DISCUSSION. and control. Barnard et al.,2009 used a

DISCUSSION.

Plant-based diets have been associated with weight loss over the recent years. Epidemiological studies have shown vegetarians and vegans to have a lower body mass compared to omnivores (Alewaetersk.,2005). This systematic review assembled the most current scientific evidence to assess the efficacy of plant-based diets in the management of obesity. The focus was mainly on weight loss but the effect on clinical characteristics associated with risk factors in obesity was also assessed. The seven trials included compared a type of plant-based diet with non-vegetarian diets, whether engineered for weight loss or just a standard habitual diet. The trials included had no restriction on energy intake and physical activity was not part of the intervention.

The studies in this review showed that plant-based diets are associated with significant weight reduction in a completer and intention-to-treat-analysis. Despite different durations of intervention, the studies by Turner et al 2007, Mishra et al.,2013, Gabrielle et al 2013, Jenkins et al 2014 and Wright et al 2015 showed significant weight reduction in the plant-based diet groups compared to the control. Two trials Burke et al 2008 and Barnard et al 2009, despite significant weight loss in both groups, found no difference in weight loss between the plant-based diet and control. Barnard et al.,2009 used a diabetes diet as control and this observation was attributed to the fact that the diabetes diet is also engineered for weight loss. Burke et al.,2008 used a Lacto-Ovo vegetarian diet which included milk and egg consumption which may have affected energy intake hence a lack of significant difference. A deduction from other trials on plant-based diets in weight management is low total energy density as the chief driving mechanism due high fiber and low-fat content, an observation in trials in this systematic review as well. The decrease in energy intake was observed in all the studies apart from 2.

Reduction in total fat, saturated fat and cholesterol intake, which subsequently influence blood lipid, in the intervention group are alike to those observed in previous trials with plant-based diets. (3 references). All studies showed a significant amount of reduction in total cholesterol apart from 1 study (Turner et al 2007) which did not have a report on the clinical outcomes.

Jenkins et al.,2014 attributed the decrease in LDL-C to cholesterol-lowering effect of soy foods and nuts(Sabate et al.,2010:Ramdath et al.,2017) and decrease in risk of coronary heart disease(CHD) to high monounsaturated fatty acids and plant protein and oil. There was no significant decrease in HDL-C in Jenkin et al.,2014’s study.Previuos research has however shown that lowering LDL-C and maintaining HDL-C reduces the risk of CHD(Grundy et al.,2004)

Barnard et al 2009, was the only study that showed an increase in HDL-C and an increase in carbohydrate intake was reported as the cause. No significant difference in saturated fat was reported by Jenkins et al 2014 due to high consumption of vegetable and soy protein. The type of protein consumed may not be a key factor in weight loss, a conclusion from a study that found no difference in effect of soy/legume and meat-based protein in control of appetite, energy expenditure and thermic effect of food because participants in this trial lost weight regardless of protein source (Li et al.,2016)

Adherence to diet is important in the management of obesity for it to be any effective. Wright et al observed a 76% adherence to the vegan diet and attributed it to satiety as participants reported ‘not feeling hungry’ enabled them to stick to the diet. In the study by Turner et al.,2007, individuals who lost the most weight were those who were most adherent. In contrast, Jenkins et al.,2014 reported an attrition rate of 50% but the reason was not stated. There was a high relation between weight loss and attendance and support for individuals receiving the vegan diet.Burke et al.,2008 reported excellent retention at 18 months(75%),attributing it to group support.This could be considered a possible way of increasing adherence in future studies.

The benefits of the plant-based diets have been strongly associated with the high amount of fiber given the consumption of whole grains, vegetables and fruits (Berkow and Barnard,2006). Gabrielle et al.,2013 showed the greatest amount of weight loss in the vegan group which had the highest amount of fiber compared to the pesco, Lacto-Ovo vegetarian and omnivorous group. Barnard et al.,2008 attributed high fiber intake to increase consumption of fruits and vegetables. While Mishra et al attributed it to consumption of whole grains.

Previous studies have attributed the plant-based diet’s ability to improve blood glucose to Its component of low GI foods as well as high fibre.Low-fat vegan diet have been suggested to increase insulin sensitivity in cells allowing quick metabolism of glucose by cells hence no fat is stored up(Banard et al.,2005)Turner et al 2007 reported a tremendous improvement in insulin sensitivity in the vegan diet group compared to the diabetes diet group, Mishra et al and Jenkins et al showed a decrease in glycated hemoglobin as well.71% of diabetics in Barnard et al’s(2009) study altered their diabetes medication. Wright et al reported a mean decrease of 5 mmol/mol-1 at 12 months in the vegan diet group and 2 diabetics no longer met diagnostic criteria. Glycated hemoglobin fell similarly in intervention and control group. Burke et al.,2008 reported a decrease in insulin.

