A spatial obesity risk score for describing the obesogenic environment using kernel density estimation: development and parameter variation.

BACKGROUND: Overweight and obesity are severe public health problems worldwide. Obesity can lead to chronic diseases such as type 2 diabetes mellitus. Environmental factors may affect lifestyle aspects and are therefore expected to influence people's weight status. To assess environmental risks, several methods have been tested using geographic information systems. Freely available data from online geocoding services such as OpenStreetMap (OSM) can be used to determine the spatial distribution of these obesogenic factors. The aim of our study was to develop and test a spatial obesity risk score (SORS) based on data from OSM and using kernel density estimation (KDE). METHODS: Obesity-related factors were downloaded from OSM for two municipalities in Bavaria, Germany. We visualized obesogenic and protective risk factors on maps and tested the spatial heterogeneity via Ripley's K function. Subsequently, we developed the SORS based on positive and negative KDE surfaces. Risk score values were estimated at 50 random spatial data points. We examined the bandwidth, edge correction, weighting, interpolation method, and numbers of grid points. To account for uncertainty, a spatial bootstrap (1000 samples) was integrated, which was used to evaluate the parameter selection via the ANOVA F statistic. RESULTS: We found significantly clustered patterns of the obesogenic and protective environmental factors according to Ripley's K function. Separate density maps enabled ex ante visualization of the positive and negative density layers. Furthermore, visual inspection of the final risk score values made it possible to identify overall high- and low-risk areas within our two study areas. Parameter choice for the bandwidth and the edge correction had the highest impact on the SORS results. DISCUSSION: The SORS made it possible to visualize risk patterns across our study areas. Our score and parameter testing approach has been proven to be geographically scalable and can be applied to other geographic areas and in other contexts. Parameter choice played a major role in the score results and therefore needs careful consideration in future applications.

Does uptake of specialty care affect HRQoL development in COPD patients beneficially? A difference-in-difference analysis linking claims and survey data.

BACKGROUND: There is an evidence gap on whether the choice of specialty care beneficially affects health-related quality of life (HRQoL) in patients with chronic obstructive pulmonary disease (COPD). This study analyzes how newly initiated pulmonologist care affects the generic and disease-specific HRQoL in COPD patients over a period of 1 year. METHODS: We linked claims data with data from two survey waves to investigate the longitudinal effect of specialty care on HRQoL using linear Difference-in-Difference models based on 1:3 propensity score matched data. Generic HRQoL was operationalized by EQ-5D-5L visual analog scale (VAS), and disease-specific HRQoL by COPD assessment test (CAT). Subgroup analyses examined COPD patients with low (GOLD AB) and high (GOLD CD) exacerbation risk. RESULTS: In contrast to routine care patients, pulmonologists' patients (n = 442) experienced no significant deterioration in HRQoL (VAS - 0.0, p = 0.9870; CAT + 0.5, p = 0.0804). Models unveiled a small comparative advantage of specialty care on HRQoL (mean change: CAT - 0.8, VAS + 2.9), which was especially pronounced for GOLD AB (CAT - 0.7; VAS + 3.1). CONCLUSION: The uptake of pulmonologist care had a statistically significant, but not clinically relevant, beneficial impact on the development of HRQoL by slowing down overall HRQoL deterioration within 1 year. Including specialty care more appropriately in COPD management, especially at lower disease stages (GOLD AB), could thus improve patients' health outcome.

The effect of differential privacy on Medicaid participation among racial and ethnic minority groups.

