Abstract
Objective
To estimate the excess burden of out-of-pocket healthcare spending associated with Alzheimer’s disease and related disorders (ADRD) among older individuals (age ≥ 65 years).
Methods
We adopted a retrospective, cross-sectional study design with data from 2012 Medicare Current Beneficiary Survey. The study sample comprised of elderly community dwelling individuals who had positive total healthcare expenditures, and enrolled in Medicare throughout the calendar year (462 with ADRD, and 7,160 without ADRD). We estimated the per-capita total annual out-of-pocket spending on healthcare and out-of-pocket spending by service type: inpatient, outpatient, home health, prescription drugs, and other services. We measured out-of-pocket spending burden by calculating the percentage of income spent on healthcare and defined high out-of-pocket spending burden as having this percentage above 10%. Multivariable analyses included ordinary least squares regressions and logistic regressions and these analyses adjusted for predisposing, enabling, need, personal healthcare practices and external environment characteristics.
Results
The average annual per-capita out-of-pocket healthcare spending was greater among individuals with ADRD compared to those without ADRD ($3,285 vs. $1,895); home health and prescription drugs accounted for 52% of total out-of-pocket spending among individuals with ADRD and 34% among individuals without ADRD. Elderly individuals with ADRD were more likely to have high out-of-pocket spending burden (AOR = 1.49; 95% CI = 1.13, 1.97) compared to those without ADRD.
Conclusion
ADRD is associated with excess out-of-pocket healthcare spending, primarily driven by prescription drugs and home healthcare use.
Introduction
Alzheimer’s disease and related disorders (ADRD) affect 8.24% of individuals in the United States (US)1. ADRD are associated with neurocognitive impairments due to its multiple etiologies, including Alzheimer’s disease, Lewy body disease, vascular disease, traumatic brain injury, HIV infection, prion disease, Parkinson’s disease, Huntington’s disease, and certain medications. Among ADRD, 80% of dementias are attributed to Alzheimer’s disease2. Individuals with ADRD may have worsening neurocognitive impairments as the disease progresses, often requiring increasing levels of medical and non-medical care, including full-time residential services. Most elderly with Alzheimer’s disease are covered by Medicare because Medicare provides coverage for nearly all of the elderly in the US3. However, not all healthcare expenditures are covered by Medicare. The patients and families bear some direct medical care costs in terms of deductibles, coinsurance, and copayments for medical care and prescription drugs, insurance premiums for supplemental coverage, amount paid for non-prescription medications; transportation to health care providers; and uncovered structural or lifestyle modifications. For example, among Medicare beneficiaries, health insurance premiums account for 42% of the total out-of-pocket spending, with payments towards cost-sharing and non-covered services and goods accounting for the remaining 58% 3. In fact, the total payments for health care, long-term care and hospice care are estimated to be $236 billion for people with ADRD in 2016, with just under half of the costs covered by Medicare 4. Therefore, high out-of-pocket spending for ADRD care may place a significant financial burden on families draining resources for the household as a whole 5.
It is important to assess the magnitude of out-of-pocket expenditures because high expenditures can lead to worse health outcomes. For example, individuals with high out-of-pocket spending may stop taking their medications 6 and may not use preventive care or outpatient services for their healthcare in order to save money 7, 8. This may exacerbate the ADRD symptoms and lead to higher rates of morbidity and mortality 9. Furthermore, with disease progression, as individuals with ADRD face worsening medical complications and declining functional status, their mix of services required may vary. For example, it has been reported that nearly one-third (34.5%) used home healthcare and the incremental total costs for paid home care in 2010 was $5,678, accounting for 20% of the incremental direct costs associated with ADRD10. These services may require co-payments leading to high out-of-pocket expenses. Therefore, it is also important to examine out-of-pocket spending for different types of care such as home health, inpatient, outpatient, medical provider, and prescription drugs. This may also reflect the trade-off decisions patients need to make on the type of services when faced with limited financial resources.
However, only a few studies have evaluating the association between ADRD and out-of-pocket healthcare spending 10-13 and these studies have limitations. Kelley et al. reported that the average out-of-pocket spending during the last five-years of life for patients with dementia was 81% higher compared to patients without dementia11. This study used Health Retirement Survey (HRS) and included only fee-for-service Medicare beneficiaries (≥ 70 years) who died between 2005 and 2010. Hurd et al. used the same dataset and found that the average annual out-of-pocket spending can be as high as $6,194 among elderly with ADRD10 However, this study used an estimated probability of dementia rather than the observed status (yes/no). Delavande et al. compared individuals with normal cognition, dementia with cognitive impairment and dementia without cognitive impaired and reported that those with dementia and cognitive impairment had 356% higher annual out-of-pocket expenditures compared to those with normal cognition 12. This study did not analyze out-of-pocket spending burden as a percentage of income spent. Another study used the Medicare Current Beneficiary Survey (MCBS) and analyzed gender differences in life-time out-of-pocket spending 13. However, that study included out-of-pocket spending only for assisted living facilities and home healthcare.
Therefore, the main objective of the current study is to estimate the excess burden of annual total direct out-of-pocket spending and out-of-pocket spending on different types of healthcare services among all community-dwelling elderly Medicare beneficiaries with ADRD by comparing them to those without ADRD.
Conceptual framework
The explanatory variables for this study were selected using the Andersen’s Expanded Behavioral Model 14. Under this model, out-of-pocket expenditures of an individual is influenced by predisposition factors (e.g., age, sex, and race), enabling factors (e.g., marital status, education, and poverty status), need factors (e.g., chronic conditions, health status) and personal health practices (e.g., smoking, obesity, physical activity).
METHODS
Study Design
We adopted a retrospective cross-sectional design.
Data source
The data source is the Medicare Current Beneficiary Survey (MCBS) for the year 2012. The MCBS is nationally representative sample of Medicare beneficiaries - the aged, disabled and institutionalized. The survey began in 1992 and is released every year. The survey directly collects data from the respondents and includes self-reported health status, height and weight, activities of daily living, functional status, living arrangement, history of medical conditions, out-of-pocket expenditures, non-Medicare utilization, expenditures, and other health-related information. Data collected from the beneficiaries are merged with Medicare claims except for Part D through an extensive and rigorous reconciliation process. The survey is designed with a multistage, stratified, random sampling of Medicare beneficiaries4. West Virginia University Institutional Review Board reviewed this project and granted exemption status to the study as all the data were de-identified.
