A Real-World Analysis of the Clinical Journey,Diagnosis,and Monitoring Patterns of Patients With Alzheimer Disease by Stage in the United States

Abstract

This study aimed to describe the clinical journey of patients with different stages of Alzheimer disease (AD). This was a cross-sectional survey of US primary care physicians (PCPs)/specialists and patients using the Adelphi Real World AD Disease Specific Programme™ (December 2022 – September 2023). Patients were stratified by disease severity and data are presented as the mean (SD) or frequencies/percentages. In the overall sample (N = 990), mean time from symptom onset to first evaluation was 31.4 (40.6) weeks and mean time from evaluation to diagnosis was 14.2 (29.0) weeks (mild cognitive impairment due to AD, 12.0 [22.7] weeks; mild AD dementia, 15.7 [31.6] weeks; moderate AD dementia, 14.0 [29.9] weeks; severe AD dementia, 5.1 [9.6] weeks). 74.5% of the overall sample was initially evaluated by their PCP and 13.8% by a neurologist. Patients with AD experience many barriers during the diagnostic journey; however, PCPs and neurologists play key roles in early diagnosis.

Plain Language Summary

Alzheimer disease (AD) is a main cause of dementia in older adults. Early diagnosis of AD is key, as it increases opportunities for treatment that may slow cognitive decline. However, underdiagnosis and diagnostic delays are common, particularly in the early stages of the disease. This study aimed to describe the clinical journey of patients with different stages of AD and to describe the diagnostic and monitoring tests used in these patients. This study also aimed to evaluate the role of primary care physicians (PCPs) and specialists in the diagnostic process. This study used survey data from the Adelphi Real World AD Disease Specific Programme™ between December 2022 and September 2023. The survey included data from US PCPs and specialists and their patients. Patient data was separated by disease severity into 4 categories: mild cognitive impairment due to AD (MCI due to AD), mild AD dementia (MAD), moderate AD dementia (MoAD), and severe AD dementia (SevAD). In the overall sample (N = 990), the average time from the beginning of symptoms to the first doctor’s appointment was 31.4 weeks, and the average time from the first doctor’s appointment to diagnosis was 14.2 weeks. The average time from the first doctor’s appointment to diagnosis was 12.0 weeks for patients with MCI due to AD, 15.7 weeks for MAD, 14.0 weeks for MoAD, and 5.1 weeks for SevAD. In the overall sample, 74.5% of patients were first evaluated by their PCP and 13.8% were first evaluated by a neurologist. Blood tests for AD-specific biomarkers were used to diagnose 9.0% of patients. This study found that PCPs and neurologists play an important role in diagnosing AD during early disease stages and that biomarker tests were not frequently used during diagnosis. In sum, patients with AD experience many barriers that may prevent quick diagnosis and potential treatment.

Keywords

Alzheimer disease Alzheimer dementia cognitive impairments cross-sectional survey diagnosis delay

Introduction

Alzheimer disease (AD) is the main cause of dementia in older individuals, affecting approximately 50 million people worldwide¹ and 6.7 million people aged ≥65 in the United States.² Pathological hallmarks of AD include the presence of amyloid β and phosphorylated tau.¹ Currently, AD is the fifth leading cause of death in US adults aged ≥65.² AD represents a substantial cost to society, including an estimated $339.5 billion in unpaid caregiving costs in 2022 and $345 billion in health care costs in 2023.² Patients with advanced AD also have higher AD-related health care resource utilization compared to patients with early AD.³

Diagnosis of AD typically begins with an assessment of cognitive status, often using a brief cognitive assessment tool such as the Mini Mental State Exam (MMSE). MMSE scores indicating cognitive impairment may lead physicians to further evaluate specific etiologies and to consider the potential contribution of AD-related pathology to the impairment.^4,5 Diagnosis may involve structural neuroimaging tests, functional imaging such as fluorodeoxyglucose positron emission tomography (FDG-PET), or more rarely amyloid-positron emission tomography (amyloid-PET).¹ However, confirming AD pathology as early as possible may be difficult due to a variety of factors including difficulties distinguishing between normal aging and cognitive impairment.⁶ Therefore, underdiagnosis and diagnostic delays are common, particularly in the early stages of disease.^7,8 Moreover, while several tools are available to establish a biomarker-confirmed diagnosis (amyloid-PET, cerebrospinal fluid [CSF] biomarkers, and emerging blood-based biomarkers), these tools are not widely used and can be difficult to access.

Early diagnosis can expand opportunities for patients to access treatments, such as anti-amyloid monoclonal antibodies, indicated for early stages of AD.⁹ A diagnostic delay could prevent patients from receiving treatment, should cognitive impairment progress during the delay or while awaiting biomarker confirmation.⁶ In order to obtain timely AD diagnoses, it is important to identify factors that may delay diagnoses.

AD may be treated with therapies that address symptoms or medications that remove amyloid. US Food and Drug Administration (FDA)-approved symptomatic treatments include cholinesterase inhibitors and N-methyl-D-aspartate receptor antagonists, which are intended to enhance cognitive function.^1,10 More recently, the FDA approved the disease-modifying therapies donanemab and lecanemab.^10-12 First approvals for lecanemab and donanemab were received in January 2023¹³ and July 2024,¹² respectively. While emerging therapies have potential benefits for patients with early stages of AD, current diagnostic and monitoring processes remain poorly understood and the roles of primary care physicians (PCPs) and specialists in these processes are unclear. The data from this study serve as a benchmark for the field as more biomarkers and therapies enter the market.

The objectives of this study were to describe the clinical journey of patients with mild cognitive impairment (MCI) due to AD, mild AD dementia (MAD), moderate AD dementia (MoAD), and severe AD dementia (SevAD) at diagnosis, understand which tests are used to diagnose and monitor patients, and understand the roles of PCPs and specialists in the diagnostic process.

Methods

Data were drawn from the Adelphi Real World AD Disease Specific Programme™ (DSP), a cross-sectional survey with elements of retrospective data collection which captured data from PCPs, neurologists, psychiatrists, geriatricians, and their patients and was conducted in the United States between December 2022 and September 2023. The DSP methodology has been previously described,^14,15 validated,¹⁶ and demonstrated to be representative and consistent over time.¹⁷

Physicians were instructed to complete patient record forms for their next 9 consecutive patients with AD or MCI due to AD. Each patient record form captured details regarding patient demographics, cognitive test results at the time of diagnosis and at the time of survey completion, comorbidities, physician specialties consulted, time to diagnosis, and testing conducted. Each patient for whom the physician completed a patient record form was invited to fill out a patient self-completion form; these were completed independently immediately following evaluation and were returned in a sealed envelope, ensuring that responses were kept confidential.

Patients included in the survey were aged ≥50 years with a current diagnosis of AD dementia or MCI due to AD based on physician assessment. MCI due to AD included prodromal AD, predementia AD, and amnestic/nonamnestic MCI due to suspected AD. AD dementia included AD and mixed dementia. Patient- or physician-reported biomarker testing was not required in this population. For all diagnosis-related results, patients were classified based on their MMSE score at diagnosis. For all other results, including demographics, comorbidities, and patient experiences, patients were classified based on their MMSE score at the time of survey completion, as their MMSE score and AD stage may have changed since their initial diagnosis. Patients with MMSE scores of 26 to 29, 21 to 25, 11 to 20, and 0 to 10 were classified as having MCI due to AD, MAD, MoAD, and SevAD, respectively.¹⁸

Post hoc analyses were performed to assess the differences between patients who had a shorter or longer than median time duration from symptom onset to first evaluation (≤21.9 or >21.9 weeks) and first evaluation to diagnosis (≤12.0 weeks or >12.0 weeks), including variables such as age, sex, comorbidities, and types of medications. The statistical testing method is indicated with footnotes in each table. Additionally, a Cox proportional hazards model was used to explore the relationship between the hazard function and a set of covariates. This model assumes that covariates multiplicatively shift the hazard function. This assumption was evaluated using Schoenfeld residuals after fitting the Cox proportional hazards model. In cases of ties, the Breslow method was applied.

An oversample of patients with a biomarker-confirmed diagnosis of MCI due to AD (eg, based on a positive CSF test or amyloid-PET scan) was also collected. These patients were included in the analyses to boost the number of patients with milder disease to better understand this patient cohort and reduce variability from individual patients. These patients were removed from analyses where their inclusion would influence sizing (eg, analyses of severity and diagnostic testing).

