Introduction
Diabetes mellitus (DM) is a growing health challenge recognised by the World Health Organisation (WHO), the global prevalence of diabetes among adults has risen from 4.7% in 1980 to 8.5% in 2014 (WHO., 2018). So, why has this increase taken place? Research has identified various factors which potentially contribute to the prevalence of diabetes including: obesity, ageing, ethnicity, exercise, diet, genetics, smoking, socioeconomic status, maternal hyperglycaemia (MH), hypertension etc. (Wu et al., 2014). Many of these factors overlap. Hu et al. (2001) found obesity to be the most important predictor of diabetes. It is possible that some risk factors of diabetes only exist due to their role in obesity.
The scope of this essay will discuss the role of obesity, ageing and MH in contribution to increasing DM over the last 20 years. DM often has long-term effects e.g. retinopathy with risk of blindness, nephropathy with risk of renal failure or amputation (WHO., 1999). The cost of diabetes worldwide was reported as $1.31 trillion (Zhang and Gregg., 2017). The ghastly side effects and economic burden on families and healthcare that this condition creates makes identifying its risk factors significant, to successfully developing preventative measures.
Obesity, Exercise & Diet
A twelve-year cross-sectional study was conducted in Turkey to identify risk factors of diabetes (Satman et al., 2013). Prevalence of diabetes in Turkey was 16.5%, compared to 7.2% in a similar study from 1997-1998 (Satman et al., 2002). This highlights the increase in diabetes over the last 20 years. Satman et al. (2013) measured fasting plasma glucose (FPG), those who had a FPG of ≥ 126mg/dL were considered to have diabetes. A questionnaire accompanied the collection of samples to identify confounding factors e.g. medical history, body measurements were also taken. Results showed an association between a one standard deviation (SD) increase in waist size, body mass index (BMI) and a 1.16 and 1.09 relative risk ratio of diabetes in women. In men, one SD increase in BMI was associated with a 1.28 increased risk of new diabetes.
These results show a strong association between being overweight/obese and diabetes prevalence. Although the questionnaire was relevant, answers were subjective so their accuracy is unknown, this could have mislead results. However, this study has many strengths; the large randomly selected sample size (n=26,499) combined with the high response rate makes the study nationally representative. Additionally, collection of data over 12-years allowed for the identification of trends, therefore, any trends concluded were made with confidence. Overall, it can be derived from this study that, in Turkey, the increase in diabetes prevalence is related to the increase in overweight/obese people. During the last 20 years there has been an epidemic in overweight/obese individuals (Hruby and Hu., 2014). The association between obesity and risk of diabetes, combined with their simultaneous increase over the last 20 years, evidences that there may be a causative link between both factors. Yang et al. (2010) also found that diabetes cases increased with the increase in weight amongst Chinese people, indicating that the association between weight and diabetes is a global trend.
As diabetes is characterised as an obesity-related disease, it is important to consider how the factors causing obesity link to causing diabetes. A controlled diet and moderate exercise are required to avoid being overweight/obese (Foster-Schubert et al., 2011). Oggioni et al. (2014) conducted an ecological analysis to detect the factors associated with global diabetes prevalence. Data of estimated diabetes prevalence was obtained from 96 countries. Their results found diabetes prevalence was significantly correlated with physical inactivity (r=0.34, p <0.01), this p-value indicates that low levels of exercise are associated with increased diabetes prevalence. The Middle-Eastern region had the highest prevalence of diabetes (13.1%) and the highest prevalence of physical inactivity (54.2%), further suggesting that both factors are linked. Conversely, the difference in these values emphasises that physical inactivity does not necessarily lead to diabetes. This study identified three dietary patterns (agricultural, transitional and westernised). Results showed an association between the agricultural pattern and a lower prevalence of diabetes. No difference was observed between the westernised and transitional patterns. While three dietary patterns may not represent worldwide food consumption, it was thought diet was not an independent risk factor for diabetes prevalence.
