| Study |
Bias |
| Randomisation process |
Deviations from intended interventions |
Missing outcome data |
Measurement of the outcome |
Selection of the reported results |
Overall |
| Authors' judgement |
Support for judgement |
Authors' judgement |
Support for judgement |
Authors' judgement |
Support for judgement |
Authors' judgement |
Support for judgement |
Authors' judgement |
Support for judgement |
Authors' judgement |
Support for judgement |
| Aude 2004 |
Some concerns |
Randomisation was done using a block design by sex to achieve an equal sex distribution. No information on allocation concealment. "At baseline, there were no significant differences in age, weight, BMI, waist‐to‐hip ratio, and any of the lipid variables between the NCEP and MLC groups" (only for completers). |
High risk of bias |
No information on blinding of participants and people delivering these dietary interventions, but blinding is unlikely. No reasons or details about deviations from intended interventions reported. A quarter of the participants were not included in the analysis: 8/30 participants in the intervention group and 7/30 in the control group were excluded from analysis; 6/60 did not complete the trial (missing), but no information about remaining 9/60 who were not included. |
High risk of bias |
8/30 (27%) participants in the intervention and 7/30 (23%) participants in the control arm did not have data available. No reasons for loss to follow‐up or withdrawal provided, so difficult to judge if related to participants' health status, but attrition was high in both groups; thus missingness could and is likely to depend on the true value. |
Low risk of bias |
Weight was measured with indoor clothing and without shoes on a single scale. No evidence that a different method was used across intervention groups to measure weight. "All anthropometric measurements were obtained by an observer blinded to the treatment assignment." |
Some concerns |
No trial registry reported and no mention of a prespecified analysis plan. Analysis intentions were not reported in sufficient detail to enable an assessment. |
High risk of bias |
No specific information on allocation concealment. A quarter of participants were not included in the analysis, with withdrawal reported for 10% of participants. No reasons were provided for the remaining missingness. Attrition was high in both groups with no reasons provided, therefore missingness could and is likely to be dependent on the true value of the outcome. No pre‐registered document available to assess analysis intentions. |
| Bales 2017 |
Low risk of bias |
We used a computerised centralised randomisation scheme blocking by race and marital or partner status. "...also blocked by functional status..." "The 1:2 allocation to the C‐WL and HP‐WL groups was chosen...to make within‐group comparisons in the protein group in subsequent analyses. This oversampling of the protein group did not cause a substantial loss of power for the between‐group comparisons; the relative power for a 2:1 split is 0.94(23)." No statements or statistical tests about differences between groups, no problematic imbalance suspected. |
Low risk of bias |
Not blinded (open‐label trial) (trial registry). Deviations reported are expected to arise in usual care: "We examined the reasons for dropout to inform future trials and found that 56% of dropouts had difficulty meeting study requirements due to family commitments or health issues; another 11% developed new health problems unrelated to the study." Number analysed unclear, but seemed to be a modified ITT: "The study was conducted under “intent to treat” criteria." |
High risk of bias |
22/51 (43%) participants in the intervention and 16/29 (55%) participants in the control arm did not have data available. "We examined the reasons for dropout to inform future trials and found that 56% of dropouts had difficulty meeting study requirements due to family commitments or health issues; another 11% developed new health problems unrelated to the study. In particular, the tendency for more dropouts among minorities and those with less education..." Very high attrition, thus missingness could and is likely to depend on the true value. |
Low risk of bias |
Weekly weigh‐ins. No evidence that a different method was used across intervention groups to measure weight. Open‐label. Weight is not an observer‐reported outcome involving judgement. |
Some concerns |
...treatment codes revealed only after the study statistician locked the database at the end of the trial. No mention of prespecified analysis plan, but trial registry at date prior to study publication mentioned the outcome and time point. All time points mentioned in trial registry reported on in publication. Analysis intentions were not reported in sufficient detail prior to study publication to enable an assessment. |
High risk of bias |
Very high attrition with a variety of reasons relating to health, therefore some missingness could be and is likely to be dependent on the true value of the outcome. Lack of information in the pre‐registered methods. |
| Bazzano 2014 |
Some concerns |
"We used a computer‐generated, blocked randomization, stratified by sex, to allocate participants to 1 of the 2 diet groups." No information on allocation concealment. "Demographic characteristics and cardiovascular risk factors were similar between groups." No formal hypothesis tests were reported, but baseline values in Table 1 showed no apparent imbalance. |
Low risk of bias |
No information about participant blinding, however, blinding of participants is highly unlikely. "...dietitians were not blinded to the study hypothesis." No reasons or details about deviations from intended interventions reported. "All analyses were conducted using the intention‐to‐treat principle." n = 148 randomised, and n = 148 analysed. |
Low risk of bias |
13/75 (17%) participants in the intervention and 9/73 (12%) participants in the control arm did not have data available. Imputed missing values, and performed sensitivity analyses to determine whether missing at random (MAR) assumptions were met. "We performed sensitivity analyses to assess the robustness of our conclusions and departures from the MAR assumption." "We used Markovchain Monte Carlo techniques to impute missing values, including additional covariates..." "Results of sensitivity analyses using multiple imputation techniques to impute missing values were consistent with those presented in our primary analyses." |
Low risk of bias |
Body weight was measured to the nearest 0.1 kg using a single calibrated scale. No evidence that a different method was used across intervention groups to measure weight. "all outcome assessors were blinded to the diet group assignment..." |
Some concerns |
No mention of a prespecified analysis plan. "Study protocol and data set: Not available." Trial registry outcomes amended after study publication; but trial registry entry at date prior to study publication mentioned the outcome and time points. All time points mentioned in trial registry at date prior to study publication, were reported on in publication. Analysis intentions were not reported in sufficient detail prior to study publication to enable an assessment. |
Some concerns |
No specific information on allocation concealment. Lack of information in the pre‐registered methods. |
| Benassi‐Evans 2009 |
Some concerns |
"Subjects were randomised to one of two dietary interventions…" No further information about allocation sequence or concealment. "Subject characteristics were not significantly different between groups at baseline." |
Low risk of bias |
No information, but blinding of participants and people delivering the interventions was highly unlikely. No reasons or details about deviations from intended interventions reported. No details about analysis were reported, but it looked to be intention‐to‐treat (number randomised per group = number analysed per group). |
High risk of bias |
No details about attrition reported, but some attrition was implied: "...in the subjects who completed the 12 weeks intensive weight loss and 9 months weight maintenance..." No further information or evidence that the result was not biased due to missing data were provided. |
Low risk of bias |
Volunteers fasted overnight and were weighed (model AMZ14; Mercury Digital Scales, Tokyo, Japan) wearing light clothing and no shoes at weeks. No evidence that a different method was used across intervention groups to measure weight. No information on blinding of outcome assessors. Weight is not an observer‐reported outcome involving judgement. |
Some concerns |
No mention of a prespecified analysis plan. Analysis intentions were not reported in sufficient detail to enable an assessment. |
High risk of bias |
No specific information on allocation concealment. Attrition was implied, with no reasons or evidence to show that results were not biased by this missingness. No pre‐registered document available to assess analysis intentions. |
