The present study uses data from a secondary source: the first wave of Longitudinal Ageing Study in India [LASI], conducted during 2017-18. The LASI data has particular importance as it’s the first study on older adults and covers more than 72 thousand participants residing across India except for the state of Sikkim. The data comprises robust information on social, economic, health, morbidity, and other covariates. In order to get a scientific and representative sample, probability-based multistage stratified area cluster sampling was used for data collection. Detailed information on LASI concerning the instrument, survey design, and data collection can be accessed from the recently published LASI India report[12].
The onset of frequently occurring chronic morbidities, old age-related cognitive decline, and its associated factors become more prominent since the age of 45 [13], from the vast array of information on older adults in India, this study only focuses on a population aged 45 and above, which accounts for a sample of 65,562 respondents. We have used multiple covariates information from the LASI data to study our objectives.
Variables
Outcome variable: As per the purposes of the study, Cognitive Impairment is taken as the primary dependent variable. It is fragmented into five broad domains: arithmetic function, memory, orientation, executive function, and object naming. The composite Cognitive Impairment Index [CII], derived from the Michigan University of Health and Retirement Study [HRS], is utilized to understand the cognitive health among older adults[12][14]. The CII has been constructed based on the responses to questions on the above-mentioned five domains of cognitive measure.
Description of domain-wise cognitive measures.
|
Domain
|
Measure
|
Total Score Range
|
Memory
|
Word recall
|
0-20
|
Orientation
|
Time and place
|
0-8
|
Arithmetic function
|
Backward counting, serial, and computation
|
0-9
|
Executive function
|
Executive [paper folding] and Pentagon drawing
|
0-4
|
Object naming
|
Object naming
|
0-2
|
Composite Cognitive Index
|
Total score cognition
|
0-43
|
Source: LASI, India Report [1].
The CII score ranges from 0-43, in which a high score denotes high cognitive functioning [12]. Moreover, to understand mental health in the most affected group, the population in the lowest 10th percentile was used to study cognitive impairment. The respondents who reported cumulative scores in the lowest 10th percentile were considered as having low cognitive health, and the rest were considered having normal cognition levels.
Explanatory Variables
The key explanatory indicators for our study are as follows: social capital, Activities of Daily Living [ADL], Instrumental Activities of Daily Living [IADL], living arrangement, multimorbidity, health behaviors like physical activity, smoking, and alcohol drinking.
Different methods exist for measuring Social Capital Index [SCI][15][16][17]. Based on previous studies and availability of SCI information in LASI data, the index was constructed based on the following questions: 1] How many meetings/regular gatherings, if any, do you attend in a year?; 2]Visit relatives /friends; 3]Attend cultural performances/shows/Cinema. 4] Attend religious functions /events such as bhajan/Satsang/prayer; 5] Attend political/community/organization group meetings. Each of the above-stated questions had four choices [0-never visited, 1- daily, 2-in a week, 3-in a month, 4-in a year]. Responses from all five questions were added, and the final scores were categorized into three equal percentile as Low, Middle, and high, each demonstrating the equivalent level of social capital.
Similarly, other important covariates such as functional abilities measured through ADL [yes/no] and IADL [yes/no], living arrangement [single/ spouse/ others], health behavior [physical activity, smoking, and alcohol drinking], and multimorbidity [no/single/multi] variables were constructed and categorized based on the available data. Further, this study has also included secondary adjustable socio-economic and demographic characteristics such as working status [yes/no], marital status [currently married/widowed/divorced or separated], age [45-59 and 60 and above], sex [male/female], place of residence [rural/urban], education [respondents with- no schooling/1-5 years of schooling/5-9 years of schooling/10 or more years of schooling], caste [Schedule Tribe [ST]/ Scheduled Caste [SC]/ other backward class[OBC]/ Other], religion [Hindu/Muslim/ Christen/ other], and Monthly Per Capita Expenditure-MPCE [poorest/poorer/middle/richer/richest]. With the advantage of available information on study covariates and as per our objective, we have applied statistical analysis as mentioned below.
Statistical Analysis
The present study systematically examines how cognitive health is affected by social and physical health, risk behaviors, and other background characteristics. The study population is divided into two age groups- 45-59 and 60+. At first descriptive analysis is conducted to understand the socio-demographic profile of the study sample. We have also used descriptive statistics and cross-tabulation to assess the status of social capital and cognitive impairment across background characteristics. For examining the relationship between the predictors and predictands, a chi-square test for association at a 5% level of significance is done. Further, to examine the association between outcome and explanatory variables, multiple logistic regression is also used. In the end, we have used the dominance analysis, which is used to determine the most contributing factors among the selected explanatory variables for cognitive impairment.