Types of studies

Published and unpublished RCTs will be eligible for inclusion. We will exclude non‐randomised studies and quasi‐randomised trials. We will include cross‐over trials if individually randomised women are the unit of analysis; we will only include data from the first phase (pre‐cross‐over data) in the meta‐analyses, as the cross‐over trial is not a valid study design in the context of subfertility.

Types of participants

1. Subfertile women in whom routine imaging did not show uterine cavity abnormalities.

2. Women undergoing screening hysteroscopy prior to IVF.

We will exclude subfertile women suspected of uterine cavity abnormalities (present on any imaging techniques).

Types of interventions

We will include the following randomised comparisons.

1. A routine, screening hysteroscopy, including hysteroscopic treatment of any detected uterine cavity abnormalities versus no hysteroscopy in subfertile women undergoing evaluation for subfertility.

2. A routine, screening hysteroscopy, including hysteroscopic treatment of any detected uterine cavity abnormalities versus no hysteroscopy before IVF.

Types of outcome measures

Electronic searches

We will search the following electronic databases, trial registers, and websites from inception to present:

• Cochrane Gynaecology and Fertility Group (GFG) Specialised Register, ProCite platform (Appendix 1);

• Cochrane Central Register of Controlled Trials (CENTRAL CRSO), web platform (Appendix 2);

• MEDLINE, Ovid platform (Appendix 3);

• Embase, Ovid platform (Appendix 4);

• PsycINFO, Ovid platform (Appendix 5);

• CINAHL (Cumulative Index to Nursing and Allied Health Literature), Ebsco platform (Appendix 6).

We will combine the MEDLINE search with the Cochrane Highly Sensitive Search Strategy for identifying randomised trials, which appears in Section 6.4.11 of the Cochrane Handbook for Systematic Reviews of Interventions (Lefebre 2011).

The Embase, PsycINFO, and CINAHL searches are combined with trial filters developed by the Scottish Intercollegiate Guidelines Network (SIGN) (www.sign.ac.uk/search‐filters.html).

Other electronic sources of trials will include:

• trial registers for ongoing and registered trials:

  • US National Institutes of Health Ongoing Trials Register ClinicalTrials.gov (www.clinicaltrials.gov);

  • World Health Organization International Trials Registry Platform (www.who.int/trialsearch/Default.aspx).

• DARE (Database of Abstracts of Reviews of Effects) on the Cochrane Library (onlinelibrary.wiley.com/o/cochrane/cochrane_cldare_articles_fs.html) for reference lists from relevant non‐Cochrane reviews;

• Web of Knowledge (wokinfo.com/) (another source of trials and conference abstracts);

• OpenGrey (www.opengrey.eu/) for unpublished literature from Europe;

• LILACS (Latin American and Caribbean Health Science Information database) (regional.bvsalud.org/php/index.php?lang=en) for trials from the Portuguese‐ and Spanish‐speaking world;

• PubMed and Google for recent trials not yet indexed in MEDLINE.

Searching other resources

Three review authors (SS, JB, and MSK) will handsearch reference lists of articles retrieved by the search and contact experts in the field to obtain additional data. We will also handsearch relevant journals and conference abstracts that are not covered in the GFG register, in liaison with the Information Specialist.

Selection of studies

One review author (MSK) will conduct an initial screen of titles and abstracts identified by the search, after which we will retrieve the full texts of all potentially eligible studies. Three review authors (SS, JB, and MSK) will independently examine these full‐text articles for compliance with the inclusion criteria and select studies eligible for inclusion in the review. We will correspond with study investigators as required to clarify study eligibility. Disagreements as to study eligibility will be resolved by discussion or by a third review author (SKS). We will document the selection process with a PRISMA flow chart.

Data extraction and management

Two review authors (one a methodologist (MSK) and one a topic area specialist (JB)) will independently extract data from eligible studies using a data extraction form designed and pilot‐tested by the review authors. Any disagreements will be resolved by discussion or by a third review author (SKS). We will extract data including study characteristics and outcome data (Appendix 7). Where studies have multiple publications, the review authors will collate multiple reports of the same study so that each study rather than each report is the unit of interest in the review, and such studies will have a single study ID with multiple references.
We will correspond with study investigators for further data on methods or results, or both, as required. We will include studies irrespective of whether outcomes are reported in a 'usable' way. In multi‐arm studies, data from arms that do not meet the eligibility criteria will be excluded.