| Brehm 2003 |
Some concerns |
"... after each block of subjects was assessed, the principal investigator used a random number table to randomly assign those subjects to one of two diets." No information on allocation concealment. No significant differences at baseline. |
Low risk of bias |
No information, however, blinding of participants and people delivering these interventions is highly unlikely. Deviations reported are expected to arise in usual care, e.g. dislike diet, could not make scheduled visit. Modified ITT: excluded only missing patients |
High risk of bias |
7/26 (27%) in the intervention and 14/27 (52%) in the control group did not have data. "In this intention to treat analysis, the initial weights for the subjects who withdrew from the study were used as their follow‐up weights at 3 and 6 months..." Lost to follow‐up and withdrawal reasons were stated for all participants, but very high attrition in the control group and differential attrition, thus missingness could and is likely to depend on the true value. |
Some concerns |
Subjects were weighed on a single electronic scale. No evidence that a different method was used across intervention groups to measure weight. No details about blinding of outcome assessors measuring weight. Weight is not an observer‐reported outcome involving judgement. |
Some concerns |
No mention of prespecified analysis plan. Analysis intentions were not reported in sufficient detail to enable assessment. |
High risk of bias |
No specific information on allocation concealment. Differential attrition and high attrition in the control group, therefore missingness could and is likely to be dependent on the true value of the outcome. It was not clear whether outcome assessors were blinded to allocation. No pre‐registered document available to assess analysis intentions. |
| Brehm 2005 |
Some concerns |
"After each block of subjects was enrolled, the principal investigator used a computerized randomization program to randomly assign those subjects to one of two diets." No information on allocation concealment. No significant differences between groups at baseline (completers only). |
Low risk of bias |
No information, however, blinding of participants and people delivering these interventions is highly unlikely. Deviations reported are expected to arise in usual care, e.g. pregnancy, disliked diet. Modified ITT: all those who completed the study were analysed. |
High risk of bias |
5/25 (20%) in the intervention and 5/25 (20%) in the control group did not have data available. Difficult to judge whether it was likely that missingness was related to the true value for weight, as reasons were not explicit enough. "The most common reason for discontinuing the study was the inability to commit to weekly counseling sessions. Other reasons included pregnancy, new out‐of‐state job, dislike of the low‐carbohydrate diet, and advice against following the low carbohydrate diet by family physician." |
Low risk of bias |
Weighed on a single electronic scale. No evidence that a different method was used across intervention groups to measure weight. No details about blinding of outcome assessors measuring weight. Weight is not an observer‐reported outcome involving judgement. |
Some concerns |
No mention of prespecified analysis plan. Analysis intentions were not reported in sufficient detail to enable assessment. |
High risk of bias |
No specific information on allocation concealment. Reasons for attrition were vague and it was difficult to judge whether it was as a result of the true value of the outcome, but with 20% attrition in both groups it is likely that some missingness was dependent on true value. No pre‐registered document available to assess analysis intentions. |
| Calleja‐Fernández 2012 |
Some concerns |
Random allocation to each dietary group was performed by the main investigator (MDBP) by using the Web site Randomization.com (http://www.randomization.com) upon the request of the study dietitian (ACF) after the eligibility of a participant was confirmed and assignment to either of the insulin resistance or insulin sensitivity groups had been performed. No information on allocation concealment. No problematic imbalance in baseline characteristics between groups suspected. |
Low risk of bias |
No information on blinding of participants, however, blinding was highly unlikely. People delivering these dietary interventions were unblinded. No reasons or details about deviations from intended interventions reported. ITT analysis conducted. |
Low risk of bias |
It seemed data were available for all participants randomised for 16‐week study, but we could not be sure (flowcharts in Fernandez et al 2012 and Ballesteros‐Pomar 2009 incongruent). |
Low risk of bias |
"Body weight was measured to the nearest 0.5 kg and then taken using calibrated digital scales…" No evidence that a different method was used across intervention groups to measure weight. "The medical assessments were blinded to the type of diet being followed." |
Some concerns |
No mention of a prespecified analysis plan. Analysis intentions were not reported in sufficient detail to enable an assessment. |
Some concerns |
No specific information on allocation concealment. No pre‐registered document available to assess analysis intentions. |
| Cornier 2005 |
Some concerns |
No information on how randomisation was done or allocation concealment. "Subject characteristics are summarized in Table 1. BW, BMI, percentage body fat, RMR, and RQ were similar in all groups." |
Low risk of bias |
No information, however, blinding of participants and people delivering these interventions was highly unlikely. "Forty‐four obese, healthy normoglycemic women 23 to 53 years old with BMI of 30 to 35 kg/m2 were screened, and 21 of them were enrolled and completed the 16‐week intervention." ITT analysis conducted: number randomised equalled number analysed; no subjects lost to follow‐up. |
Low risk of bias |
Number randomised equalled number analysed; no subjects lost to follow‐up. |
Low risk of bias |
Once a week, subjects were weighed. No evidence that a different method was used across intervention groups to measure weight. No information on blinding of outcome assessors. Weight is not an observer‐reported outcome involving judgement. |
Some concerns |
No mention of a prespecified analysis plan. Analysis intentions were not reported in sufficient detail to enable an assessment. |
Some concerns |
No information on how randomisation was done and whether, or how, allocation concealment was done. No pre‐registered document available to assess analysis intentions. |
| Dyson 2007 |
Low risk of bias |
"Randomization was undertaken by means of sealed envelopes equivalent to the number of subjects and filled fifty‐fifty with an indicator of either a low‐carbohydrate diet or healthy‐eating advice. Two separate sets of envelopes were prepared for both diabetic and non‐diabetic subjects. An independent observer witnessed randomization." Author reported in correspondence that no formal test of hypothesis was done for difference at baseline for non‐diabetic participants. Baseline characteristics table showed differences which were consistent with differences resulting by chance with very small sample sizes. |
Low risk of bias |
No information, however, blinding of participants and people delivering these dietary interventions was highly unlikely. Deviations reported are expected to arise in usual care: n = 3 of randomised participants in balanced group refused follow‐up as not allocated to low‐carbohydrate diet. Modified ITT conducted. |
High risk of bias |
No participants in the intervention group and 3/7 (43%) in the control group did not have data. Modified ITT with last‐observation‐carried‐forward. High attrition in control group, and differential attrition. Refusal of follow‐up due to not being allocated to low carbohydrate may be influenced by expected change in weight, therefore missingness could and was likely to be influenced by the true value. |
Low risk of bias |
Body weight recorded monthly. No evidence that a different method was used across intervention groups to measure weight. No details about blinding of outcome assessors. Weight is not an observer‐reported outcome involving judgement. |
Some concerns |
No mention of a prespecified analysis plan. Analysis intentions were not reported in sufficient detail to enable an assessment. |
High risk of bias |
Differential attrition and high attrition in the control group, with disappoinment at not being assigned to the low‐carbohydrate diet cited as a reason; therefore, missingness could be and was likely dependent on the true value of the outcome. No pre‐registered document available to assess analysis intentions. |
| Ebbeling 2007 |
Low risk of bias |