This article was developed in accordance with STROBE cross-sectional reporting guidelines.¹⁹

Results

Overall, 191 physicians reported data for 1365 patients with AD or MCI due to AD. Of those, 906 patients had an MMSE score at the time of data collection, including 127 with MCI due to AD, 396 with MAD, 336 with MoAD, and 47 with SevAD. In the overall group, mean patient age was 74.6 years old and 73.8% were White (Table 1). Mean (SD) patient age was 72.3 (8.0) years in the MCI due to AD cohort, 74.0 (8.2) years in the MAD cohort, 75.7 (8.2) years in the MoAD cohort, and 77.7 (8.8) years in the SevAD cohort; patients with more severe disease were significantly older (P < 0.001; Table 1). Overall, 52.5% of patients were male and 47.1% were female. In the MCI due to AD group, 56.7% of patients were male and 43.3% were female, in the MAD group 54.3% were male and 45.2% were female, and in the MoAD group 50.9% were male and 48.8% were female. The SevAD cohort was predominantly female (38.3% male, 61.7% female; Table 1). The mean (SD) BMI in the overall sample was 26.4 (4.3) kg/m² and the median was 25.9 kg/m². Mean (SD) BMI was 27.6 (4.3) kg/m² in the MCI due to AD cohort and 25.4 (4.9) kg/m² in the SevAD cohort; patients with more severe disease had a significantly lower BMI (P = 0.003; Table 1). Notably, the median BMI in the MCI due to AD cohort was 27.3 kg/m², indicating that >50% of patients in this cohort were overweight or obese.

Table 1.

Patient Demographics, Comorbidities, and Treatment Regimens Stratified by Disease Severity at Survey Completion^a

Demographics	Overall sampleN = 906	MCI due to AD n = 127	MAD n = 396	MoAD n = 336	SevADn = 47	P value
Age,^b y						<0.001
Mean (SD)	74.6 (8.3)	72.3 (8.0)	74.0 (8.2)	75.7 (8.2)	77.7 (8.8)
Median (IQR)	75.0 (68.0-81.0)	72.0 (67.0-77.0)	74.0 (68.0-80.0)	76.0 (70.0-82.0)	78.0 (72.0-85.0)
Sex,^c n (%)						0.31
Male	476 (52.5)	72 (56.7)	215 (54.3)	171 (50.9)	18 (38.3)
Female	427 (47.1)	55 (43.3)	179 (45.2)	164 (48.8)	29 (61.7)
Intersex	3 (0.3)	0 (0.0)	2 (0.5)	1 (0.3)	0 (0.0)
BMI,^b kg/m²						0.003
Mean (SD)	26.4 (4.3)	27.6 (4.3)	26.3 (4.0)	26.1 (4.5)	25.4 (4.9)
Median (IQR)	25.9 (23.6-28.7)	27.3 (24.4-29.4)	25.9 (23.7-28.5)	25.6 (23.1-28.3)	25.4 (21.1-29.7)
Race,^c,d n (%)
White	669 (73.8)	105 (82.7)	299 (75.5)	230 (68.5)	35 (74.5)	0.01
Black/African American/African or Caribbean	143 (15.8)	12 (9.4)	64 (16.2)	59 (17.6)	8 (17.0)	0.17
East or Southeast Asian	29 (3.2)	4 (3.1)	7 (1.8)	18 (5.4)	0 (0.0)	0.03
South or Central American Native	24 (2.6)	2 (1.6)	11 (2.8)	10 (3.0)	1 (2.1)	0.93
South Asian (Indian subcontinent)	16 (1.8)	2 (1.6)	6 (1.5)	8 (2.4)	0 (0.0)	0.72
Middle Eastern or North African	10 (1.1)	0 (0.0)	4 (1.0)	5 (1.5)	1 (2.1)	0.36

Comorbidities	Overall sampleN = 906	MCI due to ADn = 127	MADn = 396	MoADn = 336	SevAD n = 47	P value
Number of comorbidities^b						<0.001
Mean (SD)	3.2 (2.6)	3.0 (2.4)	2.8 (2.3)	3.7 (2.9)	3.5 (2.4)
Median (IQR)	3.0 (1.0-4.0)	3.0 (1.0-4.0)	2.0 (1.0-4.0)	3.0 (2.0-5.0)	4.0 (2.0-5.0)
Types of comorbidities,^c,e n (%)
Arterial hypertension	326 (36.0)	50 (39.4)	129 (32.6)	127 (37.8)	20 (42.6)	0.25
Anxiety	324 (35.8)	47 (37.0)	126 (31.8)	128 (38.1)	23 (48.9)	0.07
Depression	259 (28.6)	34 (26.8)	102 (25.8)	106 (31.5)	17 (36.2)	0.21
Hyperlipidemia	229 (25.3)	32 (25.2)	95 (24.0)	90 (26.8)	12 (25.5)	0.86
Arthritis	176 (19.4)	21 (16.5)	68 (17.2)	74 (22.0)	13 (27.7)	0.14
Insomnia/sleep disorders	106 (11.7)	8 (6.3)	40 (10.1)	48 (14.3)	10 (21.3)	0.01
Hypothyroidism	100 (11.0)	13 (10.2)	43 (10.9)	39 (11.6)	5 (10.6)	0.99
Back pain	99 (10.9)	19 (15.0)	40 (10.1)	34 (10.1)	6 (12.8)	0.40
Stress	85 (9.4)	17 (13.4)	32 (8.1)	30 (8.9)	6 (12.8)	0.25
Diabetes without chronic complications	81 (8.9)	7 (5.5)	29 (7.3)	38 (11.3)	7 (14.9)	0.06
Not diagnosed with any concomitant conditions	114 (12.6)	17 (13.4)	57 (14.4)	35 (10.4)	5 (10.6)	0.42

Medications^c,f n (%)	Overall sampleN = 792	MCI due to ADn = 110	MADn = 339	MoADn = 301	SevADn = 42	P value
Antidepressants	294 (37.1)	36 (32.7)	118 (34.8)	120 (39.9)	20 (47.6)	0.20
ACE inhibitors	225 (28.4)	21 (19.1)	78 (23.0)	113 (37.5)	13 (31.0)	<0.001
Statins	223 (28.2)	39 (35.5)	88 (26.0)	88 (29.2)	8 (19.0)	0.14

^aData included in this table were obtained from the physician-completed patient record form. An oversample of patients with a biomarker-confirmed diagnosis of MCI due to AD was included in these analyses to boost the number of patients with milder disease and to reduce variability from individual patients.

^bStatistical significance was assessed using ANOVA.

^cStatistical significance was assessed using Fisher’s exact test.

^dData for ‘American Indian, Indigenous American, or Alaska Native,’ ‘Native Hawaiian or Pacific Islander,’ and ‘Other’ groups were not included in the table due to a low sample size.

^eShown are the top 10 comorbidities in the overall sample.

^fMedications shown are the top 3 medication types reported in the overall sample.

ACE, angiotensin-converting enzyme; AD, Alzheimer disease; ADHD, attention deficit hyperactivity disorder; ALS, amyotrophic lateral sclerosis; ANOVA, Analysis of Variance; BMI, body mass index; CBS, corticobasal syndrome; IBS, irritable bowel syndrome; IQR, interquartile range; MAD, mild Alzheimer disease dementia; MCI, mild cognitive impairment; MND, motor neuron disease; MoAD, moderate Alzheimer disease dementia; NASH, nonalcoholic steatohepatitis; SevAD, severe Alzheimer disease dementia.

According to physician-completed patient record forms, patients experienced a mean of 3.2 comorbidities at the time of data collection (Table 1). Frequently occurring comorbidities in the overall sample included arterial hypertension (36.0%), anxiety (35.8%), depression (28.6%), and hyperlipidemia (25.3%; Table 1). The mean (SD) number of comorbidities in patients with SevAD, MoAD, MAD, and MCI due to AD was 3.5 (2.4), 3.7 (2.9), 2.8 (2.3), and 3.0 (2.4), respectively; disease severity was significantly associated with the number of comorbidities (P < 0.001; Table 1).

