Preventing Diabetes: Early Versus Late Preventive Interventions

Primary prevention of chronic noncommunicable diseases such as type 2 diabetes (T2D) must be based upon the alteration of the natural history of the disease by influencing known modifiable risk factors. A large number of risk factors for T2D are well known: some operating through insulin resistance, some through insulin secretion, and some through both. Many prevention strategies can be implemented early on in the disease’s life, while some strategies can only be introduced at a later stage. Controlled lifestyle intervention trials have provided convincing evidence that T2D can be prevented in different populations and cultural settings (1–4). The main objective of interventions applied in such “proof-of-concept” studies was to correct unhealthy lifestyle patterns among the study participants by providing individual and/or group counseling. The implementation of the interventions varied among these studies. The target groups in these intervention trials were people at a high risk of T2D. There are, however, many ways to define “high risk.” Although the high-risk approach is certainly a suitable strategy for late interventions to prevent T2D, interventions may also be applied early for some high-risk target groups. The high-risk strategy is a typical approach to be carried out within the health sector, but it can also be implemented by stakeholders in other sectors, e.g., sports, nutrition, and education.

There is controversy about the right time to start interventions for T2D prevention from the medical perspective. Risk factor modification can also be done through the population approach, in which the entire population, rather than a high-risk group alone, will be the target (5). The population approach is well suited for early interventions aiming at promoting healthy diet and physical activity at any age. Since health promotion interventions are safe and can also result in prevention of other noncommunicable diseases, the population approach can be the most desired way to implement early intervention for T2D prevention. In addition to the participation of people in the health sector, the population approach requires the participation of other stakeholders from sectors such as taxation, city planning, education, agriculture, etc.

The high risk of developing T2D can be identified in different ways. For early and late interventions, the following risk stratification may be used.

It has been known that characteristics of fetal growth retardation are associated with an increased risk of T2D during adult life (6,7). In addition, babies born small who experience the so-called catch-up growth during early infancy, which leads to a rapid increase in body weight (and thus fat mass), are at particularly high risk of T2D (8). Also, the probability of developing T2D seems to be higher in children who were born small but are the heaviest during prepubertal years. In principle, such information could be used to initiate early prevention of T2D. However, intervention studies in which children and adolescents at high risk based on their characteristics at birth and infancy are selected for the study do not exist.

At present, >80 genetic T2D susceptibility loci have been identified, but each has only a small effect on T2D risk (8). On the basis of these loci, it is possible to generate a genetic risk score to predict the development of T2D. Such a score can be applied at any age, even at birth or in childhood, and thus may offer an opportunity to select high-risk individuals for early interventions to prevent the development of the disease (9,10). Such intervention studies initiated from a genetic risk assessment have not been conducted thus far. Post hoc analyses of the Finnish Diabetes Prevention Study (DPS) and the U.S. Diabetes Prevention Program (DPP) applying the genotype data of selected known susceptibility loci have unequivocally demonstrated that carriers of high-risk alleles benefit significantly from lifestyle intervention (11–15). The DPS results also showed both that people with a high and low genetic risk score benefited from lifestyle intervention and that the intervention was beneficial in both positive– and negative–family history groups (15).

For prediction of T2D, several multivariable models that combine risk factor profiles and sometimes also measures of glucose disturbances with other biochemical or clinical variables have been published (16). Such algorithms are appropriate for etiologic investigations to search for underlying causes of T2D. However, due to their complexity they may not be practical for public health screening efforts aimed at identifying individuals at a high risk of T2D or other forms of glucose disturbances (16).

During the past decade, there has been a lot of interest on the development and validation of simple risk scores for T2D, based on either nonlaboratory parameters alone or combining such information with biochemical parameters (17–21). With use of a T2D risk score, it is possible to identify people at high risk who can then be invited to join lifestyle intervention programs. This approach has been used in the follow-up of the Finnish National Diabetes Prevention Program (FIN-D2D) (22) and the European DE-PLAN (Diabetes in Europe–Prevention Using Lifestyle, Physical Activity and Nutritional Intervention) project (23).

