| |
Medicaid Expansion Association with End-Stage Liver Disease Mortality Depends on Leniency of Medicaid Hepatitis C Virus Coverage
|
| |
| |
Dowload the PDF here
• We found that improvements in ESLD mortality and LDR were associated with both Medicaid expansion and leniency of HCV coverage under Medicaid.
• mortality improvements in expansion states occurred because increased insurance coverage through Medicaid allowed for more access to essential treatments to prevent the progression of liver disease and mitigated the need for transplant.
• Interestingly, states that expanded Medicaid with lenient coverage also had the lowest annual population-adjusted ESLD deaths and highest annual LDR throughout the study period (including before expansion), suggesting an association between better baseline liver disease care and a greater likelihood of enacting Medicaid expansion and having lenient Medicaid coverage.
• States that expanded Medicaid and had more lenient coverage tended to have the most relative improvements in ESLD mortality and LDR. Our results add to a growing body of evidence suggesting that the implementation of lenient and widespread public health insurance policies may improve liver-related outcomes.
Slide presentation of this study at AASLD 2020: AFFORDABLE CARE ACT MEDICAID EXPANSION IMPROVED LIVER TRANSPLANT WAITLIST PLACEMENT AND SLOWED RACIAL DISPARITIES - (11/28/20)
I am sharing this as well to display the healthcare economic burden presented by liver disease. PLWH have higher rates of fatty liver, NAFLD, yet this condition does not get enough attention by HUV care providers.: ECONOMIC BURDEN OF NASH-ASSOCIATED CIRRHOSIS: US PAYOR'S PERSPECTIVE - (12/01/20)
-------------------------
12 June 2021
Nabeel Wahid, MD, MA1; Jihui Lee, PhD2; Alyson Kaplan, MD1,3; Brett E. Fortune, MD, MS1,3; Monika M. Safford, MD1,4; Robert S. Brown, Jr., MD, MPH1,3; Russell Rosenblatt, MD, MS1,3 1Department of Medicine, Weill Cornell Medicine/NewYork-Presbyterian Hospital, New York, NY, US;
2Department of Population Health Sciences, Weill Cornell Medicine, New York, NY, US; 3Division of Gastroenterology and Hepatology, Weill Cornell Medicine, New York, NY, US; 4Division of General Internal Medicine, New York, NY, US
Abstract
Background and Aims
The Affordable Care Act (ACA) expanded Medicaid around the same time that direct-acting antivirals (DAAs) became widely available for the treatment of hepatitis C virus (HCV). However, there is significant variation in Medicaid HCV treatment eligibility criteria between states. We explored the combined effects of Medicaid expansion and leniency of HCV coverage under Medicaid on liver outcomes.
Methods
We assessed state-level end-stage liver disease (ESLD) mortality, listings for liver transplant (LT), and listing-to-death ratio (LDR) for adults 25-64 years old using data from UNOS and CDC WONDER. States were divided into four nonoverlapping groups based on expansion status on January 1, 2014 (expansion versus non-expansion) and leniency of Medicaid HCV coverage (lenient versus restrictive coverage). Joinpoint regression analysis evaluated for significant changes in slope over time (joinpoints) during the pre-expansion (2009-2013) and post-expansion (2014-2018) time periods.
Results
We found significant changes in annual percent change (APC) for population-adjusted ESLD deaths between 2014-2015 in all cohorts except for the non-expansion/restrictive cohort, in which deaths increased at the same APC from 2009-2018 (APC +2.5% [95% CI 1.8, 3.3]). In the expansion/lenient coverage cohort, deaths increased at an APC of +2.6% (95% CI 1.8, 3.5) until 2014 and then tended to decrease at an APC of -0.4% (95% CI -1.5, 0.8). LT listings tended to decrease over time for all cohorts. For LDR, only the expansion/lenient and expansion/restrictive cohorts had statistically significant joinpoints.
Conclusion
Improvements in ESLD mortality and LDR were associated with both Medicaid expansion and leniency of HCV coverage under Medicaid. These findings argue for the implementation of more lenient and widespread public health insurance to improve liver disease outcomes, including mortality.
Introduction
In an effort to expand health the Affordable Care Act (ACA) was signed into law in 2010 and states to The first cohort of states to expand Medicaid through the ACA did so on January 1, 2014 with several other states following suit in subsequent years.(1, 2) In addition to (3) has been associated with improvements in a variety of domains in health care.(4-11)
Although liver transplantation (LT) has historically been subject to in care,(12-14) the impact that ACA has had on liver disparities and LT outcomes is understudied. In prior analysis, Medicaid expansion has been associated with worse transplant outcomes,(15) increases in the number of LT patients using Medicaid insurance, and a differential impact on waitlisting rates by by race/ethnicity. (16-18) Recently, Kumar et. al found that states expanding Medicaid had 8.3 fewer per than would have been expected if they had not (19) however their associations were not significantly different when states with poor HCV coverage under Medicaid were excluded.
