Increased Odds of Incident Alzheimer’s Disease and Related Dementias in Presence of Common Non-Cancer Chronic Pain Conditions in Appalachian Older Adults

Abstract

Background

There is a growing concern regarding the increasing prevalence of common non-cancer chronic pain conditions (NCPCs) and their possible association with Alzheimer’s disease and related dementias (ADRD). However, large population-based studies are limited, especially in Appalachian and other predominantly rural, underserved populations who suffer elevated prevalence of both NCPCs and known ADRD risk factors.

Objectives

We investigated the relation of NCPC to risk of incident ADRD in older Appalachian Medicare beneficiaries and explored the potential mediating effects of mood and sleep disorders.

Methods

Using a retrospective cohort design, we assessed the overall and cumulative association of common diagnosed NCPCs at baseline to incident ADRD in 161,573 elders ≥65 years, Medicare fee-for-service enrollees, 2013–2015. NCPCs and ADRD were ascertained using claims data. Additional competing risk for death analyses accounted for potential survival bias.

Main Findings

Presence of any NCPC at baseline was associated with significantly increased odds for incident ADRD after adjustment for covariates [adjusted odds ratio (AOR) = 1.26 (1.20, 1.32), p < .0001]. The magnitude and strength of this association increased significantly with rising burden of NCPCs at baseline [AOR for ≥4 vs. no NCPC = 1.65 (1.34, 2.03), p-trend = .01]. The addition of depression and anxiety, but not sleep disorders, modestly attenuated these associations [AORs for any NCPC and ≥4 NCPCs, respectively = 1.16 (1.10, 1.22) and 1.39 (1.13, 1.71)], suggesting a partial mediating role of mood impairment. Sensitivity analyses, multinomial logistic regressions accounting for risk of death, yielded comparable findings.

Conclusion

In this large cohort of older Appalachian Medicare beneficiaries, baseline NCPCs showed a strong, positive, dose–response relationship to odds for incident ADRD; this association appeared partially mediated by depression and anxiety. Further longitudinal research in this and other high-risk, rural populations are needed to evaluate the causal relation between NCPC and ADRD.

Keywords

Alzheimer’s disease and related dementias Appalachia chronic pain older adults rural population retrospective cohort study West Virginia Medicare

Introduction

As of year 2021, 6.2 million US adults over age 65 are living with Alzheimer’s dementia (Alzheimer’s Association Facts and Figures, 2021), the most common condition among Alzheimer’s disease and related dementias (ADRD), a constellation of debilitating, irreversible neurodegenerative disorders (McNeill, 2011). In addition to its rising prevalence and subsequent functional impairment and disability in the older population, ADRD-related mortality has risen disproportionately over the last two decades, surpassing the mortality owing to heart diseases. ADRD is also associated with enormous economic burden both in the United States and elsewhere (Nichols et al., 2019; Takizawa et al., 2015); for example, in the United States the total Medicare and Medicaid spending toward ADRD care in the fiscal year 2021 has been estimated at $239 billion (Alzheimer’s Association Facts and Figures, 2021). Elevated risk for ADRD has been linked with factors such as older age, female sex, non-Hispanic Black, and certain genes, as well as with a number of modifiable factors, including low education, poverty, sedentary lifestyle, smoking, and alcohol consumption (Baumgart et al., 2015; Cooper et al., 2015; Moon et al., 2019). Moreover, individuals with ADRD are more likely to have other chronic disorders, including midlife hypertension and obesity, diabetes, heart and cerebrovascular diseases, chronic respiratory diseases, as well as depression, anxiety, and sleep disorders (Baumgart et al., 2015; Cooper et al., 2015; Julien et al., 2017; Zhou et al., 2015).

In addition, non-cancer chronic pain conditions (NCPCs) common in older populations have not only been associated with higher rates of disability, poor quality of life, increased healthcare utilization (Blyth et al., 2004; McWilliams et al., 2004; Nicholson & Verma, 2004; Rice et al., 2016), and a range of adverse physical health outcomes (Nicholson & Verma, 2004; Dominick et al., 2012; Gibson & Lussier, 2012; Smith et al., 2014; Fayaz et al., 2016), but may also be an important contributor to ADRD. A growing body of evidence from experimental (Moriarty et al., 2011; Cao et al., 2019) and cross-sectional (Van Der Leeuw et al., 2015; Ikram et al., 2019) studies suggests significant neurocognitive impairment, including deterioration in attention, memory, and executive function with chronic pain and certain NCPCs, as well as adverse neurostructural and neurofunctional changes paralleling those implicated in ADRD pathogenesis (Ng et al., 2018; Cao et al., 2019). Longitudinal studies from North America (Whitlock et al., 2017; van der Leeuw et al., 2018; Ezzati et al., 2019; Khalid et al., 2020), Europe (Veronese et al., 2018), and East Asia (Yang et al., 2016; Tzeng et al., 2018; Yamada et al., 2019) also suggest significant associations of chronic pain and certain NCPCs to subsequent deterioration in memory (Whitlock et al., 2017; van der Leeuw et al., 2018), accelerated cognitive decline (Whitlock et al., 2017), new onset cognitive impairment (Whitlock et al., 2017; van der Leeuw et al., 2018), and incident ADRD (Ezzati et al., 2019; Innes & Sambamoorthi, 2020; Khalid et al., 2020).

Population-based evidence primarily focused on nationally representative populations (Whitlock et al., 2017; Khalid et al., 2020) with a small number of US-based studies conducted at county level (van der Leeuw et al., 2018; Ezzati et al., 2019); however, the longitudinal association of chronic pain to subsequent adverse cognitive outcomes remains underexplored in rural, underserved populations, who exhibit historically high prevalence of both chronic pain (Johannes et al., 2010; Goldberg & McGee, 2011; Jackson et al., 2014; Dahlhamer et al., 2018; Guglielmo et al., 2019) and ADRD risk factors (Shaw et al., 2004; Serrano et al., 2007; Schwartz et al., 2009; Centers of Disease Control (CDC) West Virginia - State Nutrition, Physical Activity, and Obesity Profile, 2012; Abner et al., 2016; Akushevich et al., 2018; Mamudu et al., 2017; Mattos et al., 2017).

In the United States, chronic pain prevalence is particularly elevated in Appalachian adults (Centers for Medicare and Medicaid Services (CMS) Medicare Reports, 2018), populations characterized by poverty and poor access to health care (Seufert & Carrozza, 2004). For example, state-specific data from the Behavioral and Risk Factor Surveillance Survey (BRFSS) in 2018 indicated prevalence of arthritis in West Virginia (WV) (BRFSS, 2018), the only US state situated entirely within Appalachia, to be 39%, the highest in the nation (Dahlhamer et al., 2018). Likewise, WV adults not only include a disproportionately high percentage of elders and rural residents (US Census Bureau QuickFacts, 2019) but suffer a high burden of other ADRD risk factors (Kivipelto et al., 2001; Baumgart et al., 2015; Cooper et al., 2015; Zhou et al., 2015) including socioeconomic factors [poor educational attainment, low household income, and high unemployment (Seufert & Carrozza, 2004; Shaw et al., 2004; Behringer & Friedell, 2006; Hendryx, 2010)] as well as specific lifestyle characteristics and chronic health conditions (Serrano et al., 2007; Schwartz et al., 2009; Holt et al., 2011; Herath & Brown, 2013; Schoenberg et al., 2015; Hege et al., 2017; Mamudu et al., 2017), factors also linked to chronic pain (Verkaik et al., 2007; Shiri et al., 2010b; Dominick et al., 2012; Gibson & Lussier, 2012; Okifuji & Hare, 2015; Fayaz et al., 2016; Haaksma et al., 2017). Notably, WV has the highest prevalence in the nation of many modifiable health-related risk factors for ADRD, including physical inactivity, tobacco smoking, cardiovascular disease, chronic kidney disease, diabetes, chronic obstructive lung disease, midlife hypertension and obesity, and multimorbidity (United Health Foundation, America’s Health Rankings: Annual report, West Virginia, 2020).

Yet to our knowledge, only one study to date has investigated the association of chronic pain or NCPCs to ADRD risk in Appalachian or other impoverished, rural, medically underserved populations (Innes & Sambamoorthi, 2017). In this cross-sectional investigation of a large sample of Ohio Valley residents, osteoarthritis and related joint pain were strongly and positively associated with perceived memory loss in a dose-dependent manner, an association that appeared mediated in part by impairment in mood and sleep. Mood and sleep disorders have been linked bidirectionally with chronic pain conditions owing to a number of overlapping neurobiological mechanisms (Menefee et al., 2000; Finan & Smith, 2013; Hooten, 2016) and therefore qualify as potential effect mediators.

However, there is a scarcity of investigations on the association of common NCPCs and NCPC burden to subsequent risk of incident ADRD in older Appalachian or other rural, high-risk populations, and the potential mediating influence of mood and sleep disorders, conditions common in this population (Liu et al., 2016; BRFSS, 2018) and linked to both NCPC and ADRD (Menefee et al., 2000; Ownby et al., 2006; Bubu et al., 2016; Hooten, 2016), remains to be explored.

In this large retrospective cohort study of West Virginia Medicare beneficiaries, we investigate the association of common NCPCs and NCPC burden to risk for incident ADRD and assess, as a secondary aim, the potential mediating influence of mood and sleep disorders on these associations. The study’s primary hypotheses were as follows: 1) presence of diagnosed common NCPCs at baseline would be associated with increased risk of ADRD at follow-up and 2) the magnitude of this association would increase with rising burden of baseline NCPCs. We also hypothesized a potential role of mood and sleep disorders as mediators in these associations.