In contrast, Panagiotakos et al. (2005) investigated the association between food groups, insulin resistance and poor glycaemic control in adults without type 2 diabetes (T2DM). Their results showed a positive association between red meat consumption and hyperglycaemia (p=0.04) and hyperinsulinemia (p=0.04) signifying that higher meat consumption may lead to insulin resistance and possibly diabetes. These results were consistent with the findings of Murakami et al. (2005), adding further credibility to their results. However, the study by Panagiotakos et al. (2005) does not provide causal relationships and requires further research to validate the results.
Conflict between which food components are associated with diabetes prevalence has taken place for decades. Marshall et al. 1991 found that a high-fat, low-carbohydrate diet is associated with onset of DM. Whilst, Colditz et al. (1992) found no association between fat intake, carbohydrates and risk of diabetes. Diet is strongly linked to obesity, yet from observing the literature, further research must be conducted to directly link diet and physical inactivity to diabetes. Obesity has been highlighted as an independent risk factor of diabetes prevalence, diet and physical inactivity are risk factors of obesity. Nonetheless, recent research has shown physical inactivity may have a direct impact on risk of diabetes aside from its association through obesity (Tripathy et al., 2017).
Ageing
Advanced age is another risk factor associated with DM (Oggioni et al., 2014). For the first time in history, people from all income countries are likely to live to ≥ 60 years of age (United Nations., 2007). Part of the increased prevalence of DM over the last 20 years may be due to an increase of diagnoses made at older age because of populations living longer.
Hu et al. (2017) collected data amongst adults in Xi’an, China through a self-developed questionnaire and health examination. Advanced age was associated with having a higher rate of diabetic treatment. Alongside this, the population aged ≥ 65 years in Xi’an increased to 8.46% in 2010, supporting the idea that recent ageing has contributed to the increased rate of diabetes. One question that needs to be asked, however, is whether diabetes has been found more prevalent at older age because of the awareness the elder generation has about diabetes and therefore get diagnosed, rather than because of the implications old age brings. A positive association between age and diabetes awareness was found by Hu et al. (2017), this finding was also consistent with previous research conducted by Yu et al. (2013). Both studies are based in China, therefore, trends worldwide need to be considered.
Moreover, Shaw et al. (2010) estimated the number of people with diabetes between the years 2010-2030 by observing 133 studies from 91 countries. Global prevalence of diabetes amongst adults (aged 20-79 years) will increase from 6.4% in 2010 to 7.7% in 2030. Age-specific prevalence estimates were analysed in 5-year intervals. In developing countries, increases in diabetes are expected from each age group but will double in those aged over 60. In developed countries, an increase of 38% is only expected amongst those over the age of 60. These results suggest that ageing of populations is likely to have contributed and will continue to contribute to worldwide diabetes prevalence. The use of 91 countries in this study provides a more accurate representation of the worldwide impacts of ageing on diabetes prevalence. Oggioni et al. (2014) also found age to be an independent risk factor of T2DM diabetes corroborating the study from 2010.
Maternal Hyperglycaemia
T2DM has a strong genetic component (Poulsen et al., 2009). Independent of this, a relation between foetal exposure to MH and increased risk of developing diabetes later in life has been identified. In a study on Pima Indians, babies born to diabetic mothers had up to a 45% risk of developing diabetes, compared to 1.4% for non-diabetic mothers and 8.6% for prediabetic mothers. These results persist even when accounting for paternal diabetes, offspring’s weight/height, and the age parents develop diabetes (Pettitt et al., 1988). The findings suggest that the intrauterine environment is an important factor of diabetes development regardless of genetics. Franks et al. (2006) also found an association between MH during pregnancy and increased risk of diabetes in Pimas offspring. Nonetheless, both studies fail to demonstrate that these findings would translate globally. Pimas are a population known to be at high risk of diabetes. This would suggest there is a genetic factor involved which may affect results. However, the study by Pettitt et al. (1988) accounted for possible genetic influences making their results more translatable to global trends. Besides, it is thought that the metabolic pathways determining T2DM in Pimas, will be common to Native American and non-Native American populations proposing that the results from the studies on Pimas would appear in several populations (Schulz and Chaudhari., 2015). Despite these promising results, more in-depth studies could be done on other populations in this regard.