OBJECTIVE: To investigate how county and state-level estimates of Medicaid enrollment among the total, non-Hispanic White, non-Hispanic Black or African American, and Hispanic or Latino/a population are affected by Differential Privacy (DP), where statistical noise is added to the public decennial US census data to protect individual privacy. DATA SOURCES: We obtained population counts from the final version of the US Census Bureau Differential Privacy Demonstration Products from 2010 and combined them with Medicaid enrollment data. STUDY DESIGN: We compared 2010 county and state-level population counts released under the traditional disclosure avoidance techniques and the ones produced with the proposed DP procedures. DATA COLLECTION/EXTRACTION METHODS: Not applicable. PRINCIPAL FINDINGS: We find the DP method introduces errors up to 10% into counts and proportions of Medicaid participation rate accuracy at the county level, especially for small subpopulations and racial and ethnic minority groups. The effect of DP on Medicaid participation rate accuracy is only small and negligible at the state level. CONCLUSIONS: The implementation of DP in the 2020 census can affect the analyses of health disparities and health care access and use among different subpopulations in the United States. The planned implementation of DP in other census-related surveys such as the American Community Survey can misrepresent Medicaid participation rates for small racial and ethnic minority groups. This can affect Medicaid funding decisions.

Augmented Inverse Probability Weighting and the Double Robustness Property.

This article discusses the augmented inverse propensity weighted (AIPW) estimator as an estimator for average treatment effects. The AIPW combines both the properties of the regression-based estimator and the inverse probability weighted (IPW) estimator and is therefore a "doubly robust" method in that it requires only either the propensity or outcome model to be correctly specified but not both. Even though this estimator has been known for years, it is rarely used in practice. After explaining the estimator and proving the double robustness property, I conduct a simulation study to compare the AIPW efficiency with IPW and regression under different scenarios of misspecification. In 2 real-world examples, I provide a step-by-step guide on implementing the AIPW estimator in practice. I show that it is an easily usable method that extends the IPW to reduce variability and improve estimation accuracy.[Box: see text].

Effectiveness of the German disease management programs: quasi-experimental analyses assessing the population-level health impact.

BACKGROUND: In 2002-2003 disease management programs (DMPs) for type 2 diabetes and coronary heart disease were introduced in Germany to improve the management of these conditions. Today around 6 million Germans aged 56 and older are enrolled in one of the DMPs; however, their effect on health remains unclear. METHODS: We estimated the impact of German DMPs on circulatory and all-cause mortality using a synthetic control study. Specifically, using routinely available data, we compared pre and post-intervention trends in mortality of individuals aged 56 and older for 1998-2014 in Germany to trends in other European countries. RESULTS: Average circulatory and all-cause mortality in Germany and the synthetic control was 1.63 and 3.24 deaths per 100 persons. Independent of model choice, circulatory and all-cause mortality decreased non-significantly less in Germany than in the synthetic control; for the model with a 3 year time lag, for example, by 0.12 (95%-CI: - 0.20; 0.44) and 0.22 (95%-CI: - 0.40; 0.66) deaths per 100 persons, respectively. Further main analyses, as well as sensitivity and subgroup analyses supported these results. CONCLUSIONS: We observed no effect on circulatory or all-cause mortality at the population-level. However, confidence intervals were wide, meaning we could not reject the possibility of a positive effect. Given the substantial costs for administration and operation of the programs, further comparative effectiveness research is needed to clarify the value of German DMPs for type 2 diabetes and CHD.

The causal impact of sugar taxes on soft drink sales: evidence from France and Hungary.

Sugar-sweetened beverages (SSBs) are associated with increased body weight and obesity, which induce a wide array of health impairments such as diabetes or cardiovascular disorders. Excise taxes have been introduced to counteract SSB consumption. We investigated the effect of sugar taxes on SSB sales in Hungary and France using a synthetic control approach. For France, we found a slight decrease in SSB sales after tax implementation while overall soft drink sales increased. For Hungary, there was only a short-term decrease in SSB sales which disappeared after 2 years, leading to an overall increase in SSB sales. However, both effects are characterized by great uncertainty.

The Effect of BMI and Type 2 Diabetes on Socioeconomic Status: A Two-Sample Multivariable Mendelian Randomization Study.

