Drug utilization studies (DUS) using real-world data (RWD) are essential for understanding how medications are prescribed, dispensed, and used in everyday clinical practice. Unlike clinical trials, which are conducted under controlled conditions, RWD is derived from routine healthcare settings, including electronic health records (EHRs), insurance claims, patient registries, and pharmacy data.
These studies provide valuable insights into prescribing patterns, patient adherence, treatment effectiveness, and potential adverse effects in diverse patient populations. By leveraging RWD, researchers can assess the real-world impact of medications, identify gaps in care, and inform healthcare policy and decision-making. This approach enhances the understanding of medication use in the context of actual clinical practice, contributing to the optimization of pharmacotherapy and improved patient outcomes and safety.
With DUS, it is possible to perform basic assessments of drug utilization such as incidence and prevalence of drug use and more advanced evaluations such as treatment adherence, persistence, drug combinations, drug switching, concurrent drug use, poly-pharmacy, and potential drug misuse. However, when designing DUS using RWD researcher face several challenges due to the complexity and variability inherent to the data.
Let’s have a look to some of the key aspects to address when designing a DUS:
1. Prescriptions, dispensations, and adherence: Data available from the routine healthcare setting might provide accessed to either prescription or dispensation data. However, just because a drug is prescribed does not mean it is dispensed and even if a drug is dispensed, there is no guarantee that the patient will take the medication and adhere to the prescribed regimen. Measuring adherence may require complementary measures.
2. Persistence: Treatment persistence refers to the time from initiation of treatment until discontinuation. Treatment persistence should be based on information on the prescribed daily dose and amount of dispensed drug. However, it is often necessary to rely on assumption for this assessment, as the information might not be available. When working with assumption some of the methodologies available are: the estimation on the number of dispensed tablets or the amount of dispensed Define Daily Doses (DDDs) (1, 2); the refill gap method, that allows assessing persistence based on gaps between prescription refills; the anniversary model: a patient is considered persistent for one year if they refill their prescription within a specified interval surrounding the anniversary date of their first prescription (1, 3, 4); the proportion of patients covered (PPC) method estimates the proportion of alive patients that are covered by treatment at a given day after treatment initiation (1, 3, 4).
3. Timing of exposure: Drug exposure definitions depend on the study design and include factors like start date, frequency, duration, timing of administration and changes in medication. However, precise information about when a patient starts and stops taking a medication is often missing or unclear in RWE research. From a pharmacological prospective, it should also be taken into account that the effect of the drug may continue after treatment discontinuation.
4. Dosage Information: Dosage details may not be consistently recorded, and the availability of differences in dosages can complicate the assessment of drug exposure. The definition of drug exposure should include details of the dose for each administration (e.g., daily dose), and an estimate of the cumulative dose. It is also key to consider dose changes over time.
5. Route of administration: Many active ingredients can have different indications and formulations and have different routes of administration. The definition of drug exposure should take into account how different dosage forms (e.g., dermal, parenteral, oral) will be adding into the dose calculation if multiple forms are available.
6. Changes in Prescribing Practices: Over time, changes in clinical guidelines, drug availability, and physician prescribing habits can affect drug exposure patterns.
7. Patient Behavior: Patients may change their behavior, including their medication adherence, based on various factors like side effects, costs, or perceived effectiveness.
8. Linking Multiple Data Sources: RWE studies may use data from various sources, including EHRs, insurance claims, registries, and patient-reported outcomes. Each source has its own format, level of detail, and reliability. Accessing data from different data source and linking the data (e.g., linking EHRs with pharmacy claims) can be technically challenging and may introduce errors.
9. Data heterogeneity: Different data sources may capture drug exposure differently or with delayed timelines, making it challenging to harmonize data for analysis. RWD sources like EHRs and claims databases often suffer from inconsistent data entry practices, leading to missing or incomplete data.
10. Data Accuracy: Errors or delays in data entry, varying levels of detail, and lack of standardized coding can affect the accuracy of recorded drug exposure.
11. Confounders: Patients often take multiple medications, which can interact and affect the outcomes of interest (concurrent medications). The presence of other health conditions can confound the relationship between drug exposure and outcomes (comorbidities).
In conclusion, assessing drug exposure in RWE research requires careful consideration of data quality, product characteristics, source variability, patient behavior, and regulatory constraints. Addressing these challenges often involves using advanced methodologies, such as data harmonization techniques, sophisticated statistical methods to adjust for confounding, and robust data linkage practices to improve the reliability and validity of RWE studies.
REFERENCES
- Vrijens B, De Geest S, Hughes DA, et al. A new taxonomy for describing and defining adherence to medications. Br J Clin Pharmacol. 2012; 73(5):691-705. doi:10.1111/j.1365-2125.2012.04167.x
- World Health Organisation Collaborating Centre for Drug Statistics Methodology. ATC classification index with DDDs. Accessed from http://www.whocc.no/atc_ddd_publications/atc_ddd_index/
- Caetano PA, Lain PJMC, Morgan SG. Toward a standard definition and measurement of persistence with drug therapy: examples from research on statin and antihypertensive utilization. Clin Ther. 2006;28(9):1411.
- Grégoire JP, Moisan J. Assessment of adherence to drug treatment indatabase research. Drug Utilization Research—Methods and Applications. Wiley-Blackwell; 2016.