Multimorbidity and its associated risk factors among the older adults in India

Health at older ages is a key public health challenge especially among the developing countries. Older adults are at greater risk of vulnerability due to their physical and functional health risks. With rapidly rising ageing population and increasing burden of non-communicable diseases elderly in India are at a greater risk for multi-morbidities. Therefore, to understand this multimorbidity transition and its determinants we used a sample of older Indian adults to examine multimorbidity and its associated risk factors among the Indian elderly aged 45 and above. Using the sample of 72250 older adults this study employed the multiple regression analysis to study the risk factors of multimorbidity. Multimorbidity was computed based on the assumption of elderly having one or more than one of the diseases risks. Our results confirm the emerging diseases burden among the older adults in India. One of the significant findings of the study was the contrasting prevalence of multimorbidity among the wealthiest groups, which diverges from some earlier studies in developing countries examining the multimorbidity. Thus, given the contrasting results and rise of multimorbidity among older adults India, there is paper argues for an immediate need for proper policy measures and health system strengthening to ensure the better health of older adults in India. Keywords: Multimorbidity, older adults, Epidemiological transition, Health, Diseases burden

. CC-BY-NC-ND 4.0 International license It is made available under a is the author/funder, who has granted medRxiv a license to display the preprint in perpetuity. (which was not certified by peer review) preprint The copyright holder for this this version posted November 15, 2021. ;https://doi.org/10.1101https://doi.org/10. /2021 Introduction India is witnessing an unprecedented change in the demographic and social structure in the recent decades. India is experiencing an epidemiological transition which witnesses a rising burden of noncommunicable diseases (NCDs) (Quigley 2006;Agrawal andArokiasamy 2010, Yadav andArokiasamy 2014). NCDs are rapidly increasing in India mainly because of lifestyle changes 2 . With the ageing of population in India, which has now become a challenge for public health experts, policymakers, and other research organizations 3-5 , increasing prevalence of senility is a concern in India with the rise of NCDs. There is an urgent need to understand the burden of chronic health conditions among elderly Indians, in order to improve and develop suitable responses for the future requirements of healthcare services.
Increase in longevity and decrease in mortality leads to increase the multiple comorbid conditions which is commonly known as 'Multimorbidity'. In other words, Multimorbidity is defined as the coexistence of two or more chronic conditions which have become prevalent widely 6,7 . Multimorbidity has now emerged as major public health issues worldwide and its associated greater adverse outcome of health like-disability, mortality, poor quality of life, hospitalizations, consequent use of medical resources and health expenditure [8][9][10][11] .
A systematic review study has revealed that prevalence of multimorbidity among the elderly were found to be more than 55% in different countries 14 . Despite that, high multimorbidity prevalence have been observed in many developed nations, for instance-United states, Australia 13 , Canada 15 & Europe 9,12,16 . Interestingly, the elderly from developing nations are inadequately equipped with the multimorbidity challenge, as a result a study conducted in Vietnam 17 revealed that more than 40% of elderly had multimorbidity conditions, whereas 69% in China 11 and 52% in Bangladesh 18 . Besides that, the least developed country like Tanzania showed 25.3% multimorbidity prevalence among the elderly population.

Multimorbidity in Indian settings
Multimorbidity research in India among the elderly is still at early stage. About 23.3% multimorbidity prevalence has been observed in India in the previous study conducted in 2017 . CC-BY-NC-ND 4.0 International license It is made available under a is the author/funder, who has granted medRxiv a license to display the preprint in perpetuity. (which was not certified by peer review) preprint The copyright holder for this this version posted November 15, 2021. ;https://doi.org/10.1101https://doi.org/10. /2021 where Kerala showed the highest prevalence of multimorbidity with 42% followed by Punjab Works of literature has suggested that there exists a strong positive association between age and the prevalence of multimorbidity in India 19,23-25 . A study conducted in Odisha 26 has revealed that multimorbidity prevalence were higher among women than men and similar results have been also found in West Bengal 21 . The rich elderly group in India were more likely to have poor health due to long term multimorbidity conditions 27 . Recent studies have revealed that there exists significant associations between obesity 28 and loneliness 29 accompanied with multimorbidity in India. Another recent study has investigated in Odisha that multimorbidity increases the odds of elderly abuse 30 .There are very few studies that exists on the multimorbidity prevalence and its associated risk factors among the elderly in India. Therefore, we aim to examine the prevalence of multimorbidity among the elderly in India and its states and, we examine its associated risk factors.