Methods

Design and Data Source

Risk for incident ADRD in Appalachian population with chronic pain and NCPC burden at baseline was investigated using a retrospective cohort study design using data from ADRD-free West Virginia (WV) Medicare fee-for-service beneficiaries (years 2013–2015) with baseline (year 1, 2013) and follow-up (year 2 and 3, 2014 and 2015) periods.

The WV Medicare health insurance program is available for all WV permanent residents 65 years or older or with an eligible disability. Compared to the 19% national enrollment average, 24.4% of West Virginia’s population is enrolled in the Medicare health insurance program due to relatively high percentage of disabled (23% of the Medicare enrollees, part A and B) and older adults (77% of the Medicare enrollees, part A and B) in West Virginia (Centers for Medicare and Medicaid Services (CMS) Medicare Reports, 2018). The WV Medicare fee-for-service (FFS) files are claims-based data from inpatient (IP), skilled nursing facility (SNF), outpatient (OT), home health agency (HHA), and physician office (PO) health services utilization. Linked Medicare administrative records and FFS files were used to collect information on sociodemographics (age, sex, race, region, and Medicaid coverage) and medical claims. Zip-Code data files were linked with administrative records to measure Zip-Code–level percent education attainment and household income.

Inclusion and Exclusion Criteria

The primary study cohort included non-institutionalized, Medicare FFS beneficiaries with continuous enrollment, who were age 65 years or older at baseline and still alive at end of follow-up. Participants diagnosed with ADRD at baseline (8.4%) were excluded from the study. Stepwise application of all a priori exclusion criteria yielded a final study cohort of 161,573 individuals. The present study was approved as an exempt protocol by the WVU Institutional Review Board, which identified the research as a non-human subject research.

Study Measures

Incident ADRD

Medicare FFS claims for IP, SNF, OT, HHA, and PO for years 2013–2015 were used to identify ADRD status at baseline and follow-up years. Consistent with recommendations of the Center for Medicare and Medicaid Services (CMS) (Medicare Current Beneficiary Survey (MCBS), CMS Medicare and Medicaid Research Data, 2020), ADRD was identified as “at least one FFS claim with any of the following International Classification of Diseases, ninth Edition, clinical modification (ICD-9-CM) diagnostic codes: 290.0–290.3, 331.0–331.2, 331.7, and 331.8 (Lin et al., 2010) or ICD-10-CM codes: G30.xx and F02.xx” (Forbes et al., 2018). ADRD status identified from FFS claims was used to ascertain both prevalent ADRD (for purposes of exclusion) and incident ADRD at follow-up times 1 and 2.

NCPCs and Number of NCPCs

Medicare FFS claims for IP, SNF, OT, HHA, and PO during baseline year 2013 (01/01/2013–12/31/2013) were used to ascertain baseline NCPC status, which involved identification of five common NCPCs: back or neck pain, headache, joint pain, neuropathic pain, and osteoarthritis. Presence of any NCPC was identified using either two outpatient claims (90 days apart) or one inpatient claim using ICD-9-CM codes as recommended by CMS (Chronic Conditions Data Warehouse: Your source for national CMS Medicare and Medicaid research data, Access Yr 2020) and consistent with prior studies of NCPC (Edlund et al., 2010; Sullivan et al., 2008). Any NCPC was assessed as a binary variable (yes/no) during baseline. Relative NCPC burden was ascertained with a categorical variable (0 to ≥4 NCPCs).

Other Variables

Covariates were selected a priori based on published research documenting their association with ADRD risk and/or chronic pain (Baumgart et al., 2015; Cooper et al., 2015; Kivipelto et al., 2001; Musicco et al., 2009; Qiu et al., 2010) for inclusion in our multivariable models. Covariates included demographic factors: age-group (65–69, 70–74, 75–79, and ≥80 years), sex (female/male), and race/ethnicity (non-Hispanic white/non-Hispanic Black/other). College education and household income were measured using Zip-Code–level demographic data by linking participants’ Zip-Code information with state-wide median education or income level derived from 2000 US Census data. Participants’ status on education and income is represented as percentiles based on state-wide median data [college education and household income (1= 0–25, 2 = 26–50, 51–75, 76–100 percentile)] (Healthcare Cost and Utilization Project (HCUP). Quartile classification of patient’s zip code, 2019). Insurance and geographical region were measured as: health insurance status (Medicaid insurance (yes/no), the West Virginia Department of Health and Human Resources (WVDHHR) North (I&III) and South (II&IV) geographical regions (“WVDHHR Regions Description”, Access Year 2020). Other covariates include lifestyle factors: smoking status (current smoker/non-smoker) and obesity (yes/no); history of stroke or traumatic brain injury (TBI); and chronic health conditions, including hypertension, diabetes, heart disease, respiratory illness, and cancer, as well as specific auto-immune conditions associated with chronic pain, including rheumatoid arthritis (RA), and systemic lupus erythematosus (lupus). Analgesic use, defined as baseline use (yes/no) of non-steroidal anti-inflammatory medications (NSAIDS) and opioid analgesics, was ascertained using records of reimbursed prescription claims during the baseline year.

Potential mediators: Baseline depression and anxiety disorders as well as insomnia-related sleep disorders were ascertained using medical claims data.

Statistical Analysis

Rao–Scott chi-square tests were used to evaluate baseline study characteristics in relation to NCPC and ADRD status. The associations of NCPC presence (Y/N) and burden (i.e., number of NCPCs) to risk for incident ADRD were assessed using multivariable logistic regressions. We assessed the linear effects of NCPC burden on risk for incident ADRD using polynomial contrasts. Covariates were added block-wise to logistic regression models to allow incremental adjustment for baseline sociodemographic and related factors, lifestyle characteristics, chronic physical health conditions, and medication use, with final models including all covariates. To evaluate the potential mediating influence of mood disorders (defined as a diagnosis of depression and/or anxiety) and insomnia-related sleep disorders on the relation of NCPCs to ADRD risk, we followed a two-step process: (1) we first confirmed the association of these disorders to both baseline NCPCs and incident ADRD and (2) we then included these disorders, both separately (diagnosed depression and anxiety) and in combination (diagnosed depression, anxiety, and insomnia-related disorders), in our fully adjusted logistic regression models to assess the relative effects on increased odds of ADRD.

In addition to the primary analyses, sensitivity analyses were also conducted to account for potential survival bias. Specifically, to assess competing risk of death (with and without a diagnosis of ADRD) into our analyses, we built models using multinomial logistic regression with four outcome categories: ADRD-free and alive at the end of follow-up (n = 153,075) (reference category); ADRD-free and not alive at the end of follow-up (n = 10,599); ADRD positive and alive at the end of follow-up (n=8498); and ADRD positive and not alive at the end of follow-up (n = 2331). All analyses were performed using SAS survey procedure (SAS version 9.4, SAS Institute, Inc.).

Results

Of the 161,573 eligible Medicare beneficiaries comprising our Appalachian study cohort, the majority were female (55.7%) and non-Hispanic white (97%) (Table 1). Overall baseline prevalence of NCPC was 41%, with differing prevalence across individual chronic pain conditions (range 7.2%–50%). Our study population had a high comorbid pain burden, with 38.7% study participants with at least two concurrent chronic pain conditions whereas 17% indicated having three or more comorbid pain conditions. A higher proportion of participants with baseline NCPC versus no baseline had chronic physical and mental health conditions that are risk factors for ADRD, including hypertension (89% vs. 70%), heart disease (44% vs. 29%), stroke (7.5% vs. 3.4%), diabetes (44.5% vs. 32%), depression (28% vs. 12%) and anxiety (19.7% vs. 8%), as well as sleep disorders (32% vs. 11.7%) (Table 1).

Table 1.

Baseline Characteristics by NCPC in the Cohort of N = 161,573 Adults ≥ 65 Years, Using Fee-For-Service West Virginia Medicare Claims, 2013–2015.