Assessment of risk of bias in included studies

Three review authors (MSK, JB, and SS) will independently assess the included studies for risk of bias using the Cochrane 'Risk of bias' assessment tool (Higgins 2011). We will assess the following items: selection (random sequence generation and allocation concealment); performance (blinding of participants and personnel); detection (blinding of outcome assessors); attrition (incomplete outcome data); reporting (selective reporting); and other bias. Disagreements will be resolved by discussion or by a fourth review author. We will describe all judgements fully and present the conclusions in the 'Risk of bias' table, which will be incorporated into the interpretation of review findings by means of sensitivity analyses (see Sensitivity analysis). Selective reporting is a type of reporting bias that affects the internal validity of an individual study. It refers to the selective reporting of some outcomes (e.g. positive outcomes) and the failure to report others (e.g. adverse events). We will take care to search for within‐trial selective reporting, such as trials failing to report obvious outcomes, or reporting them in insufficient detail to allow inclusion. We will seek published protocols and compare the outcomes between the protocol and the final published study. Where identified studies fail to report the primary outcome of live birth, but do report interim outcomes such as pregnancy, we will undertake informal assessment as to whether the interim values (e.g. pregnancy rates) are similar to those reported in studies that also report live birth.

Measures of treatment effect

For dichotomous data (e.g. live‐birth rates), we will use the numbers of events in the control and intervention groups of each study to calculate risk ratios (RR). We will use Peto odds ratio (OR) for outcomes with low event rates. We will reverse the direction of effect of individual studies, if required, to ensure consistency across trials. We will present 95% confidence intervals (CI) for all outcomes. Where data to calculate RRs or ORs are not available, we will utilise the most detailed numerical data available that may facilitate similar analyses of included studies (e.g. test statistics, P values). We will compare the magnitude and direction of effect reported by studies with how they are presented in the review, taking account of legitimate differences.

Unit of analysis issues

The primary analysis will be per woman randomised; we will include per‐pregnancy data for some outcomes (for the outcome miscarriage). If studies report only 'per‐cycle' data, we will contact the study authors to request 'per‐woman' data. Data that do not allow valid analysis (e.g. per‐cycle data) will be briefly summarised in an Additional table and will not be meta‐analysed. We will count multiple live births (e.g. twins or triplets) as one live‐birth event. We will include only first‐phase data from cross‐over trials.

Dealing with missing data

We will analyse the data on an intention‐to‐treat basis to the greatest degree possible and will attempt to obtain missing data from the original authors. Where these data are unobtainable, we will undertake imputation of individual values for live birth only. We will assume live births not to have occurred in women without a reported outcome. For other outcomes, we will analyse only the available data.

Any imputation undertaken will be subjected to sensitivity analysis (see Sensitivity analysis). If studies report sufficient detail to calculate mean differences but no information on associated standard deviation, we will assume the outcome to have standard deviation equal to the highest standard deviation from other studies within the same analysis.

Assessment of heterogeneity

We will consider whether the clinical and methodological characteristics of the included studies are sufficiently similar for meta‐analysis to provide a clinically meaningful summary. We will assess statistical heterogeneity using the I² statistic. An I² measurement greater than 50% will be taken to indicate substantial heterogeneity (Higgins 2011).

Assessment of reporting biases

In view of the difficulty of detecting and correcting for publication bias and other reporting biases, the review authors will aim to minimise the potential impact of these biases by ensuring a comprehensive search for eligible studies and by being alert for duplication of data. If there are 10 or more studies in an analysis, we will use a funnel plot to explore the possibility of small‐study effects (a tendency for estimates of the intervention effect to be more beneficial in smaller studies).

Data synthesis

One review author (SS) will enter the data and perform the statistical analysis using Review Manager 5 (RevMan 2014). If the studies are sufficiently similar, we will combine the data using a fixed‐effect model for the following comparisons.

• Hysteroscopy versus no hysteroscopy for subfertile women.

• Hysteroscopy versus no hysteroscopy for women undergoing IVF.

We will display an increase in the odds of a particular outcome, which may be beneficial (e.g. live birth) or detrimental (e.g. adverse effects of the hysteroscopy) graphically in the meta‐analyses to the right of the centre‐line and a decrease in the odds of an outcome to the left of the centre‐line.

Subgroup analysis and investigation of heterogeneity

Where data are available, we will conduct subgroup analyses to determine the separate evidence within the following subgroups.

• According to the presence or absence of uterine cavity abnormalities at hysteroscopy.

We will conduct the following subgroup analyses within the IVF population.

• Women undergoing first IVF.

• Women with two or more failed IVF cycles.

The interpretation of the statistical analysis for subgroups is difficult. We will mainly use the subgroup analysis to substantiate certain hypotheses concerning the results. In case no RCTs are retrieved for some subgroup analysis, the absence of literature will be reported and identified knowledge gaps will be described.

Sensitivity analysis

We will conduct sensitivity analyses for the primary outcomes to determine whether the conclusions are robust to arbitrary decisions made regarding the eligibility and analysis. These analyses will include consideration of whether the review conclusions would have differed if:

• eligibility had been restricted to studies without high risk of bias (not at high risk of bias in any domain and at low risk for randomisation procedures);

• a random‐effects model had been adopted;

• alternative imputation strategies had been implemented;

• the summary effect measure had been risk ratio rather than odds ratios;

• the primary outcome had been limited to live birth.