Subjective cognitive decline among older adults and its link to diabetes,cardiovascular diseases,and other comorbidities: A US population-based study

Abstract

Background

Subjective cognitive decline (SCD) is the self-reported occurrence of increasing confusion or memory loss. Accumulating evidence suggests that SCD may be a prodromal phase of progressed cognitive decline stages, particularly Alzheimer's disease. The risk factors, mainly comorbidities, associated with SCD are not well known.

Objective

This study aims to examine whether diabetes, cardiovascular disease (CVD), with a focus on stroke and coronary heart disease (CHD), and other comorbidities are associated with SCD.

Methods

We conducted a quantitative analysis employing weighted analyses on cross-sectional data adopted from the 2022 Behavioral Risk Factor Surveillance System.

Results

Our results showed that out of 60,492 adults aged 45 years and older, 6423 individuals (10.45%) reported experiencing SCD, 19.49% diabetes, 11.78% CHD, and 5.67% stroke. Among diabetics, 33.41% reported insulin use. Comorbidities that were significantly associated with SCD included stroke (OR: 1.61; 95%CI: 1.24–2.08), CHD (OR:1.56; 95% CI: 1.27,1.92), and diabetes (OR: 1.25; 95%CI: 1.02–1.55). Other significant comorbidities included depressive disorder (OR: 2.56; 95% CI: 2.18–3.02), kidney disease (OR: 1.89; 95% CI: 1.24–2.87), arthritis (OR: 1.31; 95%CI: 1.07–1.61), and asthma (OR: 1.27; 95%CI: 1.06–1.52). Age of diagnosis with diabetes was younger among participants with SCD. Insulin use was significantly associated with SCD, particularly among type 2 diabetes, accounting for duration of diabetes. (OR: 1.39; 95% CI:1.02- 1.92).

Conclusions

Given the importance of SCD as a precursor to progressed cognitive impairment, these findings highlight the need for awareness and proactive screening for cognitive dysfunction among high-risk individuals with CVDs, diabetes, depression, and other comorbidities.

Keywords

Alzheimer's disease cognitive decline comorbidities coronary heart disease depressive disorder insulin kidney disease stroke

Introduction

Subjective cognitive decline (SCD) is the self-reporting of an increase in confusion, and reduction in cognitive capabilities, or memory loss, occurring more frequently or worsening over time,¹ without the presence of objective neuropsychological deficits.² SCD is increasingly recognized as a risk factor for incident dementia, particularly Alzheimer's disease (AD), manifesting before the onset of clinical impairment.³ According to SCD initiatives, SCD can be regarded as the third stage of preclinical AD, characterized by the onset of initial cognitive alterations.⁴ Evidence shows that older adults with SCD have an increased likelihood of developing abnormalities in biomarkers related to AD in cerebrospinal fluid.⁵ Drawing from the framework of a widely recognized model outlining biomarker dynamics in AD development, there is a suggestion that SCD symptoms might stem from adaptive adjustments in reaction to the buildup of amyloid-β and neurodegeneration, indicating that SCD could represent the initial indication of AD.⁶

SCD has been a subject of global attention. A study estimating the prevalence of SCD across international cohorts concluded that about 25% of cognitively unimpaired old adults have SCD.⁷ In the United States (US), around one out of nine adults aged 65 years or older and one out of ten middle-aged adults indicated experiencing frequent memory loss or confusion.¹ Moreover, Centers for Disease Control and Prevention (CDC) estimated SCD prevalence to be 11.3% among men and 10.6% among women.¹ In addition, SCD varies across racial and ethnic groups, with 10.9% of Whites reporting SCD, compared to 12.8% among Blacks/African Americans, and 11.0% among Hispanics.¹

Diabetes represents a significant factor associated with cognitive function, drawing increasing attention due to its widespread prevalence and impact.^8,9 The relationship between diabetes and cognitive health has garnered attention in existing literature. Studies have consistently demonstrated that diabetes increases the risk of dementia, attributed to both vascular and neurodegenerative alterations in the brain,^10,11 emphasizing the intricate connection between metabolic health and cognitive function. Notably, a meta-analysis reveals a substantial association between diabetes and an increased risk of dementia with a relative risk of 1.59.¹² Moreover, research has also shown an increased SCD odds of 2.29 for individuals with versus without diabetes.¹³ Investigations into the association between diabetes and cognitive function have revealed that both type 1 and type 2 diabetes mellitus are linked to decreased performance across various cognitive domains, including concentration, memory retention, processing speed, executive function, and sustained attention performance.^14–17

Cardiovascular diseases (CVD) were also found to be significantly associated with cognitive health, whereby heart disease was twice as prevalent in individuals with SCD compared to those without it.¹⁸ Moreover, approximately 34% and 27% of individuals with SCD reported having a history of stroke and heart disease, respectively, while 18% reported having diabetes, highlighting the significant burden of CVDs and diabetes among individuals experiencing SCD in the US.¹⁸ A prospective clinical study focusing on cognitive impairment in post-myocardial infarction cardiovascular patients revealed that 40% of these individuals exhibited varying severity of cognitive impairment.¹⁹ Furthermore, having both diabetes and heart disease as co-morbid conditions was shown to double the risk of cognitive impairment and its progression to dementia,²⁰ and AD dementia.^20,21

Beyond diabetes and CVDs, a growing body of evidence suggests that other chronic diseases, such as arthritis, asthma, chronic kidney disease, and depression, are also relevant to cognitive health. These conditions are common in midlife and older adults and have been consistently associated with a higher prevalence of SCD. For instance, in adults aged 45–64 years, the prevalence of SCD stands at 19.6% for arthritis, 20.9% for asthma, and 25.6% for kidney disease, indicating that cognitive concerns often emerge in the context of broader chronic disease burden.^18,22 This underscores the importance of examining a wide spectrum of comorbidities when assessing SCD risk, rather than focusing solely on isolated conditions.

In addition to all the aforementioned chronic diseases and morbidities associated with cognition, social determinants of health (SDOHs) such as socioeconomic status, education, employment, race, rural/urban residency, and other factors, including social/emotional support, are also critical in influencing cognitive functions.^23–29 Similarly, lifestyle factors such as smoking, alcohol consumption, sleep patterns, and physical activity have been linked to cognitive health.^23,30–35 These factors are strongly associated with cognitive outcomes and may contribute to an increased risk of SCD. Therefore, examining their independent associations with SCD as well as adjusting for their effects is essential when addressing cognitive health.

Strategies for delaying or preventing pathological cognitive decline may be most effective when directed towards the initial symptomatic stages of the disease, such as SCD, when cognitive functioning is still relatively preserved.³⁶ Therefore, it is essential to identify factors and comorbidities linked to SCD, not just to identify high-risk groups but also to offer early interventions to these groups before they experience advanced stages of cognitive decline.

Despite growing interest in the associations between comorbid conditions and SCD, studies in this emerging field remain limited. Most existing research has focused on cognitive impairment at later stages, such as dementia and AD, or has assessed SCD in relation to individual conditions, rarely considering the combined effects of multiple comorbidities and the wider social, behavioral, and emotional context. The present study addresses this gap by investigating the associations between diabetes (distinguishing between type 1 and type 2, and assessing any potential role of insulin), cardiovascular diseases (including stroke, coronary heart disease, myocardial infarction, and angina), and other chronic comorbidities with SCD as the primary outcome. Importantly, this study applies a comprehensive multivariable framework that simultaneously accounts for multiple comorbidities alongside a wide range of biological, behavioral, social, and emotional determinants of health, allowing assessment of multimorbidity patterns and independent associations with SCD. Using this integrative framework, based on a large nationally representative US dataset, we also aim to identify determinants of health that may also correlate with SCD.

Methods

Data sources

This study utilized the Centers for Disease Control and Prevention's (CDC) Behavioral Risk Factor Surveillance System (BRFSS) data. BRFSS annually conducts over 400,000 interviews with noninstitutionalized adults aged 18 and above, residing in the US and participating regions. The primary focus of this data is on behavioral risk factors, chronic health conditions, the use of preventative services, and access to healthcare. The BRFSS 2022 questionnaire includes 2 sections: the core section including demographics, chronic conditions, lifestyle factors, and other variables; and optional modules, not administered to all the states, including the main outcome of this study, which is SCD.

Study design/methodological approach

The BRFSS survey uses random digit dialing of both cell phones and landlines to collect data from all 50 states, the District of Columbia, the US Virgin Islands, Guam, and Puerto Rico. Each survey respondent was weighted to account for differences in the probability of selection based on whether they are part of the landline or cell phone sample, their geographic region, the number of residential telephones in the household, the number of adults in the household, non-coverage, and non-response.

Study population

The target population from whom telephone samples were collected in 2022 includes individuals aged 18 years and older who live in private homes or college housing and have a mobile phone or landline. However, data on cognitive decline, our main outcome, was restricted to individuals who are aged 45 years and older residing in different states. In our study, we have included data collected from 12 states which answered the cognitive decline module, yielding a sample size of 60,492 participants. These states can be grouped into four regions: North (Maine, Rhode Island, Vermont), South (Florida, South Carolina, Virginia), Midwest (Indiana, Wisconsin), and West (Utah, Nevada, Idaho, Oregon).

Figure 1 below illustrates the participants included in this study and those who were not eligible to be included.

Figure 1.

Diagram showing the inclusion criteria for the study participants.

