Association of chronic morbidity and early retirement in India: Evidence from Longitudinal Ageing Study in India

doi:10.21203/rs.3.rs-3226649/v1

Context:

Chronic diseases have a profound impact on productivity and job interest, often resulting in significant losses in both. Numerous scientific research has shown that chronic illnesses have a negative impact on work-related outcomes. The findings of these studies highlight the importance of studying the association of chronic morbidity and early retirement by addressing chronic diseases comprehensively in the workplace to promote employee well-being, productivity, and overall work satisfaction.

Methods

The initial phase of the Longitudinal Ageing Study in India (LASI) was utilized, focusing on outcome variables related to early retirement caused by poor health. Through multivariate analysis using a probit regression model, we evaluated the correlation between early retirement and specific chronic diseases, as well as the cumulative number of chronic diseases and other sociodemographic factors.

Findings:

Neurological problems and stroke were those showing the strongest association with early retirement, followed by chronic lung and heart diseases. Among other covariates, the risk of retirement was significantly lower among the currently married and rural sector. Considering socioeconomic characteristics, subjects with lower education were significantly at higher risk of early retirement.

Conclusions

We observed a statistically significant connection between chronic health conditions and early retirement among individuals nearing the age for receiving statutory pensions, even after accounting for potential confounding factors. The association between health issues and early retirement was particularly pronounced among subjects affected by stroke and neurological problems. The study results highlight the importance of developing interventions aimed at enhancing prevention and treatment of chronic illnesses, as well as creating workplace accommodations for individuals with health issues, in order to bolster labor force participation rates among older workers

Chronic diseases

early retirement

morbidity

labour market participation

Chronic diseases have a profound impact on productivity and job interest, often resulting in significant losses in both. Numerous scientific research has shown that chronic illnesses have a negative impact on work-related outcomes. Employees with chronic health disorders reported reduced productivity and more absenteeism, resulting in significant economic losses for firms (Lerner et al., 2004). Furthermore, chronic diseases can cause fatigue, pain, and cognitive impairments, all of which can interfere with concentration and focus, lowering work performance (Beard et al., 2016). Individuals with chronic diseases have lower motivation and involvement in their work duties (Momsen et al., 2016). These findings highlight the importance of studying the association of chronic morbidity and early retirement by addressing chronic diseases comprehensively in the workplace to promote employee well-being, productivity, and overall work satisfaction.

Early onset of chronic diseases in India adversely affect the social, economic and health well-being of individuals and the family. Evidence suggest onset of chronic diseases at 53 years in 2018 and it has been progressing to younger age (Akhtar et al., 2022; Mohanty et al., 2021). Besides ill health, increasing medical spending and the early onset of chronic diseases reduces the propensity to work and may lead to early retirement.

Poor health is one of the determinant of early retirement (Perez et al., 2022). Previous studies have predominantly linked early retirement with cancer (Carlsen et al., 2008), coronary heart disease (Kruse et al., 2009), depression (Karpansalo et al., 2005), arthritis (Alavinia, 2008), and diseases of the nervous system (Li Ranzi T, d’Errico A, 2013).

A number of studies have shown that health status is an important factor influencing retirement, demonstrating that both perceived poor health and the presence of chronic disorders or disabilities are associated with an early retirement from paid work (Alavinia, 2008; Karpansalo et al., 2005; Schuring M, Burdorf L, Kunst A, 2007; van den Berg TI, Elders LA, 2010; Von Bonsdorff ME, Huuhtanen P, Tuomi K, 2010). Though most of the researchers agree on a direct impact of poor health on early exit from paid work, a study from the US have contrasting results (Miah MS, n.d.), demonstrating that people with chronic diseases tend to retire later because they have saved fewer assets during their working life. However, there is no consensus about the magnitude of the effect of health relative to the effect of other characteristics (Deschryvere M., 2004).

India has been experiencing gradual increase in life expectancy, and accompanied by increased labour market participation among older people (Alam M, 2012). However, with the increasing life expectancy, non-communicable diseases are increasing faster and adversely impact productivity. The NCDs are the leading cause of early retirement, therefore reducing overall labour market participation (Bloom DE, 2013). Since poor health seems to be a significant factor in early retirement, it is important to examine the specific chronic conditions that cause workers to leave the workforce.

Besides poor health, various sociodemographic and employment-related factors are associated with early retirement, including being female (Alavinia, 2008), living without a partner (Alavinia, 2008; Schuring M, Burdorf L, Kunst A, 2007), lower education (Siegrist & Wahrendorf, 2009), lower socioeconomic position (Lund & Villadsen, 2005), economic incentives (Nilsson et al., 2011), working in the public sector (Siddiqui, 1997), and workplace characteristics, especially heavy physical work, high work pace, and low job control (Karpansalo et al., 2005; Siegrist & Wahrendorf, 2009).

Among manual workers, who typically face poorer working conditions and potential challenges in performing job tasks when dealing with chronic or limiting illnesses, the effect of poor health is expected to be more pronounced compared to white-collar workers. Conversely, individuals with low socioeconomic status and illnesses may be less inclined to retire early than those from higher social strata due to the relatively smaller income they would have post-retirement. Limited research has explored the interaction between poor health, social class, and early retirement, but generally, higher-educated individuals have been observed to have a slightly elevated risk (McDonough & Amick III, 2001; Schuring M, Burdorf L, Kunst A, 2007). (Chanda & Mishra, 2019) Chanda and Mishra (2019) using the cross-sectional data on 5212 elderly from the World Health Organization’s Study on global aging and adult health (Wave 1) (2007–08) in India, assessed the impact of work status on cognitive health among elderly in India. In India, the association between health and early retirement in a representative sample of the ageing population has not been assessed by any study. Hence, the impact of poor health on early retirement is significant within this country. It is also of interest to assess which chronic conditions are leading to withdraw from the workforce, in order to prevent their premature exit.