"Enrolled participants were entered sequentially onto a list of random group assignments prepared in advance by the study statistician, with stratification by sex and ethnicity/race..." "To avoid any bias in assigning participants to diet groups, staff conducting recruitment and enrollment were masked to sequence. The study director assigned participants to groups." No problematic differences in baseline characteristics between the two groups |
Low risk of bias |
No masking (open‐label trial). No reasons or details about deviations from intended interventions reported. ITT with imputation for missing values |
High risk of bias |
4/36 (11%) in the intervention and 3/37 (8%) in the control group did not have outcome data. "For body weight after dropout, an increase of 1 kg per year was imputed, starting from the last available measurement." No reasons for loss to follow‐up or withdrawal provided; could not judge if related to participants' health status and if missingness depended on its true value |
Low risk of bias |
Dietitians measured at each visit using electronic scale. No evidence that a different method was used across intervention groups to measure weight. "Data were collected by personnel who were masked to group assignment." |
Some concerns |
No mention of a prespecified analysis plan; trial registry did not mention the outcome and time point in any versions. Analysis intentions were not reported in sufficient detail to enable assessment. |
High risk of bias |
Attrition reported in both groups, with no reasons provided. Missingness was addressed through last‐observation‐carried‐forward. It was not possible to determine whether missingness could or was likely to be dependent on the true value of the outcome. No pre‐registered document available to assess analysis intentions |
| Farnsworth 2003 |
Some concerns |
No information on how the randomisation sequence was generated. No information on allocation concealment. "There were no significant differences in any of the variables between the diet groups." |
Some concerns |
No information on blinding of participants and people delivering these dietary interventions; however, blinding of these dietary interventions was highly unlikely. Deviations reported are expected to arise in usual care, e.g. work commitments, pregnancy. 66 randomised, 57 completed and 56 analysed. Per protocol analysis. "...data from one subject were excluded from the analysis because of the subject's noncompliance with the diet." 1/57 participant data excluded |
Low risk of bias |
5/33 (15%) in the intervention and 5/33 (15%) in the control group did not have data available. Data from n = 1 participant where missingness possibly influenced by true value. "Six subjects withdrew before study commencement because of family and work commitments; in addition, one subject withdrew because of pregnancy during the course of the study, the clinic lost contact with one subject, and data from one subject were excluded from the analysis because of the subject's noncompliance with the diet." "... one subject declined to undergo a DXA scan for personal reasons." |
Low risk of bias |
Body weight measured after an overnight fast. No evidence that a different method was used across intervention groups to measure weight. No information about blinding of outcome assessors. Weight is not an observer‐reported outcome involving judgement. |
Some concerns |
No mention of a prespecified analysis plan. Analysis intentions are not reported in sufficient detail to enable an assessment. |
Some concerns |
No information on how randomisation was done and whether, or how, allocation concealment was done. Data from one subject was excluded due to 'noncompliance' with the diet, but other reasons for loss to follow‐up were expected to arise in usual care. No pre‐registered document available to assess analysis intentions |
| Foraker 2014 |
Some concerns |
"Women were randomized using stratified randomization, based on BMI (< 30 kg/m2 vs. ≥ 30 kg/m2) to ensure equal distribution of overweight and obese participants into each intervention group." No information on allocation concealment. No problematic imbalance in baseline characteristics between groups |
Low risk of bias |
Double‐masked: investigator and outcomes assessor. Deviations reported were expected to arise in usual care, e.g. personal/family medical issues, time constraints. Modified ITT: completers analysed "number still under follow‐up". |
High risk of bias |
13/38 (34%) in the intervention and 15/41 (37%) in the control group did not have data available. High attrition in both groups with limited reporting of reasons, but likely that such high attrition missingness was possible and likely to depend on true value of outcome. |
Low risk of bias |
Anthropometric tests at each visit included body‐weight measurements. No evidence that a different method was used across intervention groups to measure weight. Double‐masked: investigator and outcomes assessor |
Some concerns |
No mention of prespecified data analysis plan, and trial registry at date prior to study publication did not mention the outcome and time point (mentioned only number of women who lost weight as an outcome). Analysis intentions were not reported in sufficient detail to enable an assessment; weight findings were emphasised as part of the primary aim in the paper's methods and trial registry; with limited reporting of these findings in the paper and none reported in the abstract. |
High risk of bias |
No specific information on allocation concealment. High attrition in both groups with limited reasons reported; therefore, missingness could and was likely to be dependent on the true value of the outcome. Lack of information in the pre‐registered methods. |
| Foster 2010 |
Some concerns |
"Using a random‐number generator, we randomly assigned participants within each site to treatment with either a low‐carbohydrate or low‐fat, calorie‐restricted diet for 2 years." No information on allocation concealment. No problematic differences observed between the two groups for baseline characteristics |
Low risk of bias |
Masking: none; open‐label. Deviations reported were expected to arise in usual care, e.g. dissatisfied with the programme, pregnancy, life stressors, relocated. ITT analysis with n = 307; all randomised. "We used a random‐effects linear model that was fitted to all observed data for each variable on each of the 307 participants for the primary analysis." |
Low risk of bias |
25/153 (16%) in the intervention and 19/154 (12%) in the control group did not have data available. "..longitudinal models implemented for this study relax this missing‐completely‐at‐random assumption in various ways." "Under the sensitivity analysis based on imputing missing outcomes with the highest (13.795) and lowest (‐18.355) random‐effects slopes (that is, change in weight per month) under the mixed‐effects model for weight, our qualitative findings were not sensitive to either imputation approach." Results of sensitivity analyses not provided |
Low risk of bias |
Body weight was measured at each treatment visit on calibrated scales while participants wore light clothing and no shoes. No evidence that a different method was used across intervention groups to measure weight. "Assessors were not blinded to treatment condition." Weight is not an observer‐reported outcome involving judgement. |
Some concerns |
No mention of prespecified analysis plan; trial registry at date prior to study publication mentioned outcome, but did not mention this time point. Analysis intentions were not reported in sufficient detail to enable an assessment. |
Some concerns |
No specific information on allocation concealment. Lack of information in the pre‐registered methods |
| Frisch 2009 |
Some concerns |
"Participants (n = 200) were randomly assigned by computer‐generated random number lists in two equal groups…" No information on allocation concealment. No problematic imbalance in baseline characteristics between groups suspected |
Low risk of bias |
No masking (open‐label trial). Deviations reported were expected to arise in usual care, e.g. non‐compliance. ITT analysis conducted |
High risk of bias |
5/100 (5%) in the intervention and 11/100 (11%) in the control group did not have data available. Data were evaluated by ITT and per protocol method. ITT done with baseline values carried forward. "Results did not differ when the per protocol method was used (data not shown)." 4/100 (4%) in the intervention and 10/100 (10%) in the control group withdrew due to non‐compliance, which may be related to change in weight (missingness possibly influenced by true value). |
Low risk of bias |
Participants were weighed using a calibrated electronic scale. No evidence that a different method was used across intervention groups to measure weight. Open‐label. Weight is not an observer‐reported outcome involving judgement. |
Some concerns |
