Study designs of evaluations included in the review
Only studies judged to be of a moderate or strong study design were included in the review. Strong designs included double-blind randomised controlled trials (RCTs) or crossover trials; moderate designs included unblinded RCTs, or double-blinded trials in which participants were their own controls without randomised crossover or blinded before- and-after designs; weak designs included unblinded trials in which participants acted as their own controls without randomised crossover (before-and-after) or non-randomised controlled trials. Only studies classified as being of a strong or moderate design were included in the narrative synthesis. The included studies were blinded and unblinded RCTs, and blinded before-and-after studies.
Specific interventions included in the review
Trials of dietary fibre were eligible. Dietary fibre was defined as a product or mixture including at least one food item high in dietary fibre. Non-food sources of fibre were excluded. Trials in which a laxative was given to the treatment group and not to the control group were excluded. The dose of dietary fibre ranged, where stated, from 0.7 to 10.0 g/day. The forms of fibre included finely milled wheat bran, applesauce, prune juice, corn and bran biscuits, yoghurt and guar gum, and high fibre cookies. The treatment duration ranged from 2 to 13 weeks. Cointerventions were laxatives, milk of magnesia, enemas, and suppositories.
Participants included in the review
Older patients (aged 60 years or over) with constipation were eligible. The included participants were either in hospital or long-stay care settings. No definition of constipation was specified as an inclusion criteria for the review. Patients with irritable bowel syndrome or diverticular disease were excluded.
Outcomes assessed in the review
Trials that quantified outcomes for at least one measure of bowel function were eligible. The included studies assessed stool frequency, pharmaceutical agent use, comfort of defecation, faecal incontinence, diarrhoea and amount of stool. The reviewers set a minimum clinically meaningful difference for change in pharmaceutical agent use as 20%. Differences of less than 20% were treated as no difference in the analysis.
How were decisions on the relevance of primary studies made?
The authors do not state how the papers were selected for the review, or how many of the reviewers performed the selection.