Pharmacoepidemiology analyses the use and the effect of drugs in populations. It addresses pharmacological research questions by applying epidemiological methods. The main focus of our research is the population-based evaluation of drug effects and adverse drug reactions in different patient groups. Therefore, we analyze factors that potentially influence the occurrence of adverse drug reactions or drug interactions. As basis for pharmacoepidemiological analyses, we use primary data (e.g. population-based cohort studies) as well as secondary data (e.g. health insurance data). The aim is to assess efficiency and safety of pharmacotherapy. Thus, pharmacoepidemiology contributes to drug safety.

Head

Head of research group: Prof. Dr. Britta Hänisch
Phone: +49 (0) 228 99 307 5721
E-Mail: britta.haenisch@bfarm.de

Curriculum Vitae

Staff

Dr. Cornelia Becker

Kerstin Pfeifer

Dr. Christoph Röthlein

Martin Russek

Julia Wicherski

Further members of the research group see: DZNE Group Pharmacoepidemiology

Current Research Projects

The EU Project „Real4Reg“ – Development, optimisation and implementation of artificial intelligence methods for real-world data analyses in regulatory decision-making and health
technology assessment along the product lifecycle

Background:
The use of real-world data (RWD) is becoming increasingly relevant in drug regulatory decision making, but heterogeneity in data structure and sources pose a challenge.
Moreover, there is a need for optimised tools for the appropriate application of artificial intelligence (AI) to RWD sources in the regulatory and health technology assessment (HTA) context.

Objectives:
Based on highly relevant use cases from regulatory practice and across the product lifecycle Real4Reg develops AI-based data-driven methods and tools for the assessment of medicinal products. Our results will inform training activities on good practice examples and will be informative for existing and emerging guidelines for both health regulatory authorities and health technology assessment (HTA) bodies across Europe. The use cases comprise information on the use of RWD to describe study populations, the design of historical controls and synthetic data for pre-authorisation/evaluation purposes, post-marketing safety evaluation and drug repurposing.

Methods:
The data used for the project include national register and health insurance claims data from Denmark, Finland, Germany, and Portugal. We develop user-friendly analytical solutions along four use cases, including pre- and post-authorization examples. Methodological tools involved include a common data model, a trial emulation, propensity score algorithms and a broad spectrum of AI algorithms. The project consortium consists of ten partners from six different European countries among them two patient organisations. Moreover the Real4Reg project is consulted by a multi-disciplinary advisory board consisting of experts from regulatory agencies, patient organisations, HTA, clinical experts, representatives for industry and representatives for payers.

Pharmacoepidemiological analyses on medication- and morbidity-associated risk factors in vulnerable patient groups on the progression of COVID-19

Background:
The SARS-CoV-2 pandemic is an ongoing challenge for health care systems worldwide. Particularly, people with pre-existing chronic diseases are vulnerable patient groups. Some of the drugs used for these chronic diseases such as antihypertensive drugs, immunosuppressants and antiretroviral drugs have been discussed as possible influencing factors on the progression of COVID-19.

Objectives:
The study examines the effect of medication- and morbidity-associated risk factors suspected to modify the disease course and progression of COVID-19. We analyse the risk for severe disease progression with hospitalization and factors influencing this risk. Predisposing factors regarding the occurrence of COVID-19, severity, and progression are identified. The resulting findings can be used to inform treatment decisions for vulnerable patient groups.

Methods:
The study is based on current routine data of the Techniker Krankenkasse. The data set includes insurants of the Techniker Krankenkasse with COVID-19 diagnosis from both the outpatient and inpatient sectors. The control group includes Techniker Krankenkasse insurants without COVID-19 diagnosis. In addition to multiple regression models, machine learning methods are used to include a variety of covariates and the occurrence of nonlinear complex relationships. In addition, we will conduct exploratory analyses to detect additional determinants of COVID-19 disease progression.

Serious Adverse Drug Reactions of Fluoroquinolones: pharmacoepidemiological analyses

Background:
Based on the results of the fluoroquinolone (FQ) risk assessment report from the European Medicines Agency (EMA), - which reports rare serious adverse drug reactions (ADRs), some of which are life-threatening or irreversible - restrictions and changes have been imposed on the use of FQs.

Objectives:
The study investigates whether and to what extent the use of an FQ is associated with a modified risk for the occurrence of various severe ADRs, such as cardiac arrhythmias, aortic aneurysms, acute toxic liver disease, collagen-associated diseases, or certain neurological or neuropsychiatric disorders. Thus, the results will contribute to real-world evidence for the adequate use of FQs and the occurrence of severe ADRs in routine care.

Methods:
The longitudinal cohort study is based on AOK routine data and covers the AOK insurance years 2013 to 2019. The cohort consists of all AOK insured persons who are ≥ 18 years of age at the time of the first antibiotic prescription. The individual observation period is one year. Multiple regression models with adjustment for potential influencing factors and propensity score approaches are used for the analysis.

EMPAR - Influence of metabolic profiles on drug safety in routine care

Background:
Different responses to drugs due to individual genetically determined metabolic profiles could lead to the occurrence of adverse drug reactions (ADRs) and therapy resistance, which result in inefficient use of health care structures and delay the success of the patient's therapy.

