The South Carolina Revenue and Fiscal Affairs Office, Health and Demographic Section (H & D), is a central state repository for health and human service data. Data housed at H & D and utilized for this project originated from Medicaid, the Department of Disabilities and Special Needs (DDSN), and the Department of Social Services. Through a series of statutes and agreements, agencies and organizations entrust data to H & D while retaining access control at all times. We obtained data use agreements from participating organizations, and the data linkages and analyses were performed at the H & D. Non-H & D investigators received aggregated data for review. Procedures for the protection of human subjects were reviewed and approved by the University of South Carolina Institutional Review Board.

We identified 22,463 members with IDD in the study period 2000–2014, of whom 42.2% were identified as having a hypertension diagnosis (Table 1). A total of 1,573 members met criteria for inclusion in our adherence study cohort (Table 1). Most (81%) adherence study cohort members were born between 1947 and 1980, making them slightly older than all members with IDD, of whom 73.8% were born between 1958 and 1992. Overall, 62.9% of the adherence study cohort were Black/African American, and 64.5% received SNAP for at least three months during the study period compared with half (45.9% and 50.4%, respectively) of all members with IDD. Slightly over half (52.7%) of the adherence study cohort was male, most (61.3%) lived in an urban focused area, and approximately one-third (32.7%) had a specific IDD code recorded. Certain comorbid conditions, including ADHD/anxiety, urinary retention, or ‘‘other severe mental health condition’’ were recorded for 15.5%, 2.4%, and 42.0% of the adherence study cohort, respectively (Table 1).

For both the entire study period and the adherence study measurement year, the most common type of medication prescribed to the adherence study cohort were diuretics (67.1% and 52.5%, respectively). Over the study period, 28.2% of the adherence study cohort had one medication type and 71.8% had two or more types recorded. During the measurement year, slightly over half (50.6%) had one medication type recorded (Table 2).

Approximately half (49.5%) of the adherence study cohort reached ≥80% PDC (i.e., were considered adherent to anti-hypertensive medication). Compared with those with <80% PDC, those who were adherent were older (45–54 years, 26.4% vs. 23.0%; 55–64 years: 13.2% vs. 8.6%), male (54.2% vs. 51.3%), White (34.3% vs. 28.2%), had a specific IDD code recorded (34.3% vs. 26.3%), did not receive SNAP (42.1% vs. 29.0%), and lived in community-based settings (19.4% vs. 5.9%) or nursing home or ICF/ID facility (7.6% vs. 1.9%). Most (85.0%) of those who were adherent to anti-hypertensive medication had one or more primary care visits recorded, compared with 80.5% of those who were not adherent. Over one-third (36.4%) of those who were adherent were on a waiver for at least 90 days compared with 18.9% of those who were not adherent. Those who were adherent had a higher number of medication types recorded than those who were not adherent (two types: 30.9% vs. 25.8%; three or more types: 18.4% vs. 6.3%). Rural-urban residence areas and comorbid conditions, with the exception of mental health or substance use comorbidity, which was more prevalent in those who were not adherent, were similar between the two groups (Table 3).

In multivariate analyses, Black/African American cohort members were less likely than their White counterparts to reach ≥80% PDC (aOR: 0.68, 95% CI: 0.53, 0.87), as were those who received SNAP during the measurement year (aOR: 0.74, 95% CI: 0.58, 0.94). Those with a mental health/substance use comorbid condition were also less likely to be adherent (aOR: 0.72, 95% CI: 0.55, 0.95). Adherence study cohort members who lived in community-based settings, and those who lived in nursing home or ICF/ID facilities were more likely to be adherent than those who lived in unsupervised settings (aOR: 4.20, 95% CI: 2.86, 6.19 and aOR: 5.49, 95% CI: 2.94, 10.22, respectively). Additionally, those with at least one primary care visit (aOR: 1.40, 95% CI: 1.04, 1.89), those who were on a waiver for at least 90 days (aOR: 2.68, 95% CI: 2.07, 3.47), and those taking two and three or more medication types (aOR: 1.76, 95% CI: 1.37, 2.25 and aOR: 5.02, 95% CI: 3.44, 7.33, respectively) were more likely to be adherent (Table 4).

