Medication Fill Patterns for Cognitive and Behavioral or Psychological Symptoms of Alzheimer Disease and Related Dementias.

Introduction

Alzheimer disease and related dementias (ADRDs) impact cognition and function among nearly 5 million US adults. Symptoms include changes in memory, decision-making, behavior, mood, depression, and anxiety and are often managed pharmacologically. This study describes prescription drug fills for cognitive and behavioral symptoms of ADRD among Medicare Advantage Part D (MAPD) enrollees.

Methods

This cohort study was determined to be exempt from institutional review board review and informed consent by the UnitedHealth Group Office of Human Research Affairs because it uses deidentified data. The study followed the STROBE reporting guideline for cohort studies.

We used OptumLabs deidentified administrative medical and pharmacy claims data. Structured data reflect International Statistical Classification of Diseases, Tenth Revision, Clinical Modification (ICD-10-CM) codes; Current Procedural Terminology codes; Logical Observation Identifiers Names and Codes; and National Drug Code claims. We identified adults with available demographics enrolled in MAPD and aged 65 years and older at ADRD diagnosis. We required more than 1 claim with a primary, secondary, or tertiary ADRD code between January 2016 and October 2021 and 2 or more ADRD codes during the study period. We required 12 or more months of enrollment before and after ADRD diagnosis and excluded individuals living in long-term care facilities.

To assess fills for cognitive symptoms, we focused on Food and Drug Administration–approved drugs for ADRD: acetylcholinesterase inhibitors and N-methyl-d-aspartate (NMDA) receptor antagonists.[1] To assess behavioral and psychological symptoms of dementia (BPSD) fills, we focused on antipsychotics, mood stabilizers and anticonvulsants, antidepressants, sedative-hypnotic Z-drugs, and benzodiazepine classes.[2] We included outpatient fills through the Part D benefit with 30 or more days of cumulative supply during years before and after diagnosis.

Results

The study population of 161 368 individuals (median age, 83 years; 62.2% women; 12.3% Black individuals, 2.8% Hispanic individuals, and 79.2% White individuals) was distributed across the US (Table 1). Acetylcholinesterase inhibitor fills increased from 15.3% in the prediagnosis year to 35.5% postdiagnosis, with most fills among individuals with Alzheimer disease (53.0%). NMDA receptor antagonist fills increased from 5.8% to 16.6%. Antipsychotic fills increased from 5.6% to 12.8% and were prevalent among individuals with Lewy body dementia (12.2% before and 29.7% after diagnosis). Mood stabilizers and anticonvulsants fills (15.8% before and 18.2% after diagnosis) were highest in individuals with Parkinson disease (24.0%), Lewy body (20.1%), or vascular dementias (21.2%). Antidepressants were the most common BPSD drugs (19.4% before and 24.2% after diagnosis), with the greatest prevalence in individuals with frontotemporal (21.1% before and 30.6% after diagnosis) or Lewy body dementias (23.4% before and 28.9% after diagnosis). Benzodiazepines (13.0% before and 13.6% after diagnosis) and sedative-hypnotic Z-drugs (1.9% before and 1.8% after diagnosis) remained stable (Table 2).

Table 1.

