Criteria for considering studies for this review
Types of studies
We will include randomized controlled trials (RCTs) and the first period of cross-over RCTs. We will exclude quasi-randomized controlled trials (qRCTs) (i.e. trials in which assignment to treatment was made by a non-random method such as alternation), because the method of allocation may lead to biased estimates of effect. We will not restrict study eligibility by language or publication status. We will only include health economics studies conducted alongside effectiveness studies included in the effectiveness component of the review (Shemilt 2011).
Types of participants
We will include studies with women aged 18 years or more who have been diagnosed as having urinary incontinence on the basis of symptoms, signs, or urodynamic evaluation (as defined by the trial authors). We will only include trials that have recruited both men and women if at least 90% of participants are women or if the trial reports demographic and outcome data separately for women. We will include trials with women diagnosed as having stress, urgency, or mixed urinary incontinence, as well as trials with participants who have mixed or undefined diagnoses of urinary incontinence.
We will include trials carried out in women who were experiencing urinary incontinence during the antenatal or postnatal period. However, because this group has physiological differences from women who are not pregnant or postpartum, and the natural history of incontinence in this population differs from that in non-pregnant women, the observed effects of the intervention may differ in this group. Therefore, we will not group this population with that of women who are not pregnant or postpartum, and we will analyse the studies separately throughout the review.
We will exclude studies of women who have urinary incontinence that is thought to be due to factors beyond the urinary tract, such as neurological or psychiatric disorders, cognitive impairment, or mobility problems. We will also exclude studies that focused on women who were experiencing nocturnal enuresis.
Types of interventions
We will include studies of yoga as a treatment for urinary incontinence. We will require that the study report specify that the intervention is 'yoga' and we will exclude studies of interventions based on yoga (e.g. exercises based on yoga postures) but not characterized as yoga. We will exclude studies of multimodal interventions in which yoga is only one component amongst others, such as mindfulness-based stress reduction (MBSR).
We will include interventions that belong to any yoga tradition, but we will exclude studies in which yoga does not include a physical practice component. We will include studies in which the yoga intervention is provided at any dose, with any frequency, and for any duration.
We will include studies that compare yoga to no treatment or to another active treatment. We will also include studies that compare yoga as an adjunct to other treatments versus those same treatments without yoga. The types of comparisons will be as follows.
Yoga versus no specific active intervention (e.g. usual care, a waiting list).
Yoga versus an active intervention (e.g. a lifestyle advice intervention or pelvic floor training), for which we will consider different active comparators separately (e.g. yoga versus lifestyle advice, yoga versus pelvic floor training).
Yoga plus an intervention versus the same intervention without yoga (e.g. yoga as an add-on intervention to pelvic floor training, versus pelvic floor training alone), for which we will consider different interventions separately.
Studies that compare yoga to a sham yoga intervention do not appear to be common (Park 2014), and we do not expect to find such studies. However, if we encounter such studies, we will consider them to represent an additional type of comparison (i.e. yoga versus sham yoga) and will analyse them separately.
We will include studies in which there are co-interventions provided the co-interventions are similar between intervention groups (e.g. both the yoga and the comparison groups receive lifestyle advice).
Types of outcome measures
We will use outcomes as suggested by the Standardisation Committee of the International Continence Society. The Committee recommends the following five categories of outcomes for research investigating the effect of therapeutic interventions for women with urinary incontinence (Lose 2001).
The woman’s observations (e.g. symptoms).
Quantification of the woman's symptoms (e.g. urine loss).
The clinician’s observations (anatomical and functional).
The woman's quality of life (urinary incontinence-specific and general).
In this review we will consider at least one outcome from each of the first four categories. We will include specific outcomes that are commonly found in other Cochrane Incontinence Group reviews of interventions for urinary incontinence in women, so that this review may produce results that are easily compared to or combined with those of other reviews of treatment for the same condition.
We will group all outcomes into three time points: short-term (closest to three months after randomisation), intermediate-term (closest to six months after randomisation), and long-term (closest to one year after randomisation). When an included trial presents multiple time points, we will consider the short-term time point as primary.
Number of women who report continence.
Number of women who report continence or improvement of urinary incontinence.
Urinary incontinence condition- or symptom-specific quality of life, as measured by any relevant scale (e.g. the Urinary Incontinence Quality of Life scale (Patrick 1999), or the Incontinence Impact Questionnaire (Uebersax 1995)).
Search methods for identification of studies
We will not impose any restrictions, for example language or publication status, on the literature searches described below.