Physicians reported that patients in all 4 cohorts commonly received treatments for comorbidities. In the overall sample, the most frequently received comorbidity treatments were antidepressants (37.1%), angiotensin-converting enzyme (ACE) inhibitors (28.4%), and statins (28.2%; Table 1). Antidepressants and ACE inhibitors were used by 39.9% and 37.5% of patients with MoAD and 47.6% and 31.0% of patients with SevAD, respectively. These medications were used by 34.8% and 23.0% of patients with MAD and 32.7% and 19.1% of patients with MCI due to AD, respectively. Conversely, 35.5% of patients with MCI due to AD were using statins compared to 26.0% of patients with MAD, 29.2% of patients with MoAD, and 19.0% of patients with SevAD (Table 1).

Most patients experienced some memory impairment, with physicians reporting that 85.8% of the overall sample had some short-term memory loss (34.4% mild, 51.7% moderate, and 13.9% severe) and 34.1% had some long-term memory loss (26.2% mild, 55.0% moderate, and 18.8% severe; Table 2). Severe short-term and long-term memory loss were reported in 65.2% and 58.8% of patients with SevAD and 0.0% and 0.0% of patients with MCI due to AD, respectively (Table 2). Physicians also reported that patients frequently had problems with functional abilities and that these problems worsened with disease progression. In the overall sample, 29.0% of patients had difficulties travelling outside the home, 33.3% had difficulties with driving, and 15.8% had difficulties using public transportation (Table 2; Supplemental Table 1). Difficulties with driving were mostly mild (50.0%) and moderate (40.9%) in patients with MCI due to AD; however, 90.0% of patients with SevAD had severe difficulties with driving (Table 2). Similar patterns were seen for difficulties travelling outside of the home and difficulties using public transportation (Supplemental Table 1).

Table 2.

Top Ten Symptoms Stratified by Disease Severity at Survey Completion^a

Current symptom profile	Overall sample	MCI due to AD	MAD	MoAD	SevAD	P value^b
Loss of short-term memory, n (%)	N = 777	n = 102	n = 343	n = 286	n = 46	<0.001
Mild	267 (34.4)	80 (78.4)	142 (41.4)	44 (15.4)	1 (2.2)
Moderate	402 (51.7)	22 (21.6)	178 (51.9)	187 (65.4)	15 (32.6)
Severe	108 (13.9)	0 (0.0)	23 (6.7)	55 (19.2)	30 (65.2)
Difficulties in recall of names/words, n (%)	N = 427	n = 57	n = 150	n = 188	n = 32	<0.001
Mild	163 (38.2)	49 (86.0)	70 (46.7)	43 (22.9)	1 (3.1)
Moderate	185 (43.3)	8 (14.0)	68 (45.3)	101 (53.7)	8 (25.0)
Severe	79 (18.5)	0 (0.0)	12 (8.0)	44 (23.4)	23 (71.9)
Difficulties in concentration/attention, n (%)	N = 404	n = 37	n = 141	n = 190	n = 36	<0.001
Mild	123 (30.4)	28 (75.7)	61 (43.3)	32 (16.8)	2 (5.6)
Moderate	223 (55.2)	9 (24.3)	76 (53.9)	127 (66.8)	11 (30.6)
Severe	58 (14.4)	0 (0.0)	4 (2.8)	31 (16.3)	23 (63.9)
Difficulties managing finances/paying bills, n (%)	N = 372	n = 31	n = 140	n = 169	n = 32	<0.001
Mild	94 (25.3)	25 (80.6)	49 (35.0)	20 (11.8)	0 (0.0)
Moderate	148 (39.8)	4 (12.9)	71 (50.7)	70 (41.4)	3 (9.4)
Severe	130 (34.9)	2 (6.5)	20 (14.3)	79 (46.7)	29 (90.6)
Difficulties in problem solving/calculations, n (%)	N = 348	n = 33	n = 126	n = 158	n = 31	<0.001
Mild	93 (26.7)	29 (87.9)	44 (34.9)	19 (12.0)	1 (3.2)
Moderate	163 (46.8)	4 (12.1)	74 (58.7)	83 (52.5)	2 (6.5)
Severe	92 (26.4)	0 (0.0)	8 (6.3)	56 (35.4)	28 (90.3)
Confusion with time or place, n (%)	N = 328	n = 19	n = 101	n = 171	n = 37	<0.001
Mild	90 (27.4)	13 (68.4)	40 (39.6)	36 (21.1)	1 (2.7)
Moderate	176 (53.7)	6 (31.6)	56 (55.4)	100 (58.5)	14 (37.8)
Severe	62 (18.9)	0 (0.0)	5 (5.0)	35 (20.5)	22 (59.5)
Loss of long-term memory, n (%)	N = 309	n = 12	n = 108	n = 155	n = 34	<0.001
Mild	81 (26.2)	8 (66.7)	43 (39.8)	28 (18.1)	2 (5.9)
Moderate	170 (55.0)	4 (33.3)	59 (54.6)	95 (61.3)	12 (35.3)
Severe	58 (18.8)	0 (0.0)	6 (5.6)	32 (20.6)	20 (58.8)
Difficulties with driving, n (%)	N = 302	n = 22	n = 113	n = 137	n = 30	<0.001
Mild	57 (18.9)	11 (50.0)	34 (30.1)	12 (8.8)	0 (0.0)
Moderate	107 (35.4)	9 (40.9)	40 (35.4)	55 (40.1)	3 (10.0)
Severe	138 (45.7)	2 (9.1)	39 (34.5)	70 (51.1)	27 (90.0)
Difficulties in executive function, n (%)	N = 279	n = 16	n = 94	n = 138	n = 31	<0.001
Mild	72 (25.8)	14 (87.5)	39 (41.5)	19 (13.8)	0 (0.0)
Moderate	132 (47.3)	2 (12.5)	48 (51.1)	75 (54.3)	7 (22.6)
Severe	75 (26.9)	0 (0.0)	7 (7.4)	44 (31.9)	24 (77.4)
Difficulties switching between tasks/multi-tasking, n (%)	N = 267	n = 23	n = 95	n = 121	n = 28	<0.001
Mild	77 (28.8)	23 (100.0)	40 (42.1)	13 (10.7)	1 (3.6)
Moderate	120 (44.9)	0 (0.0)	48 (50.5)	70 (57.9)	2 (7.1)
Severe	70 (26.2)	0 (0.0)	7 (7.4)	38 (31.4)	25 (89.3)

^bStatistical significance was assessed using the Kruskal-Wallis test.

AD, Alzheimer disease; MAD, mild Alzheimer disease dementia; MCI, mild cognitive impairment; MoAD, moderate Alzheimer disease dementia; SevAD, severe Alzheimer disease dementia.

Data were available for 990 patients with an MMSE score at diagnosis, including 225 with MCI due to AD, 484 with MAD, 257 with MoAD, and 24 with SevAD (Table 3).

Table 3.

Diagnostic Journey Stratified by Disease Severity at Diagnosis^a

First evaluation	Overall sample	MCI due to AD	MAD	MoAD	SevAD
Top 5 symptoms prompting initial evaluation, n (%)	N = 978	n = 223	n = 475	n = 256	n = 24
Loss of short-term memory	814 (83.2)	195 (87.4)	399 (84.0)	202 (78.9)	18 (75.0)
Difficulties in recall of names/words	386 (39.5)	101 (45.3)	181 (38.1)	95 (37.1)	9 (37.5)
Difficulties in concentration/attention	368 (37.6)	69 (30.9)	175 (36.8)	110 (43.0)	14 (58.3)
Difficulties managing finances/paying bills	280 (28.6)	43 (19.3)	145 (30.5)	80 (31.2)	12 (50.0)
Confusion with time or place	275 (28.1)	30 (13.5)	121 (25.5)	111 (43.4)	13 (54.2)
Symptom onset to first evaluation, weeks	N = 619	n = 143	n = 313	n = 151	n = 12
Mean (SD)	31.4 (40.6)	32.1 (34.1)	29.8 (40.3)	34.1 (47.7)	31.7 (23.3)
Median (IQR)	17.4 (5.4-43.6)	18.7 (8.7-52.1)	16.7 (4.4-35.0)	18.6 (5.7-44.7)	33.0 (9.6-53.3)
Min, max	0.0, 356.4	0.0, 196.0	0.0, 241.3	0.0, 356.4	0.0, 64.3
Clinician first consulted, n (%)	N = 981	n = 224	n = 477	n = 256	n = 24
PCP	731 (74.5)	172 (76.8)	364 (76.3)	179 (69.9)	16 (66.7)
Neurologist	135 (13.8)	35 (15.6)	62 (13.0)	36 (14.1)	2 (8.3)
Psychiatrist	50 (5.1)	2 (0.9)	18 (3.8)	26 (10.2)	4 (16.7)
Geriatrician	44 (4.5)	11 (4.9)	24 (5.0)	8 (3.1)	1 (4.2)
Physician assistant	9 (0.9)	1 (0.4)	4 (0.8)	3 (1.2)	1 (4.2)
Geriatric psychiatrist	4 (0.4)	2 (0.9)	1 (0.2)	1 (0.4)	0 (0.0)
Neuropsychiatrist	3 (0.3)	1 (0.4)	2 (0.4)	0 (0.0)	0 (0.0)
Nurse	1 (0.1)	0 (0.0)	0 (0.0)	1 (0.4)	0 (0.0)
Other	4 (0.4)	0 (0.0)	2 (0.4)	2 (0.8)	0 (0.0)
MMSE at first evaluation	N = 797	n = 190	n = 380	n = 208	n = 19
Mean (SD)	22.5 (4.3)	26.9 (1.2)	23.3 (2.0)	18.1 (2.9)	11.1 (6.9)
Median (IQR)	23.0 (20.0-26.0)	27.0 (26.0-28.0)	23.0 (22.0-24.0)	18.0 (16.0-20.0)	10.0 (6.0-15.0)

