• Clinical and economic impact of statin therapy compliance on hyperlipidemic patients with concomitant CHD risk factors

      Samant, Navendu Dinanath; Shaya, Fadia T. (2010)
      In cholesterol management, a common and widely used class of drug is 3-hydroxyl-3-methylglutaryl coenzyme A reductase inhibitor (statin). These drugs lower cholesterol by slowing down the production of cholesterol and by increasing the liver's ability to remove LDL cholesterol already in the blood. Statins are well tolerated, and serious side effects such as liver problems, muscle soreness, pain, and weakness are rare. Compliance to prescribed medication is an important component of appropriate pharmacotherapy. Maryland Medicaid managed care (MCO) data were used to examine risk factors of patient non-compliance to statin therapy and its impact on healthcare utilization and the risk of having a new cardiovascular event among patients with hyperlipidemia and one or more CHD risk factors. The sample included Maryland Medicaid MCO patients who were diagnosed with hyperlipidemia and an additional CHD risk factor and were statin drug users. The proportion of patients compliant to statin therapy was 38 percent. The average age of the sample was 50. Over half of the sample was African American (50.69 percent), the rest Caucasian (41.54 percent) and Others (7.77 percent). Results for predictors of statin therapy compliance indicated that age and race had a significant impact on compliance. Although statistically non-significant, compliance was higher among patients with higher Charlson Comorbidity Index (CCI) scores. The probability of being compliant to statin therapy increased with age. African Americans (odds ratio = 0.662, 95 percent CI = 0.580-0.756) and Other race (odds ratio = 0.689, 95 percent CI = 0.549-0.865) groups were less compliant to statin therapy than Caucasians. The cost model indicated that compliant patients did not differ in cardiovascular disease (CVD) related medical costs from semi-compliant or non-compliant patients. The results of the CV risk model indicated that patients who were compliant to statin therapy had lower risk of a cardiovascular event than patients not compliant to statins and the findings were statistically significant. In conclusion, the study did not find significant evidence to prove that compliant patients would have lower CVD related costs. Risk of CV event was lower in compliant patients and was statistically significant. The study was conducted from January 1, 2002 through June 31, 2005. Future studies could examine the impact of compliance over a longer period of time in the Medicaid population.
    • Community Treatment Patterns for Pediatric Hypertension in Medicaid-insured Youth

      Gbarayor, Confidence Mona'le; Shaya, Fadia T. (2011)
      Little is known about at-risk youth in terms of hypertension (HTN) treatment. This study explored the community-based prevalence of pediatric hypertension treatments among Medicaid-insured youth. We examined clinician-reported diagnoses (CR-DX) and racial/ethnic disparities in outpatient antihypertensive (AHT) medication use, days of persistence of AHT use, inpatient hospitalizations, and emergency department (ED) visits. A cross-sectional design was used for this study, employing retrospective data from personal summary, medical, pharmacy, and inpatient administrative claims files for youth continuously enrolled in 11 states Medicaid programs in the year 2003. Multivariate logistic regression adjusted for covariates measured prevalence of CR-DX of HTN, dispensed AHTs, hospitalizations, and ED visits; and Cox proportional-hazards regression was used to analyze persistence of AHT use by racial/ethnic group. A total of 7,795,395 youth, < 18 years of age, were eligible for this study. Analysis included 7,782 youth (0.10%) with ≥ 2 CR-DX of HTN. Primary HTN was 13 times more common than secondary HTN (80% vs. 6%). African American youth had the greatest likelihood of diagnosed HTN compared to Caucasian youth (OR=1.27, p<0.0001). Approximately 46% of youth diagnosed with HTN and 0.32% of youth without diagnosed HTN had dispensed AHTs. No racial/ethnic disparities in treatment were observed: African American youth (OR=1.06, p=0.43) and Hispanic youth (OR=0.96, p=0.59) were as likely as Caucasian youth to be treated with an AHT medication following diagnosis of HTN. Among youth with diagnosed primary HTN, persistence on AHT therapy did not differ among racial/ethnic groups. Older youth, aged 10 to 14 years (HR=1.40, p=0.025) and those aged 15 to 17 years (HR=1.45, p=0.011) had significantly shorter persistence with AHT therapy. Youth eligible for Supplemental Security Income (SSI), i.e. those with disabilities, had longer persistence on AHT therapy (HR=0.79; p=0.013) than those in other eligibility groups. African American and Hispanic youth with HTN were as likely as Caucasian youth to have a hospitalization or ED utilization at any time. The findings from this study highlight specific subpopulations (i.e. African American, disabled, and older youth) needs in HTN treatment that warrant further research, to assure optimal community-based care.