Study Sample
In this study, we restricted our sample to older adults (≥ 65 years), who lived in the community, who were alive, and enrolled in Medicare throughout the entire year. We excluded individuals who did not answer the relevant health questionnaires considered in our study (n =158) and those who had zero total healthcare expenditures (n=107). The final study sample included 7,622 Medicare beneficiaries with (N = 462) or without ADRD (N = 7160).
Measures
Dependent Variables: Out-of-pocket healthcare spending
Out-of-pocket healthcare spending consisted of Medicare cost sharing and non-covered services, but not insurance premiums. We examined out-of-pocket spending using several measures: absolute out-of-pocket expenditures; the log-transformed out-of-pocket spending; and out-of-pocket spending burden. We used log-transformed out-of-pocket expenditures, to reduce skewness 15, 16 Out-of-pocket spending burden was based on percent income spent out-of-pocket for healthcare. We defined an indicator of positive out-of-pocket spending, and an indicator of high out-of-pocket spending burden indicating that the percentage was above 10% of income based on prior studies 17, 18. There were seven components of out-of-pocket spending which were considered in the analyses: home health, facility charges, hospice, inpatient, outpatient, medical provider, prescription drugs, and dental care. The out-of-pocket spending was measured over a one year period in 2012.
Key Independent Variable: Alzheimer’s’ Disease and Related Dementias (ADRD)
The key explanatory variable in our study was the presence or absence of ADRD. ADRD was ascertained using self-reports or Medicare claims. Self-reported ADRD was based on giving an affirmative response to either of the following two questions: “Has a doctor (ever) told [you/(SP)] that (you/he/she) had Alzheimer’s disease?” and “Has a doctor (ever) told [(you/(SP)] that (you/he/she) had any type of dementia other than Alzheimer’s disease?”. We used the International Classification of Diseases, 9th edition Clinical Modification (ICD-9-CM) codes to derive ADRD from Medicare claims. The ICD-9-CM codes (including 3310, 33111, 33119, 3312, 3317, 2900, 29010, 29011, 29012, 29013, 29020, 29021, 2903, 29040, 29041, 29042, 29043, 2940, 29410, 29411, 29420, 29421, 2948, and 797). These codes were based on the Centers for Medicaid and Medicare Services (CMS) chronic conditions warehouse algorithm 19.
Other Explanatory Variables
Using the Andersen Model health care utilization model, we identified predisposing characteristics consisting of sex (male/female), age (65–69 years, 70–74 years, 75–79 years, and 80 years and older), race/ethnicity (White, African American, Latino, other), enabling factors comprising marital status (married, widowed, divorced/separated, or never married), education (less than high school, high school, or above high school, college), income relative to the federal poverty line (FPL) (less than 200% of FPL or at least 200% of FPL), supplementary health insurance Medicaid (yes/no), private insurance (yes/no), and prescription drug coverage (yes/no), and need factors (number of chronic conditions [considered from the following list: arthritis, cancer, diabetes, heart disease, hypertension, respiratory disease, osteoporosis, mental illness] (none, one, two to three, four or more), individual perceived health status (excellent, very good, good, fair, poor), functional limitations measured by activities of daily living (none, one to two, three or more). We also adjusted for personal health practice factors, including body mass index (BMI) (underweight, normal weight, overweight, or obese), and smoking status (never-smoker, former smoker, or current smoker). The BMI categories were based on the CDC definition: underweight (<18.5), normal (18.5 to <25), overweight (25.0 to <30), and obese (30.0 or higher). We also accounted for external environment such as socioeconomic status which included region of residence (Northeast, Midwest, South, and West), urban/rural status (metropolitan area/non-metropolitan area) and income relative to the federal poverty level (less than 200% of federal poverty level or at least 200% of federal poverty level). It has to be noted that all our independent variables were measured as categorical variables.
Statistical analyses
We tested statistically significant differences between ADRD and no ADRD groups with chi-square statistics. We used ordinary least squares for out-of-pocket spending in the whole sample and among those with positive out-of-pocket spending.
We conducted logistic regressions to examine the relationship between ADRD and having out-of-pocket spending burden above 10%. All our multivariable models adjusted for expanding number of covariates.
RESULTS
The sample consisted of 57% female, and 79% white. The age groups were equally distributed with 24% between 65 and 69 years, 27% were between 70 and 74 years, 20% between 75 and 79 years and 30% at least 80 years of age (Table 1).
Table 1.