OBJECTIVE: To assess the independent causal effect of BMI and type 2 diabetes (T2D) on socioeconomic outcomes by applying two-sample Mendelian randomization (MR) analysis. RESEARCH DESIGN AND METHODS: We performed univariable and multivariable two-sample MR to jointly assess the effect of BMI and T2D on socioeconomic outcomes. We used overlapping genome-wide significant single nucleotide polymorphisms for BMI and T2D as instrumental variables. Their causal impact on household income and regional deprivation was assessed using summary-level data from the UK Biobank. RESULTS: In the univariable analysis, higher BMI was related to lower income (marginal effect of 1-SD increase in BMI [ß = -0.092; 95% CI -0.138; -0.047]) and higher deprivation (ß = 0.051; 95% CI 0.022; 0.079). In the multivariable MR, the effect of BMI controlling for diabetes was slightly lower for income and deprivation. Diabetes was not associated with these outcomes. CONCLUSIONS: High BMI, but not diabetes, shows a causal link with socioeconomic outcomes.

COVID-19 mitigation measures and nitrogen dioxide - A quasi-experimental study of air quality in Munich, Germany.

BACKGROUND: In response to the COVID-19 pandemic, the Bavarian State government announced several COVID-19 mitigation measures beginning on March 16, 2020, which likely led to a reduction in traffic and a subsequent improvement in air quality. In this study, we evaluated the short-term effect of COVID-19 mitigation measures on NO2 concentrations in Munich, Germany. METHODS: We applied two quasi-experimental approaches, a controlled interrupted time-series (c-ITS) approach and a synthetic control (SC) approach. Each approach compared changes occurring in 2020 to changes occurring in 2014-2019, and accounted for weather-related and other potential confounders. We hypothesized that the largest reductions in NO2 concentrations would be observed at traffic sites, with smaller reductions at urban background sites, and even small reductions, if any, at background sites. All hypotheses, as well as the main and additional analyses were defined a priori. We also conducted post-hoc analyses to ensure that observed effects were not due to factors other than the intervention. RESULTS: Main analyses largely supported our hypotheses. Specifically, at the two traffic sites, using the c-ITS approach we observed reductions of 9.34 µg/m3 (95% confidence interval: -23.58; 4.90) and 10.02 µg/m3 (-19.25; -0.79). Using the SC approach we observed reductions of 15.65 µg/m3 (-27.58; -4.09) and 15.1 µg/m3 (-24.82; -9.83) at these same sites. We observed effects ranging from smaller in magnitude to no effect at urban background and background sites. Additional analyses showed that the effect was largest in the first two weeks following introduction of measures, and that a 3-day lagged intervention time also showed a larger effect. Post-hoc analyses suggested that at least some of the observed effects may have been attributable to changes in air quality occurring before the intervention, as well as unusually high concentrations in January 2020. CONCLUSION: We applied two quasi-experimental approaches in assessing the impact of the COVID-19 mitigation measures on NO2 concentrations in Munich. Taking the 2020 pre-intervention average concentrations as a reference, we observed reductions in NO2 concentrations of approximately 15-25% and 24-36% at traffic sites, suggesting that reducing traffic may be an effective measure to reduce NO2 concentrations in heavily trafficked areas by margins which could translate to public health benefits.

Effects of adherence to pharmacological secondary prevention after acute myocardial infarction on health care costs - an analysis of real-world data.

BACKGROUND: Acute myocardial infarction (AMI), a major source of morbidity and mortality, is also associated with excess costs. Findings from previous studies were divergent regarding the effect on health care expenditure of adherence to guideline-recommended medication. However, gender-specific medication effectiveness, correlating the effectiveness of concomitant medication and variation in adherence over time, has not yet been considered. METHODS: We aim to measure the effect of adherence on health care expenditures stratified by gender from a third-party payer's perspective in a sample of statutory insured Disease Management Program participants over a follow-up period of 3-years. In 3627 AMI patients, the proportion of days covered (PDC) for four guideline-recommended medications was calculated. A generalized additive mixed model was used, taking into account inter-individual effects (mean PDC rate) and intra-individual effects (deviation from the mean PDC rate). RESULTS: Regarding inter-individual effects, for both sexes only anti-platelet agents had a significant negative influence indicating that higher mean PDC rates lead to higher costs. With respect to intra-individual effects, for females higher deviations from the mean PDC rate for angiotensin-converting enzyme (ACE) inhibitors, anti-platelet agents, and statins were associated with higher costs. Furthermore, for males, an increasing positive deviation from the PDC mean increases costs for ß-blockers and a negative deviation decreases costs. For anti-platelet agents, an increasing deviation from the PDC-mean slightly increases costs. CONCLUSION: Positive and negative deviation from the mean PDC rate, independent of how high the mean was, usually negatively affect health care expenditures. Therefore, continuity in intake of guideline-recommended medication is important to save costs.