A meta-analysis by Chu et al. (2007) estimated the risk between maternal obesity and developing gestational DM (GDM), as the magnitude of previous associations was uncertain. Their findings indicated that a high maternal weight was substantially associated with a higher risk of GDM. As discussed earlier, weight gain has increased rapidly over the last 20 years. The fact that weight gain has been associated with a higher risk of maternal diabetes adds to the rise in diabetes itself. However, this association leads to a higher risk of babies born to those mothers developing diabetes at a later age, even if the offspring are not obese/overweight themselves. This further accelerates the rise in diabetes. Therefore, maternal hyperglycaemia is an important determinant in the accelerated rise in diabetes worldwide. Another angle would be to consider the relation between MH and the risk of obesity in childhood. This was studied by Hillier et al. (2007) in a multi-ethnic United States population, their results suggested that increasing hyperglycaemia in pregnancy is associated with an increased risk of childhood obesity. Thus, theoretically the risk of obesity in childhood as a result of MH in the mother could later lead to DM in the affected offspring. Would this be considered as the MH being associated with the diabetes of the offspring or the obesity itself? Further studies which consider these factors need to be undertaken. A strength of the study by Hillier et al. (2007) is that it involved a large (n=10,000) multi-ethnic population, their results were consistent amongst different ethnic groups giving a better representation of global results. This could have been further improved by providing more details on the ethnicities involved.
Other factors contributing to the increase in Diabetes Mellitus
Although studies are consistent in stating that diabetes prevalence has increased over the last 20 years. There may be other contributing factors.
In 1997, the diagnostic criteria for diabetes lowered FPG from 140mg/dL to 126 mg/dL, this could be a potential reason for the increased diabetes incidence over the last 20 years (Geiss et al., 2014). Given that the increase in diabetes began before 1997 and there has been no dramatic shift since, solely the change in diagnostic criteria does not explain the continuing increase.
Despite the health of Sub-Saharan Africa, and low-income Asia remaining below the global average, there has been a slight improvement in the utilisation of healthcare within these populations (Audibert and Mathonnat., 2013). This improvement may have potentially lead to more diagnoses of diabetes being made which previously would have remained undiagnosed. This would not be considered an increased prevalence of diabetes, rather, simply increased diagnoses. Nonetheless, the improvement has not been significant and therefore does not explain such increase in diabetes over the last 20 years.
Awareness can improve the number of undiagnosed diabetic cases. Within the last two decades there have been major technological advances, this led to a large increase in social network users. Social networking has promoted actions of “World Diabetes Day” and generally ameliorated diabetes awareness (Labate., 2013). Hu et al. (2017) found that those with more awareness were more likely to contact health professionals. If more people have been aware and sought medical assessment then this could have reduced the number of undiagnosed cases and marginally contributed to the increase in diabetes.
Though all factors mentioned above have improved over the last 20 years, they are still grossly inadequate and require further improvement, thus, most likely have not significantly impacted statistics on prevalence of diabetes.
Conclusion
Diabetes has increased over the last two decades. Literature has shown that obesity, physical inactivity, ageing and MH are potentially responsible for this. Amongst these, there are several other factors which also contribute to diabetes prevalence, however, these were beyond the scope of this essay. Obesity seems to be the most prevalent risk factor of diabetes, as it links to several other factors mentioned (physical inactivity, diet and MH), advocating the statement of Hu et al. (2001). Recent studies suggest that diabetes prevalence will continue to increase in the foreseeable future (Rowley et al., 2017). Although diabetes care has recently developed, improved medical care in the future may increase lifespan with diabetes and cause less increased prevalence in future especially with regards to ageing.