|Year : 2013 | Volume
| Issue : 3 | Page : 156-164
Effects of education and income on cognitive functioning among Indians aged 50 years and older: evidence from the Study on Global Ageing and Adult Health (SAGE) Wave 1 (2007-2010)
Department of Internal Medicine, Scott & White Healthcare/Texas A&M Health Science Center College of Medicine, Temple TX, United States of America
|Date of Web Publication||23-May-2017|
Assistant Professor, Department of Internal Medicine, Scott&White Healthcare, Texas A&M University Health Science Center College of Medicine, 2401 South 31st Street, 501 Brindley Circles (#540), Temple, Texas 76508
United States of America
Objectives: Relatively little is known about socioeconomic predictors of cognitive health among middle-aged and elderly Indians. The primary objective of this study was to examine the extent to which education and income influence cognitive functioning after adjusting for demographic characteristics, health risk factors and transgenerational factors such as parental education. The study also examined gender disparities in cognitive functioning across geographic regions in India.
Methods: Using cross-sectional data from the World Health Organization Study on Global Ageing and Adult Health (SAGE) Wave 1 (2007–2010) in a national sample of adults aged 50 years or older, a generalized linear model was used to examine the impacts of education and per-capita income on overall cognitive functioning. The generalized estimating equation approach was utilized to quantify these impacts on respondents’ overall cognitive performance score. This technique accounted for any correlation of responses of individuals within the same household.
Results: Respondents with primary or secondary education and those with education above secondary level scored 3.8 and 6 points (P < 0.001) respectively more than respondents who had no formal education. In a similar vein, individuals in higher per-capita income quartiles scored 0.4,1.0 and 1.8 (P < 0.001) more than respondents in the lowest income quartile. Although respondents in northern states scored 1.8 points higher than those from other geographic locations (P < 0.001), females in northern states had the worst cognitive performance (1.9 points lower) compared with females in other Indian states. In addition, early and adult life characteristics such as parental education, physical activity and a history of depression were found to be significant predictors of overall cognitive functioning.
Conclusion: Education and income play important roles in influencing overall cognitive performance among middle-aged and elderly Indians. In addition, cognitive performance scores varied across geographic regions, and female disadvantage was observed in northern Indian states. Policies directed towards greater educational opportunities, particularly for women in northern Indian states, or promotion of physical activity programmes, have potential to improve cognitive performance and enhance cognitive health among middle-aged and older adults in India.
Keywords: education and cognitive status, older Indians, parental education, geographicdisadvantage, genderdisparities, lifestyle factors.
|How to cite this article:|
Basu R. Effects of education and income on cognitive functioning among Indians aged 50 years and older: evidence from the Study on Global Ageing and Adult Health (SAGE) Wave 1 (2007-2010). WHO South-East Asia J Public Health 2013;2:156-64
|How to cite this URL:|
Basu R. Effects of education and income on cognitive functioning among Indians aged 50 years and older: evidence from the Study on Global Ageing and Adult Health (SAGE) Wave 1 (2007-2010). WHO South-East Asia J Public Health [serial online] 2013 [cited 2019 Sep 17];2:156-64. Available from: http://www.who-seajph.org/text.asp?2013/2/3/156/206762
| Introduction|| |
Population ageing is becoming a global issue and will have a major impact on health-care systems worldwide. According to the World Health Organization (WHO), three quarters of the estimated 1.2 billion people aged 60 years or older will be living in developing countries by the year 2025., With the ageing of the world’s population, age-related diseases, including dementia, are increasingly prevalent in the developing world and are affecting more than 50% of the global elderly population. India is one of the developing countries experiencing a rapid demographic and epidemiologic transition. The proportion of India’s population aged 60 or older is projected to increase from 7.7% in 2010 to 18.3% in 2050, according to the United Nations Department of Economic and Social Affairs Population Division. Therefore, the risk of dementia should also increase.,, Poor cognitive status is a risk factor for dementia, yet we have limited knowledge about factors that trigger poor cognitive functioning among older adults in India. Most studies on cognitive ageing in India have included small community-based samples, which raises the question of generalizability of the findings. Therefore, there is a need for larger nationally representative population-based studies for a better understating of the predictors of cognitive functioning among older Indians. Using the WHO Study on Global Ageing and Adult Health (SAGE) data, this study aims to address this gap.