No mention of prespecified analysis plan; trial registry at date prior to study publication mentioned outcome, but did not mention this time point. Analysis intentions were not reported in sufficient detail to enable an assessment. |
High risk of bias |
No specific information on allocation concealment. Attrition reported with reasons such as 'non‐compliance', therefore missingness could have and likely depended on the true value of the outcome. Both per protocol and baseline values carried forward approaches were used for missing data. Lack of information in the pre‐registered methods |
| Gardner 2007 |
Low risk of bias |
"Randomization was conducted in blocks of 24 (6 per treatment group) and occurred by having a blinded research technician select folded pieces of paper with group assignments from an opaque envelope." No problematic imbalance in baseline characteristics between groups suspected |
Low risk of bias |
Masking: single ("Clinic and laboratory staff members were blinded to treatment assignment"). Deviations reported were expected to arise in usual care, e.g. moved away, disliked diet, pregnancy. ITT analysis conducted |
High risk of bias |
6/77 (8%) in the intervention and 15/79 (19%) in the control did not have data available. ITT with baseline values carried forward. "Also for exploratory purposes, all analyses of weight and secondary outcome measures were tested using only available data, without using baseline values carried forward for missing data or other imputation methods. There were no substantive differences in any of these findings compared with the analyses with baseline values carried forward and, therefore, only the primary analyses are presented." 5/9 in the intervention and 5/18 in the control group withdrew due to health and "other" reasons (vague); but attrition was differential and thus missingness could and was likely to depend on the true value. |
Low risk of bias |
"Body weight was measured to the nearest 0.1 kg on a calibrated clinical scale." No evidence that a different method was used across intervention groups to measure weight. "Clinic and laboratory staff members were blinded to treatment assignment." |
Some concerns |
No mention of a prespecified analysis plan, trial registry at date prior to study publication mentioned outcome and time point. The time point mentioned in the trial registry at date prior to publication was reported on in the publication. Analysis intentions were not reported in sufficient detail to enable assessment. |
High risk of bias |
Differential attrition reported, with vague reasons; therefore, missingness could and was likely to depend on the true value of the outcome. Per protocol and baseline values carried forward approaches were used to account for missingness. Lack of information in the pre‐registered methods |
| Goni 2018 |
Some concerns |
"Subjects were randomly assigned to one of the two diets by a specific logarithm design for the study by MATLAB using stratified block randomization..." No information about allocation concealment. No significant differences between groups at baseline
|
Low risk of bias |
Masking: none; open‐label. The only deviations reported were dropouts. Modified ITT: looked as if only missing outcome data excluded |
High risk of bias |
21/75 (28%) in the intervention 19/72 (26%) in the control group did not have data available for this outcome. No reasons for dropouts provided but, with this high attrition in both groups, missingness was possibly and likely to depend on true value of outcome. |
Low risk of bias |
Assessed using a Tanita BC418 scale. No evidence that a different method was used across intervention groups to measure weight. Masking: none; open‐label. Weight is not an observer‐reported outcome involving judgement. |
Some concerns |
No mention of prespecified analysis plan, but trial registry at date prior to study publication mentioned the outcome and time point. The time point mentioned in the trial registry at the date prior to publication was reported on in the publication. Analysis intentions were not reported in sufficient detail to enable an assessment. |
High risk of bias |
No specific information on allocation concealment. High attrition reported in both groups with no reasons provided, therefore missingness could and was likely to be dependent on the true value of the outcome. Lack of information in the pre‐registered methods |
| Griffin 2013 |
Low risk of bias |
"The study consisted of a random block design per every four participants. Random allocation was prepared by a researcher not involved in participant enrolment or assignment." "random number table"; "sealed opaque envelopes" (trial registry). No problematic differences observed between the two groups for baseline characteristics |
Low risk of bias |
"Participants were blind to diet macronutrient composition and were informed eating plans differed only in the type of meat (i.e. red or white)." "...inability to double‐blind the intervention were limitations." Deviations reported were expected to arise in usual care, e.g. did not like diet. ITT analysis conducted |
High risk of bias |
12/36 (33%) participants in the intervention and 15/35 (43%) in the control group did not have data available. ITT with last‐observation‐carried‐forward, and sensitivity analysis comparing completers with ITT. 5/6 (83%) in the intervention and 5/9 (56%) in the control discontinued intervention due to illness, too difficult and not losing weight. Very high attrition in both groups; likely that, with such high attrition, missingness was possible and likely to depend on the true value. |
Low risk of bias |
Taken at every visit on a digital scale with light clothing and no shoes. No evidence that a different method was used across intervention groups to measure weight. "...inability to double‐blind the intervention were limitations." Single‐blind with blinded participants. Weight is not an observer‐reported outcome involving judgement. |
Some concerns |
No mention of prespecified analysis plan, but trial registry at date prior to study publication mentioned the outcome and time point. The time point mentioned in the trial registry at the date prior to publication is reported on in the publication. Analysis intentions were not reported in sufficient detail to enable assessment. |
High risk of bias |
Very high attrition reported in both groups, with reasons such as 'illness', diets being 'too difficult' and 'not losing weight'; missingness could be and was likely dependent on the true value of the outcome. Lack of information in the pre‐registered methods |
| Haufe 2013 |
Low risk of bias |
"For allocation of the subjects, a computer‐generated list of random numbers was used. The randomization sequence was created using SPSS 18 (Chicago, IL) statistical software and subjects were assigned to reduced carbohydrate or reduced fat diet with a 1:1 allocation using random block sizes of 2, 4, and 6. Study nurses and physicians screening and enrolling volunteers were blinded for the randomization sequence." "Both groups were well matched for gender, age, body weight, body mass index, blood lipid profiles, glucose metabolism, and cardiorespiratory fitness." |
High risk of bias |
"Blinding of participants for the allocated dietary intervention was impossible." "Except for the dieticians, study nurses and physicians were blinded for the treatment assignment." Deviations reported were expected to arise in usual care, e.g. lost to follow‐up, withdrew consent. Per protocol analysis conducted: 24/28 (86%) in the intervention and 30/33 (91%) in the control group excluded from the analysis including 'protocol violations' |
High risk of bias |
25/80 (31%) in the intervention and 29/83 (35%) in the control group did not have data available. ITT with last‐observation‐carried‐forward. High attrition in both groups with limited reporting of reasons, but likely that, with such high attrition, missingness was possibly and likely to depend on true value of outcome. |
Low risk of bias |
Measured body weight in a standardised manner after an overnight fast. No evidence that a different method was used across intervention groups to measure weight. "....study nurses and physicians were blinded for the treatment assignment." |
Some concerns |
No mention of prespecified analysis plan, but trial registry at date prior to study publication mentioned the outcome and time point. The time point mentioned in trial registry at the date prior to publication is reported on in the publication. Analysis intentions were not reported in sufficient detail to enable an assessment. |
High risk of bias |
The majority of data not included in the analyses were due to exclusions based on 'protocol violations'. High overall attrition was reported in both groups, with no reasons (aside from exclusions); missingness could be and was likely dependent on the true value of the outcome. Last‐observation‐carried‐forward approach used for dealing with missingness. Lack of information in the pre‐registered methods |
