Changes in chronic medication adherence, costs, and health care use after a cancer diagnosis among low-income patients and the role of patient-centered medical homes.

BACKGROUND: Approximately 40% of patients with cancer also have another chronic medical condition. Patient-centered medical homes (PCMHs) have improved outcomes among patients with multiple chronic comorbidities. The authors first evaluated the impact of a cancer diagnosis on chronic medication adherence among patients with Medicaid coverage and, second, whether PCMHs influenced outcomes among patients with cancer. METHODS: Using linked 2004 to 2010 North Carolina cancer registry and claims data, the authors included Medicaid enrollees who were diagnosed with breast, colorectal, or lung cancer who had hyperlipidemia, hypertension, and/or diabetes mellitus. Using difference-in-difference methods, the authors examined adherence to chronic disease medications as measured by the change in the percentage of days covered over time among patients with and without cancer. The authors then further evaluated whether PCMH enrollment modified the observed differences between those patients with and without cancer using a differences-in-differences-in-differences approach. The authors examined changes in health care expenditures and use as secondary outcomes. RESULTS: Patients newly diagnosed with cancer who had hyperlipidemia experienced a 7-percentage point to 11-percentage point decrease in the percentage of days covered compared with patients without cancer. Patients with cancer also experienced significant increases in medical expenditures and hospitalizations compared with noncancer controls. Changes in medication adherence over time between patients with and without cancer were not determined to be statistically significantly different by PCMH status. Some PCMH patients with cancer experienced smaller increases in expenditures (diabetes) and emergency department use (hyperlipidemia) but larger increases in their inpatient hospitalization rates (hypertension) compared with non-PCMH patients with cancer relative to patients without cancer. CONCLUSIONS: PCMHs were not found to be associated with improvements in chronic disease medication adherence, but were associated with lower costs and emergency department visits among some low-income patients with cancer.

Incorporating Prescription Drugs Into Affordable Care Act Risk Adjustment.

BACKGROUND: The 2010 Patient Protection and Affordable Care Act reformed the individual and small group health insurance markets and established a risk adjustment program to create a level playing field for competition. A new set of predictive models for measuring enrollee risk across plans was developed for the Patient Protection and Affordable Care Act-reformed markets, referred to as the Department of Health and Human Services Hierarchical Condition Category (HHS-HCC) models. Beginning in 2018, selected prescription drug classes were added to the models as risk markers. OBJECTIVE: We describe the motivations, concerns, methodology, and results of adding prescription drug utilization to the HHS-HCC models. METHODS: Separate HHS-HCC models are estimated by enrollee age and plan actuarial value. We defined and added 10 prescription drug classes, called RXCs, to the HHS-HCC adult models. RESULTS: Using selected RXCs alongside demographic and diagnostic indicators yielded modest overall improvement in HHS-HCC models' predictive power. Also, adding RXCs captures the higher costs of enrollees taking certain expensive pharmaceuticals and allows imputation of diagnoses for enrollees utilizing a drug but lacking the associated diagnosis. CONCLUSIONS: Including selected drugs in risk adjustment improved the models' predictive power. In addition, inclusion of selected drugs may discourage insurers from using formulary and drug benefit design to avoid enrollment of patients taking high-cost drugs, such as for HIV, multiple sclerosis, and rheumatoid arthritis, and improve access for enrollees taking these drugs. Adding RXCs also may improve plan risk measurement for plans with less complete diagnosis reporting.

A Comparison of Health Risk and Costs Across Private Insurance Markets.

