Types of studies

We included all randomized and quasi‐randomised controlled trials. We excluded cross‐over trials due to the uncertainty regarding carry‐over effects.

Types of participants

We included trials where participants are over 50 years of age with measurable cataract, specifying which eye is affected (or both eyes), as defined by a suitably qualified medical practitioner or researcher, which is shown to be affecting their quality of life through reduced or abnormal vision. We did not make any demographic differentiations. We excluded participants with any ocular comorbidity, as this condition may make it difficult to diagnose the cause of reduced vision.

Types of interventions

We included all trials where topical NAC is compared to a control (such as saline or artificial tear) or to cataract surgery. Included trials had a minimum treatment period of two months at any dose and frequency.

Types of outcome measures

Electronic searches

We searched CENTRAL (which contains the Cochrane Eyes and Vision Trials Register) (2016, Issue 6), Ovid MEDLINE, Ovid MEDLINE In‐Process and Other Non‐Indexed Citations, Ovid MEDLINE Daily, Ovid OLDMEDLINE (January 1946 to June 2016), Embase (January 1980 to June 2016), Allied and Complementary Medicine Database (AMED) (January 1985 to June 2016), Cumulative Index to Nursing and Allied Health Literature (CINAHL) (1982 to June 2016), the ISRCTN registry (www.isrctn.com/editAdvancedSearch), ClinicalTrials.gov (www.clinicaltrials.gov) and the World Health Organization (WHO) International Clinical Trials Registry Platform (ICTRP) (www.who.int/ictrp/search/en). We did not use any date or language restrictions in the electronic searches for trials. We last searched the electronic databases on 28 June 2016.

See Appendices for details of search strategies for CENTRAL (Appendix 1), MEDLINE (Appendix 2), Embase (Appendix 3), CINAHL (Appendix 4), AMED (Appendix 5 ), ISRCTN (Appendix 6), ClinicalTrials.gov (Appendix 7) and the WHO ICTRP (Appendix 8) .

Searching other resources

We searched the reference lists of included studies and searched the Science Citation Index database to identify any additional trials. We handsearched the society meetings of the American Society of Cataract and Refractive Surgery (ASCRS) and the European Society of Cataract and Refractive Surgeons (ESCRS) from 2005 to September 2015.

Selection of studies

Both review authors (VD, AB) searched through the results and independently extracted data in duplicate. We included studies on the basis of the criteria specified above. For studies that we labelled as 'include' or 'unsure', we obtained a full‐text report of the study. We recorded the selection process in sufficient detail to complete a PRISMA flow diagram (Moher 2009), and Characteristics of excluded studies table. The two review authors resolved any disagreement by discussion.

Data extraction and management

We designed a paper data collection form. The form consisted of study and version ID, review author ID, citation and contact details, confirmation of eligibility for review, any reasons for exclusion of participants, study design, study duration, evidence of sequence generation and allocation sequence concealment, masking and documentation of any concerns over bias.

• Participant factors: number, setting, diagnostic criteria, age, sex, country and study dates.

• Intervention data: number of groups, number of participants per group and details of intervention for all groups.

• Outcomes: quality of life (differences in scores between baseline and study endpoint), cataract severity (differences in LOCS III scores and density (Scheimpflug) between baseline and study endpoint) and visual function (differences in measured visual acuity and contrast sensitivity between baseline and study endpoint).

We planned to collect these data at short‐term (less than one year) and at long‐term (more than one year) intervals. We based the outcome criteria on change from the baseline; we recorded scales, including limits and units of measurement.

• Results: any missing participants, summary data as detailed above including means, confidence intervals and P values.

• Adverse event data: (worsening in quality of life scores, visual function, or cataract within the study period, or adverse effects from topical application of NAC drops, i.e. pain, eye infection, allergy and scarring to ocular surface) and relevant notes.

We documented sources of funding, along with key conclusions from study authors, any references to related studies and any comments from the review authors.

Both review authors independently collected all the data in duplicate. One review author (VD) entered the data from both review authors into Review Manager 5 (RevMan 2014), and the second review author (AB) then checked that the data had been correctly entered.

Assessment of risk of bias in included studies

We used Cochrane's tool for assessing risk of bias as per Chapter 8 of the Cochrane Handbook for Systematic Reviews of Interventions assessing the following (Higgins 2011): sequence generation, allocation concealment, masking (blinding) of participants, personnel and outcome assessors, incomplete outcome data, selective outcome reporting and other sources of bias. We assessed all parameters as 'low', 'high' or 'unclear' risk of bias.

Measures of treatment effect

We planned to analyse ordinal data such as visual acuity and LOCS III grading as continuous data, using their respective standardised mean differences (SMDs).

We planned to analyse continuous data such as cataract density (measured by pixel counting on a Scheimpflug camera system) and quality of life data, with respective SMDs, as the measurement scales may vary depending upon which equipment or method(s) are used in each study.

We planned to consider data on glare sensitivity and contrast sensitivity as ordinal or continuous, depending upon the measurement method used.

Unit of analysis issues

We included only studies where people are randomized to treatment. We excluded trials where one eye receives treatment and the other placebo.

Trialists may use two main methods to deal with the problem of non‐independence of data from eyes: (i) they may enrol one eye per person into the trial, or (ii) they may analyse data from both eyes making corrections for within‐person correlation.

For (i), we documented the criteria used to define which eye is to be included in the trial.

For (ii), if the trialist has not adjusted for within‐person correlation, we planned to take statistical advice, but as no trials were included this is not relevant for this version of the review.

Dealing with missing data

We planned to document the reasons why participants were not followed up, if reported. We planned to contact study trialists for further clarification as needed. At that stage we planned to be able to make judgements as to whether the data are missing at random or not. We planned to address the potential impact of missing data in the Discussion section.

Assessment of heterogeneity

We planned to investigate clinical heterogeneity by examining differences in population, intervention and outcomes. We planned to investigate statistical heterogeneity by looking at the forest plots and statistical tests (Chi² and the I² statistic) since tests for statistical heterogeneity are underpowered when the number of studies is low.

Assessment of reporting biases

We planned to use funnel plots to help in our assessment of bias as per Chapter 10 of the Cochrane Handbook for Systematic Reviews of Interventions (Sterne 2011), but as there was not enough data, we did not create funnel plots.

Data synthesis

As we cannot assume that the intervention will have the same effect in different groups of people, that is, that each included study will be measuring the same effect, we planned to pool the data using a random‐effects model (providing there are three or more included trials providing relevant data). A random‐effects model will allow for differences in the effect of the intervention between the included trials. If the heterogeneity is too great, then meta‐analysis will not be possible.

Subgroup analysis and investigation of heterogeneity

We planned to consider the following subgroups but in the event there were not enough data to do these analyses.

• Smokers versus non‐smokers.

• Diabetics versus non‐diabetics.

Sensitivity analysis

We did not plan to conduct any sensitivity analysis for this review.