    • The Creation of Objective Performance Criteria and Generation of Predictive Models among Medical Devices in a Vascular Space

      Gressler, Laura; Shaya, Fadia T.; 0000-0003-2042-2174 (2021)
      Background: Objective Performance Criteria (OPC) have been explored as a tool to address the growing pressures to expedite device approval and enhance active surveillance. Existing data infrastructures can be employed to develop OPC to evaluate the use of devices, and can be further leveraged to develop predictive models. The objective of this dissertation was to: (1) Develop a framework for the creation of OPC, (2) Compare the use of stent, atherectomy, and combination of stent and atherectomy, and (3) Formulate a predictive model used to predict the probability of undergoing a major adverse limb event (MALE) or experiencing death following the aforementioned treatments. Methods: The framework was developed in 3 phases through (1) Review of the literature, (2) Engagement of key stakeholders, and (3) Feedback from an advisory committee. Retrospective cohort studies were conducted using the Vascular Quality Initiative (2010-2018). Logistic regression and the Fine-Gray subdistribution hazard model were used to compare short- and long-term MALE, respectively. A generalized linear model (GLM), a Least Absolute Shrinkage and Selection Operator (LASSO) regularized GLM, a gradient boosted decision tree, and random forest model were compared when used to predict MALE and mortality. Results: The developed framework consisted of 5 elements: (1) Identification of Medical Devices, (2) Engagement of Key Stakeholders, (3) Selection of Data Source, (4) Performance of Appropriate Statistical Analyses, (5) Reporting of Findings. The odds of short-term MALE (0.94;95%CI:0.77-1.14) and hazards of long-term MALE (0.92;95%CI:0.82-1.04) were not significantly different in the combination stent and atherectomy group when compared to stent alone. The most effective predictive model was the gradient boosted decision tree (Area Under the Curve (AUC)= 0.7539) for MALE and the LASSO regularized GLM (AUC=0.7930) for mortality. Conclusions: The developed framework provides a guide and needed foundation for the continued generation of OPC. Applying the identified statistical steps in the framework to an existing data infrastructure showed that patients receiving combination stent and atherectomy do not experience significantly different rates of MALE compared to stent alone. Predictive models generated using the infrastructure demonstrated the ability of machine learning techniques to generate robust predictive models within the vascular space.
