Diabetes has been found to be one of the largest factors increasing the risk of mortality, morbidity, and disability worldwide. We investigate how diabetes is related to hospital admission and mortality risk among older adults in Europe.

Diabetes has been found to be one of the largest factors increasing the risk of mortality, morbidity, and disability worldwide and its economic burden is a major public health challenge to design new ways to curb diabetes healthcare expenditure (De Lagasnerie et al., 2017). Moreover, diabetes prevalence has been estimated to concern around 380 million people by 2025 (O´Shea et al., 2013), having a greater impact among older people, in whom type 2 diabetes mellitus (T2DM) is more prevalent. This increase in diabetes prevalence in the next years will be accompanied by population ageing, and together with the obesity pandemic, will make diabetes incidence greater than in the past.

The existing literature has already supported the fact that medical costs for patients with diabetes are up to three times higher than costs for patients without the disease (Clarke et al., 2010). Furthermore, poverty is associated not only with higher diabetes incidence, but also with inequality of diabetes care, despite universal health coverage (Hsu et al., 2012). Sortso et al. (2017) demonstrated that, despite the Danish universal health care model, patients with diabetes and with lower and higher education and socioeconomic status showed different healthcare use patterns. In this regard, reducing financial barriers to primary care and pharmacy services may benefit diabetic populations, especially those in worse economic conditions. However, handling health inequalities can become very complicated and will only lead to success when social determinants of health are addressed (Marmot et al., 2012).

Using data from the Survey of Health, Ageing and Retirement in Europe (SHARE), the aim of this study was to investigate how diabetes is related to hospital admission (number of hospitalizations in the previous 12 months and average length of stay per hospital admission too) and mortality risk (and age at death) among older adults in Europe. Moreover, we aimed to assess potential trends in such associations by socioeconomic groups.

By pooling data from five waves of SHARE (waves 1, 2, 4, 5 and 6) and sixteen European countries (Austria, Belgium, Czech Republic, Denmark, Estonia, France, Germany, Greece, Italy, The Netherlands, Poland, Portugal, Slovenia, Spain, Sweden and Switzerland), our results showed that, in the overall sample, diabetes was significantly associated with a higher probability of being admitted to hospital, as well as a higher number of hospital admissions and mean length of stay. However, its effect decreased when clinical and functional complications entered the analysis.

Figure 1: Countries coloured in green were included in the analysis.

When performing the same analysis by socioeconomic group and country (Figure 2), we found that diabetes had a stronger effect among women, people aged 50 to 65 years old, individuals with low and medium education (only if complications were excluded) and medium household income.

When we analysed the effect of diabetes by country group, we only found a statistically significant and positive effect on the odds of hospital admission for Southern and Western countries, being the impact of diabetes greater in case of the former group of countries. A finding that should be highlighted and was homogeneous regardless of the subsample observed is the relevance of functional impairment. When the different functional impairment severity levels were introduced, they emerged as the main predictor of higher probability of being admitted to hospital. �

Figure 2: Diabetes and probability of hospital admission, expressed in percentage points.

In case of mortality risk, diabetes had a significant and positive effect on the probability of death for the overall sample, but not on the age at death. As it was observed for the other outcome of interest, the effect of diabetes on mortality decreased when covariates were included, especially when functional impairment was introduced into the model. By subgroups, diabetes had a significant effect on the risk of death in every subsample (Table 1). However, the impact was greater among men than women, in the oldest old group (above 80 years old), in low educated individuals, among people with medium household income and in Southern countries.

Our results showed that diabetes was significantly related to both outcomes, the risk of death and hospital admission, with significant differences across socioeconomic groups and countries. Higher odds of hospital admission were found in women with diabetes, individuals between 50 to 65 years old or people with medium educational level and household income. On the other hand, males, older people, lower educated people and individuals with medium household income were found to be associated with a greater mortality risk. We have also proved that the effect of diabetes on hospital admission and risk of death is influenced by clinical and, especially, by functional complications.

The consistency of our findings across European countries would be of great relevance for governments and policymakers to be aware of diabetes burden on health outcomes among older adults. Special attention should be paid to people in advancing age and with low socioeconomic status (education and income, according to our results). Measurement and understanding of socioeconomic inequalities in health and healthcare are critical for achieving higher equity in healthcare (Sortso et al., 2016; 2017). Hence, measures of socioeconomic status may be valuable in making valid comparisons of the quality of diabetes care. However, handling health inequalities can become very complicated and will only lead to success when social determinants of health are addressed (Marmot et al., 2012). Moreover, comorbidity should not be neglected when assessing diabetes impact, as we have shown that clinical complications play a key role lowering potential bias, but functional impairment should not be excluded either.

Table 1: Results of the effect of diabetes from the random-effects logit model for having died. Reference categories: No diabetes, other marital status different from married, unemployed or homemaker, no education, normal weight, no functional impairment, wave 1, Germany.

About the authors:

Beatriz Rodríguez-Sánchez, PhD, Department of Economic Analysis, University of Castilla la Mancha, Toledo, Spain

David Cantarero-Prieto, PhD Department of Economics and GEN, University of Cantabria, Santander, Spain

The article is based on:

Beatriz Rodríguez-Sánchez and David Cantarero-Prieto: Socioeconomic differences in the associations between diabetes and hospital admission and mortality among older adults in Europe, Economics and Human Biology (2019),

Beatriz Rodríguez-Sánchez and David Cantarero-Prieto 2019-05-11T20:37:32+02:00
14 May 2019

Leave A Comment