Research in developed countries has emphasized the role of education and wealth on cognitive health among older adults.,,, More importantly, these studies consistently find strong positive relationships between higher education and improved cognitive functioning. These associations were observed regardless of the outcome measures of cognitive functioning. Other socioeconomic factors such as income and wealth are also found to be strong predictors of cognitive functioning among older adults in developed countries.,,, This study sought to examine the cognitive health of older Indians using cross-sectional data from the SAGE Wave 1 (2007–2010).
The objectives of the present study are to investigate (a) if a higher level of education and per-capita household income would each be associated with a greater overall cognitive score, adjusting for other socioeconomic factors, early life factors, health risk behaviours, and chronic ill-health; (b) whether or not there are gender disparities in cognitive functioning; and (c) how geographical differences affect cognitive functioning in India.
| Methods|| |
The analysis was performed using data from Wave 1 of SAGE. The SAGE was implemented as a face-to-face household survey using a stratified multistage cluster design to allow each household and respondent to be assigned a known nonzero probability of selection. SAGE was designed to be a longitudinal panel survey to collect information on the health and well-being of middle-aged and older adults and on the ageing process. SAGE collects data on individuals aged 50 years or older from nationally representative samples of six countries: China, Ghana, India, Mexico, Russia Federation and South Africa. The SAGE survey instruments were adapted from those used by WHO and other surveys on ageing, including the United States Health and Retirement Survey Q1RS) and the United Kingdom English Longitudinal Study of Ageing (ELSA). The description of the SAGE survey and a detailed discussion are available elsewhere. The current analysis focused on the SAGE Wave 1 survey from India that was conducted during 2007–2010. The survey was fielded in six states (Assam, Karnataka, Maharashtra, Rajasthan, Uttar Pradesh and West Bengal) selected to capture geographic variations as well as socioeconomic and cultural differences across India. The primary sampling units (PSU) were stratified across urban and rural areas in each state to capture socioeconomic differences and lifestyle behaviours. SAGE randomly sampled 10 424 households from the stratified PSUs and collected data primarily on individuals aged 50 years or older within the household.
The SAGE multi-country survey includes questions about demographic, economic, social, behavioural as well as physical and cognitive health. Face-to-face interviews were conducted in India (2007–2008). The sample in India was representative of subnational and substate levels for the six selected states. The survey consisted of two sets of questionnaires: the household questionnaire and the individual questionnaire. The household questionnaire collected information on household income, expenditure, consumption, assets and debts. It was considered that any household member over the age of 18 would be able to provide this information. The response rate for the household questionnaire was 77%. The individual questionnaire included only age-eligible (aged 50 or above) household members and their spouses. The individual questionnaire also included proxy respondents with the response rate of 88%. A total of 7750 individuals aged 50 years or older from 4473 unique households were included in the target sample. The individual sample was then further reduced by excluding the proxy as well as missing respondents (n=264). The final study sample included 6786 adults (3423 males, 3363 females) aged 50 or above.
Outcome variable: measures of cognitive function
The SAGE Wave 1 survey administered tests to measure cognitive performance against objective indicators of various aspects of cognition. The cognitive tests described below were intended to measure the cognitive domains most affected by impairment, i.e. immediate memory, concentration and attention.
Episodic memory was tested by the immediate and delayed verbal recall method where respondents were read 10 words which were repeated three times to saturate the learning curve. To test delayed recall, respondents were asked to recall the words after some time. Episodic memory as measured by verbal recall was defined as the average number of words recalled, and assessed learning capacity, memory storage and memory retrieval.,
Global cognitive function
Two tests were administered to assess global cognitive function: digit span subtests and a verbal fluency test. The digit span test was utilized to measure working memory. A series of number sequences was presented and the respondent was asked to reproduce the exact same sequence. In the second portion, the respondent was asked to repeat the sequence backwards. Following a correct recall, longer sequences were given until failure. The maximum score for the forward digit count was 9 with a range of 0–9; the score for the backward digit count ranged from 0–8 and a summary score, created by adding forward and backward counting scores, ranged from 0–17. Respondents were given a maximum of two trials; if they failed to specify correctly the number sequence after two trials, the interviewer stopped asking questions regarding digit counting and the respondent received a score of zero.
The verbal fluency test measured respondents’ ability to retrieve information from semantic memory. This was a one-minute assessment in which respondents were asked to name as many animals as they could. The verbal fluency score was defined by the number of correctly named animals. Repeated names were not counted.
The overall cognitive score was obtained by adding scores of verbal recall, digit counting and verbal fluency tests, converted to a scale of 0 (worst cognition) to 100 (best cognition).
