Body Mass Index and Mild Cognitive Impairment Among Middle-Aged and Older Adults from Low- and Middle-Income Countries

The survey

Data from the Study on Global Ageing and Adult Health (SAGE) were analyzed. These data are publicly available through: http://www.who.int/healthinfo/sage/en/. This survey was undertaken in China, Ghana, India, Mexico, Russia, and South Africa between 2007 and 2010. These countries broadly represent different geographical locations and levels of socio-economic and demographic transition. Based on the World Bank classification at the time of the survey, Ghana was the only low-income country, and China and India were lower middle-income countries although China became an upper middle-income country in 2010. The remaining countries were upper middle-income countries.

Details of the survey methodology have been published elsewhere [18]. Briefly, in order to obtain nationally representative samples, a multistage clustered sampling design method was used. The sample consisted of adults aged ≥18 years with oversampling of those aged ≥50 years. Trained interviewers conducted face-to-face interviews using a standard questionnaire. Standard translation procedures were undertaken to ensure comparability between countries. The survey response rates were: China 93%; Ghana 81%; India 68%; Mexico 53%; Russia 83%; and South Africa 75%. Sampling weights were constructed to adjust for the population structure as reported by the United Nations Statistical Division [19]. Ethical approval was obtained from the WHO Ethical Review Committee and local ethics research review boards. Written informed consent was obtained from all participants.

Mild cognitive impairment

MCI was ascertained based on the recommendations of the National Institute on Aging-Alzheimer’s Association [20]. We applied the identical algorithms used in previous SAGE publications using the same survey questions to identify MCI [21, 22]. Briefly, individuals fulfilling all of the following conditions were considered to have MCI:

Concern about a change in cognition: Individuals who replied ‘bad’ or ‘very bad’ to the question “How would you best describe your memory at present?” and/or those who answered ‘worse’ to the question “Compared to 12 months ago, would you say your memory is now better, the same or worse than it was then?” were considered to have this condition.

Body mass index

Height was measured using a stadiometer, and weight was measured using an electronic weighing scale that was periodically calibrated. Weight and height measurements were taken with participants wearing only one layer of clothes and with their shoes off; height was recorded to the nearest 0.1 cm with a stadiometer, while weight was measured to the nearest 0.1 kg with a weighting scale. BMI was calculated as weight in kilograms divided by height in meters squared based on measured weight and height and classified as underweight (BMI < 18.5 kg/m²), normal weight (BMI 18.5–24.9 kg/m²), overweight (BMI 25.0–29.9 kg/m²), and obesity (BMI ≥30 kg/m²) in line with WHO guidelines [26].

Control variables

The control variables were selected based on past literature [27] and included age, sex, wealth quintiles based on income, years of education received, physical activity, alcohol use, smoking (never, current, former), depression, anxiety, sleep problems, diabetes, hypertension, and stroke. Levels of physical activity were assessed with the Global Physical Activity Questionnaire and were classified as low, moderate, and high based on conventional cut-offs [28]. Consumers of at least four (females) or five drinks (males) of any alcoholic beverage per day on at least one day in the past week were considered to be ‘heavy’ drinkers. Those who had ever consumed alcohol but were not heavy drinkers were categorized as ‘non-heavy’ drinkers [29]. Questions based on the World Mental Health Survey version of the Composite International Diagnostic Interview [30] were used for the endorsement of past 12-month DSM-IV depression [31]. Those who claimed to have severe/extreme problems with worry or anxiety in the past 30 days were considered to have anxiety [32]. Sleep problems were assessed by the question “Overall in the last 30 days, how much of a problem did you have with sleeping, such as falling asleep, waking up frequently during the night or waking up too early in the morning?” and those answering severe or extreme were considered to have sleep problems [33]. Stroke and diabetes were based solely on self-reported lifetime diagnosis. Hypertension was defined as having at least one of: systolic blood pressure ≥140 mmHg; diastolic blood pressure ≥90 mmHg; or self-reported diagnosis.

Statistical analysis

The statistical analysis was performed with Stata 14.1 (Stata Corp LP, College station, Texas). The analysis was restricted to those aged ≥50 years. The middle-aged were also included in our study as from the point of prevention of dementia, intervening in middle-age is now considered important [34], especially that cognitive dysfunction can manifest up to 10 years before a dementia diagnosis [35]. The analysis was stratified by age (i.e., middle-aged 50–64 years and older adults ≥65 years) as a previous study showed that the association between BMI and MCI may differ between these age groups [27]. Multivariable logistic regression analysis was conducted to assess the association between BMI (independent variable with normal weight as the reference category) and MCI (dependent variable) using country-wise samples while simultaneously adjusting for age, sex, wealth, education, physical activity, alcohol use, smoking, depression, anxiety, sleep problems, diabetes, hypertension, and stroke. The Higgins’s I² based on estimates from each country was also calculated in order to assess between-country heterogeneity. The Higgins’s I² represents the degree of heterogeneity that is not explained by sampling error with a value of < 40% often considered as negligible and 40–60% as moderate heterogeneity [36]. A pooled estimate was obtained by random-effects meta-analysis.

The sample weighting and the complex study design were taken into account in the analyses. Results from the regression analyses are presented as odds ratios (ORs) with 95% confidence intervals (CIs). The level of statistical significance was set at p < 0.05.