Data source
The data for this study has been taken from Longitudinal Ageing Study in India (LASI) Wave 1, which was carried out during 2017-18. LASI is a multidisciplinary, internationally harmonized panel study of 72,250 older adults aged 45 and above including their spouses less than 45 years, representative to India and all of its states and union territories (excluding Sikkim). It is a baseline data of India's first longitudinal ageing study that provides comprehensive scientific evidence base on demographics, household economic status, chronic health conditions, symptom-based health conditions, functional health, mental health (cognition and depression), biomarkers, health insurance and healthcare utilization, family and social networks, social welfare programs, work and employment, retirement, satisfaction, and life expectations.
. CC-BY-NC-ND 4.0 International license It is made available under a is the author/funder, who has granted medRxiv a license to display the preprint in perpetuity. (which was not certified by peer review) preprint The copyright holder for this this version posted November 15, 2021. ;https://doi.org/10.1101https://doi.org/10. /2021 Analytical sample

Outcome variables
The outcome variable in the study is multimorbidity which was measured on the basis of multiple chronic diseases reported among the older adults surveyed. Respondents were asked about 10 various diseases (See supplementary file) from which the outcome variable of this study was computed. These responses were combined together into trichotomous variable with categories (0= No), (1= single) and (2=more than one morbidity) to study the prevalence. But for regression analysis we converted the variable into two categories where '0' represented no morbidity and '1' denotes multimorbidity to apply the logistic model in the study.

Independent variables
Demographic and socio-economic risk factors included in the study, such as age, gender, residence, level of education, health insurance status, MPCE Quintiles, caste-group, religion, currently working, marital status (See supplemental file).

Statistical analysis
We used frequencies, percentages and cross tabulations for prevalence of mmultimorbidity with respect to the social and demographic characteristics with 95% confidence interval. We applied chi-square test (χ2) to see the association between multimorbidity and its covariates. We then performed the logistic regression to study the determinants of multimorbidity among older adults in India

Socio-demographic characteristics
The socio-economic and demographic characteristics of the study population by the number of diseases are presented in Table 1. The mean number of illnesses per year caused by single disease in the sample was 2.58 (SD = 2.19) with 18.75% (95% CI 18.16% -19.36%), whereas it was 62.68% (95% CI 61.9% -63.45%) for more than one disease, i.e., multimorbidity.

Single & multimorbidity prevalence at state level
Map 1 shows the single morbidity prevalence among the elderly in India at state level using LASI Wave-1 data. The highest single morbidity prevalence was in Odisha (24.4%), followed by

Determinants of Multimorbidity
Multimorbidity is common at higher ages given the associated risk of physical and functional vulnerability. There are numerous other risk factors like smoking drinking, underweight obesity, physical limitations and occupational exposures that are likely to enhance the risk for multimorbidity. Multivariate associations between population characteristics and multimorbidity . CC-BY-NC-ND 4.0 International license It is made available under a is the author/funder, who has granted medRxiv a license to display the preprint in perpetuity. (which was not certified by peer review) preprint The copyright holder for this this version posted November 15, 2021. ; https://doi.org/10.1101/2021.11.12.21265083 doi: medRxiv preprint are presented in Table 2. The predicted probability of having a multiple disease showed a significant increase, and it is almost 5 times more likely in 71-80 years (Adjusted OR=5.033; 95% CI = 4.553 -5.564; reference category ≥ 40 years of age). Women were more likely than men to have more than one morbidity (Adjusted OR=1.354; 95% CI = 1.296 -1.416). Other

