Abstract
Background
The introduction of Imatinib in 2001 has brought a paradigm shift in the management of CML. Patients on TKI therapy continue to require hospitalizations, however, for progressive disease, treatment side effects and other unrelated causes. In our study we compared the cost of inpatient health care, mortality, length of stay (LOS) and complications for patients who had stem cell transplants to those on TKI therapy.
Methods
We queried the NIS database from the Healthcare Cost and Utilization Project (HCUP), Agency for Healthcare Research and Quality between 2002-2011 using ICD-9 code 205.1 for CML in the primary and secondary diagnosis fields. Patients 18 years or older were included in the analysis. Cost of hospitalization was adjusted for inflation in reference to 2011 and cost to charge ratio. We analyzed the trend in hospitalizations, cost and mortality. Linear and logistic regression models were generated to evaluate multivariate predictors of LOS, cost, mortality and complications. Odds ratios and odds estimates were generated comparing the group that underwent HSCT to the group that was treated with TKI therapy. We compared three groups: patients admitted for the transplant procedure (BMT procedure), patients readmitted post HSCT, and patients treated with TKIs. Multivariate analysis for complications from CML included splenic infarct, septic shock, splenomegaly, blast crises and DIC. Complications of graft versus host disease and graft rejection were included as they were complications of allogeneic transplant that warranted hospitalization. Age-related comorbidities, such as atrial fibrillation, congestive heart failure, and acute and chronic renal failure were also analyzed to further delineate the reason for hospitalization. A p value of <0.05 was considered significant.
Results
A total of 38,950 hospitalizations (weighted n= 19,1285) were analyzed (male 54.6% and age 65.9±0.08). There was a decrease of 81.96 % in mortality from 2002 to 2011 (p<0.0001). The average age was 66.7 years in the non-transplant group, and 45.6 years in the transplant group (p = 0.0016). 64% in the TKI group had Medicare, compared to 23.7% in the transplant group (p<0.0001). The inpatient mortality for transplant was 8.9%, but was 6.3% in the group readmitted after a successful transplant. It was 7.9 % in the TKI group (p = 0. 032). Admissions due to age-related co-morbidities was 28.5 % in the transplant group and 50.8% in the TKI group (p<0.0001). Only 14% of patients in the TKI group were admitted for CML related problems vs. 23.7% in the transplant group (0.0001). The average length of stay was 7.05 days in the TKI group and 18.4 days in the transplant group. The average length for the transplant procedure was 33.85 days (p<0.0001). The average cost of hospitalization in the transplant group was $173,780, and was $46,955 in the TKI group. The transplant procedure cost $338,229 (p<0.0001). The odds of mortality (OR) are in favor of TKI therapy with an OR of 1.9 against the transplant procedure.
Discussion
Patients on TKI therapy have a lower mortality, average length of stay and hospitalization cost compared to the transplant group. The main reasons for hospital admission for patients on TKI therapy were age-related comorbidities, rather than complications of CML. The mortality in the TKI group was lower than the HSCT group. However, the yearly cost of TKI therapy must be taken into account for health care costs of non-transplant patients. At present, Imatinib costs $92,000/ year and Dasatinib $118,000/year. Hence, Imatinib therapy for even 4 years would be more expensive than a transplant. Therefore, TKI therapy provides improved mortality and shorter length of hospital stay at the cost of a net higher expense.
No relevant conflicts of interest to declare.
Author notes
Asterisk with author names denotes non-ASH members.