Study Population Characteristics by Specific Diagnosis

Characteristic	Patients, No. (%)a
	Total (N = 161 368)	Alzheimer 33 141 (20.5)	Dementia 76 634 (47.5)	Frontotemporal dementia 866 (0.5)	Lewy body 1858 (1.2)	MCI 24 138 (15.0)	Parkinson 12 799 (7.9)	Vascular dementia 11 932 (7.4)
Age, y
Median (IQR)	83 (77-89)	84 (79-90)	85 (79-90)	79 (74-84)	81 (76-86)	79 (74-85)	81 (75-87)	83 (77-89)
65-75	30 317 (18.8)	4500 (13.6)	11 356 (14.8)	315 (36.4)	451 (24.3)	7766 (32.2)	3503 (27.4)	2426 (20.3)
76-85	65 877 (40.8)	13 772 (41.6)	29 518 (38.5)	371 (42.8)	921 (49.6)	10 973 (45.5)	5488 (42.9)	4834 (40.5)
≥86	65 174 (40.4)	14 869 (44.9)	35 760 (46.7)	180 (20.8)	486 (26.2)	5399 (22.4)	3808 (29.8)	4672 (39.2)
Sex
Men	61 029 (37.8)	11 523 (34.8)	27 691 (36.1)	391 (45.2)	1019 (54.8)	10 006 (41.5)	5655 (44.2)	4744 (39.8)
Women	100 339 (62.2)	21 618 (65.2)	48 943 (63.9)	475 (54.8)	839 (45.2)	14 132 (58.5)	7144 (55.8)	7188 (60.2)
US Census region
Midwest	26 847 (16.6)	5834 (17.6)	11 938 (15.6)	165 (19.1)	347 (18.7)	3795 (15.7)	2652 (20.7)	2116 (17.7)
Northeast	21 180 (13.1)	3884 (11.7)	10 377 (13.5)	106 (12.2)	229 (12.3)	3057 (12.7)	1533 (12.0)	1994 (16.7)
South	68 980 (42.7)	14 026 (42.3)	33 167 (43.3)	350 (40.4)	751 (40.4)	10 189 (42.2)	5320 (41.6)	5177 (43.4)
West	44 361 (27.5)	9397 (28.4)	21 152 (27.6)	245 (28.3)	531 (28.6)	7097 (29.4)	3294 (25.7)	2645 (22.2)
Residential setting
Urban	64 150 (39.8)	13 547 (40.9)	30 694 (40.1)	316 (36.5)	675 (36.3)	9741 (40.4)	4426 (34.6)	4751 (39.8)
Suburban	58 496 (36.3)	12 097 (36.5)	26 971 (35.2)	350 (40.4)	709 (38.2)	9140 (37.9)	4831 (37.7)	4398 (36.9)
Rural	38 722 (24.0)	7497 (22.6)	18 969 (24.8)	200 (23.1)	474 (25.5)	5257 (21.8)	3542 (27.7)	2783 (23.3)
Income, median (IQR), $1000b	57.2 (48.4-65.7)	57.3 (48.4-66.0)	57.2 (48.2-65.3)	58.0 (49.3-68.0)	57.3 (48.7-67.0)	57.8 (48.7-67.8)	54.8 (46.8-62.9	57.2 (48.6-65.8)
<50.0	48 872 (30.3)	9750 (29.4)	23 548 (30.7)	227 (26.2)	520 (28.0)	6995 (29.0)	4329 (33.8)	3503 (29.4)
50.0-64.9	71 411 (44.3)	14 805 (44.7)	33 786 (44.1)	392 (45.3)	843 (45.4)	10 519 (43.6)	5727 (44.7)	5339 (44.7)
65.0-74.9	17 253 (10.7)	3674 (11.1)	8138 (10.6)	110 (12.7)	209 (11.2)	2749 (11.4)	1033 (8.1)	1340 (11.2)
>75/0	23 832 (14.8)	4912 (14.8)	11 162 (14.6)	137 (15.8)	286 (15.4)	3875 (16.1)	1710 (13.4)	1750 (14.7)
Race and ethnicityc
Asian	2628 (1.6)	558 (1.7)	1322 (1.7)	13 (1.5)	32 (1.7)	327 (1.4)	191 (1.5)	185 (1.6)
Black	19 865 (12.3)	3882 (11.7)	10 165 (13.3)	67 (7.7)	153 (8.2)	2230 (9.2)	1307 (10.2)	2061 (17.3)
Hispanic	4571 (2.8)	1031 (3.1)	2209 (2.9)	15 (1.7)	31 (1.7)	652 (2.7)	368 (2.9)	265 (2.2)
Native North American	258 (0.2)	47 (0.1)	131 (0.2)	0	3 (0.2)	37 (0.2)	20 (0.2)	20 (0.2)
White	127 849 (79.2)	26 449 (79.8)	60 053 (78.4)	724 (83.6)	1555 (83.7)	19 692 (81.6)	10 366 (81.0)	9010 (75.5)
Other	3049 (1.9)	652 (2.0)	1425 (1.9)	16 (1.8)	36 (1.9)	471 (2.0)	253 (2.0)	196 (1.6)
Unavailable	3148 (2.0)	522 (1.6)	1329 (1.7)	31 (3.6)	48 (2.6)	729 (3.0)	294 (2.3)	195 (1.6)