Characteristics of Elderly Medicare Beneficiaries by ADRD Status Medicare Current Beneficiary Survey, 2012
| Total | ADRD | No ADRD | Sig | ||||
|---|---|---|---|---|---|---|---|
| N | Wt % | N | Wt % | N | Wt % | ||
| ALL | 7,622 | 100.0 | 462 | 100.0 | 7,160 | 100.0 | |
| Sex | *** | ||||||
| Women | 4,290 | 56.5 | 303 | 64.6 | 3987 | 56.0 | |
| Men | 3,332 | 43.5 | 159 | 35.4 | 3173 | 44.0 | |
| Race/Ethnicity | *** | ||||||
| White | 5,956 | 78.5 | 320 | 70.7 | 5,636 | 78.9 | |
| African American | 624 | 7.8 | 67 | 13.2 | 557 | 7.4 | |
| Latino | 699 | 9.1 | 58 | 12.1 | 641 | 8.9 | |
| Other race | 320 | 4.7 | 17 | 4.1 | 303 | 4.7 | |
| Age in Years | *** | ||||||
| 65–69 | 1,404 | 23.6 | 19 | 6.5 | 1,385 | 24.5 | |
| 70–74 | 1,779 | 27.3 | 52 | 14.7 | 1,727 | 28.0 | |
| 75–79 | 1,558 | 20.1 | 73 | 17.6 | 1,485 | 20.2 | |
| 80 + | 2,881 | 29.1 | 318 | 61.2 | 2,563 | 27.3 | |
| Marital Status | *** | ||||||
| Married | 4,028 | 54.6 | 195 | 44.4 | 3,833 | 55.2 | |
| Widowed | 2,413 | 28.4 | 211 | 43.2 | 2,202 | 27.6 | |
| Divorced/Separated | 919 | 13.4 | 41 | 9.3 | 878 | 13.7 | |
| Not Married | 260 | 3.5 | 15 | 3.1 | 245 | 3.6 | |
| Metro Status | |||||||
| Metro | 5,661 | 77.5 | 341 | 76.5 | 5,320 | 77.6 | |
| Not Metro | 1,961 | 22.5 | 121 | 23.5 | 1,840 | 22.4 | |
| Education | *** | ||||||
| Less than HS | 1,772 | 21.2 | 173 | 35.1 | 1,599 | 20.4 | |
| High School | 2,585 | 33.3 | 138 | 29.9 | 2,447 | 33.4 | |
| Above High School | 1,187 | 16.3 | 58 | 13.1 | 1,129 | 16.5 | |
| College | 2,054 | 29.2 | 90 | 21.8 | 1,964 | 29.7 | |
| Poverty Status | *** | ||||||
| Less than 200% | 3,721 | 46.1 | 306 | 65.2 | 3,415 | 45.1 | |
| At least 200% | 3,901 | 53.9 | 156 | 34.8 | 3,745 | 54.9 | |
| Medicaid | *** | ||||||
| Yes | 1,088 | 13.8 | 124 | 26.4 | 964 | 13.1 | |
| No | 6,534 | 86.2 | 338 | 73.6 | 6,196 | 86.9 | |
| Private Health Insurance | |||||||
| Yes | 4,257 | 56.2 | 269 | 58.0 | 3,988 | 56.1 | |
| No | 3,365 | 43.8 | 193 | 42.0 | 3,172 | 43.9 | |
| Prescription Drug Coverage | |||||||
| Yes | 6,199 | 81.8 | 386 | 84.0 | 5,813 | 81.7 | |
| No | 1,423 | 18.2 | 76 | 16.0 | 1,347 | 18.3 | |
| # of Chronic Conditions | *** | ||||||
| None | 335 | 5.0 | 12 | 2.7 | 323 | 5.2 | |
| One | 1,002 | 13.9 | 49 | 10.5 | 953 | 14.1 | |
| 2–3 | 3,775 | 49.8 | 222 | 47.4 | 3553 | 49.9 | |
| 4 or more | 2,510 | 31.3 | 179 | 39.4 | 2331 | 30.8 | |
| Health Status | *** | ||||||
| Excellent | 1,402 | 19.5 | 59 | 12.1 | 1,343 | 19.9 | |
| Very good | 2,532 | 33.6 | 111 | 24.4 | 2,421 | 34.1 | |
| Good | 2,270 | 29.3 | 135 | 30.5 | 2,135 | 29.2 | |
| Fair | 1,079 | 13.5 | 117 | 24.8 | 962 | 12.9 | |
| Poor | 310 | 4.1 | 38 | 8.2 | 272 | 3.9 | |
| Functional Limitations | *** | ||||||
| None | 4,970 | 67.3 | 168 | 37.7 | 4,802 | 68.9 | |
| 1 or 2 | 1,839 | 23.0 | 124 | 26.7 | 1,715 | 22.8 | |
| 3 or more | 808 | 9.8 | 170 | 35.6 | 638 | 8.3 | |
| Body Mass Index | *** | ||||||
| Underweight | 177 | 2.2 | 21 | 4.2 | 156 | 2.1 | |
| Normal | 2,597 | 33.3 | 185 | 38.8 | 2,412 | 33.0 | |
| Overweight | 2,824 | 37.7 | 168 | 38.2 | 2,656 | 37.7 | |
| Obese | 1,957 | 26.8 | 78 | 18.8 | 1,879 | 27.2 | |
| Smoking Status | * | ||||||
| Current Smoker | 644 | 9.4 | 23 | 5.5 | 621 | 9.6 | |
| Past Smoker | 3,779 | 49.5 | 215 | 47.9 | 3,564 | 49.6 | |
| Never smoked | 3,195 | 41.1 | 223 | 46.6 | 2,972 | 40.8 | |
Note: Based on 7,622 Medicare beneficiaries age 65 and older, alive during the calendar and had positive total healthcare expenditures. Significant group differences in characteristics by ADRD status was based on chi-square tests.
ADRD: Alzheimer’s disease and related dementias ; HS: High School; Sig: Significance; Wt: Weighted
p < .001;
.001 ≤ p < .01;
.01 ≤ p < .05.
There were significant group differences involving ADRD status and sex, race, age, marital status, poverty status, being on Medicaid, number of chronic conditions, perceived health status, functional limitations, BMI, and smoking status. A lower percentage of ADRD individuals had college education (21.8% versus 29.7%) and a higher percentage of individuals with ADRD were poor defined as less than 200% of the FPL (65% vs 45%), and were on Medicaid (26% vs 13%) compared to those without ADRD. A higher percentage of individuals with ADRD had 3 or more ADL (35.6% versus 8.9%) compared to those without ADRD. A higher percentage of those with ADRD had 4 or more chronic conditions compared to individuals without ADRD (39% vs 31%).
Unadjusted Differences in Out-of-Pocket Expenditures by ADRD Status
The average out-of-pocket spending by type of services and ADRD status are presented in Table 2. Elderly with ADRD had significantly higher out-of-pocket spending across all measures except dental care. The total out-of-pocket spending in ADRD group was $3,284.6 whereas the total out-of-pocket spending in no-ADRD group was $1895.0. Among those who had positive out-of-pocket spending (i.e. expenditures that the insurance did not cover), the results were similar (total out-of-pocket spending was $3,319.40 in ADRD group vs. $1,907.20 in no-ADRD group), except the difference in outpatient spending became insignificant. Results for out-of-pocket spending burden is similar however differences in outpatient, medical provider and dental spending failed to reach significance.
Table 2.