Main explanatory variables
The primary independent covariates of interest were years of completed education and per-capita household income. Education was measured using the following categories: no formal education, primary or completed secondary school, and completed high school or above. Per-capita household income was used as a measure of economic status. In the SAGE survey, household income information could be provided by any age-eligible respondent (i.e. 18 years or older) and this information was included only in the household questionnaire, not in the individual questionnaire. For the purpose of the current study, per-capita income was constructed by dividing the total household income by the number of individual members who depended on that income.
Additional explanatory variables
Chronic ill-health and health behaviour
Several measures of chronic ill-health and health behaviours were considered in the analysis. These measures included:
(1) indicators of under weight based on body mass index (BMI) that have been consistently used in developing countries;(2) physical activity, tobacco and alcohol use, and consumption of fruit andvegetables; (3) self-reported measures of subjective health, cardiovascular status based on self-reported medical conditions and biomarker assessments to capture variances in health among respondents.
Respondents with BMI < 18.5 were classified as underweight; 18.5–24.9 as normal weight; 25–29.9 as overweight; and BMI of 30 or over as obese.
Self-reported subjective health was based on the responses of very good, good, moderate, bad and very bad health status. A binary variable indicating self-reported good health status (very good and good health) was included in the analysis. A binary variable indicating self-reported diagnosis by a health professional for heart disease, stroke, high blood pressure and diabetes was created to measure cardiovascular health status. These self-reported measures were based on the following question: “Have you ever been told by a health professional that you had [a stroke, angina (heart disease), diabetes, or high blood pressure]?” Whenever possible, biomarker information was supplemented to define the presence or absence of certain chronic conditions in the sample. For example, systolic and diastolic blood pressure readings were taken three times for each respondent in addition to the self-reported information on hypertension diagnosis. Based on the biomarker measure, if respondents had average systolic readings over 140 mmHg and average diastolic readings over 90 mmHg, they were considered to have hypertension (or high blood pressure). In the regression model, a binary variable indicating the presence of any of the self-reported conditions (stroke, angina, diabetes) or a biomarker measure of hypertension, was defined as poor cardiovascular status.
Health behaviour was measured by tobacco and alcohol use, physical activity and diet. The current smoking status was based on the self-report of smoking behaviour. The majority of respondents who reported ever-smoking also reported smoking currently and thus limited the inclusion of past smoking status in the analysis. Smoking activity included the use of tobacco, cigarettes, chewing tobacco, pipes, or cigars. Alcohol drinking was measured by a binary variable indicating whether the respondent currently drank alcohol.
Physical activity was measured using the General Practice Physical Activity Questionnaire. The instrument captures information on physical activity on three domains: activity at work, travel to and from places, and recreation. For the purpose of this study, physical activity at work and for recreation were considered since the questionnaire assessed vigorous and moderate activities in these two areas. The number of days in a week on different activities as well as time spent on each activity were recorded. To classify the level of physical activity, total minutes of activity and activity volume weighted by energy requirement in metabolic equivalents (MET) for each type of activity were calculated. Total MET activity volume per week was calculated by multiplying the time spent on each activity during the week by the MET values of each level of activity. MET values of 4 for moderate physical activity and 8 for vigorous physical activity were used to calculate MET energy consumption.
Total physical activity was calculated by the sum of total moderate and vigorous activities at work and for recreation per week. Respondents were categorized into low, medium and high levels of physical activity based on the number of days and total physical activity MET minutes per week. Respondents were classified into the high physical activity group if vigorous intensity activity on > 3 days achieved 1500 MET minutes per week; or if ≥ 7 days of any combination of moderate or vigorous intensity activities achieved ≥ 3000 MET minutes per week. Moderate intensity was defined as not meeting the criteria for ‘high’ intensity, but meeting any of the following: ≥ 3 days of vigorous intensity activity af ≥ 20 minutes per day; ≥ 5 days of moderate intensity activity of≥ 30 minutes per day; or ≥ 5 days of any combination of moderate or vigorous intensity activities, achieving > 600 MET minutes per week. Finally, respondents were categorized into the low physical activity group if they did not meet criteria for high or moderate activity levels.
Healthy diet was measured by determining the number of servings of fruit or vegetables in a typical day that the respondent had eaten over the past 12 months of the survey.