Referral	Overall sample	MCI due to AD	MAD	MoAD	SevAD
Referral before diagnosis?, n (%)	N = 964	n = 221	n = 473	n = 246	n = 24
Yes	468 (48.5)	113 (51.1)	218 (46.1)	128 (52.0)	9 (37.5)
No	496 (51.5)	108 (48.9)	255 (53.9)	118 (48.0)	15 (62.5)
Physician referred to, n (%)	N = 232	n = 55	n = 117	n = 58	n = 2
Neurologist	162 (69.8)	47 (85.5)	80 (68.4)	34 (58.6)	1 (50.0)
Psychiatrist	24 (10.3)	2 (3.6)	20 (17.1)	1 (1.7)	1 (50.0)
PCP	20 (8.6)	0 (0.0)	6 (5.1)	14 (24.1)	0 (0.0)
Geriatrician	11 (4.7)	2 (3.6)	6 (5.1)	3 (5.2)	0 (0.0)
Neuropsychiatrist	8 (3.4)	4 (7.3)	2 (1.7)	2 (3.4)	0 (0.0)
Geriatric psychiatrist	5 (2.2)	0 (0.0)	2 (1.7)	3 (5.2)	0 (0.0)
Nurse	2 (0.9)	0 (0.0)	1 (0.9)	1 (1.7)	0 (0.0)

Diagnosis	Overall sample	MCI due to AD	MAD	MoAD	SevAD
First evaluation to diagnosis (all patients), weeks	N = 745	n = 173	n = 366	n = 188	n = 18
Mean (SD)	14.2 (29.0)	12.0 (22.7)	15.7 (31.6)	14.0 (29.9)	5.1 (9.6)
Median (IQR)	3.7 (0.0-13.6)	4.4 (0.0-14.4)	3.8 (0.0-14.7)	3.5 (0.0-13.0)	0.0 (0.0-9.3)
Min, max	0.0, 305.3	0.0, 160.6	0.0, 305.3	0.0, 215.6	0.0, 34.9
First evaluation to diagnosis (referred patients), weeks	N = 311	n = 78	n = 148	n = 80	n = 5
Mean (SD)	24.4 (34.8)	21.1 (29.1)	27.2 (36.2)	23.2 (38.1)	11.2 (14.3)
Median (IQR)	11.1 (3.7-29.0)	13.4 (4.4-24.7)	11.9 (4.4-40.6)	8.0 (0.0-27.4)	10.0 (0.0-11.0)
Min, max	0.0, 215.6	0.0, 160.6	0.0, 213.4	0.0, 215.6	0.0, 34.9
First evaluation to diagnosis (non-referred patients), weeks	N = 423	n = 94	n = 212	n = 104	n = 13
Mean (SD)	6.9 (21.3)	4.5 (11.2)	8.0 (25.6)	7.5 (19.8)	2.8 (6.4)
Median (IQR)	0.0 (0.0-6.4)	0.0 (0.0-6.6)	0.0 (0.0-4.8)	0.0 (0.0-7.6)	0.0 (0.0-0.0)
Min, max	0.0, 305.3	0.0, 93.6	0.0, 305.3	0.0, 168.4	0.0, 21.9
Physician who made diagnosis, n (%)	N = 988	n = 225	n = 483	n = 256	n = 24
PCP	388 (39.3)	78 (34.7)	199 (41.2)	101 (39.5)	10 (41.7)
Neurologist	432 (43.7)	126 (56.0)	205 (42.4)	97 (37.9)	4 (16.7)
Psychiatrist	87 (8.8)	4 (1.8)	40 (8.3)	35 (13.7)	8 (33.3)
Geriatrician	57 (5.8)	12 (5.3)	29 (6.0)	14 (5.5)	2 (8.3)
Neuropsychiatrist	15 (1.5)	5 (2.2)	6 (1.2)	4 (1.6)	0 (0.0)
Geriatric psychiatrist	7 (0.7)	0 (0.0)	2 (0.4)	5 (2.0)	0 (0.0)
Physician assistant	1 (0.1)	0 (0.0)	1 (0.2)	0 (0.0)	0 (0.0)
Other	1 (0.1)	0 (0.0)	1 (0.2)	0 (0.0)	0 (0.0)
MMSE at diagnosis	N = 990	n = 225	n = 484	n = 257	n = 24
Mean (SD)	22.1 (4.4)	26.9 (0.9)	23.1 (1.3)	17.4 (2.5)	7.3 (3.5)
Median (IQR)	23.0 (20.0-25.0)	27.0 (26.0-27.0)	23.0 (22.0-24.0)	18.0 (15.0-20.0)	9.5 (4.5-10.0)

AD, Alzheimer disease; IQR, interquartile range; MAD, mild Alzheimer disease dementia; MCI, mild cognitive impairment; MMSE, Mini Mental State Exam; MoAD, moderate Alzheimer disease dementia; PCP, primary care physician; SevAD, severe Alzheimer disease dementia.

According to physician-completed patient record forms, symptoms prompting a patient to seek an initial evaluation included loss of short-term memory (83.2%), difficulties recalling names/words (39.5%), and difficulties in concentration/attention (37.6%; Table 3). In the overall sample, mean (SD) time from symptom onset to first evaluation was 31.4 (40.6) weeks (Table 3). Most patients (74.5% of the overall group) initially consulted with their PCP, while 13.8% initially consulted a neurologist (Table 3). A total of 76.8% of patients with MCI due to AD and 76.3% with MAD consulted their PCP before any other physician compared to 69.9% of patients with MoAD and 66.7% with SevAD.

Approximately half (48.5%) of the overall sample was referred to a secondary provider prior to diagnosis, including 51.1% of patients with MCI due to AD, 46.1% of patients with MAD, 52.0% of patients with MoAD, and 37.5% of patients with SevAD (Table 3).

In the overall sample, 39.3% of initial diagnoses were made by a PCP, while 43.7% were made by a neurologist (Table 3). PCPs made 34.7%, 41.2%, 39.5%, and 41.7% of diagnoses in patients with MCI due to AD, MAD, MoAD, and SevAD (Table 3). Neurologists made 56.0%, 42.4%, 37.9%, and 16.7% of the diagnoses in patients with MCI due to AD, MAD, MoAD, and SevAD, respectively (Table 3).

In the overall group, mean (SD) time from the first evaluation to diagnosis was 14.2 (29.0) weeks. Time to diagnosis ranged from 0.0 (0.0 years; immediate diagnosis) to 305.3 weeks (5.9 years), with a median (IQR) of 3.7 (0.0-13.6) weeks (Table 3, Figure 1). Median (IQR) time from first evaluation to diagnosis was 4.4 (0.0-14.4) weeks for MCI due to AD, 3.8 (0.0-14.7) weeks for MAD, 3.5 (0.0-13.0) weeks for MoAD, and 0.0 (0.0-9.3) weeks for SevAD. Median (IQR) time from symptom onset to first evaluation was 18.7 (8.7-52.1) weeks in the MCI due to AD group, 16.7 (4.4-35.0) weeks in the MAD group, 18.6 (5.7-44.7) weeks in the MoAD group, and 33.0 (9.6-53.3) in the SevAD group.