Out-of-Pocket Spending among Elderly Medicare Beneficiaries by Alzheimer’s Disease and Related Dementias Medicare Current Beneficiary Survey, 2012
| ADRD | No ADRD | ||||||
|---|---|---|---|---|---|---|---|
| ALL | N | Wt. Mean |
SE | N | Wt. Mean |
SE | Sig |
| Total | 462 | 3,284.6 | 376.0 | 7,160 | 1,895.0 | 47.3 | *** |
| Inpatient | 462 | 111.4 | 35.6 | 7,160 | 42.9 | 6.4 | * |
| Outpatient | 462 | 161.2 | 65.6 | 7,160 | 120.9 | 12.5 | *** |
| Medical Provider | 462 | 981.2 | 136.6 | 7,160 | 667.1 | 20.5 | * |
| Prescription Drugs | 462 | 847.5 | 61.8 | 7,160 | 597.9 | 14.0 | *** |
| Home Health | 462 | 874.3 | 276.9 | 7,160 | 52.7 | 26.1 | ** |
| Dental | 462 | 242.2 | 62.0 | 7,160 | 404.9 | 15.8 | ** |
| Other | 462 | 66.9 | 21.5 | 7,160 | 8.8 | 1.7 | ** |
|
Among those with Out-of-Pocket Spending > 0 | |||||||
| Total | 457 | 3,319.4 | 379.6 | 7,114 | 1,907.2 | 47.8 | *** |
| Inpatient | 42 | 1,228.1 | 356.9 | 279 | 1,109.2 | 151.9 | * |
| Outpatient | 145 | 508.1 | 204.2 | 2,316 | 381.0 | 36.8 | |
| Medical Provider | 410 | 1,101.5 | 151.2 | 6,421 | 744.9 | 22.4 | * |
| Prescription Drugs | 451 | 867.7 | 64.0 | 6,855 | 627.4 | 14.6 | *** |
| Home Health | 26 | 14,523.7 | 3,492.5 | 80 | 5,060.4 | 2,431.8 | *** |
| Dental | 152 | 726.3 | 183.3 | 3,516 | 798.4 | 28.0 | |
| Other | 20 | 1,496.1 | 340.5 | 79 | 908.4 | 160.6 | *** |
Note: Based on 7,622 Medicare beneficiaries age 65 and older, alive during the calendar and had positive total healthcare expenditures. Significant group differences average out-of-pocket spending by ADRD status were based on t-tests. Among those with out-of-pocket spending greater than zero, the total spending is not sum of individual domains.
ADRD: Alzheimer’s disease and related disorders; SE: Standard error; Sig: Significance; Wt: Weighted
p < .001;
.001 ≤ p < .01;
.01 ≤ p < .05
Adjusted Differences in Out-of-Pocket Expenditures by ADRD Status
Based on multivariable OLS regression on out-of-pocket spending (Table 3), we observed that Individuals with ADRD spent $1,101 higher in total, $274 higher in prescription drugs (p < .01), $622 higher in home health (p < 0.05) than individuals without ADRD. Among those who had positive out-of-pocket spending, those with ADRD had substantially higher home health out-of-pocket spending ($5,570) compared to individuals without ADRD. Individuals with ADRD also incurred higher out-of-pocket spending on prescription drugs ($274, p-value<0.001) and total out-of-pocket spending ($1,126, p <0.01). Individuals with ADRD spent less out-of-pocket on inpatient ($401, p-value<0.001) than individuals without ADRD.
Table 3.
Regression Coefficients and Standard Errors of ADRD Status from Multivariable Linear Models on Out-of-Pocket Spending Medicare Current Beneficiary Survey, 2012
| Ordinary Least Squares Regression (Whole Sample)
| ||||
|---|---|---|---|---|
| Estimate | SE | Prob. | Sig | |
|
|
||||
| Total | $1,101.1 | $378.7 | 0.004 | ** |
| Inpatient | $42.2 | $43.3 | 0.330 | |
| Outpatient | $39.7 | $67.1 | 0.554 | |
| Medical Provider | $144.1 | $137.2 | 0.294 | |
| Prescription Drugs | $274.4 | $65.3 | <0.0001 | *** |
| Home Health | $622.2 | $285.2 | 0.030 | * |
| Dental | -$68.1 | $70.5 | 0.335 | |
| Other | $46.5 | $21.2 | 0.029 | * |
|
Ordinary Least Squares Regression (Out-of-pocket spending >0) | ||||
| Estimate | SE | Prob. | Sig | |
|
|
||||
| Total | $1,125.5 | $381.4 | 0.003 | ** |
| Inpatient | -$401.1 | $113.2 | <0.0001 | *** |
| Outpatient | $93.1 | $210.5 | 0.658 | |
| Medical Provider | $165.1 | $153.8 | 0.284 | |
| Prescription Drugs | $274.1 | $67.1 | <0.001 | *** |
| Home Health | $5,570.2 | $836.5 | <0.001 | *** |
| Dental | -$11.0 | $183.5 | 0.952 | |
| Other | $231.6 | $0.0 | <0.0001 | *** |
|
Ordinary Least Squares Regression Log-transformed Out-of-Pocket Spending | ||||
| Estimate | SE | Prob. | Sig | |
|
|
||||
| Total | 0.319 | 0.070 | <0.0001 | *** |
| Inpatient | 0.273 | 0.090 | 0.002 | ** |
| Outpatient | 0.005 | 0.135 | 0.969 | |
| Medical Provider | 0.176 | 0.110 | 0.111 | |
| Prescription Drugs | 0.498 | 0.072 | <0.001 | *** |
| Home Health | 0.304 | 0.100 | 0.002 | ** |
| Dental | −0.428 | 0.132 | 0.001 | ** |
| Other | 0.195 | 0.060 | 0.001 | ** |
Note: Based on 7,622 Medicare beneficiaries age 65 and older, alive during the calendar and had positive total healthcare expenditures. The regression models controlled for sex, age, race/ethnicity, marital status, education, poverty status, Medicaid, private insurance, prescription drug coverage, number of chronic conditions, perceived physical health, functional status, body mass index, and current smoking.
