Introduction

People in industrialised societies have a tendency to be overweight [1, 2], thus in the past century non-professional diets have focused primarily on calorie restriction [3–6]. In the last two decades, many diets that were previously followed by only a narrow social group (vegan diet), or that were initially designed for an even narrower group of mainly medical patients (ketogenic diet), have become popular [7, 8].

The dietary trends now being followed by increasing numbers of people are still based on restrictions, although not exclusively calorie restriction [9–11]. Paleo, vegetarian and vegan trends, for example, exclude certain food groups. Some trends restrict one of the three main macronutrient groups: these include low-carbohydrate, low-fat, and high-protein diets (e.g. the carnivore diet), or low-carbohydrate and high-fat diets (e.g. the ketogenic diet). Other trends are based on the manipulation of timing and include time-restricted diets and intermittent or even prolonged fasting. Self-managing dieters sometimes choose elements from this palette and combine them. They might follow a vegan-ketogenic diet or a ketogenic diet combined with prolonged fasting, for example, which are obviously highly restrictive [10, 12–14].

Furthermore, dieters commonly switch between diets after a short period if their objectives are not met immediately [15, 16], resulting in an infinite number of further combinations and variations of the modified trends. All diets potentially have temporary or long-term benefits, but all of them are accompanied by risks in the context of different conditions [13, 14, 17–20].

The goal of the present paper is to draw attention to the fact that the vast majority of dieters today follow a self-managed, “do-it-yourself” diet without consulting a professional at any stage [21, 22]. However, developing and personalising an appropriate diet, monitoring it, and minimising its adverse effects are the tasks of qualified professionals [23, 24]. The aim of our research was therefore to survey self-proclaimed ketogenic and vegan dieters to assess their motivations, their sources of information, and the side effects they experience. We chose two very different diets in order to present a broad spectrum of lessons learned. We were curious to find out what percentage of the dieters had sought the help of a dietitian, nutritionist or doctor — in other words, professional help.

The term “pseudo-diet” is used to express the dichotomy between a professionally supervised diet and a do-it-yourself diet. Followers of a pseudo-diet do not regularly consult a professional but decide for themselves which elements of a given diet to follow, how long for, and what kind of detailed nutritional parameters to include. Thus, the label used by the dieters does not necessarily correspond to the strictly professional definition of the given diet. In the present study, the diets are thus considered to be ketogenic-like or vegan-like (KLD or VLD), unless their validity as genuine ketogenic or vegan diets has been verified by a professional. While professional monitoring might be the subject of a future study, in the context of the present study we have retained the suffix “-like” in each case, although we were curious to know whether the dieters had ever asked for professional help with their diet.

Materials and methods

The research took the form of a questionnaire-based cross-sectional study, for which we administered self-edited online questionnaires. The questionnaires contained both closed (multiple choice) and open-ended questions and took about 15 min to complete. We recruited our participants from the most popular Hungarian ketogenic and vegan Facebook groups, ensuring comprehensive sampling across the country.

There were 452 participants, 249 of whom followed a KLD and 203 of whom followed a VLD. All participants were Hungarian residents, and the questionnaire was voluntary and anonymous. Exclusion criteria were being younger than 18 years of age and following a ketogenic diet for the medical therapeutic treatment of epilepsy.

The data (mean, SD) were analysed using IBM SPSS Statistics for Windows, Version 25.0 (Released 2017. Armonk, NY: IBM Corp.) and Excel 2016 (v16.0). Weight and height data were entered by the dieters in the questionnaire. The difference in mean BMI between the two groups was examined using an independent samples t-test. For the investigation of other hypotheses, we used the chi-square test. Statistical significance for all tests was P < 0.05.

Results

Baseline data

Of the 452 participants, 249 followed a diet that they considered to be ketogenic (KLD) and 203 followed what they labelled a vegan diet (VLD). The mean age of the KLD followers was 40.72 ± 10.08, and that of the VLD followers 27.61 ± 9.25. A VLD was preferred by women aged between 18 and 29, as 136 out of 166 female respondents below 30 years of age (81.9%) were following a VLD, accounting for 74.3% of the total of 183 women on a VLD. Among the 229 female respondents aged 30 years and over, 182 (79.4%) followed a KLD, accounting for 85.8% of the 212 female followers of a KLD.