The main aims of this chapter are to (1) examine the association between the chronic morbidity and early retirement in an Indian working population approaching the statutory pension age; (2) assess whether the association between premature retirement and morbidity was modified by socioeconomic status or by other sociodemographic characteristics?

Data

The unit data from large-scale population-based surveys, namely, the Longitudinal Ageing Study in India (LASI), Wave 1, conducted during 2017-18 has been used. LASI is a nationally representative study on the health, economic and social wellbeing of adults aged 45 and above. It is a collaborative study, of International Institute for Population Sciences (IIPS), Mumbai, the Harvard T. H. Chan School of Public Health, and the University of Southern California, USA. LASI is internationally harmonized with the major longitudinal studies; the Health and Retirement Survey (HRS), USA, the English Longitudinal Study of Ageing (ELSA), and the China Health and Retirement Study (CHARLS), Japanese Study of Ageing and Retirement (JSTAR) and the Korean Longitudinal Study of Aging (KLoSA) among others.

It holds the distinction of being the largest, worldwide study with a representative sample of 73,396 individuals and 43,688 households across all states and union territories of India. The age-eligible households include who have at least one member who is 45+. LASI used the multistage stratified sampling in the selection of the sample households. It provides comprehensive information on parameters like economic wellbeing, work and employment, retirement and pension, chronic health conditions, functional health, mental health, extensive biomarkers, health care utilisation, and health insurance of adults aged 45 and above. The details of the sampling procedure, the instruments used, and the findings of the survey are available in the national report (IIPS et al., 2020). We have classified the subjects into four groups such as retired, not retired, never worked and not retired but not working.

Health measurement

Two different health measures have been used to assess the influence of health problems on the risk of early retirement: number of diseases reported and the specific chronic illnesses.

In the survey, participants were asked to indicate which chronic illnesses they suffered from and whether they were diagnosed by a doctor, choosing from a list of nine chronic illnesses. In our analysis, we have taken into account all chronic illnesses except for high cholesterol. The presence of chronic illnesses was self-reported but based on doctor’s diagnoses. With the help of dichotomous answers on the specific chronic illness, a sum score was calculated as a proxy measure of general health status and classified into four groups (0, 1, 2, 3 or more chronic illnesses).

Covariates

We have used a set of demographics, economic, behavioural and health covariates in the analyses. These includes age (45–54, 55–64, 65–74, 75+), sex categorized as “male” and “female”, place of residence (rural/urban), marital status (currently married, widowed, others), educational attainment (illiterate, less than 5 years, 5–9 years completed, 10 years or more), caste (scheduled caste, scheduled tribe, other backward classes, others), religion (Hindu, Muslim, Christian, others), the monthly per capita consumption expenditure (MPCE) was categorized into five groups. i.e. “poorest”, “poorer”, “middle”, “richer”, and “richest”, the living arrangement of older adults was categorized as “living alone”, “living with spouse or others”, “living with spouse and children”, and “living with children and others” and Chronic diseases includes hypertension, diabetes, cancer, chronic lung disease, chronic heart disease, stroke, arthritis, and neurological or psychiatric problem. Based on these, individuals are categorized into three groups, i.e. older adults with no chronic disease, with one chronic disease, two chronic diseases, three chronic diseases, four chronic diseases, and more than 5 chronic diseases. Each chronic disease is dichotomized as the presence of a particular chronic condition (yes) and the absence of any chronic condition (no). All chronic conditions include hypertension, diabetes, cancer, chronic lung disease, chronic heart disease, stroke, arthritis, and neurological or psychiatric problem.

Statistical analysis

Prevalence of specific chronic diseases and other subjects’ characteristics by different work status (retired, currently working, never worked, and not retired but not working were computed as weighted proportions. The association of early retirement with specific chronic diseases, the cumulative number of chronic diseases, and other sociodemographic were assessed through multivariate analysis, by means of probit regression model. We used a multivariate probit model to estimate full adjusted difference in the probability of early retirement across chronic diseases and sociodemographic groups.

Probit models are a form of a statistical model that is used to predict the probability of an event occurring. In probit model, the cumulative standard normal distribution function Φ(⋅) is used to model the relationship between the predictors and the probability of the event occurring. The output of the Probit model ranges from 0 to 1, like that of the Logit model.

The Probit model can be represented using the following formula:

Pr(Y = 1|X) = Φ(Z)

Where,

Z = b0 + b1X1 + b2X2 + ….. + bnXn

Where, Y is the early retirement and represents the probability that the event will occur (hence, Y = 1) given the variables X. Z is the linear combination of independent variables (X) with coefficients (b0, b1, b2…bn).

