ-5D-3L anxiety/depression (71.6%). Relative informativity was higher for all EQ-5D-Y-3L dimensions (0.20-0.75) than EQ-5D-3L (0.18-0.66). Agreement was the weakest for worried/sad/unhappy vs. anxiety/depression (0.636) and the strongest for mobility (0.841). Both instruments showed medium to large effect sizes across known-groups based on level sum scores (EQ-5D-Y-3L: 0.820-2.454; EQ-5D-3L: 0.820-2.696) and index values (EQ-5D-Y-3L: 0.754-2.362; EQ-5D-3L: 0.747-2.365), with EQ-5D-Y-3L showing higher discriminatory power in 62-69% of known groups.

Conclusion: Notable differences emerged between EQ-5D-Y-3L and EQ-5D-3L in an adult general population sample, especially in the mental health dimension, suggesting that transitions between these instruments should be treated cautiously. The EQ-5D-Y-3L may offer advantages in detecting variations in mental health, even in adult populations.

Background: Many studies of the adjuvanted recombinant zoster vaccine (RZV) consider it cost-effective using efficacy estimates from randomized trials (RCTs). However, the effect magnitude differs between RCTs and observational studies, in addition to other input parameters that have shown to greatly impact cost-effectiveness. The aim of the current study is to assess the economic case of the RZV in Sweden and assess at which price the vaccine would be considered cost-effective.

Methods: A decision-analytic model was used to estimate the health economic impact of introducing RZV in Sweden. Five-year age-cohorts were modelled between ages 65 to 100+, comparing the cost-effectiveness of two-dose RZV to no vaccination from a health care perspective, using efficacy data from RCTs and observational estimates. The model was run over a lifetime time horizon with quality adjusted life years (QALYs) as the outcome. Multiple one-way and probabilistic sensitivity analyses were conducted to analyze the impact of parameter uncertainty.

Results: At a willingness-to-pay of 80,000 Euro per QALY, the RZV was cost-effective across cohorts at a price per dose of 80 to 105 Euro in basecase analyses, in contrast to the current market price at 176 Euro. However, due to parameter uncertainty, the price per dose at which the RZV may be considered cost-effective varies between as high as the current market price to less than 10% of that price, depending on which input variables are used.

Conclusion: The price at which the RZV would be considered cost-effective varies greatly, highlighting the need to explore and consider parameter uncertainty in both analyses and procurement negotiations.

Objective: Mental disorders impose a substantial economic burden, making updated cost estimates essential for informed policymaking. This study provides recent cost-ofillness estimates for mental disorders in the general population and examines trends over a 12-year period.

Method: The Netherlands Mental Health Survey and Incidence Study (NEMESIS) is a cohort representative for the Dutch population in which common DSM-IV/DSM-5 mental disorders (mood disorders, anxiety disorders, substance use disorders and attention-deficit/hyperactivity disorder) were diagnosed using the Composite International Diagnostic Interview 3.0. In baseline data of NEMESIS-3 (2019-2022, N=6,194, age 47.9±16.4 years, 50.4% women) we estimated healthcare, productivity, travel costs associated with mental disorders through weighted regression models and compared these estimates with baseline data from NEMESIS-2 (2007-2009; N=6,506, age 44.3±12.5 years, 55.2% women), assuming identical unit prices.

Results: Total annual per capita costs of any mental disorder were €5,630. Productivity costs comprised the largest share of total costs (61%-85%). In all conditions, except substance use disorder, both total annual costs and productivity costs were lower in 2019-2022 compared to 2007-2009. Healthcare costs were substantially lower for mood and substance use disorders in 2019- 2022, even though primary healthcare utilization was increased in 2019-2022.

Discussion: Productivity costs are the primary cost driver of mental disorders. While primary care utilization has increased- possibly due to the introduction of general practice mental health professionals- healthcare resource use has decreased over time. This may indicate a shift in service provision toward primary care from specialized mental healthcare, potentially contributing to more cost-effective mental healthcare delivery.

Background: Post-procedural complications after transfemoral transcatheter aortic valve implantation (TF-TAVI) lead to extended healthcare resource consumption. A comparison with resource consumption after surgical aortic stenosis valve replacement (SAVR) complications has not been conducted.

Methods: The impact of acute kidney injury (AKI), stroke, severe bleeding and permanent pacemaker implantation (PPI) on length of stay, mechanical ventilation > 48 h and reimbursement was analyzed by risk-adjusted linear and logistic regression analyses of all German aortic stenosis TF-TAVI and SAVR cases 2021/2022.

Results: 48,565 TF-TAVI and 9,803 SAVR procedures for aortic stenosis treatment were performed in Germany 2021/2022. The length of stay for TF-TAVI was shorter (10.16 ± 7.19 vs. 13.91 ± 9.82 days, p < 0.001), the rate of mechanical ventilation > 48 h was lower after TF-TAVI (1.3% vs. 7.0%, p < 0.001) and reimbursement was higher for TF-TAVI (26,483 ± 4,487 vs. 20,538 ± 11,748 €, p < 0.001). Length of stay was increased by all investigated complications after TF-TAVI and SAVR (p < 0.001) with the highest increase after bleeding in TF-TAVI. Odds ratios for mechanical ventilation > 48 h were significantly increased for stroke, severe bleeding and AKI (p < 0.001) but not for PPI after both TF-TAVI and SAVR with the highest OR increase after bleeding in TF-TAVI. Reimbursement was significantly increased after TF-TAVI and SAVR by all investigated complications (p < 0.001) finding significantly higher increases after SAVR compared to TF-TAVI for all complications. The total hospital stay after stroke, AKI and PPM was longer for SAVR (p < 0.001), while severe bleeding led to longer total hospital stay after TF-TAVI (p < 0.001). Total reimbursement remained higher for TF-TAVI after all investigated complications (p < 0.001).

Conclusion: Healthcare resource consumption differs between TF-TAVI and SAVR for aortic stenosis treatment also after procedural complications. SAVR is associated with longer hospitalization and more mechanical ventilation, while TF-TAVI is associated with higher reimbursement in the German healthcare system. Complications lead to increased resource use for both procedures with higher extra reimbursement for SAVR and more extra hospital days for TF-TAVI after bleeding reversing the length of stay advantage.