Figure 1.

Length of time spent at each stage of the diagnostic journey, with patients stratified by disease severity at diagnosis. Hazard curves show the time from symptom onset to the first evaluation (A) and first evaluation to diagnosis (B). Patients with MCI due to AD, MAD, MoAD, and SevAD at diagnosis are shown in blue, green, yellow, and red curves, respectively. AD, Alzheimer disease; MAD, mild Alzheimer disease dementia; MCI, mild cognitive impairment; MoAD, moderate Alzheimer disease dementia; SevAD, severe Alzheimer disease dementia

Patients reported several reasons for delaying an initial evaluation or receiving a delayed diagnosis (Table 4). In the overall sample, the most frequent reason for delaying an evaluation was a belief that memory problems were due to normal aging, rather than a cognitive disorder (81.0%). Other patients reported not wanting to face a serious diagnosis (33.3%) and worrying about what others might think (15.9%). Independence concerns were reported by several patients, including statements such as, “I was worried I might lose my driver’s license” (19.0%), “I was worried about losing my independence” (15.9%), and “I was worried that I would have to move into a nursing home” (14.3%; Table 4).

Table 4.

Reasons for Delays in First Evaluation and/or Diagnosis, Stratified by Disease Severity at Diagnosis^a,b

	Overall sample	MCI due to AD	MAD	MoAD
Reasons for delay in initial evaluation,^c n (%)	N = 63	n = 15	n = 39	n = 9
I initially thought my memory problems were a normal part of aging	51 (81.0)	15 (100.0)	31 (79.5)	5 (55.6)
I was scared what the doctor might tell me/I didn’t want to face the possibility that my memory problems were serious	21 (33.3)	3 (20.0)	14 (35.9)	4 (44.4)
I was worried I might lose my driver’s license	12 (19.0)	4 (26.7)	7 (17.9)	1 (11.1)
I was worried about losing my independence	10 (15.9)	2 (13.3)	7 (17.9)	1 (11.1)
I was worried what other people might think	10 (15.9)	2 (13.3)	8 (20.5)	0 (0.0)
I was worried that I would have to move into a nursing home	9 (14.3)	3 (20.0)	5 (12.8)	1 (11.1)
I was worried about losing friends/family	6 (9.5)	3 (20.0)	3 (7.7)	0 (0.0)
I was worried about not being able to qualify for long term care/life insurance	3 (4.8)	2 (13.3)	1 (2.6)	0 (0.0)
Other	1 (1.6)	0 (0.0)	1 (2.6)	0 (0.0)
Reasons for delay between first evaluation and diagnosis,^d n (%)	N = 777	n = 174	n = 393	n = 195
Time taken to schedule the next consultation	336 (43.2)	72 (41.4)	154 (39.2)	104 (53.3)
Time taken for patient’s referral to a specialist	272 (35.0)	54 (31.0)	149 (37.9)	63 (32.3)
Time taken to schedule a diagnostic test	214 (27.5)	62 (35.6)	106 (27.0)	44 (22.6)
Time taken to get test results/lack of availability of a specialized health care provider	126 (16.2)	36 (20.7)	64 (16.3)	23 (11.8)
Difficulty for patient to attend appointments	91 (11.7)	12 (6.9)	50 (12.7)	28 (14.4)
Time taken for physician to be confident in the diagnosis	48 (6.2)	8 (4.6)	25 (6.4)	14 (7.2)
Patient stopped attending/forgot their appointments	43 (5.5)	6 (3.4)	23 (5.9)	13 (6.7)
Doubts about the current usefulness or desirability of an early diagnosis	32 (4.1)	3 (1.7)	22 (5.6)	7 (3.6)
Concerns about the consequences of a misdiagnosis	22 (2.8)	5 (2.9)	14 (3.6)	3 (1.5)
Other	18 (2.3)	5 (2.9)	10 (2.5)	3 (1.5)

^aAn oversample of patients with a biomarker-confirmed diagnosis of MCI due to AD was included in these analyses to boost the number of patients with milder disease and to reduce variability from individual patients.

^bData for patients with SevAD were not included in this table due to low patient numbers. No patients with SevAD provided information on their reasons for delay in obtaining an initial evaluation and physicians provided information on reasons for delay between first evaluation and diagnosis for only 15 patients with SevAD.

^cData included in this section were obtained from the patient self-completion form. The sample size in this section is smaller than samples obtained from physician-provided data, as not every patient completed the questionnaire.

^dData included in this section were obtained from the physician-completed patient record form.

AD, Alzheimer disease; MAD, mild Alzheimer disease dementia; MCI, mild cognitive impairment; MoAD, moderate Alzheimer disease dementia; SevAD, severe Alzheimer disease dementia.

When asked about factors that contributed to a delay between the first evaluation appointment and diagnosis, physicians identified difficulties related to appointment availability, including, “time taken to schedule the next consultation” (43.2% of patients), “time taken for patient’s referral to a specialist” (35.0%), and “time taken to schedule a diagnostic test” (27.5%; Table 4).

Several post hoc analyses were performed to assess differences between patients who had a shorter or longer duration than the median for aspects of the diagnostic timeline (Supplemental Table 2). These analyses revealed that patients who experienced longer times than the median from symptom onset to first evaluation and from first evaluation to diagnosis had a statistically significantly higher number of comorbidities (Supplemental Table 2 and Supplemental Table 3). Results from the Cox proportional hazards models indicate that patients in the early stage of AD have a higher likelihood of receiving their first clinical evaluation sooner after the onset of initial symptoms. They are also more likely to receive a final diagnosis sooner after the initial evaluation, when controlling for all other covariates. However, this effect is not consistent across all model specifications (Supplemental Table 4).

In the overall sample (excluding the oversample), the most common diagnostic workup tools used to aid in diagnosis were feedback from the patient and their family (87.6%) and the patient’s family history (61.5%; Table 5). Behavioral/cognitive assessments (94.5%) and non-AD specific blood tests (90.5%) were also commonly used as part of the diagnostic process. MRI was performed as part of the evaluation in 59.1% of patients, while CT scans were used in 20.3% of patients. Other tests included AD-specific blood tests (9.0%), FDG-PET (7.7%), and amyloid-PET (4.6%). Most of the overall sample (83.7%) was diagnosed without a biomarker-based confirmation (Table 5).

Table 5.