The relationship between body mass index and health-related quality of life in COPD: real-world evidence based on claims and survey data.

BACKGROUND: Body mass index (BMI) is an important parameter associated with mortality and health-related quality of life (HRQoL) in chronic obstructive pulmonary disease (COPD). However, informed guidance on stratified weight recommendations for COPD is still lacking. This study aims to determine the association between BMI and HRQoL across different severity grades of COPD to support patient management. METHODS: We use conjunct analysis of claims and survey data based on a German COPD disease management program from 2016 to 2017. The EQ-5D-5L visual analog scale (VAS) and COPD Assessment Test (CAT) are used to measure generic and disease-specific HRQoL. Generalized additive models with smooth functions are implemented to evaluate the relationship between BMI and HRQoL, stratified by COPD severity. RESULTS: 11,577 patients were included in this study. Mean age was 69.4 years and 59% of patients were male. In GOLD grades 1-3, patients with BMI of around 25 had the best generic and disease-specific HRQoL, whereas in GOLD grade 4, obese patients had the best HRQoL using both instruments when controlled for several variables including smoking status, income, COPD severity, comorbidities, emphysema, corticosteroid use, and days spent in hospital. CONCLUSION: This real-world analysis shows the non-linear relationship between BMI and HRQoL in COPD. HRQoL of obese patients with mild to severe COPD might improve following weight reduction. For very severe COPD, a negative association of obesity and HRQoL could not be confirmed. The results hint at the need to stratify COPD patients by disease stage for optimal BMI management.

Effect of BMI on health care expenditures stratified by COPD GOLD severity grades: Results from the LQ-DMP study.

Chronic Obstructive Pulmonary Disease (COPD) is characterized by persistent respiratory symptoms and airflow limitation, which is progressive and not fully reversible. In patients with COPD, body mass index (BMI) is an important parameter associated with health outcomes, e.g. mortality and health-related quality of life. However, so far no study evaluated the association of BMI and health care expenditures across different COPD severity grades. We used claims data and documentation data of a Disease Management Program (DMP) from a statutory health insurance fund (AOK Bayern). Patients were excluded if they had less than 4 observations in the 8 years observational period. Generalized additive mixed models with smooth functions were used to evaluate the association between BMI and health care expenditures, stratified by severity of COPD, indicated by GOLD grades 1-4. We included 30,682 patients with overall 188,725 observations. In GOLD grades 1-3 we found an u-shaped relation of BMI and expenditures, where patients with a BMI of 30 or slightly above had the lowest and underweight and obese patients had the highest health care expenditures. Contrarily, in GOLD grade 4 we found an almost linear decline of health care expenditures with increasing BMI. In terms of expenditures, the often reported obesity paradox in patients with COPD was clearly reflected in GOLD grade 4, while in all other severity grades underweight as well as severely obese patients caused the highest health care expenditures. Reduction of obesity may thus reduce health care expenditures in GOLD grades 1-3.

Isolating cost drivers in interstitial lung disease treatment using nonparametric Bayesian methods.