| Josse 2011 |
Some concerns |
No information about how the randomisation sequence was generated, or allocation concealment. "Our stratification scheme was adequate, because there were no between‐group differences in any baseline variable." |
Low risk of bias |
Masking: none; open‐label. Deviations reported were expected to arise in usual care, e.g. moving house, exam time at school, sickness of a relative. Modified ITT "...we conducted an intent‐to‐treat analysis where missing data values were imputed using model‐based multiple imputation." |
Low risk of bias |
4/30 (13%) participants in the intervention and 1/30 (3%) in the control group did not have data available for this outcome. The only information reported was "...we conducted an intent‐to‐treat analysis where missing data values were imputed using model based multiple imputation." Based on the reasons provided, it was unlikely that missingness could be dependent on the true value of the outcome. |
Low risk of bias |
Digital physician's scale used to determine weight. No evidence that a different method was used across intervention groups to measure weight. Masking: none; open‐label. Weight is not an observer‐reported outcome involving judgement. |
Some concerns |
No mention of prespecified analysis plan, but trial registry at the date prior to study publication mentioned the outcome and time point. The time point mentioned in the trial registry at the date prior to publication is reported on in the publication. Analysis intentions were not reported in sufficient detail to enable an assessment. |
Some concerns |
No information on how randomisation was done and whether, or how, allocation concealment was done. Lack of information in the pre‐registered methods |
| Keogh 2007 |
High risk of bias |
Clinstat used to generate sequence; "allocation concealment: No" (trial registry). From Table 1 there appeared to be no large differences between groups at baseline. |
Low risk of bias |
Open (masking not used). Deviations reported were expected to arise in usual care, e.g. personal reasons, unable to comply. Intention‐to‐treat: all randomised were analysed. |
Low risk of bias |
All participants, 13/13 in intervention and 12/12 in control, had data available. |
Low risk of bias |
Weight was measured using a Mercury digital scale in light clothing at baseline, and every 2 weeks. No evidence that a different method was used across intervention groups to measure weight. Open (masking not used). Weight is not an observer‐reported outcome involving judgement. |
Some concerns |
No mention of prespecified analysis plan, but trial registry at date prior to study publication mentioned the outcome and time point. The time point mentioned in the trial registry at the date prior to publication is reported on in the publication. Analysis intentions were not reported in sufficient detail to enable an assessment. |
High risk of bias |
Allocation was not concealed. Lack of information in the pre‐registered methods |
| Klemsdal 2010 |
Some concerns |
No information about randomisation or allocation concealment. Footnote to Table 1: "There were no significant differences between groups." |
Low risk of bias |
Masking: None (open‐label). Deviations reported were vague, but those reported were expected to arise in usual care, e.g. non‐attendance of study visits. ITT analysis: "All analyses were done on an intention to treat basis, with the last value carried forward for non‐completers." All randomised participants were analysed. |
High risk of bias |
Short‐term outcomes: 16% of participants overall did not have data available (not reported per trial arm). "All analyses were done on an intention to treat basis, with the last value carried forward for non‐completers." No mention of sensitivity analysis. No reasons for lost to follow‐up or withdrawal provided and therefore we could not make a judgement about whether missingness was dependent on the true value of the outcome. |
Low risk of bias |
"Body weight...measured with standard methods." No evidence that a different method was used across intervention groups to measure weight. Masking: None (open ‐label). Weight is not an observer‐reported outcome involving judgement. |
Some concerns |
No mention of prespecified analysis plan, but trial registry at date prior to the study publication mentioned the outcome and time point. The time point mentioned in the trial registry at the date prior to publication was reported on in the publication. Analysis intentions were not reported in sufficient detail to enable an assessment. |
High risk of bias |
No information on how randomisation was done and whether, or how, allocation concealment was done. Attrition with no reasons reported for the overall group; it was therefore not possible to determine whether missingness could be or was likely dependent on the true value of the outcome. Missingness was addressed by using last‐observation‐carried‐forward. Lack of information in the pre‐registered methods |
| Landers 2002 |
Some concerns |
No information on randomisation or allocation concealment. Families were assigned to the same dietary intervention. Baseline characteristics were not reported, with the exception of body weight, fat mass and lean mass; all of which did not appear to be significantly different. |
Low risk of bias |
No information about blinding, however, blinding of participants and people delivering these dietary interventions was highly unlikely. Deviations reported were expected to arise in usual care, e.g. did not attend follow‐up visits. Modified ITT excluding participants with missing outcome data |
High risk of bias |
Data were not available for 12/28 (43%) participants in the intervention and 12/33 (36%) in the control group. High attrition in both groups with limited reporting of reasons, but likely that, with such high attrition, missingness was possibly and likely to depend on true value of outcome. |
Low risk of bias |
Measured in light clothing without shoes at the initial visit. No evidence that a different method was used across intervention groups to measure weight. No information about blinding of outcome assessors. Weight is not an observer‐reported outcome involving judgement. |
Some concerns |
No mention of prespecified data analysis plan. Analysis intentions were not reported in sufficient detail to enable an assessment. |
High risk of bias |
No information on how randomisation was done and whether, or how, allocation concealment was done. High attrition reported in both groups, therefore missingness could be and was likely to depend on the true value of the outcome. No pre‐registered document available to assess analysis intentions |
| Lasker 2008 |
Some concerns |
"Subjects were blocked on gender, matched on age...and fasting glucose..." No information on allocation concealment. Significant baseline imbalance for insulin; no significant differences for 13 other characteristics at baseline |
High risk of bias |
No information about blinding, however blinding of participants and people delivering these dietary interventions was highly unlikely. Deviations reported were expected to arise in usual care, e.g. training for marathon. Per protocol analysis (participants excluded for lack of adherence to protocol: 23% excluded for various reasons, including lack of adherence, but breakdown by trial arm and reason not reported) |
High risk of bias |
Data were not available for 7/32 (22%) participants in the intervention and 8/33 (24%) in the control group. High attrition with reasons including non‐compliance, which may be related to change in weight, so likely that with such high attrition and this reason, some missingness was possibly and likely to depend on true value of outcome |
Low risk of bias |
Measured using an electronic scale (Tanita). No evidence that a different method was used across intervention groups to measure weight. No information about blinding of outcome assessors. Weight is not an observer‐reported outcome involving judgement. |
Some concerns |
No mention of prespecified data analysis plan. Analysis intentions were not reported in sufficient detail to enable an assessment. |
High risk of bias |
No specific information on allocation concealment. Some participants were excluded from the analysis due to a lack of adherence, with high attrition reported in both trial arms. Based on these numbers and known exclusions for lack of adherence, missingness could be and was likely dependent on the true value of the outcome. No pre‐registered document available to assess analysis intentions |
| Layman 2005 |
Some concerns |
"Note that the randomization was blocked on age and BMI for each wave of recruitment." No information on allocation concealment. "Baseline characteristics did not differ among subjects in each of the treatment groups..."