	Total		Any NCPC*				p-value
			Yes		No
	Total		N	%	N	%		N	%
ALL ^¥	161,573	100	65,891	40.8	95,682	59.2
Sociodemographic
Sex							<.0001
Female	89,958	55.7	42,290	64.2	47,668	49.8
Male	71,615	44.3	23,601	35.8	48,014	50.2
Age in years							<.0001
65–69	57,735	35.7	22,106	33.5	35,629	37.2
70–74	42,562	26.3	17,032	25.8	25,530	26.7
75–79	29,782	18.4	12,600	19.1	17,182	18.0
80 +	31,494	19.5	14,153	21.5	17,341	18.1
Race/ethnicity							<.0001
Non-Hispanic white	156,489	96.9	63,966	97.1	92,523	96.7
Non-Hispanic Black	3427	2.10	1427	2.20	2000	2.10
Other	1134	.70	349	.53	785	.82
College education⁺⁺							<.0001
0–25%	36,133	23.6	14,197	22.7	21,936	24.2
26–50%	39,340	25.7	16,187	25.9	23,153	25.5
51–75%	39,308	25.7	16,519	26.5	22,789	25.1
76–100%	38,349	25.0	15,523	24.9	22,826	25.2
Household income⁺⁺							<.0001
0–25%	36,054	23.5	13,944	22.3	22,110	24.4
26–50%	38,679	25.3	16,014	25.7	22,665	25.0
51–75%	40,368	26.4	16,489	26.4	23,879	26.3
76–100%	38,029	24.8	15,979	25.6	22,050	24.3
Insurance
Medicaid insurance							<.0001
Yes	20,734	12.8	10,184	15.5	10,550	11.0
No	140,839	87.2	55,707	84.5	85,132	89.0
Geography
West Virginia region⁺⁺⁺							<.0001
1	43,291	27.0	16,362	25.0	26,929	28.4
2	38,809	24.2	15,575	23.8	23,234	24.5
3	43,388	27.1	18,666	28.5	24,722	26.0
4	34,881	21.8	14,808	22.6	20,073	21.1
Lifestyle characteristics
Obesity							<.0001
Yes	14,332	8.9	9115	13.8	5217	5.5
No	147,241	91.1	56,776	86.2	90,465	94.5
Smoking status							<.0001
Current	10,576	6.5	5706	8.7	4870	5.1
Non-smoker	150,997	93.5	60,185	91.3	90,812	94.9
Baseline health history
Physical health conditions
Hypertension							<.0001
Yes	125,807	77.9	58,779	89.2	67,028	70.1
No	35,766	22.1	7112	10.8	28,654	29.9
Heart disease							<.0001
Yes	57,391	35.5	29,265	44.4	28,126	29.4
No	104,182	64.5	36,626	55.6	67,556	70.6
Stroke							<.0001
Yes	8166	5.1	4924	7.5	3242	3.4
No	153,407	94.9	60,967	92.5	92,440	96.6
Diabetes							<.0001
Yes	59,909	37.1	29,291	44.5	30,618	32.0
No	101,664	62.9	36,600	55.5	65,064	68.0
Asthma							<.0001
Yes	13,954	8.6	8356	12.7	5598	5.9
No	147,619	91.4	57,535	87.3	90,084	94.1
Respiratory disease							<.0001
Yes	36,597	22.7	19,941	30.3	16,656	17.4
No	124,976	77.3	45,950	69.7	79,026	82.6
Cancer							<.0001
Yes	39,659	24.5	19,855	30.1	19,804	20.7
No	121,914	75.5	46,036	69.9	75,878	79.3
Traumatic brain injury							<.0001
Yes	3311	2.0	2452	3.7	859	0.9
No	158,262	98.0	63,439	96.3	94,823	99.1
Rheumatoid arthritis							<.0001
Yes	4282	2.7	3201	4.9	1081	1.1
No	157,291	97.3	62,690	95.1	94,601	98.9
Lupus							<.0001
Yes	1287	0.8	977	1.5	310	0.3
No	160,286	99.2	64,914	98.5	95,372	99.7
Number of NCPC*							<.0001
0	59,340	36.7	—	—	—	—
1	39,646	24.5	—	—	—	—
2	34,915	21.6	—	—	—	—
3	20,009	12.4	—	—	—	—
≥4	7663	4.74	—	—	—	—
Sleep and Mood disorders
Sleep disorder							<.0001
Yes	32,268	20.0	21,051	31.9	11,217	11.7
No	129,305	80.0	44,840	68.1	84,465	88.3
Depression							<.0001
Yes	29,958	18.5	18,325	27.8	11,633	12.2
No	131,615	81.5	47,566	72.2	84,049	87.8
Anxiety							<.0001
Yes	20,506	12.7	12,982	19.7	7524	7.9
No	141,067	87.3	52,909	80.3	88,158	92.1
Medication use
NSAIDs							<.0001
Yes	23,167	14.3	15,397	23.4	7770	8.1
No	138,406	85.7	50,494	76.6	87,912	91.9
Opioid analgesics							<.0001
Yes	35,648	22.1	25,066	38.0	10,582	11.1
No	125,925	77.9	40,825	62.0	85,100	88.9

NSAIDs, non-steroidal anti-inflammatory drugs.

*Non-cancer chronic pain conditions (NCPCs) include osteoarthritis, joint pain, headache pain or migraine, back and neck pain, and neuropathic pain conditions; ^¥Numbers within each variable category may not total of N = 161,573, due to missing data; ^§p-values are based on statistically significant difference in variable levels across NCPC status, examined with Rao–Scott chi-square test; ⁺⁺Participants’ education and income status as percentiles based on linked Zip Code and state-wide median data; ⁺⁺⁺WV region North (1 and 3), South (2 and 4).

Baseline study characteristics within incident ADRD and ADRD-free categories are summarized in Table 2. Overall incidence rate of ADRD in the study was 5.3 per 100 participants, with a total of 8,498 participants diagnosed with incident ADRD during the two follow-up years. Significant differences were observed in incident ADRD rates by baseline NCPC status (6.8 vs. 4.2 per 100 in those with vs. without NCPC, p < .0001). ADRD incidence rates were significantly higher in women (6.1% vs. 4.2% in men), non-Hispanic Black (6.0% vs. 5.3% in non-Hispanic whites), and age 80+ years (12% vs. 2.2% in 65–69 years age) (ps < .0001). Diagnosis of incident ADRD was also significantly higher in individuals with Medicaid insurance (8.4% vs. 4.8% in non-Medicaid) or residence in WVDHHR South regions (5.6–5.7 vs. <5.2 in North regions) (ps < .0001). Moreover, incident ADRD rates in this study were significantly higher in participants with a history of chronic conditions, including hypertension, heart disease, stroke, diabetes, and history of brain injury (ps < .0001). In addition, presence of sleep, depression, or anxiety disorders at baseline also corresponded with significantly higher rates of incident ADRD (ps < .0001). Particularly, as shown in Table 2, incident ADRD rates were significantly higher with increasing burden of pain conditions, that is, 5.2 per 100 participants with low pain burden to 9.2 per 100 participants in those with high pain burden (ps < .0001).

Table 2.

Baseline Characteristics by Incident ADRD in the Cohort of N = 161,573 Older Adults (≥65 Years), Using Fee-For-Service West Virginia Medicare Claims, 2013–2015.

	Incident ADRD				p value^§
	Yes		No
	N	%	N	%
ALL ^¥	8498	5.3	153,075	94.7
Sociodemographic
Sex					<.0001
Female	5487	6.1	84,467	93.9
Male	3008	4.2	68,608	95.8
Age in years					<.0001
65–69	1271	2.2	56,493	97.8
70–74	1533	3.6	41,037	96.4
75–79	1817	6.1	27,973	93.9
80+	3939	12.5	27,572	87.5
Race/ethnicity					.001
Non-Hispanic white	8297	5.3	148,254	94.7
Non-Hispanic Black	206	6.0	3223	94.0
Other	50	4.41	1084	95.6
College education⁺⁺					.054
0–25%	1843	5.1	34,302	94.9
26–50%	2086	5.3	37,274	94.7
51–75%	2082	5.3	37,208	94.7
76–100%	1878	4.9	36,453	95.1
Household income⁺⁺					.051
0–25%	1874	5.2	34,168	94.8
26–50%	2049	5.3	36,618	94.7
51–75%	1978	4.9	38,386	95.1
76–100%	1978	5.2	36,065	94.8
Insurance
Medicaid					<.0001
Yes	1742	8.4	18,991	91.6
No	6761	4.8	134,084	95.2
Geography
West Virginia region⁺⁺⁺					<.0001
1	1989	4.6	41,302	95.4
2	2200	5.7	36,609	94.3
3	2247	5.2	41,141	94.8
4	1969	5.6	32,912	94.4
Lifestyle characteristics
Obesity					.101
Yes	717	5.0	13,620	95.0
No	7805	5.3	139,455	94.7
Smoking status					.005
Current	613	5.8	9958	94.2
Non-smoker	7850	5.2	143,117	94.8
Baseline health history
Physical health conditions
Hypertension					<.0001
Yes	7419	5.9	118,325	94.1
No	1001	2.8	34,750	97.2
Heart disease					<.0001
Yes	4249	7.4	53,166	92.6
No	4271	4.1	99,909	95.9
Stroke					<.0001
Yes	1012	12.4	7151	87.6
No	7519	4.9	145,924	95.1
Diabetes					<.0001
Yes	3654	6.1	56,248	93.9
No	4882	4.8	96,827	95.2
Asthma					<.0001
Asthma	851	6.1	13,102	93.9
No Asthma	7678	5.2	139,973	94.8
Respiratory disease					<.0001
Yes	2670	7.3	33,911	92.7
No	5877	4.7	119,164	95.3
Cancer					.031
Yes	2182	5.5	37,490	94.5
No	6340	5.2	115,585	94.8
Traumatic brain injury					<.0001
Yes	493	14.9	2817	85.1
No	8075	5.1	150,258	94.9
Rheumatoid arthritis					<.0001
Yes	283	6.6	4000	93.4
No	8177	5.2	149,075	94.8
Lupus					.506
Yes	73	5.7	1214	94.3
No	8499	5.3	151,861	94.7
Any NCPC⁺⁺					<.0001
Yes	4458	6.8	61,433	93.2
No	4040	4.2	91,642	95.8
Number of NCPCs					<.0001
0	2254	3.8	57,060	96.2
1	2062	5.2	37,583	94.8
2	1955	5.6	32,955	94.4
3	1421	7.1	18,597	92.9
≥4	783	9.2	6880	90.8
Sleep and Mood disorders
Sleep disorder					<.0001
Yes	2387	7.4	29,872	92.6
No	6076	4.7	123,203	95.3
Depression					<.0001
Yes	2786	9.3	27,169	90.7
No	5657	4.3	125,906	95.7
Anxiety					<.0001
Yes	1578	7.7	18,917	92.3
No	6912	4.9	134,158	95.1
Medication use
NSAIDs					.286
Yes	1251	5.4	21,915	94.6
No	7194	5.2	131,160	94.8
Opioid analgesics					<.0001
Yes	2425	6.8	33,235	93.2
No	6042	4.8	119,840	95.2

NSAIDs, non-steroidal anti-inflammatory drugs.