Abbreviation: MCI, mild cognitive impairment.

Demographic information is reported from the month of the earliest observed Alzheimer disease and related dementia diagnosis.

Annual household income by zip code.

Race and ethnicity data were reported by the clinician or patient and principally include data provided to the Social Security Administration by individuals. Race and ethnicity were assessed because inequalities exist in access to care and treatment provided; thus, the population was assessed to begin to understand the skew and bias throughout the population and accurately describe the demographic characteristics of the sample.

Table 2.

Health Care Use Before and After Diagnosis

Health care service type	No. of uses, median (IQR)a
	Total		Alzheimer		Dementia		Frontotemporal dementia		Lewy body		MCI		Parkinson		Vascular dementia
	Before	After	Before	After	Before	After	Before	After	Before	After	Before	After	Before	After	Before	After
Visits to specialistsb	0	0 (0-1)	0	0 (0-1)	0	0 (0-1)	0 (0-1)	1 (0-1)	0 (0-1)	1 (0-1)	0 (0-1)	1 (0-1)	0	0 (0-1)	0	0 (0-1)
Specialty types visited	8 (5-12)	10 (6-15)	7 (4-11)	9 (6-13)	8 (5-12)	10 (6-15)	8 (5-11)	10 (7-14)	9 (6-13)	11 (7-16)	9 (6-13)	10 (7-15)	9 (6-14)	12 (8-17)	9 (5-13)	11 (7-16)
Visits by service type
Inpatient	0	0 (0-5)	0	0 (0-3)	0 (0-1)	0 (0-6)	0	0 (0-2.75)	0	0 (0-4)	0	0 (0-1)	0 (0-1)	0 (0-7)	0 (0-3)	0 (0-8)
Outpatient	6 (2-16)	12 (4-29)	5 (2-12)	8 (3-22)	6 (2-17)	13 (5-32)	5 (2-11)	9 (4-20)	7 (3-16)	14 (5-31.75)	6 (3-14)	9 (4-20)	8 (3-18)	15 (6-33)	8 (3-23)	17 (6-41)
Doctor	13 (7-23)	18 (10-28)	11 (6-19)	15 (9-24)	13 (6-22)	17 (10-28)	12 (6-23)	18 (11-28)	16 (9-25)	21 (13-31)	16 (9-26)	20 (12-31)	17 (9-27)	22 (13-34)	14 (7-24)	19 (11-31)
Emergency department	0 (0-1)	0 (0-2)	0 (0-1)	0 (0-1)	0 (0-1)	1 (0-2)	0 (0-1)	0 (0-1)	0 (0-1)	1 (0-2)	0 (0-1)	0 (0-1)	0 (0-1)	1 (0-2)	0 (0-1)	0 (0-2)
Distinct generic drugs prescribed	10 (6-15)	11 (7-17)	9 (5-13)	11 (7-15)	10 (6-15)	11 (7-17)	8 (5-13)	11 (7-16)	10 (6-15)	12.5 (9-18)	10 (6-16)	11 (7-17)	11 (7-17)	13 (8-19)	11 (6-16)	13 (8-18)
Annual cost by service type, $1000
Inpatient	13.2 (5.3-27.8)	14.2 (5.1-30.2)	12.0 (4.5-24.4)	12.6 (3.8-26.3)	13.2 (5.0-27.8)	14.3 (4.9-30.5)	13.4 (5.8-28.0)	14.5 (6.3-29.3)	11.5 (4.7-24.2)	14.4 (6.3-30.8)	13.4 (5.8-27.7)	13.9 (5.4-29.0)	13.7 (5.7-29.0)	15.9 (6.0-35.4)	15.2 (6.1-32.6)	15.6 (5.8-33.0)
Outpatient	0.5 (<0.1-2.7)	1.2 (<0.1-4.5)	0.3 (<0.1-2.0)	0.7 (<0.1-3.4)	0.5 (<0.1-2.8)	1.3 (<0.1-4.8)	0.5 (<0.1-2.1)	1.1 (0.2-4.0)	0.7 (<0.1-3.0)	1.7 (0.2-5.2)	0.6 (<0.1-2.6)	1.0 (0.1-3.6)	0.7 (<0.1-3.4)	2.1 (0.1-6.3)	0.8 (0.1-3.4)	1.8 (0.2-5.6)