Cochrane Incontinence Group Specialised Register
This review will draw on the search strategy developed for the Cochrane Incontinence Group. We will identify relevant trials from the Cochrane Incontinence Group Specialised Register. For more details of the search methods used to build the Specialised Register please see the Group's module in the Cochrane Library. The register contains trials identified from the Cochrane Central Register of Controlled Trials (CENTRAL), MEDLINE, MEDLINE In-Process, MEDLINE Epub Ahead of Print, CINAHL, ClinicalTrials.gov, the World Health Organization International Clinical Trials Registry Platform (WHO ICTRP) portal (http://apps.who.int/trialsearch/) and handsearches of journals and conference proceedings. Most trials in the Cochrane Incontinence Group Specialised Register are also contained in CENTRAL.
The terms that we will use to search the Incontinence Group Specialised Register are in Appendix 1.
We will search for all eligible published and unpublished trials in all languages. We will translate the non-English language abstract for potential inclusion. We will base our search strategy on concepts of types of study population, types of study design, and symptoms of urinary incontinence such as dribbling of urine.
Other electronic bibliographic databases
We will also use the terms in Appendix 1 to search the Cochrane Complementary Medicine Field Specialised Register for relevant trials of yoga. The Complementary Medicine Field Specialised Register contains trials identified from CENTRAL, Chinese databases, Korean databases, and multiple small databases that have been identified as good sources of complementary medicine trials (Wieland 2013).
We will perform searches of the NHS Economic Evaluation Database (NHS EED) in order to identify potential economic studies. As the last database searches for NHS EED were carried out in December 2014, we will use the NHS EED search filters to search MEDLINE and Embase for economics studies from January 2015 onwards. See Appendix 2 for the NHS EED search filters.
We had planned to search the following additional databases, detailed below. However, after preliminary test searches and deduplication against the Cochrane Incontinence Group Specialised Register, they did not appear to yield any additional records of relevance to the review and we will not search them for the first version of this review.
AMED (Allied and Complementary Medicine): we conducted a preliminary test search (Appendix 3) on AMED on OvidSP (1985 to November 2015) on 30 November 2015 to gauge how useful this database would be as a source of records for this review. A broad search designed to identify RCTs or qRCTs related to urinary incontinence, overactive bladder, or enuresis did not find any additional randomised (or qRCTs) in this area after deduplication against the Cochrane Incontinence Group Speciaised Register.
CINAHL (Cumulative Index to Nursing & Allied HealthLiterature): we tested a preliminary search (Appendix 4) in CINAHL (on EBSCOhost) on 2 December 2015 (December 1981 to most recent entry date 20151201) and after deduplication against the Cochrane Incontinence Group Specialized Trial Register we did not retrieve any extra records.
IndMED: we performed a preliminary search on 2 December 2015 using the strategy in Appendix 5. We only identified one irrelevant record.
Searching other resources
We will check reference lists of all included studies and systematic reviews for additional references. We will contact experts in the field and authors of included studies to identify additional unpublished studies. We will also check the proceedings of the following conferences for relevant research.
Data collection and analysis
We will conduct data collection and analysis in accordance with the Cochrane Handbook for Systematic Reviews of Interventions (Higgins 2011).
Selection of studies
Two review authors will independently screen titles and abstracts of all retrieved references. We will retrieve the full-text study reports for all citations that at least one review author considers potentially relevant. Two review authors will independently screen the full-text articles and identify studies for inclusion, and identify and record reasons for exclusion of ineligible studies in a 'Characteristics of excluded studies' table. We will resolve any disagreement through discussion or, if required, we will consult a third review author. We will identify and exclude duplicates and collate multiple reports of the same study so that each study, rather than each report, is the unit of interest in the review. We will record the selection process in sufficient detail to complete a Preferred Reporting Items for Systematic Reviews and Meta-Analyses (PRISMA) flow diagram (Moher 2009).
Data extraction and management
We will use a piloted data collection form for study characteristics and outcome data. We will extract the following study characteristics.
Methods: study design.
Participants: number randomized, number with each type(s) of incontinence and how incontinence was diagnosed, if participants were pregnant or postpartum how many were pregnant and how many were postpartum, severity and duration of urinary incontinence if reported, study participant mean age or age range, study location and setting, recruitment methods, inclusion and exclusion criteria.
Interventions: description of yoga intervention characteristics, dose and duration of yoga intervention, description of comparison intervention characteristics, adherence to the yoga and comparison interventions, description of any co-interventions, length of follow-up, number of withdrawals, and reasons for withdrawal.
Outcomes: description of primary and secondary outcomes for the review that were reported in the trial, listing of other outcomes collected in the trial.
Notes: funding for trial and notable conflicts of interest of trial authors.
'Risk of bias' assessment (see below).