Mixture modeling is a popular approach to accommodate overdispersion, skewness, and multimodality features that are very common for health care utilization data. However, mixture modeling tends to rely on subjective judgment regarding the appropriate number of mixture components or some hypothesis about how to cluster the data. In this work, we adopt a nonparametric, variational Bayesian approach to allow the model to select the number of components while estimating their parameters. Our model allows for a probabilistic classification of observations into clusters and simultaneous estimation of a Gaussian regression model within each cluster. When we apply this approach to data on patients with interstitial lung disease, we find distinct subgroups of patients with differences in means and variances of health care costs, health and treatment covariates, and relationships between covariates and costs. The subgroups identified are readily interpretable, suggesting that this nonparametric variational approach to inference can discover valid insights into the factors driving treatment costs. Moreover, the learning algorithm we employed is very fast and scalable, which should make the technique accessible for a broad range of applications.

Application of Mendelian Randomization to Investigate the Association of Body Mass Index with Health Care Costs.

Causal effect estimates for the association of obesity with health care costs can be biased by reversed causation and omitted variables. In this study, we use genetic variants as instrumental variables to overcome these limitations, a method that is often called Mendelian randomization (MR). We describe the assumptions, available methods, and potential pitfalls of using genetic information and how to address them. We estimate the effect of body mass index (BMI) on total health care costs using data from a German observational study and from published large-scale data. In a meta-analysis of several MR approaches, we find that models using genetic instruments identify additional annual costs of €280 for a 1-unit increase in BMI. This is more than 3 times higher than estimates from linear regression without instrumental variables (€75). We found little evidence of a nonlinear relationship between BMI and health care costs. Our results suggest that the use of genetic instruments can be a powerful tool for estimating causal effects in health economic evaluation that might be superior to other types of instruments where there is a strong association with a modifiable risk factor.

A greedy stacking algorithm for model ensembling and domain weighting.

OBJECTIVE: Because it is impossible to know which statistical learning algorithm performs best on a prediction task, it is common to use stacking methods to ensemble individual learners into a more powerful single learner. Stacking algorithms are usually based on linear models, which may run into problems, especially when predictions are highly correlated. In this study, we develop a greedy algorithm for model stacking that overcomes this issue while still being very fast and easy to interpret. We evaluate our greedy algorithm on 7 different data sets from various biomedical disciplines and compare it to linear stacking, genetic algorithm stacking and a brute force approach in different prediction settings. We further apply this algorithm on a task to optimize the weighting of the single domains (e.g., income, education) that build the German Index of Multiple Deprivation (GIMD) to be highly correlated with mortality. RESULTS: The greedy stacking algorithm provides good ensemble weights and outperforms the linear stacker in many tasks. Still, the brute force approach is slightly superior, but is computationally expensive. The greedy weighting algorithm has a variety of possible applications and is fast and efficient. A python implementation is provided.

The effect of bariatric surgery on health care costs: A synthetic control approach using Bayesian structural time series.

Surgical measures to combat obesity are very effective in terms of weight loss, recovery from diabetes, and improvement in cardiovascular risk factors. However, previous studies found both positive and negative results regarding the effect of bariatric surgery on health care utilization. Using claims data from the largest health insurance provider in Germany, we estimated the causal effect of bariatric surgery on health care costs in a time period ranging from 2 years before to 3 years after bariatric intervention. Owing to the absence of a control group, we employed a Bayesian structural forecasting model to construct a synthetic control. We observed a decrease in medication and physician expenditures after bariatric surgery, whereas hospital expenditures increased in the post-intervention period. Overall, we found a slight increase in total costs after bariatric surgery, but our estimates include a high degree of uncertainty.

Do alternative weighting approaches for an Index of Multiple Deprivation change the association with mortality? A sensitivity analysis from Germany.

