Types of studies

We will consider for inclusion all RCTs and quasi‐RCTs (studies in which allocation to treatment was obtained by alternation, use of alternate medical records, date of birth or other predictable methods) of home haemodialysis compared with in‐centre haemodialysis in individuals with ESKD requiring renal replacement therapy (RRT). Studies longer than three months will be considered. Non‐randomised studies will not be eligible. Cross‐over studies will be considered.

Types of participants

• All individuals with ESKD. Participants may already be established on haemodialysis or peritoneal dialysis and are then randomised to haemodialysis (home or in‐centre haemodialysis).

• Participants may also have chronic kidney disease (CKD) and are expected to commence haemodialysis within a time‐frame designated by the study investigators.

Types of interventions

We will include studies in which participants are randomised to home haemodialysis or in‐centre haemodialysis. Haemodialysis can be provided using any dialysis machine, dialysate, blood or dialysate flow rate, membrane type, dialysis dose (urea clearance), or vascular access type (central venous catheter, or arteriovenous fistula or graft).

• Home haemodialysis is defined as any type of haemodialysis, haemodiafiltration, or haemofiltration carried out by the patient, technician, or nurse, at home. We will include any duration of dialysis and any frequency in either treatment arm.

• In‐centre haemodialysis will include dialysis provided in a hospital unit, a private dialysis provider, or a satellite dialysis unit in which nursing or technical staff provide dialysis care.

Studies of haemodialysis performed at self‐care facilities other than home will not be considered. RCTs evaluating peritoneal dialysis as a home dialysis modality will not be included.

Types of outcome measures

• Mortality and cardiovascular events

  • All‐cause mortality

  • Cardiovascular mortality (fatal myocardial infarction, fatal stroke, sudden death, heart failure)

  • Composite endpoints of major adverse cardiovascular events and death

  • Non‐cardiovascular mortality

  • Non‐fatal myocardial infarction

  • Non‐fatal stroke

  • Revascularisation

  • Hospitalisation (all‐cause, cardiovascular cause, unrelated to vascular access)

• Quality of life: we will consider and tabulate where necessary all reports of quality of life outcomes using any instrument. We will conduct meta‐analyses when sufficient studies report quality of life outcomes using a single instrument. This will include measures of depression, household financial stress. We will also assess end‐of‐treatment employment status (employed, unemployed, not eligible for employment)

• Symptoms during dialysis (intradialytic cramping, hypotension, nausea, vomiting, headache)

• Complications secondary to vascular access

  • Hospitalisation due to vascular access complication or procedure (≥1 event)

  • Access‐related bacteraemia (≥1 event)

  • Insertion or replacement of dialysis central venous catheter (≥1 event)

  • Vascular access intervention including surgery or percutaneous intervention (≥1 event)

  • Vascular access angiogram (≥1 event)

  • Time to access failure; access failure (≥1 event)

• Blood pressure (pre‐dialysis systolic and diastolic blood pressure at end‐of‐treatment) (mmHg)

• Change in predialysis blood pressure (mmHg)

• Number and dose of blood pressure medications at end‐of‐treatment

• Change in number of blood pressure medications at end‐of‐treatment

• Left ventricular mass (described using any diagnostic tool including magnetic resonance imaging or echocardiography (g; g/m²), considered separately according to diagnostic tool and summarized using standardized mean differences)

• Haemoglobin at end‐of‐treatment (g/dL; m/m²)

• Serum phosphorus at end‐of‐treatment (mg/dL)

• Serum calcium at end‐of‐treatment (mg/dL)

• Serum beta‐2‐microglobulin at end of study

• Serum calcium by phosphorus product at end‐of‐treatment (mg²/dL²)

• Number and dose of phosphorus binding agents at end‐of‐treatment

• Number and dose of blood pressure lowering agents at end‐of‐treatment

• Number and dose of any other drugs at end‐of‐treatment

• Change in number of blood pressure lowering drugs at end of treatment

• Dose of vitamin D compounds at end‐of‐treatment

• Dose of epoetin at end‐of‐treatment (units or units/kg) calculated as erythropoietin‐equivalent doses

• Parathyroidectomy

• Wait‐listing for kidney transplant

• Recovery time

• Waking hours free for living

• Hospital‐acquired infection rate

We will tabulate other adverse events reported in available studies.

Electronic searches

We will search the Cochrane Renal Group's Specialised Register through contact with the Trials' Search Co‐ordinator using search terms relevant to this review.

