Schätzung kleinräumiger Krankheitshäufigkeiten für die deutsche Bevölkerung anhand von Routinedaten am Beispiel von Typ-2-Diabetes

[Utilizing Routine Health Insurance Data for Calculation of Disease Frequencies in the Project BURDEN 2020]

OBJECTIVE

The concept of disease burden enables a comprehensive analysis of the health status of a population. Key indicators are years of life lost due to mortality (YLL) and morbidity (years lived with disability, YLD), summarised in the DALY indicator (disability adjusted life years). These indicators are suitable for planning prevention, health care or provision of health services. With the project BURDEN 2020, funded by the German Federal Joint Committee's Innovation Fund, a national and regionalised calculation of burden of disease for Germany is being carried out for the first time, based on the methodology of the international "Global Burden of Disease" study.

METHODS

Calculation of YLD requires data on the frequency and severity of diseases, with routine health insurance data constituting an important data source. Case definitions for 18 selected diseases and severity levels for 11 of these diseases were developed in expert meetings. Based on these case definitions, the AOK Research Institute (WIdO) calculated disease frequencies from health utilisation data of patients insured with the AOK. A specific concept for prevalence calculation takes into account the dynamics of an open cohort of insurees. For severity levels, the results of the AOK insurees were extrapolated to the total population in Germany according to age and gender groups. For disease frequencies, the results were additionally adjusted for morbidity and estimated on regional levels.

RESULTS

Disease frequencies measured by prevalences or rates are available for 18 diseases from seven categories (cardiovascular diseases, diabetes, cancer, mental disorders, dementia, COPD and lower respiratory tract infections) at the regional levels of the 16 federal states and 96 regional planning areas. Severity distributions are provided on the national level stratified by age groups and gender. The results and documentation of methods are available at www.krankheitslage-deutschland.de (in German language).

CONCLUSION

Routine health insurance data are an important data source in the BURDEN 2020 project because regional figures and, in some cases, severity levels can be determined on the basis of a large number of cases. A comprehensive publication of results creates transparency and allows reutilisation of methods in further projects. Future research should extend burden of disease calculations to other diseases. In addition, there is an increasing demand for health data linkage.

[Algorithm for the Classification of Type 1 and Type 2 Diabetes Mellitus for the Analysis of Routine Data]

BACKGROUND

Diabetes mellitus is a disease of high public health relevance. To estimate the temporal development of prevalence, routine data of statutory health insurances (SHI) are being increasingly used. However, these data are primarily collected for billing purposes and the case definition of specific diseases remains challenging. In this study, we present an algorithm for differentiation of diabetes types analyzing SHI routine data.

METHODS

The basis for the analysis was an age and sex-stratified random sample of persons of the Barmer SHI with a continuous insurance duration from 2010 to 2018 in the magnitude of 1% of the German population. Diabetes was defined in the reporting year 2018, as documentation of (1) a "confirmed" ICD diagnosis E10.- to E14.- in at least two quarters, (2) a "confirmed" ICD diagnosis E10.- to E14.- in one quarter with an additional prescription of an antidiabetic drug (ATC codes A10), or (3) an ICD diagnosis E10.- to E14.- in the inpatient sector, outpatient surgery, or work disability. Individuals were assigned to a diabetes type based on the specific ICD diagnosis E10.- to E14.- and prescribed medications, differentiated by insulin and other antidiabetics. Still unclear or conflicting constellations were assigned on the basis of the persons' age or the frequency and observation of the diagnosis documentation over more than one year. The participation in a disease management program was considered in a sensitivity analysis.

RESULTS

The prevalence of documented diabetes in the Barmer sample was 8.8% in 2018. Applying the algorithm, 98.5% of individuals with diabetes could be classified as having type 1 diabetes (5.5%), type 2 diabetes (92.6%), or another specific form of diabetes (0.43%). Thus, the prevalence was 0.48% for type 1 diabetes and 8.1% for type 2 diabetes in 2018.

CONCLUSION

The vast majority of people with diabetes can be classified by their diabetes type on the basis of just a few characteristics, such as diagnoses, drug prescription, and age. Further studies should assess the external validity by comparing the results with primary data. The algorithm enables the analysis of important epidemiological indicators and the frequency of comorbidities based on routine data differentiated by type 1 and type 2 diabetes, which should be considered in the surveillance of diabetes in the future.

Busch M, Haller S, Hapke U, Neuhauser H, Reitzle L, Scheidt-Nave C, Schlotmann A, Steppuhn H, Thom J, Ziese T, Rommel A. The Burden of Disease in Germany at the National and Regional Level

BACKGROUND

.Summary measures such as disability-adjusted life years (DALY) are becoming increasingly important for the standardized assessment of the burden of disease due to death and disability. The BURDEN 2020 pilot project was designed as an independent burden-of-disease study for Germany, which was based on nationwide data, but which also yielded regional estimates.

METHODS

DALY is defined as the sum of years of life lost due to death (YLL) and years lived with disability (YLD). YLL is the difference between the age at death due to disease and the remaining life expectancy at this age, while YLD quantifies the number of years individuals have spent with health impairments. Data are derived mainly from causes of death statistics, population health surveys, and claims data from health insurers.

