Abstract
Background:
Although one-third of Medicare beneficiaries are enrolled in Medicare Advantage (MA) plans, there is limited information about the cost of treating Alzheimer’s disease and related dementias (ADRD) in these settings.
Objective:
To estimate direct health care costs attributable to ADRD among older adults within a large MA plan.
Research Design:
A retrospective cohort design was used to estimate direct total, outpatient, inpatient, ambulatory pharmacy and nursing home costs for three years before and after an incident ADRD diagnosis for 927 individuals diagnosed with ADRD relative to a sex- and birth year- matched set of 2,945 controls.
Subjects:
Adults, 65 years of age and older enrolled in the Kaiser Permanente Washington MA plan and the Adult Changes in Thought (ACT) Study, a prospective longitudinal cohort study of ADRD and brain aging.
Measures:
Data on monthly health service use obtained from health system electronic medical records for the period 1992-2012.
Results:
Total monthly health care costs for individuals with ADRD are statistically greater (p < .05) than controls beginning in the third month prior to diagnosis and remain significantly greater through the 8th month following diagnosis. Greater total health costs are driven by significantly (p < .05) greater nursing home costs among individuals diagnosed with ADRD beginning in the third month pre-diagnosis. While total costs were no longer significantly greater at 8 months following diagnosis, nursing home costs remained higher for the people with dementia through the three-years post-diagnosis we analyzed.
Conclusions:
Greater total health care costs among individuals with ADRD are primarily driven by nursing home costs.
Keywords: Medicare Advantage, Alzheimer’s disease, Dementia, Direct health care cost
Introduction
The increasing prevalence of Alzheimer’s Disease and related dementias (ADRD) in the United States has led to a demand for reliable estimates of the direct medical care costs for individuals diagnosed with ADRD. In 2018 approximately 5.5 million people in the United States are diagnosed with ADRD, 96% of whom are aged 65 and older.1 Demographic trends and rates of incident ADRD suggest that the number Americans with ADRD will more than double by 2030, with 11 to over 13 million1 people diagnosed with ADRD by mid-century.
The literature reporting ADRD-attributable health care costs in the United States relies primarily on the experience of older adults in Medicare’s fee-for-service (FFS) program.2 The Center for Medicare and Medicaid Services (CMS) has made service use data available for research within Medicare FFS for decades. Access to this data has supported a large number of studies addressing the cost of ADRD within Medicare FFS.3,4
There remains a critical gap however, in our understanding of the costs of ADRD among older adults insured through Medicare’s managed care program. A Medicare managed care option has been available since 1983 when the Part C program, now known as Medicare Advantage (MA) was established. As of this writing, CMS has only made data on health care use within the MA program available for research for the year 2015. The lack of data creates a barrier to conducting ADRD specific research within MA.
There are three reasons why better information about ADRD-attributable spending among MA plans is needed. First, while MA enrollment was relatively stable from the mid-1990s through 2010 at between 15 and 20 percent of Medicare beneficiaries,5 there has been significant MA growth in recent years. As of 2017, almost one-third of beneficiaries receive care through managed care plans and CMS forecasts greater absolute and relative MA enrollment growth over the next decade.5 Second, while MA enrollees have traditionally been younger and in relatively better health than Medicare FFS beneficiaries, a new risk adjustment formula introduced between 2004 and 2006 creates incentives for less favorable MA risk selection.6,7 There is evidence that beneficiaries with ADRD may are now more likely to switch to MA plans as well as less likely to switch out of MA plans after being diagnosed.8,9 Finally, while MA plans must cover the same basic services as Medicare FFS, some provide services in addition to traditional Medicare requirements, such as nursing home care beyond 30 days post discharge from an inpatient stay.10–12 The different service mix may result in different costs of care between Medicare FFS and MA for patients with similar disease profiles.13,14 Therefore, relying primarily on Medicare FFS data may lead to an incomplete understanding of the current and future costs of ADRD.
A recent review15 identified only seven16–22 peer-reviewed papers reporting original data on ADRD costs relative to controls within an MA population, and only 1 used data more recent than 2011.21 Differences in populations, study periods, means of case ascertainment, costing methods and analytic approach result in a wide divergence in the estimates of ADRD attributable costs within this sparse literature. In the first year following diagnosis, Suehs, et al.21 estimated that individuals with ADRD incur an additional $8,938 ($17,712 vs $8,774) in medical expenses relative to individuals without ADRD, while Joyce, et al. report first year incident ADRD costs of $38,794 ($48,985 vs $10,191) in 2018 US dollars. Both studies used claims data to assess health service use and cost but different diagnostic criteria to identify people with incident cases of ADRD. Estimates of prevalent ADRD costs show similar variation, ranging from $27,948 ($8,726 greater than controls) to $9,33417($3,429 greater than controls), with the variation attributable to both differences in ADRD definitions and which components of service use are included in analyses.
The limited number of MA specific ADRD cost studies highlights the need for additional research on the economic impact of ADRD within the MA program. In support of this goal, we estimate the direct attributable medical care costs among older adults diagnosed with ADRD that participated in the Adult Changes in Thought (ACT) study, a longitudinal study of aging and the incidence of and risk factors for dementia that has been conducted since 1994 at Kaiser Permanente Washington (KPWA), which from its founding in 1947 until 2016 operated as Group Health Cooperative (GH).
Data and Methods
We estimate the incremental direct medical care costs of ADRD from the perspective of a managed health care delivery system. We examine the experience of adults, 65 years of age and older, enrolled in the KPWA MA plan who participated in the ACT study, an NIA-funded (U01AG006781) longitudinal cohort study of dementia incidence. KPWA is a large, integrated health care and insurance system headquartered in Seattle, WA, that provides comprehensive health care and insurance for a stable year over year population of approximately 600,000 people in Washington State, 11% of whom are aged 65 or older.
KPWA’s integrated delivery system includes 25 owned and operated primary care clinics throughout Washington State, including 20 in the Puget Sound region of Western Washington, most with onsite pharmacies, laboratories, and radiology suites. Four specialty clinics and six urgent care/emergency departments support these primary care clinics staffed by salaried providers. Until 2002 KPWA also owned a nursing home. Services are also provided through a network of providers and facilities, with data on service use captured through claims. The integration of service delivery and finance with KPWA incents greater coordination of care across providers and care settings as well as for reducing expenditures. Data on health service use provided by staff and contracted providers is merged into a unified patient record that is used for clinical and research applications. This study was approved by the KPWA and University of Washington Institutional Review Boards.
The ACT study began in 1994 with a randomly sampled cohort of 2,581 community dwelling KPWA members aged 65 and older found not to have dementia. An additional 811 persons entered the cohort in 2000-2002 and through ongoing enrollment to replace persons who die, become demented or drop out, 120 to180 people have been added annually since 2004 to maintain a cohort size of approximately 2000. ACT participants are evaluated when they join the study and screened every two years thereafter for evidence of ADRD. Those screening positive undergo a complete dementia diagnostic evaluation and multidisciplinary consensus diagnosis process yielding classifications of no dementia, Alzheimer’s disease (AD), vascular dementia (VaD), dementias with multiple etiologies (usually AD + VaD), dementia due to a general medical condition or medications/drugs, and other dementias (e.g. frontotemporal, Lewy body). Persons who continue to screen negative or who do not meet diagnostic criteria for dementia are evaluated every two years. Those who are diagnosed with incident dementia are assigned an onset date equal to the midpoint between the two bi-annual evaluations and are then followed annually to confirm the accuracy of the dementia diagnosis and stability of the diagnostic subtype. Details on the ACT study may be found in previously published research.
