Diet is one of the most influential risk factors for metabolic, cardiovascular and several other disease groups. Optimizing nutrition is therefore of enormous importance in prevention and therapy. For the traditional Mediterranean diet and the low-fat concept, there is epidemiological and, above all, interventional evidence for the improvement of all metabolic axes, the lowering of the body fat percentage and the reduction of relevant long-term risks such as type 2 diabetes, coronary heart disease, stroke and certain cancers.
You can take the CME test in our learning platform after recommended review of the materials. Please click on the following button:
Diet is one of the most influential risk factors for metabolic, cardiovascular and several other disease groups. Optimizing nutrition is therefore of enormous importance in prevention and therapy. For the traditional Mediterranean diet and the low-fat concept, there is epidemiological and, above all, interventional evidence for the improvement of all metabolic axes, the lowering of the body fat percentage and the reduction of relevant long-term risks such as type 2 diabetes, coronary heart disease, stroke and certain cancers [1,2]. Low-carb or vegetarian-vegan diets are on a par with or even superior to the aforementioned concepts in certain metabolic axes, but have been too poorly studied in others. There is also no long-term data on low-carb or vegetarian-vegan diets from randomized controlled trials (RCTs). The effectiveness of low-GI, intermittent fasting and other concepts for surrogate metabolic parameters is in the middle range or unclear due to a lack of sufficient studies.
Personalization strategies
With the emergence of cluster subtypes for prediabetes and type 2 diabetes, our knowledge of individual pathogenesis and thus also of individual treatment needs is becoming increasingly refined [3,4]. While certain individual characteristics such as body mass index (BMI), blood glucose and patient age allow a prediction of metabolic success, at least in some studies, such a prediction based on the cluster subtypes is not yet ready for application despite their pathomechanistic concept. This is because all approaches to the personalization of nutritional therapies share one hurdle: Before the effectiveness of a dietary change in RCT subgroups can be taken as a basis, actual compliance with the therapy must also be ensured in these subgroups. However, this compliance is difficult to define, difficult to measure and therefore difficult to use as a predictor of treatment success.
Compliance – how to define, how to measure?
Compliance generally refers to adherence to the therapeutically prescribed guidelines, i.e. pharmacotherapy, noxious substances or specific nutritional treatment. Non-compliance, i.e. the deliberate discontinuation of treatment by patients, is desirable if a therapy is not effective, cannot be implemented or even has unacceptable side effects.
Nevertheless, it is difficult to measure compliance precisely. With drug treatment (including food supplements) there is “only” the possibility of overtreatment (the patient takes too much active ingredient) or undertreatment (too little active ingredient). However, lifestyle measures are usually based on several approaches at the same time: diet and exercise; exercise intensity and quality, diet quantity, pattern and frequency. Individual elements can even hinder each other in terms of feasibility. Thus, severe calorie restriction with conventional foods reduces the chance of eating “low fat”, as the few remaining calories may only contain unrealistically small amounts of dietary fat. It is difficult to fall below a hypocaloric limit of e.g. 30 kcal%. It is also unlikely that you will be able to eat a high protein diet without losing weight. Complete compliance with all instructions given is hardly to be expected with complex nutritional therapies.
Another dilemma concerns the quantification of compliance. While pharmacotherapy provides an objective marker for treatment compliance by determining the level of the active ingredient, only a few foods and nutrients offer comparable biomarkers: alkylresorcinols for whole grains, methylhistidine for red meat, essential fatty acids for their food sources and some other metabolites that are only determined in studies [5]. However, these measurement methods are not established in clinical practice and are too complex and expensive even for most research projects. Subjective recording through food logs or food frequency questionnaires are more common, but also more prone to errors – from over- and underreporting (incorrect documentation) to over- and undereating (i.e. the distortion of eating behavior through logging) [6].
Incidentally, weight loss is not an ideal compliance parameter. Although many patients and therapists strive to lose weight, not all patients can or should lose a significant amount of weight. The ideal weight is higher for older patients; people who start therapy at an ideal weight or underweight should not lose weight at all. Body weight and BMI do not specifically reflect the targeted reduction of (visceral) adipose tissue; weight loss can also reflect cachexia or sarcopenia as a result of well-intentioned nutritional therapy.