*Non-cancer chronic pain conditions (NCPCs) include osteoarthritis, joint pain, headache pain or migraine, back and neck pain, and neuropathic pain conditions; ^¥Numbers within each variable category may not total of N = 8,498, due to missing data; ^§p-values are based on statistical difference across ADRD status, examined with Rao–Scott chi-square test; ⁺⁺Participants’ education and income status as percentiles based on linked Zip Code and state-wide median data; ⁺⁺⁺WV region North (1 and 3), South (2 and 4).

Association of ADRD Risk to Baseline NCPC and NCPC Burden

The likelihood of developing ADRD was 66% higher in participants with any baseline NCPC compared to those without baseline NCPC [unadjusted odds ratio (OR) = 1.66; 95% confidence interval (CI) = 1.58–1.73; p < .0001] (Table 3). Covariates adjustment for sociodemographic factors attenuated the odds for developing ADRD to 48% [adjusted OR (AOR) = 1.48; CI = 1.41–1.56; p < .0001]. The association was further diluted after additional adjustment for lifestyle factors, chronic physical health conditions, and analgesics use but remained statistically significant (AOR = 1.26; CI 1.20–1.32; p < .0001). As shown in Table 3, the association of NCPC to incident ADRD showed a significant incremental rise in strength and magnitude with increasing number of comorbid NCPCs at baseline (AOR for ≥4 NCPCs vs. no NCPC = 1.65; CI = 1.34–2.03; p < .0001). Inclusion of depression and anxiety in the adjusted analyses further attenuated the association between NCPC and ADRD risk (AOR = 1.16; CI = 1.10–1.22; p < .0001) (Table 3), whereas additional inclusion of sleep disorders only modestly reduced the magnitude of this association (AOR = 1.14; CI = 1.08–1.20; p < .0001). When sleep disorders were included without mood disorders, only a mild dilution in the association was observed [AOR = 1.22; CI = 1.16–1.28; p < .0001 (not shown in Table 3)]. Similar effects of sleep and mood disorders on the association between number of chronic pain conditions at baseline and risk of ADRD were observed in the adjusted analyses (Table 3).

Table 3.

Association of Baseline Non-Cancer Chronic Pain Conditions (NCPCs) and NCPCs’ Burden to Incident Alzheimer’s Disease and Related Dementias (ADRD) in West Virginia Medicare Beneficiaries Cohort of N = 161,573 Adults*, 2013–2015 [Odds Ratios (ORs) and Adjusted Odds Ratios (AORs) With 95% Confidence Intervals (CIs)].

Any NCPC**	Unadjusted model		Adjusted for sociodemographics^¥		Also adjusted for lifestyle factors^‡, chronic conditions, analgesics^¥¥		Also adjusted for mood disorders^‡‡		Also adjusted for sleep disorders^‡‡‡
Any NCPC**	OR (95% CI)	p	AOR (95% CI)	p	AOR (95% CI)	p	AOR (95% CI)	p	AOR (95% CI)	p
No	1.00		1.00		1.00		1.00		1.00
Yes	1.66 (1.58, 1.73)	<.0001	1.48 (1.41, 1.56)	<.0001	1.26 (1.20, 1.32)	<.0001	1.16 (1.10, 1.22)	<.0001	1.14 (1.08, 1.20)	<.0001
Number of NCPC
None	1.00		1.00		1.00		1.00		1.00
1	.92 (.87, .99)	.0216	.93 (.87, 1.00)	.0418	.97 (.90, 1.04)	.3511	1.01 (.94, 1.07)	.9554	1.01 (.93, 1.08)	.8022
2	1.27 (1.18, 1.37)	<.0001	1.24 (1.15, 1.34)	<.0001	1.17 (1.08, 1.26)	<.0001	1.12 (1.04, 1.21)	.0044	1.10 (1.02, 1.19)	.0113
3	1.85 (1.68, 2.04)	<.0001	1.80 (1.63, 1.99)	<.0001	1.52 (1.37, 1.68)	<.0001	1.34 (1.21, 1.49)	<.0001	1.30 (1.17, 1.44)	<.0001
4+	2.34 (1.93, 2.84)	<.0001	2.30 (1.88, 2.82)	<.0001	1.65 (1.34, 2.03)	<.0001	1.39 (1.13, 1.71)	.002	1.30 (1.06, 1.61)	.0136
Test for linear trend		.0160		.0170		.0130		.0100		.016

ADRD, Alzheimer’s disease and related dementias.

*Age ≥ 65 years, continuously enrolled in fee-for-service West Virginia Medicare. **At least one baseline condition: osteoarthritis, headache, migraine, chronic back or neck pain; joint pain, neuropathic pain. ^¥Including sex, age, race/ethnicity, education, income, Medicaid insurance, private insurance, marital status, region. ^‡Including smoking status, obesity; ^¥¥Including chronic physical health conditions (hypertension, diabetes, heart disease, cancer, rheumatoid arthritis, lupus, history of stroke, traumatic brain injury, analgesics); ^‡‡Including depression and anxiety disorders; ^‡‡‡insomnia-related sleep disorders.

Competing Risk of Death Analysis

The results from the sensitivity analyses accounting for competing risk of death confirmed findings from our primary analyses (Table 4). Participants with any NCPC were significantly more likely to be living with incident ADRD at follow-up than living without incident ADRD after adjustment for demographics, lifestyle characteristics, chronic physical conditions, and analgesics, relative to participants with no baseline NCPC (AOR = 1.24; CI = 1.19–1.31; p < .0001). Risk of death with or without incident ADRD at follow-up was unrelated to baseline NCPC status, suggesting survival bias is an unlikely explanation for the observed associations between NCPC and incident ADRD.

Table 4.

Association of Baseline Non-Cancer Chronic Pain Conditions (NCPCs) and NCPCs’ Burden to Incident Alzheimer’s Disease and Related Dementias (ADRD) in West Virginia Medicare Beneficiaries Cohort of Older Adults*, 2013–2015 (Odds Ratios (ORs) and Adjusted Odds Ratios (AORs) With 95% Confidence Intervals (CIs): Study of Competing Risk of Death Using Multinomial Logistic Regression.

Any non-cancer chronic pain condition (NCPC)**	Alive at follow-up, no ADRD (referent)		Alive at follow-up, incident ADRD		Died before follow-up, incident ADRD		Died before follow-up, no ADRD
Any non-cancer chronic pain condition (NCPC)**	OR	p	OR (95% CI)	p	OR (95% CI)	p	OR (95% CI)	p
Model 1. Unadjusted model	1.00	—	1.66 (1.59, 1.74)	<.0001	1.64 (1.51, 1.78)	<.0001	1.44 (1.39, 1.50)	<.0001
Model 2. Adjusted for sociodemographics^¥	1.00	—	1.49 (1.42, 1.56)	<.0001	1.52 (1.40, 1.66)	<.0001	1.45 (1.39, 1.51)	<.0001
Model 3. Also adjusted for lifestyle, physical health conditions, and analgesics^¥¥	1.00	—	1.24 (1.19, 1.31)	<.0001	1.13 (1.03, 1.23)	.0072	1.04 (.99, 1.08)	.0917
Model 4. Also adjusted for mood disorders^‡‡	1.00	—	1.15 (1.09, 1.20)	<.0001	1.07 (.98,1.17)	.0144	1.01 (.96,1.05)	.7971
Model 5. Also adjusted for sleep disorders^‡‡‡	1.00	—	1.12 (1.07, 1.18)	<.0001	1.03 (.94, 1.13)	.4836	.99 (.94, 1.03)	.5765

ADRD, Alzheimer’s disease and related dementias.

*Age ≥ 65 years, continuously enrolled in fee-for-service West Virginia Medicare. **At least one baseline condition: osteoarthritis, headache, migraine, chronic back or neck pain; joint pain, neuropathic pain. ^¥Including sex, age, race/ethnicity, education, income, Medicaid insurance, private insurance, marital status, region. ^‡Including smoking status, obesity. ^¥¥Including chronic physical health conditions (hypertension, diabetes, heart disease, cancer, rheumatoid arthritis, lupus, history of stroke, traumatic brain injury, analgesics); ^‡‡Including depression and anxiety; ^‡‡‡insomnia-related sleep disorders.

Subsidiary Findings

Other than NCPCs, other variables that were found to be significantly associated with risk for incident ADRD in the study cohort included sex (female vs. male, AOR = 1.09, CI = 1.04–1.15); advanced age (80+ vs. 65–69 years, AOR 6.61, CI = 6.14–7.11); WV regions (North vs. South, AOR .86, CI .76–.97); Medicaid insurance (AOR 1.77, CI = 1.66–1.86); history of hypertension versus no hypertension, AOR = 1.24, CI = [1.15, 1.34]; diabetes versus no diabetes, AOR = 1.13, CI = [1.07, 1.19]; cardiovascular disease (CVD) versus no CVD, AOR = 1.19, CI = [1.13, 1.25]; stroke versus no stroke, AOR = 1.74, CI = [1.60, 1.88]; COPD versus no COPD, AOR = 1.19, CI = [1.12, 1.26]; TBI versus no TBI, AOR = 1.71, CI = [ 1.53, 1.92];depression versus no depression, AOR = 1.94, CI = [1.83, 2.05]; anxiety versus no anxiety, AOR = 1.1, CI = [1.03, 1.18]; and any sleep disorders versus none, AOR = 1.14, CI = [1.07, 1.21]. Moreover, in this study, high prevalence of baseline NCPC as well as high rate of incident ADRD among female versus male suggests a potential modifying role of sex in the association between NCPC and ADRD.