We identified ACT participants with a confirmed ADRD diagnosis as of December 31, 2012, who were enrolled in KPWA, and provided consent for the study team to access their medical records for at least one year following their ADRD diagnosis. To estimate the incremental costs of incident and prevalent ADRD we randomly assigned without replacement ACT participants without ADRD to cases based on exact matches of sex and ACT cohort, matches of age within 3 years at entry into the ACT study and remaining in the study at the time of the case’s ADRD diagnosis with controls were assigned a pseudo-onset date of the ADRD diagnosis date for their matched positive case. The ACT protocol requires that participants have at least 1 study visit after their onset date; we similarly require that potential controls have at least 1 study visit at which no dementia diagnosis was made following the onset date of the case to whom they are matched. Individuals subsequently diagnosed with dementia remain in the pool of potential controls if their dementia onset date was at least five years after the onset date of the case to whom they matched.
We assign unit costs to services using an approach KP researchers developed to support comparisons of the KP experience to other managed care plans and to FFS Medicare. This model is based on how CMS reimburses providers for inpatient (Part A) and outpatient services (Part B), ambulatory pharmacy (Part D) and nursing home care that may or may not be covered by Medicare FFS. This work was funded by the National Cancer Institute [R01CA114204] and has been used in several MA specific cost analyses.14,23,24
We collected monthly total, inpatient, outpatient, ambulatory pharmacy, and nursing home services costs for up to 3 years before and 3 years after each individual index date - for the calendar period January, 1992 through December, 2012 - while study participants were enrolled in KPWA and provided consent for the study team to access their medical records. Total costs include other components of service use in addition to the four we report. Nominal cost values were adjusted to reflect 2018 prices using the Medical Care Component of the Consumer Price Index.
We use repeated measures generalized estimating equations (GEE) to estimate the incremental monthly cost of incident and prevalent ADRD. Monthly cost estimates allow for greater detail on the cost impact of incident ADRD. Repeated measures GEE supports corrections for correlated data that are common in analyses of health cost over time and also allows for missing data in a time series design.25,26 Thus, we may include individuals in our analysis for whom we have less than the full 72 months of data because they either dis-enrolled from KPWA, their follow-up overlapped with the end of the study, or they died within three years following the onset date. We estimated separate regressions using the person month as the unit of analysis for each cost measure and assume errors follow a normal distribution with an identity link.26,27 Our goal is to estimate adjusted monthly mean costs for persons with and without ADRD rather than test hypotheses related to specific or combinations of coefficients. This specification produces unbiased coefficient estimates with asymptotically efficient standard errors.28 Each regression includes adjustments for individual age, sex, marital status, race, education, and the presence of a diagnosis for the following co-morbid or co-occurring conditions: acute myocardial infarction, anemia, arthritis, atrial fibrillation, chronic kidney disease, cataracts, chronic obstructive pulmonary disease, chronic kidney disease, depression, diabetes, heart failure, hip or pelvic fracture, hypertension, hypotension, lung cancer, osteoporosis and stroke in the year prior to ADRD onset. Robust standard errors are calculated using the ‘sandwich’ method described by Huber.29
Results
As of December 31, 2012, 5,097 ACT participants had completed at least one biannual assessment. Among these individuals 1,080 had a confirmed ADRD diagnosis of whom 927 were enrolled in KPWA, provided consent for the study team to access their medical records for at least one year following their diagnosis and were successfully matched to at least one control. Using the criteria described above, 2,945 ACT participants without an ADRD diagnosis were identified as controls or 3.2 controls per case on average. A descriptive summary of the sample in Table 1 reveals that cases are more likely to be women and are slightly older and have less education on average although none of these differences are statistically significant. Cases were more likely to be widowed and less likely to have been married (p .05) at ACT study enrollment. While we required all subjects be alive one year following the index date, controls have greater mortality (15.8% vs 9.9%) two years post index, which is a statistically significant difference (p < 0.01), but the mortality gap narrows (23.4% vs 20.0%) by the end of the third year post index. There are similar prevalence rates for most co-morbid and co-occurring medical conditions but cases have significantly higher rates of depression (p < .0001), heart failure (p = .0002), hip/pelvic fractures (p = .0002), hypotension (p = .0075) and stroke (p < .0001).
Table 1:
Sample Descriptives
| Variable | Total n= 3872 | Case n= 927 | Control n= 2945 | Probability of chi square test for group difference | |||
|---|---|---|---|---|---|---|---|
|
| |||||||
| Mean Age (SD) at Onset | 75.5 | 6.3 | 76.9 | 6.3 | 75.1 | 6.2 | .56 |
|
| |||||||
| Gender - N/% Male | 1,504 | 38.8% | 340 | 36.7% | 1164 | 39.5% | .13 |
|
| |||||||
| Race | N | % | N | % | N | % | |
|
| |||||||
| White | 3,487 | 90.1% | 845 | 91.2% | 2,642 | 89.7% | .52 |
| Black | 154 | 4.0% | 34 | 3.7% | 120 | 4.1% | |
| Asian | 142 | 3.7% | 27 | 2.9% | 115 | 3.9% | |
| American Indian/Alaska Native | 7 | 0.2% | 3 | 0.3% | 4 | 0.1% | |
| Native Hawaiian/Other Pacific Islander | 1 | 0.0% | - | 0.0% | 1 | 0.0% | |
| Other including. Mixed | 81 | 2.1% | 18 | 1.9% | 63 | 2.1% | |
|
| |||||||
| Education | .08 | ||||||
|
| |||||||
| No Degree | 424 | 11.0% | 116 | 12.5% | 308 | 10.5% | |
| GED/HS Diploma | 1,795 | 46.4% | 461 | 49.7% | 1,334 | 45.3% | |
| College Degree | 821 | 21.2% | 170 | 18.3% | 651 | 22.1% | |
| Master’s Degree | 393 | 10.1% | 93 | 10.0% | 300 | 10.2% | |
| Doctoral Degree | 158 | 4.1% | 27 | 2.9% | 131 | 4.4% | |
| Other | 281 | 7.3% | 60 | 6.5% | 221 | 7.5% | |
|
| |||||||
| Mortality in follow-up year | p. < .01 | ||||||
|
| |||||||
| Year 2 | 556 | 14.4% | 92 | 9.9% | 464 | 15.8% | |
| Year 3 | 317 | 8.2% | 94 | 10.1% | 223 | 7.6% | |
| Total | 873 | 22.5% | 186 | 20.1% | 687 | 23.3% | |
|
| |||||||
| Marital Status, first visit | .05 | ||||||
|
| |||||||
| Never Married | 121 | 3.1% | 24 | 2.6% | 97 | 3.3% | |
| Married/living as Married | 2,064 | 53.3% | 481 | 51.9% | 1,583 | 53.8% | |
| Divorced | 490 | 12.7% | 105 | 11.3% | 385 | 13.1% | |
| Widowed | 1,078 | 27.8% | 292 | 31.5% | 786 | 26.7% | |
| Other | 119 | 3.1% | 25 | 2.7% | 94 | 3.2% | |
|
| |||||||
| Chronic Illness Prevalence in Year Prior to Index Date | |||||||
|
| |||||||
| Acute Myocardial Infarction | 46 | 1% | 15 | 2% | 31 | 1% | .17 |
| Anemia | 458 | 12% | 122 | 13% | 336 | 11% | .15 |
| Arthritis | 1,099 | 28% | 261 | 28% | 838 | 28% | .86 |
| Asthma | 267 | 7% | 52 | 6% | 215 | 7% | .08 |
| Atrial Fibrillation | 530 | 14% | 139 | 15% | 391 | 13% | .19 |
| COPD | 420 | 11% | 95 | 10% | 325 | 11% | .50 |
| Cataracts | 2,037 | 53% | 473 | 51% | 1564 | 53% | .27 |
| Chronic Kidney Disease | 341 | 9% | 79 | 9% | 262 | 9% | .73 |
| Depression | 424 | 11% | 173 | 19% | 251 | 9% | <.0001 |
| Diabetes | 568 | 15% | 131 | 14% | 437 | 15% | .59 |
| Glaucoma | 746 | 19% | 186 | 20% | 560 | 19% | .48 |
| Heart Failure | 506 | 13% | 155 | 17% | 351 | 12% | 0.0002 |
| Hip or Pelvic Fracture | 86 | 2% | 35 | 4% | 51 | 2% | 0.0002 |
| Hypertension | 1,627 | 42% | 413 | 45% | 1214 | 41% | 0.07 |
| Hypothyroidism | 410 | 11% | 120 | 13% | 290 | 10% | 0.0075 |
| Lung Cancer | 23 | 1% | 2 | 0.2% | 21 | 1% | .09 |
| Osteoporosis | 267 | 7% | 69 | 7% | 198 | 7% | .45 |
| Stroke | 238 | 6% | 103 | 11% | 135 | 5% | <.0001 |
Regression-adjusted monthly total health care costs for three years before and after the index date for individuals with ADRD relative to controls are shown in Figure 1, nursing home costs are shown in Figure 2 and outpatient , inpatient, and ambulatory pharmacy costs are shown in Figures 3, panels A – C. Months in which cost differences between cases and controls are statistically significant are identified in each figure for specific months or for a range of months as appropriate. Detailed results for monthly adjusted costs and tests of significance for ADRD status for each cost component are reported in Table 2.