BACKGROUND: The Patient Protection and Affordable Care Act (PPACA) established new parameters for the individual and small group health insurance markets starting in 2014. We study these 2 reformed markets by comparing health risk and costs to the more mature large employer market. STUDY DATA: For 2017, claims data for all enrollees in PPACA-compliant individual and small group market plans as well as claims data from a sample of large employer market enrollees. VARIABLES AND METHODOLOGY: Risk scores and total (unadjusted and risk-adjusted) per-member-per-month (PMPM) allowed charges. Differences across markets in enrollment duration, age, and geographic distribution are addressed. The analysis is descriptive. RESULTS: Compared with large employer market enrollees, health risk was 3% lower among PPACA small group market enrollees and 20% higher among PPACA individual market enrollees. After adjusting for differences in health risk, enrollees in the PPACA individual market had 27% lower PMPM allowed charges than enrollees in the large employer market and enrollees in the PPACA small group market had 12% lower PMPM allowed charges than enrollees in the large employer market. CONCLUSIONS: On average, the PPACA individual market enrolls sicker individuals than the 2 group markets. But this does not translate to higher health costs; in fact, enrollees in the PPACA individual market accumulate lower allowed charges than enrollees in the large employer market. Lower-income enrollees particularly accumulate lower allowed charges. Narrower networks and increased enrollee cost-sharing among individual market plans, though they may reduce the value of coverage, likely significantly reduce allowed charges.

Changes in chronic medication adherence in older adults with cancer versus matched cancer-free cohorts.

OBJECTIVES: A cancer diagnosis can influence medication adherence for chronic conditions by shifting care priorities or reinforcing disease prevention. This study describes changes in adherence to medications for treating three common chronic conditions - diabetes, hyperlipidemia, and hypertension - among older adults newly diagnosed with non-metastatic breast, colorectal, lung, or prostate cancer. METHODS: We identified Medicare beneficiaries aged ≥66 years newly diagnosed with cancer and using medication for at least one chronic condition, and similar cohorts of matched individuals without cancer. To assess medication adherence, proportion of days covered (PDC) was measured in six-month windows starting six-months before through 24 months following cancer diagnosis or matched index date. Generalized estimating equations were used to estimate difference-in-differences (DID) comparing changes in PDCs across cohorts using the pre-diagnosis window as the referent. Analyses were run separately for each cancer type-chronic condition combination. RESULTS: Across cancer types and non-cancer cohorts, adherence was highest for anti-hypertensives (90-92%) and lowest for statins (77-79%). In older adults with colorectal and lung cancer, adherence to anti-diabetics and statins declined post-diagnosis compared with the matched non-cancer cohorts, with estimates ranging from a DID of -2 to -4%. In older adults with breast and prostate cancer cohorts, changes in adherence for all medications were similar to non-cancer cohorts. CONCLUSION: Our findings highlight variation in medication adherence by cancer type and chronic condition. As many older adults with early stage cancer eventually die from non-cancer causes, it is imperative that cancer survivorship interventions emphasize medication adherence for other chronic conditions.

Providers' mediating role for medication adherence among cancer survivors.

BACKGROUND: We conducted a mediation analysis of the provider team's role in changes to chronic condition medication adherence among cancer survivors. METHODS: We used a retrospective, longitudinal cohort design following Medicare beneficiaries from 18-months before through 24-months following cancer diagnosis. We included beneficiaries aged ≥66 years newly diagnosed with breast, colorectal, lung or prostate cancer and using medication for non-insulin anti-diabetics, statins, and/or anti-hypertensives and similar individuals without cancer from Surveillance, Epidemiology, and End Results-Medicare data, 2008-2014. Chronic condition medication adherence was defined as a proportion of days covered ≥ 80%. Provider team structure was measured using two factors capturing the number of providers seen and the historical amount of patient sharing among providers. Linear regressions relying on within-survivor variation were run separately for each cancer site, chronic condition, and follow-up period. RESULTS: The number of providers and patient sharing among providers increased after cancer diagnosis relative to the non-cancer control group. Changes in provider team complexity explained only small changes in medication adherence. Provider team effects were statistically insignificant in 13 of 17 analytic samples with significant changes in adherence. Statistically significant provider team effects were small in magnitude (<0.5 percentage points). CONCLUSIONS: Increased complexity in the provider team associated with cancer diagnosis did not lead to meaningful reductions in medication adherence. Interventions aimed at improving chronic condition medication adherence should be targeted based on the type of cancer and chronic condition and focus on other provider, systemic, or patient factors.