|
Low risk of bias |
No information about blinding, however blinding of participants and people delivering these dietary interventions was highly unlikely. No deviations or dropouts reported. Appeared to be a full ITT with number randomised equal to number analysed |
Low risk of bias |
All randomised participants appeared to have been analysed. |
Low risk of bias |
Measured using an electronic scale (Tanita). No evidence that a different method was used across intervention groups to measure weight. No information about blinding of outcome assessors. Weight is not an observer‐reported outcome involving judgement. |
Some concerns |
No mention of prespecified data analysis plan. Analysis intentions were not reported in sufficient detail to enable an assessment. |
Some concerns |
No specific information on allocation concealment. No pre‐registered document available to assess analysis intentions |
| Layman 2009 |
Some concerns |
No information on how the randomisation sequence was generated. No information about allocation concealment. "Randomization after stratifying participants for gender, age, BMI, and TC was successful with no significant differences between treatment groups in baseline characteristics." |
Low risk of bias |
"Due to study design of the diet treatment, it was not possible to blind subjects and research staff to group assignment." Deviations were expected to arise in usual care, e.g. personal reasons, unable to comply. "Prior to the 4‐mo data collection, 15 participants withdrew from the CHO group (9 stated personal reasons and 6 stated unable to comply with diet protocol) and 12 participants withdrew from the PRO group (11 stated personal reasons and 1 stated inability to comply with diet protocol)." Modified ITT conducted |
High risk of bias |
12/64 (19%) in the intervention and 15/66 (23%) in the control group did not have data available. No methods used to correct for bias, or sensitivity analyses conducted. 6/15 (40%) withdrawals in the intervention and 1/12 (8%) withdrawals in the control group were due to inability to comply with diet. Inability to comply may have been related to change in weight, and high attrition in the control group, thus missingness could and was likely to depend on the true value. |
Low risk of bias |
Measured using an electronic scale (Tanita). No evidence that a different method was used across intervention groups to measure weight. "Due to study design of the diet treatment, it was not possible to blind subjects and research staff to group assignment." Weight is not an observer‐reported outcome involving judgement. |
Some concerns |
No mention of prespecified data analysis plan. Analysis intentions were not reported in sufficient detail to enable an assessment. |
High risk of bias |
No information on how randomisation was done and whether, or how, allocation concealment was done. Attrition reported in both groups, with high attrition in the control group, and some attrition in both groups due to an 'inability to comply with diet'; therefore missingness could and was likely to depend on the true value of the outcome. No pre‐registered document available to assess analysis intentions |
| Lean 1997 |
Some concerns |
No information on randomisation or allocation concealment. "All variables were not significantly different between the two diet treatment groups." |
Low risk of bias |
No information about blinding, however blinding of participants and people delivering these dietary interventions was highly unlikely. No reasons or details about deviations from intended interventions reported. ITT with last‐observation‐carried‐forward conducted |
High risk of bias |
13/53 (25%) in the intervention and 15/57 (26%) in the control group did not contribute data to this outcome. ITT with last‐observation‐carried‐forward conducted. High attrition in both groups with no reasons reported but, with this high attrition, some missingness was possibly and likely to depend on true value of outcome. |
Low risk of bias |
"Measurements were made with subjects wearing light clothes using regularly calibrated scales..." Reference to WHO procedures (classification and description of anthropometric data). No evidence that a different method was used across intervention groups to measure weight. No information on blinding of outcome assessors. Weight is not an observer‐reported outcome involving judgement. |
Some concerns |
No mention of prespecified data analysis plan. Analysis intentions were not reported in sufficient detail to enable an assessment. |
High risk of bias |
No information on how randomisation was done and whether, or how, allocation concealment was done. High attrition in both groups, with no reasons reported; therefore missingness could and was likely to be dependent on the true value of the outcome. No pre‐registered document available to assess analysis intentions |
| Lim 2010 |
Some concerns |
No information about randomisation or allocation concealment. Baseline characteristics table showed small differences, consistent with differences resulting by chance with small sample sizes. |
Low risk of bias |
No information about blinding, however blinding of participants and people delivering these dietary interventions was highly unlikely. Deviations reported were expected to arise in usual care, e.g. travel, work commitment, did not like diet. Modified ITT with all partial data contributed by non‐completers included in linear mixed effects model |
High risk of bias |
6/30 (20%) in the intervention and 8/30 (27%) in the control group did not have data available. ITT with all partial data contributed by non‐completers included in linear mixed effects model. Results of ITT not presented and no sensitivity analyses performed (author correspondence). High attrition in control group and some missingness could possibly and was likely to depend on true value of outcome. |
Low risk of bias |
Recorded in light clothing on Mercury digital scales. No evidence that a different method was used across intervention groups to measure weight. No information about blinding. Weight is not an observer‐reported outcome involving judgement. |
Some concerns |
No mention of prespecified data analysis plan. Analysis intentions were not reported in sufficient detail to enable an assessment. |
High risk of bias |
No information on how randomisation was done and whether, or how, allocation concealment was done. High attrition reported in both groups, therefore missingness could be and was likely to depend on the true value of the outcome. Partial data from missing participants included in linear mixed effects model, but results of this ITT analysis not reported. No pre‐registered document available to assess analysis intentions |
| Liu 2013 |
Low risk of bias |
No information about sequence generation. "Randomisation was conducted by a statistician who was not involved in any other aspects of the study by using block randomisation, stratified by 10‐year age categories." "Although the participants were randomised by age, baseline BMI levels of the two groups were similar after randomisation." |
Low risk of bias |
No information about blinding, however blinding of participants and people delivering these dietary interventions was highly unlikely. Deviations reported were expected to arise in usual care, e.g. personal reasons, busy schedule. Modified ITT: one dropout due to busy schedule in control group excluded |
Low risk of bias |
1/25 (4%) in the intervention and 1/25 (4%) in the control group did not have data available for this outcome. |
Low risk of bias |
"Body weight was measured to the nearest 0.1 kg at the beginning and then biweekly during the trial using an electronic scale." No evidence that a different method was used across intervention groups to measure weight. Masking: single (outcomes assessor) |
Some concerns |
No mention of prespecified analysis plan, but trial registry at date prior to study publication mentioned the outcome and time point. The time point mentioned in the trial registry at the date prior to publication is reported on in the publication. Analysis intentions were not reported in sufficient detail to enable an assessment. |
Some concerns |
Lack of information in the pre‐registered methods |
| Marco‐Benedi 2019 |
Low risk of bias |
"Randomization was performed by using an online software..." Paper reported "...participants and the staff, except for the nutritionists, were blinded to the type of diet individuals were assigned." Likely referring to allocation concealment, not blinding (trial registry reported 'Open‐label'). Only one out of numerous outcomes was significantly different between groups at baseline. |
Low risk of bias |
Trial registry stated "Masking: None (open‐label)", but paper reported "...participants and the staff, except for the nutritionists, were blinded to the type of diet individuals were assigned." Likely referring to allocation concealment, not blinding. Deviations reported were expected to arise in usual care, e.g. change of residence, personal reasons. Modified ITT conducted: "All subjects who completed the study were included in the data analysis, independent of reported dietary compliance, as indicated by food records, or weight loss according to intention‐to‐treat analysis." |
High risk of bias |
5/40 (12.5%) in the intervention and 8/40 (20%) in the control group did not have data available for this outcome. Reason for withdrawal was only available for 1/13 excluded participants. Difficult to judge whether it was likely that missingness was related to true value for 12/13 participants, based on reasons provided (unknown, personal reasons) |
Low risk of bias |
