• Lars Børty Nielsen
4. term, Science in Economics, Master (Master Programme)
BACKGROUND: Over the last 20 years, expenditures on hospital medicine have seen a large
increase in most countries. In Denmark, it rose from 1.6b DKK in 1997 to 9.1b DKK in 2019.
Most of this increase has come from medicine used in the treatment of cancer. It remains unclear
what the exact benefit from these expenses are, and the factors driving the increase are difficult
to discern. The board of Danish Regions has warned that a higher degree of rationing is coming.
This thesis presents detailed data on: drivers of increase, distribution of costs, estimates of
average costs and factors influencing costs.
METHODS: Data on chemotherapy treatments from Department of Oncology, Aalborg
University Hospital is combined with pharmacy billings of 1.2b DKK, in order to individually
price 267k treatments, of 13k patients from 2008-19. To account for censored data, mean
costs was estimated by the weighted available sample estimator, and the Gini coefficient with
bootstrapped CIs is utilized to measure statistical disparity. Factors influencing costs are modeled
by linear and linear mixed models.
RESULTS: Within the period, there has been a growth in inequality of expenditures on patients
as the Gini coefficient [95% CI] rose from 0.586 [0.572;0.598] in 2008 to 0.738 [0.727;0.747]
in 2019. Over the whole period, 6.9% of the most expensive patients accounted for 50% of
costs, ranging from 5.4% for lung cancer to 13.5% for brain cancer.
The estimated mean 3-year cost in DKK [95% CI] is: 74.863 [72k;77k], and within diagnoses,
it ranges from 118.331 [108k;129k] for breast cancer to 19.846 [17k;23k] for pancreas cancer.
Age is shown not to influence the price of individual treatments, but the number of treatments,
and therefore age influences total cost. A lower treatment intensity seems to set in around
75+ years. The average age at treatment weighted by cost was 62.4 years with wide variation
between diagnoses, ranging from 55.3 for brain cancer to 70.0 years for prostate cancer. Within
the study period, it rose from 59.2 to 66.0 years.
INTERPRETATION: The introduction of new expensive drugs is responsible for most of the
increase in total expenditures, while the increased number of patients and treatment intensity
accounts for a smaller portion. The major part of expenditures go increasingly towards fewer
patients, which is most likely due to an increased use of targeted treatment, where expensive
drugs only benefit few. Resources are increasingly spent on older patients, which is likely due to
use of new, less harsh treatments where health at a high age was previously too fragile.
KEYWORDS: Cost of cancer drugs; Censored cost estimation; Expensive hospital medicine
JEL: C5, D4, D6, I11, I14, I18
Publication date2 Jun 2020
Number of pages77
ID: 333460086