Abstract
Background/Aims
Timeliness of care is one of 4 main indicators of quality of care cited by the Institute of Medicine. In the US, an individual’s type of insurance or lack thereof has been implicated as a barrier to obtaining timely treatment, including HSCT. We compared the time to insurance approval between private and public payers among those who were undergoing evaluation for HSCT. Additionally, we evaluated if time to insurance approval is associated with survival after HSCT in patients (pts) with hematologic malignancies.
Methods
This is a retrospective cohort study that used the Insurance Transplant Database of an academic medical center. All pts evaluated for possible HSCT between 2007 and 2011 were included. Time to insurance approval (index of timeliness) was operationally defined in 3 ways: 1) payer approval – from request for approval to actual payer approval, 2) transplant speed – from payer approval to time of actual transplant, and 3) total time – from request for approval to transplant. Multivariate regression analysis was used to evaluate differences in time to approval between public and private payers. The pts who underwent HSCT were compared for pt-, disease- and transplant-related factors according to type of payer and speed of payer approval. The 3 indices of timeliness were dichotomized (using median) to evaluate if shorter (lower half) or longer (upper half) times were associated with 1 year overall survival (OS) using multivariate Cox proportional hazards regression.
Results
Of the 1389 pts evaluated for possible HSCT during the study period, 830 (60%) did not proceed to transplant: of these, 454 (55%) were not recommended for HSCT because transplant MD felt transplant was not beneficial, 119 (14%) were referred to other centers, 113 (14%) expired during the evaluation process, 89 (11%) did not want HSCT, 48 (6%) became ineligible because of significant risks due to mix of age, disease stage and comorbidities, and only 7 (1%) were denied by insurance. Of the 559 (40%) who underwent HSCT, 521 underwent first transplant: of these, 421 (80%) had private insurance, 97 (19%) had public payers (Medicare n=74, Medicaid n=23), and 3 (1%) self-pay. Pts with private insurance are likely to: be younger (53y vs 58y), whites, have higher income, reside in urban area, and have no comorbidities. Cohorts were similar in distributions of disease type, disease stage and type of transplant. Time to payer approval was longer in pts with private insurance than public payers [4 d (range 0-90) vs 0 d (0-28), p<0.0001], but time from approval to actual transplant was longer in pts with public payers than private insurance [65 d (14-277) vs 39 d (1-402), p<0.0001]. Total time to transplant was longer for public payers than private insurance [66 d (14-277) vs 48 d (1-407), p<0.001]. These differences persisted in multivariate analyses adjusting for significant covariates. In a subset analysis of the 509 HSCT pts (public or private payers) with hematologic malignancies, we tested if shorter vs longer approval times in the 3 indices of timeliness were associated with pt characteristics and 1 year OS. Pt characteristics did not differ between the groups with fast vs. slow approval times. Multivariate Cox regression adjusting for age, type of payer, and disease stage showed no significant differences in risk of death between slow and fast approval in the 3 indices of timeliness in the models that used: a) all pts, b) autologous HSCT in lymphoma (n=278), c) autologous HSCT in multiple myeloma (n=121); and d) allogeneic HSCT (n=110).
Conclusion
Insurance approval is generally fast, although the speed varies between public and private payers in HSCT. Among the cohort who successfully proceeded to HSCT, within the range of approval times observed, we did not see a difference in 1 year overall survival between shorter vs. longer approval times. While insurance approval may cause delays in timeliness of transplant, this study failed to show a significant association with survival.
Disclosures:
No relevant conflicts of interest to declare.
