This well-conducted review found that a higher intake of total dietary fibre or whole grains was associated with a reduction in the risk of mortality and in the incidence of a wide range of non-communicable diseases and their risk factors.
Overall summary Low risk of bias in the review
All domains were considered at low risk of bias suggesting no limitations with the review process.
|A. Did the interpretation of findings address all of the concerns identified in Domains 1 to 4?||Probably yes|
|B. Was the relevance of identified studies to the review's research question appropriately considered?||Probably yes|
|C. Did the reviewers avoid emphasizing results on the basis of their statistical significance?||Probably yes|
|Risk of bias in the review||Low|
|Number of studies||243|
|Number of participants||4,635|
|Last search date||28 Feb 2018|
|Objective||To determine the predictive potential of several markers, or which markers are most useful, and to establish an evidence base for quantitative recommendations for intakes of dietary fibre.|
|Population||Adults and children in low-, middle- and high-income countries without acute or chronic disease. Studies could include people with prediabetes, mild to moderate hypercholesterolaemia, mild to moderate hypertension, or metabolic syndrome.
Trials of weight loss trials or trials involving dietary fibre supplements were excluded.
|Interventions||Prospective cohort studies: Varying carbohydrate quality by increasing total dietary fibre (consuming >25g/d relative to <25g/d total dietary fibre; or consuming >30g/d relative to <30g/d total dietary fibre); or naturally occurring fibres
Randomised controlled trials (RCTs): Replacing rapidly digested starches with slowly digested starches, or carbohydrates with higher glycaemic index, or refined carbohydrate, or lower resistant starch intake or lower intakes of pulses, or higher starch plus sugar
Weight loss trials and trials involving the provision of dietary fibre supplements in powder form and trials including people on medications known to eﬀect the outcomes, or who were pregnant or in situations in which regular eating habits were likely to change (eg, individuals with eating disorders or who were breastfeeding) were excluded
|Comparator||Prospective cohort studies: Higher/normal/usual/lower fibre intake or extracted and synthetic dietary fibres
RCTs: Carbohydrates with lower glycaemic index, or wholegrain carbohydrate, or higher resistant starch intake, or higher intakes of pulses, or lower starch plus sugar
|Outcome||Prospective studies: All-cause mortality, coronary heart disease mortality and stroke mortality; incidence of coronary heart disease, stroke, type 2 diabetes, and colorectal cancer, cardiovascular disease incidence and mortality, incidence of adiposity-related cancers.
Randomised controlled trials (RCTs): Adiposity, fasting glucose, fasting insulin, insulin sensitivity, glycated haemoglobin A1c, triglycerides, cholesterol and blood pressure.
|Study design||Prospective cohort studies and randomised controlled trials of at least four weeks' duration.|
|PP factor||Carbohydrate quality (ie, differences in dietary fibre, whole grains or pulses, dietary glycaemic index, or glycaemic load).|
Prospective cohort studies
The pooled analysis of observational studies demonstrated that a higher intake of total dietary fibre was significantly associated with decreased all-cause mortality (risk ratio [RR] 0·85, 95% confidence interval [CI] 0·79 to 0·91, 10 studies, n=80,139), coronary heart disease (mortality: RR 0·69, 95% CI 0·60 to 0·81, 10 studies, n=7,243; and incidence: RR 0·76, 95% CI 0·69 to 0·83, 9 studies, n=7,155), stroke incidence (RR 0·78, 95% CI 0·69 to 0·88, 9 studies, n=13,134), type 2 diabetes incidence (RR 0·84, 95% CI 0·78 to 0·90, 17 studies, n=48,468), colorectal cancer incidence (RR 0·84, 95% CI 0·78 to 0·89, 22 studies, n=22,920) and cancer mortality (RR 0·87, 95% CI 0·79 to 0·95, 5 studies, n=29,593). However, no significant difference was observed between higher and lower intake of total dietary fibre for stroke mortality (RR 0·80, 95% CI 0·56 to 1·14, 2 studies, n=1,103).
Similarly, a higher intake of whole grains was significantly associated with decreased all-cause mortality (RR 0·81, 95% CI 0·72 to 0·90, 9 studies, n=99,224), coronary heart disease (mortality: RR 0·66, 95% CI 0·56 to 0·77, 2 studies, n=1,588; and incidence: RR 0·80, 95% CI 0·70 to 0·91, 6 studies, n=7,697), stroke mortality (RR 0·74, 95% CI 0·58 to 0·94, 2 studies, n=694), type 2 diabetes incidence (RR 0·67, 95% CI 0·58 to 0·78, 8 studies, n=14,686), colorectal cancer incidence (RR 0·87, 95% CI 0·79 to 0·96, 7 studies, n=8,803) and cancer mortality (RR 0·84, 95% CI 0·76 to 0·92, 5 studies, n=32,727) compared to the lower intake of whole grains. However, no significant difference was observed between higher and lower intake of whole grains for stroke incidence (RR 0·86, 95% CI 0·61 to 1·21, 3 studies, n=1,247).