Table 1

Socioeconomic and demographic characteristics of sample population among middle aged and elderly in India, 2017–2018
	Retired		Currently Working		Never Worked		Not working
	Number	Percent	Number	Percent	Number	Percent	Number	Percent
MPCE Quintile
Poorest	254	8.52	6,867	21.35	3,559	17.03	3,531	23.5
Poorer	379	12.72	6,774	21.06	3,887	18.6	3,177	21.14
Middle	549	18.42	6,412	19.93	4,155	19.88	3,098	20.62
Richer	700	23.49	6,240	19.4	4,463	21.36	2,813	18.72
Richest	1,098	36.85	5,874	18.26	4,832	23.12	2,406	16.01
Educational attainment
Illiterate	321	10.61	14,237	43.75	10,146	47.66	8,474	55.32
Less than 5 years	199	6.58	3,855	11.85	2,084	9.79	1,907	12.45
5–9 years completed	676	22.34	7,939	24.4	5,373	25.24	2,028	13.24
10 years or more	1,830	60.48	6,508	20	3,684	17.31	2,028	13.24
Age
45–54	152	5.02	15,214	46.75	6,445	30.27	2,271	14.83
55–64	904	29.87	9,685	29.76	5,597	26.29	3,930	25.66
65–74	1,360	44.94	4,082	12.54	3,920	18.41	5,196	33.92
75+	602	19.89	749	2.3	2,024	9.51	3,360	21.94
Sex
Male	2,563	84.7	19,453	59.78	1,264	5.94	7,242	47.28
Female	463	15.3	13,088	40.22	20,025	94.06	8,075	52.72
Residence
Rural	1,299	42.93	23,151	71.14	11,670	54.82	10,382	67.78
Urban	1,727	57.07	9,390	28.86	9,619	45.18	4,935	32.22
Caste
Scheduled Tribes	439	14.53	6,973	21.45	2,832	13.33	2,252	14.73
Scheduled Castes	362	11.98	5,851	18	3,016	14.2	2,805	18.34
OBC	929	30.75	12,385	38.1	7,513	35.36	6,332	41.41
Others	1,291	42.73	7,295	22.44	7,884	37.11	3,903	25.52
Religion
Hindu	2,243	74.12	24,582	75.54	14,395	67.63	11,702	76.4
Muslim	279	9.22	2,921	8.98	3,782	17.77	1,677	10.95
Christian	373	12.33	3,668	11.27	1,839	8.64	1,324	8.64
Others	131	4.33	1,370	4.21	1,270	5.97	613	4
Marital Status
Currently married	2,444	80.77	27,678	85.06	15,356	72.14	9,863	64.4
Widowed	506	16.72	3,748	11.52	5,337	25.07	4,983	32.53
Others	76	2.51	1,115	3.43	594	2.79	470	3.07

Table 1 presents the socioeconomic and demographic profile of the study samples of retired, currently working, not working and never worked. Of the total surveyed individuals, 4% had retired, 45% are currently working, 30% had never worked and 21% are not currently working but are not retired. The samples of retired individual was predominantly richer and richest whereas higher percentage of individuals in lower quintile are currently working. Samples for currently working, not working and never worked were proportionately distributed across the MPCE quintile.