Although evidence suggests that cognitive performances tend to decline after hospitalization, this was not controlled in the analysis because there was no significant bivariate relationship between overnight hospitalization and cognitive performance scores among sample members.
Respondents were grouped based on the geographic location of the states where they lived: southern (Karnataka), northern (Uttar Pradesh, Rajasthan), western, (Maharashtra) and eastern (Assam and West Bengal). To examine gender disparity across states, interactions between gender and state variables were also included in the model since there is evidence that females in northern Indian states face socioeconomic disadvantages and experience poor health outcomes.
Early life characteristics
The SAGE survey captures the characteristics of respondents’ early life that are associated with cognitive performance in later life. Examples of these are parental education and occupation, and the area where they lived. This study used two early life conditions: parental education (both father’s and mother’s completed education) and where they lived. Due to a high proportion of missing values for the father’s occupation, this variable was not used in the analysis. While a father’s education is an indicator of a respondent’s childhood socioeconomic status (as education is associated with occupation and hence income), a mother’s education is assumed to influence health in later life through positive health-seeking behaviour. Parental education was grouped into three categories: no formal education; completed primary education, and secondary education.
The second variable that captured early life disadvantage was whether respondents lived in a rural or urban area during childhood. Due to limited access to heaIth-care services, children living in rural areas are likely to have poorer health status, which may have a long-term impact on cognitive health in later age.
Demographic and other socioeconomic characteristics
A set of demographic variables such as age, gender, caste and marital status was included in the analyses. Castes were grouped according to respondents who belonged to scheduled castes, scheduled tribes, or any other caste affiliation. Caste was measured based on respondents’ self-identification of affiliation. Scheduled castes and scheduled tribes are considered the most disadvantaged in traditional Indian society. Scheduled castes are socially segregated and economically disadvantaged by their lower status in the traditional Hindu caste hierarchy. Scheduled tribes are geographically isolated with limited economic and social interaction with the rest of the population. Other cases, known as backward classes, are less stigmatized than scheduled castes or tribes, but these individuals also are in lower socioeconomic groups due to barriers in education and earning opportunities.
In the empirical analysis, an indicator variable of whether respondents were from a scheduled caste was included. Social engagement was measured by asking respondents how often they had engaged in various social activities in the preceding 12 months (e.g. attended any group, club, society, union or organizational meeting). Response options ranged from 1 (never) to 5 (daily). For the purpose of the study, responses were recoded to 0 (never) and (4) daily. The total score for this variable was obtained by accumulating individual scores with the range from 0–35, where a higher score indicates better social engagement.
Finally, respondents’ occupation was classified into three categories: professional, sales or clerical, and manual labour, based on the International Standard Classification of Occupation code ISCO-88 used in the SAGE survey. The analysis included a binary variable to indicate if information on the respondent’s occupation was missing in the study sample. All explanatory variables except per-capita household income came from the individual questionnaire.
| Analysis|| |
Multivariate linear regression models were used to examine the impacts of education and income influencing cognitive functions among older Indians. Because of the clustered design of the sample, robust variance estimates (Huber-White sandwich estimator) are reported for the correction of standard errors to adjust for the correlation among responses within the same household. The generalized estimating equation technique was used to estimate the regression model accounting for correlation of responses within the same household as described by Liang and Zegar.,
| Results|| |
[Table 1] outlines the characteristics of the study sample by gender. The sample of 6786 Indians aged ≥ 50 years comprised 49.5% women and 51.5% men. The educational level of most participants (84%) was lower than secondary school and a large proportion of individuals in the northern region (69%) had no formal school education compared with other regions. A significantly higher proportion of females (67%) than males (33%) had no formal education, which may indicate gender disparity in accessing formal education in India. Almost three quarters (75%) lived in rural areas. On average, males had a higher cognitive performance score than females (29 versus 25) and the mean difference of overall performance was statistically significant based on one-way analysis of variance analysis. More men than women were current alcohol and tobacco users, but a higher proportion of women (62%) engaged in high and medium levels of physical activity than men (54%). The prevalence of chronic health conditions were quantitatively similar in both groups, except a higher percentage of women (19%) than men (14%) were diagnosed with hypertension, and 25% of men had depression compared with 22% of women. No difference in average per-capita income was observed between men and women in the sample.
Predictors of cognitive functioning
In the bivariate analysis, a higher overall cognitive performance score was positively associated with the following indicators: primary/secondary level education or higher, per-capita income, male gender, younger age, being married, having a professional occupation, residing in an urban area and in a northern region, high or moderate physical activity, high fruit and vegetable consumption, optimal weight, no history of depression, self-report of a good or very good subjective health status, high social engagement, and having parents with primary education.
