Abstract

Gaps in knowledge and systematic tracking of the prevalence of intellectual and developmental disabilities (IDD) and characteristics that may affect the health of this disability group limits our ability to address the health disparities they experience in comparison to people without disability. The purpose of this study is to begin to fill one relevant critical gap in knowledge: understanding the demographics and health outcomes of adults with IDD who receive services under Medicare Fee-for-Service (FFS), many of who are also eligible for Medicaid. Using 2016 Medicare administrative claims, we examined the prevalence and characteristics of five diagnosis groups of IDD, in those under 65 and those 65 and over, as well as their health outcomes. We found that the IDD Medicare FFS group had high prevalence rates for chronic physical and mental health conditions, overuse of emergency departments, and high rate of 30-day readmission. These findings highlight the need for evidence-based health care coordination, improved and increased public health interventions, and continued surveillance.

Introduction

People with intellectual or developmental disabilities (IDD) experience poorer health outcomes than people without disability, in part due to more limited access to necessary health care (Havercamp & Scott, 2015; Krahn & Fox, 2014; Krahn, Hammond, & Turner, 2006). It is commonly accepted that, to fully address these disparities, there is a need to establish a public health surveillance system for tracking the incidence of IDD and identifying the characteristics that may affect the health of people with IDD (Krahn, Fox, Campbell, Ramon, & Jesien, 2010; Office of the Surgeon General, National Institute of Child Health and Human Development, and Centers for Disease Control and Prevention, 2002). However, there are multiple challenges when it comes to administering surveys with people with IDD, such as maintaining confidentiality, communication processes, obtaining accurate proxy reports, and overcoming stigmatization (Fox et al., 2015). Additionally, there are no standardized methods or questions that can be used to identify the IDD population in national surveys. Furthermore, creating a sampling frame that includes people with IDD can be difficult due to the relatively low population prevalence (Fox, Bonardi, & Krahn, 2015; Krahn et al., 2010; McDermott et al., 2018; Phillips et al., 2019; Shireman, Reichard, & Hunt, 2013). Analyzing administrative claims data (e.g., Medicaid, Medicare, private health insurance) has been offered as one integral approach in increasing understanding of the health of larger samples of people with IDD in a sustainable way that applies the principles that underlie successful surveillance systems: simplicity, brevity, and uniformity (Fox et al., 2015; Phillips et al., 2019).

Recent literature documents the success in using nationally representative surveys to study people with IDD. For example, Havercamp and Scott (2015) combined data from the Behavioral Risk Factor Surveillance System (BRFSS) with National Core Indicators (NCI) to compare people with IDD to adults with physical disability or no disability, regarding health outcomes and health care utilization. However, the sample produced by this combination of data was limited to a sample of adults receiving state developmental disabilities services and supports in one state, and, as a result, the findings have limited generalizability. Similarly, several studies have used large nationally representative surveys (e.g., Medical Expenditures Panel Survey and National Health Interview Survey) to understand people with cognitive limitations (e.g., Horner-Johnson et al, 2013; Reichard & Fox, 2013; Reichard & Stolzle, 2011; Reichard, Stolzle, & Fox, 2011). However, these studies identify people with cognitive limitations rather than IDD, based on a question that serves as a loose proxy for identifying people with IDD when used in combination with other survey questions, and the cognitive limitations question also identifies people with other conditions (e.g., dementia, stroke, etc.) that do not meet the definitions of IDD. It is impossible to identify which of those in the sample have IDD and which do not, limiting the accuracy and representativeness of the information.

Several recent studies have provided some understanding about the success of using administrative claims data and smaller nationally representative surveys to understand the prevalence, health outcomes, and health care access of subpopulations of people with IDD. McDermott et al. (2018) recently used Medicaid diagnostic codes in administrative claims data to describe demographics and other characteristics of beneficiaries with IDD from five states (DE, IA, MA, NY, and SC) who were enrolled in Medicaid for at least one calendar year from 2008 to 2013. Another recent study used the same method with all-payer claims data from 2010 to 2014 to examine all New Hampshire adults with IDD who were covered by Medicaid or private insurance. This study was the first to show that examining all-payer claims data is a feasible method for state-level surveillance of people with IDD (Phillips et al., 2019). In addition, Royer et al. (2014) linked one state's Medicaid claims data with other state data systems to understand a subpopulation with rare conditions. Others have also used diagnostic codes in Medicaid and Medicare subpopulations to conduct surveillance activities regarding people with IDD about health care utilization and expenditures (Shireman et al., 2013), emergency department and inpatient use (McDermott et al., 2015), and quality of care (Shireman et al., 2010). Using Medicaid administrative claims data for research has several known limitations (i.e., complexity, coverage restrictions, generalizability; Riley, 2009), but at the state level or combined across several states or nationally, this method can provide a great deal of information about beneficiaries with IDD supported by this type of health care financing.

Until we can establish a broader public health surveillance system for people with IDD, these recent studies demonstrate the potential value in using administrative claims data for understanding the health of a substantial number of people with IDD. Each of these studies provides us with valuable evidence of the unique strength of its methodologic approach. Additionally, viewing the findings of these studies in combination begins to establish a knowledge base about a larger proportion of the IDD population than ever before. Yet, gaps remain and there is much room for continuing to expand and improve the use of administrative claims data for research. The purpose of this study is to fill a critical gap by identifying all Medicare Fee-for-Service (FFS) adult beneficiaries with IDD, as well as IDD diagnostic subgroups, and to characterize their demographics and health outcomes. A second purpose is to draw comparisons of these variables between Medicare FFS beneficiaries with IDD and those with no disability.