Concepts and measures

Predictors (diabetes and CVD)

The main predictors in our study include diabetes, CVDs, and other comorbidities. Cardiovascular conditions were assessed by posing 3 separate questions: “Has a doctor, nurse, or other health professional ever informed you of having any of the following conditions: i) heart attack, also known as a myocardial infarction, ii) coronary heart disease or angina, iii) a stroke.” The possible responses for each of these questions were “Yes”, “No”, “don’t know/Not sure”, or “refused” to answer. Those who answered “Yes” to one of questions i or ii were considered as having CHD. Similarly, the data on other comorbidities (including asthma, arthritis, kidney disease, and history of depressive disorder) were collected by a “Yes/No” question.

A similar question was also used for diabetes: “Has a doctor, nurse, or other health professional ever told you that you had diabetes?” Responses included no, prediabetes/borderline diabetes, yes, yes (for women who reported on incident diabetes only during pregnancy, known as gestational diabetes), don’t know/Not sure, or refusal to answer. Those responding “yes” were then asked, “According to your doctor or other health professional, what type of diabetes do you have?” and “How old were you when you were first told you had diabetes?” to identify the age of diagnosis with diabetes, and to differentiate between type 1 and type 2 diabetes. Moreover, the diabetic individuals were also asked: “Are you now taking insulin?” to check for insulin use. Diabetes duration was calculated for individuals with diabetes by subtracting age at diagnosis with diabetes from their current age, yielding the number of years lived with diabetes.

Respondents indicating pre-diabetes or borderline diabetes were excluded, in addition to women reporting gestational diabetes, as it is a temporary condition.

Outcome: SCD-inclusion/exclusion based on cognition

SCD was collected in the cognitive decline module (detailed below). Before asking the SCD-related question, the interviewers began with an introductory text describing the type of cognitive changes the module addresses to distinguish it from forgetfulness. Participants were told, “The next few questions ask about difficulties in thinking or remembering that can make a big difference in everyday activities; this does not refer to occasionally forgetting your keys or the name of someone you recently met, which is normal; this refers to confusion or memory loss that is happening more often or getting worse, such as forgetting how to do things you've always done or forgetting things that you would normally know; we want to know how these difficulties impact you.” The participants were then asked: “During the past 12 months, have you experienced confusion or memory loss that is happening more often or is getting worse?”. Those who answered “Yes” were considered as experiencing SCD. Participants who responded with “do not know/not sure” or refused to answer had their responses set to missing and were subsequently excluded from the analysis. This characterization of SCD was also employed in a previous study, which based presence of SCD on the affirmative response to this question.³⁴

Inclusion/exclusion criteria

The cognitive decline module captures self-reported perceptions of worsening memory rather than clinically confirmed diagnoses of AD, mild cognitive impairment (MCI), or other dementias. The survey is administered via telephone interview and requires independent self-reporting. Only non-institutionalized, community-dwelling adults are sampled, and proxy respondents are not permitted.^37–39 Participants who can articulate themselves clearly and without any difficulty while completing the telephone interview and who can understand and answer the questions during the telephone interview are the only ones included in the study. Meanwhile, respondents who are unable to complete the survey due to physical or cognitive limitations are removed from the sample, along with their entire household.^37–39 Individuals with moderate to severe AD or other dementias are unlikely to reliably complete unsupervised telephone surveys due to impairments in attention, comprehension, and communication,⁴⁰ and are therefore systematically excluded from the population-based telephone surveillance systems. Thus, the survey's methodological design and the selection and screening criteria for inclusion into the study employed by BRFSS inherently filters out individuals with moderate-to-severe cognitive impairments. Consequently, the study population primarily reflects individuals with subjective or early cognitive changes, and those with advanced cognitive impairment such as AD or other dementias are most likely excluded from this surveillance system.

Covariates

Demographic covariates include sex, race, educational attainment, housing arrangement, annual income, employment status, and marital status. Additional covariates include health-related factors such as health insurance and personal health care provider availability, lifestyle factors such as alcohol consumption, smoking habits, physical activity, and sleep patterns; with inadequate sleep determined as being fewer than 7–9 h for adults under 64, and less than 7–8 h for individuals over the age of 64. Overweight/obese and emotional support were also included in our study. Details on these variables and distribution among the general population and stratified by the outcome are presented in the descriptive table (Supplemental Table 1a).

Analysis plan

We initiated our analysis by assessing the distribution of SCD by state and region. Frequencies and weighted percentages of SCD were calculated for each state using BRFSS survey weights, and regional means were derived by averaging state-level percentages. This was followed by weighted descriptive analyses, conducted to generate summary statistics pertinent to the studied population. We carried out an unadjusted weighted analysis, employing the weighted Chi-squared test and simple weighted logistic regression to determine the crude associations between each predictor and SCD as the outcome. Then, multivariable models were built using multiple weighted logistic regressions. One model to determine the association between diabetes, CVDs, other comorbidities, determinants of health, and SCD; and the other models to determine these associations among diabetics in general and type 2 diabetic participants, in particular. In the models for diabetics, we included insulin use, age of diagnosis with diabetes, diabetes type, and diabetes duration. Details are in the Results section and the Supplemental Material.

Results

Distribution of SCD by states

Among the 12 states included in the analysis, the mean weighted percentages of SCD were generally similar across regions. The West had the highest mean weighted percentage of SCD at 11.91%, driven primarily by higher values in Nevada (13.36%) and Oregon (12.38%). The North and South showed comparable mean weighted percentages of 10.62% and 10.60%, respectively, while the Midwest had the lowest mean at 10.35%. Overall, the differences between regions were modest, with no region exhibiting a substantially higher or lower burden of SCD (Supplemental Table 1).

Characteristics of the study population:

Characteristics of the study population are presented in Supplemental Table 2. Our sample size comprised 60,492 participants, the majority of whom resided in urban areas (n = 49,034; 92.15%), were married or in a relationship (n = 35,861; 63.08%), and belonged to the white race (n = 51,092; 74.44%). Females slightly outnumbered males, constituting approximately 53% of the participants (n = 33,391). Education levels were high, with 32.46% (n = 16,714) having attended some college or technical school, and 31.77% (n = 26,647) being college graduates. Given the study's focus on older adults, the majority fell into the oldest age group (65 or older), comprising about 45% (n = 32,815), while 28.74% were between 55 and 64 years old (n = 15,895). A substantial proportion of the participants (n = 46,051; 92.31%) reported receiving emotional support. Current smokers constituted only 12.05% of the participants (n = 6534), while 56.1% (n = 33,674) had never smoked. Approximately half of the study population reported consuming at least one alcoholic drink in the past 30 days (n = 29,952; 49.98%). Sleep duration varied, with about 32% (n = 17,667) sleeping for 7–9 h, while the majority (n = 39,786; 63.52%) slept less than 7 h, considered inadequate for their age (Supplemental Table 2).

As for the frequency distribution of SCD and the comorbid conditions, our results indicated that among the 60,492 total participants, 6423 individuals, accounting for 10.45%, reported experiencing SCD (Table 1A). Regarding diabetic status, 19.49% were diagnosed with diabetes (n = 11,016), predominantly type 2 diabetes (n = 4964; 92.07%), with fewer having type 1 diabetes (n = 404; 7.93%) (Table 1B). Insulin use for diabetes management was reported by 33.41% of diabetic participants (n = 1862), and the mean age of diagnosis with diabetes was 50.85 years. The mean duration of diabetes is 14.93 years among the total population (Table 1B).

Table 1A.

Distribution of the study sample by outcome, SCD, main exposure variables, and other comorbidities.

	Total sample (60,492)		SCD		p
			Yes (6423)	No (54,069)
	N	Weighted %	N (Weighted column %)
Outcome Variable
Subjective Cognitive Decline
No	54,069	89.55	–
Yes	6423	10.45	–
Exposure Variables
Diabetes
No	49,476	80.51	4662 (69.33)	44,814 (81.81)	<0.001*
Yes	11,016	19.49	1761 (30.67)	9255 (18.19)
CHD
No	53,141	88.22	5043 (76.44)	48,098 (89.6)	<0.001*
Yes	7351	11.78	1380 (23.56)	5971 (10.4)
Stroke
No	57,002	94.33	5628 (86.31)	51,374 (95.27)	<0.001*
Yes	3490	5.67	795 (13.69)	2695 (4.73)
Other Comorbidities
Asthma
No	51,647	86.07	4969 (77.3)	46,678 (87.09)	<0.001*
Yes	8661	13.93	1409 (22.7)	7252 (12.91)
Arthritis
No	32,570	56.05	2378 (38.26)	30,192 (58.12)	<0.001*
Yes	27,646	43.95	3999 (61.74)	23,647 (41.88)
Depressive Disorder
No	48,789	81,63	3630 (56.49)	45,159 (84.55)	<0.001*
Yes	11,466	18.37	2742 (43.51)	8724 (15.45)
Kidney Disease
No	56,559	93.78	5691 (86.32)	50,868 (94.65)	<0.001*
Yes	3703	6.22	676 (13.68)	3027 (5.35)

Table 1B.

Distribution of diabetic individuals by diabetes type, insulin use, age at diagnosis with diabetes, and diabetes duration.