Doctor	1.2 (0.3-2.6)	1.7 (0.5-3.3)	1.0 (0.2-2.2)	1.4 (0.4-2.8)	1.1 (0.2-2.5)	1.5 (0.4-3.1)	1.3 (0.4-2.7)	2.0 (0.8-3.6)	1.6 (0.5-3.0)	2.1 (0.9-3.9)	1.6 (0.5-3.3)	2.2 (0.8-4.0)	1.5 (0.4-3.3)	2.2 (0.7-4.1)	1.4 (0.5-2.8)	1.8 (0.8-3.4)
ED	0.6 (0.2-1.1)	0.6 (0.2-1.2)	0.5 (0.2-1.0)	0.60 (0.2-1.1)	0.6 (0.2-1.1)	0.6 (0.2-1.2	0.6 (0.2-1.0)	0.7 (0.2-1.3)	0.7 (0.2-1.1)	0.7 (0.2-1.3)	0.5 (0.2-1.0)	0.6 (0.2-1.1)	0.6 (0.2-1.2)	0.7 (0.2-1.4)	0.6 (0.2-1.1)	0.7 (0.2-1.3)
Prescription	1.1 (0.4-3.8)	1.5 (0.5-4.8)	0.9 (0.3-3.1)	1.2 (0.5-4.1)	1.1 (0.4-3.9)	1.5 (0.5-4.8)	0.9 (0.3-3.4)	1.3 (0.5-4.3)	1.3 (0.5-4.1)	2.0 (0.8-5.4)	1.2 (0.5-4.1)	1.5 (0.5-4.9)	1.4 (0.5-4.5)	1.8 (0.6-5.4)	1.5 (0.5-4.6)	2.0 (0.7-5.6)
Elixhauser comorbidity score, mean (SD)	10.0 (10.2)	11.0 (10.9)	9.0 (9.4)	11.0 (10.1)	10.0 (10.5)	12.0 (11.2)	7.0 (8.2)	8.0 (9.4)	9.0 (8.9)	10.0 (9.9)	8.0 (9.1)	9.0 (9.6)	12.0 (11.3)	14.0 (12.2)	11.0 (10.9)	12.0 (11.4)
Cognitive drug fills, No. (%)c
Acetylcholinesterase inhibitor	24 628 (15.3)	57 205 (35.5)	7417 (22.4)	17 560 (53.0)	12 923 (16.9)	27 718 (36.2)	117 (13.5)	341 (39.4)	291 (15.7)	869 (46.8)	1877 (7.8)	5959 (24.7)	606 (4.7)	1350 (10.5)	1397 (11.7)	3408 (28.6)
NMDA receptor antagonist	9351 (5.8)	26 807 (16.6)	3020 (9.1)	9005 (27.2)	4826 (6.3)	12 989 (16.9)	42 (4.8)	157 (18.1)	99 (5.3)	336 (18.1)	588 (2.4)	2059 (8.5)	234 (1.8)	617 (4.8)	542 (4.5)	1644 (13.8)
BPSD drug
Antipsychotic	9096 (5.6)	20 730 (12.8)	1571 (4.7)	4388 (13.2)	4897 (6.4)	11 432 (14.9)	47 (5.4)	140 (16.2)	227 (12.2)	551 (29.7)	1079 (4.5)	1535 (6.4)	512 (4.0)	899 (7.0)	763 (6.4)	1785 (15.0)
Mood stabilizer or anticonvulsant	25 560 (15.8)	29 380 (18.2)	4051 (12.2)	4813 (14.5)	11 843 (15.5)	13 714 (17.9)	131 (15.1)	158 (18.2)	328 (17.7)	373 (20.1)	4389 (18.2)	4716 (19.5)	2681 (20.9)	3077 (24.0)	2137 (17.9)	2529 (21.2)
Antidepressant	31 276 (19.4)	39 130 (24.2)	6078 (18.3)	8038 (24.3)	14 826 (19.3)	18 575 (24.2)	183 (21.1)	265 (30.6)	435 (23.4)	537 (28.9)	5013 (20.8)	5865 (24.3)	2327 (18.2)	2710 (21.2)	2414 (20.2)	3140 (26.3)
Z-hypnotic	3088 (1.9)	2865 (1.8)	503 (1.5)	479 (1.4)	1302 (1.7)	1203 (1.6)	22 (2.5)	19 (2.2)	28 (1.5)	16 (0.9)	650 (2.7)	582 (2.4)	368 (2.9)	375 (2.9)	215 (1.8)	191 (1.6)
Benzodiazepine	21 048 (13.0)	21 938 (13.6)	3671 (11.1)	3968 (12.0)	9883 (12.9)	10 557 (13.8)	108 (12.5)	123 (14.2)	276 (14.9)	335 (18.0)	3594 (14.9)	3442 (14.3)	2053 (16.0)	2033 (15.9)	1463 (12.3)	1480 (12.4)