"Body weight was measured...with a calibrated scale." No evidence that a different method was used across intervention groups to measure weight. Trial registry stated "Masking: None (open‐label)". Weight is not an observer‐reported outcome involving judgement. |
Some concerns |
No mention of prespecified analysis plan, but trial registry at the date prior to the study publication mentioned the outcome and time point. The time point mentioned in the trial registry at the date prior to publication was reported on in the publication. Analysis intentions were not reported in sufficient detail to enable an assessment. |
High risk of bias |
Attrition reported in both groups, with vague reasons for loss to follow‐up and one exclusion; therefore, it was not possible to judge whether missingness could be or was likely dependent on the true value of the outcome. Lack of information in the pre‐registered methods |
| Mateo‐Gallego 2017 |
Some concerns |
"We carried out the randomization process by generating a number sequence (including 93 correlative numbers because this was the number of subjects we planned to randomize) in the following webpage: https://www.random.org. The number for each participant was correlatively assigned (e.g; the first randomized women was assigned with number 1)" (from author correspondence). No information about allocation concealment. "There were no significant between‐group differences at baseline." |
Low risk of bias |
"Participants were blinded to their assigned macronutrient composition." "The dietician who formulated the diets and carried out the individual consultations was aware of each participant's group assignment, but the rest of the staff was blinded to that information." Study was reported as single‐blind (masking: single (participant)). Deviations reported were expected to arise in usual care, e.g. change of residence, personal reasons. Modified ITT conducted: "All subjects who completed the study were included in the data analysis, independent of reported dietary compliance, as indicated by food records, or weight loss according to intention‐to‐treat analysis." |
High risk of bias |
3/30 (10%) in the intervention and 6/30 (20%) in the control group did not have data available for this outcome. Limited reporting of reasons for missingness, therefore difficult to judge whether missingness could or was likely to be dependent on the true value of the outcome |
Low risk of bias |
"Body weight was measured in subjects without shoes to the nearest 0.1 kg with a calibrated scale." No evidence that a different method was used across intervention groups to measure weight. "The dietician who formulated the diets and carried out the individual consultations was aware of each participant's group assignment, but the rest of the staff was blinded to that information." |
Some concerns |
No mention of prespecified analysis plan, but trial registry at the date prior to the study publication mentioned the outcome and time point. The time point mentioned in the trial registry at the date prior to publication was reported on in the publication. Analysis intentions were not reported in sufficient detail to enable an assessment. |
High risk of bias |
No specific information on allocation concealment. Attrition reported in both groups with limited reporting of reasons, therefore, it is not possible to judge if missingness could be or was likely dependent on the true value of the outcome. Lack of information in the pre‐registered methods |
| Ooi 2021 |
Some concerns |
"This was a randomized single‐blinded, placebo‐controlled study where subjects were block randomly assigned by gender and BMI groups…" Allocation concealment: no trial registry available and no information in paper. No significant differences between groups at baseline |
High risk of bias |
Subjects blinded only to supplements. No information about blinding of diets, however, blinding of participants highly unlikely, the study was stated as single‐blind, thus people delivering these dietary interventions were unblinded. Deviations reported were expected to arise in usual care e.g family, work or personal commitments. Per protocol analysis, with exclusion of available data for reasons such as non‐compliance and inadequate weight loss. Vague reasons for exclusions and specific numbers excluded per trial arm were unclear, with 8% of subjects being excluded by the principal investigator. |
High risk of bias |
3/43 (7%) in the intervention and 7/44 (16%) in the control group did not have data available. "Study subjects with available information for each analysis were included and missing data were not imputed." Study subjects that did not comply with the diets or did not lose a specific amount of weight were excluded from analyses by the investigators; thus missingness in the outcome could and was likely to depend on its true value. |
Low risk of bias |
"Height and weight were taken using a SECA 763 digital scale…" No evidence that a different method was used across intervention groups to measure weight. The study was single‐blinded for participants to supplements. Weight is not an observer‐reported outcome involving judgement. |
Some concerns |
No mention of prespecified analysis plan, no trial registry. Analysis intentions were not reported in sufficient detail to enable an assessment. |
High risk of bias |
No specific information on allocation concealment. Subjects were likely not blinded to dietary allocation. Per protocol analysis conducted with non‐compliant participants excluded. Missing data were not imputed. Exclusions of available data were done based on non‐compliance with diet or inadequate weight loss, thus missingness in the outcome could and was likely to depend on its true value. No pre‐registered document available to assess analysis intentions |
| Parr 2016 |
High risk of bias |
"The randomization will be initially stratified by baseline BMI values (27–29, 30–34, and 35–40 kg/m2) and sex. Subsequently, simple randomisation by roll of dice within the stratified segregation will be used to allocate individuals to each of the 3 groups." "Allocation is not concealed" "No significant differences were observed among groups at baseline for any variable (P > 0.05)." |
High risk of bias |
"From then on, apart from the study dietitian, all investigators were blinded to the assigned dietary group by using participant codes unrelated to their dietary assignment." From registry: open (masking not used). Deviations reported were expected to arise in usual care, e.g. time commitment, ankle injury. Per protocol (participants excluded from analysis due to compliance). 5/10 (50%) excluded due to non‐compliance in the intervention group and 3/8 (43%) in the control group due to non‐compliance and 1/8 (12.5%) due to being an outlier |
High risk of bias |
10/39 (14%) in the intervention and 8/40 (20%) in the control group did not have data available for this outcome. 5/10 (50%) excluded due to non‐compliance in the intervention group and 3/8 (43%) in the control group due to non‐compliance, which may have been related to change in weight; thus, missingness could and was likely dependent on the true value. |
Low risk of bias |
Measured with DEXA. No evidence that a different method was used across intervention groups to measure weight. "From then on, apart from the study dietitian, all investigators were blinded to the assigned dietary group by using participant codes unrelated to their dietary assignment.", but from the registry: "open (masking not used)". |
Some concerns |
No mention of prespecified analysis plan, but trial registry at date prior to study publication mentioned the outcome and time point. The time point mentioned in the trial registry at the date prior to publication was reported on in the publication. Analysis intentions were not reported in sufficient detail to enable an assessment. |
High risk of bias |
Allocation was not concealed. Attrition was reported in both groups, with nearly half of all missing data due to participants being excluded from the per protocol analysis due to 'non‐compliance' and being an 'outlier'. Missingness could and was likely to be dependent on the true value of the outcome. Lack of information in the pre‐registered methods |
| Sacks 2009 |
Low risk of bias |
"Random assignments to one of four diet groups were generated by the data manager at the coordinating center on request of a study dietitian, after eligibility of a participant was confirmed." Independent trial unit, fully funded by public funds and large trial; it was likely the sequence was random. " Baseline characteristics were similar among participants assigned to the four diets..." |
Low risk of bias |
"Blinding was maintained by the use of similar foods for each diet. Staff and participants were taught that each diet adhered to principles of a healthful diet..." True blinding is a known challenge with dietary interventions. Deviations reported were expected to arise in usual care, e.g. moved away, did not appear at scheduled appointments. Full ITT: "We performed an intention‐to‐treat analysis in which long‐term weight loss for persons who withdrew from the study early (after at least 6 months of participation) was imputed on the basis of a rate of 0.3 kg per month of regained weight..." |
High risk of bias |
Linked paper reported 39/201 (19%) in the intervention and 47/204 (23%) in the control group did not have data available for this outcome at 6 months (De Jonge et al 2012); 2‐year attrition (Sacks 2009) reported less attrition per group. Single imputation "We performed an intention‐to‐treat analysis in which long‐term weight loss for persons who withdrew from the study early (after at least 6 months of participation) was imputed on the basis of a rate of 0.3 kg per month of regained weight..." No sensitivity analyses (author correspondence). No detailed reasons for loss to follow‐up or withdrawals provided; high attrition in the control group thus missingness could and was likely to depend on the true value. |