Two review authors will independently extract outcome data from included studies into Covidence (Covidence 2017), and use the Covidence software to compare the identify any discrepancies in data entry. We will resolve disagreements by consensus or by involving a third review author. We will note in the 'Characteristics of included studies’ table if a trial did not report outcome data in a usable way. We will transfer data from Covidence into Cochrane’s statistical software, Review Manager (RevMan) (Review Manager 2014).
When both endpoint and change data are available, we will use endpoint data for our primary analysis, and will carry out a sensitivity analysis to check whether the results vary according to endpoint versus change data. In cases where study participants are lost to follow-up and intention-to-treat analyses are conducted using imputation alongside available-case analyses, we will use the observed data for our primary analysis, and perform a sensitivity analysis to check whether the results vary according to imputed versus available case data. In cases where both unadjusted and adjusted data are available, we will use the unadjusted data for our primary analysis, and undertake a sensitivity analysis to check whether the results vary according to adjusted versus unadjusted data. See the 'Sensitivity analysis' section.
In addition to extracting data from included studies as described above, we will include an appendix of bibliographic detail of potential economic studies. We will not conduct a further review of potential economic studies.
Assessment of risk of bias in included studies
Two review authors will independently assess the risk of bias for each included trial using the criteria outlined in the Cochrane Handbook for Systematic Reviews of Interventions (Higgins 2011). We will resolve any disagreements by discussion or by involving a third review author. We will assess the risk of bias according to the following domains.
Random sequence generation.
Blinding of participants and personnel.
Blinding of outcome assessment for each outcome.
Incomplete outcome data.
Selective outcome reporting.
Other bias such as validity of outcome measure and baseline comparability.
We will assess each potential source of bias as either high, low, or unclear and we will provide a quote from the study report together with a justification for our judgment in the 'Risk of bias’ tables. We will summarize the 'Risk of bias' judgements across different studies for each of the domains listed.
The participants and personnel in most yoga trials are not blinded. As the primary outcomes of this review depend upon the perception and reporting of the study participant, we will not include blinding of personnel, participants, or outcome assessors in our dichotomizing of trials into high versus low risk of bias. Instead, for sensitivity analyses of the primary outcomes by high versus. low risk of bias, we will classify studies as having a low risk of bias if they have a low risk of bias for random allocation, allocation concealment, and incomplete outcome assessment, and do not have a high risk of bias for selective outcome reporting or other bias (Sensitivity analysis).
Measures of treatment effect
We will upload the outcome data for each study into the data tables in RevMan to calculate the treatment effects (Review Manager 2014). We will use the risk ratio (RR) for dichotomous outcomes related to benefit of treatment (e.g. cure). We will use the absolute risk difference (RD) for adverse events. We will use the mean difference (MD) for continuous outcomes reported on the same scale, and the standardized mean difference (SMD) for continuous outcomes reporting the same outcome but measured on different scales in different trials included in the same meta-analysis. We will express the uncertainty with 95% confidence intervals (CIs) for all effect estimates. If the included studies only report effect estimates and their CIs, standard errors, or exact P values, we will upload these data into RevMan so as to use the generic inverse variance method (Review Manager 2014). We will ensure that higher scores for continuous outcomes have the same meaning for the particular outcome and explain the direction to the reader. When we are unable to enter the results in either way, we will describe them in the 'Characteristics of included studies’ tables or enter the data into the ’Additional tables’ section.
Unit of analysis issues
We will include cross-over studies, but we will include only the first period of the trials in order to avoid carry-over effects. We do not expect to find cluster-randomized studies on this topic. However, if we find such studies, we will follow the guidance in Chapter 16.3 of the Cochrane Handbook for Systematic Reviews of Interventions to assess their suitability and will include these studies in the analysis if appropriate (Higgins 2011a).
Dealing with missing data
We will contact investigators or study sponsors in order to verify key study characteristics and obtain missing numerical outcome data where possible (e.g. when we identify a study as an abstract only). If we do not find a full report even after we contact the study authors, we will list such an abstract as a 'study awaiting classification’. If numerical outcome data are missing, such as standard deviations (SDs) or correlation coefficients, and we cannot obtain these from the study authors, we will calculate them from other available statistics, such as P values, according to the methods described in the Cochrane Handbook for Systematic Reviews of Interventions (Higgins 2011).
Assessment of heterogeneity
We will assess clinical heterogeneity (i.e. differences in study populations, interventions, and outcomes) between studies qualitatively. For studies that we judge to have sufficient clinical homogeneity to combine in a meta-analysis, we will assess heterogeneity in three ways: visual examination of the forest plots; Chi² test (P ≤ 0.10) for heterogeneity; and the I² statistic, which describes the percentage of the variability in the estimate that is due to clinical or methodological heterogeneity rather than to chance. We will consider the implications of the observed value of the I² statistic as follows.
0% to 40%: as perhaps not important.
30% to 60%: as perhaps representing moderate heterogeneity.