Monitoring and Diagnostic Tests^a

	Overall sample	MCI due to AD	MAD	MoAD	SevAD
Diagnostic tests,^b n (%)	N = 849	n = 192	n = 414	n = 222	n = 21
Non-AD specific blood test	768 (90.5)	187 (97.4)	366 (88.4)	195 (87.8)	20 (95.2)
Feedback from patient/patient’s family	744 (87.6)	168 (87.5)	361 (87.2)	198 (89.2)	17 (81.0)
Patient’s family history	522 (61.5)	123 (64.1)	256 (61.8)	131 (59.0)	12 (57.1)
MRI	502 (59.1)	136 (70.8)	226 (54.6)	131 (59.0)	9 (42.9)
CT	172 (20.3)	26 (13.5)	101 (24.4)	38 (17.1)	7 (33.3)
EEG/MEG	110 (13.0)	47 (24.5)	48 (11.6)	10 (4.5)	5 (23.8)
AD-specific blood test	76 (9.0)	15 (7.8)	38 (9.2)	21 (9.5)	2 (9.5)
CSF	67 (7.9)	4 (2.1)	33 (8.0)	28 (12.6)	2 (9.5)
FDG-PET	65 (7.7)	10 (5.2)	25 (6.0)	28 (12.6)	2 (9.5)
Amyloid PET	39 (4.6)	6 (3.1)	17 (4.1)	15 (6.8)	1 (4.8)
Neurofilament light blood test	23 (2.7)	4 (2.1)	12 (2.9)	7 (3.2)	0 (0.0)
Tau PET	14 (1.6)	0 (0.0)	6 (1.4)	7 (3.2)	1 (4.8)
Other imaging/scan	9 (1.1)	2 (1.0)	3 (0.7)	3 (1.4)	1 (4.8)
Other blood test	4 (0.5)	1 (0.5)	3 (0.7)	0 (0.0)	0 (0.0)
Patient has a biomarker confirmed diagnosis,^b n (%)	N = 849	n = 192	n = 414	n = 222	n = 21
Yes - AD-specific blood tests only	48 (5.7)	13 (6.8)	28 (6.8)	6 (2.7)	1 (4.8)
Yes - CSF only	36 (4.2)	4 (2.1)	20 (4.8)	11 (5.0)	1 (4.8)
Yes - PET scan only	18 (2.1)	4 (2.1)	8 (1.9)	6 (2.7)	0 (0.0)
Yes - CSF and AD-specific blood tests	12 (1.4)	0 (0.0)	5 (1.2)	7 (3.2)	0 (0.0)
Yes - CSF, AD-specific blood test, and PET scan	11 (1.3)	0 (0.0)	3 (0.7)	7 (3.2)	1 (4.8)
Yes - CSF and PET scan	8 (0.9)	0 (0.0)	5 (1.2)	3 (1.4)	0 (0.0)
Yes - AD-specific blood test and PET scan	5 (0.6)	2 (1.0)	2 (0.5)	1 (0.5)	0 (0.0)
No biomarker confirmed test conducted	711 (83.7)	169 (88.0)	343 (82.9)	181 (81.5)	18 (85.7)
Monitoring tests,^c,d n (%)	N = 879	n = 125	n = 382	n = 326	n = 46
MMSE	879 (100.0)	125 (100.0)	382 (100.0)	326 (100.0)	46 (100.0)
Feedback from patient/patient’s family	770 (87.6)	115 (92.0)	336 (88.0)	273 (83.7)	46 (100.0)
Clock Draw test only	179 (20.4)	28 (22.4)	68 (17.8)	79 (24.2)	4 (8.7)
MRI – Volumetric	179 (20.4)	36 (28.8)	69 (18.1)	69 (21.2)	5 (10.9)
MoCA	149 (17.0)	7 (5.6)	85 (22.3)	54 (16.6)	3 (6.5)
Mini-Cog	82 (9.3)	9 (7.2)	36 (9.4)	31 (9.5)	6 (13.0)
MRI – Diffusion	74 (8.4)	12 (9.6)	27 (7.1)	30 (9.2)	5 (10.9)
CT	59 (6.7)	4 (3.2)	28 (7.3)	26 (8.0)	1 (2.2)
EEG/MEG	53 (6.0)	18 (14.4)	26 (6.8)	5 (1.5)	4 (8.7)
None	269 (30.6)	42 (33.6)	122 (31.9)	91 (27.9)	14 (30.4)

^aData included in this table were obtained from the physician-completed patient record form.

^bPatients are stratified by their disease severity at diagnosis. Oversample patients were not included in these data.

^cPatients are stratified by their disease severity at the time of survey completion. An oversample of patients with a biomarker-confirmed diagnosis of MCI due to AD was included in these analyses to boost the number of patients with milder disease and to reduce variability from individual patients.

^dOnly those tests used in ≥5.0% of patients in the overall sample are listed in this table. Tests used in <5% of patients in the overall sample included FAQ, Florbetapir/Amyvid, ADAS-Cog, Lumipulse, FDG-PET, PrecivityAD^®, fMRI, ADCS-ADL, Elecsys, CIBIC, Elecsys Amyloid Plasma Panel, neurofilament light blood test, CDR, ptau-181 CSF assay, AD8 Dementia Screening Interview, Flortaucipir/Tauvid, s-MoCA, SLUMS, ptau-181, CSF test (assay unknown), QUEST AD-Detect^® Amyloid Beta 42/40 ratio, MRS, GPCOG, CAMDEX, computerized cognitive test battery, Araclon Biotech SL (Grifols) Elisa, other blood tests, DTA, Florbetaben/Neuraceq, CDR-SOB, Diadem SRL AlzoSure Predict U-p53AZ kit, Alzohis Noratest, other CSF assays, other MRI, Flutemetamol/Vizamyl, SPECT, and 123I MIBG cardiac scintigraphy.

AD, Alzheimer Disease; ADAS-Cog, Alzheimer Disease Assessment Scale–Cognitive Subscale; ADCS-ADL, Activities of Daily Living for Mild Cognitive Impairment; CAMDEX, Cambridge Mental Disorders of the Elderly Examination; CDR, Clinical Dementia Rating; CDR-SOB, Clinical Dementia Rating - Sum of Boxes; CIBIC, Clinician Interview-Based Impression of Change; CSF, cerebrospinal fluid; CT, computed tomography; DTI, diffusion tensor imaging tractography; EEG, electroencephalogram; FAQ, Functional Activities Questionnaire; FDG-PET, fluorodeoxyglucose-positron emission tomography; fMRI, functional magnetic resonance imaging; GPCOG, General Practitioner Assessment of Cognition; MAD, mild Alzheimer disease dementia; MCI, mild cognitive impairment; MEG, magnetoencephalogram; MIBG, metaiodobenzylguanidine; MMSE, Mini Mental State Exam; MoAD, moderate Alzheimer disease dementia; MoCA, Montreal Cognitive Assessment; MRI, magnetic resonance imaging; MRS, magnetic resonance spectroscopy; PET, positron emission tomography; ptau-181, Simoa^® phospho-Tau 181; SevAD, severe Alzheimer disease dementia; SLUMS, Saint Louis Mental Status Exam; s-MoCA, Short Montreal Cognitive Assessment; SPECT, single-photon emission computed tomography.

In most patients with an AD diagnosis, disease progression was monitored using feedback from the patient and their family (87.6% of the overall sample at the time of data capture; Table 5). Several other cognitive tests were used to monitor disease progression, including the Clock Draw test (20.4%), the Montreal Cognitive Assessment (MoCA; 17.0%), and the Mini-Cog test (9.3%). Laboratory tests such as PrecivityAD^® and the Elecsys^® Amyloid Plasma Panel were infrequently used to monitor patients (3.2% and 1.7% of the overall sample, respectively). Imaging was not typically used for monitoring, with CT scans and FDG-PET used in 6.7% and 3.3% of the overall sample, respectively, although volumetric and diffusion MRI were used in 20.4% and 8.4% of patients. Physicians reported that 30.6% of the overall sample did not undergo any monitoring tests. The percentage of patients who did not receive monitoring tests was 33.6% in patients with MCI due to AD, 31.9% in patients with MAD, 27.9% in patients with MoAD, and 30.4% in patients with SevAD.

Some patients included in this study completed a survey describing their feelings and experiences with AD. Patients in all disease severity groups reported problems with memory, sleep, and mood (Supplemental Table 5).

Discussion

This study demonstrates that delays may be present throughout the AD diagnostic journey. On average, 31.4 weeks elapsed between symptom onset and initial evaluation, while an additional 14.2 weeks elapsed between the first evaluation and diagnosis. The maximum time from symptom onset to first evaluation was 356.4 weeks (6.8 years), while the maximum time from first evaluation to diagnosis was 305.3 weeks (5.9 years). These findings are similar to those of previous studies that also demonstrated that AD and dementia may be subject to delayed diagnosis.^7,8,20 Underdiagnosis and diagnostic delays are common in patients with AD.^7,8 However, it is important to diagnose patients in the early stages of AD, when individuals may be eligible for disease-modifying therapies and when therapeutic and lifestyle interventions may have the most impact on cognitive decline and functioning.^2,4,6,9,21 Individuals who are diagnosed in earlier dementia stages are less likely to report engaging in potentially unsafe activities, such as driving, preparing hot meals, and managing medications.²² Individuals with milder AD may also have lower AD-related health care resource utilization compared with those with more severe AD, demonstrating the impact of early diagnosis and access to treatment on health care burden.³

In this study, patients with MCI due to AD reported some comorbidities and mild symptoms, while patients with severe disease reported several comorbidities including mental health-related concerns. This reinforces the importance of prompt treatment in patients with early stages of AD. Not only are patients in early disease stages eligible for disease-modifying therapies, but treatment at this stage may also help to slow the progression of symptoms.