Low risk of bias |
"Body weight, the primary outcome variable, was measured by calibrated hospital scales, in the morning before breakfast and after urinating, clothed in a hospital gown, on two nonconsecutive days at baseline, and at 6 and 24 months; and on a single day at 12 and 18 months." No evidence that a different method was used across intervention groups to measure weight. "Investigators and staff who measured outcomes were unaware of the diet assignment of the participants." |
Some concerns |
No mention of prespecified analysis plan; trial registry at date prior to study publication mentioned outcome, but did not mention this time point. Analysis intentions were not reported in sufficient detail to enable an assessment. |
High risk of bias |
Attrition reported in both groups, with high attrition in the control group, and limited reporting of reasons; therefore missingness could be and was likely dependent on the true value of the outcome. Simple imputation was performed, but no sensitivity analyses. Lack of information in the pre‐registered methods |
| Tay 2008 |
High risk of bias |
From registry: "random allocation using computer software clinstat" "allocation is not concealed". "Baseline characteristics between diet groups were not different". |
High risk of bias |
From registry: "Open (masking not used)". Deviations reported were expected to arise in usual care, e.g. work commitments, personal reasons. Per protocol analysis, ITT with last‐observation‐carried‐forward also performed but data not reported. 4/10 (40%) participants in the intervention group and 2/11 (18%) in the control group were excluded from analysis due to non‐compliance or unclear reasons. |
High risk of bias |
10/55 (18%) in the intervention and 11/54 (20%) in the control group did not have data available for this outcome. ITT with last‐observation‐carried‐forward: "The ITT with either baseline values carried forward or the last follow‐up visit carried forward for those who did not complete the study also showed no difference in weight loss between the diets (P <= 0.23)." 4/10 (40%) participants in the intervention group and 2/11 (18%) in the control group were excluded from analysis due to non‐compliance or unclear reasons, which may be related to change in weight; thus missingness could or was likely dependent on the true value of the outcome. |
Low risk of bias |
"Body mass was measured using calibrated electronic digital scales...to the nearest 0.05 kg, with subjects wearing light clothing and no footwear." No evidence that a different method was used across intervention groups to measure weight. From registry: "Open (masking not used)". Weight is not an observer‐reported outcome involving judgement. |
Some concerns |
No mention of prespecified analysis plan; trial registry at date prior to study publication mentioned outcome, but did not mention this time point. Analysis intentions were not reported in sufficient detail to enable an assessment. |
High risk of bias |
Allocation was not concealed. Attrition was reported in both trial arms, with high attrition in the control group due to unclear reasons and 'non‐compliance'; therefore, missingness could and was likely to be related to the true value of the outcome. ITT analyses with baseline values and last‐observations‐carried‐forward were conducted, but neither were considered appropriate to address potential bias due to missingness. Lack of information in the pre‐registered methods |
| Veum 2017 |
Some concerns |
"The list of prescreening appointments was used to randomly assign the participants by drawing ballots and blocking with block sizes of 2." No information about allocation concealment. No problematic imbalance in baseline characteristics between groups suspected |
Low risk of bias |
From registry: "Masking: None (open‐label)" "All participants were informed of group allocation after the baseline measurements and sampling. The nature of the trial required an open intervention with no blinding of the trial participants or the investigators." Deviations reported were expected to arise in usual care, e.g. busy job, fracture. "We also performed an intention‐to‐treat (ITT) analysis that included all randomly assigned participants except 2 cases that were lost before baseline (44 eligible participants in total). These 2 were unaware of the group allocation and did not provide any dietary or clinical data." |
Low risk of bias |
4/24 (17%) in the intervention and 4/22 (18%) in the control group did not have data available for this outcome. "Notably, the results from the PP and ITT analyses did not differ in nominal significance for any of the primary or secondary outcomes. Our missing data analysis (data not shown), including Little’s missing completely at random test, indicated that the dropouts were missing (completely) at random, and we found no significant differences between dropouts and completers in baseline characteristics, except for significant higher diastolic blood pressure, total cholesterol (TC), and LDL cholesterol in completers. In the reported PP and ITT analyses, we used mixed models without imputation to handle missing data. However, we also conducted an ITT analysis based on a full dataset in which missing values in the original data were replaced by values from multiple imputation (data not shown)." |
Low risk of bias |
Taken barefoot and wearing light clothing and asked to use restroom shortly before. No evidence that a different method was used across intervention groups to measure weight. "The nature of the trial required an open intervention with no blinding of the trial participants or the investigators." Weight is not an observer‐reported outcome involving judgement. |
Some concerns |
No mention of prespecified analysis plan; trial registry at date prior to study publication mentioned outcome, but did not mention this time point. Analysis intentions were not reported in sufficient detail to enable an assessment. |
Some concerns |
No specific information on allocation concealment. Lack of information in the pre‐registered methods |
| Volek 2009 |
Some concerns |
No information about randomisation or allocation concealment. No apparent imbalances between two groups at baseline |
Low risk of bias |
No information about blinding, however blinding of participants and people delivering these dietary interventions was highly unlikely. No deviations or dropouts reported. Appeared to be a full ITT with number randomised equal to number analysed |
Low risk of bias |
Number randomised equalled number analysed. |
Low risk of bias |
"Body mass was measured in the morning after an overnight fast to the nearest 100 g on a calibrated digital scale." No evidence that a different method was used across intervention groups to measure weight. "Analyses were performed by the same blinded technician." |
Some concerns |
No mention of prespecified data analysis plan. Analysis intentions were not reported in sufficient detail to enable an assessment. |
Some concerns |
No information on how randomisation was done and whether, or how, allocation concealment was done. No pre‐registered document available to assess analysis intentions |
| Wycherley 2012 |
High risk of bias |
"...participants were blocked, matched for age and body mass index, then randomised by the trial coordinators using computer‐generated random number allocation..." Allocation concealment: No (trial registry). From Table 3, only triglycerides were significantly different between groups at baseline (P < 0.05), of 17 characteristics measured. |
Some concerns |
Open (masking not used). Deviations reported were expected to arise in usual care, e.g. relocated overseas, unable to commit to study protocol/time requirements. Per protocol analysis: 3/59 and 1/64 participants excluded due to being unable to commit to study protocol or did not receive intervention. 3% of total not included (4/123) |
High risk of bias |
7/59 (12%) in the intervention and 6/64 (9%) in the control group did not have data available. Modified ITT only done at 52 weeks using maximal likelihood mixed model analysis. "Overall, the participants who withdrew after week 12 were those who lost less weight during weeks 0–12 than those who completed the full 52 weeks..."; missingness could and was likely to depend on the true value. |
Low risk of bias |
"... body weight was measured using calibrated electronic digital scales (Mercury; AMZ 14, Tokyo, Japan)." No evidence that a different method was used across intervention groups to measure weight. Open (masking not used). Weight is not an observer‐reported outcome involving judgement. |
Some concerns |
No mention of prespecified analysis plan, but trial registry at date prior to study publication mentioned the outcome and time point. The time point mentioned in the trial registry at the date prior to publication was reported on in the publication. Analysis intentions were not reported in sufficient detail to enable an assessment. |
High risk of bias |
Allocation was not concealed. Attrition was reported in both groups, and authors remarked that "...participants who withdrew after week 12 were those who lost less weight during weeks 0–12 than those who completed the full 52 weeks…; missingness was therefore likely dependent on the true value of the outcome. Furthermore, participants were excluded for an inability to commit to the study protocol or not receiving the intervention. Lack of information in the pre‐registered methods |
