Occurrence and Risk Factors for Cognitive Decline Shared by Couples: A Systematic Review and Meta-Analysis

Search strategy

We conducted an extensive search of electronic databases, including CNKI, Wanfang, VIP, PubMed, Web of Science, CINAHL, the Cochrane Library, and Embase, spanning from the inception of these databases up until 20 October 2023. Our search strategy utilized subject headings and keywords, such as ‘cognitive impairment’, ‘cognition disorders’, ‘mild cognitive impairment’, ‘spouse’, ‘couple’, ‘partner’, ‘wife’, ‘husband’, ‘married’, and ‘risk factors’, ‘factor’, ‘influence factors’, ‘correlates’. The search was conducted by two investigators, and the detailed strategy is shown in the Supplementary Material. Additionally, we performed a manual examination of the reference lists of identified studies to identify any additional potentially relevant publications through citation tracking. Data management was facilitated using Endnote, and the screening process involved independent evaluation of titles and/or abstracts of studies retrieved from the systematic literature search and other sources by two reviewers to identify potentially eligible studies.

Inclusion and exclusion criteria

Study eligibility was meticulously determined by two independent researchers, with any discrepancies resolved through consensus by a third reviewer. The inclusion criteria for this study were as follows: 1) The participants consisted of long-standing married couples residing together; 2) The primary outcome measure focused on the cognitive function of the couples; 3) The control group comprised spouses exhibiting normal cognitive function. Conversely, the exclusion criteria encompassed the following aspects: 1) Inability to access the full text of the study; 2) Repetition of literature, reviews, or case reports; 3) Whether the caregiver was the spouse of the person with cognitive impairment was not reported; 4) Insufficient or unobtainable data within the original research.

Screening and data extraction

Data extraction was undertaken independently by two authors (MHY and LJR). Discrepancies, if any, were meticulously identified and effectively resolved through comprehensive discussions and ultimately by a third reviewer (ZJX). Studies were reviewed initially on the basis of title and abstract, and then full text were reviewed for those deemed relevant studies. Data were abstracted in duplicate (by MHY and LJR) from ultimately included studies. For the purpose of this review, cognitive impairment was defined as mild cognitive impairment or dementia. The extracted information comprised a comprehensive array of key elements, including:

Thorough documentation of the study particulars, encompassing essential details such as the first author, year of publication, country of origin, source population, age restrictions, study design, and analytical sample size.

Quality assessment

The quality assessment of the included studies was conducted meticulously by two authors (MHY and LJR), with any discrepancies being discussed and resolved through consultation with a third reviewer (ZJX). Case-control or cohort studies were assessed using the widely recognized Newcastle– Ottawa Scale (NOS) [15], which incorporates crucial aspects such as study population selection, comparability, and outcome assessment. A maximum score of 9 points can be achieved, indicating high quality, with a total score of ≥6 considered indicative of high quality. For cross-sectional studies, the bias risk evaluation standard recommended by the Agency for Healthcare Research and Quality (AHRQ) [14, 16] was utilized. This standard comprises 11 scoring items, with responses categorized as “yes,” “no,” or “unclear” based on the content of each item. The results were compiled and presented using the Cochrane assessment tool, with responses replaced by “low risk,” “high risk,” or “unclear risk.” This rigorous quality assessment process ensures the reliability and credibility of the included studies in this review.

Data synthesis and analysis

The data synthesis and analysis were performed using Stata 15.1 software, employing robust statistical methods to examine the cognitive impairment in couples. Cohen’s d value and its corresponding confidence interval were utilized to estimate the differences in cognitive function between the two groups, one with cognitive impairment and the other without. To represent the effect size of the risk of cognitive impairment in these groups, odds ratios (OR) with their respective 95% confidence intervals were employed. Statistical heterogeneity was assessed using I square (I ²) values, where an I ² greater than 50% indicated significant heterogeneity, leading to the utilization of a random effects model for calculating the pooled effect size. Conversely, a fixed effects model was employed when the I ² value was below 50%. In cases where data extraction or synthesis was not feasible from the original articles or when risk factors were reported in only one study, a descriptive approach was employed to present the results. Descriptive analysis was employed due to the limited number of studies available for each risk factor and the considerable differences in influencing factors across the studies. Publication bias was assessed using Egger’s test, providing insights into the potential bias in the reported findings.

Literature search process and results

The literature search process yielded a total of 821 relevant articles from each database. After removing 70 duplicate articles, the remaining articles underwent a thorough screening process, which involved reviewing the titles, abstracts, and full texts. Following this screening process, 11 studies met the inclusion criteria and were selected for further analysis (see Fig. 1).