Methods

Medicare and Medicaid were created in statute to provide publicly funded health care to improve access to health care for people who are unlikely to have health insurance or to be underinsured. Medicare is primarily designed to provide coverage for people 65 years or older and people with disabilities of any age who are entitled to Social Security cash benefits because they are unable to work. Medicaid provides states with funds to pay for health care for eligible individuals,who have limited financial resources and cannot afford health insurance, under the state's Medicaid plan (Klees, Eckstein, & Curtis, 2018). Some people, referred to as “dually eligible,” qualify for and receive both Medicare and Medicaid.

Data Sources

The Centers for Medicare & Medicaid Services (CMS; 2017) compile and store administrative claims data for both Medicare and Medicaid in various databases. CMS's Chronic Conditions Data Warehouse (CCW) Master Beneficiary Summary File (MBSF), the Chronic Conditions data, and the Geographic Variation Database (GVDB) beneficiary file were used for this analysis. The MBSF file was merged to all the other files using the Beneficiary Identification Number, which is the same for a single beneficiary across all CCW files. Chronic condition categories and potentially disabling conditions are based on administrative claims data, and the algorithms were developed by CMS.

For this study, calendar year 2016 data were used, specifically administrative claims for Medicare FFS beneficiaries with both Medicare Parts A and B coverage during the full year. Beneficiaries who received care through a Medicare Advantage (Medicare Part C) plan during any of the months in 2016 were excluded from analysis because the services they receive are collected by private companies and are not included in Medicare administrative claims data. There were 40,131,362 Medicare FFS beneficiaries, comprising 67.1% of the all Medicare beneficiaries. When including only those with Medicare Parts A and B benefits for the full year, the resulting final sample size was 30,987,261 beneficiaries.

Five disabling conditions related to IDD were selected and analyzed: autism spectrum disorder (ASD), cerebral palsy (CP), intellectual disability and related conditions (e.g., chromosomal anomalies, congenital syndromes such as Down and fetal alcohol syndromes, etc.), learning disabilities, and other developmental delays. These conditions were created based on a CMS algorithm of International Classification of Diseases-9th edition (ICD-9) or 10th edition (ICD-10) codes (see Centers for Medicare & Medicaid Services, 2019, for a detailed list). The comparison group with no disabilities was comprised of adults who qualify for Medicare due to age (65 and older). Those eligible for Medicare because of a disability and/or having a diagnosis of any of the following potentially disabling conditions were excluded from the comparison group: ASD, CP, traumatic brain injury, cystic fibrosis, epilepsy, fibromyalgia (chronic pain and fatigue), deafness and hearing impairment, viral hepatitis, HIV/AIDS, intellectual disability and related conditions, learning disabilities, leukemia and lymphoma, liver disease (cirrhosis and other liver conditions), serious mental illness, migraine and other chronic headache, mobility impairment, multiple sclerosis and transverse myelitis, muscular dystrophy, other developmental delays, peripheral vascular disease, spina bifida and other congenital anomalies of the nervous system, spinal cord injury, pressure ulcers and chronic ulcers, and blindness and visual impairment.

Basic descriptive data are reported, such as number (percent). Given that the population we are describing is the whole Medicare FFS population with IDD, all differences are real differences; thus, bivariate analysis was not necessary to determine statistically significant differences. Statistical Analysis System (SAS) Enterprise Guide 7 was used for data aggregation and analytics.

Results

Among the 30,987,261 adults receiving care through Medicare FFS in all of 2016, the majority (74.97%) were eligible for coverage due to age, 24.68% due to having a disability or a disability with end stage renal disease (ESRD), and 0.35% due to ESRD only. Also, among this population, 1.56% (483,595) had at least one of the IDD-related conditions for which we analyzed.

In this section, we first present results describing beneficiaries with IDD receiving care through Medicare FFS in 2016. This includes how they qualified for Medicare, prevalence by type of IDD, demographics, and health characteristics. Next, we present results for the older adult (ages 65+) Medicare FFS beneficiaries, comparing beneficiaries with IDD to those with no disability on demographics and chronic conditions. Finally, we present findings about differences in emergency department use, and health outcomes following hospital care.

Overall Medicare FFS Beneficiaries With IDD

Within those with IDD diagnoses, the largest diagnosis group was intellectual disability and related conditions (69.9%), followed by CP (22.6%), ASD (14.4%), learning disabilities (10.7%), and other developmental delays (8.5%). The majority of beneficiaries with IDD had only one IDD diagnosis (78.0%), but 18.5% had two IDD diagnoses, and 3.0% had three diagnoses.

Table 1 presents the demographics for the Medicare FFS beneficiaries by type of IDD. Overall, the greatest proportion of beneficiaries with IDD were younger than 65 years of age (77.7%), male (56.1%), non-Hispanic white (75.9%), dual eligible (qualify for both Medicare and Medicaid; 78.4%), and lived in a metropolitan area (78.7%). These characteristics varied by type of IDD. Most notably, over 90% of those with ASD were less than 65 years of age and 73.3% were male, and only 50.6% of those with learning disabilities were under age 65 and only 50.0% were male. Those with learning disabilities were least likely (52.8%), and people with intellectual disability and related conditions were most likely (85.3%), to have full dual eligibility.

Table 1

Demographics of Medicare FFS by IDD-Related Conditions, CY2016

Demographics of Medicare FFS by IDD-Related Conditions, CY2016
Demographics of Medicare FFS by IDD-Related Conditions, CY2016

Figure 1 shows the distribution of age groups for the Medicare beneficiaries with IDD compared to the general population, using data from the U.S. Census Bureau. A large percentage of Medicare FFS beneficiaries with IDD were in the 50–64 age group (33.2%) as compared to 19% of the general U.S. population. Fewer Medicare beneficiaries with IDD were in the under 30 age group (10.3%) as compared to 40.7% of the general U.S. population.