Differing dietary and subsequently nutrient intake contributes to high diet quality of plant-based diet compared to omnivorous diet.in dietary intake in the plant-based diet (clary’s et al). This systematic review has mirrored this conclusion from previous studies. All the studies in this review showed an increased intake of dietary fiber contributing factor to weight loss. Most studies reported a significant decrease in animal protein intake which subsequently decreased the amount of saturated fat hence low HDL and LDL levels. Overall, there was a decrease in fat intake which subsequently lowers the risk of cardiovascular disease(ref).

Determining whether data and results from included trials are valid both externally and internally aids in drawing clear conclusions on the effectiveness of the plant-based diet. All included studies are Random control trials. The method of sequence generation has been in all apart from the study by Jenkins et al., 2014.this, however, is not enough precaution against bias in allocation of intervention unless allocation concealment was carried out and none of the above studies had allocation concealment hence there is a risk of bias. Burke et al.,2008, had participants choose their preferred diet to compare the difference between them and those who were allocated a diet hence lack of allocation concealment was inevitable.

Blinding was difficult due to the nature of the study as treatment cannot be hidden from participant and professionals because trials were dietary intervention/advice based. An absence of blinding has shown to lead to overestimation of the effects of the intervention (Higgins 2008). However, Burke et al.,2008, Barnard et al.,2009, Jenkins et al.,2014 and Wright et al.,2015 blinded statisticians in the analysis of outcome hence a low risk of detection bias as they were unaware of the treatment assigned.

Jenkins et al.,2014, Mishra et al.,2013 and Burke et al.,2008 and Gabrielle et al 2013 had a low risk of attrition bias as details on number and reasons for attrition were given. The rest of the studies did not give details on reasons for attrition hence increasing chances for attrition bias. However, All the seven trials were on an intention-to-treat analysis with an aim to account for missing data on outcome and to avoid biases in non-random dropout rates.

Power analysis, apart from reducing type 2 error, increases the validity of a research as it shows a significant effect will not be missed. All the included trials apart from one were able to achieve the sample size according to their power analysis calculation. Mishra et al,.2013 had a lower sample size than that required from power analysis but showed significant results for key variables. This may mean that the power analysis gave an overestimation of sample size.

Some observational studies using the plant-based diet included physical activity as part of intervention which invalidates the assessment of effects of diet on body weight (Barnard et al 2015) Participant in all the included studied were advised not to alter habitual physical activity. There was no information on whether this was followed or not or even whether physical activity was present. This made it difficult to assess whether physical activity, if any, had any contribution to weight loss. This is one of the limitations

Results in this systematic review are also limited to the variation in the design of study, different sample sizes and control groups. A separate analysis was not done for studies that included individuals from different races and ethnicity to establish whether it has any effect on the outcome.

Quantification of energy intake was assessed from diet records. This means dietary intake was self-reported increasing the risk of information bias especially underreporting of energy intake.

Trials were also limited to little or no follow up periods with the longest follow-up period being only two years hence difficulty in commenting on the effectiveness of plant-based diets long term.

Most studies included individuals from different races at baseline, but separate analysis was not done to establish its effect,if any,on weight loss.

Conclusion.

Overall, the evidence from the clinical trials above shows plant-based diets have a great significant positive effect on weight loss and weight management. It has also shown to improve weight-related conditions such as diabetes and hence, may be helpful in the management of obesity. Additional benefits include increase consumption of plant phytochemicals, polyphenols and antioxidants that contribute to the health benefits of a plant-based diet.There is, however, more room for research especially on the long-term effects of the plant-based diet. Future studies should consider having longer follow up periods. This will aid in establishing whether plant-based diets are useful for the maintenance of weight.

Future studies should also consider ways of reducing attrition rate and improving adherence for the sake of evidence-based recommendations.

The above evidence has shown significant weight loss despite no energy restriction a deduction attributed to the low fat, high fiber nature of plant-based diets.Other possible mechanisms suggested are increases or decreases in macronutrient e.g meat protein,fat and energy are decreased while carbohydrates are increased.The mechanism of a plant-based diet in weight reduction therefore requires further exploration to enable identification of optimal plant-based diet for weight management.