CHICAGO–An analysis of Surveillance, Epidemiology, and End Results (SEER) and Medicare Part D data suggests that coverage barriers are not the only factors that can lead to non-adherence to tyrosine kinase inhibitors (TKIs) in patients with chronic myeloid leukemia (CML). High out-of-pocket costs and heavy subsidies for low-income patients significantly increased the risk of non-adherence among patients with continuous insurance coverage.
"TKIs have transformed CML from a deadly cancer to a chronic disease," lead author Chan Shen, PhD, MS, from the University of Texas MD Anderson Cancer Center in Houston, Texas, noted in his presentation of the results at the 2016 ASCO Annual Meeting. "While these drugs dramatically improve survival, non-adherence could decrease efficacy."
Previous research found that privately insured patients with high cost-sharing had lower adherence, while Medicare Part D beneficiaries receiving low-income subsidies to cover prescription drugs may have better adherence. Dr. Shen and authors analyzed data from the SEER registry linked with Medicare Part D subsidy data from 2007-2012 to determine the effect of insurance coverage on TKI adherence.
The cohort included 836 CML patients who were new TKI users with continuous Medicare Part D insurance coverage for the time period starting six months prior to beginning TKI therapy.
Patients were deemed non-adherent if the number of days of TKI supply covered less than 80 percent of the 180-day period after therapy initiation, meaning patients took medication as directed less than 80 percent of the time.
Patients who qualified for low-income subsidies were divided into three categories:
- Heavily subsidized (100% premium subsidy with no or low co-payment)
- Moderately subsidized (100% premium subsidy with high co-payment)
- No subsidy
The authors examined group differences in non-adherence according to patient and insurance plan characteristics and used logistic regression analysis to determine the relationship between treatment non-adherence and out-of-pocket costs per 30-day TKI supply. For comparison, they also collected patient characteristics for the subset of Medicare Part D participants who did not receive any low-income insurance subsidies.
Overall, 29 percent (n=244/836) of CML patients were non-adherent to TKI therapy during the study period.
Heavily subsidized patients had lower out-of-pocket costs (falling below the median costs of the entire cohort) than patients who did not receive low-income subsidies; despite the lower out-of-pocket costs, these patients had the highest rates of non-adherence (36.41%), which was significantly higher than the non-adherence rate in patients without subsidies (27.40%). See TABLE 1 for more results.
According to multivariable analysis, the factors strongly associated with non-adherence included heavily subsidized status and high out-of-pocket costs (TABLE 2).
"This population-based study found a strong association between out-of-pocket costs and non-adherence among CML patients enrolled in Medicare Part D without subsidy," Dr. Shen and authors concluded. "The higher rate of non-adherence among heavily subsidized patients, despite their substantially lower out-of-pocket costs, suggests the need for future research to explore factors other than coverage barriers in this vulnerable population."
Reference
Shen C, Zhao B, Zhou S, et al. Low income subsidy status outweighs insurance cost-sharing in predicting adherence to tyrosine kinase inhibitors among Medicare Part D beneficiaries with chronic myeloid leukemia. Abstract #6531. Presented at the 2016 ASCO Annual Meeting, June 4, 2016; Chicago, IL.
TABLE 1. Adherence Among the Study Participants | |||
Non-adherence, N (%) | Adherence, N (%) | p Value | |
Overall | 244 (29.19%) | 592 (70.81%) | |
Low-Income Subsidy Status | |||
Heavily subsidized | 71 (36.41%) | 124 (63.59%) | 0.036 |
Moderately subsidized | 33 (25.38%) | 97 (74.62%) | |
No subsidy | 140 (27.40%) | 371 (72.60%) | |
Payment for TKIs per Month Supply (quartile) | |||
First quartile (<$2.10) | 46 (22.22%) | 161 (77.78%) | <0.001 |
Second quartile ($2.10-$419.70) | 87 (41.23%) | 124 (58.77%) | |
Third quartile ($419.80-$1,007.60) | 19 (9.09%) | 190 (90.91%) | |
Fourth quartile (>$1,007.60) | 92 (44.02%) | 117 (55.98%) |
TABLE 2. Multivariable Logistic Model for Non-Adherence to TKI Therapy | |||
Odds Ratio | 95% CI | p Value | |
Low-Income Subsidy Group | |||
Heavily subsidized | 7.77 | 3.39-17.8 | <0.001 |
Moderately subsidized | 1.45 | 0.74-2.85 | 0.295 |
No subsidy | Reference group | ||
Payment for TKIs per Month Supply (quartile) | |||
First quartile (<$2.10) | 0.06 | 0.02-0.14 | <0.001 |
Second quartile ($2.10-$419.70) | 0.53 | 0.31-0.92 | 0.023 |
Third quartile ($419.80-$1,007.60) | 0.11 | 0.06-0.19 | <0.001 |
Fourth quartile (>$1,007.60) | Reference group |