Assessing the impact of ACA on liver disease is complicated by the advent of direct-acting antivirals (DAAs)) for hepatitis C virus (HCV), which started to become widely available around the same time as in 2014.(20, 21) DAAs have been credited with decreasing HCV mortality and decreasing the burden of HCV patients on the LT waitlist.(21, 22) Although Medicaid expansion has been associated with higher numbers of DAA prescriptions,(20) there is wide-spread state-level variation in access to HCV treatment under in access to HCV treatment under Medicaid. Restrictive criteria for HCV coverage under Medicaid vary across states based on fibrosis stage, sobriety period, and prescriber specialty (20, 23) Nonetheless, the combined effect of expansion status and leniency of HCV coverage under Medicaid has not been fully assessed.
Together, ACA and the release of DAAs for HCV were two potential boons to the care of liver patients in the last decade. The impacts of and DAA therapy on the field of hepatology have largely been in prior studies, but the two are likely interrelated and their impacts may be synergistic. The need for studies that not only assesses the effects of expanding public insurance but also the leniency of coverage for is paramount to guide future health policy. We used a state-level ecological design to study two important questions:
1) Did the impact of Medicaid expansion on end-stage liver (ESLD) mortality vary by leniency of HCV coverage under Medicaid? 2) What were the impacts of Medicaid expansion and Medicaid HCV coverage on listings for LT relative to ESLD mortality rates?
We analyzed the cohort at the level of the state and divided the states into four separate groups based on expansion status (expansion versus non-expansion) and leniency of Medicaid coverage for HCV (lenient coverage versus restrictive coverage). The 18 states with transplant centers that expanded Medicaid on January 1, 2014 were defined as "expansion states," and the 14 states with transplant centers that did not expand Medicaid between 2014-2018 were defined as "non-expansion states". States that expanded Medicaid after January 1, 2014 but before December 31, 2018 were excluded and states without a liver transplant center were also excluded. HCV treatment "leniency" was defined based on the Hepatitis C: State of Medicaid Access 2017 National Summary Report, which assigns state-specific grades from A-F based on restrictive HCV coverage policies under Medicaid in 2017.(24) Restrictive HCV coverage policies that were factored into grades included fibrosis requirements prior to treatment, sobriety periods prior to treatment, and limitations on prescriber eligibility. States with grades A-B (lenient coverage, n=21 states) were compared to states with grades C-F (restrictive coverage, n=11 states) in both expansion states and non-expansion states (Figure 1).
The study period was January 1, 2009 to December 31, 2013 (pre-expansion period) and January 1, 2014 to December 31, 2018 (post-expansion period). We included all individuals between the ages of 25-64 years. Individuals over 64 years old were excluded because many of these individuals were already Medicare beneficiaries prior to ACA Medicaid expansion. Individuals under 25 years old were excluded because many were covered under other insurance programs such as by the Children's Health Insurance Program or parental insurance plans. As such, this analysis was not designed to assess children or older adults.
Outcomes
Our primary outcome in this study was state-level ESLD mortality. Secondary outcomes included LT listings and listing-to-death ratio (LDR). LDR is a metric that was recently reported by our group and reflects access to transplantation—measuring how effectively states listed patients for LT relative to their mortality rate from ESLD.(25) That is, higher LDR may reflect higher relative access to LT listing. LT listings and ESLD mortality data were derived from the UNOS STAR dataset and CDC WONDER dataset, respectively.
Discussion
Medicaid expansion has resulted in millions of newly insured Americans, but its impact on liver disease has been underreported. Prior studies have shown an association between Medicaid expansion and changes in the insurance makeup of the LT waitlist, differences in waitlisting rates by race/ethnicity, and improvements in ESLD mortality.(16-19) Outside of liver disease, Medicaid expansion has been associated with improvements in all-cause mortality, cardiovascular mortality, and mortality for patients with certain malignancies.(30-33) Our study adds to the literature by conducting a uniquely designed 2x2 cohort analysis allowing us to analyze the combined impacts of the two most substantial changes in healthcare to impact liver disease over the last decade—Medicaid expansion and the availability of DAAs for HCV. We found that improvements in ESLD mortality and LDR were associated with both Medicaid expansion and leniency of HCV coverage under Medicaid.
Not surprisingly, expansion states with lenient HCV coverage experienced the best outcomes, as evidenced by declining ESLD deaths after expansion, while non-expansion states with restrictive coverage had the worst outcomes without any significant improvement in ELSD deaths. The other two cohorts (expansion/restrictive and non-expansion/lenient) both also had joinpoints in 2014 followed by relative improvements in the APC of ESLD deaths. Although ESLD deaths from HCV+ALD were small, only non-expansion states had significant increases in deaths between the two time intervals. Taken together, our findings support the possibility that Medicaid expansion and leniency of HCV coverage under Medicaid each impacted ESLD mortality and may have acted synergistically.