Gestational diabetes mellitus (GDM) is defined as any degree of glucose intolerance with signs or first recognition occurring during pregnancy (24). After delivery, glucose levels return to the nondiabetes range. The prevalence of GDM may range from 1 to 14% of all pregnancies depending on the population studied and diagnostic criteria used in defining GDM (25). Women with GDM are likely to develop impaired glucose tolerance (IGT) or T2D during the postnatal period (26) or later in life (27–29). Subgroup analyses among women with previous GDM participating in the large lifestyle intervention and pharmacological T2D prevention trials have yielded substantial reductions in the risk of T2D, making such relatively young women an attractive clientele for early intervention to prevent the development of T2D. The Tianjin Gestational Diabetes Mellitus Prevention Program (29) is currently testing the efficacy of the prevention of T2D in women with a history of GDM in China. Another ongoing study is the GIFTS (Genomic and Lifestyle Predictors of Foetal Outcome Relevant to Diabetes and Obesity and Their Relevance to Prevention Strategies in South Asian Peoples) project in Bangladesh, Pakistan, and India (30). GDM is a pathophysiological state identified in an increasing number of young and middle-aged women that can be a starting point for early interventions to improve insulin sensitivity and thereby prevent the development of T2D.

Several studies have taken the approach of only inviting individuals with IGT and/or impaired fasting glucose (IFG) to interventions (1–4,31). These people, by definition, have significant pathophysiological disturbances and thus should be considered targets for later interventions. Thus far, no clinical trial evidence exists to show that the development of T2D can be prevented by lifestyle intervention in people with isolated IFG. There is only one lifestyle intervention trial that recruited people with IFG (31). This trial showed a 59% relative risk reduction of T2D in people who had IFG combined with IGT, an effect similar to that in other trials in people with IGT (1,2), but no beneficial effect in people with isolated IFG. Reasons for the lack of response to lifestyle intervention in people with isolated IFG are not known, but what is known is that pathophysiology between IFG and IGT is different (32). This question needs to be addressed in future studies. The American Diabetes Association is recommending lifestyle intervention and even metformin for people with IFG and “high-risk HbA_1c” (5.7–6.4%) (33) and states that this recommendation is based on “expert consensus.” Healthy lifestyle can be recommended to everyone, but without evidence that the intervention provided was efficient, the use of health care resources may be difficult to justify. Thus far, no T2D prevention studies have been carried out specifically recruiting people with HbA_1c <6.5%.

During aging, several pathophysiological processes can promote the development of T2D. In particular, the pancreatic β-cell mass and insulin production capacity decrease, muscles develop sarcopenia, and physical fitness reduces. As is typical for endocrine cells, the β-cell loss with aging may not be reversible, although the functioning of existing β-cells might be improvable (34). A progressive loss of muscle mass and strength with aging called “sarcopenia” has a complex etiology involving neuronal, hormonal, immunological, nutritional, and physical activity mechanisms (35,36) and is associated with an increased fat mass infiltration. Sarcopenia is associated with worsening insulin sensitivity (37), and there is a strong inverse association between insulin resistance and relative muscle mass (38). The results from T2D prevention trials in people with IGT in both DPS and DPP were most pronounced among the oldest people (34). Thus, late interventions targeted to elderly people at risk for T2D are warranted.

Weight gain in adults is associated with increased T2D risk. In the Health Professionals Follow-Up Study (HPFS), in people aged 40–75 years at baseline, a weight gain of >10–15 kg was associated with a significantly increased risk of T2D and also all-cause mortality (39). These results are in agreement with those from other studies that have also detected some ethnicity-specific differences in this association (40–44). A working group of experts in the fields of pathophysiology, genetics, clinical trials, and clinical care of obesity and/or T2D prepared a summary and made recommendations based on published literature and their own data (45) at a conference in 2011. The impact of obesity on T2D risk is determined not only by BMI but also by the location where fat accumulates in the body. Increased upper-body fat, indicating visceral adiposity, reflected as increased waist size or waist-to-hip ratio, is strongly associated with T2D (46), although the underlying mechanisms remain uncertain. Whether subcutaneous fat lacks the pathological effects of visceral fat or is simply a more neutral storage location remains unclear (47). Visceral obesity especially and, probably even more so, the amount of liver fat define the pathophysiological background for the causal role of obesity in the development of T2D (35). The difference between obese people at risk for T2D and those who are obese but do not develop diabetes risk is probably due to body fat composition, including visceral and hepatic fat. Overweight/obesity is an underlying risk factor for T2D with epidemic dimensions, but weight alone is not a sufficient indicator or sole target for diabetes prevention. Visceral obesity and a high amount of hepatic fat can be directly translated into increased T2D risk, highlighting the relevance of waist circumference as a surrogate for increased T2D risk (36,37). While obesity prevention obviously deals with early intervention to prevent T2D, too, reduction of obesity is one of the most common late T2D prevention strategies.

Comparing different intervention strategies to prevent T2D reveals differences in characteristics such as pathophysiology, personalized profiling, acceptance, and biological and psychological adverse effects, as well as costs. The main intervention strategies comprise physical activity, advice on eating behavior, weight reduction, and behavioral goal setting. For each of these characteristics and strategies, there are some differences between when the interventions started early and when they started late.