Figure 1

Distribution of Medicare FFS beneficiaries with IDD (CY2016) compared to the total U.S. population (CY2010), by age group. FFS = Fee-for-Service; IDD = intellectual and developmental disabilities. *A Medicare FFS beneficiary with IDD-related conditions has been diagnosed with autism spectrum disorder (ASD), cerebral palsy (CP), intellectual development, learning disability, or other development delay. **U.S. population age group totals (See U.S. Census Bureau, n.d.).

Figure 1

Distribution of Medicare FFS beneficiaries with IDD (CY2016) compared to the total U.S. population (CY2010), by age group. FFS = Fee-for-Service; IDD = intellectual and developmental disabilities. *A Medicare FFS beneficiary with IDD-related conditions has been diagnosed with autism spectrum disorder (ASD), cerebral palsy (CP), intellectual development, learning disability, or other development delay. **U.S. population age group totals (See U.S. Census Bureau, n.d.).

Nearly 73% of the IDD Medicare FFS beneficiaries had at least one of the chronic conditions for which we analyzed; 30.5% had three or more, and 7% had five or six chronic conditions. Prevalence for chronic conditions among beneficiaries with IDD of all ages varied by type of condition (Figure 2). Beneficiaries with IDD had high rates for all chronic conditions. Hypertension prevalence was the highest (55.8%), followed by diabetes (32.4%), chronic kidney disease (26.5%), chronic obstructive pulmonary disease (COPD; 20.6%), obesity (21.7%), chronic obstructive pulmonary disease (20.6%), and congestive heart failure (20.3%). The prevalence rate for all of these conditions were higher for the full population (all ages) of beneficiaries with IDD than those with no disability, with the exception of hypertension (results available upon request).

Figure 2

Prevalence of chronic conditions for all Medicare Fee-for-Service (FFS) beneficiaries with intellectual and developmental disabilities (IDD), calendar year (CY) 2016. A Medicare FFS beneficiary with IDD-related conditions has been diagnosed with autism spectrum disorder (ASD), cerebral palsy (CP), intellectual development, learning disability, or other development delay.

Figure 2

Prevalence of chronic conditions for all Medicare Fee-for-Service (FFS) beneficiaries with intellectual and developmental disabilities (IDD), calendar year (CY) 2016. A Medicare FFS beneficiary with IDD-related conditions has been diagnosed with autism spectrum disorder (ASD), cerebral palsy (CP), intellectual development, learning disability, or other development delay.

As shown in Figure 3, beneficiaries with IDD had very high prevalence rates for all mental health conditions. Almost 59.0% of this group had at least one of the mental health conditions we analyzed, and 17.1% had three or more mental health conditions. Prevalence for anxiety was the highest (31.5%), followed by major depressive affective disorder (28.9%), schizophrenia and other psychotic disorders (20.4%), bipolar disorder (18.4%), attention-deficit hyperactivity disorder (ADHD) and other conduct disorders (14.3%), personality disorders (5.6%), and post-traumatic stress disorder (PTSD; 2.8%).

Figure 3

Prevalence of mental health disorders for all Medicare Fee-for-Service (FFS) beneficiaries with intellectual and developmental disabilities (IDD), calendar year (CY) 2016. A Medicare FFS beneficiary with IDD-related conditions has been diagnosed with autism spectrum disorder (ASD), cerebral palsy (CP), intellectual development, learning disability, or other development delay.

Figure 3

Prevalence of mental health disorders for all Medicare Fee-for-Service (FFS) beneficiaries with intellectual and developmental disabilities (IDD), calendar year (CY) 2016. A Medicare FFS beneficiary with IDD-related conditions has been diagnosed with autism spectrum disorder (ASD), cerebral palsy (CP), intellectual development, learning disability, or other development delay.

The number and proportion of adult Medicare FFS beneficiaries with IDD-related conditions varied by state (available upon request). The size of each state's population of Medicare FFS beneficiaries with IDD-related conditions ranged from a minimum of 733 in Hawaii (0.2% of all adult Medicare FFS beneficiaries with IDD-related conditions) to the maximum of 41,488 in New York (8.6% of all adult Medicare FFS beneficiaries with IDD related conditions). A gross calculation using the State Health Facts Population Distribution (Kaiser Family Foundation, 2016) estimates that the mean proportion of the adult population comprised of Medicare FFS beneficiaries with IDD-related conditions is 0.2%, with a range of 0.7% (Hawaii) to 0.35% (Iowa, Nebraska, and North Dakota).

Younger Versus Older Adult Medicare FFS Beneficiaries With IDD

Younger (under 65) Medicare FFS beneficiaries were more likely than older beneficiaries with IDD to be male (58.2% vs. 48.8%) and to have full dual status (83.0% vs. 62.3%); however, %), but they were less likely to be White (73.8% vs. 83.1%). A larger proportion of younger Medicare FFS beneficiaries had autism (17.2 vs. 4.9) and intellectual disability and related conditions (72.6% vs 60.5%), and a smaller proportion had learning disabilities (6.9% vs. 23.6%). The prevalence rates for CP and other developmental delays were more similar between the two groups (all results available upon request).

Overall, the prevalence rates for chronic conditions were much lower among younger beneficiaries with IDD than older beneficiaries with IDD. All but one of the chronic conditions prevalence rates were approximately half of those of older adults supported by Medicare FFS. The one exception was obesity, where more Medicare FFS beneficiaries with IDD under 65 (23.0%) had the diagnosis than those 65+ with IDD (17.3%).