50% to 90%: as perhaps representing substantial heterogeneity.
75% to 100%: as perhaps representing considerable heterogeneity (Deeks 2011).
We will assess the importance of the value of the I² statistic in light of the range and direction of effects as observed from the forest plots and the strength of the evidence for heterogeneity based upon the Chi² test (Deeks 2011). We will seek and discuss plausible explanations for observed statistical heterogeneity. We will investigate causes of heterogeneity between trials, using qualitative assessment of any differences in the populations and interventions between individual trials. We will also use subgroup analyses (Subgroup analysis and investigation of heterogeneity) and sensitivity analyses (Sensitivity analysis) to explore heterogeneity.
Assessment of reporting biases
We will search study registries of prospectively registered trials to identify completed, but not published, trials. If we are able to pool more than 10 trials in any meta-analysis, we will use a funnel plot and the Egger’s test to assess the potential for publication bias (Egger 1997; Sterne 2011).
If trials are clinically comparable (i.e. clinical comparability of population, intervention, comparison, outcome, and timing of measurement), we will combine the outcome data from the individual trials in a meta-analysis using RevMan (Review Manager 2014). We will analyse pregnant and postpartum women separately at all times. However for non-pregnant, non-postpartum populations, we will group together women with stress urinary incontinence (SUI), urgency urinary incontinence (UUI), mixed urinary incontinence (MUI), and mixed or unclear urinary incontinence diagnoses in the primary analysis, and examine different types of urinary incontinence diagnoses using subgroup analyses (see the 'Subgroup analysis and investigation of heterogeneity' section).
As yoga is a complex intervention with variations in practice components and implementation, and we expect some between-study variation due to these factors, we plan to use the random-effects model for meta-analysis. We will include 95% CIs for all estimates.
Should we consider that the outcome data from individual trials is not sufficiently similar to be combined quantitatively, we will narratively describe the results from clinically comparable trials.
GRADE and 'Summary of findings' tables
We will create a ’Summary of findings’ table using the Grading of Recommendations Assessment, Development and Evaluation (GRADE) criteria (Guyatt 2011a; Guyatt 2011b). Two review authors (LSW and ZSL) will undertake GRADE assessments independently and will compare results. If there is disagreement on GRADE assessments, we will reach consensus through discussion. We will justify all decisions to downgrade the certainty of the evidence using footnotes and we will make comments to aid readers’ understanding of the review where necessary.
We will consider the following factors when we assess the certainty of evidence.
Limitations in the study design and conduct (i.e. risk of bias).
Inconsistency of results.
Indirectness of evidence.
We will downgrade the certainty of the evidence for a specific outcome by one level according to the performance of the included studies against each of the five factors.
The GRADE working group recommends including up to seven critical outcomes in a systematic review (Guyatt 2011a; Guyatt 2011b). In this systematic review, we will consider six critical outcomes for assessing the quality of evidence.
Number of women who report urinary continence.
Number of women who report urinary continence or improvement of urinary incontinence symptoms.
Urinary incontinence condition-specific or symptom-specific quality of life, as measured on any relevant scale.
Urine loss as measured with a pad or paper towel weight test.
Quality of life as measured on a scale that is not condition- or symptom-specific.
We will prepare separate Summary of findings tables for trials carried out with women who were pregnant or postpartum.
Subgroup analysis and investigation of heterogeneity
As the mechanisms of SUI and UUI are different, the effect of yoga may vary across types of urinary incontinence. Therefore, if there is sufficient data we will carry out subgroup analyses by type of urinary incontinence in the following groups.
If trials contain women with different types of urinary incontinence, and we are unable to extract outcome data separately for women diagnosed with the different types of incontinence, but most women included in the trial are diagnosed with one of SUI, UUI, or MUI, we will consult with the Cochrane Incontinence Group editors and statisticians on how to include the study data appropriately in the subgroup analysis.
We will also conduct subgroup analyses of trials conducted in older (mean age 65 years or greater) versus younger populations as populations from different age groups may vary in their ability to perform yoga. Among women who are pregnant or postpartum, we will perform subgroup analyses of trials conducted in women who are pregnant versus postpartum, as the natural history of incontinence is different in the two states. If the data permit, we will carry out subgroup analyses by style (e.g. Iyengar yoga, Viniyoga, Yin yoga), dose, and duration of yoga intervention, to see whether type of yoga, amount of yoga, or duration of yoga modifies the effect of the yoga intervention.
We will assess the robustness of our conclusions by excluding studies that we judge to have a high risk of bias from our meta-analyses for the primary outcomes. We will also carry out sensitivity analyses to compare analyses using imputed and available case data, analyses using change and endpoint data, and analyses using adjusted and unadjusted data.