Moreover, a lower glycaemic index was significantly associated with decreased stroke mortality (RR 0·63, 95% CI 0·52 to 0·77, 3 studies, n=951), stroke incidence (RR 0·84, 95% CI 0·72 to 0·99, 5 studies, n=5,527) and type 2 diabetes incidence (RR 0·89, 95% CI 0·82 to 0·97, 14 studies, n=36,908) compared to the higher glycaemic index. However, no significant differences were found between lower and higher glycaemic index with respect to all-cause mortality (3 studies, n=7,698), coronary heart disease mortality (1 study) or incidence (10 studies), colorectal cancer incidence (10 studies, n=11,245) and cancer mortality (1 study, n=1,401).
Randomised controlled trials (RCTs)
The pooled analysis of RCTs demonstrated that a higher intake of total dietary fibre was significantly associated with a decreased change in body weight (mean difference [MD] -0·37 Kg, 95% CI -0·63 to -0·11, 27 RCTs, n=2,495), change in total cholesterol (MD -0·15 mmol/L, 95% CI -0·22 to -0·07, 36 RCTs, n=3,503), and change in systolic blood pressure (MD -1·27 mm Hg, 95% CI -2·50 to -0·04, 15 RCTs, n=2,052) compared to lower intake of total dietary fibre. However, no significant difference was reported between higher and lower intake of total dietary fibre for change in glycated haemoglobin A1c (standard mean difference [SMD] -0·35%, 95% CI -0·73 to 0·03, 6 RCTs, n=380).
Similarly, a higher intake of whole grains was significantly associated with a decreased change in body weight compared to the lower intake of total dietary fibre (MD -0·62 Kg, 95% CI -1·19 to -0·05, 11 RCTs, n=919). However, no significant difference was reported between higher and lower intake of total dietary fibre for change in glycated haemoglobin A1c (3 RCTs, n=282), change in total cholesterol (17 RCTs, n=1473) and change in systolic blood pressure (8 RCTs, n=925). Moreover, no significant differences were found between lower and higher glycaemic index with respect to change in body weight (8 RCTs, n=799), change in glycated haemoglobin A1c (2 RCTs, n=81), change in total cholesterol (8 RCTs, n=1083) and change in systolic blood pressure (4 RCTs, n=916).
The research objective was clearly stated and appropriate inclusion criteria were defined. No restrictions were applied to eligibility criteria based on study characteristics and sources of information.
|1.1 Did the review adhere to pre-defined objectives and eligibility criteria?||Yes|
|1.2 Were the eligibility criteria appropriate for the review question?||Probably yes|
|1.3 Were eligibility criteria unambiguous?||Probably yes|
|1.4 Were all restrictions in eligibility criteria based on study characteristics appropriate (e.g. date, sample size, study quality, outcomes measured)?||Probably yes|
|1.5 Were any restrictions in eligibility criteria based on sources of information appropriate (e.g. publication status or format, language, availability of data)?||Probably yes|
|Concerns regarding specification of study eligibility criteria||Low|
Ovid MEDLINE, EMBASE, PubMed, Web of Science and Scopus were searched for prospective cohort studies and PubMed, Ovid MEDLINE, EMBASE, and the Cochrane Central Register of Controlled Trials, Cochrane Database of Systematic Reviews, and Food Science and Technology Abstracts databases for randomised controlled trials were searched for relevant studies. In addition, the authors hand searched a range of previous publications. The search strategy was reported in full and appeared appropriate. Searches were not restricted by date, publication format or language. At least two review authors were independently involved in study selection and discrepancies were resolved with an additional reviewer.
|2.1 Did the search include an appropriate range of databases/electronic sources for published and unpublished reports?||Probably yes|
|2.2 Were methods additional to database searching used to identify relevant reports?||Yes|
|2.3 Were the terms and structure of the search strategy likely to retrieve as many eligible studies as possible?||Probably yes|
|2.4 Were restrictions based on date, publication format, or language appropriate?||Yes|
|2.5 Were efforts made to minimise error in selection of studies?||Yes|
|Concerns regarding methods used to identify and/or select studies||Low|
At least two review authors were independently involved in the data extraction and risk of bias assessment; discrepancies were resolved with an additional reviewer. Sufficient study characteristics appear to have been extracted to allow interpretation of results. Relevant study results appeared to have been extracted. Methodological quality of the included studies was assessed using the Newcastle–Ottawa Scale for prospective cohort studies and the Cochrane criteria for randomised controlled trials.