Figure 1:
Total Costs by Month and ADRD Status
Figure 2:
Nursing Home Costs by Month and ADRD Status
Figure 3.
Panel A: Outpatient Pharmacy Costs by Month and ADRD Status
Panel B: Outpatient Service Costs by Month and ADRD Status
Panel C: Inpatient Costs by Month and ADRD Status
Table 2:
Regression Adjusted Monthly Costs
| Total Cost | Nursing Home Cost | Outpatient Office Costs | Ambulatory Pharmacy | Inpatient Costs | |||||||||||
|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|
|
| |||||||||||||||
| Adjusted Mean(SE) | p Value | Adjusted Mean(SE) | p Value | Adjusted Mean(SE) | p Value | Adjusted Mean(SE) | p Value | Adjusted Mean(SE) | p Value | ||||||
| Month | Controls | Cases | Controls | Cases | Controls | Cases | Controls | Cases | Controls | Cases | |||||
| 36th Pre | $1,124.71 (165.94) |
$1,364.94 (198.33 ) |
0.58 | $14.42 (5.48) |
$110.81 (34.01) |
0.88 | $660.26 (50.50) |
$682.84 (60.66) |
0.87 | $132.97 (94.99) |
$235.51 (116.70) |
0.38 | 151.19 (18.93) |
148.23 (19.80) |
0.35 |
| 35th Pre | $1,168.41 (166.72) |
$1,213.47 (181.62 ) |
0.25 | $22.69 (7.52) |
$78.73 (30.47) |
0.27 | $647.80 (49.99) |
$668.40 (60.39) |
0.89 | $171.71 (97.73) |
$109.80 (96.67) |
0.09 | 142.89 (18.83) |
171.02 (21.13) |
0.07 |
| 34th Pre | $1,180.96 (167.07) |
$1,229.54 (186.40 ) |
0.30 | $21.50 (7.68) |
$67.49 (26.87) |
0.11 | $655.11 (50.90) |
$659.22 (59.65) |
0.81 | $168.81 (96.43) |
$176.13 (112.29) |
0.33 | 145.50 (18.91) |
162.29 (19.89) |
0.76 |
| 33rd Pre | $1,154.51 (168.35) |
$1,273.84 (209.55 ) |
0.75 | $19.28 (8.19) |
$70.40 (26.53) |
0.16 | $623.68 (50.66) |
$635.80 (56.95) |
0.95 | $201.11 (100.32) |
$202.12 (126.46) |
0.35 | 142.99 (18.76) |
187.85 (44.54) |
0.77 |
| 32nd Pre | $1,083.34 (166.99) |
$1,319.62 (191.96 ) |
0.58 | $18.88 (7.57) |
$83.73 (31.10) |
0.42 | $626.13 (50.44) |
$653.95 (60.29) |
0.77 | $150.14 (97.44) |
$189.15 (103.59) |
0.09 | 141.32 (18.75) |
224.72 (47.90) |
0.88 |
| 31st Pre | $1,111.46 (165.49) |
$1,246.24 (200.09 ) |
0.80 | $20.59 (8.18) |
$73.32 (28.44) |
0.20 | $609.04 (50.52) |
$656.10 (59.58) |
0.47 | $123.27 (94.36) |
$189.15 (110.95) |
0.97 | 158.45 (20.94) |
164.26 (20.10) |
0.60 |
| 30th Pre | $1,297.54 (182.38) |
$1,176.98 (186.83 ) |
0.04 | $24.26 (12.47) |
$75.40 (30.04) |
0.20 | $664.94 (51.93) |
$600.33 (55.99) |
0.05 | $224.14 (111.52) |
$194.91 (116.92) |
0.15 | 162.56 (19.87) |
151.02 (19.74) |
0.55 |
| 29th Pre | $1,277.76 (169.37) |
$1,367.85 (209.09 ) |
0.59 | $42.35 (9.96) |
$94.56 (32.64) |
0.23 | $661.25 (51.06) |
$685.12 (60.33) |
0.86 | $199.97 (98.67) |
$267.28 (140.40) |
0.78 | 163.14 (19.87) |
165.98 (19.64) |
0.72 |
| 28th Pre | $1,182.63 (170.42) |
$1,487.03 (199.28 ) |
0.31 | $33.44 (8.16) |
$108.31 (34.72) |
0.64 | $634.22 (52.67) |
$731.60 (61.01) |
0.08 | $148.02 (97.71) |
$229.33 (108.44) |
0.02 | 142.86 (18.89) |
195.00 (25.98) |
0.45 |
| 27th Pre | $1,223.35 (173.29) |
$1,491.12 (244.75 ) |
0.62 | $22.69 (8.52) |
$109.56 (35.08) |
0.90 | $629.08 (51.14) |
$614.04 (56.80) |
0.45 | $176.46 (101.74) |
$273.11 (127.10) |
0.98 | 159.06 (19.06) |
165.06 (19.89) |
0.50 |
| 26th Pre | $1,242.06 (168.43) |
$1,565.03 (210.95 ) |
0.31 | $26.75 (10.81) |
$117.89 (33.51) |
1.00 | $648.73 (51.18) |
$681.46 (60.12) |
0.70 | $160.82 (97.07) |
$260.27 (114.21) |
0.32 | 163.48 (19.06) |
181.19 (20.18) |
0.38 |
| 25th Pre | $1,228.53 (170.65) |
$1,403.86 (193.79 ) |
0.95 | $41.04 (11.89) |
$112.06 (32.66) |
0.57 | $629.39 (50.14) |
$720.79 (59.89) |
0.08 | $201.21 (100.98) |
$212.51 (111.47) |
0.83 | 162.58 (19.30) |
166.46 (19.94) |
0.81 |
| 24th Pre | $1,373.36 (173.29) |
$1,261.23 (185.92 ) |
0.02 | $46.81 (10.56) |
$107.06 (30.64) |
0.38 | $670.30 (51.12) |
$649.08 (60.08) |
0.42 | $253.85 (103.24) |
$171.09 (107.64) |
0.96 | 169.06 (22.18) |
174.42 (20.38) |
0.14 |
| 23rd Pre | $1,345.71 (175.65) |
$1,571.08 (207.68 ) |
0.71 | $36.58 (10.48) |
$92.48 (30.52) |
0.28 | $672.52 (51.23) |
$723.00 (62.59) |
0.47 | $199.94 (103.90) |
$339.72 (129.88) |
0.80 | 170.84 (22.59) |
181.90 (22.05) |
0.30 |
| 22nd Pre | $1,318.40 (175.64) |
$1,442.90 (197.05 ) |
0.76 | $36.98 (11.97) |
$94.56 (35.82) |
0.30 | $689.75 (53.37) |
$667.45 (58.05) |
0.40 | $187.89 (102.91) |
$218.09 (107.27) |
0.88 | 167.44 (19.14) |
175.50 (20.08) |
1.00 |
| 21st Pre | $1,360.72 (176.58) |
$1,520.58 (208.90 ) |
0.96 | $44.71 (12.11) |
$133.72 (39.77) |
0.95 | $676.60 (53.92) |
$655.03 (58.54) |
0.41 | $246.65 (104.82) |
$320.78 (132.08) |
0.90 | 180.96 (22.16) |
185.15 (20.59) |
0.68 |
| 20th Pre | $1,355.57 (169.58) |
$1,427.36 (197.69 ) |
0.47 | $54.55 (10.95) |
$158.72 (40.89) |
0.77 | $683.20 (51.38) |
$641.92 (56.50) |
0.16 | $201.31 (98.25) |
$220.75 (120.09) |