In this study, most patients saw a PCP or neurologist at their initial evaluation, demonstrating the key role of these health care providers in the diagnosis of AD. Notably, PCPs made few diagnoses in patients with MCI due to AD, highlighting a diagnostic gap within this provider type. This suggests that increased education and involvement of PCPs in the AD diagnostic journey, especially in early disease stages, could lead to increased diagnosis rates, more effective referrals, and shorter diagnostic journeys.^20,23 Awareness of the signs and symptoms of the early stages of AD should be emphasized in educational efforts. This may help patients and providers to understand and differentiate between age-related and AD-related memory problems and help patients access critical treatments early in the disease process.^7,20 Additionally, the potential emergence of blood-based biomarker tests for clinical use may reduce the need for referrals and specialized expertise in early detection among PCPs, thus reducing diagnostic delays.^2,4 Finally, clinical practice guidelines should be established to help health care providers evaluate disease progression and implement follow-ups and referrals.²³

Patients in this study reported that delays in scheduling an initial evaluation were frequently due to thinking memory problems were a normal part of aging, concerns around perception from others, not wanting to face a serious diagnosis, and the potential for a loss of independence. Physicians reported that appointment availability was an important contributor to diagnostic delays. Diagnostic barriers reported in the literature, including time constraints, limited specialist availability, and lack of confidence from non-specialists in their diagnostic capabilities, are in agreement with those identified in this study.^20,24,25 Insurance coverage and reimbursement issues have also been reported as significant barriers.^20,24 These findings call for patient support programs to better understand AD dementia, overcome emotional burden, and support timely diagnosis.

The data from this study indicate that the number of patient comorbidities may be associated with delays in initial evaluation and diagnosis. Patients with a greater number of comorbidities may be unable to dedicate attention, money, or resources to seeking medical care for memory-related issues. Thus, the disease burden associated with certain comorbidities may hinder the diagnostic process. During our analysis, findings from the Cox proportional hazards model were not consistent. Our results suggest that patients in the early stage of AD have a higher likelihood of receiving both an initial evaluation and a final diagnosis sooner than those in later stages of the disease. However, these conclusions are limited by the small sample sizes and the nature of survey-based data. Therefore, larger studies will be necessary to confirm these findings.

The data from this study reaffirm that functional abilities may be preserved by maintaining a patient in earlier stages of AD. For example, 90.3% of patients with SevAD had severe difficulties with managing finances while mild, moderate, and severe difficulties were reported for 80.6%, 12.9%, and 6.5% of patients with MCI due to AD, respectively. In this study, the severity of multiple symptoms (eg, loss of short-term memory, difficulties in recall of names/words, and others listed in Table 2) was significantly associated with disease severity; however, it is difficult to draw firm conclusions about the data due to the small sample number. Patients self-reported the functional and emotional burden of their disease, including problems with sleeping, remembering names, and misplacing objects. These results demonstrate the importance of diagnosis and preservation of functional abilities in the early disease stages, when maximal benefit may be seen from treatment and lifestyle changes.^4,26

Historically, achieving a diagnosis of AD was limited by access to appropriate cognitive testing and limited tools for biomarker confirmation. However, the landscape is changing rapidly. There is a growing infrastructure for formal cognitive testing, including the potential for digital cognitive testing. In addition, the barriers to achieving a biomarker-confirmed diagnosis seem to be diminishing. Amyloid-PET imaging^27,28 is now covered by the Centers for Medicare and Medicaid Services.²⁹ The future of blood-based biomarkers (eg, plasma phosphorylated tau^30-32) to confirm an AD diagnosis looks promising. There is also growing interest in other non-invasive tools, such as retinal imaging and other digital biomarkers.

Moreover, as more AD diagnoses are made, there will be growing interest in developing new methods to monitor disease progression beyond cognitive assessments (eg, blood-based biomarkers or imaging tools).

Several methodological details limit the generalizability of these findings. First, the DSP is based on a pseudo-random sample of physicians and patients. While minimal inclusion criteria governed the selection of participating physicians, participation was influenced by their willingness to complete the survey. The physician sample may not fully represent the proportion of PCPs and specialists treating patients with AD, and the data will likely be influenced by the relative number of each physician type recruited into the survey. Participating patients may not reflect the general AD population, since the DSP only includes patients who are consulting with their physician; therefore, patients who consult more frequently were more likely to be included in the study. Further, to minimize selection bias, physicians were asked to provide data for a consecutive series of eligible patients, thus ensuring that all patients had an equal chance of being included. Patient eligibility was based on the judgement of the responding physician and not on a formal diagnostic checklist; however, it is representative of the physician’s real-world classification of their patients. This selection bias might explain the higher likelihood of initial evaluation and diagnosis among patients with early-stage AD, as individuals who remain undiagnosed or who were not evaluated may not be represented in the sample. Recall bias, a common survey limitation, might also have affected physician and patient responses. However, physicians were able to refer to patients’ records while completing the patient record form, thus minimizing the possibility of recall bias. Furthermore, data were collected at the time of each patient’s appointment to reduce the likelihood of recall bias where the opinion of the physician was required.

Several sample size limitations exist. First, a small number of patients with SevAD at diagnosis was available, as most patients present to their health care provider before the disease progresses to this stage. MMSE tests are also less likely to be performed on patients with severe disease; therefore, these data should be interpreted with caution. Second, the sample size obtained from voluntarily completed surveys is relatively small, particularly in patients with SevAD, likely biasing some of the results; therefore, caution is required when interpreting these results. Finally, the cross-sectional nature of this study allows for identification of association but not causation.

Conclusions

Patients with AD face a complicated diagnostic journey. Diagnostic delays may be due to factors such as time constraints and availability of appointments. It is important to strive for early diagnosis of AD, as currently available disease-modifying therapies are indicated for patients with earlier disease stages. These therapies may help slow disease progression and preserve patient functional abilities. PCPs and neurologists play a key role in this process and further work should be done to provide improved tools for them to better distinguish between early AD and normal aging. Recent advancements in blood-based biomarker tests may help reduce the need for referrals and speed up the diagnostic process.

Supplemental Material

Supplemental Material - A Real-World Analysis of the Clinical Journey, Diagnosis, and Monitoring Patterns of Patients With Alzheimer Disease by Stage in the United States

Supplemental Material for A Real-World Analysis of the Clinical Journey, Diagnosis, and Monitoring Patterns of Patients With Alzheimer Disease by Stage in the United States by Elnara Fazio-Eynullayeva, MA, Paul Mystkowski, MD, Lei Lv, PhD, MPH, Abdalla Aly, PhD, Sarah Cotton, MA, Nikisha Grant, MSc, Will Johnson, BSc, Soeren Mattke, MD in American Journal of Alzheimer's Disease & Other Dementias®

Footnotes

Acknowledgements

The authors thank Adele Musicant, PhD, of Precision AQ (Bethesda, Maryland) for providing writing and editing assistance in accordance with Good Publication Practice (GPP 2022) guidelines. This assistance was financially supported by Novo Nordisk, Inc.

ORCID iDs

Elnara Fazio-Eynullayeva

Lei Lv

Abdalla Aly

Soeren Mattke

Ethical Considerations

Data collection was undertaken in line with European Pharmaceutical Marketing Research Association guidelines (EPHMRA. Code of conduct. September 2024. Accessed January 21, 2025. ) and, as such, it did not require ethics committee approval. Each survey was performed in full accordance with relevant legislation at the time of data collection. No patients or physicians can be directly identified in the data. Data were aggregated before being shared with the subscriber and/or for publication. This research was submitted to the Pearl Institutional Review Board, study protocol number #22-ADRW-172/3.

Consent to Participate

All participants provided informed consent to take part in the Adelphi DSP survey.

Consent for Publication

All participant data were aggregated and deidentified before receipt by Adelphi Real World.

Author Contributions

EFE: Conceptualization, formal analysis, investigation, methodology, project administration, writing – review & editing.

PM: Conceptualization, formal analysis, investigation, methodology, visualization, writing – review & editing.

LL: Conceptualization, supervision, validation, writing – review & editing.

AA: Conceptualization, formal analysis, investigation, methodology, resources, writing – review & editing.

SC: Formal analysis, validation, visualization, writing – review & editing.

NG: Formal analysis, validation, visualization, writing – review & editing.

WJ: Formal analysis, validation, visualization, writing – review & editing.

SM: Conceptualization, formal analysis, investigation, methodology, validation, writing – review & editing.

Funding

This study was funded by Novo Nordisk Inc. The authors disclosed receipt of the following financial support for the research, authorship, and/or publication of this article: Data collection was undertaken by Adelphi Real World as part of an independent survey, entitled the Adelphi Real World AD DSP. The DSP is a wholly owned Adelphi Real World product. Novo Nordisk is one of multiple subscribers to the DSP. Publication of survey results was not contingent on the subscriber’s approval or censorship of the publication.