    • Medicare Disabled Patients with Hepatitis C: Determinants of Quality of Care Receipt, Peg-Interferon Treatment Initiation, and Risk of Metabolic and Vascular Disorders

      Chirikov, Viktor; Shaya, Fadia T.; 0000-0002-9480-0580 (2015)
      Background: Due to years of undetected hepatitis C virus (HCV) infection, the burden of liver and extrahepatic disorders will continue to increase in the US. HCV patients receiving Social Security Disability Benefits represent~70% of HCV patients in Medicare and are an understudied population facing numerous barriers to HCV management. We explored pre-treatment quality of care (QC) patterns, determined the factors associated with differential QC receipt and peg-interferon treatment initiation, and examined the effectiveness of peg-interferon therapy for ?24 weeks at reducing metabolic/vascular risk in Medicare disabled HCV patients. Methods: Medicare claims (2006-2009) linked to the Area Health Resource Files were used. We used a random forest model of conditional inference trees to aggregate QC indicators into high, good, fair, and low QC groupings. Ordinal partial proportional odds regression modelled the receipt of differential QC levels. Modified Poisson regressions, propensity-score weighted for the level of QC received, examined the association between treatment initiation and patient- and county-level characteristics. Poisson regression with weights for treatment selection, discontinuation, and informative censoring due to mortality quantified the effect of peg-interferon treatment on the risk of incident mild, severe, or mild and severe metabolic/vascular events, compared to the untreated. Results: We identified 1,936 patients with continuous enrolment, of whom 10.4% initiated peg-interferon. The five strongest QC metrics predicting treatment included "having received liver biopsy", "HCV genotype testing", "visit to specialist", "confirmation of HCV viremia", and "iron overload testing". While county of residence had no effect on QC receipt, residence in rural counties with high screening capacity was associated with higher prevalence ratio [PR] of treatment initiation (PR=1.42, p=0.09). High QC (PR=5.61, p<0.01) and good QC (PR=2.46, p<0.01) were associated with higher treatment rates. Multiple comorbidities were associated with lower odds of QC receipt (OR=0.76, p=0.05) and treatment initiation (PR=0.27, p<0.01). Over two years of follow-up, there was no difference in metabolic/vascular risk between those treated≥24 weeks (n=43) and untreated (n=879) patients. Conclusion: As barriers to eradicating the HCV infection would likely persist even with novel interferon-free regimens, future research should use our findings to better characterize and optimize treatment in HCV patients with disabilities.
    • Medication Adherence, Cardiovascular Disease Hospitalizations, and Health Care Utilization among Medicaid Populations with Diabetes, Hypertension, and/or Hyperlipidemia

      Yan, Xia; Shaya, Fadia T. (2011)
      Few studies have been conducted that explore the impact of cardiovascular disease (CVD) hospitalizations on patient adherence to CVD drugs. Medicaid beneficiaries at high-risk for CVD, a vulnerable population, had not been a focus of prior studies in exploring the impact of patient adherence to cardioprotective drugs on health utilization. The study population was Maryland Medicaid beneficiaries with hypertension, hyperlipidemia, or diabetes. Adherence rates to antidiabetic (AD), antihypertensive (AH), or antihyperlipidemic (AL) drugs were measured in the 6-month pre-index and post-index periods. Patients with and without CVD hospitalization were matched on gender, race, age (±5 years), and pre-index adherence rate (±5%). Adherence rate to cardioprotective drugs among patients with a CVD event was measured in the 6-month, one-year, or two-year post-index periods. Patients with and without adherence to cardioprotective drugs were matched using propensity score matching. Conditional logistic regression analysis was conducted to explore the impact of CVD hospitalization on patient adherence to AD, AH, or AL drugs. The association of patient adherence to cardioprotective drugs with hospital utilization was explored using Cox regression analysis. Patients with CVD hospitalization were more likely than those without CVD hospitalization to be adherent to AD or AH drugs in the 6-month post-index period. Improved adherence to AD, AH, or AL drugs comparing the 6-month post-index to the 6-month pre-index period was more likely to be shown in patients with CVD hospitalization than in those without CVD hospitalization. A substantial proportion of patients were not adherent to AD, AH, or AL drugs during the 6-month post-CVD hospitalization period. Patients who were adherent to at least one cardioprotective drug were less likely than those who were not adherent to be associated with the risk of experiencing any all-cause hospitalizations, CVD-related hospitalizations, or CVD-related emergency room (ER) visits. Hospital-based education might provide an opportunity for patients to improve their awareness of the importance of medication adherence. Follow-up interviews by health professionals could help patients maintain adherence to medication after hospital discharge. Medicaid beneficiaries at high-risk for CVD are vulnerable, and improving adherence to cardioprotective drugs in this population might contribute to a reduction in hospital utilization.