In the multivariate analysis, higher overall cognitive functioning was positively correlatedwith primary/secondary level education or higher, per-capita income, male gender, younger age, being married, high or moderate physical activity, nohistory of depression, normal weight, high fruit and vegetable consumption, high social engagement, and having a mother with primary education.
[Table 2] presents bivariate associations of education and per-capita income on the overall cognitive performance score, and the muhivariate regression results after accounting for confounding factors that might significantly influence the relationship of education and per-capita income with cognitive function. The QIC (quasi-likelihood under the independence model criterion) statistic proposed by Pan was used to assess the ‘goodness of fit’ of all regression inodels. The QIC is similar to Akaike’s Information Criteria for likelihood-based models. Adjusting for important individual characteristics, education significantly predicted the overall cognitive performance score. Compaied with no formal education, respondents with primary or secondary education, and higher secondary or college education scoied 3.8 and 6 points higher respectively.
|Table 2: Bivariate and multivariate analyses of cognitive performance score (N=6786)|
Click here to view
Per-capita household income was positively associated with the overall cognitive performance score. For example, respondents inper-capitaincome quartiles above 75%, 50–75% and 25–50% scored 1.8, 1.0 and 0.4 respectively higher than respondents in the lowest per-capita income quartile (below 25%).
Among chronic health conditions, respondents with depression had a lower performance score (1.3 point less) than individuals with no history of depression. Regular physical activity, BMI, and fruit and vegetable consumption were found to be significantly associated with overall cognitive performance. For example, individuals engaged in a high or medium level of regulai physical exercise scored 1.6 and 1.2 points more respectively on overall cognitive functioning compared with those engaged in low physical activity. Being underweight or obese scored 0.9 and 4.4 points respectively less in cognitive functioning than those who had normal weight. Finally, higher fruit and vegetable consumption was associated with an increased points score of 0.3 in the overall cognitive performance test.
Among early life characteristics, parental education was found to be a significant predictor of overall cognitive functioning in the national sample of adults aged ≥ 50 years. For example, respondents whose father had no formal education scored 1.2 points less than those whose father had at least secondary education, and respondents whose mother had completed primary education scored 1.3 points higher in the overall cognitive functioning test.
Gender disparities and regional differences in cognitive functioning were observed. For example, individuals residing in northern regions scored 1.9 points more than individuals in other regions,while women scored 0.5 points less than men in the overall cognitive performance test. Furthermore, women in northern states performed the worst (scored 1.9 points less) compared with women in other regions.
Among other demographic characteristics, marital status and occupation were significantly correlated with overall cognitive functioning. For example, married individuals had 0.8 points higher thanthose who were never married, divorced, widowed or separated, and those engaged in a professional occupation scored 1.6 points more than respondents in other occupations. Finally, respondents engaged in many social activities performed better in overaii cognitive functioning.
| Discussion|| |
In a national sample of middle-aged and older adults, the current study found that higher education and higher per-capita income were independently associated with a greater cognitive functioning score. Consistent with other studies,,,,,”  the current article found that education had a strong and significant impact on cognitive performance in later life. After adjusting for a set of individual characteristics such as current income, it is suggested that education and socioeconomic measures are not interchangeable in regard to overall cognitive functioning. Advantages of education are that it may support brain reserves and contribute to understanding and mentally stimulating activities throughout life.
Examining correlates of cognitive functioning such as father’s education or respondents’ occupation, this study provides strong evidence of the unique contribution of education and current per-capita income towards a positive cognitive status in later life in Indians. However, the mechanism by which education contributes to cognitive health still needs further investigation. This mechanism may be the educational process, mental stimulation, a healthy lifestyle, or a combination of these factors that helps to maintain cognitive health in later life. Furthermore, the magnitude of the impact of education and per-capita income highlights the leading role that education plays in influencing cognitive function in later life.
Gender disparities in overall cognitive functioning are evident in the growing body of literature on cognitive health in developing countries, which shows that women perform worse than men on a variety of cognitive measures., This contrasts the observation in developed countries that women outperform men.,
Consistent with other studies, the current study found evidence that parental education made a significant independent contribution to respondents’ overall cognitive functioning., Modifiable lifestyle-related factors such as physical activity and fruit and vegetable consumption also contributed to overall cognitive functioning. The rate of respondents’ acknowledgement of depression indicates that poor psychological and mental health experienced by middle-aged and older Indians may sap cognitive reserves required for good memory functioning.