ADRD: Alzheimer’s disease and related dementias; Prob: Probability; SE: Standard error; Sig: Significance; Wt: Weighted.
p < .001;
.001 ≤ p < .01;
.01 ≤ p < .05
In ordinary least squares estimation for log-transformed out-of-pocket spending analyses, individuals with ADRD had higher home health (β = .304 p <.01), prescription drugs, (β = .498 p <.001), inpatient (β = .273 p <.01) and overall out-of-pocket spending (β = .319 p < .001) compared to those without ADRD.
Out-of-Pocket Spending Burden by ADRD Status
We also examined the out-of-pocket spending burden by ADRD status using percent income spent on healthcare services (Table 4). Individuals with ADRD spent a significantly higher percentage of their income on medical services among those without ADRD (at 12% vs 7%). They also spent a significantly higher percentage on home health, prescription drugs, and medical provider visits. Similar results were obtained using the ordinary least squares regression (e.g., higher percentage of their income spent on home health, prescription drugs and overall).
Table 4.
Out-of-pocket Spending Burden (Percent Income Spent Out-of-pocket) among Elderly Medicare Beneficiaries Medicare Current Beneficiary Survey, 2012
| ADRD | No ADRD | ||||||
|---|---|---|---|---|---|---|---|
| N | Wt. Mean | SE | N | Wt. Mean | SE | Sig | |
| Total | 462 | 12.44 | 1.13 | 7,160 | 6.85 | 0.15 | *** |
| Inpatient | 462 | 0.68 | 0.25 | 7,160 | 0.19 | 0.04 | |
| Outpatient | 462 | 0.67 | 0.21 | 7,160 | 0.53 | 0.05 | |
| Medical Provider | 462 | 4.33 | 0.56 | 7,160 | 2.79 | 0.10 | ** |
| Prescription Drugs | 462 | 4.28 | 0.44 | 7,160 | 2.57 | 0.10 | *** |
| Home Health | 462 | 2.23 | 0.70 | 7,160 | 0.07 | 0.02 | ** |
| Dental | 462 | 1.11 | 0.28 | 7,160 | 1.35 | 0.08 | |
| Other | 462 | 0.30 | 0.10 | 7,160 | 0.04 | 0.00 | * |
|
Regression Co-efficient and Standard Errors of ADRD status from Ordinary Least Squares Regression on Out-of-pocket Spending Burden (Percent Income Spent Out-of-pocket) | |||||||
| Beta | SE | Prob | Sig | ||||
|
| |||||||
| Total | 3.376 | 1.122 | 0.003 | ** | |||
| Inpatient | 0.370 | 0.247 | 0.135 | ||||
| Outpatient | 0.048 | 0.229 | 0.834 | ||||
| Medical Provider | 0.136 | 0.556 | 0.806 | ||||
| Prescription Drugs | 1.194 | 0.446 | 0.008 | ** | |||
| Home Health | 1.854 | 0.693 | 0.008 | ** | |||
| Dental | −0.160 | 0.328 | 0.625 | ||||
| Other | 0.182 | 0.096 | 0.058 | ||||
Note: Based on 7,622 Medicare beneficiaries age 65 and older, alive during the calendar and had positive total healthcare expenditures. The regression models controlled for sex, age, race/ethnicity, marital status, education, poverty status, Medicaid, private insurance, prescription drug coverage, number of chronic conditions, perceived physical health, functional status, body mass index, and current smoking.
ADRD: Alzheimer’s disease and related dementias; Prob: Probability; SE: Standard error; Sig: Significance; Wt: Weighted
p < .001;
.001 ≤ p < .01;
.01 ≤ p < .05
We examined the number and percentages of individuals with out-of-pocket spending burden above 10% (high out-of-pocket spending burden) (Table 5). Individuals with ADRD were more likely to have spent greater than 10% of their income on healthcare (30% vs. 17%; P<.0001). We expanded the list of covariates adjusted in the model and found that although the adjusted odds ratio were attenuated, a significant association of ADRD and higher out-of-pocket remained in the model (Adjusted odds ratio: 1.49; 95% confidence interval: 1.13, 1.97).
Table 5.
High Out-of-pocket Spending Burden among Elderly Medicare Beneficiaries Medicare Current Beneficiary Survey, 2012
| High OOP Spending Burden | Low OOP Spending Burden | Sig | |||
|---|---|---|---|---|---|
| N | Wt % | N | Wt % | ||
| ALL | 1,409 | 17.6 | 6,213 | 82.4 | |
| ADRD | *** | ||||
| Yes | 139 | 29.9 | 323 | 70.1 | |
| No | 1,270 | 16.9 | 5890 | 83.1 | |
|
Adjusted Odds Ratios (AOR) and 95% Confidence Intervals (CI) of ADRD Status from Logistic Regressions on High Out-of-pocket Spending Burden | |||||
| AOR | 95% CI | Sig | |||
|
Model 1 – Unadjusted
| |||||
| ADRD | |||||
| Yes | 2.12 | [ 1.67, 2.68] | *** | ||
| No | |||||
|
Model 2 ADRD + Sex + Race + Age | |||||
| ADRD | |||||
| Yes | 1.78 | [ 1.39, 2.28] | *** | ||
| No | |||||
|
Model 3 ADRD + Sex + Race + Age + marital status + education + health insurance | |||||
| ADRD | |||||
| Yes | 1.73 | [ 1.34, 2.23] | *** | ||
| No | |||||
|
Model 4 ADRD + Sex + Race + Age + marital status + education + health insurance + life style | |||||
| ADRD | |||||
| Yes | 1.59 | [ 1.21, 2.09] | *** | ||
| No | |||||
|
Model 5 ADRD + Sex + Race + Age + marital status + education + health insurance + life style + health + functional status | |||||
| ADRD | |||||
| Yes | 1.49 | [ 1.13, 1.97] | ** | ||
| No | |||||
Note: Based on 7,622 Medicare beneficiaries age 65 and older, alive during the calendar and had positive total healthcare expenditures. The regression models controlled for sex, age, race/ethnicity, marital status, education, poverty status, Medicaid, private insurance, prescription drug coverage, number of chronic conditions, perceived physical health, functional status, body mass index, and current smoking. High out-of-pocket spending burden is defined as spending greater than 10% of family income out-of-pocket for medical care.