    • Modeling cardiovascular health outcomes in Medicaid hypertensive patients - Effect of patient adherence

      Gu, Sen; Shaya, Fadia T. (2008)
      Background. Low adherence rates to antihypertensive pharmacotherapy have been documented everywhere, especially among people of lower socioeconomic status such as Medicaid beneficiaries. Existing studies tend to focus on hypertensive patients who received mono-therapy or measure adherence at a fixed point in time, which may generate external validity issues or not be suitable for patients whose survival periods vary considerably. Purpose. To examine the relationship between patient adherence to antihypertensive medication and subsequent risk of cardiovascular events by using a new approach to measure adherence. Methods. A secondary data analysis of a mid-Atlantic Medicaid Managed Care Organization (MCO) data was conducted. We used Ordinary Least Squares (OLS) models and logistic regression models to examine predictors of patients' adherence at one year post index date. We also used Cox's Proportional Hazard models and pooled logistic regressions to investigate the relationship between adherence and subsequent risk of cardiovascular events. Adherence was first measured as a Cumulative Medication Acquisition (CMA) at six months post index date and second as CMA at each month since six months post index date until the date of disenrollment, date of a cardiovascular event, or the end of the study time frame, whichever comes first. We compared the area under Receiver Operating Characteristic (ROC) curves of the time-constant and time-varying approaches in terms of their utility in assessing risk of cardiovascular events. Results. A total of 3091 patients were included for the fist study aim, with an estimated mean annual CMA score of 0.649 (median = 0.652). For other study aims, 7939 patients met the inclusion criteria and 140 cardiovascular events were found after six months post index date. A higher CMA score was found to be protective against cardiovascular events, when controlling for potential confounders. For each 10% increase of CMA score, the hazard of cardiovascular events decreased by approximately 14% and the effect was significant (P = 0.0134). The model with CMA measured with time-varying approach also yielded by similar results and a Mann-Whitney test indicated there was no significant difference in terms of utilities of time-constant and time-varying approach in predicting risk of cardiovascular events. Conclusions. In this Medicaid population, patient adherence to antihypertensive medications is affected by sociodemographic factors and health status. Adherence at six months post index date is a significant predictor of subsequent risk of cardiovascular events. Adherence measured at a fixed point in time and by time varying method are similar in predicting cardiovascular health outcomes.
    • Predictors and subsequent healthcare utilization associated with CDC-guideline opioid thresholds among commercially insured new chronic opioid users

      Calabrese, Martin Joseph; Shaya, Fadia T.; 0000-0003-4304-396X (2022)
      Background: The receipt and subsequent healthcare utilization surrounding new chronic opioid users (NCOUs) is multifactorial and includes clinical, demographic, and state-level factors. This study evaluated i) predictors for receipt of chronic opioid therapy informed by CDC-guideline morphine milligram equivalent (MME)/day recommendations and the short-term healthcare utilization measured by ii) total healthcare costs and iii) all-cause hospitalization after new chronic opioid use. Methods: We conducted a retrospective cohort study using IQVIA PharMetrics® Plus for Academics commercial claims with NCOUs identified between January 2014 through March 2015. NCOUs were defined as having at least 60-days coverage of opioids within a 90-day period with at least a 30-day opioid-free period prior to the date of the first qualifying opioid prescription. The short-term healthcare observation period began the 91st day or the day after last day coverage of the chronic opioid period, whichever is sooner. We placed NCOUs in one of three-tiered risk-based opioid thresholds categories: low (> 0 to < 50 MME/day), medium (≥ 50 to < 90 MME/day), and high (≥ 90 MME/day). A multinomial logistic regression was used to evaluate the impact of prescription drug monitoring program (PDMP) rigor on the receipt of respective opioid thresholds. A generalized linear model and multivariable logistic regression was utilized to evaluate the incremental total healthcare costs (ITHC) and odds of incurring a hospitalization between the thresholds, respectively. Results: A total of 16,684 NCOUs were identified. Among the NCOUs, a state with high PDMP robustness had lower odds of receiving medium (0.74; 0.62-0.90) and high (0.74; 0.59-0.92) thresholds when compared to low. When compared to low, medium and high were found to have higher ITHC, (US$, 95% Confidence Interval [CI]) $1,429 (947-1,911) and $1,775 (1,183-2,368), respectively. When compared to medium, the ITHC for high $267 (-310-844) was non-significant. When evaluating odds of all-cause hospitalization (adjusted odds; 95% CI), when compared to low, no difference was identified with medium (1.01; 0.94-1.28) or high (1.01; 0.84-1.22). Conclusion: Among NCOUs, PDMP robustness was found to decrease the odds of subsequent receipt of higher thresholds. However, short-term healthcare costs and all-cause hospitalization did not differ among the thresholds.