Main findings

Results from this large multi-country sample of middle-aged and older age adults showed that in middle-aged adults, underweight, overweight, and obesity are significantly associated with higher odds for MCI when compared to normal weight. In contrast, in older adults, underweight and overweight were associated with significantly lower odds for MCI. Although obesity was associated with 0.74 (95% CI = 0.49–1.14) times lower odds for MCI in older adults, this was not statistically significant. The level of between-country heterogeneity in these associations was in general low.

Interpretation of the findings

One meta-analysis including longitudinal data found that overweight and obesity in middle-age was associated with a non-significant 1.10 (95% CI =0.99–1.22) and a significant 1.41 (95% CI = 1.20–1.60) times higher risk for dementia onset, respectively. In older adults, overweight was non-sig-nificantly (RR = 0.88; 95% CI = 0.74–1.02) and obesity significantly (RR = 0.83; 95% CI = 0.74–0.94) associated with lower odds for dementia onset [5]. Our results are somewhat similar to the results of this meta-analysis and add to this previous literature by suggesting that weight reduction in middle-aged adults who are overweight may potentially lead to a reduced dementia risk in LMICs. In addition, our finding that underweight was associated with higher odds for MCI in middle-aged adults supports one previous longitudinal study that identified that underweight in mid-life is positively associated with dementia [37]. More research is needed to test whether addressing overweight and obesity may lead to decreased risk for future MCI/dementia onset, and whether maintenance of normal weight in middle-aged adults may be particularly important in LMICs.

The finding that overweight/obesity was associated with higher odds for MCI in the middle-aged may be explained by mechanisms that cause brain volume reduction such as inflammation, cardiovascular risk factors, and low physical activity [7]. As for the lower odds for MCI in older individuals with overweight observed in our study, there are several hypotheses that may explain this. First, overweight that occurs in older adulthood (following periods of relatively normal weight throughout middle-age) may impact cognition less than being overweight starting in middle age or earlier in the life course [38]. Second, older adults who have a higher BMI are often less likely to have frailty or other comorbidities, and this may have a protective effect on cognitive impairment [39].

The higher odds for MCI among underweight middle-aged individuals may be explained by several factors. First, underweight in middle-age may be associated with eating disorders, and these have been found to affect memory function. For example, anorexia nervosa is associated with disruption of functioning of the visuo-spatial sketch pad and central executive components of working memory [40]. Next, underweight may be associated with undernutrition, which is associated with cognitive impairment via micronutrient and macronutrient deficiency [41, 42]. Finally, underweight may be a sign of the presence of other comorbidities or frailty, which can increase risk for cognitive decline, but were not assessed in the present study [43]. As for the finding that underweight was associated with lower odds for MCI in older adults, mechanistic and epidemiologic evidence suggests that MCI or dementia may cause involuntary weight loss well before its clinical onset [37], and low BMI may spuriously appear to be protective against MCI or dementia.

Public health implications and areas for future research

Findings from the present study suggest that maintenance of normal weight only in mid-life may possibly lead to lower MCI and dementia risk in LMICs. In the context of LMICs and to achieve weight reduction in mid-life for people with overweight/obesity, lifestyle interventions that includes both diet and physical activity components may be most effective [44]. Moreover, action planning and self-monitoring of outcomes of behavior have been found to achieve best results, and thus any lifestyle intervention to achieve weight loss in this setting should consider embedding such techniques [45]. The Exercise is Medicine Global Initiative is designed to support health care professionals in prescribing exercise for patients by training providers to assess patient physical activity levels, imparting behavioral counseling to increase activity using change models, and referring patients to resources to facilitate physical activity [46]. Importantly, this initiative already exists in LMICs, and evaluation of its impact will provide important data on how to adapt and scale up this program to encourage healthy activity levels and weight status [47].

Strength and limitations

The large sample of middle-aged and older adults across six LMICs is a clear strength of the present study. However, findings from this study must be interpreted in light of the study limitations. First, the study was cross-sectional in nature, and therefore, temporal associations or causality cannot be established. Clearly, future longitudinal and intervention studies are necessary to understand the mechanisms that underlie the associations observed and to clarify the effect of weight modification on future risk of MCI or dementia.

Second, we did not have information on weight trajectories over the life course, despite the fact that this has also been reported to be important in determining risk for MCI/dementia. Third, participants with mild forms of dementia could have been included in our study sample as the study did not include a clinical assessment of dementia. Fourth, we did not have information on nutritional factors, while the data on physical activity used in our study was only on current levels of physical activity. Physical activity during the life course and nutritional factors are known to influence BMI, and also risk for MCI or dementia [48, 49] and thus, residual confounding is possible. Fifth, the use of BMI as an exposure may be a further limitation as BMI does not measure adiposity (although heavily correlated) or fat free mass [50]. Indeed, this may lead to some data distortion in older adults as more fat free muscle mass is associated with less cognitive issues [51]. Finally, the countries included in the study were not randomly selected. Thus, the results are not generalizable to all LMICs or high-income countries

Conclusions

In this large sample of middle-aged and older adults from six LMICs, it was found that underweight, overweight, and obesity are significantly associated with higher odds for MCI in the middle-aged, but that underweight and overweight are associated with significantly lower odds for MCI in older adults. These findings add to the current debate on whether weight modification has a role in the prevention of dementia.