Similarly, men in the 18 to 29 age group preferred a VLD, with 11 of the 17 male respondents aged 30 years and above following a VLD (64.7%). These 11 men accounted for 55% of the total of 20 men following a VLD. Older male respondents tended to follow a KLD, with 31 of the 40 male respondents aged 30 years and above following a KLD (77.5%). Of the 37 male respondents following a KLD, 31 (83.7%) were over 30 years of age. The detailed statistics are shown by age group in Table 1.

Table 1.

Age and gender distribution of participants

Age group	Female (n = 395)		Male (n = 57)
	Ketogenic-like (n = 212)	Vegan-like (n = 183)	Ketogenic-like (n = 37)	Vegan-like (n = 20)
18–29 (n = 183)	30 (14.1%)	136 (74.3%)	6 (16.2%)	11 (55.0%)
30–39 (n = 111)	69 (32.6%)	21 (11.5%)	15 (40.6%)	6 (30.0%)
40–49 (n = 100)	72 (34.0%)	18 (9.8%)	8 (21.6%)	2 (10.0%)
50–64 (n = 56)	41 (19.3%)	8 (4.4%)	6 (16.2%)	1 (5.0%)
65+ (n = 2)	–	–	2 (5.4%)	–

The weight of the KLD followers varied between 51 and 150 kg, the average weight in the KLD group being 79.98 kg ± 19.02. The weight of the VLD followers fell between 41 and 175 kg, with an average weight of 61.23 kg ± 13.28. Body mass index (BMI) was calculated from the participants' height and body weight data. As shown in Fig. 1, participants were categorised into BMI groups based on the WHO guidelines [25]. In the KLD group, 0.4% of participants were moderately thin, 1.2% were mildly thin, 34.54% had a normal BMI, 30.12% were overweight, 19.68% were class I obese, 9.64% were class II obese, and 4.42% were class III obese. Among the VLD followers, 9.85% were mildly thin, 77.34% had a normal BMI, 9.36% were overweight, and 3.45% were class I obese, while the higher categories of obesity were absent from this group.

View Full Size

Proportions of participants according to the World Health Organisation BMI categories

Citation: Developments in Health Sciences 5, 1; 10.1556/2066.2022.00058

• Download Figure
• Download figure as PowerPoint slide

The average BMI in the KLD group was 28.1 kg m⁻² ± 6.2, while the average BMI in the VLD group was 21.79 kg m⁻² ± 3.94. We found a statistically significant difference between the BMI values of the two groups using an independent sample t-test (P < 0.001). The data show that the KLD followers had higher BMI values than the VLD followers. The difference was statistically significant, as was the association revealed by the chi-square test (P < 0.00).

Adherence, motivations

The members of the KLD group had been on the diet for an average of nine months (±14.76 months) and the members of the VLD group for 2.5 years (±2.37 years) at the time of the survey. In the case of the KLD followers, 85.14% were motivated by the goal of losing excess kilos, while 53.41% were also motivated by the goal of improving their health and 36.55% by the goal of enhancing their well-being. “Other” motivations were rare and included treating diseases and gaining muscle and weight. Respect for animals and ethical considerations were the primary motivations cited by 56.16% of the VLD followers. Only 29.56% cited the goal of improving their health as a motivation, while 7.39% mentioned environmental protection. The differences in the quality of motivation between the two groups are shown in Fig. 2.

View Full Size

Proportions of participants by motivation in the KLD and VLD groups

Citation: Developments in Health Sciences 5, 1; 10.1556/2066.2022.00058

• Download Figure
• Download figure as PowerPoint slide

Information sources

We found that a higher percentage of VLD than KLD followers had consulted a professional (e.g. doctor, dietitian or nutritionist) for information at some stage (33.99% compared to 11.64%). The chi-square test yielded a P value of <0.001, which indicates a significant association between diet type and the seeking of professional advice.

Among the KLD followers, 71.49% mentioned the Internet or social media portals, 20.48% books, and 18.08% family and friends, while 11.64% had sought professional advice and 10.05% had consulted other sources of diet-related information. Among the VLD followers, 98.03% had sought advice from the Internet or social media portals, 54.68% from books, 42.86% from various sources available at vegan events, 28.08% from family members and friends, 33.99% from a professional, and 11.33% from other sources. Figure 3 shows the proportions of the different information sources mentioned by the respondents.

View Full Size

Proportions of participants by information sources used in the KLD and VLD groups

Citation: Developments in Health Sciences 5, 1; 10.1556/2066.2022.00058

• Download Figure
• Download figure as PowerPoint slide

Diet outcome: Adverse and desirable effects

We found a statistically significant association between type of diet and frequency of experienced adverse and desirable effects (chi-square P < 0.001). The frequency of adverse effects was more than four times higher (413%) in the KLD group than in the VLD group. At the same time, the frequency of desirable effects was almost 1.5 times higher (149%) in the VLD group than in the KLD group.