0.31 | 174.08 (21.04) |
196.28 (21.22) |
0.91 |
| 19th Pre | $1,313.33 (169.67) |
$1,545.10 (228.50 ) |
0.71 | $49.83 (12.65) |
$145.39 (39.55) |
0.92 | $658.80 (50.51) |
$682.53 (62.26) |
0.86 | $201.11 (99.74) |
$254.17 (133.09) |
0.41 | 180.77 (21.39) |
261.21 (94.43) |
0.82 |
| 18th Pre | $1,298.53 (169.32) |
$1,637.60 (216.50 ) |
0.28 | $41.44 (11.20) |
$162.47 (39.03) |
0.48 | $658.81 (50.27) |
$642.39 (56.78) |
0.42 | $194.55 (97.91) |
$388.81 (141.53) |
0.49 | 178.38 (22.75) |
173.58 (19.91) |
0.15 |
| 17th Pre | $1,340.69 (172.43) |
$1,380.05 (189.51 ) |
0.31 | $48.39 (12.50) |
$141.22 (36.23) |
0.97 | $654.34 (50.58) |
$645.81 (56.22) |
0.55 | $243.86 (101.55) |
$162.29 (108.51) |
0.57 | 191.33 (28.95) |
183.00 (19.99) |
0.15 |
| 16th Pre | $1,338.78 (170.11) |
$1,253.81 (179.59 ) |
0.02 | $39.73 (12.92) |
$131.64 (39.24) |
0.99 | $661.14 (51.94) |
$675.63 (58.84) |
0.99 | $158.90 (97.34) |
$112.85 (98.58) |
0.46 | 194.93 (28.83) |
182.08 (20.50) |
0.15 |
| 15th Pre | $1,397.57 (174.15) |
$1,427.66 (214.46 ) |
0.39 | $46.29 (13.62) |
$121.64 (41.05) |
0.68 | $690.22 (53.39) |
$626.37 (57.60) |
0.08 | $243.78 (102.13) |
$250.53 (140.79) |
0.25 | 201.80 (29.38) |
178.07 (20.04) |
0.84 |
| 14th Pre | $1,236.20 (168.65) |
$1,418.35 (190.08 ) |
0.91 | $44.85 (13.41) |
$137.05 (38.24) |
0.98 | $658.99 (53.29) |
$654.70 (62.33) |
0.71 | $127.59 (95.83) |
$177.44 (102.95) |
0.61 | 195.37 (28.82) |
188.44 (20.45) |
0.73 |
| 13th Pre | $1,574.33 (182.21) |
$1,415.46 (194.10 ) |
0.03 | $61.89 (16.05) |
$144.55 (41.52) |
0.84 | $684.88 (51.56) |
$646.70 (59.67) |
0.25 | $287.90 (106.26) |
$229.49 (108.73) |
0.18 | 211.38 (29.16) |
182.60 (20.40) |
0.27 |
| 12th Pre | $1,359.68 (168.81) |
$1,478.02 (213.40 ) |
0.75 | $64.51 (14.65) |
$139.14 (43.01) |
0.68 | $651.58 (50.30) |
$639.52 (61.58) |
0.57 | $196.60 (97.04) |
$316.08 (136.42) |
0.32 | 202.78 (29.02) |
183.56 (20.42) |
0.39 |
| 11th Pre | $1,418.10 (170.94) |
$1,557.14 (214.54 ) |
0.86 | $77.63 (15.99) |
$164.97 (50.16) |
0.93 | $664.00 (51.48) |
$672.95 (58.83) |
0.92 | $224.78 (98.19) |
$261.20 (139.13) |
0.62 | 203.12 (28.99) |
196.58 (20.67) |
0.95 |
| 10th Pre | $1,370.94 (173.89) |
$1,290.30 (180.19 ) |
0.02 | $73.82 (16.74) |
$161.22 (46.54) |
0.93 | $622.34 (50.14) |
$640.52 (55.75) |
0.90 | $218.42 (102.26) |
$139.14 (100.97) |
0.65 | 190.22 (19.72) |
203.04 (21.57) |
0.07 |
| 9th Pre | $1,386.92 (171.49) |
$1,519.74 (200.85 ) |
0.80 | $62.15 (14.37) |
$163.30 (47.66) |
0.82 | $669.72 (53.03) |
$651.50 (58.25) |
0.47 | $193.68 (99.02) |
$290.46 (120.11) |
0.90 | 190.28 (19.53) |
194.93 (20.86) |
0.43 |
| 8th Pre | $1,492.47 (174.73) |
$1,595.42 (208.89 ) |
0.66 | $81.56 (15.59) |
$234.53 (48.43) |
0.23 | $692.04 (53.82) |
$606.36 (57.39) |
0.02 | $279.93 (103.04) |
$320.31 (125.05) |
0.12 | 198.91 (19.67) |
186.26 (20.01) |
0.91 |
| 7th Pre | $1,536.08 (175.37) |
$1,463.05 (184.37 ) |
0.04 | $89.30 (15.60) |
$197.04 (52.94) |
0.73 | $705.24 (53.12) |
$654.37 (55.92) |
0.11 | $298.15 (105.12) |
$113.25 (103.29) |
0.45 | 194.80 (19.65) |
211.33 (20.77) |
0.00 |
| 6th Pre | $1,353.98 (167.11) |
$1,726.70 (207.56 ) |
0.15 | $65.04 (8.55) |
$211.21 (29.17) |
0.26 | $696.30 (50.69) |
$711.04 (61.41) |
0.97 | $165.91 (96.77) |
$273.35 (113.26) |
0.52 | 192.09 (19.27) |
208.58 (22.69) |
0.29 |
| 5th Pre | $1,433.58 (170.21) |
$1,757.59 (207.29 ) |
0.29 | $76.97 (14.45) |
$221.20 (55.84) |
0.29 | $681.26 (50.56) |
$683.25 (64.43) |
0.81 | $218.51 (98.20) |
$245.90 (108.47) |
0.87 | 198.55 (19.70) |
207.05 (21.09) |
0.97 |
| 4th Pre | $1,488.03 (179.30) |
$1,651.99 (196.84 ) |
0.98 | $74.09 (13.03) |
$249.12 (42.68) |
0.12 | $687.60 (51.28) |
$692.04 (62.46) |
0.84 | $244.34 (108.58) |
$189.92 (103.83) |
0.84 | 217.83 (23.57) |
220.19 (21.61) |
0.23 |
| 3rd Pre | $1,403.74 (173.13) |
$2,300.34 (337.02 ) |
0.01 | $67.92 (15.30) |
$283.27 (60.28) |
0.03 | $683.83 (50.89) |
$720.33 (64.92) |
0.65 | $187.49 (99.32) |
$531.30 (205.55) |
0.28 | 196.22 (19.29) |
351.95 (146.11) |
0.07 |
| 2nd Pre | $1,497.75 (180.99) |
$2,223.34 (261.98 ) |
0.01 | $72.38 (15.27) |
$319.52 (42.46) |
0.01 | $699.07 (52.22) |
$807.76 (72.94) |
0.10 | $272.09 (115.45) |
$463.62 (158.56) |
0.26 | 195.03 (19.23) |
228.21 (28.74) |
0.25 |
| 1st Pre | $1,510.22 (175.47) |
$2,243.56 (230.68 ) |
0.00 | $74.22 (14.84) |
$411.58 (67.86) |
0.00 | $693.64 (51.87) |
$740.19 (60.08) |
0.47 | $224.76 (98.76) |
$439.38 (129.61) |
0.06 | 198.62 (19.32) |
268.98 (39.58) |
0.06 |
| ADRD | $1,475.42 (184.77) |
$2,237.92 (222.27 ) |
0.22 | $75.40 (15.86) |
$394.09 (64.82) |
0.00 | $698.61 (52.69) |
$752.24 (62.37) |
0.89 | $178.14 (111.55) |
$532.90 (125.24) |
0.17 | 211.40 (19.34) |
237.10 (36.53) |