Discussion
Multimorbidity is emerging as a critical public health challenges, especially, in the developing countries such as India. Owing to the life style changes and shift in disease pattern and rise in out-of-pocket expenditure (OOPE), multimorbidity is resulting into an economic burden for countries. In parallel to the rise in multimorbidity, ageing of population with increase in life expectancy has become a major public health challenge. The ageing of population further manifests the multifold vulnerability in old ages caused by these diseases' risks. In the view to rise in risk of diseases, we examined the prevalence and risk factors of multimorbidity in 45 and above years using data provided in the LASI wave-1 in India.
An individual suffers from multimorbidity due to multiple reasons ranging from comorbidities that may arise due to a common risk factor or due to the outcome of a particular diseases leading to other diseases 31 . This risk likely enhances with age due to physical and functional vulnerabilities. Research shows ageing contributes to multimorbidity through the loss of physical and functional health including frailty, which later results into greater complications like falls, disability, immobility, and mortality 32,33 .
. CC-BY-NC-ND 4.0 International license It is made available under a is the author/funder, who has granted medRxiv a license to display the preprint in perpetuity. (which was not certified by peer review) preprint The copyright holder for this this version posted November 15, 2021. ; https://doi.org/10. 1101/2021 Our results showed the significant association between multimorbidity and its associated demographic and socio-economic risk factors like age, income, education and place of residence.
The results corroborate with the earlier findings where significant association was found between multimorbidity and socio-economic outcomes 34 .
This study showed the significant and positive relation of multimorbidity in urban areas. The risk associated with multimorbidity is higher in 45 and above years in urban areas as compared to rural areas. This higher risk in urban areas is likely attributable to increasing life style changes 35 .
This higher risk of disease in urban areas is also appreciated due to the imbalance in medical care that exists in weak health care facilities 36 .
One of the significant findings of this paper is the contrasting prevalence of multimorbidity among the most wealthiest groups, which diverges from some earlier studies from developing countries examining the multimorbidity 37,38 . One of the most likely reasons maybe self-reporting of morbidity given the fact that elderly belonging to better socio-economic classes have greater access to health care service provisions, which increases the likelihood of their diagnosis and care for a particular diseases 39 .
Multimorbidity increases likely to ageing risk as shown by various studies 33,40 . These findings are also well reflected through our results where increase in age likely enhances the risk for more than one morbidity. Therefore, increasing longevity has a likely consequences of morbidity pattern of older adults, which needs an immediate policy attention to avert the challenges of morbidity, disability and death at older ages. Furthermore, strong measures can ensure the active and healthy ageing interventions to avert the diseases burden with greater concentration of older adults in upper ages.

Conclusions
This study provides an evidence of emerging diseases burden among the older adults in India.
The study highlights the need for better interventions for older adults to avert the health crisis in later years of life. As the findings of this research specifically indicate the growing burden of multimorbidity, there is an immediate need for proper policy measures and health system strengthening to ensure the health ageing in India. Moreover emphasis should be given on . CC-BY-NC-ND 4.0 International license It is made available under a is the author/funder, who has granted medRxiv a license to display the preprint in perpetuity. (which was not certified by peer review) preprint The copyright holder for this this version posted November 15, 2021. ; https://doi.org/10. 1101/2021 workforce training and quality improvement strategies which can ensure the better physical and functional health of older adults. There is also an immediate need for improving the financial incentives for elderly at older ages given the challenges they face in terms of health and social security provisions in India.
Agrawal G, Arokiasamy P. . CC-BY-NC-ND 4.0 International license It is made available under a is the author/funder, who has granted medRxiv a license to display the preprint in perpetuity. (which was not certified by peer review) preprint NCDS_and_Associated_Determinants_Among_Elderly_in_India . CC-BY-NC-ND 4.0 International license It is made available under a is the author/funder, who has granted medRxiv a license to display the preprint in perpetuity. (which was not certified by peer review) preprint is the author/funder, who has granted medRxiv a license to display the preprint in perpetuity. (which was not certified by peer review) preprint The copyright holder for this this version posted November 15, 2021. ;https://doi.org/10.1101https://doi.org/10. /2021 28. . CC-BY-NC-ND 4.0 International license It is made available under a is the author/funder, who has granted medRxiv a license to display the preprint in perpetuity. (which was not certified by peer review) preprint The copyright holder for this this version posted November 15, 2021. ; https://doi.org/10.1101/2021.11.12.21265083 doi: medRxiv preprint Table 1 Mean score of morbidity among the older-adults in India (N=72250).

Indicators
Mean Score of morbidity   . CC-BY-NC-ND 4.0 International license It is made available under a is the author/funder, who has granted medRxiv a license to display the preprint in perpetuity. (which was not certified by peer review) preprint

MPCE quintile
. CC-BY-NC-ND 4.0 International license It is made available under a is the author/funder, who has granted medRxiv a license to display the preprint in perpetuity. (which was not certified by peer review) preprint

Place of Residence
. CC-BY-NC-ND 4.0 International license It is made available under a is the author/funder, who has granted medRxiv a license to display the preprint in perpetuity. (which was not certified by peer review) preprint The copyright holder for this this version posted November 15, 2021. ;https://doi.org/10.1101https://doi.org/10. /2021