Socio-economic inequalities in C-reactive protein levels: Evidence from longitudinal studies in England and Indonesia

Highlights

• Higher education attainment and wealth were linked to lower C-reactive protein levels in England.

• Socio-economic status had no significant association with C-reactive protein in Indonesia.

• Engaging in physical activities is related to lower CRP levels in both countries.

Abstract

Objective

The levels of health enjoyed by individuals vary according to socio-economic status, and inflammation has been proposed as one pathway through which socio-economic status influences health in high-income countries. However, little is known regarding the social determinants of inflammation in low- and middle-income countries. This study investigated the association between socio-economic status and C-reactive protein (CRP) in England and Indonesia.

Methods

This study used data from 4923 respondents aged 50+ in the English Longitudinal Study of Ageing and from 1812 respondents from the Indonesian Family Life Survey at baseline. The sources of information on individuals’ education attainment, wealth, CRP and other covariates were ELSA Waves 2, 4 and 6, and IFLS Waves 4 and 5. Linear mixed models were performed to identify the longitudinal relationships between socio-economic status and CRP, and joint models were used to deal with bias in longitudinal ageing surveys due to attrition.

Results

Respondents with high school education (β = −0.036, p < 0.01) or college degree or higher (β = −0.127, p < 0.01) in England had lower CRP levels. Being wealthy is correlated with lower CRP levels in England (middle: β = −0.073, p < 0.01; wealthiest: β = −0.133, p < 0.01). Being wealthy and having higher education were related with higher level of CRP in Indonesia. However, those significant relationships vanished when health behaviour and health status were included in the analysis.

Conclusions

The relationships between socio-economic status and CRP proved to be different in England and Indonesia. Encouraging healthy lifestyles among older adults with low socio-economic status may be the main focus of efforts toward healthy ageing in England and other developed countries. However, in Indonesia and other developing countries, attention is also required to older adults in wealthier socio-economic groups.

Introduction

Despite considerable efforts to alleviate health inequalities in the past three decades (Murray et al., 1999), poor health remains associated with lower socio-economic status (SES) both in high-income and in low- and middle-income countries (LMICs) (Gwatkin et al., 2007, Pickett and Wilkinson, 2015). Lower SES has been associated with higher morbidity (Clark et al., 2009) and mortality due to cardiovascular diseases (Mackenbach et al., 2000, Méjean et al., 2013). The literature has proposed inflammation as a physiological process linking SES to cardiovascular risk (Gémes et al., 2008, Pollitt et al., 2008, Emerging Risk Factors Collaboration, 2010). C-reactive protein (CRP), a protein produced by the liver, is commonly used as a clinical marker of inflammation (Sproston and Ashworth, 2018). An elevated CRP level is a strong independent predictor of first cardiovascular events (Ridker et al., 2002), cardiovascular death (Emerging Risk Factors Collaboration, 2010), stroke, and peripheral vascular disease, suggesting that it plays an active role in the pathophysiology of cardiovascular diseases (Blake and Ridker, 2002).

A link between socio-economic status and CRP levels has been established in high-income countries (Alley et al., 2006, Gimeno et al., 2007, Deverts et al., 2012, Davillas et al., 2017). Lower CRP levels have been found among those with higher educational attainment (Panagiotakos et al., 2004, Loucks et al., 2006) and household income (Jousilahti et al., 2003). A longitudinal study with a thirteen-year follow-up found that higher educational attainment and family income were associated with lower increases in CRP in the US (Deverts et al., 2012). Using data from the NHANES IV, Alley et al. (2006) showed that the relationship between SES and CRP is evident only at very high CRP levels (>10.0 mg/L). The literature has proposed several possible mechanisms of the association between SES and inflammation. Firstly, individuals of lower SES are more likely to engage in unhealthy behaviours such as smoking, heavy drinking and physical inactivity (Contoyannis and Jones, 2004, Hiscock et al., 2012), which may lead to higher levels of inflammatory markers (Fröhlich et al., 2003, Colbert et al., 2004). Secondly, individuals in socio-economically disadvantaged groups are more susceptible to both infection and chronic diseases (Dalstra et al. 2005; Krieger et al., 2003), and inflammatory processes are thought to be involved in the development of those diseases. Finally, stress-mediated factors and psychosocial processes might account for the link between low SES and higher levels of inflammatory markers. The capacity of a synthetic glucocorticoid hormone to suppress IL-6 production is significantly reduced in individuals with chronic psychological stress (Miller et al. 2002). Thus inflammation could be a common biological process explaining the social inequalities in health observed in many populations.

To date, research has not addressed socio-economic differences in inflammation levels in LMICs (Nazmi and Victora, 2007). This is a serious omission given the high prevalence of cardiovascular diseases in those countries. At present, LMICs contribute a greater share to the global burden of cardiovascular diseases than high-income countries (Gersh, 2012, Naghavi et al., 2017). By 2030, cardiovascular diseases are expected to represent the third leading cause of death in middle-income countries (Mathers and Loncar, 2006). Low SES is one of the most important cardiovascular risk factors in LMICs, including China (Yu et al., 2000), India (Gupta and Gupta, 2009), and Indonesia (Kisjanto et al., 2005). This study thus aimed to identify the consequences of SES on inflammation over time among older adults in the contexts of high-income countries and LMICs using longitudinal data from England and Indonesia. Few studies have incorporated SES gradients in longitudinal patterns of inflammation. However, the studies performed have rarely addressed the pervasive issue of attrition in longitudinal studies (Hardy et al., 2009). Attrition may lead to bias in analysis as respondents are prone to selective dropout due to poor health or death. This study addresses the attrition problem by jointly modelling trajectories of inflammation with attrition when estimating the effect of SES and other covariates (Graham et al., 2011). This study contributes to the existing literature in a number of ways. It is one of the first to draw longitudinal inflammation trajectories according to SES based on national representative surveys in England and Indonesia in a similar time frame. It further has uncovered the contradictory findings on the link between SES and inflammation in England and Indonesia. Lastly, it used joint models to address the issue of attrition in longitudinal surveys.