Declaration of Conflicting Interests

The authors declared the following potential conflicts of interest with respect to the research, authorship, and/or publication of this article: EFE, PM, LL, and AA are employed by and hold stocks in Novo Nordisk. SC, NG, and WJ are employees of Adelphi Real World. SM serves on the board of directors for Senscio Systems; serves on scientific advisory boards for ALZpath and Boston Millennia Partners; and has received consulting and/or speaker fees from Biogen, C2N Diagnostics, Eisai, Eli Lilly, Novartis, Novo Nordisk, and Genentech/Roche.

Data Availability Statement

All data (ie, methodology, materials, data, and data analysis) that support the findings of this survey are the intellectual property of Adelphi Real World. All requests for access should be addressed directly to Sarah Cotton at sarah.cotton@adelphigroup.com. Sarah Cotton is an employee of Adelphi Real World.*

Supplemental Material

Supplemental material for this article is available online.

References

Scheltens

De Strooper

Kivipelto

, et al. Alzheimer's disease. Lancet. 2021;397(10284):1577-1590.

Alzheimer’s Association . 2023 Alzheimer’s disease facts and figures. Alzheimers Dement. 2023;19(4):1598-1695.

Fazio-Eynullayeva

Cunnington

Mystkowski

, et al. Real-world healthcare resource utilization of Alzheimer's disease in the early and advanced stages: a retrospective cohort study. J Med Econ. 2025;28(1):81-88.

Liss

Seleri Assuncao

Cummings

, et al. Practical recommendations for timely, accurate diagnosis of symptomatic Alzheimer's disease (MCI and dementia) in primary care: a review and synthesis. J Intern Med. 2021;290(2):310-334.

Knopman

Amieva

Petersen

, et al. Alzheimer disease. Nat Rev Dis Primers. 2021;7(1):33.

Tahami Monfared

Byrnes

White

Zhang

. Alzheimer's disease: epidemiology and clinical progression. Neurol Ther. 2022;11(2):553-569.

Lang

Clifford

Wei

, et al. Prevalence and determinants of undetected dementia in the community: a systematic literature review and a meta-analysis. BMJ Open. 2017;7(2):e011146.

White

Ingraham

Larson

Fishman

Park

Coe

. Observational study of patient characteristics associated with a timely diagnosis of dementia and mild cognitive impairment without dementia. J Gen Intern Med. 2022;37(12):2957-2965.

Cummings

Rabinovici

Atri

, et al. Aducanumab: appropriate use recommendations update. J Prev Alzheimers Dis. 2022;9(2):221-230.

10.

Varadharajan

Davis

Ghosh

, et al. Guidelines for pharmacotherapy in Alzheimer’s disease - a primer on FDA-approved drugs. J Neurosci Rural Pract. 2023;14(4):566-573.

11.

Cummings

Zhou

Lee

Zhong

Fonseca

Cheng

. Alzheimer’s disease drug development pipeline: 2024. Alzheimers Dement. 2024;10:e12465.

12.

US Food & Drug Administration . FDA approves treatment for adults with Alzheimer's disease. Updated July 2, 2024. Accessed July 11, 2024.https://www.fda.gov/drugs/news-events-human-drugs/fda-approves-treatment-adults-alzheimers-disease#:∼:text=The_U.S._Food_and_Drug,studied_in_the_clinical_trials

13.

Hoy

. Lecanemab: first approval. Drugs. 2023;83(4):359-365.

14.

Anderson

Benford

Harris

Karavali

Piercy

. Real-world physician and patient behaviour across countries: Disease-Specific Programmes - a means to understand. Curr Med Res Opin. 2008;24(11):3063-3072.

15.

Anderson

Higgins

Courcy

, et al. Real-world evidence generation from patients, their caregivers and physicians supporting clinical, regulatory and guideline decisions: an update on Disease Specific Programmes. Curr Med Res Opin. 2023;39(12):1707-1715.

16.

Babineaux

Curtis

Holbrook

Milligan

Piercy

. Evidence for validity of a national physician and patient-reported, cross-sectional survey in China and UK: the Disease Specific Programme. BMJ Open. 2016;6(8):e010352.

17.

Higgins

Piercy

Roughley

, et al. Trends in medication use in patients with type 2 diabetes mellitus: a long-term view of real-world treatment between 2000 and 2015. Diabetes Metab Syndr Obes. 2016;9:371-380.

18.

Perneczky

Wagenpfeil

Komossa

Grimmer

Diehl

Kurz

. Mapping scores onto stages: mini-mental state examination and clinical dementia rating. Am J Geriatr Psychiatry. 2006;14(2):139-144.

19.

von

Altman

Egger

, et al. The Strengthening the Reporting of Observational Studies in Epidemiology (STROBE) statement: guidelines for reporting observational studies. Int J Surg. 2014;12(12):1495-1499.

20.

Liu

Jun

Becker

Wallick

Mattke

. Detection rates of mild cognitive impairment in primary care for the United States Medicare population. J Prev Alzheimers Dis. 2024;11(1):7-12.

21.

Budd

Burns

Guo

L’Italien

Lapuerta

. Impact of early intervention and disease modification in patients with predementia Alzheimer’s disease: a Markov model simulation. Clinicoecon Outcomes Res. 2011;3:189-195.

22.

Amjad

Roth

Samus

Yasar

Wolff

. Potentially unsafe activities and living conditions of older adults with dementia. J Am Geriatr Soc. 2016;64(6):1223-1232.

23.

Mattke

Batie

Chodosh

, et al. Expanding the use of brief cognitive assessments to detect suspected early-stage cognitive impairment in primary care. Alzheimers Dement. 2023;19(9):4252-4259.

24.

Mattke

Wang

. Implications of Alzheimer's treatment for organization and payment of medical practices in the United States . https://cesr.usc.edu/sites/default/files/Implications_20of_20Alzheimer_27s_20Treatment_20for_20Organization_20and_20Payment_20of_20Medical_20Practices_20in_20the_20United_20States_20_282020_29.pdf. Accessed 11 July 2024.

25.

Drabo

Barthold

Joyce

Ferido

Chang Chui

Zissimopoulos

. Longitudinal analysis of dementia diagnosis and specialty care among racially diverse Medicare beneficiaries. Alzheimers Dement. 2019;15(11):1402-1411.

26.

Isaacson

Hristov

Saif

, et al. Individualized clinical management of patients at risk for Alzheimer’s dementia. Alzheimers Dement. 2019;15(12):1588-1602.

27.

Amadoru

Dore

McLean

, et al. Comparison of amyloid PET measured in Centiloid units with neuropathological findings in Alzheimer's disease. Alzheimers Res Ther. 2020;12(1):22.

28.

Chapleau

Iaccarino

Soleimani-Meigooni

Rabinovici

. The role of amyloid PET in imaging neurodegenerative disorders: a review. J Nucl Med. 2022;63(Suppl 1):13S-19S.

29.

Alzheimer's Association . Alzheimer’s association applauds CMS decision to cover PET imaging for Alzheimer’s disease diagnosis. Updated October 13, 2023. https://www.alz.org/news/2023/medicare-covers-pet-imaging-alzheimers-diagnosis#:∼:text=Amyloid_PET_imaging_enables_clinicians,of_one_scan_per_lifetime. Accessed 12 July 2024.

30.

Gonzalez-Ortiz

Kac

Brum

Zetterberg

Blennow

Karikari

. Plasma phospho-tau in Alzheimer's disease: towards diagnostic and therapeutic trial applications. Mol Neurodegener. 2023;18(1):18.

31.

Kac

Gonzalez-Ortiz

Simren

, et al. Diagnostic value of serum versus plasma phospho-tau for Alzheimer's disease. Alzheimers Res Ther. 2022;14(1):65.

32.

Palmqvist

Tideman

Cullen

, et al. Prediction of future Alzheimer's disease dementia using plasma phospho-tau combined with other accessible measures. Nat Med. 2021;27(6):1034-1042.

Supplementary Material

Please find the following supplemental material available below.

For Open Access articles published under a Creative Commons License, all supplemental material carries the same license as the article it is associated with.

For non-Open Access articles published, all supplemental material carries a non-exclusive license, and permission requests for re-use of supplemental material or any part of supplemental material shall be sent directly to the copyright owner as specified in the copyright notice associated with the article.

0.00 MB

0.41 MB