Overall, we found that just over 40% of adults with IDD had a diagnosis of hypertension, and approximately half of the adults in our adherence study cohort were considered to be adherent to anti-hypertension medication. We identified several factors supportive of adherence to anti-hypertensive medication refills among adults with IDD, including living in a supervised residence, being on a Medicaid waiver, and having more frequent contact with primary care providers.

We found that 42% of adults with IDD had hypertension, which is higher than prevalence estimates among similarly aged cohorts in the US general population (Yoon et al., 2015). While still higher, it is similar to estimates for adults with functional cognitive limitation in the U.S. (Stevens et al., 2014). The high proportion of Black/African Americans in our analysis cohort may have contributed to this higher percentage given that hypertension is more common among Black/African Americans than other racial/ethnic groups (Chobanian et al., 2003; National Center for Health Statistics, 2016). Only half of our adherence study cohort was considered adherent to prescribed anti-hypertensive medication, which is lower than the nearly 7 in 10 U.S. adults with hypertension who report taking medication (Centers for Disease Control and Prevention, 2011), but similar to the 55% adherence noted in a study of Kansas Medicaid recipients with developmental disabilities (Vacek et al., 2013). However, the adherence rate in our study differs in a substantial way from studies that included a cohort with broadly defined disability. A recent study of Medicare Part D beneficiaries over the age of 65 years reported a 68.0% adherence rate in those whose initial entitlement reason was any type of disability (Ritchey et al., 2016). A 2009 study by Shaya et al., reported 24.8% adherence in an analysis of continuously enrolled, nonelderly Maryland Medicaid recipients; the Shaya et al., study did not look specifically at disability, but the Medicaid population is more likely to have a disability than the non-Medicaid population (Anderson, Armour, Finkelstein, & Wiener, 2010).

In our study, Black/African American cohort members were less likely to be adherent, as were those who received SNAP during the measurement year, and those with a noted mental health or substance use comorbidity. SNAP participation was not included in our study as a proxy for nutritional status. Rather, we used it as a surrogate for poverty, since only people at or below the federal poverty level are eligible for the benefit. Prior studies have identified race, substance use, and depression as risk factors for non-adherence among Medicaid (Bailey et al., 2014) and veteran (Siegel et al., 2007) populations, and higher non-adherence rates have been noted among Medicaid Part D enrollees living in poverty (Ritchey et al., 2016).

Findings from our study showed that those living in more supervised settings (nursing home, ICF/ID, or supervised community setting) had statistically significantly higher adherence to anti-hypertensive medications compared with those who lived at home or in an unsupervised setting without Medicaid services. Our study explored the association of adherence with additional support and found those with at least one primary care visit during the measurement year and those on a waiver for 90 days or more (indications of higher levels of support) were more likely to be adherent. Although few studies have looked at indicators of community support or residential type as factors related specifically to anti-hypertensive medication adherence in adults with IDD, other research has found associations between mutable characteristics and adherence, including community and clinical supports. One research group posited that the better anti-hypertensive medication adherence they found in younger ages was a result of better community-based support for these adults (Vacek et al., 2013). Another study of anti-epileptic medication use in adults with developmental disabilities found higher adherence rates among those living in group homes compared with those living in semi-independent settings and those in family homes (Hom et al., 2015). Additionally, in a study of psychotropic medication use among those with developmental disabilities and mental illness, there was a positive association between outpatient visits and adherence, which the authors felt was due to a better support system (Tan et al., 2015). In the general population modifiable factors associated with improved adherence to hypertension management include education, communication, and reducing pill burden (Klootwyk & Sanoski, 2011).