Table 2

Age-sex adjusted estimates of labour force participation by socioeconomic and demographic characteristics among middle aged and elderly in India
	N = 3,058	N = 32,827	N = 15,451	N = 21,948
	Retired	Currently Working	Not Working	Never Worked
	Prevalence (95% CI)	Prevalence (95% CI)	Prevalence (95% CI)	Prevalence (95% CI)
India	4.82 (4.58, 5.10)	43.73 (43.23, 44.22)	20.73 (20.28, 21.17	30.70 (30.24, 31.16)
mpce_quintile
1 Poorest	1.94 (1.71, 2.18)	48.72 (47.95, 49.49)	22.45 (21.77, 23.13)	26.86 (26.17, 27.55)
2 Poorer	2.82 (2.54, 3.10)	47.49 (46.73, 48.26)	21.39 (20.72, 22.06)	28.21 (27.52, 28.91)
3 Middle	4.01 (3.68, 4.34)	46.27 (45.50, 47.04)	20.95 (20.28, 21.62)	28.71 (28.01, 29.41)
4 Richer	4.53 (4.18, 4.89)	45.17 (44.40, 45.94)	20.47 (19.80, 21.14)	29.74 (29.03, 30.45)
5 Richest	7.23 (6.79, 7.67)	41.60 (40.83, 42.38)	19.61 (18.93, 20.28)	31.59 (30.87, 32.32)
education
Illiterate	0.99 (0.87, 1.10)	52.12 (51.62, 52.63)	23.29 (22.83, 23.75)	24.34 (23.93, 24.76)
Less than 5 years	2.12 (1.82, 2.43)	47.01 (45.99, 48.02)	22.12 (21.22, 23.03)	29.06 (28.06, 30.05)
5–9 years completed	4.19 (3.86, 4.52)	41.43 (40.75, 42.11)	18.77 (18.13, 19.40)	36.47 (35.78, 37.17)
10 years or more	12.89 (12.28, 13.50)	36.59 (35.86, 37.31)	16.55 (15.87, 17.24)	35.65 (34.83, 36.47)
residence
1 Rural	2.65 (2.50, 2.80)	50.66 (50.23, 51.08)	21.57 (21.19, 21.94)	25.07 (24.69, 25.45)
2 Urban	6.98 (6.65, 7.31)	36.79 (36.23, 37.36)	19.88 (19.37, 20.39)	36.32 (35.78, 36.86)
caste
SC	3.11 (2.78, 3.44)	48.85 (48.01, 49.69)	23.61 (22.84, 24.38)	24.37 (23.62, 25.12)
ST	3.56 (3.22, 3.90)	55.33 (54.52, 56.14)	17.90 (17.22, 18.58)	22.98 (22.27, 23.69)
OBC	3.24 (3.02, 3.46)	46.10 (45.55, 46.66)	22.93 (22.43, 23.43)	27.64 (27.14, 28.15)
Others	6.08 (5.74, 6.42)	37.35 (36.70, 37.99)	18.74 (18.19, 19.29)	37.75 (37.14, 38.37)
religion
Hindu	4.09 (3.91, 4.26)	47.12 (46.72, 47.52)	21.95 (21.60, 22.31)	26.78 (26.42, 27.14)
Muslim	3.35 (2.95, 3.75)	34.85 (33.88, 35.82)	19.64 (18.77, 20.51)	42.15 (41.22, 43.07)
Christian	4.97 (4.44, 5.49)	52.07 (50.98, 53.16)	17.45 (16.56, 18.34)	25.57 (24.60, 26.53)
Others	3.55 (2.92, 4.19)	40.51 (38.93, 42.09)	16.99 (15.70, 18.28)	38.80 (37.31, 40.29)
Marital status
Currently married	4.24 (4.06, 4.41)	45.80 (45.42, 46.19)	19.59 (19.23, 19.96)	30.43 (30.04, 30.82)
Widowed	3.43 (3.10, 3.77)	46.28 (45.35, 47.20)	25.05 (24.31, 25.80)	25.04 (24.37, 25.71)
Others	3.71 (2.88, 4.53)	45.71 (43.80, 47.63)	22.19 (20.43, 23.95)	28.43 (26.64, 30.22)
Smoke/Substance use
1 Yes	3.21 (3.02, 3.40)	51.69 (51.05, 52.33)	22.54 (22.00, 23.07)	20.88 (20.25, 21.52)
2 No	4.97 (4.73, 5.21)	42.90 (42.46, 43.34)	20.09 (19.70, 20.48)	31.50 (31.12, 31.88)
Practicing exercise
Yes	8.44 (7.81, 9.08)	40.67 (39.57, 41.77)	19.97 (18.99, 20.96)	30.79 (29.67, 31.92)
Rarely/Never	3.55 (3.40, 3.70)	46.42 (46.06, 46.78)	21.10 (20.78, 21.42)	28.84 (28.51, 29.17)
Health insurance
No	4.08 (3.91, 4.25)	44.32 (43.93, 44.72)	20.64 (20.30, 20.98)	30.86 (30.50, 31.22)
Yes	4.00 (3.68, 4.32)	50.93 (50.21, 51.65)	22.25 (21.60, 22.90)	22.61 (21.98, 23.25)
Chronic conditions
No disease	3.04 (2.86, 3.21)	50.83 (50.39, 51.26)	18.43 (18.05, 18.81)	27.27 (26.88, 27.66)
Diabetes	7.52 (6.92, 8.11)	34.50 (33.40, 35.59)	20.76 (19.80, 21.72)	36.78 (35.69, 37.86)
Cancer	6.63 (3.99, 9.27)	35.04 (30.40, 39.68)	27.13 (22.66, 31.60)	31.07 (26.95, 35.18)
Chronic lung disease	3.36 (2.80, 3.93)	38.31 (36.63, 39.99)	28.25 (26.70, 29.80)	29.24 (27.62, 30.87)
Chronic heart disease	7.54 (6.59, 8.50)	30.61 (28.62, 32.61)	23.38 (21.64, 25.12)	36.27 (34.21, 38.34)
Stroke	5.02 (3.97, 6.07)	23.86 (21.26, 26.46)	36.12 (33.28, 38.96)	30.63 (27.61, 33.66)
Arthritis	4.30 (3.91, 4.69)	40.49 (39.59, 41.39)	24.81 (24.02, 25.60)	29.88 (29.11, 30.65)
Neurological problem	5.04 (4.08, 6.01)	33.12 (30.97, 35.26)	28.62 (26.58, 30.66)	31.51 (29.49, 33.53)
Number of chronic conditions
0	3.03 (2.86, 3.20)	50.86 (50.42, 51.29)	18.44 (18.06, 18.81)	27.23 (26.84, 27.63)
1	4.08 (3.72, 4.43)	42.35 (41.49, 43.22)	24.95 (24.16, 25.74)	28.33 (27.53, 29.13)
2	6.20 (5.75, 6.64)	33.63 (32.70, 34.57)	24.18 (23.36, 24.99)	34.32 (33.45, 35.18)
3	6.59 (5.82, 7.37)	28.82 (27.23, 30.41)	25.84 (24.44, 27.24)	35.69 (34.23, 37.14)
4	7.20 (5.78, 8.62)	22.48 (19.54, 25.43)	27.17 (24.56, 29.79)	36.97 (34.22, 39.72)
5+	9.49 (6.60, 12.39)	18.39 (13.29, 23.50)	23.90 (19.38, 28.42)	41.56 (36.54, 46.59)

Table 2 presents the Age-sex adjusted estimates of labour participation by socioeconomic and demographic characteristics. We estimated that 4.82% (95% CI: 4.58-5.10) older adults in India retired from work, 43.73% (95% CI: 43.23, 44.22) were currently working, 20.73% (95% CI: 20.28, 21.17) were not working and 30.70% (95% CI: 30.24, 31.16) never worked. The proportion of individuals retired increases with the level of education and with the number of chronic conditions. However, the proportion of individuals currently working declines with level of education and number of chronic conditions. It was higher in urban areas (6.98%, 95%CI: 6.65-7.31), among those who did not smoke (4.97%, 95% CI: 4.73-5.21) and among those who were practicing exercise (8.44%, 7.81-9.08). However, no difference in prevalence of retired was observed across caste, religion and marital status.