	Diabetics (11,016)		SCD		p
	N	Weighted %	Yes (6423)	No (54,069)	p
Diabetes Type
Type 2	4964	92.07	761 (90.25)	4203 (92.4)	0.1895^a
Type 1	404	7.93	75 (9.75)	329 (7.6)
Insulin Use
No	3872	66.59	557 (59.58)	3315 (67.88)	0.0015^a,c
Yes	1862	33.41	341 (40.42)	1521 (32.12)

Age of diagnosis with Diabetes (Mean)	50.85		49.18	51.17	0.014^b,c
Duration of Diabetes (Mean)	14.93		16.84	14.57	0.001^b,c

Weighted Chi-square test.

weighted simple logistic regression.

p ≤ 0.05 indicating significant results.

The prevalence of CHD was 11.78% (n = 7351) while the prevalence of stroke was 5.67% (n = 3490). In regards to other morbidities, only 18% reported suffering from a depressive disorder (n = 11,466), few participants had kidney disease (n = 3703; 6.22%) or asthma (n = 8661; 13.93%), while many reported arthritis (n = 27,646; 43.95%) (Table 1A).

Distribution of diabetes, CVDs, and other comorbidities based on SCD status

Table 1A and 1B also illustrate the prevalence of the comorbid diseases, which are our primary predictors, cross-classified by SCD status, displaying weighted column percentages alongside p-values, calculated using the weighted Chi-square test. Among participants experiencing SCD, 30.67% (n = 1761) reported having diabetes, with 90.25% (n = 761) of them diagnosed with type 2 diabetes, and only 9.75% (n = 75) with type 1. The prevalence of diabetes is lower for participants with no SCD at 18.19% (9255). The Chi-square test yielded a p-value of <0.001, indicating a significant association between SCD and diabetes. A significant association was also captured between insulin use among diabetics and SCD (p = 0.0015), with 40.42% of diabetics with SCD being on insulin, while a lower percentage of 32.12% of insulin use was observed for diabetics with no SCD (Table 1B). The mean age of diagnosis with diabetes for those with SCD was 49.18 years compared to a higher average of 51.17 years for diabetics without SCD. This result indicates that individuals with SCD have a younger age of diagnosis with diabetes on average compared to those with no SCD, with a significant p-value = 0.014. Similarly, participants with SCD had a longer mean duration of diabetes compared with those without SCD (16.84 versus 14.57 years; p = 0.001).

In terms of CVDs, the prevalence of CHD is 10.4% (n = 5971), more than the prevalence of stroke at 4.73% (2699) among participants who did not report experiencing SCD. Similarly, among SCD participants, CHD emerged as more prevalent at 23.56% (n = 1380), compared to stroke which was present in approximately 13.7% (n = 795) of individuals with SCD. The distribution of CVDs among SCD patients demonstrates significance, with a Chi-square p-value of <0.001. In addition, for the other comorbidities, arthritis seems to be the most prevalent among participants with SCD, accounting for 61.74% (n = 3999), followed by depressive disorder at 43.51% (n = 2742), asthma at 22.7% (1409), and kidney disease at 13.68% (n = 676), with a significant p-value <0.001.

SCD, CVDs, diabetes, other comorbidities, and other covariates

The frequency distribution and unadjusted associations between SCD and other covariates, including SDOHs and other behavioral variables, along with their corresponding 95% confidence intervals (CI) and p-values, are presented in Supplemental Tables 2 and 3, respectively. By introducing these crude associations, we identify significant variables with a p-value <0.05, which will serve as the basis for constructing our multivariable models and will be integrated into the adjusted full models (Tables 2 –4) to account for their effects. Table 2 was constructed to investigate the associations between diabetes, CVDs, other comorbidities, and SCD among all participants, while Table 3 examined the associations between CVDs, insulin use, age of diagnosis with diabetes, other comorbidities, and SCD among diabetic patients. Similar to Table 3, Table 4 examined the associations among diabetic patients; but included the duration of diabetes with diabetes type, to better capture the SCD-insulin association. The independent adjusted effects of SDOH and behavioral factors on SCD, as included in our model of predictors (comorbidities), are reported in Supplemental Tables 4 and 5, for the whole sample and the diabetic subgroup, respectively. Interpretation of our results will focus on the comorbid diseases presented in the multivariable models (Tables 2 –4).

Table 2.

Logistic regression to examine association between diabetes, CHD, Stroke, and other comorbidities with SCD among total participants (60,492).

Subjective cognitive decline	Weighted unadjusted OR (95% CI)	p	Weighted adjusted^a OR (95% CI)	p
Predictors
Diabetes
No	Ref		Ref
Yes	1.99 (1.70- 2.31)	<0.001^b	1.25 (1.02–1.55)	0.031^b
Cardiovascular Diseases
CHD
No	Ref		Ref
Yes	2.65 (2.29- 3.07)	<0.001^b	1.56 (1.27- 1.92)	<0.001^b
Stroke
No	Ref		Ref
Yes	3.19 (2.63- 3.85)	<0.001^b	1.61 (1.24- 2.08)	<0.001^b
Other Comorbidities
Depressive Disorder
No	Ref		Ref
Yes	4.21 (3.69–4.80)	<0.001^b	2.56 (2.18–3.02)	<0.001^b
Kidney Disease
No	Ref		Ref
Yes	2.80 (2.16–3.62)	<0.001^b	1.89 (1.24–2.87)	0.003^b
Asthma
No	Ref		Ref
Yes	1.98 (1.69–2.31)	<0.001^b	1.27 (1.06–1.52)	0.008^b
Arthritis
No	Ref		Ref
Yes	2.24 (1.96–2.55)	<0.001^b	1.31 (1.07–1.61)	0.008^b

Adjusting for socio-demographic factors and SDOHs (sex, age, race, education, employment, income level, place of residence, health insurance, home ownership, marital status, emotional support), and behavioral factors (smoking, alcohol use, exercise, sleep duration).

p ≤ 0.05 indicating significant results.

Table 3.

Logistic regression to examine association between insulin use, age of diabetes diagnosis, diabetes type, CHD, Stroke, and other comorbidities with SCD, among diabetics (11,016).

Subjective cognitive decline	Weighted unadjusted OR (95% CI)	p	Weighted adjusted^a OR (95% CI)	p
Predictors
Age of diagnosis with diabetes	0.98 (0.98–0.99)	0.014^b	1.00 (0.99- 1.02)	0.247
Insulin Use
No	Ref		Ref
Yes	1.43 (1.14–1.79)	0.002^b	1.36 (1.00- 1.84)	0.044^b
Diabetes Type
Type 2	Ref		Ref
Type 1	1.31 (0.87–1.97)	0.191	1.22 (0.75–1.98)	0.417
Cardiovascular Diseases
CHD
No	Ref		Ref
Yes	1.88 (1.45- 2.44)	<0.001^b	1.52 (1.12- 2.06)	0.007^b
Stroke
No	Ref		Ref
Yes	2.22 (1.64- 2.99)	<0.001^b	1.88 (1.34- 2.65)	<0.001^b
Other Comorbidities
Depressive Disorder
No	Ref		Ref
Yes	3.35 (2.56–4.37)	<0.001^b	3.15 (2.39–4.15)	<0.001^b
Kidney Disease
No	Ref		Ref
Yes	1.95 (1.27–2.98)	0.002^b	1.11 (0.78–1.58)	0.542

Asthma
No	Ref		Ref
Yes	1.59 (1.24–2.05)	<0.001^b	1.51 (1.12–2.04)	0.006^b
Arthritis
No	Ref		Ref
Yes	1.59 (1.15–2.20)	0.005^b	1.50 (1.13–1.98)	0.004^b

p ≤ 0.05 indicating significant results.

Table 4.

Logistic regression to examine the association between insulin use, diabetes duration, diabetes type, CHD, stroke, and other comorbidities with SCD.

Subjective cognitive decline	Weighted adjusted OR (95% CI), among diabetics (11,016)	p	Weighted adjusted^a OR (95% CI), among type-2 diabetics (4964)	p
Predictors
Duration of Diabetes	0.99 (0.98–1.00)	0.422	0.99 (0.98–1.01)	0.672
Insulin Use
No	Ref		Ref
Yes	1.34 (0.99–1.81)	0.057^c	1.39 (1.02- 1.91)	0.036^b
Diabetes Type
Type 2	Ref		-	-
Type 1	1.19 (0.74–1.94)	0.463
Cardiovascular Diseases
CHD
No	Ref		Ref
Yes	1.52 (1.12- 2.07)	0.006^b	1.52 (1.11- 2.08)	0.009^b
Stroke
No	Ref		Ref
Yes	1.89 (1.35- 2.66)	<0.001^b	2.13 (1.50- 3.02)	<0.001^b
Other Comorbidities
Depressive Disorder
No	Ref		Ref
Yes	3.14 (2.38–4.14)	<0.001^b	2.08 (2.10–3.72)	<0.001^b
Kidney Disease
No	Ref		Ref
Yes	1.11 (0.78–1.57)	0.561	1.13 (0.78–1.63)	0.508
Asthma
No	Ref		Ref
Yes	1.51 (1.12–2.03)	0.006^b	1.49 (1.10–2.04)	0.010^b
Arthritis
No	Ref		Ref
Yes	1.50 (1.13–1.98)	0.004^b	1.51 (1.12–2.02)	0.006^b

p ≤ 0.05 indicating significant results.

bordeline significance.