OBJECTIVES: This study aimed to assess the impact of using different weighting procedures for the German Index of Multiple Deprivation (GIMD) investigating their link to mortality rates. DESIGN AND SETTING: In addition to the original (normative) weighting of the GIMD domains, four alternative weighting approaches were applied: equal weighting, linear regression, maximization algorithm and factor analysis. Correlation analyses to quantify the association between the differently weighted GIMD versions and mortality based on district-level official data from Germany in 2010 were applied (n=412 districts). OUTCOME MEASURES: Total mortality (all age groups) and premature mortality (<65 years). RESULTS: All correlations of the GIMD versions with both total and premature mortality were highly significant (p<0.001). The comparison of these associations using Williams's t-test for paired correlations showed significant differences, which proved to be small in respect to absolute values of Spearman's rho (total mortality: between 0.535 and 0.615; premature mortality: between 0.699 and 0.832). CONCLUSIONS: The association between area deprivation and mortality proved to be stable, regardless of different weighting of the GIMD domains. The theory-based weighting of the GIMD should be maintained, due to the stability of the GIMD scores and the relationship to mortality.

Identifying and interpreting subgroups in health care utilization data with count mixture regression models.

Inpatient care is a large share of total health care spending, making analysis of inpatient utilization patterns an important part of understanding what drives health care spending growth. Common features of inpatient utilization measures such as length of stay and spending include zero inflation, overdispersion, and skewness, all of which complicate statistical modeling. Moreover, latent subgroups of patients may have distinct patterns of utilization and relationships between that utilization and observed covariates. In this work, we apply and compare likelihood-based and parametric Bayesian mixtures of negative binomial and zero-inflated negative binomial regression models. In a simulation, we find that the Bayesian approach finds the true number of mixture components more accurately than using information criteria to select among likelihood-based finite mixture models. When we apply the models to data on hospital lengths of stay for patients with lung cancer, we find distinct subgroups of patients with different means and variances of hospital days, health and treatment covariates, and relationships between covariates and length of stay.

Performance of the UKPDS Outcomes Model 2 for Predicting Death and Cardiovascular Events in Patients with Type 2 Diabetes Mellitus from a German Population-Based Cohort.

BACKGROUND AND OBJECTIVE: Accurate prediction of relevant outcomes is important for targeting therapies and to support health economic evaluations of healthcare interventions in patients with diabetes. The United Kingdom Prospective Diabetes Study (UKPDS) risk equations are some of the most frequently used risk equations. This study aims to analyze the calibration and discrimination of the updated UKPDS risk equations as implemented in the UKPDS Outcomes Model 2 (UKPDS-OM2) for predicting cardiovascular (CV) events and death in patients with type 2 diabetes mellitus (T2DM) from population-based German samples. METHODS: Analyses are based on data of 456 individuals diagnosed with T2DM who participated in two population-based studies in southern Germany (KORA (Cooperative Health Research in the Region of Augsburg)-A: 1997/1998, n = 178; KORA-S4: 1999-2001, n = 278). We compared the participants' 10-year observed incidence of mortality, CV mortality, myocardial infarction (MI), and stroke with the predicted event rate of the UKPDS-OM2. The model's calibration was evaluated by Greenwood-Nam-D'Agostino tests and discrimination was evaluated by C-statistics. RESULTS: Of the 456 participants with T2DM (mean age 65 years, mean diabetes duration 8 years, 56% male), over the 10-year follow-up time 129 died (61 due to CV events), 64 experienced an MI, and 46 a stroke. The UKPDS-OM2 significantly over-predicted mortality and CV mortality by 25% and 28%, respectively (Greenwood-Nam-D'Agostino tests: p < 0.01), but there was no significant difference between predicted and observed MI and stroke risk. The model poorly discriminated for death (C-statistic [95% confidence interval] = 0.64 [0.60-0.69]), CV death (0.64 [0.58-0.71]), and MI (0.58 [0.52-0.66]), and failed to discriminate for stroke (0.57 [0.47-0.66]). CONCLUSIONS: The study results demonstrate acceptable calibration and poor discrimination of the UKPDS-OM2 for predicting death and CV events in this population-based German sample. Those limitations should be considered when using the UKPDS-OM2 for economic evaluations of healthcare strategies or using the risk equations for clinical decision-making.