ADRD: Alzheimer’s disease and related dementias; SE: Standard error; Sig: Significance; Wt: Weighted
p < .001;
.001 < p < .01;
.01 < p < .05
Discussion
In unadjusted analyses, we found that having ADRD was associated with a doubling of total out-of-pocket spending over those without ADRD. Even after adjusting for other factors, those with ADRD had 37.5% higher out-of-pocket spending compared to those without ADRD. The care of individuals with ADRD is complex because the care includes increasing dependency on others for basic daily care needs, the management of comorbid conditions and the need for appropriate end-of-life care 20. Their range of needs span from minor assistance at the outset of the dementia to comprehensive services to meet activities of daily living and other care to address issues from the disease progression or other co-morbidities. The usual course of the disease is 5 to 10 years with the majority of the care focused on keeping the individual in community rather than a nursing facility.
We are not able to directly compare our estimates of out-of-pocket spending on healthcare with published studies due to differences in samples, time period, and components of out-of-pocket spending. However, when examined as proportions, our estimates were considerably smaller (73%) compared to the 356% higher out-of-pocket spending reported by Delavande et al. 12. We speculate that our estimates are lower because our study sample included elderly who were living in the community and did not include nursing home spending. It has to be noted that our finding of higher out-of-pocket spending among individuals with ADRD is contrary to the published study by Delavande et al, who found no significant differences among those with dementia and cognitive impairment or normal cognition21.
In our study, we found higher levels of out-of-pocket spending on home healthcare among those with ADRD compared to elderly without ADRD. This was not unexpected because home health is an important component of overall health management as this level of care often provides some assistance with self-care and opportunities for social engagement for the person with dementia and prevents behavioral outbursts, falls, injuries, or individuals from getting lost while the individual is receiving skilled nursing care. The increased utilization of home health and other skilled care has been noted in other studies, particularly in Lin et al., 2016, where they found, when compared to matched controls, individuals with ADRD were shown to use more home health before and after diagnosis 22. These authors concluded that recognition of the ADRD diagnosis takes special attention, It is worth noting that there is a recognition of the impact of the ADRD crisis by the policy-makers at the U.S. Department of Health and Human Services who created the National Plan to Address Alzheimer’s Disease to assist family members and persons with ADRD through research, care, and governmental collaboration 23.
We observed that elderly with ADRD had higher out-of-pocket spending on prescription drugs compared to those without ADRD, consisted with prior studies 24. The cost of prescription medications are on the rise in the U.S. In 2016, the total U.S. prescription sales were $448.2 billion, a 5.8% increase compared with 2015. Prescription expenditures in clinics and nonfederal hospitals totaled $63.7 billion (an 11.9% increase from 2015) and $34.5 billion (a 3.3% increase from 2015), respectively 25. This increase may disproportionately impact those with ADRD because they have higher number of multiple chronic conditions that require medications compared to patients without ADRD, which has been observed in our study as well as other published studies 24. Multimorbidity and polypharmacy complicate the needs of persons with ADRD and may also result in potentially inappropriate medications, adverse events from medications, both of which may also increase the out-of-pocket costs for these patients with ADRD 26.
Another contributing factor to increased levels of out-of-pocket expenditures for those with ADRD is the presence of the “donut-hole” gap (i.e. prescription drug coverage gap) in Medicare coverage. Patients with ADRD are particularly vulnerable to this gap; 39.5% of patients with ADRD experienced “donut-hole”, which may partially explain the high OOP 27.
These findings have implications for clinical management of chronic conditions. Although not specific to ADRD, a systematic review of cost sharing for prescription drugs found that in 85% of the studies cost sharing for prescription drugs had a negative effect on adherence28. The same review indicated that many studies (86% of the total studies reviewed) found improved adherence to be associated with improved health outcomes, suggesting that cost-sharing may lead to poor health outcomes 28.
Given that the number of older Americans with ADRD will likely increase significantly in the future, changes in public funding and healthcare policy aimed at reducing out-of-pocket spending for can reduce the financial burden of individuals with ADRD and their families. It is estimated that a tax credit for insured Americans who spent more than 5% of their income on healthcare can decrease spending up to 33% 29. Setting copays based on the level of sickness such as those adopted in France can also be considered 30-32. As cost-sharing is a major barrier for chronic care33, waiving or reducing cost-sharing for those with chronic illnesses such as ADRD may not only decrease the financial burden on the patients and their families but also improve chronic illness care.
Study Limitations
Several limitations to our study should be noted. Misclassification bias is possible with MCBS data. MCBS may not have captured all of the individuals with ADRD since, as a medical claims data source, if individuals have not sought care for ADRD, they would not have been identified as having ADRD. We could not measure the severity of ADRD as MCBS did not collect severity information. Also, some components of treatments may not be included in charges (and therefore not in the claims data) if reimbursement rates are very low, even if the treatment is provided or if out-of-pocket costs resulted. Medicare’s home healthcare benefit is limited. Medicare does not cover 24-hour care at home, meals delivered to the individual’s home, and homemaker or custodial care services (i.e. cooking, shopping, and laundry) unless such care is part of the skilled nursing or skilled therapy services individuals receive from a home health aide 34. Finally, income and asset data are not available for people with ADRD in MCBS dataset.
Considerations for future research
Individuals with ADRD often have family members or significant others provide uncompensated care. It is estimated that informal caregivers provide 70 hours/week to individuals with ADRD35, 36 Caregivers often experience lost earnings as a result 37, 38 The Alzheimer’s Association reported 17.4 billion hours of unpaid care was provided for individuals with ADRD in 2011, valued at $210 billion 39. In addition, more than 60% of formal services provided for individuals with ADRD are financed by the family, regardless of care setting 40. The growing elderly population and the possible shortage of informal caregivers increase individuals’ healthcare needs and costs 41. Policymakers recommend that people plan and save for the likelihood of requiring long-term care, and that tax proposals should be in place to limit the financial burdens.