Effect of Medicaid Policy Changes on Medication Adherence: Differences by Baseline Adherence.

BACKGROUND: In 2001, the North Carolina (NC) Medicaid program reduced the number of days prescription supply that enrollees could fill from 100 days to 34 days and increased copayments for brand-name medications. Previous work has shown that a change in these policies led to a decrease in medication adherence from 2.9 to 8.0 percentage points in specific populations with chronic conditions. Studies have also shown that days supply limits and copayment increases have heterogeneous effects based on enrollees' baseline characteristics, including baseline adherence. However, this phenomenon has not been studied in the Medicaid population. We undertook this study to assess the heterogeneous effect of the NC Medicaid policy changes in groups with varying levels of baseline adherence. OBJECTIVE: To examine whether restrictions on days supply had heterogeneous effects in subgroups defined by medication adherence before the policy changes. METHODS: A partial difference-in-difference-in-differences model with fixed effects was used to compare medication adherence before and after the NC Medicaid policy changes among Medicaid enrollees subject to the policy changes because of their use of long prescriptions (> 40 days) as compared with (a) NC Medicaid enrollees using short prescriptions (< 40 days) before policy adoption, as well as (b) Medicaid enrollees in Georgia restricted to a 31 days supply through the study period. Medicaid enrollees were included if they filled a prescription for 1 of the following medication classes: antihypertensives, lipid-lowering drugs, or antipsychotics. The effect of the policy changes on medication adherence, calculated using the proportion of days covered (PDC) each quarter by baseline adherence level and clinical condition group, was studied. Average adherence levels over the 18-month prechange period were used to stratify individuals into 3 baseline adherence groups: fully adherent (PDC ≥ 80%), partially adherent (50%-79%), and nonadherent (PDC ≤ 50%). RESULTS: Enrollees fully adherent at baseline observed a 2.0 (P = 0.001) and 1.2 (P < 0.001) percentage-point decline in adherence for the lipid-lowering drug and antihypertensive cohorts, respectively, in the period after the policy changes. The nonadherent and partially adherent cohorts in the statin group observed an increase in adherence by 1.7-2.6 (P < 0.05) percentage points in the post-index period. CONCLUSIONS: Adherence changes after cost containment policies have a heterogeneous effect on individuals with varying baseline adherence in the Medicaid population. Individuals fully adherent at baseline decreased adherence following policy changes, while individuals partially adherent and nonadherent at baseline either had no change or showed increases in adherence, possibly because of increased contact with pharmacists and clinicians required by shorter prescription lengths. Managed care strategies to control costs should take into consideration the heterogeneity of responses by the enrollees to these policies. Furthermore, policies that consider baseline characteristics of enrollees may be more effective in improving adherence. DISCLOSURES: This study was partly funded by a grant from the Robert Wood Johnson Foundation for use in data creation. Maciejewski was supported by a Research Career Scientist Award from the Department of Veterans Affairs (RCS 10-391) and owns stock in Amgen. Farley reports consultancy fees from Daiichi Sankyo outside of the conduct of this study. The other authors report no financial or other conflicts of interest related to the subject of this article. The views expressed in this article are those of the authors and do not reflect the position or policy of the Centers for Medicare & Medicaid Services, University of North Carolina at Chapel Hill, Department of Veteran Affairs, or Duke University. Study design and concept were contributed by Amin and Domino, along with Farley and Maciejewski. Domino collected the data, and data interpretation was performed primarily by Amin, along with Domino, with assistance from Farley and Maciejewski. The manuscript was primarily written by Amin, along with Domino, and revised by all the authors.

Group-based trajectory analysis applications for prognostic biomarker model development in severe TBI: a practical example.