Most mental health conditions were more prevalent among the younger group of people with IDD than the older group with IDD. These included ADHD (16.4% vs. 6.8%), bipolar disorder (19.9 vs. 13.3%), personality disorders (6.1% vs 4.0%), and PTSD (3.4% vs. 1.0%). Only depressive disorder was lower among the younger IDD beneficiaries than the older IDD beneficiaries (27.1% vs. 35.1%). The prevalence rates for anxiety and schizophrenia/psychotic disorders were similar.

Older Adult (65+) Medicare FFS Beneficiaries With IDD Compared to Those With No Disability

Table 2 presents the demographics for the older adult Medicare FFS beneficiaries, comparing beneficiaries with IDD and related conditions to those with no disability by age group. The group with IDD included 107,699 beneficiaries who were 65 years and older and 375,896 under age 65; the no disability group included 14,752,465 beneficiaries who were 65 years and older. The IDD group was proportionally younger than the no disability group overall.

Table 2

Demographics of Medicare FFS Beneficiaries Stratified by Age, 65+ and <65, CY2016

Demographics of Medicare FFS Beneficiaries Stratified by Age, 65+ and <65, CY2016
Demographics of Medicare FFS Beneficiaries Stratified by Age, 65+ and <65, CY2016

Compared to those with no disability, older adults with IDD and related conditions were more likely to be male (48.8% vs. 44.1%) and to be full dual eligible (62.3% vs. 4.6%). Race and ethnicity composition differed across these two groups, with the IDD and related conditions group comprised of more Whites (83.1% vs. 82.5%), Blacks (10.2% vs. 6.6%), and Hispanics (4.2% vs. 0.4%) than the no disability group, but fewer Asian and Pacific Islanders (0.9% vs. 3.0%). Geographic distribution varied little across the two groups.

Among older adult Medicare FFS beneficiaries, those with IDD and related conditions had substantially higher prevalence rates for all chronic conditions than beneficiaries with no disability (Figure 4). Hypertension prevalence was very high in both groups but nearly 15 percentage points higher among beneficiaries with IDD (84.7% vs. 70.1%). Diabetes was nearly 20 percentage points higher among the IDD group than the no disability group (47.0% vs. 27.9%). Congestive heart failure among beneficiaries with IDD (41.0%) was 2.9 times higher than for those with no disability (14.3%), and the prevalence of COPD was 2.4 times higher among the IDD and related conditions group (34.9%) compared to the no disability group (14.4%). Obesity was over 7 percentage points higher among those with IDD (17.1%) than those with no disability (10.0%).

Figure 4

Chronic condition prevalence among older adult Medicare Fee-for-Service (FFS) beneficiaries, comparing intellectual and developmental disabilities (IDD) to no disability, calendar year (CY) 2016. COPD = chronic obstructive pulmonary disease. *A Medicare FFS beneficiary with IDD-related conditions has been diagnosed with autism spectrum disorder (ASD), cerebral palsy (CP), intellectual development, learning disability, or other development delay. **Beneficiaries were excluded if they were eligible for Medicare because of a disability and/or had any potentially disabling conditions.

Figure 4

Chronic condition prevalence among older adult Medicare Fee-for-Service (FFS) beneficiaries, comparing intellectual and developmental disabilities (IDD) to no disability, calendar year (CY) 2016. COPD = chronic obstructive pulmonary disease. *A Medicare FFS beneficiary with IDD-related conditions has been diagnosed with autism spectrum disorder (ASD), cerebral palsy (CP), intellectual development, learning disability, or other development delay. **Beneficiaries were excluded if they were eligible for Medicare because of a disability and/or had any potentially disabling conditions.

Emergency Department Use and Health Outcomes Following Hospital Care

Table 3 displays emergency department use and health outcomes following hospital care for beneficiaries with IDD-related conditions, by age group (under 65, 65+), and as compared to beneficiaries with no disabilities (65+). Outpatient emergency department visit rates were 2.3 times higher for those with IDD-related conditions under 65 and almost 2.7 times higher for those with IDD-related conditions 65+, compared to beneficiaries with no disabilities (38.1% vs. 43.8% vs. 16.4%, respectively). Similarly, a significantly higher proportion of those with IDD-related conditions from both age groups were high utilizers of emergency department visits, compared to beneficiaries with no disability (7.6% vs. 6.5% vs. 0.5%, respectively).

Table 3

Number and Percent of Health Outcomes for Beneficiaries' IDD-Related Conditions, by Age Group, CY2016

Number and Percent of Health Outcomes for Beneficiaries' IDD-Related Conditions, by Age Group, CY2016
Number and Percent of Health Outcomes for Beneficiaries' IDD-Related Conditions, by Age Group, CY2016

Among Medicare FFS beneficiaries who are 65 and older, mortality within 30 days of hospitalization was nearly twice as high for those with disabilities (1.7%) compared to beneficiaries with no disabilities (0.9%); however, this rate was 22.2% lower among people with IDD under 65 years old (0.7%). Additionally, more than one out of five people with IDD-related conditions under 65 and close to one out of five who were 65 or older were readmitted to the hospital within 30 days of discharge, compared to less than one out of 10 for beneficiaries with no disability (Figure 5).

Figure 5

Unplanned 30-day hospital readmission rates. IDD = intellectual and developmental disabilities. *A Medicare FFS beneficiary with IDD-related conditions has been diagnosed with autism spectrum disorder (ASD), cerebral palsy (CP), intellectual development, learning disability, or other development delay. **Beneficiaries were excluded if they were eligible for Medicare because of a disability and/or had any potentially disabling conditions.

Figure 5

Unplanned 30-day hospital readmission rates. IDD = intellectual and developmental disabilities. *A Medicare FFS beneficiary with IDD-related conditions has been diagnosed with autism spectrum disorder (ASD), cerebral palsy (CP), intellectual development, learning disability, or other development delay. **Beneficiaries were excluded if they were eligible for Medicare because of a disability and/or had any potentially disabling conditions.