|3.1 Were efforts made to minimise error in data collection?||Yes|
|3.2 Were sufficient study characteristics considered for both review authors and readers to be able to interpret the results?||Probably yes|
|3.3 Were all relevant study results collected for use in the synthesis?||Probably yes|
|3.4 Was risk of bias (or methodological quality) formally assessed using appropriate criteria?||Yes|
|3.5 Were efforts made to minimise error in risk of bias assessment?||Yes|
|Concerns regarding methods used to collect data and appraise studies||Low|
The synthesis included all of the relevant studies. The method of analysis was explained and appeared appropriate. There was significant evidence of heterogeneity. Subgroup analysis, sensitivity analysis and meta-regression analysis were performed to address the heterogeneity and to assess the robustness of results. The quality of individual studies was considered in the synthesis.
|4.1 Did the synthesis include all studies that it should?||Probably yes|
|4.2 Were all pre-defined analyses reported or departures explained?||Probably yes|
|4.3 Was the synthesis appropriate given the degree of similarity in the research questions, study designs and outcomes across included studies?||Probably yes|
|4.4 Was between-study variation minimal or addressed in the synthesis?||Probably yes|
|4.5 Were the findings robust, e.g. as demonstrated through funnel plot or sensitivity analyses?||Probably yes|
|4.6 Were biases in primary studies minimal or addressed in the synthesis?||Probably yes|
|Concerns regarding synthesis and findings||Low|
Summary Background Previous systematic reviews and meta-analyses explaining the relationship between carbohydrate quality and health have usually examined a single marker and a limited number of clinical outcomes. We aimed to more precisely quantify the predictive potential of several markers, to determine which markers are most useful, and to establish an evidence base for quantitative recommendations for intakes of dietary fibre. Methods We did a series of systematic reviews and meta-analyses of prospective studies published from database inception to April 30, 2017, and randomised controlled trials published from database inception to Feb 28, 2018, which reported on indicators of carbohydrate quality and non-communicable disease incidence, mortality, and risk factors. Studies were identified by searches in PubMed, Ovid MEDLINE, Embase, and the Cochrane Central Register of Controlled Trials, and by hand searching of previous publications. We excluded prospective studies and trials reporting on participants with a chronic disease, and weight loss trials or trials involving supplements. Searches, data extraction, and bias assessment were duplicated independently. Robustness of pooled estimates from random-effects models was considered with sensitivity analyses, meta-regression, dose-response testing, and subgroup analyses. The GRADE approach was used to assess quality of evidence. Findings Just under 135 million person-years of data from 185 prospective studies and 58 clinical trials with 4635 adult participants were included in the analyses. Observational data suggest a 15–30% decrease in all-cause and cardiovascular related mortality, and incidence of coronary heart disease, stroke incidence and mortality, type 2 diabetes, and colorectal cancer when comparing the highest dietary fibre consumers with the lowest consumers Clinical trials show significantly lower bodyweight, systolic blood pressure, and total cholesterol when comparing higher with lower intakes of dietary fibre. Risk reduction associated with a range of critical outcomes was greatest when daily intake of dietary fibre was between 25 g and 29 g. Dose-response curves suggested that higher intakes of dietary fibre could confer even greater benefit to protect against cardiovascular diseases, type 2 diabetes, and colorectal and breast cancer. Similar findings for whole grain intake were observed. Smaller or no risk reductions were found with the observational data when comparing the effects of diets characterised by low rather than higher glycaemic index or load. The certainty of evidence for relationships between carbohydrate quality and critical outcomes was graded as moderate for dietary fibre, low to moderate for whole grains, and low to very low for dietary glycaemic index and glycaemic load. Data relating to other dietary exposures are scarce. Interpretation Findings from prospective studies and clinical trials associated with relatively high intakes of dietary fibre and whole grains were complementary, and striking dose-response evidence indicates that the relationships to several non-communicable diseases could be causal. Implementation of recommendations to increase dietary fibre intake and to replace refined grains with whole grains is expected to benefit human health. A major strength of the study was the ability to examine key indicators of carbohydrate quality in relation to a range of non-communicable disease outcomes from cohort studies and randomised trials in a single study. Our findings are limited to risk reduction in the population at large rather than those with chronic disease. Funding Health Research Council of New Zealand, WHO, Riddet Centre of Research Excellence, Healthier Lives National Science Challenge, University of Otago, and the Otago Southland Diabetes Research Trust.