A framework to evaluate the effects of small area variations in healthcare infrastructure on diagnostics and patient outcomes of rare diseases based on administrative data.

INTRODUCTION: Small area variations in healthcare infrastructure may result in differences in early detection and outcomes for patients with rare diseases.
METHODS: It is our aim to provide a framework for evaluating small area variations in healthcare infrastructure on the diagnostics and health outcomes of rare diseases. We focus on administrative data as it allows (a) for relatively large sample sizes even though the prevalence of rare diseases is very low, and (b) makes it possible to link information on healthcare infrastructure to morbidity, mortality, and utilization.
RESULTS: For identifying patients with a rare disease in a database, a combination of different classification systems has to be used due to usually multiple diseases sharing one ICD code. Outcomes should be chosen that are (a) appropriate for the disease, (b) identifiable and reliably coded in the administrative database, and (c) observable during the limited time period of the follow-up. Risk adjustment using summary scores of disease-specific or comprehensive risk adjustment instruments might be preferable over empirical weights because of the lower number of variables needed.
CONCLUSION: The proposed framework will help to identify differences in time to diagnosis and treatment outcomes across areas in the context of rare diseases.