Over the last decade, biomarker research has identified potential biomarkers for the diagnosis, prognosis, and management of traumatic brain injury (TBI). Several cerebrospinal fluid (CSF) and serum biomarkers have shown promise in predicting long-term outcome after severe TBI. Despite this increased focus on identifying biomarkers for outcome prognostication after a severe TBI, several challenges still exist in effectively modeling the significant heterogeneity observed in TBI-related pathology, as well as the biomarker-outcome relationships. Biomarker data collected over time are usually summarized into single-point estimates (e.g., average or peak biomarker levels), which are, in turn, used to examine the relationships between biomarker levels and outcomes. Further, many biomarker studies to date have focused on the prediction power of biomarkers without controlling for potential clinical and demographic confounders that have been previously shown to affect long-term outcome. In this article, we demonstrate the application of a practical approach to delineate and describe distinct subpopulations having similar longitudinal biomarker profiles and to model the relationships between these biomarker profiles and outcomes while taking into account potential confounding factors. As an example, we demonstrate a group-based modeling technique to identify temporal S100 calcium-binding protein B (S100b) profiles, measured from CSF over the first week post-injury, in a sample of adult subjects with TBI, and we use multivariate logistic regression to show that the prediction power of S100b biomarker profiles can be superior to the prediction power of single-point estimates.

S100b as a prognostic biomarker in outcome prediction for patients with severe traumatic brain injury.

As an astrocytic protein specific to the central nervous system, S100b is a potentially useful marker in outcome prediction after traumatic brain injury (TBI). Some studies have questioned the validity of S100b, citing the extracerebral origins of the protein as reducing the specificity of the marker. This study evaluated S100b as a prognostic biomarker in adult subjects with severe TBI (sTBI) by comparing outcomes with S100b temporal profiles generated from both cerebrospinal fluid (CSF) (n = 138 subjects) and serum (n = 80 subjects) samples across a 6-day time course. Long-bone fracture, Injury Severity Score (ISS), and isolated head injury status were variables used to assess extracerebral sources of S100b in serum. After TBI, CSF and serum S100b levels were increased over healthy controls across the first 6 days post-TBI (p ≤ 0.005 and p ≤ 0.031). Though CSF and serum levels were highly correlated during early time points post-TBI, this association diminished over time. Bivariate analysis showed that subjects who had temporal CSF profiles with higher S100b concentrations had higher acute mortality (p < 0.001) and worse Glasgow Outcome Scale (GOS; p = 0.002) and Disability Rating Scale (DRS) scores (p = 0.039) 6 months post-injury. Possibly as a result of extracerebral sources of S100b in serum, as represented by high ISS scores (p = 0.032), temporal serum profiles were associated with acute mortality (p = 0.015). High CSF S100b levels were observed in women (p = 0.022) and older subjects (p = 0.004). Multivariate logistic regression confirmed CSF S100b profiles in predicting GOS and DRS and showed mean and peak serum S100b as acute mortality predictors after sTBI.

CSF Bcl-2 and cytochrome C temporal profiles in outcome prediction for adults with severe TBI.

The biochemical cascades associated with cell death after traumatic brain injury (TBI) involve both pro-survival and pro-apoptotic proteins. We hypothesized that elevated cerebrospinal fluid (CSF) Bcl-2 and cytochrome C (CytoC) levels over time would reflect cellular injury response and predict long-term outcomes after TBI. Cerebrospinal fluid Bcl-2 and CytoC levels were measured for 6 days after injury for adults with severe TBI (N=76 subjects; N=277 samples). Group-based trajectory analysis was used to generate distinct temporal biomarker profiles that were compared with Glasgow Outcome Scale (GOS) and Disability Rating Scale (DRS) scores at 6 and 12 months after TBI. Subjects with persistently elevated temporal Bcl-2 and CytoC profiles compared with healthy controls had the worst outcomes at 6 and 12 months (P≤0.027). Those with CytoC profiles near controls had better long-term outcomes, and those with declining CytoC levels over time had intermediate outcomes. Subjects with Bcl-2 profiles that remained near controls had better outcomes than those with consistently elevated Bcl-2 profiles. However, subjects with Bcl-2 values that started near controls and steadily rose over time had 100% good outcomes by 12 months after TBI. These results show the prognostic value of Bcl-2 and CytoC profiles and suggest a dynamic apoptotic and pro-survival response to TBI.