Discussion

Not surprisingly, the IDD Medicare FFS group had high prevalence rates for chronic conditions. In addition, they had dramatically high prevalence rates for mental health disorders. The findings for chronic conditions from this study were substantially higher than previous findings. For example, compared to our finding of 32.4% (ages 18+), Balogh, Lake, Lin, Wilton, and Lunsky (2015) found a diabetes prevalence of 16.3% among Ontarian adults with IDD within a more circumscribed age group (30–69). An estimate using 2006 Medical Expenditure Panel Survey (MEPS) data found a diabetes prevalence of 19.4% among a national sample of individuals with cognitive limitations (ages 18–65; Reichard & Stolzle, 2011). The age range differences in these two studies accounts for some of the difference in prevalence, especially given that risk of diabetes increases with age. In addition, the Balogh et al. (2015) and Reichard and Stolzle (2011) studies also estimated the prevalence within broader populations. The prevalence rates for other chronic conditions found in this study among Medicare FFS adults with IDD-related conditions were similarly high in comparison to those found in other studies using broader national samples (Horner-Johnson, Dobbertin, Lee, & Andresen, 2013; Reichard et al., 2011). Our hypothesis affirms what is known about people who qualify for Medicare due to any type of disability, especially those who also receive services under Medicaid: They are more likely to have poorer health and highly complex health profiles compared to those who qualify for Medicare based on age, including higher prevalence of chronic health conditions (Bynum, Austin, Carmichael, & Meara, 2017; Cubanski & Neuman, 2010; Iezzoni, 2006; Joynt et al., 2017; Ruiz, Giuriceo, Caldwell, Snyder, & Putnam, 2019).

A limited number of studies regarding the prevalence of mental health conditions among adults with IDD, overall, have been published. One Australian study found a prevalence of any psychiatric illness among adults with developmental disabilities of 31.7% (Morgan, Leonard, Bourke, & Jablensky, 2008). This is much lower than our finding of 59.0% prevalence for at least one mental health disorder among the Medicare FFS population with IDD and related conditions and Iezzoni's (2006) finding of 53.4% Medicare beneficiaries with disabilities under age 65 reporting at least one mental illness. This difference in findings between the broader IDD population and those receiving care through Medicare FFS is that, similar to chronic physical conditions, Medicare eligibility criteria likely result in it covering less mentally healthy groups.

The 2016 30-day readmission rate reported for Medicare beneficiaries under 65 overall (Bailey, Weiss, Barrett, & Jiang, 2019) aligns closely with the same rate from this study for beneficiaries with IDD and related conditions in that age group (21.2% vs. 21.0%, respectively). This rate is substantially higher than for all other types of health insurance coverage, demonstrating more room for improvements in quality of and access to care for this group. Similar to the high 30-day readmission rates, the dramatically high rate of overuse of emergency departments for routine ambulatory care among the study population is also concerning and requires further investigation into the underlying causes, especially among high utilizers.

For many people in the IDD population, chronic health conditions start at a younger age than in the general population, due to accelerated aging (Dykens, 2013; Perkins & Berkman, 2012; Ruparelia, Pearn, & Mobley, 2013; Tsao, Kindelberger, Fréminville, Touraine, & Bussy, 2015), poorer access to health care (Krahn & Fox, 2014), limited receipt of preventive care (Havercamp & Scott, 2015), sedentary habits, poor nutrition, and obesity (Hinckenson, Dickinson, Water, Sands, & Penman, 2013; Melville et al., 2017; Rimmer & Rowland, 2008). Less is known about the onset and prevalence of mental health conditions to date.

Addressing high prevalence of physical and mental health conditions among beneficiaries with IDD in the Medicare FFS will require a concomitant level of public health and environmental interventions. Public health strategies may include health screenings, targeted health education, community health promotion, and evidence-based interventions for weight loss, healthy eating, and physical activity (Carmeli & Imam, 2014; Reppermund & Trollor, 2016). Further, to maximize community inclusion and minimize the time required for developing these programs, we recommend using a tool such as Guidelines, Recommendations, Adaptations Including Disability (GRAIDs; Rimmer et al., 2014) to adapt existing public health programs originally intended for the general population to include people with disabilities. Most people with IDD require supports and services across the lifespan, and most live at home with their parents as caregivers (Braddock et al., 2015). Therefore, engaging caregivers in these interventions will be critical to their success. Additionally, now that people with IDD have begun to achieve longer lifespans but also have higher risk of chronic conditions, there is an urgent need for greater research to understand the unique physical and mental health care needs of aging adults with IDD.

Similar to other populations with complex health care profiles, beneficiaries with IDD supported by Medicare FFS could also benefit from evidence-based health care coordination techniques (Bowers, Owen, & Heller, 2017; Cooper et al., 2018; Reichard, Gulley, Rasch, & Chan, 2015). Some Health Care Innovation Awards to translate some techniques, awarded under the CMS Center for Medicare and Medicaid Innovation, have tested the translation of some techniques that have been effective for other populations for application with the IDD population. Although not without challenges, these programs have demonstrated promise in improving quality outcomes (Ruiz et al., 2019). Previous studies have shown similar promise on improving health appraisals and project participants' satisfaction with outcomes (Bowers et al., 2017; Williamson et al., 2017). Progress is being made to support and implement these types of intervention as many states have begun participating in managed long-term services and support systems (Lewis, Eiken, Amos, & Saucier, 2018).

