What is the empirical evidence for an association between the consumption of water as a beverage and body weight outcomes in adults?

Four electronic databases are searched for references with no restriction on publication date: MEDLINE, EMBASE, CINAHL and COCHRANE. The search strategy in MEDLINE included the following major text words as well as synonyms and closely related words and MeSH terms: “beverages”, “water consumption”, “drinking behavior” and "body weight and measures”, “overweight”, and “obesity”. The search was limited to humans and adults using a highly specific method, and to publications in English, German, Spanish or French language. No limit on study or publication type was set. The search strategies for each of the other three databases were adapted from the search in MEDLINE using the indexing system for subject headings specific to the databases. Duplications were identified by the reference software Endnote X5 and deleted.

Further studies were identified by hand-searching the reference lists of all included review articles and relevant articles. Experts are contacted via email asking for relevant studies known to them. Selected experts are authors of relevant research and experts in this field known to us. We also consider articles which were related to the included publications or which cited them using the online functions of PubMed “Related citations” and “Cited by PubMed Central articles“. We retrieved the first ten references of the list of related citations, ranked by relevance, and all of the references that cited the included studies.

Inclusion criteria: Interventional studies, observational studies (cross-sectional, cohort, case-control), review, meta-analysis.

Exclusion criteria: case report, case series, non-systematic, narrative reviews, conference abstracts, articles from non-scientific magazines or newspaper.

Any body weight outcome such as body weight, body mass index, classifications of weight status, e.g. overweight and obesity, and measures for body fat or lean mass

Inclusion criteria: adults aged older than 18 years.

Exclusion criteria: exclusively patients with psychotic diseases, eating disorders defined by the international codes of diseases (ICD) code F50, or patients receiving haemodialysis or a fluid replacement therapy, athletes or participants performing extreme exercise.

Increased/higher consumption of water as a beverage and as part of the regular diet with a measured or estimated intake period of at least one day.

Lower consumption of water as a beverage and as part of the regular diet with a measured or estimated intake period of at least one day.

Any setting was included.

Primary outcomes

Any variation in body weight outcome such as body weight, body circumferences, body mass index, classifications of weight status, body fat and lean mass.

Secondary outcomes

None.

All citations retrieved from the electronic search strategies will be reviewed independently by two reviewers, one expert in the research field and one who was not (RM and AG). Based on titles, abstracts and keywords of the citations, the reviewers will determine in an over selective manner if the citation is possibly relevant for further investigation, using defined inclusion criteria. In the next step, the full texts of the citations will be retrieved. Each of the two reviewers will screen the full texts of the references they selected in the first screening by applying all inclusion and exclusion criteria. Disagreements in study inclusion will be arbitrated and final decision will be made by discussion with two other researchers (JMN and GSM). Citations and full texts identified by methods other than the electronic search will be reviewed by one reviewer (RM). The reviewers will not be blinded to the journal and authors name. Articles in Spanish will be reviewed by RM and FB, articles in French by FB and GSM.

The reviewing process will be documented for the evaluation of the level of agreement between the two reviewers. Data from included studies will be extracted by one reviewer (RM) using a standardized form and checked by another (GSM). Disagreements will be resolved by discussion. Articles will be merged if they described the same study. The extraction sheet is based on the recommendations of the Centre for Reviews and Dissemination. Because of the diversity of the designs and reported results of the included studies, the data extraction sheet will be refined during the course of data extraction. Extracted data will include, where possible, description of the participants, study type and setting, primary aim of the study, outcome measures and the assessment tool, the measure of association between water consumption and body weight measure, the applied statistics and the source of funding. Reported outcomes will be taken directly from studies if possible. If information is missing, the authors of the article will be contacted by email. If the authors provide data, we will perform descriptive and statistical analyses ourselves.

Study quality will be assessed using assessment tools specific to each study design. The tools for systematic reviews, intervention and observation studies and were adapted from the corresponding critical appraisal skills program (CASP) checklist for systematic reviews, RCTs and cohort studies. Cross-sectional studies will be assessed with the tool for cohort studies where applicable. According to study type, study quality will be evaluated using four selected criteria of the critical appraisal tools suggested by CASP.

Descriptive synthesis of the data.

Subgroup analaysis by study population defined by body weight status and dieting behaviour.

Publication in an English language scientific journal, press communication

Rebecca Muckelbauer

Berlin School of Public Health

Charite - Universitätsmedizin Berlin (University Medical Center)

Seestr. 73 - Haus 10

D-13347 Berlin

Germany

rebecca.muckelbauer@charite.de

Berlin School of Public Health Charite - Universitätsmedizin Berlin (University Medical Center), Department of Epidemiology / Public Health

http://bsph.charite.de/en/about_us/

Dr Rebecca Muckelbauer, Berlin School of Public Health Charite - Universitätsmedizin Berlin (University Medical Center), Department of Epidemiology / Public Health
Dr Giselle Sarganas-Margolis, Berlin School of Public Health Charite - Universitätsmedizin Berlin (University Medical Center), Department of Epidemiology / Public Health
Professor Jacqueline Müller-Nordhorn, Berlin School of Public Health Charite - Universitätsmedizin Berlin (University Medical Center), Department of Epidemiology / Public Health
Mrs Anke Grüneis, Berlin School of Public Health Charite - Universitätsmedizin Berlin (University Medical Center), Department of Epidemiology / Public Health
Dr Flavia Barbieri, Berlin School of Public Health Charite - Universitätsmedizin Berlin (University Medical Center), Department of Epidemiology / Public Health

01 September 2011

31 August 2012

Berlin School of Public Health Charite - Universitätsmedizin Berlin (University Medical Center), Department of Epidemiology / Public Health

The reviewer (RM) has given talks on water consumption at conferences which were sponsored by Danone. The other authors declare that they have no known conflicts of interest.

English

Germany

Subject indexing assigned by CRD

Body Weight; Drinking Water; Humans

06 July 2012

06 July 2012