Background
Urinary incontinence imposes a considerable burden on individuals and on society. Although a range of treatments is available, alterations in lifestyle are frequently recommended for the treatment of urinary incontinence, as they are relatively low in cost and have few unwanted side-effects. Advice commonly given includes losing weight, changes in diet, adjusting volume of fluid intake, decreasing caffeine or alcohol consumption, avoiding constipation and straining (when passing faeces), stopping smoking, and being more physically active - though restricting excessive heavy activity.
What we wanted to find out
We (a team of Cochrane researchers) wanted to see whether lifestyle interventions have a beneficial effect on any type of urinary incontinence in adults
What we did
We searched the medical literature extensively up to July 2013 for studies that compared the effects of community-based lifestyle alterations with either no treatment, or other non-surgical treatments, or medical (medicine) treatment, on urinary incontinence in adults.
What we found
We identified 11 studies, with 5974 participants (nearly all women, only 20 were men), that investigated the effect of lifestyle alterations on urinary incontinence. Four investigated weight loss; one compared a soy-rich diet with a soy-free diet; three investigated changes in volume of fluid intake; and three investigated the effect of reducing caffeine intake. We identified no trials that investigated reducing alcohol intake, avoiding constipation and straining, stopping smoking or levels of physical activity.
Findings from four studies suggested that weight loss may reduce incontinence among overweight women and this merits further research. However, it should be noted that a large proportion of the participants contributing to this result were part of two diabetes studies that, while they recorded the effect of weight loss on urinary incontinence, did not record how many participants suffered from it at the start of the study. The duration of the weight loss programmes in these studies ranged from three to 12 months.
A small amount of very low quality evidence from the studies that investigated volume of fluid intake suggested that symptoms of urinary incontinence may reduce when fluid intake is reduced, although some participants in the studies reported headaches, constipation or thirst.
We could not combine the findings from other studies that investigated a similar treatment (e.g. caffeine reduction) because they measured their results in different ways, and/or were of poor quality, which means their results may be unreliable. Much more well-designed research is needed, so that lifestyle recommendations for the treatment of incontinence can be based on good evidence. At present there is not enough evidence to establish whether any lifestyle treatments work.
Evidence for the effect of weight loss on urinary incontinence is building and should be a research priority. Generally, there was insufficient evidence to inform practice reliably about whether lifestyle interventions are helpful in the treatment of urinary incontinence.
Low cost, non-invasive alterations in lifestyle are frequently recommended by healthcare professionals or those presenting with incontinence. However, such recommendations are rarely based on good evidence.
The objective of the review was to determine the effectiveness of specific lifestyle interventions (i.e. weight loss; dietary changes; fluid intake; reduction in caffeinated, carbonated and alcoholic drinks; avoidance of constipation; stopping smoking; and physical activity) in the management of adult urinary incontinence.
We searched the Cochrane Incontinence Group Specialised Register, which contains trials identified from the Cochrane Central Register of Controlled Trials (CENTRAL), MEDLINE and MEDLINE in process, and handsearching of journals and conference proceedings (searched 3 July 2013), and the reference lists of relevant articles. We incorporated the results of these searches fully in the review. We undertook an updated search of the Specialised Register, which now includes searches of ClinicalTrials.gov and WHO ICTRP, on 27 October 2014; potentially eligible studies from this search are currently awaiting classification.
Randomised and quasi-randomised studies of community-based lifestyle interventions compared with no treatment, other conservative therapies, or pharmacological interventions for the treatment of urinary incontinence in adults.
Two authors independently assessed study quality and extracted data. We collected information on adverse effects from the trials. Data were combined in a meta-analysis when appropriate. We assessed the quality of the evidence using the GRADE approach.
We included 11 trials in the review, involving a total of 5974 participants.
Four trials involving 4701 women compared weight loss programmes with a control intervention. Low quality evidence from one trial suggested that more women following weight loss programmes reported improvement in symptoms of incontinence at six months (163/214 (76%) versus 49/90 (54%), risk ratio (RR) 1.40, 95% confidence interval (CI) 1.14 to 1.71), and this effect was sustained at 18 months (N = 291, 75% versus 62%, RR not estimable, reported P value 0.02). No data were available for self-reported cure and quality of life. One of the weight loss trials involving 1296 women reported very low quality evidence for a reduction in weekly urinary incontinence a mean of 2.8 years after following a lifestyle weight loss intervention that had been compared with a pharmacological weight loss intervention.
Three trials involving 181 women and 11 men compared change in fluid intake with no change. Limited, very low quality evidence suggested that symptom-specific quality of life scores improved when fluid intake was reduced, although some people reported headaches, constipation or thirst. A further three trials involving 160 women and nine men compared reduction in caffeinated drinks with no change, and one trial involving 42 women compared a soy-rich diet with soy-free diet. However, it was not possible to reach any conclusions about the effects of these changes, due to methodological limitations, that resulted in very low quality evidence.
Adverse effects appeared relatively uncommon for all interventions studied.
All included studies had a high or unclear risk of bias across all bias parameters, but most notably for allocation concealment. The main factors for our downgrading of the evidence were risk of bias, indirect evidence (less than 12 months of follow-up; and not all participants having confirmed urinary incontinence at baseline in some studies), and imprecise results with wide confidence intervals.
Other interventions such as reduction in consumption of sweetened fizzy or diet drinks; reduction in alcohol consumption; avoiding constipation; smoking cessation; restricting strenuous physical forces; or reducing high levels of, or increasing low levels of, physical activity, could not be assessed in this review, as no evidence from randomized controlled trials or quasi-randomised trials was available.