Limitations

The primary purpose of CMS administrative data is to document eligibility and code claims for payment purposes rather than research. Nevertheless, use of administrative data in research offers rich information, despite the limitations of such data for this purpose. For this research, the data are restricted to only beneficiaries eligible for Medicare FFS and, therefore, does not include the whole universe of the IDD population. The potentially disabling conditions in the administrative claims data are created using ICD-9 and ICD-10 codes reported in the claims. If a condition is undiagnosed, there is no other way to capture and identify the disabling condition of a beneficiary. As some services are not covered by Medicare or beneficiaries might use other types of insurance for services, some beneficiaries may not be captured and included in our IDD population. Additionally, the accuracy of coding (or coding errors) could be a limitation of administrative data.

Accurately identifying race and ethnicity is a critical element of public health surveillance. To improve the correct identification of Medicare beneficiaries who are Hispanic and Asian/Pacific Islanders, CMS uses an imputation algorithm developed by RTI International (Eicheldinger & Bonito, 2008). However, this algorithm prevents distinguishing the race (Black or White) of the Hispanic population. Despite its imperfections, this algorithm has helped improve the identification of race and ethnicity among Medicare beneficiaries.

The comparison group, beneficiaries with no disability, excludes all Medicare beneficiaries eligible for Medicare due to disability as determined by Social Security Administration (SSA) or beneficiaries with any potentially disabling conditions as listed in the methodology. Because the SSA definition of disability includes those eligible due to mental illness, people with mental illness are excluded from the comparison group (Centers for Medicare & Medicaid Services, 2019). Therefore, we were not able to compare prevalence rates for mental health conditions between the IDD group to the no disability group.

There is a need to establish a broad surveillance system to better understand and monitor the health of people with IDD. Various research methods have been used to identify, count, and better understand IDD subpopulations until a broad surveillance system can be established for the full IDD population. Most recently, researchers in the field have particularly acknowledged the value of administrative claims data for this purpose, and several studies have confirmed the feasibility and utility of using state Medicaid claims data toward this end (Fox et al., 2015; McDermott et al., 2018; Phillips et al., 2019). Our study adds to this area of work by using diagnosis codes from Medicare administrative claims data to identify beneficiaries with IDD who were Medicare eligible for both Parts A and B for a full year. In addition, it provides the first comprehensive look at the prevalence, demographics, and health outcomes for the population, and compares these characteristics and chronic conditions among older adults with IDD to older adults with no disability supported by Medicare FFS. This study also is the first to our knowledge to present data for smaller minority populations, including Native Americans and Alaska Natives with IDD.

Conclusion

Much remains to be learned about the population of people with IDD to understand and address the health disparities they experience. Although several surveillance approaches have been published in the literature, including those using administrative claims data, none has achieved a comprehensive understanding of the size and characteristics of the full population. This study adds to what we know from other documented studies by providing prevalence, demographics, disease prevalence, emergency room use, and health outcomes following hospital care, for beneficiaries with IDD supported by Medicare FFS in one year and comparing these statistics to those of Medicare beneficiaries with no disability. Importantly, it demonstrates that Medicare administrative claims data can be used as a surveillance tool to better understand Medicare FFS beneficiaries with IDD. Our findings of very high prevalence across the board for very debilitating diseases also strongly highlight the urgent need for prevention, chronic disease management, and research and resources to address mental health in this highly unique population. Future studies should use Medicare administrative claims data to continue this work by taking a more fine-grained look at chronic conditions and mental health disorders among the IDD subgroups, as well as examining health trends over time using longitudinal data, and make further comparisons to other Medicare populations to inform policy, practice, and fiscal planning.

Currently, the CMS Office of Minority Health (OMH) is in the process of further refining how the ICD-9 and ICD-10 codes are used to identify each of the specific disabling conditions. Once refined, larger categories of disabilities such as IDD will be created and validated, comparing CMS administrative claims data to survey data. Finally, CMS OMH will use these new groupings to examine the population's utilization of health care, such as hospitalizations, emergency department visits, hospital readmissions, and mortality.

In addition, the Administration on Community Living and CMS are collaborating to determine the specific questions to include in the Medicare Current Beneficiary Survey (MCBS) to identify Medicare FFS IDD beneficiaries. This will also allow for surveillance of a sample of Medicare FFS beneficiaries with IDD by allowing for an examination of such outcomes as access to health care, usual source of care, barriers to care, and use of preventive services.