On a pathophysiological level, the interplay between muscle, fat tissue, gut, and brain is of relevance (35). By secreting adipokines and hepatokines, increased visceral and hepatic fat accumulation is an important driver for an increased T2D risk (38). People with a high amount of visceral and liver fat may profit less from lifestyle intervention and may thus require intensified lifestyle prevention strategies or even pharmacological approaches to improve insulin sensitivity (35). Increased physical activity improves muscle mass and reduces fat mass, resulting in improved insulin sensitivity. A recent meta-analysis provides strong evidence for an inverse relationship between physical activity and risk of T2D, which may partly be mediated by reducing adiposity (48). Daily physical activity is a central element of early intervention for T2D prevention. Among strategies focusing on eating behavior, reduction of saturated fatty acids and total fat consumption together with an increased fiber consumption and reduction of sugar-sweetened beverages are the most important approaches for preventing T2D (34). Consuming high amounts of fat and alcohol leads to the accumulation of visceral fat and boosters the development of fatty liver disease. Consuming soft drinks with artificial sweeteners may influence gut microbiota and thus increase glucose intolerance (49). Changes in dietary habits are effective in preventing T2D if started early and performed in a sustained manner. For the short-term, interventions focusing on weight reduction can influence T2D risks effectively. They are also preferable for later interventions, as mathematic models show (39). The preferred weight-reduction strategies in T2D prevention are those that effectively reduce visceral and hepatic fat (40).

It is challenging to educate people on healthy eating behavior, especially those with a low health literacy (41). Finding the intervention that has the highest probability for success in preventing T2D in an individual is one of the major challenges. In design of interventions for T2D prevention, following a behavior change model improves the success rate (3,42). In the past, the application of educational programs was mostly driven by disease characteristics. Developing profiles, including individual dimensions such as the degree of motivation and personal preferences, would increase the probability of the success (43). For instance, SweetSmart is a concept that provides an assessment to identify the most appropriate intervention applicable to a personal profile and is applicable to both early and late interventions (43).

Personal profiling also means the efficient identification of people who are at a high T2D risk. There are several T2D risk prediction tools that include information on behavioral characteristics, biomarkers, anthropometric markers, and genetic markers (34,44). Compared with clinical risk factors alone, common genetic variants associated with the risk of T2D have only a minor effect on the ability to predict the development of T2D (45). A combination of clinical, anthropometric, and behavioral parameters offers a better prediction of the future T2D risk (46,47). Emerging information regarding metabolomics, lipidomics, and proteomics may help in the future to elucidate T2D risk increasingly in a personalized manner (50,51). A recent study showed that history of daily physical activity was a robust proxy indicator of the risk of chronic diseases today (52). Most personal risk profiles are associated with clinically relevant risk factors and are therefore most relevant for late interventions. Goal-setting strategies addressing behavioral stigmata are applicable throughout the whole diabetes risk journey.

Today, we live in a toxic food environment, which provides us with energy-dense and cheap food 24 h a day and exposes us to aggressive marketing campaigns that promote the consumption of unhealthy food items (53). We are all consumers and influenced by our peer environment marketing, our own experiences, and educational level. These factors are driven by cultural, religious, ethnic, and social aspects, all of which influence acceptance and adherence and also the efficacy of T2D prevention. Our eating behavior is reflected in our daily food choices, where our cognitive knowledge on healthy eating competes with the emotional arguments of marketing campaigns of food industry. Such marketing campaigns often twist unhealthy food items into images of healthy lifestyle. There might be an increased acceptance of a healthy eating behavior by applying strategies such as higher taxes on unhealthy food items, plain packaging, intuitive food labeling, or a policy of liability of food and beverage companies for adverse health events associated with the use of their products. These strategies would help the consumer to make healthier choices and trigger the acceptance of healthy nutritional behavioral strategies (54). Goal-setting strategies are by intention—individual strategies. Goal-setting strategies can help people at high risk of T2D accept and adhere to healthier lifestyles, and they can provide a more useful approach for sustainable T2D prevention (44).

The intensity of sports and physical activity correlates directly with a probability of adverse effects: injuries, organ failure, overexhaustion, and psychological resistance. Some studies show that the higher the disease risk and the better the expected intervention effects, the more people are willing to accept and endure potential side effects (55). Daily walking or similar leisure-time activities may result in fewer metabolic effects in the short-term, but these activities have only minimal side effects and show a positive dose response for the prevention of T2D in the long-term (48). Walking 5 and 7 h per week is relatively easy to achieve and can be applied as a part of both early and late intervention for diabetes prevention.