    • Utilization and Cost of the Biologic Disease Modifying Anti-rheumatic Drugs among Medicare Beneficiaries with Rheumatoid Arthritis

      Gaitonde, Priyanka; Shaya, Fadia T. (2018)
      Background: Disease modifying anti-rheumatic drugs (DMARDs) are essential for symptom control among rheumatoid arthritis (RA) patients. Biologic DMARDs are expensive and typically used among moderate to severe RA patients. The prevalence of RA is higher among Medicare beneficiaries compared to the rest of the population in the U.S (2% vs. 0.6%). The coverage rules of Medicare, in addition to access factors and patient preferences, may influence the use of facility-administered, infusible biologics (Part B covered) and home-administered self-injectable/oral biologics (Part D covered). However, there is limited information about utilization patterns of biologic DMARDs by route of administration and their impact on Medicare spending overall. The goals of this dissertation were to identify patient factors and healthcare expenditure associated with biologic DMARD use by route of administration among Medicare beneficiaries with RA. Methods: The study population consisted of Medicare beneficiaries with RA from the 5% random sample of the Chronic Conditions Warehouse database from 2006-2015. First, the study analyzed patient-level factors associated with biologic DMARD use by route of administration using generalized estimating equations. Second, adherence (PDC>80%), discontinuation, and switching patterns for biologic DMARDs were measured, accounting for patient level-factors , using logistic regression, Cox proportional hazards models, and chi-square analyses, respectively. Third, the study compared annualized average healthcare costs of patients who were adherent to versus non-adherent to biologic DMARDs. Results: Among Medicare beneficiaries diagnosed with RA who received DMARD treatment (n=46,002), 71.8% (n=33,028) used traditional DMARDs, and among biologic DMARD users (n=12,931), twice as many used infusible biologics (18.3%, n=8,436) compared to self-injectable/oral (9.9%, n=4,538). Beneficiaries who were low-income subsidy (LIS) recipients i.e. who had lower out-of-pocket costs for using biologics, (compared to non-LIS) had 4.54 times higher odds of using self-injectable/oral biologics (95% CI: 4.2 - 5.0) and 5% lower odds of discontinuing biologic DMARDs (OR=0.94, 95% CI: 0.91-0.97). The total healthcare cost was lower among adherent compared to non-adherent infusible biologic DMARD users ($33,797 vs. $90,181; p<0.001) and among adherent vs. non-adherent self-injectable/oral biologic DMARD users ($64,977 vs. $80.908; p<0.05). Conclusions: Adherence and cost savings generated, as a result, varied by the biologic DMARD route of administration. Additionally, beneficiaries' LIS status was associated with the route of administration used and the discontinuation rates, indicating an association with their out-of-pocket costs. These findings are relevant to the discussion about the proposed transition of Part B covered infusible medications under the Part D which is projected to increase the beneficiary out-of-pocket cost. The evidence on adherence can also be used for value-based insurance design involving RA biologics. Future research could leverage the findings from this study to additionally explore how biosimilar products may impact overall biologic utilization and RA treatment spending.