The current study has some limitations that merit discussion. First, the SAGE survey relied on the self-reporting of health variables such as chronic disease (depression, self-rated health and cardiovascular conditions) and lifestyle-related behaviours (tobacco or alcohol use, physical activity). These variables should be interpreted with caution. Second, information on per-capita income was obtained only from age-eligible members in the household and therefore may be subject to reporting bias. This bias could be due to the fact that household income earners might not necessarily be those who reported income in the household survey. However, given that income and asset information in the SAGE survey are available only at the household level, these variables had to be converted to the individual level to account for the number of household members for the individual-level analysis. About 35% of household members who reported household income were the main income earners; therefore, the per-capita income variable might also be subject to a reporting bias. Finally, given that the present study was based on data collected in a cross-sectional survey, causality cannot be ascribed to any of the associated factors in the study.
Despite these limitations, the current study makes a positive contribution to the emerging literature on effects of education and per-capita household income on cognitive function in older Indians. While previous findings on cognitive functioning in this age group were limited to small and geographically restricted samples,, this study expands our understanding of how education and income impact different dimensions of cognitive functioning among older Indians. It also examines the extent to which results can be compared with similar studies in high-income countries using population-based data sources such as the United States Health and Retirement study and the English Longitudinal Study on Ageing. Finally, this study takes advantage of a large nationally representative sample from economically and geographically diverse Indian states to quantify the relative importance of education and income on cognitive functioning among middle-aged and older adults in India. Further studies utilizing longitudinal data and focusing on factors over the life span could help us better understand mechanisms by which education and income influence not only current cognitive function but cognitive trajectory over time. Policies directed towards greater educational opportunities, particularly for women in northern Indian states, or wellness programmes promoting regular physical activity, may have potential to improve cognitive well-being among middle-aged and older Indians.
| Acknowledgements|| |
This study is based on SAGE data that are supported by the United States National Institute on Aging’s Division of Behavioral and Social Research through interagency agreements and research grants, and the World Health Organization’s Department of Health Statistics and Information Systems. I want to thank the anonymous reviewer for insightful comments on the earlier version that improved the clarity of the paper greatly.
Source of Support: Nil
Conflict of Interest: None declared.
Contributorship: DSP, ZK, AKD, BCD conceived the idea and planned the study. DSP, ZK, JPS did the review of literature. JPS, DSP, ZK and BCD, performed data extraction, statistical analysis and tabulation. DSP and ZK prepared the manuscript. DSP, ZK, AKD and BCD provided technical suggestions; DSP and AKD supervised the study. All the authors read and approved the final draft manuscript.
| References|| |
Kalaria RN, Maestre GE, Arizaga R, et al. Alzheimer’s disease and vascular dementia in developing countries: prevalence, management, and risk factors. Lancet Neurology. 2008;7(9):812–826.
Prince M. Dementia in developing countries: a consensus statement from the 10/66 Dementia Research Group. International Journal of Geriatric of Psychiatry. 2000;15(l):14–20.
Alladi S, Kaul S, Mekala S. Vascular cognitive impairment: current concepts and Indian perspective. Annals of Indian Academy of Neurology. 2010;13(Suppl 2):S104-S108.
Das SK, Bose P, Biswas A, Dutt A, Banerjee TK, Hazra AM, Raut DK, Chaudhuri A, Roy T. An epidemiologic study of mild cognitive impairment in Kolkata, India. Neurology. 2007;68(23):2019–2026.
Suh GH, Shah A. A review of the epidemiological transition in dementia—cross-national comparisons of the indices related to Alzheimer’s disease and vascular dementia. Acta Psychiatrica Scandinavica. 2001;104(1):4–11.
Cagney KA, Lauderdale DS. Education, wealth, and cognitive function in later life. Journal of Gerontology Series B: Psychological Sciences and Social Sciences. 2002;57(2):163–172.
Kubzansky LD, Berkman LF, Glass TA, Seeman TE. Is educational attainment associated with shared determinants of health in the elderly? Findings from the McArthur Studies of Successful Aging. Psychometric Medicine. 1998 Sept l;60(5):578–585.
Evans DA, Hebert LE, Beckett LA, Scherr PA, Albert MS, Chown MJ, Pilgrim DM, Taylor JO. Education and other measures of socioeconomic status and risk of incident Alzheimer disease in a defined population of older persons. Archives of Neurology. 1997;54(11):1399–1405.