Figure 1 shows reasons for early retirement among older adults in India. Poor health is the major reason for early retirement. Write the question asked in LASI

Table 3

Sample distribution of sociodemographic characteristics and chronic disease by retirement status and type of retirement
Among those retired (N = 35,885)
	Retired (N = 3,058)	Not Retired (N = 32,827)	Early retirement (N = 375)	Others (N = 2,390)
	(%)	(%)	(%)	(%)
Retirement Status	4.17	95.83	13.56	86.44
MPCE_quintile
1 Poorest	10.01	20.89	9.07	10.13
2 Poorer	14.16	20.69	13.6	14.18
3 Middle	19.52	20.23	21.33	19.21
4 Richer	22.86	19.97	21.6	22.76
5 Richest	33.45	18.23	34.4	33.72
Education
illiterate	10.69	47.42	13.6	8.41
Less than 5 years	6.67	12.05	7.47	6.19
5–9 years completed	22.37	22.69	28	21.51
10 years or more	60.27	17.83	50.93	63.89
Sex
1 Male	84.7	55.89	84.27	86.74
Female	15.3	44.11	15.73	13.26
Residence
Rural	43.17	70.13	44.53	42.22
Urban	56.83	29.87	55.47	57.78
Caste
SC	11.92	18	11.5	11.69
ST	14.84	19.4	20.59	13.57
OBC	30.85	39.35	27.54	31.5
Others	42.38	23.26	40.37	43.23
Religion
Hindu	74	75.69	72.53	74.23
Muslim	9.12	9.53	5.87	9.79
Christian	12.3	10.43	17.87	11.3
Others	4.58	4.35	3.73	4.69
Marital status
Currently married	80.71	78.44	78.93	82.01
Widowed	16.78	18.23	17.07	15.73
Others	2.52	3.33	4	2.26
Smoke/Substance abuse
1 Yes	40.24	42.49	43.82	39.82
2 No	59.76	57.51	56.18	60.18
Practicing exercise
Yes	22.43	9.25	20.75	23.04
Rarely/Never	77.57	90.75	79.25	76.96
Health Insurance
No	77.94	74.59	74.12	77.73
Yes	22.06	25.41	25.88	22.27
No disease	36.12	65.84	34.69	35.25
Hypertension	15.25	11.29	15.72	15.62
Diabetes	17.10	6.17	13.01	17.98
Cancer	0.66	0.31	0.54	0.72
Chronic lung disease	4.16	2.91	5.42	3.93
Chronic heart disease	6.77	1.51	6.50	7.01
Stroke	2.61	0.62	4.61	2.36
Arthritis	14.23	9.88	14.36	14.31
Neurological problem	3.10	1.46	5.15	2.83
Number of chronic condition
0	35.92	58.57	34.32	35.05
1	29.84	26.15	30.29	30.56
2	22.32	10.79	23.59	22.25
3	8.04	3.34	9.12	8.02
4	2.76	0.91	2.14	2.94
5+	1.12	0.24	0.54	1.18

Table 3 shows the descriptive statistics of the variables included in the analysis. Four percent of the subjects (about 8% among men and 1% among women) were retired, as of the survey data. Approximately 57% of the subjects who were retired and 30% of those still working reported residing in urban area. Analysing for the type of retirement, about 14% of the subjects retired early whereas the others constituted mandatory and later retirements. Proportion of early retirement increases with the years of education and MPCE quintile. About 34% and 9% of the subjects who retired early belonged to the richest and poorest quintile respectively. Approximately 44% of the subject who smoked and 80% of the subjects who rarely/never practiced exercise retired early. The highest percentage of subjects who retired and those who not retired were suffering from hypertension. The percentage of subjects suffering from hypertension who retired early or due to other reasons where almost the same. Approximately 30% of the subjects who were retired and 26% of those who were still working reported suffering from at least one chronic disease.

Table 4

Prevalence of specific chronic disease by different work status among middle aged and elderly in India, 2017–2018
	Retired (N = 3,058)	Not Retired (N = 32,827)	Never Worked (N = 21, 948)	Not working (N = 15,451)
	Prevalence (95% CI)	Prevalence (95% CI)	Prevalence (95% CI)	Prevalence (95% CI)
No chronic disease	0.24 (0.23, 0.26)	0.51 (0.50, 0.51)	0.40 (0.39, 0.40)	0.32 (0.31, 0.33)
Hypertension	0.27 (0.25, 0.29)	0.26 (0.26, 0.27)	0.26 (0.26, 0.27)	0.25 (0.25, 0.26)
Diabetes	0.17 (0.16, 0.19)	0.06 (0.06, 0.06)	0.09 (0.09, 0.10)	0.09 (0.08, 0.09)
Cancer	0.01 (0.00, 0.01)	0.00 (0.00, 0.00)	0.01 (0.00, 0.01)	0.01 (0.01, 0.01)
Chronic lung disease	0.04 (0.04, 0.05)	0.03 (0.03, 0.03)	0.03 (0.03, 0.03)	0.06 (0.06, 0.06)
Chronic heart disease	0.07 (0.06, 0.08)	0.02 (0.01, 0.02)	0.02 (0.02, 0.03)	0.03 (0.03, 0.04)
Stroke	0.03 (0.02, 0.03)	0.01 (0.01, 0.01)	0.01 (0.01, 0.01)	0.03 (0.03, 0.03)
Arthritis	0.14 (0.13, 0.16)	0.10 (0.10, 0.10)	0.15 (0.15, 0.16)	0.18 (0.17, 0.19)
Neurological problem	0.03 (0.03, 0.04)	0.01 (0.01, 0.02)	0.02 (0.02, 0.03)	0.03 (0.03, 0.04)

Table 4 presents the Age-sex adjusted estimates of labour participation by specific chronic disease. The prevalence of hypertension is highest among retired (27%, 95% CI: 25–29), not retired (26%, 95% CI: 25–27), not working (26%, 95% CI: 26–27) and never worked (25%, 95% CI: 25–27).