In the overall multivariable model (Table 2), diabetes, CHD, stroke, and all other comorbidities showed significant associations with SCD (p < 0.05). Individuals with stroke had 1.61 higher odds of SCD (95%CI: 1.24–2.08), followed by CHD with an odds ratio of 1.56 (95% CI: 1.27,1.92), and diabetes (OR: 1.25; 95%CI: 1.02–1.55). For other comorbidities, depressive disorder stood out as a strong correlate of SCD with a 2.56 times increased odds (95% CI: 2.18–3.02), followed by kidney disease (OR: 1.89; 95% CI: 1.24–2.87), arthritis (OR: 1.31; 95%CI: 1.07–1.61), and asthma (OR: 1.27; 95%CI: 1.06–1.52), in a model that accounts for indices of SDOH and behavioral variables. To assess the robustness of the main findings, sensitivity analyses excluding participants with stroke or cardiovascular disease resulted in estimates that are consistent with the main multivariable model. Excluding participants with stroke, diabetes (OR = 1.30; 95% CI: 1.05–1.61; p = 0.016), and cardiovascular disease (OR = 1.30; 95% CI: 1.19–1.86; p = 0.001) remained significantly associated with SCD. When participants with cardiovascular disease were excluded, diabetes (OR = 1.29; 95% CI: 1.03–1.63; p = 0.027) and stroke (OR = 1.44; 95% CI: 1.01–2.07; p = 0.044) also remained significant. Interaction terms between diabetes and stroke (p = 0.203) and between diabetes and cardiovascular disease (p = 0.263) were not statistically significant, indicating no evidence of effect modification.

In Table 3, the analysis was focused on diabetic individuals. Insulin use, age of diagnosis with diabetes, and diabetes type were added to the previous model to examine their associations with SCD among diabetics. Our results showed that insulin use is significantly associated with elevated SCD odds (OR: 1.36; 95% CI: 1.00–1.84). However, the age of diabetes diagnosis and diabetes type were not statistically significant (p = 0.247 and 0.417, respectively) in this model, noting that the age of diabetes diagnosis was significant in the unadjusted association. CVDs in this model also have an increased OR, with stroke (OR: 1.88; 95%CI: 1.34–2.65) and CHD (OR: 1.52; 95%CI: 1.12–2.06) being significantly associated with SCD. Additionally, other comorbidities remain significant, except for kidney disease, which is no longer significant for diabetics (p = 0.542). In this model, the magnitude of the association between depressive disorder and SCD increased compared to the previous model, where the whole sample was considered in the analysis (OR: 3.15; 95%: 2.39–4.15).

Given the importance of diabetes duration in the association between insulin use and SCD, we constructed two additional models: one including all participants with diabetes along with diabetes type as predictor, and one restricted to type 2 diabetics (Table 4). Both models included diabetes duration, CHD, stroke, other comorbidities, and all the remaining covariates that were previously included in the multivariable model, presented in Table 3. Among all diabetics, insulin use continued to have higher SCD odds (OR = 1.34; 95% CI: 0.99–1.84); however, this association was borderline significant (p = 0.057) (Table 4). In the type 2 diabetes subgroup, the association reached statistical significance (p = 0.036), whereby those who were on insulin were associated with 1.39 (95% CI: 1.02–1.91) times higher odds of SCD. Associations for the other comorbidities were similar to those observed in the previous model (Table 3).

Our findings also identified several key social and behavioral determinants associated with SCD. As shown in Supplemental Table 4, divorced or separated participants had higher odds of SCD compared with those who were married or partnered (OR = 1.51; 95% CI: 1.10–2.06). Employment status was similarly important: participants who were unemployed, unable to work, or retired had increased odds of SCD (OR = 1.44, 1.32, and 2.45, respectively; all p-values < 0.05) compared with those who were employed. Receiving emotional support was associated with substantially lower odds of SCD (OR = 0.52; 95% CI: 0.42–0.65), and inadequate sleep (<7 h) was linked to higher SCD odds (OR = 1.41; 95% CI: 1.17–1.70) relative to those sleeping 7–9 h. These results highlight the significance of social and lifestyle factors in shaping SCD risk in addition to medical comorbidities.

Discussion

SCD has been increasingly recognized as an early marker of more advanced cognitive impairment; however, prior studies have generally examined SCD within limited analytic frameworks. Existing studies have focused on selected modifiable risk factors or specific populations, single comorbidities rather than multimorbidity patterns, often adjusting for a limited set of sociodemographic variables or reporting only descriptive prevalence estimates.^{13,18,41–43} In contrast, in the present study, we examined the associations between SCD and a cluster of comorbidities assessed within a broad framework of biological, behavioral, social, and emotional indices of health. We have also assessed SCD in relation to type 1 and type 2 diabetes, age of diagnosis with diabetes, duration of diabetes, and insulin use among all diabetics in general and type 2 diabetes in particular, research that has not been much highlighted in the literature. In addition, our analysis identified several predictors that have been infrequently examined in prior SCD research, including sleep duration, emotional support, employment status, and marital status, which remained significantly associated with SCD after comprehensive adjustment. This approach helps us determine multimorbidity patterns rather than focusing on single comorbidities and provide a more integrated assessment of SCD risk within a holistic framework of several determinants of health.

In this study, we have also carried out frequency distribution of SCD stratified by state and region. Our results showed modest geographical differences in the weighted percentages of SCD that may be attributed to the limited number of states included in each region. Geographic variations likely reflect differences in the study participants’ socioeconomic and social characteristics, such as health awareness, access to care, and demographic composition, as well as broader social, cultural, and healthcare-related factors that shape regional patterns of SCD.^42,44–47

In terms of comorbidities, our results highlight that diabetes was more prevalent among participants experiencing SCD, compared to CHD and stroke. Stroke had a higher risk effect on SCD in our main multivariable model compared to CHD and diabetes. Moreover, upon narrowing our focus to diabetics, insulin use was significantly associated with elevated SCD odds, particularly exhibited among type 2 diabetics. Interestingly, depression emerged as the most influential factor compared to the main and other comorbid diseases, whereby individuals with depressive disorder having more than double the odds of SCD compared to those not having depressive disorder. Beyond the associations between comorbidities and SCD, several social and behavioral determinants of health were also significantly associated with SCD odds. Specifically, being married or partnered (versus divorced or separated), being employed, receiving adequate emotional support, and maintaining sufficient sleep were all associated with lower odds of SCD, highlighting the protective role that these factors may have on SCD.

CVDs as risk factors for cognitive decline

Our results highlighted that CHD is significantly associated with SCD. Compared to adults who did not report experiencing CHD, the estimated adjusted and unadjusted odds of SCD are 1.56 and 2.86, respectively. In a similar study analyzing BRFSS data of SCD module from 2017–2021 for adults aged 45–64 years, the crude OR of SCD for individuals who reported CHD was 3.26. It was reduced to 2.23 in the adjusted model. Their results showed a relatively stronger OR compared to our findings. However, it is important to note that in their study, the adjusted model included only race, income, and education, while our study was more comprehensive as it incorporated a wider range of relevant variables and adjustments.⁴⁸ Additionally, their analysis examined CHD, stroke, and diabetes in separate models alongside race, income, and education,⁴⁸ whereas our study included these comorbidities in a single model to better capture their combined influence.

In agreement with our results showing elevated SCD among participants with CHD, a prospective longitudinal cohort study of primary care patients in Germany assessed cognitive decline using the Mini-Mental State Examination (MMSE) and the Clinical Dementia Rating Sum of Boxes (CDR-SoB). The study reported that, compared with patients without CHD, those with CHD experienced a 66% accelerated rate of cognitive decline (2.5 versus 1.5 MMSE points per year) and an approximately 83% greater annual increase in cognitive-functional impairment (2.2 versus 1.2 CDR-SoB points per year) throughout the study period.⁴⁹ Furthermore, a prospective cohort study concluded that angina is significantly associated with a moderately elevated cognitive decline with a 1.45 hazard ratio (HR).⁵⁰ In contrast, the findings of a long-term follow-up study within the Cardiovascular Risk Factors, Ageing, and Dementia (CAIDE) did not find any evidence of the association between CHD occurring in mid-life or later life and an elevated risk of dementia.⁵¹

Evidence has shown that the association between CHD and cognitive decline may involve direct mediation. CHD might induce cerebral hypoxia, leading to a reduced supply of oxygen to brain tissues, along with the development of silent brain lesions such as white matter hyperintensities and ventricular enlargement.^52–54 Additionally, findings from a postmortem study have indicated that ischemic heart disease, a common manifestation of CHD, is associated with cerebral microinfarcts, which are tiny areas of tissue damage in the brain. This association suggests a potential link between CHD and cerebral small vessel disease, which could contribute to cognitive impairment.⁵⁵

Vascular risk factors (VRFs), such as smoking, hypertension, obesity, physical inactivity, and diabetes, were also identified in the literature as common risk factors between CHD and cognitive decline,^56–58 that may contribute to the development of neurodegenerative lesions in the brain, such as amyloid-β.⁵² VRFs long exposure starting in early life is considered serious and might contribute to cognitive decline via different mechanisms, including oxidative stress, immune responses, and endothelial dysfunction.^52,59 This comes in support of our findings linking diabetes, smoking status, alcohol consumption, and exercise as significant correlates of SCD.