Evans DA, Beckett LA, Albert MS, Hebert LE, Scherr PA, Funkenstein HH,Taylor JO. Level of education and change in cognitive function in a community population of older persons. Annals of Epidemiology. 1993;3:71–77.
Lee S, Kawachi B, Berkman F, Grodstein F. Education, other socioeconomic indicators, and cognitive function. American Journal of Epidemiology. 2003;157(8):712–20.
Koster A, Penninx W, Boseman H, et al. Socioeconomic differences in cognitive decline and the role of biomedical factors. Annals of Epidemiology. 2005;15(8):564–571.
Kowal P, Chatterji S, Naidoo N, et al. Data resource profile: the World Health Organization Study on global AGEing and adult health (SAGE). International Journal of Epidemiology. 2012;41:1639–49.
Ganguli M, Chandra V Gilby JE, Ratcliff G, Sharma SD, Pandav R Seaberg EC, Belle S. Cognitive test performance in a community-based nondemented elderly 27 sample in rural India: The Indo-U.S. Cross National Dementia Epidemiologic Study. International Psychogeriatrics. 1996;8(4):507–524.
Mathuranath PS, George A, Cherian PJ, Alexander A, Sarma SG, Sarma PS. Effects of age, education and gender on verbal fluency. Journal of Clinical and Experimental Neuropsychology. 2003;25(8):1057–1064.
Sharma VK, Das S, Mondol S, Goswami u, Gandhi A. Effect of sahaj yoga on neuro-cogntive functions in patients suffering from major depression. Indian Journal of Physiology and Pharmacology. 2006;50(4):375–383.
Rosen WG. Verbal fluency in aging and dementia. Journal of Clinical Neuropsychology. 1980;2(2):135–146.
Maika A, Mittinty M, Brinkman S, Harper S, Straiwan S, Lynch J. Changes in socioeconomic inequality in Indonesian children’s cognitive function from 2000–2007: a decomposition analysis. PLOS one. 2013;8(10):e78809.
Peltzer K, Mafuys P. Cognitive functioning and associated factors in older adults in South Africa. SAJP. 2012;18(4):157–163.
Wilson RS, Hebert LE, Scherr PA, Dong X, Leurgens SE, Evans DA. Cognitive decline after hospitalization in community population of older persons. Neurology. 2012;78(13):950–956.
Sen AK. Missing women – revisited. British Medical Journal. 2003;327(7427):1297–1298.
Rajaram S, Zottarelli LK, Sunil TS. Individual, household, programme and community effects on childhood malnutrition in rural Indią Maternal and Child Malnutrition. 2007;3 (2):129–140.
Subramanian sv, Ackerson LK, Subramanyam MA, Sivaramakrishnan K. Health inequalities in India: the axes of stratification. The Brown Journal of World Affairs. 2008;14(2):127–138.
Rogers WH. Regression standard errors inclustered samples. Stata Tech Bull. 1993;13:19–23.
Liang KY, Zegar SL. Longitudinal data analysis using generalized linear model. Biometrika. 1986;73(l):13–22. doi:10.1093/biomet/73.1.13.
Crouchley R, Davis RB. A comparison of population average and random effects models for the analysis of random effects count data with base-line information. Journal of Royal Statistical Society. 1999;162(3):331–347.
Pan w. Akaike’s information criterion in generalized estimating equations. Biometrics. 2001;57(1):120–125.
ott A, Breteler MM, Van Harskamp F, Claus JJ, Van der Cammen TJ, Grobbee DE, Hofman A. Prevelance of Alzheimer’s disease and vascular dementia: Association with education. The Rotterdam study. British medical Journal. 1995;310:970–973.
Stern Y, Albert SM, Sano M, et al. Assessing patient dependence in Alzheimer’s disease. Journal of Gerontology. 1994;49:M216-M222.
Hultsch DF, Hertzog C, Small G, Dexon RA. Use it or lose it: engaged lifestyle as a buffer of cognitive decline in aging? Psychology & Ag1ng. 1999;14:245–263.
Maurer J. Education and male-female differences in later-life cognition: international evidence from Latin America and the Caribbean. Demography. 2011;48(3):915–930.
Langa KM, Llewellyn DJ, Lang IA, Wei, DR, Wallace RB, Kabeto MU, Huppert FA. (2009). Cognitive health among older adults in the United States and in England. BMC Geriatrics. 2009;9:23. doi:10.1186/1471-2318-9-23.