References

References
Balogh
,
R. S.
,
Lake
,
J. K.
,
Lin
,
E.
,
Wilton
,
A.
, &
Lunsky
,
Y.
(
2015
).
Disparities in diabetes prevalence and preventable hospitalizations in people with intellectual and developmental disability: A population-based study
.
Diabetic Medicine
,
32
(
2
),
235
242
.
Bailey
,
M. K.
,
Weiss
,
A. J.
,
Barrett
,
M. L.
, &
Jiang
,
H. J.
(
2019
).
Characteristics of 30-day all-cause hospital readmissions, 2010-2016 [Statistical Brief #248]
.
Rockville, MD
:
Agency for Healthcare Research and Quality
.
Bowers
,
A.
,
Owen
,
R.
, &
Heller
,
T.
(
2017
).
Care coordination experiences of people with disabilities enrolled in Medicaid managed care
.
Disability and Rehabilitation
,
39
(
21
),
2207
2214
.
Braddock
,
D. L.
,
Hemp
,
R.
,
Rizzolo
,
M. C.
,
Tanis
,
E. S.
,
Haffer
,
L.
, &
Wu
,
J.
(
2015
).
The state of the states in intellectual and developmental disabilities: Emerging from the great recession
.
Washington, DC
:
American Association on Intellectual and Developmental Disabilities
.
Bynum
,
J.
,
Austin
,
A.
,
Carmichael
,
D.
, &
Meara
,
E.
(
2017
).
High-cost dual eligibles' service use demonstrates the need for supportive and palliative models of care
.
Health Affairs
,
36
(
7
),
1309
1317
.
Carmeli
,
E.
, &
Imam
,
B.
(
2014
).
Health promotion and disease prevention strategies in older adults with intellectual and developmental disabilities. Frontiers in Public Health, 2, 31
.
Centers for Medicare & Medicaid Services
. (
2017
).
Chronic conditions data warehouse, condition categories
.
Centers for Medicare & Medicaid Services
. (
2019
).
Chronic conditions data warehouse, data dictionaries
.
Cooper
,
S. A.
,
Hughes-McCormack
,
L.
,
Greenlaw
,
N.
,
McConnachie
,
A.
,
Allan
,
L.
,
Baltzer
,
M.
, … &
Morrison
,
J.
(
2018
).
Management and prevalence of long-term conditions in primary health care for adults with intellectual disabilities compared with the general population: A population-based cohort study
.
Journal of Applied Research in Intellectual Disabilities
,
31
,
68
81
.
Cubanski
,
J.
, &
Neuman
,
P.
(
2010
).
Medicare doesn't work as well for younger, disabled beneficiaries as it does for older enrollees
.
Health Affairs
,
29
(
9
),
1725
1733
.
Dykens
,
E. M.
(
2013
).
Aging in rare intellectual disability syndromes
.
Developmental Disabilities Research Reviews
,
18
(
1
),
75
83
.
Eicheldinger
,
C.
, &
Bonito
,
A.
(
2008
).
More accurate racial and ethnic codes for Medicare administrative data
.
Health Care Financing Review
,
29
(
3
),
27
42
.
Fox
,
M. H.
,
Bonardi
,
A.
, &
Krahn
,
G. L.
(
2015
).
Expanding public health surveillance for people with intellectual and developmental disabilities in the United States
.
International Review of Research in Developmental Disabilities
,
48
(
4
),
73
114
.
Havercamp
,
S. M.
, &
Scott
,
H. M.
(
2015
).
National health surveillance of adults with disabilities, adults with intellectual and developmental disabilities, and adults with no disabilities
.
Disability and Health Journal
,
8
(
2
),
165
172
.
Hinckson
,
E. A.
,
Dickinson
,
A.
,
Water
,
T.
,
Sands
,
M.
, &
Penman
,
L.
(
2013
).
Physical activity, dietary habits and overall health in overweight and obese children and youth with intellectual disability or autism
.
Research in Developmental Disabilities
,
34
(
4
),
1170
1178
.
Horner-Johnson
,
W.
,
Dobbertin
,
K.
,
Lee
,
J. C.
, &
Andresen
,
E. M.
(
2013
).
Disparities in chronic conditions and health status by type of disability
.
Disability and Health Journal
,
6
(
4
),
280
286
.
Iezzoni
,
L. I.
(
2006
).
Quality of care for Medicare beneficiaries with disabilities under the age of 65 years
.
Expert Review of Pharmacoeconomics & Outcomes Research
,
6
(
3
),
261
273
.
Joynt
,
K. E.
,
Figueroa
,
J. F.
,
Beaulieu
,
N.
,
Wild
,
R. C.
,
Orav
,
E. J.
, &
Jha
,
A. K.
(2017
,
March)
.
Segmenting high-cost Medicare patients into potentially actionable cohorts
.
Healthcare
,
5
(
1–2
),
62
67
.
Klees
,
B.
,
Eckstein
,
E.
, &
Curtis
,
C.
(
2018
).
Brief summaries of Medicare and Medicaid: Title XVIII and Title XIX of The Social Security Act. Retrieved from https://www.cms.gov/Research-Statistics-Data-and-Systems/Statistics-Trends-and-Reports/MedicareProgramRatesStats/SummaryMedicareMedicaid.html
Krahn
,
G. L.
, &
Fox
,
M. H.
(
2014
).
Health disparities of adults with intellectual disabilities: What do we know?
What do we do? Journal of Applied Research in Intellectual Disabilities
,
27
(
5
),
431
446
.
Krahn
,
G.
,
Fox
,
M. H.
,
Campbell
,
V. A.
,
Ramon
,
I.
, &
Jesien
,
G.
(
2010
).
Developing a health surveillance system for people with intellectual disabilities in the United States
.
Journal of Policy and Practice in Intellectual Disabilities
,
7
(
3
),
155
166
.
Krahn
,
G. L.
,
Hammond
,
L.
, &
Turner
,
A.
(
2006
).
A cascade of disparities: Health and health care access for people with intellectual disabilities
.
Mental Retardation and Developmental Disabilities Research Reviews
,
12
(
1
),
70
82
.
Lewis
,
E.
,
Eiken
,
S.
,
Amos
,
A.
, &
Saucier
,
P.
(
2018
).
The growth of managed long-term services and supports programs: 2017 update
.
Cambridge, MA
:
Truven Health Analytics
.
McDermott