Our results showed that the OR of SCD is 1.61 times higher for participants who have a history of stroke. In agreement, a recent study found that the OR of SCD among individuals who have a history of stroke is about 3 times higher, noting that the stronger magnitude in their study might be due to a few variables in their main model as compared to our model.⁴⁸ Furthermore, a meta-analysis of nine longitudinal hospital-based cohorts, investigating long-term cognitive decline following stroke, found that compared to individuals without a history of stroke, cognitive decline accelerates faster within the first three years post-stroke.⁶⁰ At the time of diagnosis, stroke primarily affects attention and executive functions more profoundly than memory, although memory impairment may manifest at different intervals following the stroke event.^61–64

The link between cognitive functions and stroke is not well known. Findings from the literature suggest that the neuroanatomical lesions resulting from a stroke can have a significant impact on cognitive function.⁶⁵ Specifically, strokes often affect strategic areas of the brain, such as the hippocampus, which plays a crucial role in memory formation and retrieval.⁶⁵ Additionally, strokes can lead to white matter lesions, which can disrupt communication between different brain regions, impairing cognitive processes such as attention and information processing.⁶⁵ Strokes can also cause cerebral microbleeds as a result of small cerebrovascular diseases; these microbleeds involve tiny hemorrhages in the brain, which can further contribute to cognitive impairment by damaging surrounding brain tissue and disrupting neural connections.⁶⁵

Diabetes, insulin use, and cognitive decline

Diabetes emerges as a significant predictor in our study, correlating with elevated SCD odds. Among adults reporting SCD, approximately 30.67% were diagnosed with diabetes, and nearly 90% of them identified as having type 2 diabetes. Although previous evidence has demonstrated discrepancies in the impact of diabetes type on cognitive functions such as verbal episodic memory, executive function, and psychomotor processing speed,^66,67 our study was unable to examine this distinction due to the limited number of participants with type 1 diabetes (n = 404, 7.93%), presenting a notable limitation of our study.

Our results showed that the crude association between diabetes and SCD is significant, with an unadjusted OR of 1.99. In the multivariable model, individuals with diabetes have 1.25 times higher odds of SCD compared to those without diabetes. Similarly, an analysis of BRFSS 2017–2021 data involving 110,305 participants aged 45–64 showed an unadjusted OR of 2.23 quantifying the association between diabetes and SCD, and a 1.92 adjusted OR in a model that accounts only for race, income, and education,⁴⁸ whereas our study involves a more comprehensive model. Additionally, a prospective cohort study conducted within a large community setting tracked participants for 20 years, revealed that diabetes diagnosed during midlife correlates with a 19% increase in cognitive decline.⁶⁸ Prior studies revealed that the association between diabetes and cognitive decline is related to hyperglycemia, elevated blood sugar levels, which can result in the formation of advanced glycosylated end products and oxidative stress, both of which are recognized as contributors to neuronal and vascular endothelial damage, ultimately leading to cognitive impairment.⁶⁹ Along the same lines, previous findings have shown that diabetes significantly influences how the brain manages amyloid-β and tau proteins, potentially accelerating the development of senile plaques and neurofibrillary tangles, characteristic signs of AD.⁷⁰

Effective management of blood glucose levels is closely linked to cognitive function. A community-based cohort study revealed that diabetic patients experience a more pronounced decline in cognitive abilities when their glucose levels remain uncontrolled.⁶⁸ Moreover, a study utilizing data from the 2018 Korea Community Health Survey found that better blood glucose control was associated with a lower likelihood of SCD among diabetic patients.⁷¹

However, the link between blood glucose control and cognitive health is dependent on the type of antidiabetic medication being used. For instance, metformin and dipeptidyl-peptidase-4 inhibitors (DPP-4i) were found to be correlated with a slower decline of cognition assessed using the MMSE scores, while the use of insulin and sulfonylureas was associated with a larger decline in cognition.⁷² The latter result is in concordance with our finding, which highlights that users of insulin experience a decline in their cognitive functions, identifying insulin use as a marker associated with increased odds of SCD. Nonetheless, findings concerning the effect of insulin use on cognition were not consistent. In this regard, recent randomized controlled trials have demonstrated no significant difference between the intranasal insulin and placebo groups concerning the cognition of patients with MCI or dementia.⁷³ Additionally, a pooled analysis from five population-based cohorts revealed that insulin use showed no association with cognitive function or brain MRI measures.⁷⁴

To delve deeper into the association between insulin use and SCD, we conducted an additional analysis among individuals with diabetes (Table 4), in which we fitted a fully adjusted model that included diabetes duration and diabetes type in addition to comorbidities and social and behavioral determinants of health. In this model, the direction and magnitude of the association between insulin use and SCD remained consistent with the primary analysis, although the statistical significance was attenuated after accounting for diabetes duration, suggesting that disease chronicity may partially explain this relationship. To further address potential confounding by indication, we performed a subgroup analysis restricted to participants with type 2 diabetes, among whom insulin therapy is typically initiated when glycemic control cannot be achieved with lifestyle modification or oral glucose-lowering medications and when pancreatic insulin production becomes insufficient.^75,76 In this context, insulin use is more likely to reflect more advanced disease. Using the same fully adjusted model, insulin use remained significantly associated with higher odds of SCD in individuals with type 2 diabetes. Together, these findings support the interpretation that insulin use may serve as a marker for a prolonged diabetes duration and underlying diabetes severity rather than a direct independent contributor to SCD, while acknowledging the potential for residual confounding due to the absence of glycemic control measures such as HbA1c in the BRFSS dataset.

Findings from the literature suggest that the increase in SCD among diabetics being treated with insulin could be related to insulin resistance.^77,78 For instance, patients with AD often exhibit insulin resistance in neuronal cells, suggesting that this resistance may trigger neuronal dysfunction, contributing to cognitive impairment and the eventual onset of dementia.⁷⁷ Moreover, peripheral insulin resistance may lead to neuronal damage via pathways involving amyloid-β and cytokines. Peripheral hyperinsulinemia has the potential to elevate levels of amyloid-β and inflammatory cytokine levels, which could contribute to neuronal loss, the formation of amyloid-β plaques, and the development of neurofibrillary tangles, thus affecting cognitive functions.⁷⁹

Other comorbidities and cognitive decline

Depression emerged as a significant risk factor in our study, with individuals suffering from depressive disorder showing a 2.56-fold increase in the odds of developing SCD. This OR was the highest among all the predictors studied, highlighting the importance of depression in correlating with cognitive outcomes. Our findings align with the results of a study conducted using BRFSS 2015–2018 data, which found an increased 3.12 adjusted RR for SCD among individuals who have a depressive disorder.⁸⁰ This association can be explained by consistent evidence showing that individuals with Major Depressive Disorder perform worse on neuropsychological assessments.⁸¹ In addition, a recent study among patients with major depression shows that subjective cognitive complaints were associated with altered connectivity and entropy in the insula and superior temporal region, correlating with executive function deficits.⁸²

Given the major influence of diabetes, CVD, and depression on SCD risk as concluded in our study, it is crucial to delve into the possible interactions of these factors in correlation with cognitive outcomes. For instance, studies consistently find that diabetics who also have depression suffer faster cognitive deterioration. A systematic review pooled data across multiple cohorts found that, compared to diabetics without depression, those with depression had a significantly larger decline in executive function, language, and memory scores, and exhibited nearly double the risk of progressing to dementia (HR≈1.82)⁸³ Similarly, a longitudinal study reports that type 2 diabetic patients with depressive symptoms lose cognitive functions more quickly on tests of processing speed and memory.⁸⁴ In addition, depressive symptoms after stroke frequently co-occur with impairments in multiple cognitive domains, particularly executive function, attention, memory, and processing speed. Moreover, those with higher depression scores often perform worse on neuropsychological tests than non-depressed stroke patients..^85,86 Furthermore, in patients with late-life depression, comorbid CVD or related risk factors were associated with a significantly higher rate of global cognitive deterioration over one year compared to depressed patients without these comorbidities, suggesting synergistic effects on cognitive decline.⁸⁷ All of these findings suggest a potential interplay between diabetes, CVD, and depression in relation to cognitive decline.