Kaplan GA, Turrell G, Lynch JW, Everson SA, Helkala E, Salonen, JT. Childhood socioeconomic position and cognitive function in adulthood. International Journal of Epidemiology. 2001;30(2):256–263.
Luo Y Waite LJ. The impact of childhood and adult SES on physical, mental and cognitive well-being in later life. Journal of Gerontology: Social Sciences. 2005;60B(2):S93-S101.
MacDonald SW, Hultsch DF, Bunce D. Intraindividual variability in vigilance performance: does degrading stimuli mimic age-related “neural noise”? Journal of Clinical and Experimental Neuropsychology. 2006;28(5):655–675.
Gerstorf D, Hoppmann CA, Kadlec KM, McArdle JJ. Memory and depressive symptoms are dynamically linked among married couples: longitudinal evidence from the AHEAD Study. Developmental Psychology. 2009;45(6):1595–1610.
English Longitudinal Study Of Ageing. Insight into a maturing population. University of London. www.ifs.org.uk/elsa/ - accessed 13 February 2014.
[Table 1], [Table 2]
|This article has been cited by|
||Letter Verbal Fluency in Spanish-, Basque-, and Catalan-Speaking Individuals: Does the Selection of the Letters Influence the Outcome?
| ||Laiene Olabarrieta-Landa,Itziar Benito-Sánchez,Montserrat Alegret,Anna Gailhajanet,Esther Landa Torre,Juan Carlos López-Mugartza,Juan Carlos Arango-Lasprilla |
| ||Journal of Speech, Language, and Hearing Research. 2019; 62(7): 2400 |
|[Pubmed] | [DOI]|
||Effects of 2-Year Cognitive–Motor Dual-Task Training on Cognitive Function and Motor Ability in Healthy Elderly People: A Pilot Study
| ||Emiko Morita,Hisayo Yokoyama,Daiki Imai,Ryosuke Takeda,Akemi Ota,Eriko Kawai,Yuta Suzuki,Kazunobu Okazaki |
| ||Brain Sciences. 2018; 8(5): 86 |
|[Pubmed] | [DOI]|
||Women’s mid-life health in Low and Middle Income Countries: a comparative analysis of the timing and speed of health deterioration in six countries
| ||Tiziana Leone |
| ||SSM - Population Health. 2018; : 100341 |
|[Pubmed] | [DOI]|
||Long-term exposure to ambient PM 2.5 associated with fall-related injury in six low- and middle-income countries
| ||Yanfei Guo,Hualiang Lin,Yan Shi,Yang Zheng,Xing Li,Jianpeng Xiao,Tao Liu,Weilin Zeng,Michael G. Vaughn,Lenise A. Cummings-Vaughn,Erik J. Nelson,Zhengmin (Min) Qian,Wenjun Ma,Fan Wu |
| ||Environmental Pollution. 2017; |
|[Pubmed] | [DOI]|
||Dengue Contingency Planning: From Research to Policy and Practice
| ||Silvia Runge-Ranzinger,Axel Kroeger,Piero Olliaro,Philip J. McCall,Gustavo Sánchez Tejeda,Linda S. Lloyd,Lokman Hakim,Leigh R. Bowman,Olaf Horstick,Giovanini Coelho,Duane J. Gubler |
| ||PLOS Neglected Tropical Diseases. 2016; 10(9): e0004916 |
|[Pubmed] | [DOI]|
||Measuring adolescent friendly health services in India: A scoping review of evaluations
| ||Andrea J. Hoopes,Paras Agarwal,Sheana Bull,Venkatraman Chandra-Mouli |
| ||Reproductive Health. 2016; 13(1) |
|[Pubmed] | [DOI]|
||Cross-sectional Serosurvey of Crimean-Congo Hemorrhagic Fever Virus IgG in Livestock, India, 2013–2014
| ||Devendra T. Mourya,Pragya D. Yadav,Anita M. Shete,Padmakar S. Sathe,Prasad C. Sarkale,Bramhadev Pattnaik,Gaurav Sharma,Kamlesh J. Upadhyay,Surekha Gosavi,Deepak Y. Patil,Gouri Y. Chaubal,Triparna D. Majumdar,Vishwa M. Katoch |
| ||Emerging Infectious Diseases. 2015; 21(10): 1837 |
|[Pubmed] | [DOI]|