Using the Causal Inference Framework to Support Individualized Drug Treatment Decisions Based on Observational Healthcare Data

Abstract

When healthcare professionals have the choice between several drug treatments for their patients, they often experience considerable decision uncertainty because many decisions simply have no single “best” choice. The challenges are manifold and include that guideline recommendations focus on randomized controlled trials whose populations do not necessarily correspond to specific patients in everyday treatment. Further reasons may be insufficient evidence on outcomes, lack of direct comparison of distinct options, and the need to individually balance benefits and risks. All these situations will occur in routine care, its outcomes will be mirrored in routine data, and could thus be used to guide decisions. We propose a concept to facilitate decision-making by exploiting this wealth of information. Our working example for illustration assumes that the response to a particular (drug) treatment can substantially differ between individual patients depending on their characteristics (heterogeneous treatment effects, HTE), and that decisions will be more precise if they are based on real-world evidence of HTE considering this information. However, such methods must account for confounding by indication and effect measure modification, eg, by adequately using machine learning methods or parametric regressions to estimate individual responses to pharmacological treatments. The better a model assesses the underlying HTE, the more accurate are predicted probabilities of treatment response. After probabilities for treatment-related benefit and harm have been calculated, decision rules can be applied and patient preferences can be considered to provide individual recommendations. Emulated trials in observational data are a straightforward technique to predict the effects of such decision rules when applied in routine care. Prediction-based decision rules from routine data have the potential to efficiently supplement clinical guidelines and support healthcare professionals in creating personalized treatment plans using decision support tools.

Keywords:

• claims data
• decision-making
• heterogeneous treatment effects
• effect modification
• confounding by indication
• prediction-based decision rules

Author Contributions

ADM was responsible for data generation, data analysis, and visualization of results. ADM and WEH drafted the manuscript, all authors critically revised the manuscript. All authors made substantial contributions to conception and design, acquisition of data, or analysis and interpretation of data; took part in drafting the article or revising it critically for important intellectual content; agreed to submit to the current journal; gave final approval of the version to be published; and agree to be accountable for all aspects of the work.

Disclosure

Andreas D. Meid is funded by the Physician-Scientist Programme of the Medical Faculty of Heidelberg University. The funding body did not play any role in the design of the study and collection, analysis, and interpretation of data and in writing the manuscript.

Hanna M Seidling reports grants from Chambers of pharmacists Baden-Württemberg, Chamber of pharmacists Nordrhein, Chamber of pharmacists Hessen, Chamber of pharmacists Niedersachsen, Klaus-Tschira-Stiftung, Förderinitiative Pharmazeutische Betreuung e.V., Dosing GmbH, ABDA – Bundesvereinigung Deutscher Apotheker e.V., g-BA, BMBF, European Commission Horizon 2020, AOKPlus, Dr. August und Dr. Anni Lesmüller Stiftung, personal fees and non-financial support from ADKA e.V.; EAHP; Chamber of Pharmacists, Hessen; Chamber of pharmacists, Baden-Württemberg, Chamber of pharmacists, Westfalen-Lippe, Chamber of pharmacists Nordrhein, Chamber of pharmacists Bavaria, DPhG, ABDA, omnicell, Chamber of pharmacists Niedersachsen, and Chamber of pharmacists Thüringen, Govi Verlag, Deutscher Apotheker Verlag, Wissenschaftliche Verlagsgesellschaft Stuttgart, and Bundesgesundheitsblatt, and non-financial support from VKliPha; AkdÄ; GSASA, APS e.V., NHS, ESCP, BAK, ÄZQ, SFPC, Dosing GmbH, Karolinska Institutet, University of Bonn, and University Hospital Hamburg, outside the submitted work.

Walter E Haefeli reports grants from Landesapothekerkammer BW, Dosing GmbH, Heidelberg, Gemeinsamer Bundesaussschuss (HIOPP, Proper med, PIM Stop), and Smooth ClinicalTrials Ltd., research funding from Sumaya Biotec GmbH & Co. KG, Klaus Tschira Stiftung, Vaxxim GmbH, AOK Baden-Württemberg, Basilea Pharmaceutica Ltd., and Elsevier GmbH, EU Horizon 2020 (PRESTIGE-AF), personal fees from CHIESI GmbH, PIQUR Basel, and Regierungspräsidium Stuttgart, personal fees and non-financial support from BayerPharma AG, personal fees, non-financial support, speaker fee, and traveling expenses from Aqua Institute Göttingen, speaker fee and traveling expenses from Aspen Europe GmbH, Diaplan, MSD Sharp & Dohme GmbH, AstraZeneca GmbH, Grünenthal GmbH, KWHC GmbH, Novartis, Berlin-Chemie AG, and Doctrina Med GmbH & Co.KG, grants and non-financial support from Actelion GmbH, speaker fee and research funding from GSK GER/UK/Slovakia/France/, grants, research funding, speaker fee, and traveling expenses from Daiichi Sankyo GmbH, research funding, speaker fee, and traveling expenses from Bristol-Myers Squibb GmbH, speaker fee, traveling expenses, and consultancy services from Boehringer GmbH, non-financial support, speaker fee, and traveling expenses from Pfizer GmbH, traveling expenses from University Frankfurt, research funding and consultancy services from Heidelberg ImmunoTherapeutics GmbH, personal fees, consultancy services, and traveling expenses from Fresenius Medical Care, consultancy services from Ligatur Verlag and Stiftung Warentest, editorship/authorship royalties from Thieme Verlag and Wissenschaftliche Verlagsgesellschaft Stuttgart, grant numbers TI02.001, TTU05.901,O1ET10048, 01GK0702 from the German Federal Ministry of Education and Research (BMBF) (projects entitled DZIF, ESTHER, Agewell, LipORa, Prima LiveR) J

The authors report no other potential conflicts of interest for this work.