In clinical studies, there is another parameter that indicates how stressful a therapeutic intervention is: the drop-out rate. The more intensive the therapy, the more likely patients are to discontinue treatment. In individual cases, of course, this parameter does not help to predict treatment success, but in larger cohorts it could be a good surrogate parameter of compliance, since precise treatment adherence (to calorie balance, nutrient requirements and other aspects) is not or only insufficiently reported in many nutritional intervention studies. However, the number of treatment discontinuations is published relatively reliably.
How high is compliance on average?
Compliance with diets cannot be deduced from observational studies; all the people included in the study show a lifestyle pattern that is free of systematic guidelines, i.e. is primarily based on individual preferences, religious rules, personal acceptance and therapeutic recommendations received in individual cases. Thus, a small proportion of “voluntary” vegans in a cohort study – mostly recruited decades ago – in no way proves a low acceptance of this form of nutrition if it were to be prescribed as a standard treatment to a group of non-vegans today.
The average compliance with dietary therapies can therefore only be estimated from RCTs and we do not have a sufficient number of such studies for all dietary approaches. Long-term intervention studies are rarities in nutrition research anyway, but even for shorter periods the evidence is variable. For low-carb, low-fat, high-protein, Mediterranean diet, low-GI, vegetarian-vegan diet and intermittent fasting, rough estimates can be determined from studies with a study duration of up to 6 months – based on the drop-out rate presented above in these publications. 6 months is a favorable threshold, because by this time the compliance of most subjects has already fallen to a relatively stable plateau [7] (Fig. 1).
As a general trend, it can be seen that studies on carbohydrate-restrictive methods have higher drop-out rates than studies that use dietary guidelines without a focus on carbohydrate quantity (Table 1). However, this does not mean that “low carb” is more difficult to implement. Studies on “low carb” more frequently include (older) patients with type 2 diabetes, the average duration of the study is longer, and the average age and gender distribution also vary greatly between all studies and the dietary methods tested in them. There are many reasons for dropping out: Intolerance or side effects, lack of dietary success, lack of variety on the plate, individual stress in the private environment, new illnesses and many more. However, the main factor is likely to be obstacles to compliance.
Can compliance (diet-specific) be predicted?
Even the drop-out rate therefore collects a potpourri of drop-out scenarios, only some of which could contribute to the predictability of compliance. Random events and rare complications can always influence the persistence of a therapy. Factors of the dietary intervention itself (duration, intensity, guidelines), the target group (age, BMI, gender) and the further therapeutic setting (offers of help, financial support, supplementary therapeutic guidelines or options) affect all patients, are documented in many studies and therefore allow a systematic analysis of their influence on compliance.
RCTs on low-carb (compared to low-fat) show in a comprehensive statistical analysis from 2018 that diet duration and diet severity correlate perfectly plausibly with the drop-out rate. In addition, RCTs with younger participants and those with particularly obese patients have particularly high dropout rates. In addition, too frequent monitoring of treatment adherence through dietary protocols appears to have a deterrent effect and is associated with a higher drop-out rate. The fact that neither the proportion of diabetes patients nor smokers has a statistical influence strengthens the significance of the drop-out rate as a compliance marker, which does not represent health-related reasons for dropping out (e.g. diabetes complications or similar), but primarily a behavioral pattern [8].
There are also some RCTs on vegetarian-vegan diets that are accessible for such an evaluation. Once again, RCTs with a longer study duration show a higher drop-out rate, and stricter diets perform worse. Here, too, too-close nutritional monitoring is a hindrance. In contrast to “low carb/low fat”, older patient age and higher BMI are not significant influencing factors that make treatment adherence less likely. Studies with a higher proportion of smokers and those with additional requirements for physical activity had higher dropout rates. Veganism and vegetarianism scored similarly [9].
In the analyses of low-carb/low-fat and vegetarian/vegan diets, studies with a high proportion of participants with pre-existing conditions (e.g. type 2 diabetes) did not show an increased dropout rate. The gender distribution also had no influence [8,9]. Similar analyses are possible for all other diets (from formula to Mediterranean, from low-GI to high-protein), but have not yet been published.
Outlook
The determination and – ideally – the prediction of compliance are essential in order to adequately design nutritional therapies for clinical trials and clinical routine and to assess their prospects of success. The desired personalization of therapies in diabetology also requires a precise statement on compliance. However, defining and measuring compliance is complex and technically difficult, especially with objective parameters. Only RCTs in which personal preferences and other factors do not play a role in the allocation of the respective dietary form (although they do play a role in the willingness to participate in such a study at all and possibly to be allocated an unattractive diet) can be used as the basis for the study. For many reasons, RCTs are needed in nutrition research in larger numbers, for longer durations and with broader recruitment; actual compliance must also be better recorded and published in future.