Additional research is needed to examine caregiver burden including the extent to which individuals are providing care to those with ADRD, the nature of the care that is being provided, the impact on the family psychosocially by providing the care, and the impact on the family financially. Previously most of the research has focused upon the person with ADRD, additional information is needed about the impact more broadly.
Conclusion
Medicare beneficiaries with ADRD have higher out-of-pocket expenditures as compared with Medicare beneficiaries without ADRD. The financial burden as a percent of income is higher with Medicare beneficiaries with ADRD as compared with Medicare beneficiaries without ADRD.
Footnotes
The authors report no conflicts of interest.
All authors contributed to the conception, and design of the research. US conducted the statistical analyses. ND wrote the first draft. All authors contributed to the manuscript and approved the final version.
Contributor Information
Nilanjana Dwibedi, Department of Pharmaceutical Systems and Policy, West Virginia University School of Pharmacy, Robert C. Byrd Health Sciences Center [North], P.O. Box 9510 Morgantown, WV 26506-9510.
Patricia A. Findley, Rutgers University, School of Social Work, 536 George Street, New Brunswick, NJ 08901.
R. Constance Wiener, Department of Dental Practice and Rural Health, School of Dentistry, 104A Health Sciences Addition, P.O. Box 9448, West Virginia University, Morgantown, WV 26506-9448.
Chan Shen, Departments of Health Services Research and Biostatistics, University of Texas MD Anderson Cancer Center, 1400 Pressler St, Houston, TX 77030.
Usha Sambamoorthi, Department of Pharmaceutical Systems and Policy, West Virginia University School of Pharmacy, Robert C. Byrd Health Sciences Center [North], P.O. Box 9510 Morgantown, WV 26506-9510.
References
- 1.Koller D, Bynum JP. Dementia in the USA: state variation in prevalence. J Public Health (Oxf) 2015;37:597–604. doi: 10.1093/pubmed/fdu080. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 2.2016 Alzheimer’s disease facts and figures. Alzheimers Dement. 2016;12:459–509. doi: 10.1016/j.jalz.2016.03.001. [DOI] [PubMed] [Google Scholar]
- 3.Cubanski J, Swoope C, Damico A, et al. How much is enough? Out-of-pocket spending among Medicare beneficiaries: a chartbook. 2014 Available from URL: http://kff.org/report-section/how-much-is-enough-out-of-pocket-spending-among-medicare-beneficiariessection-10.
- 4.CMS. Medicare Coverage – General Information 2016. 2016 Available from https://www.cms.gov/Medicare/Coverage/CoverageGenInfo/
- 5.Miller KD, Siegel RL, Lin CC, et al. Cancer treatment and survivorship statistics, 2016. CA Cancer J Clin. 2016;66:271–289. doi: 10.3322/caac.21349. [DOI] [PubMed] [Google Scholar]
- 6.Tamblyn R, Eguale T, Huang A, Winslade N, Doran P. The incidence and determinants of primary nonadherence with prescribed medication in primary care: a cohort study. Annals of Internal Medicine. 2014;160:441–450. doi: 10.7326/M13-1705. [DOI] [PubMed] [Google Scholar]
- 7.Hoffman C, Schwartz K. Eroding access among nonelderly US adults with chronic conditions: ten years of change. Health Affairs. 2008;27:w340–w348. doi: 10.1377/hlthaff.27.5.w340. [DOI] [PubMed] [Google Scholar]
- 8.Thorpe KE, Ogden LL, Galactionova K. Chronic conditions account for rise in Medicare spending from 1987 to 2006. Health Affairs. 2010;29:718–724. doi: 10.1377/hlthaff.2009.0474. [DOI] [PubMed] [Google Scholar]
- 9.Sansoni J, Anderson KH, Varona LM, Varela G. Caregivers of Alzheimer’s patients and factors influencing institutionalization of loved ones: some considerations on existing literature. Ann Ig. 2013;25:235–246. doi: 10.7416/ai.2013.1926. [DOI] [PubMed] [Google Scholar]
- 10.Hurd MD, Martorell P, Delavande A, Mullen KJ, Langa KM. Monetary costs of dementia in the United States. N Engl J Med. 2013;368:1326–1334. doi: 10.1056/NEJMsa1204629. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 11.Kelley AS, McGarry K, Gorges R, Skinner JS. The burden of health care costs for patients with dementia in the last 5 years of life. Ann Intern Med. 2015;163:729–736. doi: 10.7326/M15-0381. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 12.Delavande A, Hurd MD, Martorell P, Langa KM. Dementia and out-of-pocket spending on health care services. Alzheimers Dement. 2013;9:19–29. doi: 10.1016/j.jalz.2011.11.003. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 13.Yang Z, Levey A. Gender Differences: A Lifetime Analysis of the Economic Burden of Alzheimer’s Disease. Womens Health Issues. 2015;25:436–440. doi: 10.1016/j.whi.2015.06.001. [DOI] [PubMed] [Google Scholar]
- 14.Andersen RM. Revisiting the behavioral model and access to medical care: does it matter? Journal of Health and Social Behavior. 1995:1–10. [PubMed] [Google Scholar]
- 15.Gregori D, Petrinco M, Bo S, Desideri A, Merletti F, Pagano E. Regression models for analyzing costs and their determinants in health care: an introductory review. Int J Qual Health Care. 2011;23:331–341. doi: 10.1093/intqhc/mzr010. [DOI] [PubMed] [Google Scholar]
- 16.Mihaylova B, Briggs A, O’Hagan A, Thompson SG. Review of statistical methods for analysing healthcare resources and costs. Health Econ. 2011;20:897–916. doi: 10.1002/hec.1653. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 17.Galbraith AA, Wong ST, Kim SE, Newacheck PW. Out-of-Pocket Financial Burden for Low-Income Families with Children: Socioeconomic Disparities and Effects of Insurance. Health Services Research. 2005;40:1722–1736. doi: 10.1111/j.1475-6773.2004.00421.x. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 18.Sambamoorthi U, Shea D, Crystal S. Total and out-of-pocket expenditures for prescription drugs among older persons. Gerontologist. 2003;43:345–359. doi: 10.1093/geront/43.3.345. [DOI] [PubMed] [Google Scholar]
- 19.Chronic Conditions Data Warehouse. Available from URL: https://www.ccwdata.org/web/guest/condition-categories.