Although we suspect that leniency of Medicaid HCV coverage is a surrogate for the overall quality of Medicaid coverage in each state, DAA coverage is likely a key contributor to the improvements noted in ESLD mortality. DAAs have been previously shown to improve mortality in patients with HCV(22, 34) and as such, the number of individuals registering on the LT waitlist with HCV has decreased dramatically since 2014.(35) Nephew et al. found decreases in waitlisting rates for Black patients with HCV within expansion states after ACA Medicaid expansion. This was postulated to be from increased access to DAAs in expansion states compared to non-expansion states, resulting in fewer decompensation events necessitating LT.(18) Although we assessed different outcomes, our results suggest that incorporating state-specific Medicaid coverage leniency for DAAs is important in assessing the clinical impacts of Medicaid expansion.
Our study also assesses the association between Medicaid expansion/HCV coverage leniency with listings for LT relative to deaths from ESLD (LDR). As recently reported by our group, the advantage of using LDR is that it incorporates both listings for transplantation and mortality from ESLD into a single metric with a higher LDR suggesting more favorable listing rates relative to the number of liver disease deaths (as a measure of need).(25)In our analysis of LDR, states that expanded Medicaid had improvements in the APC of LDR after expansion while the two cohorts of states that did not expand Medicaid (either with lenient or restrictive coverage) failed to have improvements. Notably, the relative improvements seen in ESLD mortality and LDR within expansion states did not occur due to increased listings for LT. Rather, listings for LT tended to remain relatively steady or decrease over the entire study period in all cohorts. Thus, the relative improvements in APC of LDR within expansion states after 2015 are due to improvements in ESLD deaths (lower denominator) rather than increasing listings (higher numerator). We postulate, therefore, that the mortality improvements in expansion states occurred because increased insurance coverage through Medicaid allowed for more access to essential treatments to prevent the progression of liver disease and mitigated the need for transplant. Such treatments include DAA therapy for HCV (particularly in states with lenient HCV coverage) but also treatment of comorbidities such as diabetes, obesity, and mental health coverage such as alcohol cessation therapy.
Interestingly, states that expanded Medicaid with lenient coverage also had the lowest annual population-adjusted ESLD deaths and highest annual LDR throughout the study period (including before expansion), suggesting an association between better baseline liver disease care and a greater likelihood of enacting Medicaid expansion and having lenient Medicaid coverage. Prior studies have found that Medicaid non-expansion states have populations with lower average income, more comorbidities, and a higher proportion of obese patients than expansion states.(36) Thus, the impact that Medicaid expansion had on liver disease may have been greater in states with relatively less medical and socioeconomic need. As such, both Medicaid expansion and more lenient DAA therapy could have enormous benefits in the states that need it most but have not yet adopted such policies.
Although initial Medicaid expansion occurred on January 1, 2014, the majority of the joinpoints in our study occurred one year after expansion (2015), suggesting a one-year delay in accruing the impact of early ACA Medicaid expansion on ESLD deaths and LDR. The exhibited one-year delay is not surprising given that there should be an expected lag between policy implementation and its effect on clinical outcomes, including mortality. The clinical effect of ACA Medicaid expansion is not immediate and future studies assessing the impact Medicaid expansion on clinical outcomes should factor in this delay into their analysis rather than setting the year that states expanded as a preset changepoint.
Limitations of our study include its state-level design, which may be prone to ecological fallacies. We also lacked detailed information beyond ICD codes in the CDC WONDER database for assessing causes of death. As such, we aimed to capture deaths from ESLD overall but could not capture deaths from specific etiologies of liver disease (other than alcohol and hepatocellular carcinoma) as the
"underlying causes of death" and thus could not robustly estimate how many deaths were due to HCV directly. We did attempt to parse out deaths from HCV+ALD, but these values are likely underestimated because we only included one out of multiple overlapping ways that that HCV+ALD may be coded on death certificates. Additionally, we based state-by-state HCV coverage on "The Hepatitis C: State of Medicaid Access National Summary Report", which assessed HCV restrictions in 2017, but our analysis does not account for changes in state-specific HCV restrictions during our post-expansion period. Unfortunately, many states did not report publicly available data regarding HCV coverage under Medicaid in 2014, so no report card was published until 2017 when all states reported criteria.(24) Our study used the most comprehensive reported data for HCV Medicaid access available during our post-expansion period.
In conclusion, our uniquely designed 2x2 cohort analysis showed an association between ACA Medicaid expansion and ESLD mortality that varied depending on leniency of Medicaid HCV coverage. States that expanded Medicaid and had more lenient coverage tended to have the most relative improvements in ESLD mortality and LDR. Our results add to a growing body of evidence suggesting that the implementation of lenient and widespread public health insurance policies may improve liver-related outcomes.
| |
| |
| |
|
|
|