Abbreviations: BPSD, behavioral and psychological symptoms of dementia; ED, emergency department; MCI, mild cognitive impairment; NMDA, N-methyl-d-aspartate.

Health care use is presented in the year before and the year after diagnosis, including month of diagnosis.

Includes neurology, geriatric, psychiatric, and other specialist clinicians.

Prescription use was included if an individual had any fill of more than a 30-day supply of a given drug during the 12-month observation window.

Discussion

This cohort study found that prescription fills were present preceding ADRD diagnosis, suggesting that acetylcholinesterase or NDMA receptor antagonists were being used as preemptive medication prior to a formal ADRD diagnosis or off label for other conditions. Further study is needed to understand implications of fills preceding ADRD diagnosis. Increases in antidepressant, mood stabilizer, and anticonvulsant fills suggest off-label use for BPSD, despite limited clinical effectiveness evidence.[3] Benzodiazepine fills were stable, while antipsychotic fills increased, despite evidence that these medications are associated with worse cognitive symptoms and increased risk of falls, stroke, and hospitalization.[4] Low fills for sedative-hypnotic-Z-drugs may reflect a clinical understanding of their appropriate place in therapy.

Evidence of polypharmacy (median, 11 prescriptions postdiagnosis) suggests opportunities to simplify medication regimens and minimize Beers criteria medications. Care across many clinician specialties (median [IQR], 10 [6-15] visits postdiagnosis) suggests the need for ADRD care coordination. Primary care clinicians are uniquely positioned to manage complex medical regimens from multiple specialist consultants. Using evidenced-based recommendations, giving specific attention to deprescribing, may be associated with improved outcomes and reduced harm. Our findings are limited in that our sample did not represent all US payers or individuals without insurance. Our methodology did not distinguish among dementia subtypes or off-label drug use for BPSD vs on-label use for comorbidities (eg, depression, psychosis, or seizures). Fills do not equate to prescriptions written or medication adherence.