- 20.Bynum JP. The long reach of alzheimer’s disease: Patients, practice, and policy. Health Affairs. 2014;33:534–540. doi: 10.1377/hlthaff.2013.1247. [DOI] [PubMed] [Google Scholar]
- 21.Delavande A, Hurd MD, Martorell F, Langa KM. Dementia and Out-of-Pocket Spending on Health Care Services. Alzheimers Dement. 2013;9:19–29. doi: 10.1016/j.jalz.2011.11.003. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 22.Lin P-J, Zhong Y, Fillit HM, Chen E, Neumann PJ. 30-DAY HOSPITAL READMISSIONS AMONG MEDICARE BENEFICIARIES WITH ALZHEIMER’S DISEASE AND RELATED DEMENTIAS PRIOR TO AND FOLLOWING DIAGNOSIS. Alzheimer’s & Dementia: The Journal of the Alzheimer’s Association. 2016;12:811–812. [Google Scholar]
- 23.Borson S, Boustani MA, Buckwalter KC, et al. Report on milestones for care and support under the US National Plan to Address Alzheimer’s Disease. Alzheimer’s & Dementia. 2016;12:334–369. doi: 10.1016/j.jalz.2016.01.005. [DOI] [PubMed] [Google Scholar]
- 24.Hunter CA, Kirson NY, Desai U, Cummings AKG, Faries DE, Birnbaum HG. Medical costs of Alzheimer’s disease misdiagnosis among US Medicare beneficiaries. Alzheimer’s & Dementia. 2015;11:887–895. doi: 10.1016/j.jalz.2015.06.1889. [DOI] [PubMed] [Google Scholar]
- 25.Schumock GT, Li EC, Wiest MD, et al. National trends in prescription drug expenditures and projections for 2017. American Journal of Health-System Pharmacy. 2017:ajhp170164. doi: 10.2146/ajhp170164. [DOI] [PubMed] [Google Scholar]
- 26.Patel T, Slonim K, Lee L. Use of potentially inappropriate medications among ambulatory home-dwelling elderly patients with dementia: A review of the literature. Canadian Pharmacists Journal/Revue des Pharmaciens du Canada. 2017;150:169–183. doi: 10.1177/1715163517701770. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 27.Ettner SL, Steers N, Duru OK, et al. Entering and exiting the Medicare part D coverage gap: role of comorbidities and demographics. Journal of General Internal Medicine. 2010;25:568–574. doi: 10.1007/s11606-010-1300-6. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 28.Eaddy MT, Cook CL, O’Day K, Burch SP, Cantrell CR. How Patient Cost-Sharing Trends Affect Adherence and Outcomes: A Literature Review. Pharmacy and Therapeutics. 2012;37:45–55. [PMC free article] [PubMed] [Google Scholar]
- 29.Eibner C, N S, Liu J. Hillary Clinton’s Health Care Reform Proposals: Anticipated Effects on Insurance Coverage, Out-of-Pocket Costs, and the Federal Deficit. [accessed 09, 2016] [Google Scholar]
- 30.Rodwin VG, Sandier S. Health care under French national health insurance. Health Aff (Millwood) 1993;12:111–131. doi: 10.1377/hlthaff.12.3.111. [DOI] [PubMed] [Google Scholar]
- 31.Fund TC. International Profiles of Health Care Systems. 2015;2015 [Google Scholar]
- 32.Foundation THJKF. Cost Sharing for Health Care: France, Germany, and Switzerland. 2009 [Google Scholar]
- 33.Iuga AO, McGuire MJ. Adherence and health care costs. Risk Manag Healthc Policy. 2014;7:35–44. doi: 10.2147/RMHP.S19801. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 34.Medicareinteractive.org. Types of home health care Medicare will not pay for. Available from URL: http://www.medicareinteractive.org/get-answers/medicare-covered-services/home-health-care-benefit-part-a-and-b/types-of-home-health-care-medicare-will-not-pay-for [accessed 11/04/2016]
- 35.Rice DP, Fox PJ, Max W, et al. The economic burden of Alzheimer’s disease care. Health Aff (Millwood) 1993;12:164–176. doi: 10.1377/hlthaff.12.2.164. [DOI] [PubMed] [Google Scholar]
- 36.Max W, Webber P, Fox P. Alzheimer’s disease. The unpaid burden of caring. J Aging Health. 1995;7:179–199. doi: 10.1177/089826439500700202. [DOI] [PubMed] [Google Scholar]
- 37.Small GW, McDonnell DD, Brooks RL, Papadopoulos G. The impact of symptom severity on the cost of Alzheimer’s disease. J Am Geriatr Soc. 2002;50:321–327. doi: 10.1046/j.1532-5415.2002.50065.x. [DOI] [PubMed] [Google Scholar]
- 38.Moore MJ, Zhu CW, Clipp EC. Informal costs of dementia care: estimates from the National Longitudinal Caregiver Study. J Gerontol B Psychol Sci Soc Sci. 2001;56:S219–228. doi: 10.1093/geronb/56.4.s219. [DOI] [PubMed] [Google Scholar]
- 39.Thies W, Bleiler L, Alzheimer’s Association 2011 Alzheimer’s disease facts and figures. Alzheimers Dement 2011. 2011;7:208–44. doi: 10.1016/j.jalz.2011.02.004. [DOI] [PubMed] [Google Scholar]
- 40.Rice DP, Fox PJ, Max W, et al. The economic burden of Alzheimer’s disease care. Health Affairs. 1993;12:164–176. doi: 10.1377/hlthaff.12.2.164. [DOI] [PubMed] [Google Scholar]
- 41.Zhu CW, Sano M. Economic considerations in the management of Alzheimer’s disease. Clinical interventions in aging. 2006;1:143. doi: 10.2147/ciia.2006.1.2.143. [DOI] [PMC free article] [PubMed] [Google Scholar]