0.83 |
| 1st Post | $1,537.74 (171.37) |
$2,241.98 (243.25 ) |
0.00 | $70.97 (26.14) |
$430.74 (84.14) |
0.00 | $669.36 (51.64) |
$690.12 (65.69) |
0.45 | $259.40 (96.04) |
$572.01 (153.00) |
0.19 | 203.06 (19.71) |
229.30 (21.84) |
0.01 |
| 2nd Post | $1,657.61 (177.85) |
$2,269.66 (281.55 ) |
0.03 | $89.27 (15.55) |
$415.33 (69.86) |
0.00 | $698.99 (50.79) |
$727.86 (58.98) |
0.95 | $256.92 (107.04) |
$434.73 (211.79) |
0.24 | 255.63 (21.22) |
234.45 (21.64) |
0.16 |
| 3rd Post | $1,625.31 (184.41) |
$2,025.32 (228.94 ) |
0.02 | $106.99 (27.52) |
$448.24 (96.91) |
0.00 | $641.02 (52.64) |
$647.64 (60.88) |
0.86 | $319.42 (101.31) |
$368.26 (131.73) |
0.57 | 198.46 (52.18) |
249.20 (21.72) |
0.16 |
| 4th Post | $1,744.72 (178.28) |
$2,408.89 (220.88 ) |
0.19 | $125.34 (17.71) |
$500.31 (67.11) |
0.00 | $685.31 (51.02) |
$688.06 (58.33) |
0.66 | $338.27 (105.66) |
$498.85 (127.04) |
0.00 | 206.34 (19.39) |
240.71 (22.12) |
0.86 |
| 5th Post | $1,605.16 (185.84) |
$2,366.59 (271.22 ) |
0.04 | $118.17 (21.04) |
$514.89 (88.13) |
0.00 | $643.14 (51.77) |
$747.24 (61.76) |
0.58 | $257.92 (116.36) |
$411.62 (157.43) |
0.12 | 210.63 (19.32) |
251.56 (23.70) |
0.38 |
| 6th Post | $1,742.78 (177.68) |
$2,413.01 (228.64 ) |
0.00 | $125.18 (18.94) |
$489.07 (71.51) |
<.0001 | $707.36 (51.43) |
$721.69 (62.09) |
0.10 | $308.38 (103.90) |
$485.53 (124.96) |
0.04 | 220.64 (19.77) |
243.41 (22.73) |
0.22 |
| 7th Post | $1,771.18 (180.34) |
$2,540.84 (313.49 ) |
0.08 | $151.06 (20.94) |
$600.29 (74.11) |
0.00 | $676.03 (52.75) |
$675.65 (75.44) |
0.75 | $351.66 (104.26) |
$504.46 (204.95) |
0.38 | 202.44 (21.14) |
246.62 (21.92) |
0.45 |
| 8th Post | $1,755.65 (180.80) |
$2,270.62 (247.87 ) |
0.00 | $161.67 (20.04) |
$609.87 (72.78) |
<.0001 | $643.49 (52.20) |
$596.67 (63.85) |
0.44 | $309.85 (106.11) |
$392.52 (129.75) |
0.02 | 221.17 (19.31) |
257.25 (22.43) |
0.27 |
| 9th Post | $2,000.53 (181.23) |
$2,563.87 (238.34 ) |
0.09 | $179.50 (20.33) |
$656.11 (73.42) |
<.0001 | $697.90 (51.03) |
$669.94 (60.14) |
0.04 | $399.76 (108.52) |
$637.67 (133.95) |
0.07 | 218.66 (19.66) |
250.31 (22.20) |
0.66 |
| 10th Post | $2,047.14 (187.52) |
$2,264.31 (239.42 ) |
0.06 | $183.93 (23.63) |
$709.85 (79.18) |
<.0001 | $707.63 (53.02) |
$614.32 (60.77) |
0.12 | $433.21 (110.36) |
$347.36 (146.44) |
0.19 | 222.25 (20.99) |
250.65 (22.48) |
0.11 |
| 11th Post | $2,328.79 (201.68) |
$2,392.02 (216.27 ) |
0.84 | $214.16 (24.82) |
$691.11 (83.54) |
<.0001 | $727.76 (55.69) |
$576.51 (57.94) |
0.00 | $570.23 (123.71) |
$343.17 (117.21) |
0.20 | 238.54 (20.05) |
256.39 (21.95) |
0.29 |
| 12th Post | $2,179.18 (207.11) |
$2,372.52 (248.06 ) |
0.63 | $213.23 (25.91) |
$711.94 (88.41) |
<.0001 | $711.59 (54.97) |
$609.47 (57.85) |
<.0001 | $416.58 (134.00) |
$247.88 (142.84) |
0.67 | 276.10 (22.59) |
266.97 (23.62) |
0.09 |
| 13th Post | $2,196.59 (195.03) |
$2,621.82 (225.97 ) |
0.94 | $208.19 (26.81) |
$735.68 (87.62) |
<.0001 | $688.85 (55.83) |
$653.33 (58.70) |
0.00 | $500.01 (109.65) |
$417.37 (106.61) |
0.71 | 274.98 (52.12) |
305.06 (23.57) |
0.02 |
| 14th Post | $2,525.21 (201.27) |
$2,377.36 (238.05 ) |
0.26 | $243.32 (27.57) |
$736.10 (89.57) |
<.0001 | $707.59 (56.78) |
$588.87 (62.19) |
0.02 | $653.64 (119.18) |
$305.88 (131.22) |
0.73 | 237.58 (47.83) |
275.21 (44.82) |
0.35 |
| 15th Post | $1,988.75 (207.73) |
$2,474.77 (220.52 ) |
0.10 | $210.07 (27.15) |
$782.75 (90.25) |
<.0001 | $597.46 (60.78) |
$562.10 (60.01) |
<.0001 | $345.03 (125.04) |
$248.98 (110.85) |
0.09 | 220.93 (20.25) |
268.08 (22.64) |
0.00 |
| 16th Post | $2,418.41 (186.90) |
$2,834.96 (240.26 ) |
0.14 | $178.80 (31.12) |
$809.00 (90.37) |
<.0001 | $636.92 (54.86) |
$585.72 (60.07) |
0.01 | $616.42 (107.71) |
$552.12 (109.13) |
0.01 | 232.13 (19.90) |
280.59 (21.91) |
0.13 |
| 17th Post | $2,113.30 (249.99) |
$2,647.44 (293.95 ) |
0.42 | $175.52 (27.19) |
$799.83 (91.86) |
<.0001 | $609.58 (68.11) |
$577.21 (60.00) |
0.00 | $389.32 (138.87) |
$296.24 (202.59) |
0.02 | 230.75 (20.15) |
277.52 (22.47) |
0.63 |
| 18th Post | $2,378.50 (209.58) |
$2,965.27 (251.71 ) |
0.13 | $179.70 (25.22) |
$819.83 (96.13) |
<.0001 | $610.46 (60.88) |
$473.46 (60.16) |
0.00 | $710.32 (128.42) |
$673.83 (109.58) |
0.03 | 225.63 (20.13) |
286.50 (22.87) |
0.24 |
| 19th Post | $2,206.95 (231.09) |
$2,655.40 (338.10 ) |
0.23 | $198.42 (25.43) |
$830.24 (97.01) |
<.0001 | $538.59 (65.67) |
$484.73 (56.91) |
<.0001 | $533.29 (154.54) |
$497.66 (244.24) |
0.00 | 236.97 (20.31) |
273.26 (23.28) |
0.79 |
| 20th Post | $2,360.61 (200.50) |
$2,990.30 (259.89 ) |
0.26 | $167.65 (25.11) |
$816.50 (96.65) |
<.0001 | $536.97 (58.07) |
$474.68 (60.11) |
0.00 | $646.40 (118.53) |
$648.42 (156.07) |
