Despite the growing international popularity of the use of health care claim data for research purposes, its use in patients with kidney disease in the Netherlands was still in its infancy. This observation has led to Manon van Oosten’s dissertation.
Healthcare expense data is a relatively new source of data that offers unique possibilities for research on large numbers of patients in a non-experimental setting. Healthcare claim databases contain very extensive and detailed data on large numbers of insured persons in a country. In addition, they contain unique data on healthcare use and healthcare costs.
Van Oosten’s dissertation has two objectives. The first concerns the investigation of possibilities and limitations of care declaration data for nephrological research. The second objective relates to the investigation of health care costs and health care resource use of chronic kidney damage (CNS) patients, being stage G4-G5 CNS, dialysis and kidney transplantation.
Features and usage
Worldwide (ten countries) thirteen databases were identified that were used for nephrological research. The Dutch health care claim data is managed by Vektis. The Vektis database contains declared care for almost all residents of the Netherlands as of 2012. This makes the database unique in its size and content and offers many opportunities for research to improve healthcare.
Validity of Dutch health care claim data
Unfortunately, the Vektis database contains no clinical data, so identification of CNS patients had to be based on diagnosis codes. Dialysis and transplant patients can be identified very accurately with care declaration data. However, for CNS patients without renal replacement therapy this is more challenging. Research results showed that the accuracy of care declaration data is remarkably higher in young patients than in older patients. The data have a low sensitivity for estimating CNS prevalence in the general population, especially in elderly CNS patients and with less advanced CNS, the data appear to be less accurate. Nevertheless, health care claim data may be valuable for estimating the prevalence of CNS in specific subgroups, especially in younger patients and patients with advanced CNS.
Healthcare costs and medication use
Unique to the Vektis database is the detailed information about healthcare costs and medication use. This made it possible to calculate the average annual healthcare costs of dialysis and transplant patients. These costs ranged from 77,566 euros for CAPD patients to 105,833 euros for patients with different treatment modalities in one year. Costs for kidney transplant patients were 85,127 euros in the year of transplantation and decreased rapidly in the first (29,612 euros) and second (15,018 euros) years after transplantation.
The health care utilization and associated health care costs of CNS patients (without renal replacement therapy) were also significantly higher than those of the general population. This difference was present at an early age and in the earlier stages of CNS. Notably, although health care costs increase with age in the general population, this study showed a decrease in health care costs in CNS patients 75 years of age or older. This was mainly attributable to lower hospital costs and lower medication costs.
Prescribed medication from patients can also be researched in a unique way using Vektis data. For example, polypharmacy was found to be a common problem in CNS patients (66-75%) as opposed to matched controls (7-18%). Analysis also revealed a markedly lower rate of antidepressant use in Dutch CNS patients than reported on CNS populations from other countries.
Conclusion
The thesis shows that the Vektis database has the potential to be an important data source for nephrological research. The database offers possibilities to analyze unique information at the national level as well as to make comparisons with matched subgroups from the general population. The results from this thesis have been used for the development of the Nieratlas website (www.nieratlas.nl), which was developed during the PhD research. This website maps CNS patients in the Netherlands in a new way using health care declaration data.
In order to improve the research potential of the Vektis database and stimulate its use, a sounding board group has been established to support future researchers in working with the Vektis database.
The promotion
Image: Maarten Nauw
Manon van Oosten obtained his PhD on 1 April 2022 with a thesis entitled ‘The use of health claims data for epidemiology and costs of chronic kidney disease’ at the University of Amsterdam. The promoters were Prof. Dr. KJ de Jager, affiliated with AMC-UvA, and Prof. Dr. HJG Bilo from the University of Groningen. The co-supervisors were Dr VS Stel, AMC-UvA, and Dr SJJ Logtenborg, Diakonessenhuis Utrecht.