OPEN IN VIEWER

Fig. 1

Selection strategy and results.

Literature basic characteristics and quality evaluation

The included studies in this review consisted of six studies that examined the comparison of cognitive scores between elderly individuals whose spouses had dementia (DCG) and those whose spouses did not have dementia (NCG). These six studies encompassed a total of 687 cases in the DCG group and 1628 cases in the NCG group. Additionally, three studies focused on assessing cognitive function in couples, involving a total of 5,071 couples. The overall sample size across all included studies comprised 12,457 elderly individuals. The study locations encompassed the United States, Netherlands, and Korea. One study specifically explored the perceptions of patients with mild cognitive impairment and their spouses regarding the content validity, practicality, and acceptability of the Enhanced Meaningful Activities of Daily (DEMA) program. Another evaluation article was included as it examined other potential risk factors for cognitive decline shared by couples. The reported outcomes of cognitive impairment in all studies primarily included mild cognitive impairment and dementia.

In our included studies, there were variations in the assessment of cognitive functioning as the primary outcome measure. These assessments included clinical examinations conducted by geriatric neurologists and/or neuropsychologists, telephone interviews, and questionnaires. Additionally, different assessment tools were chosen, but they generally covered various aspects of cognitive functioning, such as memory, attention, orientation, and language executive functions. During clinical examinations, geriatric neurologists and/or neuropsychologists may employ a range of cognitive tests to evaluate individuals’ cognitive status. These tests may include assessments of memory, such as the ability to recall words or images; attention tests, such as sequence recall or abstract reasoning; orientation tests, involving awareness of time, place, and current events; and language executive function tests, such as vocabulary and grammar comprehension. Telephone interviews and questionnaires provide a more convenient way to assess individuals’ cognitive functioning. By asking questions about memory, attention, language abilities, and other aspects of daily life, researchers can gain preliminary insights into individuals’ cognitive functioning. Although there may be variations in assessment methods and tools, the overall aim of the selected evaluation tools was to comprehensively assess different aspects of cognitive functioning. This helps us gain a better understanding of the relationship between cognitive decline and risk factors. Study characteristics and spousal association effect estimates are shown in Table 1.

Among the 11 included articles, seven were cohort studies, and one was a case-control study. The quality of cohort and case-control studies was assessed using the Newcastle-Ottawa Scale (NOS) criteria [14], with one article being of moderate quality and eight articles being of high quality according to the evaluation results. The quality of the cross-sectional studies was evaluated according to the Agency for Healthcare Research and Quality (ARHQ) criteria, which indicated moderate quality. Additionally, the included studies comprised one review and one qualitative study, both of which did not undergo a formal quality evaluation.

Occurrence of shared cognitive decline in couples

Cognitive decline was more rapid in spouses of patients with cognitive impairment.

The results of the analysis of baseline data, as depicted in Table 1, revealed that spouses of patients with cognitive impairment (DCG) exhibited worse cognitive function, particularly in vocabulary performance, compared to age-matched controls. The effect sizes (Cohen’s d) ranged from 0.18 to 0.62 across the five studies, suggesting small to moderate effects. The heterogeneity test showed an I ² value of 27.2%. A fixed effects model was employed for the meta-analysis, yielding an overall effect size (Cohen’s d) of –0.35 (–0.46, –0.23), indicating a moderate effect (see Fig. 2). However, one study did not demonstrate a significant difference between groups at baseline and was therefore not included in the meta-analysis [19].

OPEN IN VIEWER

Fig. 2

Forest plot of meta-analysis of cognitive function between spouses of patients with cognitive impairment and spouses of controls.

Regarding cognitive scores after follow-up, two studies [19, 20] reported lower scores in the DCG group compared to the NCG group, with Cohen’s d values of –0.16 and –0.38, respectively. These findings indicate that DCG experiences a greater degree of cognitive decline than NCG individuals of the same age. In the study by Vitaliano et al. [19], DCG showed a decrease of 1 point (0.01 points) relative to NCG, with a statistically significant difference. Similarly, Dassel et al. [22] found that DCG exhibited a decrease of 1.77 points (0.87 points) compared to NCG.