Our findings of an association between adherence and supervised living situations and community supports are also notable in that there has been a twenty-year trend toward increasing community and other support programs for people with IDD. Nationally, between 1999 and 2009, there was 17.3% increase in use of community group homes, 90.2% increase of assistance in an individual’s own home, and 28.5% increase in services in host or foster care homes for people with IDD (Smith, Lakin, Larson, & Salmi, 2011). On the national level, in fiscal year 2009, the majority of Medicaid spending on long-term supports and assistance for those with IDD was for the Home and Community-Based Services (HCBS) Waiver program (Rizzolo, Friedman, Lulinski-Norris, & Braddock, 2013). The primary service category to be funded in fiscal year 2010 was residential habilitation services, representing 53% of total proposed waiver spending (Rizzolo et al., 2013). In South Carolina in 2009, only 810 people with IDD resided in public/private institutions (a decrease of 34.0% from 1999), while 3,271 people resided in community group homes, 142 in host homes, and 662 in their own or family homes (increases of 30.5%, 21.4%, and 5.9% from 1999, respectively; Smith et al., 2011). A subset of the South Carolina adults with IDD who live in their own or family homes have a Medicaid waiver that covers services such as residential habilitation, companion/personal assistance/supported living, adult day health, community transition supports, care coordination, transportation, prevocational, supported employment, assistive and medical technologies, and health and professional services (Rizzolo et al., 2013).

Many strategies have been evaluated to improve anti-hypertensive medication adherence. Some effective strategies include encouraging patients, caregivers, and healthcare providers to work together to set therapeutic goals, design and implement a treatment plan, and adjust based on patient progress (Tan et al., 2015); consistently incorporating regular follow-up with the health care system (Ho et al., 2009); and simplifying the treatment regimen (e.g., use of fixed-dose combination medicines, use of long-acting drugs, increasing day’s supply per fill, etc.; Weber et al., 2014; Ritchey et al., 2016). The effectiveness of any of these strategies in the population of adults with IDD is unknown. However, our study suggests that interventions that incorporate additional structure and a community support component may aid in increasing anti-hypertensive medication adherence in adults with IDD. While our study results seem promising in supporting this approach, some caution is warranted as clinical outcomes that may result from such interventions in this population are unknown. An updated Cochrane review about improving adherence to prescribed medication regimens in general noted that current methods used are mostly complex (i.e., include interventions with multiple components) and not as effective for determining both adherence and clinical outcomes (Nieuwlaat et al., 2014). Further investigation of supportive systems within the IDD population should include not only adherence measures, but also outcome measures.

This study is subject to at least five limitations. First, we used administrative claims data and therefore were unable to assess measured blood pressure levels or blood pressure control. Second, PDC only assesses availability of medication and not whether or not the medication was taken as prescribed; however, PDC is a standard and accepted method of determining adherence with these types of data (Ritchey et al., 2016; Yang et al., 2017). On the other hand, we could assess that medication was purchased at a pharmacy, which is more accurate than medical records for determining if the prescription was filled. Third, we measured adherence in the year following an individual’s first anti-hypertensive medication prescription; this may or may not represent adherence in subsequent years. Fourth, results of this study pertain to the Medicaid population with IDD in South Carolina and may not necessarily be generalizable to other geographic areas or other disability types, including non-Medicaid populations. However, since adults with IDD represent a high cost group for the Medicaid program (The Pew Charitable Trusts & The McArthur Foundation, 2014), the identification of potential intervention strategies to improve health and reduce cost is important. Finally, we did not assess adherence by specific IDD type, as that was considered beyond the scope of this study; further work in this area may be warranted.

Our study identifies community and other supports as a strategic area which can be leveraged to improve medication adherence in people with IDD. Adults with IDD may live in a supervised residence, be on a Medicaid waiver (which confers eligibility for numerous services and supports), and have more frequent contact with a primary care provider. Unfortunately, an individual’s ability to receive a Medicaid waiver is usually impacted by long waiting lists, averaging nearly 4 years for those with IDD in 2014 (The Kaiser Commission on Medicaid and the Uninsured, 2015). Nonetheless, it is likely feasible for organizations that serve people with IDD to provide more training about the issues related to medication adherence to family members, caregivers, primary care providers, and adults with IDD themselves, to support improved adherence in the least restrictive residential environments.