Our results showed that arthritis is more prevalent among SCD participants (61.74%) compared to non-SCD participants (41.88%). This finding is in concordance with recently reported CDC data, which revealed that the prevalence of arthritis was higher for SCD individuals aged 45 to 64 years (60%) compared to individuals within the same age group but with no SCD (30%).⁸⁸ In our study, the adjusted OR for SCD risk is 1.31 times for those with arthritis. Previous research has found that individuals with arthritis often exhibit poorer cognitive functions, particularly significant declines in immediate and delayed word recall, reduced semantic verbal fluency, and decreased numeric abilities.⁸⁹ The chronic pain associated with arthritis can disrupt concentration, memory, and attention.^89,90 Similarly, a recent study concluded that out of the individuals diagnosed with severe asthma, a considerable proportion exhibited cognitive decline, with verbal learning and memory functions being highly impacted.⁹¹ Patients with asthma display heightened AD pathology characterized by reduced levels of amyloid-β and elevated phosphorylated-tau, along with increased concentrations of synaptic degeneration biomarkers.⁹² All of these findings come in support of our results showing elevated 1.27 SCD odds for participants with asthma. Furthermore, we also found a doubled adjusted OR for SCD risk among individuals who suffer from kidney disease. Comparably, a recent study concluded that the odds of SCD is tripled for chronic kidney disease patients, aligning with evidence suggesting that individuals with kidney disease are more susceptible to cognitive disorders.⁹³

Cognitive decline and determinants of health

An important aspect of our study was identifying determinants of health, beyond comorbidities, as correlates of SCD. Our findings showed that inadequate sleep is significantly associated with increased SCD odds. Consistent with our results, population-based studies, focusing solely on SCD in relation to sleep, have reported that poor sleep quality, including short duration and sleep disturbances, is linked to higher odds of SCD and impaired cognitive performance.^94,95 Other studies have shown that insufficient or disrupted sleep may impair memory consolidation and reduce the brain's clearance of metabolic waste, including neurotoxic proteins, which together can contribute to an increase in cognitive decline.^96–98 Our results suggest that being married or coupled may be protective, as it is associated with lower odds of SCD compared to being divorced or separated. These results are in line with previous research demonstrating a protective association between partnership and cognitive health. Evidence suggests that divorced or separated individuals experience poorer cognitive outcomes compared with those who are married, likely reflecting lower levels of social engagement and networking.⁹⁹ Similarly, a recent study has shown that having a spouse is associated with better cognitive functioning, enhanced by informal social support.¹⁰⁰ Additionally, our results showed that receiving emotional support was associated with lower odds of SCD. This is consistent with evidence indicating that individuals reporting higher levels of emotional support experience better cognitive outcomes.^101,102 A large study conducted in the US, which incorporated 17,206 adults aged 46 years and older, concluded that 21.15% SCD cases were among participants who did not receive emotional/social support, as compared to a lower percent of 8.45% among those who received it.²⁸ Evidence has shown that emotional support may buffer the effects of stress, enhance psychological well-being, and promote engagement in healthy behaviors, all of which can help preserve cognitive function.^102,103 Furthermore, our results showed that individuals who were unemployed, unable to work, or retired had higher odds of SCD compared with those who were employed, suggesting that engagement in work may improve cognitive health. Employment provides cognitive stimulation, social interaction, and structured routines, all of which can promote positive neuroplasticity that can potentially help sustain cognitive health over time.¹⁰⁴ Supporting this, evidence indicates that individuals with complex or cognitively demanding jobs tend to maintain better cognitive function with age and have a lower risk of dementia.^105,106 Taken together, our findings highlight the importance of the social and emotional aspects in determining cognitive health.

Strengths and limitations

Our study stands out due to several key factors that contribute to its importance and strength. Firstly, there is a scarcity of research exploring SCD, making our study one of the few to focus on this critical issue. Moreover, we delve into common comorbidities that encompass CVDs, diabetes, asthma, arthritis, kidney disease, and depressive disorder, as correlates of SCD, resulting in a comprehensive multimorbidity examination of cognitive decline. A key strength of this study is the broader assessment of SCD across diabetes type, age at diagnosis, duration of diabetes, and insulin use, with particular attention to type 2 diabetes, factors that have received limited attention in previous research. Another aspect that contributes to the robustness of our findings is the assessment of these relationships within a comprehensive framework that encompasses various social, biological, and behavioral indices of health, highlighting novel factors associated with cognitive health.

While this study offers valuable insights and significant findings, it is essential to acknowledge some limitations that warrant consideration. Firstly, although the focus is on SCD, incorporating objective measures of cognitive performance would enhance the accuracy and reliability of the results. Even though in the methodological approach for the selection of the participants, those with advanced cognitive impairment were most likely inherently excluded from the study, relying on a diagnostic test to identify these cases would have been more affirmative. Moreover, the inclusion of the BRFSS cognitive decline data from only 12 states, limits geographic coverage, and reduces variability in the study population and generalizability to the broader US population. Along the same lines, in view of the fact that not all the states administered the optional cognitive decline module, response bias could be introduced, potentially leading to an underestimation of SCD prevalence and discrepancies in its prevalence between different regions. In addition, the self-reporting of data can also have some embedded recall bias and an over- or under-reporting of information. Lastly, due to the small number of individuals with type 1 diabetes in the study sample, conducting an analysis that distinguishes between the effects of diabetes type on SCD was not achievable.

Conclusions

Our study highlights the significant associations between multimorbidities and SCD, emphasizing the necessity for deeper exploration of the mechanistic links between these conditions and cognitive functions in future research. The association between insulin use and elevated odds of SCD suggests the need for future studies to disentangle whether this relationship reflects underlying diabetes severity or a potential independent association of insulin with decreased cognitive function, particularly among those with type 2 diabetes.

Moreover, our findings suggest that SCD reflects not only underlying chronic disease burden but also broader modifiable social and behavioral determinants, highlighting the importance of incorporating these factors into cognitive health surveillance and intervention strategies. Future longitudinal studies are needed to clarify temporal relationships and to determine whether modifying these factors may reduce the burden of SCD.

Given the potential impact of these comorbidities on SCD, individuals with such health conditions should undergo regular proactive cognitive screenings and receive comprehensive care that addresses both their physical and psychological needs. In conclusion, recognizing SCD as a critical public health concern is essential given its role as a potential precursor to dementia and AD, underscoring the need for greater attention to mitigate its burden and subsequent risk on cognitive health.

Supplemental Material

sj-docx-1-alz-10.1177_13872877261441275 - Supplemental material for Subjective cognitive decline among older adults and its link to diabetes, cardiovascular diseases, and other comorbidities: A US population-based study

Supplemental material, sj-docx-1-alz-10.1177_13872877261441275 for Subjective cognitive decline among older adults and its link to diabetes, cardiovascular diseases, and other comorbidities: A US population-based study by Marwa Shouman, Ayad A. Jaffa, Abla M. Sibai, Kelly Hunt and Miran A. Jaffa in Journal of Alzheimer's Disease

Footnotes

Acknowledgements

ChatGPT, version GPT-5, a language model developed by OpenAI (), was used to assist in language refinement and improvement of the manuscript's text. All content was reviewed and verified by the authors to ensure accuracy and integrity.

ORCID iD

Miran A. Jaffa

Ethical considerations

The data for this study is retrieved from a secondary source that is publicly available, thus the ethical approval from the Institutional Review Board (IRB) from the American University of Beirut (AUB) was deemed not required. To protect the data's privacy and confidentiality, de-identification techniques were used, and all participants gave their informed consent. This study was carried out in agreement with the 2013 revised version of the Declaration of Helsinki.

Consent to participate

Not applicable

Consent for publication

Not applicable

Author contribution(s)

Marwa Shouman: Conceptualization; Data curation; Formal analysis; Software; Visualization; Writing – original draft; Writing – review & editing.

Ayad A. Jaffa: Conceptualization; Data curation; Formal analysis; Investigation; Methodology; Project administration; Software; Supervision; Validation; Writing – review & editing.

Abla M. Sibai: Supervision; Visualization; Writing – review & editing.

Kelly Hunt: Supervision; Validation; Writing – review & editing.

Miran A. Jaffa: Conceptualization; Data curation; Formal analysis; Investigation; Methodology; Project administration; Software; Supervision; Validation; Writing – review & editing.

Funding

The authors received no financial support for the research, authorship, and/or publication of this article.

Declaration of conflicting interests

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

Data availability statement

The data supporting the findings of this study are openly available in the Behavioral Risk Factor Surveillance System (BRFSS) at the Centers for Disease Control and Prevention (CDC) website, accessible at .

Supplemental material

Supplemental material for this article is available online.

References

CDC. Subjective Cognitive Decline — A Public Health Issue, https://www.cdc.gov/aging/data/subjective-cognitive-decline-brief.html#:∼:text=The%20prevalence%20of%20subjective%20cognitive,or%201%20in%209%20adults (2019).

Jessen

Amariglio

Van Boxtel

, et al. A conceptual framework for research on subjective cognitive decline in preclinical Alzheimer's disease. Alzheimers Dement 2014; 10: 844–852.

Marafioti

Culicetto

Latella

, et al. Neural correlates of subjective cognitive decline in Alzheimer’s disease: a systematic review of structural and functional brain changes for early diagnosis and intervention. Front Aging Neurosci 2025; 17: 1549134.

Jessen

Amariglio

van Boxtel

, et al. A conceptual framework for research on subjective cognitive decline in preclinical Alzheimer's disease. Alzheimers Dement 2014; 10: 844–852.

Kuhn

Klinger

Amariglio

, et al. SCD-plus features and AD biomarkers in cognitively unimpaired samples: a meta-analytic approach for nine cohort studies. Alzheimers Dement 2025; 21: e14307.

Sun

Yang

Lin

, et al. Biochemical and neuroimaging studies in subjective cognitive decline: progress and perspectives. CNS Neurosci Ther 2015; 21: 768–775.

Röhr

Pabst

Riedel-Heller

, et al. Estimating prevalence of subjective cognitive decline in and across international cohort studies of aging: a COSMIC study. Alzheimers Res Ther 2020; 12: 167.

Ong

Stafford

McLaughlin

, et al. Global, regional, and national burden of diabetes from 1990 to 2021, with projections of prevalence to 2050: a systematic analysis for the Global Burden of Disease Study 2021. Lancet 2023; 402: 203–234.

CDC. Diabetes Statistics, https://diabetesresearch.org/diabetes-statistics/ (2022).

10.

Munshi

. Cognitive dysfunction in older adults with diabetes: what a clinician needs to know. Diabetes Care 2017; 40: 461–467.

11.