Table 5

Prevalence of retired and early retirement due to poor health by socioeconomic characteristics and specific chronic disease (among which sample. Write N of each row
	Retired (%) (N = 3,058)	Among those retired (Early retirement due to poor health (%) (N = 375)
Chronic disease
No disease	97.99	2.01
Hypertension	96.75	3.25
Diabetes	97.49	2.51
Chronic lung disease	90.48	9.52
Chronic heart disease	94.63	5.37
Stroke	86.08	13.92
Arthritis	95.13	4.87
Neurological problem	85.11	14.89
MPCE quintile
1 Poorest	96.41	3.59
2 Poorer	94.69	5.31
3 Middle	96.65	3.35
4 Richer	96.71	3.29
5 Richest	95.7	4.3
Education
Illiterate	92.66	7.34
Less than 5 years	92.16	7.84
5–9 years completed	95.61	4.39
10 years or more	97.23	2.77
Sex
Male	96.06	3.94
Female	95.94	4.06
Residence
Rural	95.91	4.09
Urban	96.14	3.86
Caste
SC	95.33	4.67
ST	94.7	5.3
OBC	95.97	4.03
Others	96.75	3.25
Religion
Hindu	95.98	4.02
Muslim	96.77	3.23
Christian	96.01	3.99
Others	95.71	4.29
Marital status
Currently married	96.35	3.65
Widowed	95.71	4.29
Others	88.31	11.69

Table 5 shows the prevalence of retired and early retirement among those who retired by chronic disease and sociodemographic characteristics. The highest percentage (15%) of the subjects suffering from neurological problems have retired early, followed by stroke (14%) as compared to other chronic diseases. However, only 2% of the subjects among the ones who have no chronic disease retired early. Those who are not currently married are more likely to retire early due to poor health than the ones who are currently married. Approximately 3% of the subjects with an education level of more than 10 years retired early whereas, about 7% of the subjects among illiterate retired early due to poor health.

Table 6

Adjusted marginal effect for those with early retirement, adults aged 45+, and their spouses in India
	Prevalence
	Marginal effect (95% CI)
Chronic conditions
No chronic disease	0.017 (0.010, 0.024)
Hypertension	0.038 (0.020, 0.056)
Diabetes	0.028 (0.014, 0.043)
Chronic lung disease	0.098 (0.047, 0.149)
Chronic heart disease	0.061 (0.027, 0.095)
Stroke	0.167 (0.083, 0.250)
Arthritis	0.054 (0.033, 0.076)
Neurological problem	0.123 (0.060, 0.187)
mpce_quintile
1 Poorest	0.034 (0.014, 0.053)
2 Poorer	0.047 (0.028, 0.066)
3 Middle	0.032 (0.018, 0.046)
4 Richer	0.035 (0.022, 0.049)
5 Richest	0.046 (0.032, 0.059)
education
Illiterate	0.085 (0.051, 0.119)
Less than 5 years	0.086 (0.046, 0.125)
5–9 years completed	0.042 (0.027, 0.057)
10 years or more	0.027 (0.020, 0.034)
residence
1 Rural	0.038 (0.028, 0.048)
2 Urban	0.041 (0.032, 0.051)
caste
SC	0.036 (0.018, 0.054)
ST	0.053 (0.029, 0.076)
OBC	0.038 (0.026, 0.050)
Others	0.037 (0.026, 0.048)
religion
Hindu	0.043 (0.034, 0.052)
Muslim	0.024 (0.008, 0.041)
Christian	0.032 (0.014, 0.051)
Others	0.047 (0.011, 0.083)
Marital status
Currently married	0.036 (0.028, 0.043)
Widowed	0.047 (0.026, 0.067)
Others	0.142 (0.061, 0.223)

The multivariate probit regression model examining the association of early retirement with chronic diseases and other socioeconomic characteristics. Regarding the association with the specific chronic diseases, from multivariate analysis (Table 6), neurological problems and stroke were those showing the strongest association with early retirement, followed by chronic lung and heart diseases. Among other covariates, the risk of retirement was significantly lower among currently married and rural sector. Considering socioeconomic characteristics, subjects with lower education were at significant higher risk of early retirement.

The results of the present study indicate that in India, early retirement was significantly positively associated with the presence of chronic diseases, lower education, male gender and those not currently married; on the contrary, number of chronic disease was inversely associated with the outcome. Regarding the specific chronic diseases, early retirement was significantly associated with stroke, neurological problems, chronic lung diseases and chronic heart diseases. The same connection could potentially exist for other diseases as well. However, despite the extensive sample size, certain illnesses like cancer did not show any association with the outcome due to the limited number of cases that were exposed, preventing statistical significance from being achieved. Additionally, the findings presented evidence that the link between health and retirement is not consistent across all socioeconomic groups. the probability of early retirement for people with poor health was lower among people with lower socioeconomic status, likely because they lacked sufficient financial resources to cope with a reduction in earnings resulting from retirement.

Only a small number of studies utilized the count of diseases as an indicator of health status, yet all of them consistently demonstrate that a higher prevalence of chronic diseases corresponds to a decrease in the number of years’ individuals remain active in the workforce. By employing linear regression, Yen et al. (2011) estimated that for individuals over the age of 50, the age of retirement decreased by approximately 1 year for each additional illness (Yen et al., 2011). Schofield et al. (2008) findings revealed a noteworthy trend in the impact of the cumulative number of illnesses on the likelihood of individuals being out of the labor force (Schofield et al., 2013). The risks ranged from 1.63 to 16.10 for individuals reporting one illness and six or more diseases, respectively, indicating a significant association. The Australian studies exhibited stronger correlations compared to our own, which can be attributed to a higher proportion of subjects with chronic disorders among those who retired early. This pattern is consistent with the prevailing trend in developed countries, including Australia, where health conditions or disability predominantly contribute to reasons for early retirement. In India, because of the unavailability of social security, workers who retire early are not expected to be, on average, healthier than those retiring prematurely in other developed countries and poor working condition makes it difficult for the subjects to continue working even with one chronic condition.