The Cochrane Renal Group’s Specialised Register contains studies identified from:

1. Quarterly searches of the Cochrane Central Register of Controlled Trials CENTRAL;

2. Weekly searches of MEDLINE OVID SP;

3. Handsearching of kidney‐related journals and the proceedings of major renal conferences;

4. Searching of the current year of EMBASE OVID SP;

5. Weekly current awareness alerts for selected kidney journals;

6. Searches of the International Clinical Trials Register (ICTRP) Search Portal and ClinicalTrials.gov

Studies contained in the specialised register are identified through search strategies for CENTRAL, MEDLINE and EMBASE based on the scope of the Cochrane Renal Group. Details of these strategies as well as a list of handsearched journals, conference proceedings and current awareness alerts are available in the 'Specialised Register' section of information about the Cochrane Renal Group.

See Appendix 1 for search terms used in strategies for this review.

Searching other resources

1. Reference lists of nephrology textbooks, review articles and relevant studies.

2. Letters seeking information about unpublished or incomplete studies to investigators known to be involved in previous studies.

Selection of studies

The search strategy described will be used to obtain titles and abstracts of studies that may be relevant to the review. The titles and abstracts will be screened independently by two authors, who will discard studies that are not applicable; however studies and reviews that might include relevant data or information on studies will be retained initially. Two authors will independently assess retrieved abstracts and, if necessary the full text, of these studies to determine which studies satisfy the inclusion criteria.

Studies reported in non‐English language journals will be translated before assessment.

Data extraction and management

Data extraction will be carried out independently by two authors using standard data extraction forms. Studies reported in non‐English language journals will be translated before assessment. Where more than one publication of one study exists, reports will be grouped together and only the publication with the most complete data will be included. Where relevant outcomes are only published in earlier versions these data will be used. Any discrepancy between published versions will be highlighted. Any further information required from the original author will be requested by written correspondence and any relevant information obtained in this manner will be included in the review. We will seek unpublished disaggregated data for outcomes among subgroups of patients on home or in‐centre haemodialysis from study investigators of potentially eligible studies and included these data in analyses when received. Disagreements will be resolved by consultation with all authors.

Two independent authors will use standardised data forms to extract data for:

• Study design: parallel or cross‐over, risks of bias, duration, sample size, non‐randomised co interventions, location of study, number of centres

• Participants: source, time on dialysis, previous dialysis modality, age, gender, ethnicity, comorbidities (hypertension, diabetes mellitus, cardiovascular disease), baseline biochemistry and haemoglobin), occupational status, quality of life scores, educational attainment, medications used, distance from dialysis centre, inter‐dialytic weight gain

• Interventions: make and model of dialysis machine, dialysis duration/week, dialysis duration/session, number of dialysis sessions/week, dialysis membrane type (synthetic, cellulose, modified cellulose), dialysis flow rate, dialysis composition (acetate/bicarbonate), dialysis membrane reuse, ultrafiltration (volume/rate)

• Outcomes: as described in Types of outcome measures

Assessment of risk of bias in included studies

The following items will be independently assessed by two authors using the risk of bias assessment tool (Higgins 2011) (seeAppendix 2).

• Was there adequate sequence generation (selection bias)?

• Was allocation adequately concealed (selection bias)?

• Was knowledge of the allocated interventions adequately prevented during the study (detection bias)?

  • Participants and personnel

  • Outcome assessors

• Were incomplete outcome data adequately addressed (attrition bias)?

• Are reports of the study free of suggestion of selective outcome reporting (reporting bias)?

• Was the study apparently free of other problems that could put it at a risk of bias?

We will make explicit judgements regarding whether studies are at high risk of bias according to the criteria given in the Cochrane Handbook for Systematic Reviews of Interventions (Higgins 2011). We will explore the impact of risks of bias by undertaking sensitivity analyses where possible.

Measures of treatment effect

For dichotomous outcomes (mortality, cardiovascular events, hospitalisation, vascular access adverse events) results will be expressed as risk ratio (RR) with 95% confidence intervals (CI). Where continuous scales of measurement are used to assess the effects of treatment (quality of life scale, blood pressure, doses of medication, haemoglobin, biochemical variables), the mean difference (MD) will be used, or the standardised mean difference (SMD) if different scales have been used. We will be cautious when providing summary estimates of treatment when high‐level heterogeneity between studies cannot be explained.

Unit of analysis issues

We will include only data from the first period of treatment in cross‐over studies (Higgins 2011). Data in different metrics will be analysed by converting reported values to SI units. The final results will be presented in International System (SI) units with conventional units in parentheses.