Discussion

Incident ADRD was significantly associated with baseline NCPC in this large retrospective cohort study of older Appalachian adults. The significance of this association remained after adjustment for sociodemographic, lifestyle factors, chronic physical health conditions, and analgesic use. We also observed an incremental rise in the odds of incident ADRD with increasing number of baseline NCPCs. Attenuation of the observed associations with inclusion of baseline depression and anxiety suggests a possible mediating role of these mood disorders on the relationship of NCPC to incident ADRD. In contrast, inclusion of sleep disorders did not appreciably affect estimates, perhaps in part due to the potential under-ascertainment of sleep disorders in medical claims data.

This study is the first rigorous population-based study to evaluate the relation of common NCPCs and NCPC burden to incident ADRD in a cohort of predominantly rural, poor, underserved elders and to explore, in this population, the potential mediating influence of mood and sleep disorders on this association. Appalachian populations not only suffer disproportionately high burden of chronic pain conditions, but also a high concomitant burden of both non-modifiable and modifiable risk factors for ADRD (Kivipelto et al., 2001; Qiu et al., 2010; Baumgart et al., 2015; Zhou et al., 2015). For example, West Virginia has almost twice the national percentage of rural-living population (Pollard & Jacobsen, 2011); the third highest proportion of adults 65+ years of age in the United States (Centers for Medicare and Medicaid Services (CMS) Medicare Reports, 2018); and the highest prevalence in the nation of most chronic physical health conditions [e.g., cardiovascular diseases, chronic kidney disease, diabetes, chronic obstructive lung disease, and multimorbidity (Schwartz et al., 2009; Holt et al., 2011; Herath & Brown, 2013; Mamudu et al., 2017; United Health Foundation, America’s Health Rankings: Annual report, West Virginia, 2020)] and adverse lifestyle factors [e.g., tobacco smoking, obesity, and physical inactivity (Behringer & Friedell, 2006; Herath & Brown, 2013; Schoenberg et al., 2015; Hege et al., 2017; United Health Foundation, America’s Health Rankings: Annual report, West Virginia, 2020)] linked to increased ADRD risk (Baumgart et al., 2015); these risk factors are also strongly associated with chronic pain (Blyth et al., 2004; Day & Thorn, 2010; Ohayon & Schatzberg, 2010; Shiri et al., 2010b; Dominick et al., 2012; Smith et al., 2014; Fayaz et al., 2016; Guglielmo et al., 2019). Thus, understanding the relation of chronic pain conditions to incident ADRD in WV and other high-risk rural populations is of particular importance.

To our knowledge, published studies regarding the association of pain to adverse cognitive outcomes in rural, underserved populations are limited to a cross-sectional investigation of 21,982 older Appalachian adults residing in the mid-Ohio Valley near Parkersburg, WV (Innes & Sambamoorthi, 2017). Consistent with our findings, this study documented a strong, positive, dose–response association of perceived memory loss to osteoarthritis and associated joint pain, an association that appeared mediated in part by sleep and mood impairment. Likewise, published research regarding the collective and incremental relation of common NCPCs to incident ADRD remains sparse, to date limited to our recent study in a large sample of US elders enrolled in fee-for-service Medicare (Khalid et al., 2020). In this prior investigation, we found risk of ADRD to rise significantly with increasing number of pain conditions at baseline after adjusting for sociodemographic, lifestyle and comorbid conditions, and medication use (Khalid et al., 2020), with risk estimates similar overall to those observed in the current study of WV elders.

Although longitudinal investigations of rural, disadvantaged populations are lacking, our findings from this study of Appalachian adults are also in broad agreement with those from other previously published observational studies examining the association of chronic pain and/or chronic pain conditions to subsequent adverse cognitive outcomes. For example, findings from recent cross-sectional (Van Der Leeuw et al., 2015; Innes & Sambamoorthi, 2017; Ikram et al., 2019) and cohort studies in East Asia (Huang et al., 2015; Yang et al., 2016; Tzeng et al., 2018; Yamada et al., 2019), Europe (Hagen et al., 2014; Røttereng et al., 2015), and North America (Morton et al., 2019; Innes & Sambamoorthi, 2020; Khalid et al., 2020) suggest that those with specific chronic pain conditions including fibromyalgia (Tzeng et al., 2018), headache and migraine (Hagen et al., 2014; Yang et al., 2016; Morton et al., 2019), and osteoarthritis/knee (Yamada et al., 2019; Innes & Sambamoorthi, 2020) may be at increased risk for incident ADRD. Likewise, studies of chronic pain symptoms in US (Whitlock et al., 2017; van der Leeuw et al., 2018; Ezzati et al., 2019) and British (Veronese et al., 2018) adults have documented associations of persistent pain (Whitlock et al., 2017), severe chronic pain (van der Leeuw et al., 2018; Veronese et al., 2018), and/or reported pain interference (Ezzati et al., 2019) to subsequent deterioration in memory (Whitlock et al., 2017; van der Leeuw et al., 2018), accelerated cognitive decline (Whitlock et al., 2017), new onset cognitive impairment (Whitlock et al., 2017; van der Leeuw et al., 2018), and incident dementia (Whitlock et al., 2017; Ezzati et al., 2019).

While the factors underlying the observed link between NCPCs and ADRD risk remain little understood, several biological, behavioral, health, and economic factors may play a role. For example, posited biological mechanisms have included the exhaustion of neural reserves for attention and memory due to frequent processing of pain signals, chronic pain-mediated inflammation and eventual neuronal dysplasia, and altered neurochemistry (Moriarty et al., 2011; Mutso et al., 2012; Malfliet et al., 2017; Cao et al., 2019). Contributing behavioral factors may include the use of maladaptive pain coping strategies, negative pain perception, and lower health-seeking behavior (Hoffman et al., 2002; Seufert & Carrozza, 2004; Behringer & Friedell, 2006; Day & Thorn, 2010). Health and economic factors associated with both chronic pain and ADRD risk, such as lower access to health care, poor education, and low household income (Seufert & Carrozza, 2004; Shaw et al., 2004; Behringer & Friedell, 2006), may also help explain observed associations. Finally, certain lifestyle factors linked to both chronic pain and ADRD (Shiri et al., 2010b; Baumgart et al., 2015; Cooper et al., 2015; Okifuji & Hare, 2015; Fayaz et al., 2016), including tobacco smoking (Shiri et al., 2010b), physical inactivity (West Virginia Behavioral Risk Factor Surveillance Survey Report 2017, 2017), and mid-life obesity (Shiri et al., 2010a), may also help explain the association between NCPCs and ADRD risk. Consistent with prior studies (Shiri et al., 2010b; Okifuji & Hare, 2015), we found an increased prevalence of any chronic pain condition at baseline in individuals with obesity and smoking. As has been documented previously (Kivipelto et al., 2001; Musicco et al., 2009; Qiu et al., 2010; Baumgart et al., 2015; Cooper et al., 2015; Zhou et al., 2015), the prevalence of other ADRD-related risk factors including diabetes, stroke, hypertension, and history of heart disease was also significantly higher in study participants with baseline NCPC. Notably, adjustment for socioeconomic factors, lifestyle characteristics, and chronic physical health conditions attenuated, although did not eliminate the positive association between NCPCs and incident ADRD risk, suggesting these factors may in part underlie this relationship.

Our study specifically investigated the potential mediating role of sleep and mood disorders on the association between NCPCs and ADRD risk. Depression, anxiety, and sleep disorders are significant independent predictors of ADRD (Ownby et al., 2006; Bubu et al., 2016; Gulpers et al., 2016) and have been shown to have strong, reciprocal relationships with chronic pain (Hooten, 2016; Kroenke et al., 2011; Menefee et al., 2000). Chronic pain can lead to poor sleep quality (Sayar et al., 2002; Finan et al., 2013); conversely, sleep deprivation is also documented to affect pain perception and trigger hyperalgesia (Gopalakrishnan et al., 2004; Lautenbacher et al., 2006; Monti & Monti, 2007; Berridge et al., 2012). Similarly, evidence exists in support of a bilateral relationship between mood disorders and chronic pain (Wilson et al., 2002; Ohayon & Schatzberg, 2010; Hooten, 2016). For example, a cohort study of 500 primary-care patients showed a reciprocal relationship between depression and pain severity, with equal and strong effects on each other when assessed longitudinally (Kroenke et al., 2011). Evidence suggests that the shared neurobiology underlying chronic pain and mood disorders such as depression and anxiety may contribute to the observed bidirectionality (Finan & Smith, 2013; Hooten, 2016; Malfliet et al., 2017). In our study, inclusion of depression and anxiety disorders in the adjusted models weakened the association between NCPCs and incident ADRD, suggesting that these mood disorders may in part mediate this relationship. In contrast to findings of recent studies in nation-wide (Khalid et al., 2020) and Appalachian (Innes & Sambamoorthi, 2017) Medicare beneficiaries, inclusion of sleep disorders in the adjusted model only modestly attenuated the relation of NCPCs to incident ADRD in this study of WV elders. While reasons for this discrepancy are unclear, our results may in part reflect under-ascertainment of insomnia and related sleep disorders in WV medical claims data.