In the majority of prior studies, the evaluation of health status was confined to general subjective assessments of health, and only a limited number of authors focused on identifying the particular chronic diseases linked to the decision to retire early. A Danish study reported, determinants associated with a higher risk of voluntary early retirement included poorer self-rated health, more depressive symptoms, a higher physical workload, lower job satisfaction and lower influence at work (Sewdas et al., 2020) (Sewdas et al. 2020).

Based on our findings, it is found that the strength of the relationship between retirement and health is comparatively weaker when compared to social class indicators like having a low level of education or economic status. Research has shown that individuals with higher levels of education tend to delay their retirement because they enter the labor market at a later stage and have higher salary expectations (Focarelli & Zanghieri, 2005) (Focarelli and Zanghieri 2005). Furthermore, individuals employed in blue-collar occupations, who are typically subjected to more physically demanding work and poorer working conditions, may encounter greater challenges in carrying out their job responsibilities as they age. This could elucidate their increased likelihood of early withdrawal from the labor force. Conversely, the elevated risks observed among those with lower education, even after adjusting for chronic diseases, indicate that social disparities in early retirement are not solely due to variations in health status among different social classes. The lower risk of early retirement associated with female gender appears in agreement with the results of other studies, where higher risk of early retirement was found predominantly among women (Alavinia, 2008; Cardano et al., 2004; Coppola & Wilke, 2010; Disney et al., 2006) (Alavinia and Burdorf 2008; Disney et al. 2006; Coppola and Wilke 2010; Cardano et al. 2004). Consistent with prior studies that have reported a significantly higher risk of early retirement among individuals without a partner, our findings indicate the presence of an association between marital status and premature retirement (Alavinia, 2008; Schuring M, Burdorf L, Kunst A, 2007) (Schuring et al. 2007; Alavinia and Burdorf 2008).

In summary, we observed a statistically significant connection between chronic health conditions and early retirement among individuals nearing the age for receiving statutory pensions, even after accounting for potential confounding factors. The association between health issues and early retirement was particularly pronounced among subjects affected by stroke and neurological problems. Various specific diseases were identified as factors influencing retirement decisions, and these associations align closely with previous research findings. These study results highlight the importance of developing interventions aimed at enhancing prevention and treatment of chronic illnesses, as well as creating workplace accommodations for individuals with health issues, in order to bolster labor force participation rates among older workers.

Acknowledgments

Not applicable

Authors' contributions

Conceptualization of study: SKM, Analysis and/or interpretation of data: SA; SA participated in the writing of the manuscript; drafting the manuscript: SA; revising the manuscript critically for important content: SKM. The manuscript was prepared under the overall supervision of SKM. Both the authors read and approved the final manuscript.

Funding

The authors did not receive any funding to carry out this research.

Availability of data and material

The datasets generated and/or analysed during the current study are available with the International Institute for Population Sciences, Mumbai, India repository and could be accessed from the following link: https://iipsindia.ac.in/sites/default/files/LASI_DataRequestForm_0.pdf. Those who wish to download the data have to follow the above link. This link leads to a data request form designed by International Institute for Population Sciences. After completing the form, it should be mailed to: [email protected] for further processing. After successfully sending the mail, individual will receive the data in a reasonable time.

Ethical Approval and Consent to participate

The authors were not involved in data collection process and therefore they did not require any ethical approval or consent to participate. The LASI data is secondary in nature. The data is freely available on request and survey agencies that conducted the feld survey for the data collection have collected a prior consent from the respondent. The ethical clearance was provided by Indian Council of Medical Research (ICMR), India. The survey agencies that collected data followed all the protocols. To maximize the cooperation of the sampled HHs and individuals, participants were provided with information brochures explaining the purpose of the survey, ways of protecting their privacy, and the safety of the health assessments as part of the ethics protocols. As per ethics protocols, consent forms were administered to each HH and age-eligible individual. In accordance with Human Subjects Protection, four consent forms were used in the LASI: Household Informed Consent, Individual Informed Consent, Consent for Blood Samples Collection for Storage and Future Use (DBS), and Proxy Consent. For each survey participant, the study protocol was described and the steps of each biomarker test were demonstrated by the trained health investigators. Participant’s informed consent was obtained for the interviews. Since, the survey obtained either signed or oral consent, it was feasible for each participant to provide his/her consent. All methods were performed in accordance with the relevant guidelines and regulations.

Consent for publication

Not applicable.

Competing interests

The authors declare that they have no competing interests.

Author details

¹International Institute for Population Sciences, Mumbai, Maharashtra 400088, India. ²Department of Fertility Studies, International Institute for Population Sciences, Govandi Station Road, Deonar, Mumbai 400088, India.