Dealing with missing data

If possible, data for each prespecified outcome will be evaluated regardless of whether the analysis is based on intention‐to‐treat or completeness to follow‐up. In particular, dropout rates will be investigated and reported in detail (e.g. drop‐out due to discontinuation of dialysis modality, treatment failure, death, transplantation, withdrawal of consent or loss to follow‐up). When data are unavailable or not reported in an extractable format, we will contact the original investigators to request the missing data. We will assess all studies for risks of bias due to incomplete reporting of results.

Assessment of heterogeneity

We will test for heterogeneity with the Cochran Q test which follows a Chi² distribution with N‐1 degrees of freedom, with an alpha of 0.05 used for statistical significance. The extent of heterogeneity will be assessed with I², which ranges between 0% and 100% and expresses the proportion of between group variability that is attributable to heterogeneity rather than chance (Higgins 2003). I² values of above 75% are typically held to signify extreme heterogeneity, whereas 25% and 50% correspond to low and medium levels of heterogeneity, respectively.

Assessment of reporting biases

We will test for asymmetries in the inverted funnel plots (i.e. for systematic differences in the effect sizes between more precise and less precise studies) using the original and modified Egger tests (Egger 1997) and the Begg and Mazumdar correlation test (Begg 1994). There are many potential explanations for why an inverted funnel plot may be asymmetric, including chance, heterogeneity, publication and reporting bias (Terrin 2005). We will refrain from judging funnel plot asymmetries based on visual inspection as this has been shown to be misleading in empirical research (Lau 2006). Publication bias will also be evaluated by testing the robustness of the results according to publications, namely publication as a full manuscript in a peer reviewed journal versus studies published as abstracts/text/letters/editorials.

Data synthesis

Data will be pooled using the random‐effects model but the fixed‐effect model will also be used to ensure robustness of the model chosen and susceptibility to outliers. We will qualitatively summarise data where insufficient data are available for meta‐analysis. Qualitative review will be conducted for adverse events and quality of life outcomes.

Subgroup analysis and investigation of heterogeneity

Subgroup analysis will be used to explore possible sources of heterogeneity (e.g. participants, interventions and study quality). Heterogeneity among participants could be related to age and haemodialysis methods. Heterogeneity in treatments could be related to prior agent(s) used and the agent, dose and duration of therapy. Adverse effects will be tabulated and assessed with descriptive techniques, as they are likely to be different for the various agents used. Where possible, the risk difference (RD) with 95% CI will be calculated for each adverse effect, either compared to no treatment or to another agent.

Heterogeneity will be investigated by analysing the data using subgroups according to the following parameters:

• Population characteristics

  • Presence or absence of co‐morbidities (diabetes, hypertension, dyslipidaemia, smoking, obesity, family history of cardiovascular disease, baseline cardiovascular disease); percentage of patients with these co‐morbidities in each study

  • Age (adult, paediatric)

  • Gender

  • Mean systolic blood pressure

  • Ethnicity (White, Afro‐American, Asian, other)

  • Time on dialysis (less than three years versus three years or more)

• Intervention characteristics

  • Duration of home haemodialysis/session

  • Duration of home haemodialysis/week

  • Prescribed blood flow rate

  • Prescribed dialysis dose (Kt/V)

  • Hospital versus satellite clinic comparator

  • Acetate or bicarbonate dialysis

  • Number of home haemodialysis sessions/week

  • Duration of intervention (less than 6 months, 6 to 12 months, more than 12 months)

  • Treatment dropout rate

If sufficient studies are available, we will conduct sensitivity analysis based on allocation concealment, blinding of participants, investigators and outcome assessors, attrition (above or below 10%), ITT analysis, and premature discontinuation of the study.

We will perform univariate meta‐regression according to previously described methods if sufficient studies are identified (Palmer 2007).

Sensitivity analysis

Sensitivity analyses will be undertaken to explore the robustness of findings to key decisions in the review process. These will be determined as the review process takes place (Higgins 2011). Sensitivity analyses will be undertaken to explore the influence of a study's risk of bias on the results including:

• repeating the analysis excluding unpublished studies;

• repeating the analysis taking account of risk of bias, as specified above;

• repeating the analysis excluding any very long or large studies to establish how much they dominate the results;

• repeating the analysis excluding studies using the following filters: diagnostic criteria, language of publication, source of funding (industry versus other), country.