0.10 | 250.36 (20.58) |
306.72 (21.71) |
0.64 |
| 21st Post | $2,141.65 (240.32) |
$2,965.05 (308.75 ) |
0.16 | $161.89 (24.68) |
$852.74 (94.33) |
<.0001 | $474.79 (60.64) |
$486.63 (56.32) |
<.0001 | $543.38 (177.45) |
$554.48 (213.20) |
0.02 | 245.90 (21.72) |
297.09 (24.48) |
0.90 |
| 22nd Post | $1,915.14 (229.24) |
$2,673.30 (299.40 ) |
0.04 | $145.01 (24.32) |
$830.66 (98.90) |
<.0001 | $465.42 (55.34) |
$459.37 (57.78) |
0.01 | $330.08 (160.62) |
$408.82 (196.03) |
0.06 | 234.42 (23.63) |
285.80 (23.03) |
0.92 |
| 23rd Post | $1,990.49 (194.35) |
$2,825.95 (257.36 ) |
0.01 | $130.83 (23.30) |
$801.50 (97.45) |
<.0001 | $478.38 (55.95) |
$464.06 (60.11) |
0.00 | $425.30 (109.85) |
$509.88 (148.26) |
0.02 | 242.63 (20.94) |
301.93 (22.68) |
0.73 |
| 24th Post | $1,901.11 (197.26) |
$2,665.97 (275.50 ) |
0.01 | $156.90 (21.68) |
$744.85 (96.83) |
<.0001 | $449.08 (56.94) |
$417.08 (61.07) |
0.00 | $347.14 (115.16) |
$376.57 (164.01) |
0.01 | 225.17 (20.69) |
329.52 (23.53) |
0.73 |
| 25th Post | $1,808.27 (186.91) |
$2,708.90 (258.36 ) |
0.01 | $136.30 (27.28) |
$734.43 (92.20) |
<.0001 | $404.80 (54.64) |
$420.61 (58.06) |
<.0001 | $321.86 (106.92) |
$465.90 (134.57) |
0.00 | 236.83 (20.68) |
345.97 (33.41) |
0.98 |
| 26th Post | $2,045.15 (188.24) |
$2,986.50 (271.23 ) |
0.00 | $137.25 (24.10) |
$739.85 (89.95) |
<.0001 | $432.62 (53.04) |
$443.00 (59.59) |
0.02 | $352.16 (111.15) |
$651.82 (160.78) |
0.00 | 231.76 (20.71) |
343.40 (35.17) |
0.45 |
| 27th Post | $2,108.39 (196.07) |
$2,533.84 (294.32 ) |
0.01 | $135.47 (24.01) |
$741.51 (90.33) |
<.0001 | $438.98 (54.70) |
$389.82 (60.65) |
0.01 | $431.81 (110.33) |
$294.90 (190.06) |
0.01 | 232.87 (20.60) |
333.86 (42.16) |
0.14 |
| 28th Post | $2,048.85 (198.64) |
$2,800.68 (249.56 ) |
0.29 | $134.55 (22.92) |
$745.68 (91.36) |
<.0001 | $446.54 (54.25) |
$412.47 (61.47) |
<.0001 | $368.71 (117.39) |
$520.51 (119.38) |
0.02 | 228.60 (20.91) |
322.65 (41.92) |
0.11 |
| 29th Post | $2,275.77 (201.68) |
$2,740.28 (262.18 ) |
0.02 | $114.77 (23.27) |
$679.44 (91.17) |
<.0001 | $433.55 (57.05) |
$402.79 (63.34) |
0.00 | $502.40 (122.40) |
$254.73 (141.98) |
<.0001 | 278.33 (20.92) |
336.66 (24.98) |
0.38 |
| 30th Post | $2,166.92 (239.15) |
$2,664.09 (269.25 ) |
0.32 | $90.79 (18.75) |
$666.95 (86.82) |
<.0001 | $467.16 (58.33) |
$388.13 (59.13) |
<.0001 | $487.21 (146.46) |
$463.51 (123.79) |
0.19 | 248.97 (36.04) |
326.40 (28.98) |
0.05 |
| 31st Post | $1,986.56 (208.44) |
$2,910.20 (250.47 ) |
0.19 | $88.16 (18.32) |
$692.36 (86.92) |
<.0001 | $392.28 (68.18) |
$399.43 (67.57) |
<.0001 | $410.97 (126.61) |
$713.45 (138.76) |
0.02 | 336.71 (30.42) |
323.90 (25.49) |
0.68 |
| 32nd Post | $2,232.37 (245.71) |
$2,893.09 (300.54 ) |
0.02 | $96.97 (19.31) |
$686.11 (86.29) |
<.0001 | $382.59 (58.37) |
$389.21 (65.12) |
0.02 | $339.90 (159.06) |
$401.15 (218.17) |
0.78 | 546.69 (89.72) |
328.71 (23.51) |
0.24 |
| 33rd Post | $2,029.49 (347.20) |
$2,791.94 (259.70 ) |
0.21 | $95.11 (18.99) |
$764.01 (93.40) |
<.0001 | $360.46 (58.20) |
$336.83 (64.00) |
0.01 | $421.95 (113.94) |
$418.39 (125.70) |
0.43 | 334.33 (281.68) |
332.82 (24.65) |
0.80 |
| 34th Post | $1,876.72 (220.60) |
$2,993.30 (287.54 ) |
0.04 | $94.99 (20.36) |
$731.10 (90.98) |
<.0001 | $354.47 (56.93) |
$358.36 (62.18) |
0.00 | $278.00 (127.12) |
$417.76 (157.04) |
0.85 | 282.84 (78.56) |
372.33 (24.10) |
0.82 |
| 35th Post | $1,910.34 (204.63) |
$2,721.69 (280.39)) |
0.00 | $96.89 (20.01)) |
$753.18 (94.55 |
<.0001 | $329.16 (59.04 |
$332.15 (64.59) |
0.02 | $397.41 (119.74) |
$338.51 (134.39) |
0.06 | 248.12 (33.32) |
322.40 (34.09) |
0.39 |
Total monthly costs in the two and a half years prior to an ADRD diagnosis are similar for both cases and controls with values of $1,439 (SD 886) and $1,314 (766) respectively and a significant (p <0.05) difference between groups exists in only two months - the 30th and 24th months pre-diagnosis. Beginning in the 6th month pre diagnosis, total monthly health care costs among cases are higher than those seen in controls with this difference achieving statistical significance (p < 0.05) three months before diagnosis and remaining significant through the eighth month post diagnosis during which mean monthly health care costs among cases are $2,271 (886) and $1,595 (762) for controls. After several months of higher, but statistically insignificant higher costs among cases, mean monthly costs among cases are significantly higher relative to controls for 9 months in the third year following diagnosis. For the entire three-year post diagnosis period, individuals with ADRD have mean monthly costs of $2,589 (914) relative to $1,984 (798) among controls.