The risk of cognitive impairment is higher in spouses with cognitive impairment

Within the scope of our analysis, we included three studies that specifically examined the prevalence of cognitive impairment in couples. The findings consistently indicated that the risk of cognitive impairment was higher in spouses with cognitive impairment (SCD+) compared to spouses without cognitive impairment (SCD-). In total, 5071 couples were evaluated across the three studies, with 857 spouses classified as SCD+ and 4214 spouses classified as SCD-. The assessment of heterogeneity among the studies revealed no significant variation (I ² = 5.6%). Consequently, a fixed effects model was employed for the meta-analysis. The results of the meta-analysis demonstrated a statistically significant difference in the prevalence of cognitive impairment between the two groups (see Fig. 3), with an odds ratio (OR) of 1.42 and a 95% confidence interval (CI) ranging from 1.15 to 1.76 (P-value = 0.001).

OPEN IN VIEWER

Fig. 3

Forest plot of cognitive impairment in spouses and risk of developing cognitive impairment.

Risk factors for cognitive decline in elderly couples

In terms of spouses acting as caregivers with cognitive impairment, several studies have highlighted that caregivers of individuals with cognitive impairment exhibit significant cognitive deficits compared to non-caregiving spouses (NCG). Apart from age and education level, caregiving status [17 –22] has been identified as a predictor of cognitive scores or changes in scores. This association may be influenced by the composite effects of distress rather than solely by depression. Two studies [20, 21] have indicated that this association is mediated by psychological complaints such as anxiety and depression, and that the subjective competence of caregivers is linked to cognitive impairment [18]. However, in Vitaliano et al.’s study [19], the association was mediated by hostile attribution and metabolic risk. Pertl et al.’s findings suggest that caregivers, particularly men, and individuals with lower incomes are more likely to experience adverse cognitive outcomes [21]. In addition to older age, lower socioeconomic status, and possessing the APOE ɛ4 allele, cognitive impairment in the spouse has been identified as an independent risk factor for cognitive impairment, with gender acting as a moderator [23].

In terms of people with cognitive impairment, the current study reveals that certain characteristics, such as the severity of dementia and presence of behavioral symptoms like hyperactivity, increase the risk of developing cognitive impairment in caregivers [18].

One study [24] found that common risk factors mediate the consistency of cognitive function in couples. Risk factors for cognitive impairment, including history of exposure to smoking, Physical inactivity, a history of head injury, and major depressive disorder (MDD) were consistent among spouses. And physical inactivity, a history of head injury, and MDD have been found to mediate the association between spousal cognitive disorders and cognitive functions, including memory and executive function [24]. A recent study [25] suggests that differences in risk factors during midlife may explain cognitive impairment in spouses, but having a cognitively impaired spouse is not necessarily associated with an increased risk of incident dementia. Further research is recommended in three areas of relevance to neurology and psychiatry, as proposed by Peter et al.: assortative mating, dyadic lifestyles, and spouse caregiving [26]. The shared risk factors between spouses provide new ideas and methods for early prevention and intervention.

Limitations

This review acknowledges several limitations that should be considered when interpreting the findings. Firstly, the search strategy and inclusion criteria were limited to English-language papers, which may have resulted in the exclusion of relevant studies in other languages. Secondly, the included studies predominantly consisted of observational case-control and cross-sectional designs, with limited longitudinal studies. This makes it challenging to establish causal relationships and draw definitive conclusions about the causal effects of the identified risk factors. Furthermore, the heterogeneity in cognitive assessment tools and diagnostic criteria across different studies posed challenges in conducting a comprehensive analysis. Only total cognitive scores were included in the meta-analysis, while specific domains of cognitive impairment such as orientation, memory, and language function were not individually assessed. Additionally, this review qualitatively describes studies related to risk factors for cognitive decline in couples, but due to insufficient data availability, a meta-analysis could not be performed.

Conclusion

This systematic review and meta-analysis provide evidence on cognitive decline occurrence and risk factors in couples. Spousal cognitive impairment is identified as an independent risk factor, with lower cognitive function scores observed in spouses with cognitive impairment. The presence of cognitive impairment in one partner increases the risk of cognitive decline in the other partner. Mediating factors include the physical and mental stress of caregiving and shared risk factors between couples. However, further research is needed to understand the impact of caregiver stress and shared risk factors and to develop targeted interventions. Therefore, it is crucial to implement early screening and assessment of cognitive function in both spouses within a family unit, enabling the timely identification of these risk factors and the implementation of appropriate intervention measures. Additionally, further research efforts should be directed towards exploring additional factors that may contribute to cognitive decline in couples, ultimately leading to the development of targeted interventions aimed at improving cognitive health within this population.