Protocol for a systematic review and individual patient data meta-analysis of prognostic factors of foot ulceration in people with diabetes: the international research collaboration for the prediction of diabetic foot ulcerations (PODUS)
1 Department of vascular Surgery, Freeman Hospital, Newcastle upon Tyne Hospitals NHS Foundation Trust, High Heaton, NE7 7DN, Newcastle, UK
2 Centre for Population Health Sciences, Medical School, The University of Edinburgh, Teviot Place, EH8 9AG, Edinburgh, UK
3 School of Surgical & Reproductive Sciences, 3rd Floor, William Leech Building, Faculty of Medical Sciences, Framlington Place, NE2 4HH, Newcastle upon Tyne, UK
4 Department of Diabetes, Royal Infirmary of Edinburgh, 51 Little France Crescent, Old Dalkeith Road, EH16 4SA, Edinburgh, UK
5 Centre for Endocrinology & Diabetes, Institute of Human Development, Core Technology Facility, room 3.30, University of Manchester, 46 Grafton Street, M13 9NT, Manchester, UK
6 Manchester Royal Infirmary, Division of Medicine, Oxford Road, M13 9WL, Manchester, UK
7 Epidemiologic Research and Information Center, VA Puget Sound Health Care System, University of Washington, 1100 Olive Way, Suite 1400, 98101, Seattle, WA, USA
8 Science Consulting & Clinical Monitoring SCCM, Hessegasse 30/15, A-1220, Wien, Australia
9 Department of Diabetes and Endocrinology, Ninewells Hospital and Medical School, DD1 9SY, Dundee, UK
10 Medico dirigente I livello, Sezione diabetologia, Cardiologia Geriatrica, AOU Careggi, Florence, Italy
11 Serviço de Endocrinologia–Pé Diabético, Centro Hospitalar de Vila Nova de Gaia/Espinho EPE, Unidade 1, Rua Conceição Fernandes, 4434-502, Vila Nova de Gaia, Portugal
12 Cass Lake Indian Health Services Hospital, 425 7th St NW, MN 56633, Cass Lake, USA
13 Harvard Medical School, 1563 Mass. Ave # 324, 02138, Cambridge, MA, USA
14 Newcastle hospitals community health, Podiatry Department, Geoffrey Rhodes clinic, Algernon Road, Byker, NE6 2UZ, Newcastle upon Tyne, UK
15 Department of Diabetes and Endocrinology, Nottingham City Hospital, Hucknall Road, NG5 1PB, Nottingham, UK
16 Ward 31, Royal Victoria Hospital, NE1 4LP, Newcastle, UK
17 Royal College of Surgeons in Ireland, 123 St Stephens Green, Dublin 2, Ireland
18 MRC Clinical Trials Unit, Aviation House, 125 Kingsway, WC2B 6NH, London, UK
BMC Medical Research Methodology 2013, 13:22 doi:10.1186/1471-2288-13-22Published: 15 February 2013
Diabetes–related lower limb amputations are associated with considerable morbidity and mortality and are usually preceded by foot ulceration. The available systematic reviews of aggregate data are compromised because the primary studies report both adjusted and unadjusted estimates. As adjusted meta-analyses of aggregate data can be challenging, the best way to standardise the analytical approach is to conduct a meta-analysis based on individual patient data (IPD).
There are however many challenges and fundamental methodological omissions are common; protocols are rare and the assessment of the risk of bias arising from the conduct of individual studies is frequently not performed, largely because of the absence of widely agreed criteria for assessing the risk of bias in this type of review. In this protocol we propose key methodological approaches to underpin our IPD systematic review of prognostic factors of foot ulceration in diabetes.
1. What are the most highly prognostic factors for foot ulceration (i.e. symptoms, signs, diagnostic tests) in people with diabetes?
2. Can the data from each study be adjusted for a consistent set of adjustment factors?
3. Does the model accuracy change when patient populations are stratified according to demographic and/or clinical characteristics?
MEDLINE and EMBASE databases from their inception until early 2012 were searched and the corresponding authors of all eligible primary studies invited to contribute their raw data. We developed relevant quality assurance items likely to identify occasions when study validity may have been compromised from several sources. A confidentiality agreement, arrangements for communication and reporting as well as ethical and governance considerations are explained.
We have agreement from the corresponding authors of all studies which meet the eligibility criteria and they collectively possess data from more than 17000 patients. We propose, as a provisional analysis plan, to use a multi-level mixed model, using “study” as one of the levels. Such a model can also allow for the within-patient clustering that occurs if a patient contributes data from both feet, although to aid interpretation, we prefer to use patients rather than feet as the unit of analysis. We intend to only attempt this analysis if the results of the investigation of heterogeneity do not rule it out and the model diagnostics are acceptable.
This review is central to the development of a global evidence-based strategy for the risk assessment of the foot in patients with diabetes, ensuring future recommendations are valid and can reliably inform international clinical guidelines.