Figure 2 reveals that nursing home costs are the primary driver of the total cost difference between cases and controls following an ADRD diagnosis. Mean monthly nursing home costs are greater among cases for the two and a half years prior to an ADRD diagnosis [$151 (238) vs $46 (187)] but this difference is not statistically significant until the 3rd month pre-diagnosis. From this point through the remainder of the three-year post diagnosis follow up period, individuals with ADRD have significantly (p < 0.05) greater monthly mean nursing home costs [$677 (263) vs $144 (191)].
Monthly health care costs for ambulatory pharmacy, outpatient and inpatient costs are reported in Panels A, B and C of Figure 2. Monthly ambulatory pharmacy costs increase for cases and controls over the six-year study period but there is no immediate impact of an ADRD diagnosis on this trend. Outpatient pharmacy costs increase at a faster rate among cases relative to controls, which leads to several months in which these differences are statistically significant (p < 0.05). We note several months in which outlier costs among both cases and controls result in spikes in group means, which is an artifact of the relatively small sample included in our analyses. The large variance in costs does not lead to these differences being statistically significant.
Outpatient costs reported in Figure 2, Panel B are similar for cases and controls in the pre-diagnostic period. In the post-diagnostic period, mean monthly costs among cases and controls are similar, averaging $670 and $664 respectively over this three-year period. In the years following the ADRD diagnosis costs for both groups fall absolutely but at a greater relative rate among cases. This leads to greater costs among controls achieving statistical significance (p < 0.05) for several months in the second-year post diagnosis. Lower outpatient costs for both cases and controls over time mirrors the increase in nursing home costs among both groups.
Inpatient costs increase over time for both groups with significant (p < .05) differences by group present with greater costs among controls in the 6th month pre-diagnosis and among cases in the 1st month before and 1st month after ADRD was diagnosed. As shown in Panel C, there is a substantial variation in inpatient costs both over time and between groups, which contributes to the relatively few time points for which differences between cases and controls achieve statistical significance.
Conclusion
We estimated total, inpatient, outpatient, ambulatory pharmacy and nursing home costs for older adults for three years before and after an incident ADRD diagnosis relative to a group of older adults without an ADRD diagnosis receiving care from an MA plan. We found that significantly greater total health care costs for older adults with ADRD begins in the 3 months prior to diagnosis and remains significantly greater for each month in the first post-diagnosis year. The trend in greater mean monthly costs among older adults with ADRD relative to controls continues for the second- and third-year post-diagnosis although these differences are not statistically significant in the second post-diagnosis year. Our finding of greater ADRD attributable costs appears driven by the greater use of nursing homes.
Variation in study design, populations, and cost components included in previous studies reporting costs of ADRD within Medicare managed care plans makes it difficult to draw direct comparisons with our findings. Based on our review of the literature, only two studies17,22 included skilled nursing care in their cost estimates, which we found to the be the driver of greater ADRD attributable costs. The two analyses of incident cost in the year following diagnosis ranged from $38,79420 to $8,938,21 neither of which included skilled nursing care and are still substantially greater than our annualized costs of $6,703.
We considered whether higher mortality rates among controls in the first two years post diagnosis help explain the insignificant differences in cost between cases and controls we found in the period one year after ADRD onset. We re-estimated our models requiring all subjects to survive for at least two years following onset and found that the relative cost increase among controls was no longer present and higher costs among cases were significant for each month in the three-year post onset period. This sensitivity analysis highlights the need for more research on the intersection of mortality and ADRD on health care costs.
We did not find greater relative inpatient costs among cases with the exception of the months immediately before and after ADRD was diagnosed. Although prior research has found greater hospitalizations use among older adults with ADRD, our results may reflect lower inpatient utilization within MA plans30 with individuals more likely to receive institutional care in nursing homes.
This work adds to the limited literature on the cost of ADRD within Medicare managed care in several ways. First, the existing literature relies primarily on diagnosis or pharmacy data from claims to identify ADRD cases, which may lead to inaccurate estimates of the number of older adults with ADRD. In contrast, ADRD diagnoses for our analyses are based on the multidisciplinary ACT team’s rigorous clinical assessment of cognitive impairment. Second, our analyses use data from an integrated health care delivery and finance system that includes a more complete set of health service use and cost data than is available from other managed care plans. Specifically, we are able to include information about outpatient pharmacy use prior to the introduction of Medicare Part D as well as skilled nursing services beyond the post-acute care period available through the Medicare FFS program. We therefore demonstrate the critical role that nursing home care plays in the cost of older adults with ADRD, with the potential to inform policy regarding the need to reconsider health care resource allocation to address the growing prevalence of ADRD. Finally, we examined health care costs for three years before and after an ADRD diagnosis and provide more detail regarding the trajectory of monthly costs over time attributable to ADRD.
Our study is limited by its reliance on data from a single managed care plan and participants in a single research study. We do not know if KPWA is representative of the wider MA program, but ACT participants are more educated and less racially diverse than the US population. Thus, our results may not be generalizable to other MA plans. KPWA’s finance and care delivery model does create incentives for improved care coordination relative to other MA plans but for the entire period we studied KPWA did not have any ADRD interventions or protocols and there was no coordination between the ACT study and KP providers. Future research should use data from multiple managed care plans to support tests of whether there are cost differences related to care delivery among the types of managed care organizations. CMS’s efforts to make data for older adults within the MA program available for research will allow future studies to draw from a wider range of health care settings, though there may still be challenges with respect to valid identification of ADRD in those data.
The challenge of providing appropriate resources to care for the large and growing population with ADRD requires robust estimates of ADRD attributable health care costs in all of the settings in which older adults receive health care and health services. Enrollment growth within the MA program requires that analyses of the economic impact of ADRD focus on the role that managed care plans have on the overall cost of care. Previous research on the economics of cancer has reported differences in direct medical care costs for otherwise comparable populations receiving care from managed care relative to FFS providers.13,14 We contribute to the limited research on the impact of ADRD within MA plans, which is a first step to comparative analyses of the cost of ADRD care between Medicare FFS and MA plans. Our current over-reliance on FFS estimates of the cost of ADRD care may result in incorrect projections and risk a misallocation of resources required to deal with a large and growing burden on the American health care system. Our study highlights the need for greater research specifically within managed care plans and for studies that directly compare costs between Medicare FFS and MA plans.