The variety of adverse effects was far greater in the KLD group than in the VLD group (11 and 4 different adverse effects respectively). The adverse effects most commonly experienced by followers of a KLD were constipation (40.96%), headaches (39.36%), and lowered energy levels (31.73%). Diarrhoea (22.09%), lethargy (17.67%), hunger (17.67%), reduced mental function (13.65%), malaise (12.45%), impaired performance (8.43%), bloating (6.43%), disturbed vision (6.43%), hair loss (6.02%), loss of muscle mass (2.41%), weight gain (1.2%) and elevated blood pressure (0.8%) were also mentioned in the KLD group. Among followers of a VLD, weight gain (9.85%), tiredness (4.43%), poor digestion (4.93%), and loss of appetite (0.98%) were reported as adverse effects. Desirable effects experienced by KLD followers included weight loss (79.92%), general well-being/high energy levels (63.86%), mental freshness (47.79%), and general satiety (43.78%). Members of the VLD group reported better digestion/more frequent bowel movements (75.86%), increased energy levels (56.65%), weight loss (34.48%), and improved appetite (29.56%) among the desirable effects.

To assess the correlation between frequency of desirable/adverse effects and use of professional help, we divided the two groups of dieters into two further subgroups: those who used at least some personalised professional help (WP-KLD and WP-VLD); and those who did not (NP-KLD and NP-VLD.) The percentages of participants experiencing desirable and adverse effects in the four subgroups are shown in Fig. 4, including the minor rates of those who experienced no effects from the dietary changes they introduced.

View Full Size

Proportions of participants experiencing cumulative desirable, adverse, and zero effects, with and without professional help (WP, NP)

Citation: Developments in Health Sciences 5, 1; 10.1556/2066.2022.00058

• Download Figure
• Download figure as PowerPoint slide

In the NP-KLD subgroup, 46.64% reported achieving one or more desirable effects. The achievement of desirable effects rose to 63.46% among members of the WP-KLD subgroup, meaning an increase of 16.82%. The figure shows that 8.73% more VLD followers who sought professional help (WP-VLD subgroup) experienced desirable effects than VLD followers who did not (NP-VLD subgroup) (96.15% compared to 87.42%). We found that 3.68% fewer members of the WP-KLD subgroup experienced adverse effects compared to the NP-KLD subgroup (55.47% and 51.79% respectively). However, the figure also shows that over half of KLD followers experienced adverse effects in both the WP and NP conditions. In the NP-VLD subgroup, 19.21% experienced adverse effects compared to 13.46% of participants who did consult with a professional, representing a decrease of 5.75% in the WP-VLD group.

Discussion

The modern ketogenic diet was developed in the 1920s by physicians at Mayo Clinic [26] to treat epilepsy. The diet has recently been rediscovered as a trend for weight loss, and even for the treatment of type 2 diabetes [19]. The diet is very low in carbohydrates (around 5% of calorie intake) and high in fat, making it a ketone-producing diet. The emerging underlying mechanisms of the ketogenic diet are still proving successful [17, 27–29]. However, its widespread applicability is surrounded by a great deal of controversy, mainly due to its potentially serious side effects. In the short term, the diet can cause fatigue, headache, nausea, constipation, hypoglycaemia and acidosis, while in the longer term it may result in a variety of conditions including reduced bone density, nephrolithiasis, cardiomyopathy and anaemia [30].

Originally, the vegan diet excluded all foods of animal origin for religious, ethical, or ecological reasons. In recent decades, a growing number of people have adopted vegan principles for the purposes of health enhancement [8]. In terms of macronutrients, the vegan diet is more balanced than the ketogenic diet. A well-constructed vegan diet may have several health benefits, such as reducing the risk of metabolic syndrome [31] and certain cancers [32] and supporting a healthy gut flora [33]. However, the most common risk associated with a poorly implemented vegan diet is that it may be deficient in various micronutrients (vitamin B12, zinc, calcium and selenium) [34].

