Identification of Drug Side Effects Based on Prescription Data
Name
Kairi Käiro
Abstract
This master thesis analyzed whether it is possible to identify potential drug side effects based on digital prescription data. The dataset used included 10% of the randomly selected Estonian residents, more precisely, the data of 139 987 patients who purchased 7 207 212 prescriptions between 2012 and 2019. When forming pairs of active ingredients that cause possible side effects and active ingredients that potentially alleviate them, a principle was used that the timeframe for purchasing the drugs should be within 90 days. The results showed that metoprolol, frequently prescribed for reducing blood pressure and heart rate mostly in the age group 51+, could potentially be associated with side effects belonging to different groups (central nervous system, lung diseases, psychiatric conditions, digestive tract issues, and immune-modulated effects). In the 31+ age group, a range of antidepressants and sedatives could cause potential side effects, including sleep disturbances or withdrawal symptoms upon suddenly stopping the medication, prompting the prescription of sleeping pills. The method used in this study made it possible to identify numerous pairs within the database, showing a probable connection. However, it is substantial that combinations devoid of any adverse drug reactions could also be found among the registered cases. Pairs of active ingredients that may have induced and alleviated possible side effects cannot be correctly identified in all cases solely based on package leaflets of pharmaceutical products. The results demonstrated the need to validate the results, which requires specialists with medical expertise. The method used in this study needs to be improved to consider more factors like the frequency of prescriptions, medical diagnosis codes and epicrisis to achieve more accurate results.
Graduation Thesis language
Estonian
Graduation Thesis type
Master - Conversion Master in IT
Supervisor(s)
Marek Oja, Sirli Tamm
Defence year
2024