Strengths and Limitations

Our study has several strengths, including the population-based design and use of longitudinal data from a large retrospective cohort of Appalachian elders that included all noninstitutionalized WV elders enrolled in FFS Medicare insurance program during the study period. Information was available on a range of potential confounders, including sociodemographic and lifestyle characteristics; history of chronic health conditions; and use of analgesic medications, allowing us to assess the potential influence of these factors on the relation of NCPC’s to incident ADRD. Likewise, information on diagnosed depression, anxiety, and sleep disorders permitted us to evaluate the possible mediating effects of these risk factors. ADRD and NCPC were ascertained using claims data and recommended algorithms, as were other physical health conditions, history of stroke and TBI, and mood and sleep disorders.

However, the findings of our study should be interpreted in the light of several limitations, including relatively short follow-up period. ADRD, a condition that is often underdiagnosed in part due to its typically slow and insidious progression, is likely to be under-ascertained more so in our study’s Appalachian population where underdiagnosis of ADRD has been documented (Abner et al., 2016), therefore, potentially biasing our risk estimates toward the null. Identification of NCPCs in this study was dependent on healthcare claims data, and hence, some degree of underreporting is expected (Beaudet et al., 2013; Tian et al., 2013; Ward, 2013; Treede et al., 2015). Additionally, use of a conservative case definition for chronic pain ascertainment, that is, any inpatient claim or two outpatient claims for any chronic pain conditions 90 days apart (Tonelli et al., 2015), may have further contributed to under-ascertainment of NCPCs in this study population, potentially attenuating risk estimates. We lacked information on certain factors associated with NCPC and/or ADRD, including occupational history, physical activity, alcohol consumption, dietary habits, genetic predisposition, early childhood experiences, and social isolation (Kivipelto et al., 2001; Baumgart et al., 2015; Cooper et al., 2015). In addition, our assessment of income and education relied on proxy measures, possibly resulting in under-adjustment for these factors. While both ADRD and chronic pain have been associated with increased risk for mortality (Hui et al., 2003; Smith et al., 2014; Taylor et al., 2017; Zissimopoulos et al., 2018), our sensitivity analyses for competing risk of death yielded findings similar to those of our primary analyses, suggesting that survival bias is unlikely to explain the observed association of NCPC to ADRD in this study.

While Medicare coverage for West Virginia is more than 24%, which is higher than the 19% national coverage rates, restriction of our study to continuously enrolled, noninstitutionalized FFS Medicare beneficiaries may limit generalizability of our findings to population groups who are not enrolled continuously in FFS Medicare, are institutionalized, or do not have health insurance. Likewise, our study population was limited to residents of WV as well as a predominantly white population in our study restricts the generalizability to minority races and populations from other Appalachian regions. Last, any causal inference of our study findings is limited by the short follow-up period in this study and the insidious nature of ADRD development and progression. However, a growing body of research regarding chronic pain’s involvement in neurocognitive impairments and heightened risk of cognitive decline and incident dementia (Moriarty et al., 2011; Malfliet et al., 2017; Cao et al., 2019) makes a sound case against reverse causality.

Conclusion

In this large retrospective cohort of older Appalachian Medicare beneficiaries, risk for incident ADRD was significantly associated with baseline NCPCs and increased as the number of NCPCs accrued in a dose-dependent fashion. Depression and anxiety but not sleep disorders potentially partially mediate this relationship. In the backdrop of limited healthcare resources and the absence of either effective disease-modifying or curative therapies for ADRD, additional studies are needed to elucidate the contribution of chronic pain as an emerging ADRD risk factor. Further large, prospective studies in this high-risk population with longer follow-up duration and additional information on potential confounding factors are needed to improve understanding of the role of chronic pain in ADRD pathogenesis, confirm, and extend our findings.

Footnotes

Declaration of Conflicting Interests

The author(s) declared no potential conflicts of interest with respect to the research, authorship, and/or publication of this article.

Funding

The author(s) disclosed receipt of the following financial support for the research, authorship, and/or publication of this article: This work was supported by Foundation for the National Institutes of Health (Grant No. 2U54GM104942-02).

ORCID iD

Sumaira Khalid

References

Abner

E. L.

Jicha

G. A.

Christian

W. J.

Schreurs

B. G.

(2016). Rural‐urban differences in Alzheimer’s disease and related disorders diagnostic prevalence in Kentucky and West Virginia. The Journal of Rural Health, 32(3), 314-320.

Akushevich

Yashkin

Kravchenko

Yashin

A. I.

(2018, April). Racial and geographic disparities in risk and survival of Alzheimer’s disease and related dementias. In PAA 2018 Annual Meeting. Denver, Colorado, April, pp. 26-28.

Alzheimer’s Association Facts and Figures (2021). Alzheimer's & dementia. Retrieved from https://www.alz.org/media/Documents/alzheimers-facts-and-figures.pdf.

Baumgart

Snyder

H. M.

Carrillo

M. C.

Fazio

Kim

Johns

(2015). Summary of the evidence on modifiable risk factors for cognitive decline and dementia: a population-based perspective. Alzheimer’s & Dementia, 11(6), 718-726.

Beaudet

Courteau

Sarret

Vanasse

(2013). Prevalence of claims-based recurrent low back pain in a Canadian population: A secondary analysis of an administrative database. BMC Musculoskeletal Disorders, 14(1), 151.

Behringer

Friedell

G. H.

(2006). Appalachia: where place matters in health. Preventing Chronic Disease, 3(4), A113.

Berridge

C. W.

Schmeichel

B. E.

España

R. A.

(2012). Noradrenergic modulation of wakefulness/arousal. Sleep Medicine Reviews, 16(2), 187-197.

Blyth

F. M.

March

L. M.

Brnabic

A. J.

Cousins

M. J.

(2004). Chronic pain and frequent use of health care. Pain, 111(1-2), 51-58.

BRFSS (2018). Behavioral risk factor surveillance system, prevalence and trends data. Retrieved from https://www.cdc.gov/brfss/brfssprevalence/index.html

10.

Bubu

O. M.

Brannick

Mortimer

Umasabor-Bubu

Sebastiao

Y. V.

Wen

Schwartz

Borenstein

A. R.

Morgan

(2016). Sleep, cognitive impairment, and Alzheimer’s disease: a systematic review and meta-analysis. Sleep, 40(1), zsw032.

11.

Cao

Fisher

D. W.

Dong

(2019). The link between chronic pain and Alzheimer’s disease. Journal of Neuroinflammation, 16(1), 1-11.

12.

Centers for Medicare and Medicaid Services (CMS) (2018). Medicare reports. March 03, 2020.

13.

Centers of Disease Control (CDC) West Virginia - State Nutrition, Physical Activity, and Obesity Profile (2012). Retrieved from https://www.cdc.gov/obesity/stateprograms/fundedstates/pdf/west-virginia-state-profile.pdf

14.

Chronic Conditions Data Warehouse . Your source for national CMS Medicare and Medicaid research data. Retrieved from https://www2.ccwdata.org/web/guest/condition-categories. (Access Year 2020).

15.

Cooper

Sommerlad

Lyketsos

C. G.

Livingston

(2015). Modifiable predictors of dementia in mild cognitive impairment: a systematic review and meta-analysis. American Journal of Psychiatry, 172(4), 323-334.

16.

Dahlhamer

Lucas

Zelaya

Nahin

Mackey

DeBar

Kerns

Von Korff

Porter

Helmick

(2018). Prevalence of chronic pain and high-impact chronic pain among adults—United States, 2016. MMWR Morbidity and Mortality Weekly Report, 67(36), 1001.

17.

Day

M. A.

Thorn

B. E.

(2010). The relationship of demographic and psychosocial variables to pain-related outcomes in a rural chronic pain population. PAIN®, 151(2), 467-474.

18.

Dominick

C. H.

Blyth

F. M.

Nicholas

M. K.

(2012). Unpacking the burden: Understanding the relationships between chronic pain and comorbidity in the general population. Pain, 153(2), 293-304.

19.

Edlund

M. J.

Martin

B. C.

Devries

Fan

M.-Y.

Braden

J. B.

Sullivan

M. D.

(2010). Trends in use of opioids for chronic non-cancer pain among individuals with mental health and substance use disorders: the TROUP study. The Clinical Journal of Pain, 26(1), 1.

20.

Ezzati

Wang

Katz

M. J.

Derby

C. A.

Zammit

A. R.

Zimmerman

M. E.

Pavlovic

J. M.

, Sliwinski

M. J.

Lipton

R. B.

(2019). The temporal relationship between pain intensity and pain interference and incident dementia. Current Alzheimer Research, 16(2), 109-115.

21.

Fayaz

Ayis

Panesar

S. S.

Langford

R. M.

Donaldson

L. J.

(2016). Assessing the relationship between chronic pain and cardiovascular disease: A systematic review and meta-analysis. Scandinavian Journal of Pain, 13, 76-90.

22.

Finan

P. H.

Goodin

B. R.

Smith

M. T.

(2013). The association of sleep and pain: An update and a path forward. The Journal of Pain, 14(12), 1539-1552.

23.

Finan

P. H.

Smith

M. T.

(2013). The comorbidity of insomnia, chronic pain, and depression: dopamine as a putative mechanism. Sleep Medicine Reviews, 17(3), 173-183.

24.

Forbes

Langan

Smeeth

(2018). ICD-10 codes-Dementia. Available at https://datacompass.lshtm.ac.uk/id/eprint/999/1/hes_dementia_codes_ind_prev.txt.

25.

Gibson

S. J.

Lussier

(2012). Prevalence and relevance of pain in older persons. Pain Medicine, 13(suppl_2), S23-S26.

26.

Goldberg

D. S.

McGee

S. J.

(2011). Pain as a global public health priority. BMC Public Health, 11(1), 1-5.

27.