Sato

Morishita

. Brain alterations and clinical symptoms of dementia in diabetes: aβ/tau-dependent and independent mechanisms. Front Endocrinol (Lausanne) 2014; 5: 143.

12.

Cao

Yang

, et al. The relationship between diabetes and the dementia risk: a meta-analysis. Diabetol Metab Syndr 2024; 16: 101.

13.

Monahan

Heath

Santos

, et al. Comorbidities associated with symptoms of subjective cognitive decline in individuals aged 45-64. J Osteopath Med 2024; 124: 277–283.

14.

Zilliox

Chadrasekaran

Kwan

, et al. Diabetes and cognitive impairment. Curr Diab Rep 2016; 16: 87.

15.

Kan

Wang

, et al. A review of type 2 diabetes mellitus and cognitive impairment. Front Endocrinol (Lausanne) 2025; 16: 1624472.

16.

Wen

, et al. Diabetes-related cognitive impairment: mechanisms, symptoms, and treatments. Open Med (Wars) 2025; 20: 20241091.

17.

Zou

Fan

, et al. Insulin resistance-related features are associated with cognitive decline: a cross-sectional study in adult patients with type 1 diabetes. Diabetol Metab Syndr 2024; 16: 13.

18.

Taylor

Bouldin

Greenlund

, et al. Comorbid chronic conditions among older adults with subjective cognitive decline, United States, 2015–2017. Innov Aging 2020; 4: igz045.

19.

Kasprzak

Kaczmarek-Majer

RzeźNiczak

, et al. Cognitive impairment in cardiovascular patients after myocardial infarction: prospective clinical study. J Clin Med 2023; 12: 4954.

20.

Dove

Shang

, et al. The impact of diabetes on cognitive impairment and its progression to dementia. Alzheimers Dement 2021; 17: 1769–1778.

21.

Olesen

KKW

Thrane

Gyldenkerne

, et al. Diabetes and coronary artery disease as risk factors for dementia. Eur J Prev Cardiol 2025; 32: 477–484.

22.

Zlatar

Muniz

Galasko

, et al. Subjective cognitive decline correlates with depression symptoms and not with concurrent objective cognition in a clinic-based sample of older adults. J Gerontol B Psychol Sci Soc Sci 2018; 73: 1198–1202.

23.

Lin

L-H

Wang

S-B

W-Q

, et al. Subjective cognitive decline symptoms and its association with socio-demographic characteristics and common chronic diseases in the southern Chinese older adults. BMC Public Health 2022; 22: 127.

24.

Lee

Sung

Choi

. The factors associated with subjective cognitive decline and cognitive function among older adults. J Adv Nurs 2020; 76: 555–565.

25.

Majoka

Schimming

. Effect of social determinants of health on cognition and risk of Alzheimer disease and related dementias. Clin Ther 2021; 43: 922–929.

26.

Peterson

Carvajal

McGuire

, et al. State inequality, socioeconomic position and subjective cognitive decline in the United States. SSM Popul Health 2019; 7: 100357.

27.

Nerobkova

Park

Shin

, et al. Marital transition and cognitive function among older adults: the Korean Longitudinal Study of Aging (2006–2020). BMC Geriatr 2022; 22: 1003.

28.

Weng

George

Jiang

, et al. Association between subjective cognitive decline and social and emotional support in US adults. Am J Alzheimers Dis Other Demen 2020; 35: 153331752092239.

29.

Muhammad

. Life course rural/urban place of residence, depressive symptoms and cognitive impairment among older adults: findings from the Longitudinal Aging Study in India. BMC Psychiatry 2023; 23: 391.

30.

Yen

Wang

Lin

, et al. The impact of alcohol consumption on cognitive impairment in patients with diabetes, hypertension, or chronic kidney disease. Front Med (Lausanne) 2022; 9: 861145.

31.

Rajczyk

Ferketich

Wing

. Relation between smoking status and subjective cognitive decline in middle age and older adults: a cross-sectional analysis of 2019 behavioral risk factor surveillance system data. J Alzheimers Dis 2023; 91: 215–223.

32.

Kim

Chang

Kim

, et al. Association between various types or statuses of smoking and subjective cognitive decline based on a community health survey of Korean adults. Front Neurol 2022; 13: 810830.

33.

Omura

Brown

McGuire

, et al. Cross-sectional association between physical activity level and subjective cognitive decline among US adults aged ≥45 years, 2015. Prev Med 2020; 141: 106279.

34.

Zhang

Cheng

, et al. The relationship between physical activity and subjective cognitive decline: evidence from the behavioral risk factor surveillance system (BRFSS). J Affect Disord 2023; 328: 108–115.

35.

Exalto

Hendriksen

HMA

Barkhof

, et al. Subjective cognitive decline and self-reported sleep problems: the SCIENCe project. Alzheimers Dement (Amst) 2022; 14: e12287.

36.

Rabin

Smart

Amariglio

. Subjective cognitive decline in preclinical Alzheimer's disease. Ann Rev Clin Psychol 2017; 13: 369–396.

37.

CDC. Behavioral Risk Factor Surveillance System (BRFSS) Cognitive Decline Module, https://www.cdc.gov/healthy-aging-data/brfss/cognitive-decline.html (2024).

38.

Taylor

Bouldin

McGuire

. Subjective Cognitive Decline Among Adults Aged ≥45 Years — United States, 2015–2016, https://www.cdc.gov/mmwr/volumes/67/wr/mm6727a1.htm (2018).

39.

Olivari

Baumgart

Taylor

, et al. Population measures of subjective cognitive decline: a means of advancing public health policy to address cognitive health. Alzheimers Dement (N Y) 2021; 7: e12142.

40.

Abdulrahman

Jansen

Hoevenaar-Blom

, et al. Diagnostic accuracy of the telephone interview for cognitive status for the detection of dementia in primary care. Ann Fam Med 2022; 20: 130–136.

41.

Schliep

Barbeau

Lynch

, et al. Overall and sex-specific risk factors for subjective cognitive decline: findings from the 2015–2018 Behavioral Risk Factor Surveillance System Survey. Biol Sex Differ 2022; 13: 16.

42.

Huang

Cuevas

Zuñiga

, et al. Predictors of subjective cognitive decline among people with diabetes: data from the behavioral risk factor surveillance system. Sci Diabetes Self Manag Care 2021; 47: 207–215.

43.

Subjective cognitive decline: chronic health conditions, 2013 BRFSS. Gerontologist 2016; 56: 250.

44.

Amin

Jesmin

. The social determinants of subjective cognitive decline among adults age 45 and over: implications for action. Innov Aging 2025; 9: igaf122.2470.

45.

Xue

Hao

, et al. High prevalence of subjective cognitive decline in older Chinese adults: a systematic review and meta-analysis. Front Public Health 2023; 11: 1277995.

46.

de Havenon

Stulberg

Littig

, et al. Socioeconomic and medical determinants of state-level subjective cognitive decline in the United States. Alzheimers Dement 2024; 20: 7567–7579.

47.

Wooten

McGuire

Olivari

, et al. Racial and ethnic differences in subjective cognitive decline — United States, 2015–2020. MMWR Morb Mortal Wkly Rep 2023; 72: 249–255.

48.

Monahan

Heath

Santos

, et al. Comorbidities associated with symptoms of subjective cognitive decline in individuals aged 45–64. J Osteopath Med 2023; 124: 277–283.

49.

Bleckwenn

Kleineidam

Wagner

, et al. Impact of coronary heart disease on cognitive decline in Alzheimer’s disease: a prospective longitudinal cohort study in primary care. Br J Gen Pract 2017; 67: e111–e117.

50.

Haring

Leng

Robinson

, et al. Cardiovascular disease and cognitive decline in postmenopausal women: results from the Women's Health Initiative Memory Study. J Am Heart Assoc 2013; 2: e000369.

51.

Rusanen

Kivipelto

Levälahti

, et al. Heart diseases and long-term risk of dementia and Alzheimer's disease: a population-based CAIDE study. J Alzheimers Dis 2014; 42: 183–191.

52.

Qiu

Fratiglioni

. A major role for cardiovascular burden in age-related cognitive decline. Nat Rev Cardiol 2015; 12: 267–277.

53.

Jochems

ACC

Muñoz Maniega

del C Valdés Hernández

, et al. Contribution of white matter hyperintensities to ventricular enlargement in older adults. Neuroimage Clin 2022; 34: 103019.

54.

Louizi

Khadhraoui

Lotz

, et al. Association of cervical artery stenosis with common cerebral microvascular lesions and coronary artery calcifications. Front Neuroimaging 2025; 4: 1559481.

55.

Richardson

Stephan

Ince

, et al. The neuropathology of vascular disease in the Medical Research Council Cognitive Function and Ageing Study (MRC CFAS). Curr Alzheimer Res 2012; 9: 687–696.

56.

Mahinrad

Sorond

Gorelick

. The role of vascular risk factors in cognitive impairment and dementia and prospects for prevention. Clin Geriatr Med 2023; 39: 123–134.

57.

Takeda

Rakugi

Morishita

. Roles of vascular risk factors in the pathogenesis of dementia. Hypertens Res 2019; 43: 162–167.

58.

Kalaria

Akinyemi

Paddick

S-M

, et al. Current perspectives on prevention of vascular cognitive impairment and promotion of vascular brain health. Expert Rev Neurotherap 2024; 24: 25–44.

59.

Tsao

Seshadri

Beiser

, et al. Relations of arterial stiffness and endothelial function to brain aging in the community. Neurology 2013; 81: 984–991.

