NOTE: This article is the sixth in a series of 10 articles and is part of our Economic Evaluation in Healthcare 101 course. You can find a course overview and links to all 10 course modules here:

Data Sources for Economic Evaluation in Healthcare

Economic evaluation in healthcare relies heavily on the availability and quality of data. Robust data inputs are essential for accurately estimating costs, outcomes, and comparative effectiveness across interventions. These data come from a variety of sources, each with its strengths and limitations. The most common sources include clinical trials, real-world data, systematic reviews and meta-analyses, and administrative and claims databases. However, using these sources also presents challenges related to data collection, standardization, and quality.

1. Clinical Trials vs. Real-World Data

Randomized controlled trials (RCTs) are the gold standard for assessing efficacy. They provide high internal validity by minimizing bias through randomization, blinding, and strict protocols. In economic evaluations, RCTs offer reliable estimates of treatment effectiveness, adverse event rates, and short-term resource use.

Strengths of RCTs:

  1. Controlled environments ensure causal inference.
  2. Rich clinical detail and standardized outcome measurement.
  3. Ideal for short-term cost-effectiveness analyses.

Limitations for economic evaluation:

  1. Limited external validity due to strict inclusion/exclusion criteria.
  2. Short follow-up periods restrict long-term outcome estimation.
  3. Cost data are often not collected or generalizable across settings.

RCTs are also expense and time-consuming.

To overcome these limitations, researchers increasingly use real-world data (RWD), which reflect the actual use and performance of interventions in routine clinical practice.

Sources of RWD include:

  • Electronic health records (EHRs)
  • Disease registries
  • Insurance claims
  • Patient-reported outcomes

Advantages of RWD:

  • Captures broader and more diverse patient populations.
  • Enables analysis of long-term effectiveness, safety, and resource utilization.
  • Supports budget impact and pragmatic evaluations.

However, RWD may suffer from missing data, confounding, and lack of randomization, which can affect the validity of findings.

References:

  • Makady A, de Boer A, Hillege H, et al. (2017). What is real-world data? A review of definitions. Value in Health, 20(7), 858–865.
  • Garrison LP, Towse A, Neumann PJ. (2007). Using real-world data for coverage and payment decisions. Health Affairs, 26(5), 1238–1247.

2. Systematic Reviews and Meta-Analyses

Systematic reviews and meta-analyses are critical tools for summarizing evidence from multiple studies and providing inputs for decision models. They combine data from diverse sources to improve the precision and generalizability of estimates used in economic evaluation.

Applications include:

  • Synthesizing clinical effectiveness (e.g., relative risk of mortality or hospitalization).
  • Informing transition probabilities in Markov models.
  • Providing pooled utility values or adverse event rates.

Meta-analyses enhance statistical power and reduce random error by aggregating data. In network meta-analyses, interventions that have not been directly compared can still be assessed relative to a common comparator—valuable for evaluating multiple treatment options in cost-effectiveness models.

However, the quality of the output depends on the quality of the included studies, and heterogeneity in populations, interventions, and outcomes must be carefully addressed.

Reference:

  • Higgins JPT, Thomas J, Chandler J, et al. (2019). Cochrane Handbook for Systematic Reviews of Interventions, 2nd ed. Wiley.

3. Administrative and Claims Databases

Administrative databases and insurance claims data provide a rich source of information for healthcare utilization, costs, and outcomes at the population level.

Common data sources include:

  • Medicare and Medicaid claims in the U.S.
  • Commercial insurance databases (e.g., MarketScan)
  • Hospital Episode Statistics (HES) in the UK
  • National Health Insurance databases (e.g., in Taiwan, South Korea)

Strengths:

  • Large sample sizes and long follow-up periods.
  • Useful for estimating real-world costs, resource use, and epidemiology.
  • Enables subgroup analysis and stratification by demographics or comorbidities.

Limitations:

  • Limited clinical detail—diagnoses may rely on coding (ICD codes), which may be inaccurate or inconsistent.
  • Lack of outcome measures like HRQoL, symptoms, or patient preferences.
  • Time lags in data availability.

Claims data are particularly useful in budget impact analysis and retrospective cost studies, though supplementary clinical data are often needed to assess effectiveness.

Reference:

  • Schneeweiss S, Avorn J. (2005). A review of uses of health care utilization databases for epidemiologic research on therapeutics. J Clin Epidemiol, 58(4), 323–337.

4. Challenges in Data Collection and Quality

Despite the abundance of data sources, several challenges can compromise the validity and applicability of economic evaluations:

a. Data Completeness and Accuracy

  • Missing or incomplete data can bias estimates, especially in observational datasets.
  • Inconsistencies in reporting or measurement tools (e.g., HRQoL instruments) affect comparability.

b. Generalizability

  • RCTs may not reflect real-world patients or practice patterns.
  • Claims data may not generalize beyond specific payer systems or geographies.

c. Data Linkage

  • Linking datasets (e.g., EHRs with claims, or registries with mortality databases) can enhance completeness but requires privacy safeguards and technical capacity.

d. Timeliness

  • Health technologies evolve rapidly, and data used in models may be outdated by the time of decision-making.

e. Transparency and Reproducibility

  • Analysts must clearly document data sources, assumptions, and methods to ensure transparency and peer scrutiny, particularly in models used for HTA and reimbursement.

Reference:

  • Ramsey SD, Willke RJ, Glick H, et al. (2015). Cost-effectiveness analysis alongside clinical trials II: An ISPOR Good Research Practices Task Force report. Value in Health, 18(2), 161–172.

Conclusion

Economic evaluations draw on diverse data sources to inform healthcare decision-making. While clinical trials offer rigor and internal validity, real-world data and claims databases provide essential insights into actual practice, long-term outcomes, and costs. Systematic reviews and meta-analyses are critical for synthesizing evidence, while data quality and collection challenges must be navigated carefully. The integration of multiple high-quality data sources—combined with methodological rigor—ensures that economic evaluations can effectively guide resource allocation and policy development in healthcare systems worldwide.

References

  • Garrison LP, Towse A, Neumann PJ. (2007). Using real-world data for coverage and payment decisions. Health Affairs, 26(5), 1238–1247.
  • Makady A, de Boer A, Hillege H, et al. (2017). What is real-world data? A review of definitions. Value in Health, 20(7), 858–865.
  • Higgins JPT, Thomas J, Chandler J, et al. (2019). Cochrane Handbook for Systematic Reviews of Interventions, 2nd ed. Wiley.
  • Schneeweiss S, Avorn J. (2005). A review of uses of health care utilization databases. J Clin Epidemiol, 58(4), 323–337.
  • Ramsey SD, Willke RJ, Glick H, et al. (2015). ISPOR CEA Task Force report. Value in Health, 18(2), 161–172.
ValueVitals
ValueVitals

Value Vitals helps healthcare leaders meet and exceed their goals. With over two decades of experience in healthcare consulting, Value Vitals leverages the power of the Science of Value, Technology & Programming, and Industry Know-how to overcome barriers and drive results that exceed expectations.