Acknowledgments
Research supported by:
The Centers for Disease Control and Prevention (Direct and Indirect Costs of Dementia Care, SIP 14-005, Coe, N. PI) and The National Institute on Aging, Current and Future Costs of Alzheimer’s and Dementia Care, AG049815, Coe, N, PI Adult Changes in Thought Study, AG006781, Larson, E. PI
Footnotes
None of the authors have any conflict of interest, financial or otherwise, associated with this research.
Literature Cited
- 1.Hebert LE, Scherr PA, Bienias JL, Bennett DA, Evans DA. Alzheimer disease in the US population: prevalence estimates using the 2000 census. Arch Neurol. 2003;60(8):1119–1122. [DOI] [PubMed] [Google Scholar]
- 2.Deb A, Thornton JD, Sambamoorthi U, Innes K. Direct and indirect cost of managing alzheimer’s disease and related dementias in the United States. Expert Review of Pharmacoeconomics & Outcomes Research. 2017;17(2):189–202. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 3.Taylor DH Jr., Sloan FA How much do persons with Alzheimer’s disease cost Medicare? J Am Geriatr Soc. 2000;48(6):639–646. [DOI] [PubMed] [Google Scholar]
- 4.Hurd MD, Martorell P, Delavande A, Mullen KJ, Langa KM. Monetary costs of dementia in the United States. N Engl J Med. 2013;368(14):1326–1334. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 5.Neuman P, Jacobson GA. Medicare Advantage Checkup. N Engl J Med. 2018;379(22):2163–2172. [DOI] [PubMed] [Google Scholar]
- 6.Park S, CN, Basu A, Khalil F. Service-level selection: strategic risk selection in Medicare Advantage in response to risk adjustment. . NBER Working Paper. 2017;No. 24038. [Google Scholar]
- 7.McGuire TG, Newhouse JP, Sinaiko AD. An economic history of Medicare part C. Milbank Q. 2011;89(2):289–332. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 8.Jacobson GA, Neuman P, Damico A. At least half of new Medicare advantage enrollees had switched from traditional Medicare during 2006–11. Health Aff (Millwood). 2015;34(1):48–55. [DOI] [PubMed] [Google Scholar]
- 9.Park S FP, White L, Larson EB, Coe NB. . Disease-specific plan switching between traditional Medicare and Medicare Advantage. The Permanente Journal.Forthcoming. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 10.Daras LC, Feng Z, Wiener JM, Kaganova Y. Medicare Expenditures Associated With Hospital and Emergency Department Use Among Beneficiaries With Dementia. Inquiry. 2017;54:46958017696757. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 11.Lin PJ, Zhong Y, Fillit HM, Chen E, Neumann PJ. Medicare Expenditures of Individuals with Alzheimer’s Disease and Related Dementias or Mild Cognitive Impairment Before and After Diagnosis. J Am Geriatr Soc. 2016;64(8):1549–1557. [DOI] [PubMed] [Google Scholar]
- 12.2018 Alzheimer’s disease facts and figures. Alzheimer’s & Dementia. 2018;14(3):367–429. [Google Scholar]
- 13.Fishman PA, Hornbrook MC, Ritzwoller DP, O’Keeffe-Rosetti MC, Lafata JE, Salloum RG. The challenge of conducting comparative effectiveness research in cancer: the impact of a fragmented U.S. health-care system. J Natl Cancer Inst Monogr. 2013;2013(46):99–105. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 14.Banegas MP, Yabroff KR, O’Keeffe-Rosetti MC, et al. Medical Care Costs Associated With Cancer in Integrated Delivery Systems. J Natl Compr Canc Netw. 2018;16(4):402–410. [DOI] [PubMed] [Google Scholar]
- 15.Fishman P, Coe NB, White L, et al. Cost of dementia in Medicare managed care: a systematic literature review. Am J Manag Care. 2019;25(8):e247–e253. [PMC free article] [PubMed] [Google Scholar]
- 16.Frytak JR, Henk HJ, Zhao Y, Bowman L, Flynn JA, Nelson M. Health service utilization among Alzheimer’s disease patients: evidence from managed care. Alzheimers Dement. 2008;4(5):361–367. [DOI] [PubMed] [Google Scholar]
- 17.Richards KM, Shepherd MD, Crismon ML, Snyder EH, Jermain DM. Medical services utilization and charge comparisons between elderly patients with and without Alzheimer’s disease in a managed care organization. Clin Ther. 2000;22(6):775–791. [DOI] [PubMed] [Google Scholar]
- 18.Fillit H, Hill J. The costs of vascular dementia: a comparison with Alzheimer’s disease. J Neurol Sci. 2002;203-204:35–39. [DOI] [PubMed] [Google Scholar]
- 19.Gutterman EM, Markowitz JS, Lewis B, Fillit H. Cost of Alzheimer’s disease and related dementia in managed-medicare. J Am Geriatr Soc. 1999;47(9):1065–1071. [DOI] [PubMed] [Google Scholar]
- 20.Joyce AT, Zhao Y, Bowman L, Lb JAF, Carter CT, Ollendorf DA. Burden of illness among commercially insured patients with Alzheimer’s disease. Alzheimers & Dementia. 2007;3(3):204–210. [DOI] [PubMed] [Google Scholar]
- 21.Suehs BT, Davis CD, Alvir J, et al. The Clinical and Economic Burden of Newly Diagnosed Alzheimer’s Disease in a Medicare Advantage Population. Am J Alzheimers Dis. 2013;28(4):384–392. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 22.Hill J, Fillit H, Shah SN, del Valle MC, Futterman R. Patterns of healthcare utilization and costs for vascular dementia in a community-dwelling population. J Alzheimers Dis. 2005;8(1):43–50. [DOI] [PubMed] [Google Scholar]
- 23.Ritzwoller DP, Goodman MJ, Maciosek MV, et al. Creating standard cost measures across integrated health care delivery systems. J Natl Cancer Inst Monogr. 2005(35):80–87. [DOI] [PubMed] [Google Scholar]
- 24.O’Keeffe-Rosetti MC, Hornbrook MC, Fishman PA, et al. A standardized relative resource cost model for medical care: application to cancer control programs. J Natl Cancer Inst Monogr. 2013;2013(46):106–116. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 25.Mullahy J Econometric modeling of health care costs and expenditures: a survey of analytical issues and related policy considerations. Med Care. 2009;47(7 Suppl 1):S104–108. [DOI] [PubMed] [Google Scholar]
- 26.Blough DK, Madden CW, Hornbrook MC. Modeling risk using generalized linear models. J Health Econ. 1999;18(2):153–171. [DOI] [PubMed] [Google Scholar]
- 27.Manning WG, Basu A, Mullahy J. Generalized modeling approaches to risk adjustment of skewed outcomes data. J Health Econ. 2005;24(3):465–488. [DOI] [PubMed] [Google Scholar]
- 28.Diehr P, Yanez D, Ash A, Hornbrook M, Lin DY. Methods for analyzing health care utilization and costs. Annu Rev Public Health. 1999;20:125–144. [DOI] [PubMed] [Google Scholar]
- 29.Huber PJ, Ronchetti E. Robust statistics. 2nd ed. Hoboken, N.J.: Wiley; 2009. [Google Scholar]
- 30.Raetzman SO, Hines Anika L., Barrett Marguerite L., Karaca Zeynal. Hospital Stays in Medicare Advantage Plans Versus the Traditional Medicare Fee-for-Service Program, 2013. In: HCUP, ed. Vol Statistical Brief # 198. Rockville, MD: Agency for Health Care Rersearch and Quality; 2015:1–18. [PubMed] [Google Scholar]