Gopalakrishnan

L. L.

Cirelli

(2004). Sleep deprivation and cellular responses to oxidative stress. Sleep, 27(1), 27-35.

28.

Guglielmo

Murphy

L. B.

Boring

M. A.

Theis

K. A.

Helmick

C. G.

Hootman

J. M.

Odom

E. L.

Carlson

S. A.

Liu

(2019). State-specific severe joint pain and physical inactivity among adults with arthritis—United States, 2017. MMWR Morbidity and Mortality Weekly Report, 68(17), 381.

29.

Gulpers

Ramakers

Hamel

Köhler

Voshaar

R. O.

Verhey

(2016). Anxiety as a predictor for cognitive decline and dementia: a systematic review and meta-analysis. The American Journal of Geriatric Psychiatry, 24(10), 823-842.

30.

Haaksma

M. L.

Vilela

L. R.

Marengoni

Calderón-Larrañaga

Leoutsakos

J.-M. S.

Rikkert

M. G. O.

Melis

R. J.

(2017). Comorbidity and progression of late onset Alzheimer’s disease: a systematic review. Plos One, 12(5), e0177044.

31.

Hagen

Stordal

Linde

Steiner

T. J.

Zwart

J.-A.

Stovner

L. J.

(2014). Headache as a risk factor for dementia: a prospective population-based study. Cephalalgia, 34(5), 327-335.

32.

Healthcare Cost and Utilization Project (HCUP) (2019). Quartile classification of patient’s zip code. Retrieved from https://www.hcup-us.ahrq.gov/db/vars/zipinc_qrtl/nisnote.jsp.

33.

Hege

Christiana

R. W.

Battista

Parkhurst

(2017). Active living in rural Appalachia: Using the rural active living assessment (RALA) tools to explore environmental barriers. Preventive Medicine Reports, 8, 261-266.

34.

Hendryx

(2010). Poverty and mortality disparities in central Appalachia: Mountaintop mining and environmental justice. Journal of Health Disparities Research and Practice, 4(3), 6.

35.

Herath

Brown

(2013). An analysis of adult obesity and hypertension in Appalachia. Global Journal of Health Science, 5(3), 127.

36.

Hoffman

P. K.

Meier

B. P.

Council

J. R.

(2002). A comparison of chronic pain between an urban and rural population. Journal of Community Health Nursing, 19(4), 213-224.

37.

Holt

J. B.

Zhang

Presley-Cantrell

Croft

J. B.

(2011). Geographic disparities in chronic obstructive pulmonary disease (COPD) hospitalization among Medicare beneficiaries in the United States. International Journal of Chronic Obstructive Pulmonary Disease, 6, 321.

38.

Hooten

W. M.

(2016). Chronic pain and mental health disorders: shared neural mechanisms, epidemiology, and treatment. Clin Proc, 91, 955.

39.

Huang

S.-W.

Wang

W.-T.

Chou

L.-C.

Liao

C.-D.

Liou

T.-H.

Lin

H.-W.

(2015). Osteoarthritis increases the risk of dementia: a nationwide cohort study in Taiwan. Scientific Reports, 5, 10145.

40.

Hui

Wilson

Bennett

Bienias

Gilley

Evans

(2003). Rate of cognitive decline and mortality in Alzheimer’s disease. Neurology, 61(10), 1356-1361.

41.

Ikram

Innes

Sambamoorthi

(2019). Association of osteoarthritis and pain with Alzheimer’s diseases and related dementias among older adults in the United States. Osteoarthritis and Cartilage, 27, 1470.

42.

Innes

K. E.

Sambamoorthi

(2017). The association of perceived memory loss with osteoarthritis and related joint pain in a large Appalachian population. Pain Medicine, 19(7), 1340-1356.

43.

Innes

K. E.

Sambamoorthi

(2020). The association of osteoarthritis and related pain burden to incident Alzheimer’s disease and related dementias: A retrospective cohort study of US Medicare beneficiaries. Journal of Alzheimer's disease (Preprint), 1-17.

44.

Jackson

T. P.

Stabile

V. S.

McQueen

K. K.

(2014). The global burden of chronic pain. ASA Newsletter, 78(6), 24-27.

45.

Johannes

C. B.

T. K.

Zhou

Johnston

J. A.

Dworkin

R. H.

(2010). The prevalence of chronic pain in United States adults: results of an Internet-based survey. The Journal of Pain, 11(11), 1230-1239.

46.

Julien

Joubert

Ferland

M.-C.

Frenette

Boudreau-Duhaime

Malo-Véronneau

De Guise

(2017). Association of traumatic brain injury and Alzheimer disease onset: A systematic review. Annals of Physical and Rehabilitation Medicine, 60(5), 347-356.

47.

Khalid

Sambamoorthi

Innes

K. E.

(2020). Non-cancer chronic pain conditions and risk for incident Alzheimer’s disease and related dementias in community-dwelling older adults: A population-based retrospective cohort study of United States Medicare beneficiaries, 2001–2013. International Journal of Environmental Research and Public Health, 17(15), 5454.

48.

Kivipelto

Helkala

E.-L.

Laakso

M. P.

Hänninen

Hallikainen

Alhainen

Soininen

Tuomilehto

Nissinen

(2001). Midlife vascular risk factors and Alzheimer’s disease in later life: Longitudinal, population based study. BMJ: British Medical Journal, 322(7300), 1447-1451.

49.

Kroenke

Bair

M. J.

Krebs

E. E.

Damush

T. M.

(2011). Reciprocal relationship between pain and depression: a 12-month longitudinal analysis in primary care. The Journal of Pain, 12(9), 964-973.

50.

Lautenbacher

Kundermann

Krieg

J.-C.

(2006). Sleep deprivation and pain perception. Sleep Medicine Reviews, 10(5), 357-369.

51.

Lin

P.-J.

Kaufer

D. I.

Maciejewski

M. L.

Ganguly

Paul

J. E.

Biddle

A. K.

(2010). An examination of Alzheimer’s disease case definitions using Medicare claims and survey data. Alzheimer’s & Dementia, 6(4), 334-341.

52.

Liu

Wheaton

A. G.

Chapman

D. P.

Cunningham

T. J.

Croft

J. B.

(2016). Prevalence of healthy sleep duration among adults—United States, 2014. MMWR. Morbidity and Mortality Weekly Report, 65(6), 137-141.

53.

Malfliet

Coppieters

Van Wilgen

Kregel

De Pauw

Dolphens

Ickmans

(2017). Brain changes associated with cognitive and emotional factors in chronic pain: A systematic review. European Journal of Pain, 21(5), 769-786.

54.

Mamudu

H. M.

Paul

T. K.

Wang

Veeranki

S. P.

Panchal

H. B.

Alamian

Budoff

(2017). Association between multiple modifiable risk factors of cardiovascular disease and hypertension among asymptomatic patients in central Appalachia. Southern Medical Journal, 110(2), 90-96.

55.

Mattos

M. K.

Snitz

B. E.

Lingler

J. H.

Burke

L. E.

Novosel

L. M.

Sereika

S. M.

(2017). Older rural‐and urban‐dwelling Appalachian adults with mild cognitive impairment. The Journal of Rural Health, 33(2), 208-216.

56.

McNeill

A. S.

(2011). Neurodegeneration: Theory, disorders and treatments. Nova Science Publishers.

57.

McWilliams

L. A.

Goodwin

R. D.

Cox

B. J.

(2004). Depression and anxiety associated with three pain conditions: results from a nationally representative sample. Pain, 111(1-2), 77-83.

58.

Medicare Current Beneficiary Survey (MCBS), CMS Medicare and Medicaid Research Data (2020).Retrieved from https://www.cms.gov/Research-Statistics-Data-and-Systems/Research/MCBS.

59.

Menefee

L. A.

Cohen

M. J.

Anderson

W. R.

Doghramji

Frank

E. D.

Lee

(2000). Sleep disturbance and nonmalignant chronic pain: a comprehensive review of the literature. Pain Medicine, 1(2), 156-172.

60.

Monti

J. M.

Monti

(2007). The involvement of dopamine in the modulation of sleep and waking. Sleep Medicine Reviews, 11(2), 113-133.

61.

Moon

Badana

A. N.

Hwang

S.-Y.

Sears

J. S.

Haley

W. E.

(2019). Dementia prevalence in older adults: Variation by race/ethnicity and immigrant status. The American Journal of Geriatric Psychiatry, 27(3), 241-250.

62.

Moriarty

McGuire

B. E.

Finn

D. P.

(2011). The effect of pain on cognitive function: A review of clinical and preclinical research. Progress in Neurobiology, 93(3), 385-404. doi:10.1016/j.pneurobio.2011.01.002

63.

Morton

R. E.

St John

P. D.

Tyas

S. L.

(2019). Migraine and the risk of all‐cause dementia, Alzheimer's disease, and vascular dementia: A prospective cohort study in community‐dwelling older adults. International Journal of Geriatric Psychiatry, 34(11), 1667-1676.

64.

Musicco

Palmer

Salamone

Lupo

Perri

Mosti

Cravello

(2009). Predictors of progression of cognitive decline in Alzheimer’s disease: the role of vascular and sociodemographic factors. Journal of Neurology, 256(8), 1288.

65.

Mutso

A. A.

Radzicki

Baliki

M. N.

Huang

Banisadr

Centeno

M. V.

Apkarian

A. V.

(2012). Abnormalities in hippocampal functioning with persistent pain. Journal of Neuroscience, 32(17), 5747-5756.

66.