RESULTS

In 2017, there were approximately 12 million DALY in Germany, or 14 584 DALY per 100 000 inhabitants. Conditions which caused the greatest number of DALY were coronary heart disease (2321 DALY), low back pain (1735 DALY), and lung cancer (1197 DALY). Headache and dementia accounted for a greater disease burden in women than in men, while lung cancer and alcohol use disorders accounted for a greater disease burden in men than in women. Pain disorders and alcohol use disorders were the leading causes of DALY among young adults of both sexes. The disease burden rose with age for some diseases, including cardiovascular diseases, dementia, and diabetes mellitus. For some diseases and conditions, the disease burden varied by geographical region.

CONCLUSION

The results indicate a need for age- and sex-specific prevention and for differing interventions according to geographic region. Burden of disease studies yield comprehensive population health surveillance data and are a useful aid to decision-making in health policy.

$$\ell _2$$-penalized approximate likelihood inference in logit mixed models for regional prevalence estimation under covariate rank-deficiency

Regional prevalence estimation requires the use of suitable statistical methods on epidemiologic data with substantial local detail. Small area estimation with medical treatment records as covariates marks a promising combination for this purpose. However, medical routine data often has strong internal correlation due to diagnosis-related grouping in the records. Depending on the strength of the correlation, the space spanned by the covariates can become rank-deficient. In this case, prevalence estimates suffer from unacceptable uncertainty as the individual contributions of the covariates to the model cannot be identified properly. We propose an area-level logit mixed model for regional prevalence estimation with a new fitting algorithm to solve this problem. We extend the Laplace approximation to the log-likelihood by an $$\ell _2$$ ℓ 2 -penalty in order to stabilize the estimation process in the presence of covariate rank-deficiency. Empirical best predictors under the model and a parametric bootstrap for mean squared error estimation are presented. A Monte Carlo simulation study is conducted to evaluate the properties of our methodology in a controlled environment. We further provide an empirical application where the district-level prevalence of multiple sclerosis in Germany is estimated using health insurance records.

Prevalence and Therapy Rates for Stuttering, Cluttering, and Developmental Disorders of Speech and Language: Evaluation of German Health Insurance Data

Purpose

To evaluate the prevalence and treatment patterns of speech and language disorders in Germany.

Methods

A retrospective analysis of data collected from 32% of the German population, insured by the statutory German health insurance (AOK, Local Health Care Funds). We used The International Statistical Classification of Diseases and Related Health Problems, 10th revision, German Modification (ICD-10 GM) codes for stuttering (F98.5), cluttering (F98.6), and developmental disorders of speech and language (F80) to identify prevalent and newly diagnosed cases each year. Prescription and speech therapy reimbursement data were used to evaluate treatment patterns.

Results

In 2017, 27,977 patients of all ages were diagnosed with stuttering (21,045 males, 75% and 6,932 females, 25%). Stuttering prevalence peaks at age 5 years (boys, 0.89% and girls, 0.40%). Cluttering was diagnosed in 1,800 patients of all ages (1,287 males, 71.5% and 513 females, 28.5%). Developmental disorders of speech and language were identified in 555,774 AOK-insurants (61.2% males and 38.8% females). Treatment data indicate a substantial proportion newly diagnosed stuttering individuals receive treatment (up to 45% of 6-year-old patients), with slightly fewer than 20 sessions per year, on average. We confirmed a previous study showing increased rates of atopic disorders and neurological and psychiatric comorbidities in individuals with stuttering, cluttering, and developmental disorders of speech and language.

Conclusion

This is the first nationwide study using health insurance data to analyze the prevalence and newly diagnosed cases of a speech and language disorder. Prevalence and gender ratio data were consistent with the international literature. The crude prevalence of developmental disorders of speech and language increased from 2015 to 2018, whereas the crude prevalence for stuttering remained stable. For cluttering, the numbers were too low to draw reliable conclusions. Proportional treatment allocation for stuttering peaked at 6 years of age, which is the school entrance year, and is later than the prevalence peak of stuttering.

[Prevalence and incidence of documented diabetes based on health claims data-reference analysis for diabetes surveillance in Germany]

BACKGROUND

The prevalence and incidence of documented diabetes are two essential indicators intended to be reported on a periodic basis within the framework of diabetes surveillance in Germany.

METHODOLOGY

Data provided based on the Data Transparency Act were analyzed. The data contain information on outpatient and inpatient care for all approximately 70 million persons with statutory health insurance. The case definition for the prevalence of documented diabetes comprises a confirmed outpatient diagnosis in at least two quarters of a year or an inpatient diagnosis in at least one quarter of a year in accordance with ICD-10 codes E10.- to E14.-. The incidence was calculated based on the same definition and with one year of diagnosis-free lead time.

RESULTS

In 2011, a prevalence of 9.7% (women: 9.4%, men: 10.1%) was observed for persons with statutory health insurance. There are considerable differences in prevalence between the federal states and the maximum gap is 7.1 percentage points (age standardized: 4.0 percentage points). Type 2 and type 1 diabetes show a documented prevalence of 7.5% and 0.28%, respectively. Unspecified diabetes is documented relatively frequently with 1.9%. In 0.21% of persons, the diagnosis diabetes is documented via one inpatient secondary diagnosis. In addition, 0.17% of people without documented diabetes have at least one prescription of an antidiabetic drug. In 2012, 565,040 insured persons were newly diagnosed with diabetes; this corresponds to 1.0% of the insured persons (women: 1.0%, men: 1.1%).

DISCUSSION

The developed reference analysis is suitable for reporting the prevalence and incidence of documented diabetes within the framework of diabetes surveillance. The differentiation of diabetes types is difficult due to coding practice.