In some prevention studies, depression scores increased a little in some of the participants (56). This effect was related to the feeling of nonadherence to nutritional intervention, especially in weight-reduction programs. Changing dietary behavior requires a supportive environment, as well as food and nutrition policies and strategies in the community (54). Behavioral goal-setting strategies are focusing on the assessment of individual motivation with a model of understanding the disease risk, developing an action plan, and developing routines to maintain the desired effect (44). Such strategies translate into a sustained behavior change and are best when applied as early as possible. Relapse can happen and needs to be considered to be part of the iterative goal setting and not as an adverse effect (44).

The cost-effectiveness analysis of T2D prevention programs has shown that the prevention of T2D is cost-efficient (57). The long-term follow-up data show even greater benefits, since beneficial effects of lifestyle intervention on T2D risk reduction seem to remain for many years after the intervention program is over (4). The dilemma is that the different stakeholders (individuals, employers, health plan or insurance, state and society, etc.) often have different perspectives on the concept of cost efficacy (58). The most direct cost benefit from T2D prevention can be seen in the occupational health setting, in which the employers invest in prevention programs for their employees in order to keep them healthy and active in the workforce (59). It is not clear which intervention can generate the greatest benefit with least cost. It seems that early T2D prevention programs are individually more cost-efficient, since the magnitude of change in lifestyle habits may be easier to achieve than during later stages in the natural history of T2D (Table 1).

We have obtained a substantial amount of scientific evidence on T2D prevention and developed prevention practice recommendations that can be used to define which interventions are effective among people at high risk (40). A sustainable implementation of intervention programs to reach millions of people at risk does not require more evidence from medical research work. Rather, it requires policy and entrepreneur-like efforts (60).

The perspective in T2D prevention should be in the implementation of scalable prevention programs. In particular, it should be in the development of sustainable business models for T2D prevention. This can be reached not only by building a policy frame to encourage the development of T2D prevention but also by developing policies, including tax on unhealthy foods, food labeling, and liability for adverse health effects on the food and beverage industry (61). Additionally, reducing taxes for entrepreneurs to build up business models in T2D prevention would foster implementation. Tax exemption for noncommunicable disease prevention initiatives can be bound to reach a certain number of individuals or a specified health effect with a documented quality. Another implementation strategy can be applied in the occupational health care sector. In some countries, corporations deduct the amount of money they spend investing in their employees’ health from income tax. This can become very cost-efficient if the investment in employees’ health promotion reduces absenteeism at the workplace. The participation and uptake of prevention activities are, in most of the cases, very individual personal decisions. If interventions do not match individual preferences and needs, they will not succeed. We have to learn from the marketing strategies of the big food companies, as well as accept that we have to sell and market our prevention strategy better than they do, in order to reach our clients who are at risk for developing T2D (53).

There are a number of arguments favoring early activities in the prevention of T2D, as well as concepts and strategies of interventions that suit best for later stages (34). T2D prevention is a success story if implemented in a sustainable manner. Sustainability within a T2D prevention program is more important than the actual point in time or stage in the natural history in the progression to T2D at which prevention activities may start. The quality of intervention, as well as its intensity, will vary with the degree of the observed risk. Interventions to prevent T2D should start as early as possible to allow a wide variety of low- and moderate-intensity programs (40). However, the results from the major T2D prevention trials have shown that the best relative risk reduction in T2D incidence was obtained in older participants (1,2) and in those who had the highest estimated T2D risk (62).

Public health interventions for T2D prevention represent an optimal model for early intervention. The later the people at high risk are identified, the more aggressive and more adapted to the pathophysiological stage of T2D development the intervention should be. This may also include the use of pharmacological agents for interventions. Furthermore, sustainability of the intervention should be combined with effectiveness, and this requires quality management. All intervention programs should include an embedded quality-management strategy to assure the timely and efficient delivery of interventions. For the implementation of a variety of T2D prevention programs, new concepts are needed to achieve a sustainable program delivery. Business planning in T2D prevention, together with environmental public health strategy, including all public health and political tools, is needed in order to build a friendly environment for T2D prevention. This may also include the use of T2D prevention as a model for targeting growing risks for other chronic diseases and developing prevention strategies for noncommunicable diseases in general. Early versus late prevention may be distinguished by different characteristics of intervention strategies, but all are following one goal: to develop sustainable prevention approaches in order to halt the pathophysiological process through which T2D is developing.

Funding. This article was partly supported by grants from the European Commission FP7 (grant agreement numbers Health-2011-F2-279074 and Health-2011-278917).

Duality of Interest. No potential conflicts of interest relevant to this article were reported.