S. K.

Urquhart

D. M.

Fitzgerald

P. B.

Cicuttini

F. M.

Hussain

S. M.

Fitzgibbon

B. M.

(2018). The relationship between structural and functional brain changes and altered emotion and cognition in chronic low back pain brain changes. The Clinical Journal of Pain, 34(3), 237-261.

67.

Nicholson

Verma

(2004). Comorbidities in chronic neuropathic pain. Pain Medicine, 5(suppl_1), S9-S27.

68.

Nichols

Szoeke

C. E.

Vollset

S. E.

Abbasi

Abd-Allah

Abdela

Asgedom

S. W.

(2019). Global, regional, and national burden of Alzheimer’s disease and other dementias, 1990–2016: A systematic analysis for the Global Burden of Disease Study 2016. The Lancet Neurology, 18(1), 88-106.

69.

Ohayon

M. M.

Schatzberg

A. F.

(2010). Chronic pain and major depressive disorder in the general population. Journal of Psychiatric Research, 44(7), 454-461.

70.

Okifuji

Hare

B. D.

(2015). The association between chronic pain and obesity. Journal of Pain Research, 8, 399.

71.

Ownby

R. L.

Crocco

Acevedo

John

Loewenstein

(2006). Depression and risk for Alzheimer disease: systematic review, meta-analysis, and metaregression analysis. Archives of General Psychiatry, 63(5), 530-538.

72.

Pollard

Jacobsen

L. A.

(2011). The Appalachian region in 2010: A census data overview chartbook. Retrieved from http://www.arc.gov/research/researchreportdetails.asp.

73.

Qiu

Fratiglioni

(2010). Vascular and psychosocial factors in Alzheimer’s disease: Epidemiological evidence toward intervention. Journal of Alzheimer's disease, 20(3), 689-697.

74.

Rice

A. S.

Smith

B. H.

Blyth

F. M.

(2016). Pain and the global burden of disease. Pain, 157(4), 791-796.

75.

Røttereng

A. K. S.

Bosnes

Stordal

Zwart

J.-A.

Linde

Stovner

L. J.

Hagen

(2015). Headache as a predictor for dementia: The HUNT Study. The Journal of Headache and Pain, 16(1), 89.

76.

Sayar

Arikan

Yontem

(2002). Sleep quality in chronic pain patients. The Canadian Journal of Psychiatry, 47(9), 844-848.

77.

Schoenberg

N. E.

Huang

Seshadri

Tucker

T. C.

(2015). Trends in cigarette smoking and obesity in Appalachian Kentucky. Southern Medical Journal, 108(3), 170-177.

78.

Schwartz

Ruhil

Denham

Shubrook

Simpson

Boyd

S. L.

(2009). High self‐reported prevalence of diabetes mellitus, heart disease, and stroke in 11 counties of rural Appalachian Ohio. The Journal of Rural Health, 25(2), 226-230.

79.

Serrano

Leiferman

Dauber

(2007). Self-efficacy and health behaviors toward the prevention of diabetes among high risk individuals living in Appalachia. Journal of Community Health, 32(2), 121-133.

80.

Seufert

R. L.

Carrozza

M. A.

(2004). Economic advances and disadvantages in Appalachia: Occupation, labor force participation, and unemployment. Journal of Appalachian Studies, 10(3), 331-339.

81.

Shaw

T. C.

DeYoung

A. J.

Rademacher

E. W.

(2004). Educational attainment in Appalachia: Growing with the nation, but challenges remain. Journal of Appalachian Studies, 10(3), 307-329.

82.

Shiri

Karppinen

Leino-Arjas

Solovieva

Viikari-Juntura

(2010a). The association between obesity and low back pain: A meta-analysis. American Journal of Epidemiology, 171(2), 135-154.

83.

Shiri

Karppinen

Leino-Arjas

Solovieva

Viikari-Juntura

(2010b). The association between smoking and low back pain: A meta-analysis. The American Journal of Medicine, 123(1), 87.e7-87.e35.

84.

Smith

Wilkie

Uthman

Jordan

J. L.

McBeth

(2014). Chronic pain and mortality: A systematic review. Plos One, 9(6), e99048.

85.

Sullivan

M. D.

Edlund

M. J.

Fan

M.-Y.

DeVries

Braden

J. B.

Martin

B. C.

(2008). Trends in use of opioids for non-cancer pain conditions 2000–2005 in commercial and Medicaid insurance plans: the TROUP study. Pain, 138(2), 440-449.

86.

Takizawa

Thompson

P. L.

van Walsem

Faure

Maier

W. C.

(2015). Epidemiological and economic burden of Alzheimer's disease: a systematic literature review of data across Europe and the United States of America. Journal of Alzheimer's disease, 43(4), 1271-1284.

87.

Taylor

C. A.

Greenlund

S. F.

McGuire

L. C.

Croft

J. B.

(2017). Deaths from Alzheimer’s disease—United States, 1999–2014. MMWR. Morbidity and Mortality Weekly Report, 66(20), 521.

88.

Tian

T. Y.

Zlateva

Anderson

D. R.

(2013). Using electronic health records data to identify patients with chronic pain in a primary care setting. Journal of the American Medical Informatics Association, 20(e2), e275-e280.

89.

Tonelli

Wiebe

Fortin

Guthrie

Hemmelgarn

B. R.

James

M. T.

Ronksley

(2015). Methods for identifying 30 chronic conditions: application to administrative data. BMC Medical Informatics and Decision Making, 15(1), 31.

90.

Treede

R.-D.

Rief

Barke

Aziz

Bennett

M. I.

Benoliel

Wang

S.-J.

(2015). A classification of chronic pain for ICD-11. Pain, 156(6), 1003-1007. doi:10.1097/j.pain.0000000000000160

91.

Tzeng

N. S.

Chung

C. H.

Liu

F. C.

Chiu

Y. H.

Chang

H. A.

Yeh

C. B.

Chien

W. C.

(2018). Fibromyalgia and risk of dementia-A nationwide, population-based, cohort study. The American Journal of the Medical Sciences, 355(2), 153-161. doi:10.1016/j.amjms.2017.09.002

92.

United Health Foundation, America's Health Rankings (2020). Annual report. West Virginia. Retrieved from https://www.americashealthrankings.org/explore/annual/measure/Overall/state/WV.

93.

US Census Bureau QuickFacts (2019). Retrieved from https://www.census.gov/quickfacts/fact/table/WV/PST045219.

94.

van der Leeuw

Ayers

Leveille

S. G.

Blankenstein

A. H.

van der Horst

H. E.

Verghese

(2018). The effect of pain on major cognitive impairment in older adults. The Journal of Pain, 19(12), 1435-1444.

95.

Van Der Leeuw

Eggermont

L. H.

Shi

Milberg

W. P.

Gross

A. L.

Hausdorff

J. M.

Leveille

S. G.

(2015). Pain and cognitive function among older adults living in the community. Journals of Gerontology Series A: Biomedical Sciences and Medical Sciences, 71(3), 398-405.

96.

Verkaik

Nuyen

Schellevis

Francke

(2007). The relationship between severity of Alzheimer’s disease and prevalence of comorbid depressive symptoms and depression: A systematic review. International Journal of Geriatric Psychiatry: A journal of the psychiatry of late life and allied sciences, 22(11), 1063-1086.

97.

Veronese

Koyanagi

Solmi

Thompson

Maggi

Schofield

Stubbs

(2018). Pain is not associated with cognitive decline in older adults: A four-year longitudinal study. Maturitas, 115, 92-96.

98.

Ward

M. M.

(2013). Estimating disease prevalence and incidence using administrative data: Some assembly required. The Journal of Rheumatology, 40, 1241.

99.

West Virginia Behavioral Risk Factor Surveillance Survey Report 2017 (2017). Retrieved from https://dhhr.wv.gov/hpcd/data_reports/Documents/BRFSS2017.pdf.

100.

Whitlock

E. L.

Diaz-Ramirez

L. G.

Glymour

M. M.

Boscardin

W. J.

Covinsky

K. E.

Smith

A. K.

(2017). Association between persistent pain and memory decline and dementia in a longitudinal cohort of elders. JAMA Intern Med, 177(8), 1146-1153. doi:10.1001/jamainternmed.2017.1622

101.

Wilson

K. G.

Eriksson

M. Y.

Joyce

Mikail

S. F.

Emery

P. C.

(2002). Major depression and insomnia in chronic pain. The Clinical Journal of Pain, 18(2), 77-83.

102.

WVDHHR Regions Description . Retrieved from https://dhhr.wv.gov/bms/Pages/default.aspx. (Access Yr 2020)

103.

Yamada

Kubota

Tabuchi

Shirai

Iso

Kondo

(2019). A prospective study of knee pain, low back pain, and risk of dementia: the JAGES project. Scientific Reports, 9(1), 1-7.

104.

Yang

F.-C.

Lin

T.-Y.

Chen

H.-J.

Lee

J.-T.

Lin

C.-C.

Kao

C.-H.

(2016). Increased risk of dementia in patients with tension-type headache: A nationwide retrospective population-based cohort study. Plos One, 11(6), e0156097.

105.

Zhou

J.-T.

Wang

H.-F.

Meng

X.-F.

Tan

C.-C.

Wang

Tan

(2015). Association between stroke and Alzheimer’s disease: Systematic review and meta-analysis. Journal of Alzheimer’s disease, 43(2), 479-489.

106.

Zissimopoulos

J. M.

Tysinger

B. C.

Clair

St. P. A.

Crimmins

E. M.

(2018). The impact of changes in population health and mortality on future prevalence of Alzheimer’s disease and other dementias in the United States. The Journals of Gerontology: Series B, 73(suppl_1), S38-S47.