As stated above, in the present paper the KLD and VLD are considered pseudo-diets until their authenticity has been confirmed by a qualified expert. By analysing the baseline data, we sought to identify the kind of people who start following a KLD and VLD, and why. Our results show that dieting trends are reversed in the two sexes among those over 30 years of age. To summarise, members of the younger generation (18–29 years of age) tend to follow the vegan trend, while the ketogenic trend becomes more attractive after the age of 30. An investigation of age distribution might be the subject of a follow-up study. The significant difference in mean BMI (by 1.28 times) between the VLD and KLD groups suggests that KLD followers struggle with their weight more often. Figure 1 shows that the KLD followers were indeed more likely to fall into the WHO overweight or obese categories than the VLD followers, meaning that the KLD followers who pursued the diet in order to lose weight were realistic in the assessment of their condition. Figure 2 shows the motivations of the two groups, which tend to be polarised and overlap less often. All the goals set by the dieters appear to be health promoting, including the simple need to lose weight, which can be explained by differences in BMI; ethical considerations, which can be interpreted as improving psychological health; and environmental reasons, which can be seen as improving the health of the whole biosphere. However, the individual's current condition always influences the health-related consequences of any change introduced. The degree of motivation can be measured by the extent to which followers are able to remain committed. Our cross-sectional survey does not show how many people gave up their diet prematurely, although it can be taken as an indirect indicator that a VLD is likely to be easier to adhere to. Adherence to a diet is helpful only for professionally recognised goals and results. It is encouraging that 33.9% of the VLD followers asked for professional help and worrying that only 11.64% of the KLD followers did so. The adherence data may explain this difference — the higher rate among VLD followers may be due to the fact they have been on their diet for years and have had more time to consult a specialist. Unfortunately, we have no information about the length of the professional monitoring, since our survey provides only a snapshot of who had sought professional help ever up until the time of the survey.

As shown in Fig. 3, the Internet was the primary source of non-professional information. Although professional advice may occasionally be accessible on the Internet, the non-professional sources are unlikely to provide personalised, ongoing monitoring and dieters are more likely to come across random, anecdotal, often commercial information [35, 36].

The correlation between type of diet and frequency of effects indicates that the vegan trend is more beneficial in terms of both desirable and adverse effects. To summarise what we can learn from the exact percentages of desirable and adverse effects, the ketogenic trend appears to be suitable for weight loss but has a number of potentially dangerous side effects, while the vegan trend is more likely to achieve its goals without adverse effects. We also investigated whether any difference in effects could be observed when professional help was sought. Figure 4 shows that professional help was more critical in achieving desirable effects than in avoiding adverse effects. More members of the WP-KLD subgroup experienced desirable effects than members of the NP-KLD subgroup (by 16.82%), and more members of the WP-VLD subgroup experienced desirable effects than members of the NP-VLD subgroup (by 8.73%). Adverse effects were experienced by only 3.68% fewer members of the WP-KLD subgroup than the NP-KLD subgroup. Members of the WP-VLD subgroup experienced fewer adverse effects (by 5.75%) than members of the NP-VLD subgroup, suggesting that adverse effects were avoided more easily with professional advice. However, it should be noted that very few adverse effects were experienced in the VLD group (a quarter of the number experienced in the KLD group)., Since VLD followers are more persistent in their diet, it is possible that they consult a specialist only in the case of adverse effects that occur over months.

Conclusions

The adverse effects of a poorly implemented diet in either group can have serious medical consequences, thus every effort should be made to avoid them. Perseverance, and adherence to even seemingly reasonable goals, can be dangerous in the event of hidden physical or mental health problems, such as metabolic or body image disorders or very restrictive orthorexia. In other words, precise, detailed and personalised health goals can be formulated only with a knowledge of the individual's current condition, which means that regular consultations with a specialist are essential.

As the ketogenic trend in our survey appears to be riskier in terms of adverse effects, we recommend closer professional monitoring of the followers of this diet. According to Crosby and her colleagues [30], blood ketones, glucose, cardiac, and other parameters should be investigated once or twice a month. It is also advisable to seek professional support before starting to follow the vegan trend in order to avoid deficiencies in the first place, and to monitor them, more or less closely, later on.

Authors' contribution

BV and USV summarised the theoretical background to the paper. RMA and ZSP collected data and performed the necessary analyses. BV and USV summarised and drew conclusions from the results of the study. BV, UVS, and ME supervised the final content.

Ethical approval

All procedures were in accordance with the ethical standards of the institutional and/or national research committee and with the 1964 Helsinki Declaration and its later amendments or comparable ethical standards. Respondents received written information about the study.

Conflicts of interest/funding

The author declares no conflicts of interest. No financial support was received for this study.