Akhtar S, Mohanty SK, Singh RR, Sen S (2022) Chronic diseases and productivity loss among middle-aged and elderly in India. BMC Public Health 22(1):2356
Alam M, M. A (2012) Labour market vulnerabilities and health outcomes: Older workers in India. J Popul Ageing 5:241–256
Alavinia S (2008) Unemployment and retirement and ill-health: a cross-sectional analysis across European countries. Burdorf A 82:39–45
Beard JR, Officer A, De Carvalho IA, Sadana R, Pot AM, Michel J-P, Lloyd-Sherlock P, Epping-Jordan JE, Peeters GG, Mahanani WR (2016) The World report on ageing and health: a policy framework for healthy ageing. The Lancet 387(10033):2145–2154
Bloom DE (2013) S.-P. A. Ageing and productivity: Introduction.
Cardano M, Costa G, Demaria M (2004) Social mobility and health in the Turin longitudinal study. Soc Sci Med 58(8):1563–1574
Carlsen K, Oksbjerg Dalton S, Frederiksen K, Diderichsen F, Johansen C (2008) Cancer and the risk for taking early retirement pension: a Danish cohort study. Scand J Public Health 36(2):117–125
Chanda S, Mishra R (2019) Impact of transition in work status and social participation on cognitive performance among elderly in India. BMC Geriatr 19(1):1–10
Coppola M, Wilke CB (2010) How sensitive are subjective retirement expectations to increases in the statutory retirement age? The German case
Deschryvere M (2004) Labour force behaviour of elderly two adult households: Evidence from EU-countries. ETLA Discussion Papers
Disney R, Emmerson C, Wakefield M (2006) Ill health and retirement in Britain: A panel data-based analysis. J Health Econ 25(4):621–649
Focarelli D, Zanghieri P (2005) Labour force participation of older workers in Italy: trends, causes and policy issues. The Geneva Papers on Risk and Insurance-Issues and Practice 30:711–723
IIPS N, MoHFW H (2020) & USC. Longitudinal Ageing Study in India (LASI) Wave 1, 2017–18, India Report. International Institute for Population Sciences, Mumbai: International Institute for Population Sciences, National Programme for Health Care of Elderly, MoHFW, Harvard TH Chan School of Public Health and the University of Southern California
Karpansalo M, Kauhanen J, Lakka TA, Manninen P, Kaplan GA, Salonen JT (2005) Depression and early retirement: prospective population based study in middle aged men. J Epidemiol Community Health 59(1):70–74
Kruse M, Sørensen J, Davidsen M, Gyrd-Hansen D (2009) Short and long-term labour market consequences of coronary heart disease: a register-based follow-up study. Eur J Prev Cardiol 16(3):387–391
Lerner D, Adler DA, Chang H, Lapitsky L, Hood MY, Perissinotto C, Reed J, McLaughlin TJ, Berndt ER, Rogers WH (2004) Unemployment, job retention, and productivity loss among employees with depression. Psychiatric Serv 55(12):1371–1378
Li Ranzi T, d’Errico A, C. G (2013) Association between chronic morbidity and early retirement in Italy. Int Arch Occup Environ Health 86:295–303
Lund T, Villadsen E (2005) Who retires early and why? Determinants of early retirement pension among Danish employees 57–62 years. Eur J Ageing 2:275–280
McDonough P, Amick III, B. C (2001) The social context of health selection: a longitudinal study of health and employment. Soc Sci Med 53(1):135–145
Miah MS W.-G. V. (n.d.). Do the sick retire early? Chronic illness, asset accumulation and early retirement. Appl Econ 39(15), 1921–1936
Mohanty SK, Rodgers J, Singh RR, Mishra RS, Kim R, Khan J, Behera P, Subramanian SV (2021) Morbidity compression or expansion? A temporal analysis of the age at onset of non-communicable diseases in India. Geroscience 43:409–422
Momsen A-MH, Stapelfeldt CM, Nielsen CV, Nielsen MBD, Aust B, Rugulies R, Jensen C (2016) Effects of a randomized controlled intervention trial on return to work and health care utilization after long-term sickness absence. BMC Public Health 16:1–11
Nilsson K, Hydbom AR, Rylander L (2011) Factors influencing the decision to extend working life or retire. Scand J Work Environ Health, 473–480
Perez FP, Perez CA, Chumbiauca MN (2022) Insights into the social determinants of health in older adults. J Biomed Sci Eng 15(11):261
Schofield DJ, Callander EJ, Shrestha RN, Passey ME, Percival R, Kelly SJ (2013) Multiple chronic health conditions and their link with labour force participation and economic status. PLoS ONE, 8(11), e79108
Schuring M, Burdorf L, Kunst A, M. J (2007) The effects of ill health on entering and maintaining paid employment: evidence in European countries. J Epidemiol Community Health 61(7):597–604
Sewdas R, Thorsen SV, Boot CRL, Bjørner JB, Van der Beek AJ (2020) Determinants of voluntary early retirement for older workers with and without chronic diseases: A Danish prospective study. Scand J Public Health 48(2):190–199
Siddiqui S (1997) The impact of health on retirement behaviour: empirical evidence from West Germany. Health Econ 6(4):425–438
Siegrist J, Wahrendorf M (2009) Participation in socially productive activities and quality of life in early old age: findings from SHARE. J Eur Social Policy 19(4):317–326
van den Berg TI, Elders LA, B. A (2010) Influence of health and work on early retirement. J Occup Environ Med, 576–583
Von Bonsdorff ME, Huuhtanen P, Tuomi K, S. J (2010) Predictors of employees’ early retirement intentions: an 11-year longitudinal study. Occup Med 60(2):94–100
Yen L, McRae I, Jeon Y, Essue B, Herath P (2011) The impact of chronic illness on workforce participation and the need for assistance with household tasks and personal care by older Australians. Health Soc Care Commun 19(5):485–494

No competing interests reported.

Association of chronic morbidity and early retirement in India: Evidence from Longitudinal Ageing Study in India

Status:

Version 1

Abstract

Context:

Methods

Findings:

Conclusions

Figures

Introduction

Materials and methods

Data

Health measurement

Covariates

Statistical analysis

Results

Discussion

Conclusion

Declarations

References

Additional Declarations

Status:

Version 1