OBJECTIVES

Health technology assessment programs are increasingly using longitudinal, patient-level data to assess the effect of new medicines on total cost of care. We analyzed such data to measure the impact of new pancreatic cancer therapies on other, non-drug medical expenditures.

METHODS

The latest update of the MEPS data, including the 21 Longitudinal Data Files which track subjects across the 5 rounds of interviews over 2 years, were obtained.

The Medical Conditions File was used to identify all patients who were diagnosed with a cancer of the pancreas, including International Classification of Diseases, Ninth Revision, Clinical Modification (ICD-9-CM) codes 157.0, 157.1, 157.2, 157.3, 157.4, 157.8, 157.9. Included subjects (N = 80) had a diagnosis of pancreatic cancer and available prescription data. Individual age, type of insurance coverage, and employment status were accounted for as covariates.

All analyses were performed using R version 3.6.1 on Ubuntu 19.04. Means and standard deviations were computed for the raw and inflation-adjusted total health care costs (excluding drug spending). Prescription drug costs for the period between 2009-2016 to include approval and use of (nab)-paclitaxel and erlotinib were also obtained and adjusted for inflation. The prescription drug expenses and proportion of healthcare spending were plotted along with a LOESS curve using the same parameters.

RESULTS

Between 2009 and 2016, inflation-adjusted per patient spending on pancreatic cancer care declined from $37,500 to $10,000, respectively. Prescription drug spending increased during the same time period from $2,100 to $6,100 per person (inflation-adjusted).

In effect, for every additional dollar spent on drugs for pancreatic cancer in 2016, there was a reduction in non-drug spending of $9.

CONCLUSIONS

Analysis of longitudinal, patient-level, real-world data provides a dynamic assessment of the impact of new medicines on well-being and medical utilization in those with pancreatic cancers.