,
S.
,
Royer
,
J.
,
Cope
,
T.
,
Lindgren
,
S.
,
Momany
,
E.
,
Lee
,
J. C.
, …
Armour
,
B. S.
(
2018
).
Using Medicaid data to characterize persons with intellectual and developmental disabilities in five US states
.
American Journal on Intellectual and Developmental Disabilities
,
123
(
4
),
371
381
.
Melville
,
C. A.
,
Oppewal
,
A.
,
Elinder
,
L. S.
,
Freiberger
,
E.
,
Guerra-Balic
,
M.
,
Hilgenkamp
,
T. I.
, …
Cuesta-Vargas
,
A.
(
2017
).
Definitions, measurement and prevalence of sedentary behaviour in adults with intellectual disabilities—A systematic review
.
Preventive Medicine
,
97
,
62
71
.
Morgan
,
V. A.
,
Leonard
,
H.
,
Bourke
,
J.
, &
Jablensky
,
A.
(
2008
).
Intellectual disability co-occurring with schizophrenia and other psychiatric illness: Population-based study
.
The British Journal of Psychiatry
,
193
(
5
),
364
372
.
Office of the Surgeon General, National Institute of Child Health and Human Development, and Centers for Disease Control and Prevention
. (
2002
).
Closing the gap: A national blueprint to improve the health of persons with mental retardation
.
Washington, DC
:
U.S. Department of Health and Human Services
.
Perkins
,
E. A.
, &
Berkman
,
K. A.
(
2012
).
Into the unknown: Aging with autism spectrum disorders
.
American Journal on Intellectual and Developmental Disabilities
,
117
(
6
),
478
496
.
Phillips
,
K. G.
,
Houtenville
,
A. J.
, &
Reichard
,
A.
(
2019
).
Using all-payer claims data for health surveillance of people with intellectual and developmental disabilities
.
Journal of Intellectual Disability Research
,
16
(
4
),
327
337
.
Reichard
,
A.
, &
Fox
,
M. H.
(
2013
).
Using population-based data to examine preventive services by disability type among dually eligible (Medicare/Medicaid) adults
.
Disability and Health Journal
,
6
(
2
),
75
86
.
Reichard
,
A.
,
Gulley
,
S. P.
,
Rasch
,
E. K.
, &
Chan
,
L.
(
2015
).
Diagnosis isn't enough: Understanding the connections between high health care utilization, chronic conditions and disabilities among US working age adults
.
Disability and Health Journal
,
8
(
4
),
535
546
.
Reichard
,
A.
, &
Stolzle
,
H.
(
2011
).
Diabetes among adults with cognitive limitations compared to individuals with no cognitive disabilities
.
Intellectual and Developmental Disabilities
,
49
(
3
),
141
154
.
Reichard
,
A.
,
Stolzle
,
H.
, &
Fox
,
M. H.
(
2011
).
Health disparities among adults with physical disabilities or cognitive limitations compared to individuals with no disabilities in the United States
.
Disability and Health Journal
,
4
(
2
),
59
67
.
Reppermund
,
S.
, &
Trollor
,
J. N.
(
2016
).
Successful ageing for people with an intellectual disability
.
Current Opinion in Psychiatry
,
29
(
2
),
149
154
.
Riley
,
G. F.
(
2009
).
Administrative and claims records as sources of health care cost data
.
Medical Care
,
S51
S55
.
Rimmer
,
J. A.
, &
Rowland
,
J. L.
(
2008
).
Physical activity for youth with disabilities: A critical need in an underserved population
.
Developmental Neurorehabilitation
,
11
(
2
),
141
148
.
Rimmer
,
J. H.
,
Vanderbom
,
K. A.
,
Bandini
,
L. G.
,
Drum
,
C. E.
,
Luken
,
K.
,
Suarez-Balcazar
,
Y.
, &
Graham
,
I. D.
(
2014
).
GRAIDs: A framework for closing the gap in the availability of health promotion programs and interventions for people with disabilities. Implementation Science, 9(1), 100
.
Royer
,
J. A.
,
Hardin
,
J. W.
,
McDermott
,
S.
,
Ouyang
,
L.
,
Mann
,
J. R.
,
Ozturk
,
O. D.
, &
Bolen
,
J.
(
2014
).
Use of state administrative data sources to study adolescents and young adults with rare conditions
.
Journal of General Internal Medicine
,
29
(
3
),
732
738
.
Ruiz
,
S.
,
Giuriceo
,
K.
,
Caldwell
,
J.
,
Snyder
,
L. P.
, &
Putnam
,
M.
(
2019
).
Care coordination models improve quality of care for adults aging with intellectual and developmental disabilities
.
Journal of Disability Policy Studies
.
Ruparelia
,
A.
,
Pearn
,
M. L.
, &
Mobley
,
W. C.
(
2013
).
Aging and intellectual disability: Insights from mouse models of Down syndrome
.
Developmental Disabilities Research Reviews
,
18
(
1
),
43
50
.
Shireman
,
T. I.
,
Reichard
,
A.
, &
Hunt
,
S. L.
(
2013
).
Learning your ABDs: Variation in health care utilization across Kansas Medicaid disability groups
.
Disability and Health Journal
,
6
(
3
),
220
226
.
Shireman
,
T. I.
,
Reichard
,
A.
,
Nazir
,
N.
,
Backes
,
J. M.
, &
Greiner
,
K. A.
(
2010
).
Quality of diabetes care for adults with developmental disabilities
.
Disability and Health Journal
,
3
(
3
),
179
185
.
Tsao
,
R.
,
Kindelberger
,
C.
,
Fréminville
,
B.
,
Touraine
,
R.
, &
Bussy
,
G.
(
2015
).
Variability of the aging process in dementia-free adults with Down syndrome
.
American Journal on Intellectual and Developmental Disabilities
,
120
(
1
),
3
15
.
U.S. Census Bureau
. (
n.d.
).
American fact finder: 2010 Census
.
Williamson
,
H. J.
,
Perkins
,
E. A.
,
Levin
,
B. L.
,
Baldwin
,
J. A.
,
Lulinski
,
A.
,
Armstrong
,
M. I.
, &
Massey
,
O. T.
(
2017
).
Implementation of Medicaid managed long-term services and supports for adults with intellectual and/or developmental disabilities in Kansas
.
Intellectual and Developmental Disabilities
,
55
(
2
),
84
96
.