60.

Crawford

Desmond

, et al. Long-term cognitive decline after stroke: an individual participant data meta-analysis. Stroke 2022; 53: 1318–1327.

61.

Al-Qazzaz

Ali

Ahmad

, et al. Cognitive impairment and memory dysfunction after a stroke diagnosis: a post-stroke memory assessment. Neuropsychiatr Dis Treat 2014; 10: 1677–1691.

62.

Boutros

Khazaal

Taliani

, et al. Factors associated with cognitive impairment at 3, 6, and 12 months after the first stroke among Lebanese survivors. Brain Behav 2023; 13: e2837.

63.

Rost

Brodtmann

Pase

, et al. Post-stroke cognitive impairment and dementia. Circ Res 2022; 130: 1252–1271.

64.

Elendu

Amaechi

Elendu

, et al. Stroke and cognitive impairment: understanding the connection and managing symptoms. Ann Med Surg (Lond) 2023; 85: 6057–6066.

65.

Sun

Tan

. Post-stroke cognitive impairment: epidemiology, mechanisms and management. Ann Transl Med 2014; 2: 80.

66.

Lacy

Moran

Gilsanz

, et al. Comparison of cognitive function in older adults with type 1 diabetes, type 2 diabetes, and no diabetes: results from the Study of Longevity in Diabetes (SOLID). BMJ Open Diabetes Res Care 2022; 10: e002557.

67.

Van Gemert

Wölwer

Weber

, et al. Cognitive function is impaired in patients with recently diagnosed type 2 diabetes, but not type 1 diabetes. J Diabetes Res 2018; 2018: 1470476.

68.

Rawlings

Sharrett

Schneider

, et al. Diabetes in midlife and cognitive change over 20 years: a cohort study. Ann Intern Med 2014; 161: 785–793.

69.

Strachan

. R D Lawrence Lecture 2010. The brain as a target organ in Type 2 diabetes: exploring the links with cognitive impairment and dementia. Diabet Med 2011; 28: 141–147.

70.

Sims-Robinson

Kim

Rosko

, et al.

How does diabetes accelerate Alzheimer disease pathology?

Nat Rev Neurol 2010; 6: 551–559.

71.

Koh

Rho

Lee

, et al. Association between blood glucose control and subjective cognitive decline in Korean patients with diabetes aged over 50 years. Int J Environ Res Public Health 2022; 19: 7267.

72.

Secnik

Schwertner

, et al. The association of antidiabetic medications and Mini-Mental State Examination scores in patients with diabetes and dementia. Alzheimers Res Ther 2021; 13: 197.

73.

Long

Han

Yang

, et al. Efficacy of intranasal insulin in improving cognition in mild cognitive impairment or dementia: a systematic review and meta-analysis. Front Aging Neurosci 2022; 14: 963933.

74.

Weinstein

Davis-Plourde

Conner

, et al. Association of metformin, sulfonylurea and insulin use with brain structure and function and risk of dementia and Alzheimer’s disease: pooled analysis from 5 cohorts. PLoS One 2019; 14: e0212293.

75.

Galdón Sanz-Pastor

Justel Enríquez

Sánchez Bao

, et al. Current barriers to initiating insulin therapy in individuals with type 2 diabetes. Front Endocrinol (Lausanne) 2024; 15: 1366368.

76.

Brutsaert

. Type 2 Diabetes Mellitus, https://www.msdmanuals.com/professional/endocrine-and-metabolic-disorders/diabetes-mellitus-and-hypoglycemia/type-2-diabetes-mellitus (2025).

77.

Ponce-Lopez

. Peripheral inflammation and insulin resistance: their impact on blood–brain barrier integrity and glia activation in Alzheimer’s disease. Int J Mol Sci 2025; 26: 4209.

78.

Chapple

Bayliss

Woodfin

, et al. Type 3 diabetes: linking insulin resistance to cognitive decline. Diseases 2025; 13: 359.

79.

Raza

Saleem

Imran

, et al. From metabolic dysregulation to neurodegenerative pathology: the role of hyperglycemia, oxidative stress, and blood-brain barrier breakdown in T2D-driven Alzheimer’s disease. Metab Brain Dis 2025; 40: 276.

80.

Schliep

Barbeau

Lynch

, et al. Overall and sex-specific risk factors for subjective cognitive decline: findings from the 2015-2018 Behavioral Risk Factor Surveillance System Survey. Biol Sex Differ 2022; 13: 16.

81.

Mackin

Jin

Burns

, et al. Association of major depressive disorder with remotely administered measures of cognition and subjective report of cognitive difficulties across the adult age spectrum. J Affect Disord 2023; 326: 198–205.

82.

Yulug

Yalcinkaya

Safa

, et al. Subjective cognitive decline in major depressive patients is associated with altered entropy and connectivity changes of temporal and insular region. Transl Psychiatry 2025; 15: 335.

83.

Chow

Verdonschot

McEvoy

, et al. Associations between depression and cognition, mild cognitive impairment and dementia in persons with diabetes mellitus: a systematic review and meta-analysis. Diabetes Res Clin Pract 2022; 185: 109227.

84.

Sullivan

Katon

Lovato

, et al. Association of depression with accelerated cognitive decline among patients with type 2 diabetes in the ACCORD-MIND trial. JAMA Psychiatry 2013; 70: 1041–1047.

85.

Kang

. Predictors of post-stroke cognition among geriatric patients: the role of demographics, pre-stroke cognition, and trajectories of depression. Front Psychol 2021; 12: 717817.

86.

Shi

Lenze

Mohr

, et al. Post-stroke depressive symptoms and cognitive performances: a network analysis. Arch Phys Med Rehabil 2024; 105: 892–900.

87.

Zhong

Ouyang

, et al. Cardiovascular diseases and related risk factors accelerated cognitive deterioration in patients with late-life depression: a one-year prospective study. Int Psychogeriatr 2019; 31: 1483–1489.

88.

CDC. Chronic Diseases and Cognitive Decline — A Public Health Issue, https://www.cdc.gov/aging/publications/chronic-diseases-brief.html (2023).

89.

Kang

Whelan

Malvaso

. Understanding cognitive deficits in people with arthritis. Healthcare 2023; 11: 1337.

90.

Małdyk

. Ortopedia I Traumatologia. 2008.

91.

Bernhoff

Jespersen

Dyhre-Petersen

, et al. Cognitive impairment is common in patients with severe astma that are commenced on a biological treatment. Eur Respir J 2022; 60: 4644.

92.

Nair

Van Hulle

Bendlin

, et al. Impact of asthma on the brain: evidence from diffusion MRI. CSF biomarkers and cognitive decline. Brain Commun 2023; 5: fcad180.

93.

Kelly

Rothwell

. Disentangling the relationship between chronic kidney disease and cognitive disorders. Front Neurol 2022; 13: 830064.

94.

Wang

Heizhati

Wang

, et al. Poor sleep quality is negatively associated with low cognitive performance in general population independent of self-reported sleep disordered breathing. BMC Public Health 2022; 22: 3.

95.

Kim

Ahn

Min

, et al. Association between sleep quality and subjective cognitive decline: evidence from a community health survey. Sleep Med 2021; 83: 123–131.

96.

Buongiorno

Granell

Caruana

, et al. Impairments in sleep and brain molecular clearance in people with cognitive deterioration and biological evidence of AD: a report of four cases. BMC Neurol 2023; 23: 417.

97.

Hamel

Mary

Rauchs

. Sleep and memory consolidation in aging: a neuroimaging perspective. Rev Neurol (Paris) 2023; 179: 658–666.

98.

Chen

Liu

, et al. Effects of sleep on the glymphatic functioning and multimodal human brain network affecting memory in older adults. Mol Psychiatry 2025; 30: 1717–1729.

99.

Liu

Zhang

Burgard

, et al. Marital status and cognitive impairment in the United States: evidence from the National Health and Aging Trends Study. Ann Epidemiol 2019; 38: 28–34.e22.

100.

Zhang

Zheng

. The relationship between marital status and cognitive impairment in Chinese older adults: the multiple mediating effects of social support and depression. BMC Geriatr 2024; 24: 367.

101.

Mueller

Zucchetto

Siedlecki

. The relationship between social support and subjective cognitive functioning across adulthood. Int J Aging Hum Dev 2023; 96: 174–200.

102.

Weng

George

Jiang

, et al. Association between subjective cognitive decline and social and emotional support in US adults. Am J Alzheimers Dis Other Demen 2020; 35: 1533317520922392.

103.

Costa-Cordella

Arevalo-Romero

Parada

, et al. Social support and cognition: a systematic review. Front Psychol 2021; 12: 637060.

104.

Vance

Bail

Enah

, et al. The impact of employment on cognition and cognitive reserve: implications across diseases and aging. Nursing (Auckl) 2016; 6: 61–71.

105.

Coleman

Roessler

MEH

Peng

, et al. Social enrichment on the job: complex work with people improves episodic memory, promotes brain reserve, and reduces the risk of dementia. Alzheimers Dement 2023; 19: 2655–2665.

106.

Soh

Eng

Mayeda

, et al. Association of primary lifetime occupational cognitive complexity and cognitive decline in a diverse cohort: results from the KHANDLE study. Alzheimers Dement 2023; 19: 3926–3935.

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.

0.06 MB

0.00 MB