In today’s clinical reality, the personalization of metabolic nutrition therapies therefore continues to rest on three pillars. First: trial-and-error on the therapeutic side, whereby after the possible failure of the most promising diet, the next diet candidate follows. Secondly, the patient’s individual conviction that the therapy is helping. Health belief is a decisive factor in whether guidelines are initially implemented or immediately rejected. Thirdly, patients pre-select which diet is acceptable on the basis of ethical, religious and social factors. However, the low household income of many patients with metabolic diseases limits the use of all evidence-based nutritional therapies. A healthy diet, in whatever form, is not affordable for large sections of the population in Western countries [10].
Take-Home-Messages
- Nutritional therapies have a wide range of preventive and curative effects, probably particularly strong in specific patient groups or disease subtypes.
- A prerequisite for effectiveness is good compliance, i.e. the patient’s adherence to all elements of the dietary guidelines.
- The diet-specific definition of compliance and non-compliance is difficult, and their measurement is often only possible using subjective parameters.
- In addition to diet intensity and duration, patient factors (age, BMI, etc.) presumably play a role, and the strictness of compliance monitoring also influences the willingness to adhere to the diet.
- The generally insufficient data on the effectiveness of nutritional therapies from RCTs also extends to the issue of compliance; more numerous, larger and longer intervention studies are needed.
CoI: Stefan Kabisch has received funding from the German Center for Diabetes Research (DZD), the German Diabetes Society, the Almond Board of California, the California Walnut Commission, the Wilhelm Doerenkamp Foundation, J. Rettenmaier & Söhne and Beneo Südzucker, as well as personal contributions from Lilly Deutschland, Sanofi, Berlin Chemie, Boehringer-Ingelheim and the JuZo Academy.
Literature:
- Estruch R, Ros E, Salas-Salvadó J,et al.: Primary Prevention of Cardiovascular Disease with a Mediterranean Diet Supplemented with Extra-Virgin Olive Oil or Nuts. N Engl J Med 2018; 378(25): e34.
- Gong Q, Zhang P, Wang J, et al.: Morbidity and mortality after lifestyle intervention for people with impaired glucose tolerance: 30-year results of the Da Qing Diabetes Prevention Outcome Study. Lancet Diabetes Endocrinol 2019; 7(6): 452–461.
- Ahlqvist E, Storm P, Käräjämäki A, et al.: Novel subgroups of adult-onset diabetes and their association with outcomes: a data-driven cluster analysis of six variables. Lancet Diabetes Endocrinol 2018; 6(5): 361–369.
- Wagner R, Heni M, Tabák AG, et al.: Pathophysiology-based subphenotyping of individuals at elevated risk for type 2 diabetes. Nat Med 2021; 27(1): 49–57.
- Marklund M, Magnusdottir OK, Rosqvist F, et al.: A dietary biomarker approach captures compliance and cardiometabolic effects of a healthy Nordic diet in individuals with metabolic syndrome. J Nutr 2014; 144(10): 1642–1649.
- Schoeller DA: Validation of habitual energy intake. Public Health Nutr 2002; 5(6A): 883–888.
- Dansinger ML, Gleason JA, Griffith JL, et al: Comparison of the Atkins, Ornish, Weight Watchers, and Zone diets for weight loss and heart disease risk reduction: a randomized trial. JAMA 2005; 293(1): 43-53.
- Schmidt I: Analyse zur Diätcompliance bei «Low-Carb»- und «Low-Fat»-Studien. Doktorarbeit; Charité – Universitätsmedizin Berlin 2017.
- Keller J: Metaanalyse zu Diätcompliance und Drop-out-Rate in RCTs zur vegetarischen/veganen Ernährung; Masterarbeit, Universität Potsdam 2022.
- Kabisch S, Wenschuh S, Buccellato P, et al.: Affordability of Different Isocaloric Healthy Diets in Germany – An Assessment of Food Prices for Seven Distinct Food Patterns. Nutrients 2021; 13(9): 3037.
InFo DIABETOLOGIE & ENDOKRINOLOGIE 2024; 1(2): 6–9