Objectives:
Using metabolic profiles, risk factors significant for drug therapy are analyzed for healthcare-relevant endpoints on the basis of routine data from statutory health insurance. In particular, the study investigates the extent to which individual genetic differences have an influence on the utilization of health care services, such as hospital admissions or drug prescriptions. The long-term goal is to implement the use of pre-emptive tests of metabolic profiles in routine health care. The collection of patient-relevant metabolic risk profiles for ADRs or therapy resistance serves to improve health care, both in terms of quality and cost-effectiveness.

Methods:
The study population comprises approximately 10,000 insured persons of the Techniker Krankenkasse who have been prescribed at least one drug for which there is clinical evidence that genetic polymorphisms can lead to a high variability in drug exposure at normal dosage. For the study, we include prescriptions of anticoagulants and statins as well as patients with an ADR diagnosis (ICD-10: Y57.9!). Pharmacogenetic markers will be determined from buccal swabs. These data will be analyzed pharmacogenetically, pharmacoepidemiologically and pharmacoeconomically together with cross-sectoral routine data of the Techniker Krankenkasse.

N-nitrosodimethylamine-contaminated valsartan and the risk of cancer

Background:
N-nitrosodimethylamine (NDMA) is classified as a probable human carcinogen and was detected as an impurity in the antihypertensive drug valsartan in 2018. The impurity is suspected to have resulted from a change in the manufacturing process of a major producer of the active ingredient valsartan in 2012. To date, no large cohort studies exist that evaluate the potential carcinogenic effects associated with the use of NDMA-contaminated valsartan.

Objectives:
The study analyzes the association between NDMA-contaminated valsartan and cancer risk. The results of the study are relevant to regulatory agencies worldwide by providing information to assess the impact of NDMA contamination in valsartan drug products. As an example, the study shows how "real world" data from health insurance companies, using pharmacoepidemiological methods, can help answer drug safety questions.

Methods:
The data set of the cohort study consists of health insurance data from the AOK. It includes all patients who were ≥ 40 years of age at baseline 2012 and had filled at least one prescription for valsartan during 2012-2017. Incident cancer diagnosis is considered the endpoint. Using Cox regression models with time-dependent variables as well as adjustment for potential influencing factors, hazard ratios for cancer in general and various individual cancer types are determined.

Selected Publications

• Gomm W, Röthlein C, Schüssel K, Brückner G, Schröder H, Heß S, Frötschl R, Broich K, Haenisch B (2021) N-nitrosodimethylamine-contaminated valsartan and the risk of cancer-a longitudinal cohort study based on German health insurance data. Dtsch Arztebl Int. doi: 10.3238/arztebl.m2021.0129
• Huebner T, Steffens M, Linder R, Fracowiak J, Langner D, Garling M, Falkenberg F, Roethlein C, Gomm W, Haenisch B, Stingl J (2020). Influence of metabolic profiles on the safety of drug therapy in routine care in Germany: protocol of the cohort study EMPAR. BMJ Open. 10: e032624.
• Benda N, Haenisch B (2020) Enrichment design using placebo non-responders. Pharm Stat. 19: 303-314.
• Nerius M, Haenisch B, Gomm W, Doblhammer G, Schneider A (2020) Glucocorticoid Therapy Is Associated With a Lower Risk of Dementia. J Alzheimers Dis. 73: 175-183.
• Heser K, Pohontsch NJ, Scherer M, Löffler A, Luck T, Riedel-Heller SG, Maier W, Parker D, Haenisch B, Jessen F (2018) Perspective of elderly patients on chronic use of potentially inappropriate medication - Results of the qualitative CIM-TRIAD study. PLoS One. 13: e0202068.
• Engel B, Gomm W, Broich K, Maier W, Weckbecker K, Haenisch B (2018) Hyperuricemia and dementia - a case-control study. BMC Neurol. 18: 131.
• Taipale H, Gomm W, Broich K, Maier W, Tolppanen AM, Tanskanen A, Tiihonen J, Hartikainen S, Haenisch B (2018) Use of Antiepileptic Drugs and Dementia Risk-an Analysis of Finnish Health Register and German Health Insurance Data. J Am Geriatr Soc. 66: 1123-1129.
• Pohontsch NJ, Löffler A, Luck T, Heser K, Parker D, Haenisch B, Riedel-Heller SG, Jessen F, Scherer M (2018) Informal caregivers' perspectives on health of and (potentially inappropriate) medication for (relatively) independent oldest-old people - a qualitative interview study. BMC Geriatr. 18: 169.
• Teipel SJ, Fritze T, Ellenrieder M, Haenisch B, Mittelmeier W, Doblhammer G (2018) Association of joint replacement surgery with incident dementia diagnosis in German claims data. Int Psychogeriatr. 30: 1375-1383.
• Nerius M, Johnell K, Garcia-Ptacek S, Eriksdotter M, Haenisch B, Doblhammer G (2018) The Impact of Antipsychotic Drugs on Long-term Care, Nursing Home Admission, and Death in Dementia Patients. J Gerontol A Biol Sci Med Sci. 73: 1396-1402.
• Gomm W, von Holt K, Thomé F, Broich K, Maier W, Weckbecker K, Fink A, Doblhammer G, Haenisch B (2016) Regular Benzodiazepine and Z-Substance Use and Risk of Dementia: An Analysis of German Claims Data. J Alzheimers Dis. 54: 801-808.
• Gomm W, von Holt K, Thomé F, Broich K, Maier W, Fink A, Doblhammer G, Haenisch B (2016) Association of Proton Pump Inhibitors With Risk of Dementia: A Pharmacoepidemiological Claims Data Analysis. JAMA Neurol 73: 410-416.