Multiple imputation to minimise bias from missing stage information in estimates of early cancer diagnosis in England: a population-based study |
| |
| Institution: | 1. Institute of Medical Informatics, Biometry and Epidemiology, University Hospital Essen, Germany;2. School of Public Health, Department of Epidemiology, Boston University, Boston, USA;3. Cancer Registry of North Rhine-Westphalia, Bochum, Germany;4. Saarland Cancer Registry, Saarbrücken, Germany;5. CaritasKlinikum Saarbrücken St. Theresia, Saarbrücken, Germany;1. Cancer Control Office, King Hussein Cancer Center, Amman 11941, Jordan;2. Section of Pulmonary and Critical Care, Department of Internal Medicine, King Hussein Cancer Center, Amman 11941, Jordan;3. Pharmacy Student, University of Jordan, Amman 11972, Jordan;4. Volunteer Research Program at King Hussein Cancer Center, Amman 11941, Jordan;5. Medical Student, University of Jordan, Amman 11972, Jordan;1. Maccabi Institute for Research and Innovation (Maccabitech), Maccabi Healthcare Services, HaMered 27, Tel Aviv, 68125, Israel;2. MSD Israel, Merck Sharp & Dohme (Israel-1996) Company Ltd. 34 Hacharash St. P.O.B 7340, Hod Hasharon 45240, Israel;3. Institute of Oncology, Sheba Medical Center, Tel Hashomer, Ramat Gan, Israel;4. Department of Health Systems Management, Ben-Gurion University of the Negev, Beer-Sheva, Israel;5. Sackler Faculty of Medicine, Tel Aviv University, Israel;1. ICES, Toronto, Canada;2. Manitoba Centre for Health Policy, Department of Community Health Sciences, University of Manitoba, Winnipeg, Canada;3. Faculty of Health, Dalhousie University, Halifax, Canada;4. Dalla Lana School of Public Health, University of Toronto, Toronto, Canada;5. School of Rehabilitation Therapy, Queen’s University, Kingston, Canada;6. Institute for Health Policy, Management, and Evaluation, University of Toronto, Toronto, Canada;1. School of Public Health and Preventive Medicine, Monash University, 553 St Kilda Road, Melbourne, VIC 3004, Australia;2. Victorian Melanoma Service, Alfred Health, 55 Commercial Road, Melbourne, VIC 3004, Australia;3. Department of Anatomical Pathology, Alfred Health, 55 Commercial Road, Melbourne, VIC 3004, Australia;4. Gastrointestinal and Other Cancers Research Group, Division of Cancer Prevention, National Cancer Institute, NIH, Bethesda, MD, USA;5. Walter Reed National Military Medical Center (WRNMM) Uniformed Services University (USU) Department of Surgery, Bethesda, MD, USA;6. The Walter & Eliza Hall Institute of Medical Research, University of Melbourne,1 G Royal Parade, Parkville, Victoria 3052, Australia;7. Clinical and Translational Epidemiology Unit, Massachusetts General Hospital, Massachusetts, USA;8. Department of Medical Oncology, Alfred Health, 55 Commercial Road, Melbourne, VIC 3004, Australia |
| |
| Abstract: | IntroductionMonitoring early diagnosis is a priority of cancer policy in England. Information on stage has not always been available for a large proportion of patients, however, which may bias temporal comparisons. We previously estimated that early-stage diagnosis of colorectal cancer rose from 32% to 44% during 2008–2013, using multiple imputation. Here we examine the underlying assumptions of multiple imputation for missing stage using the same dataset.MethodsIndividually-linked cancer registration, Hospital Episode Statistics (HES), and audit data were examined. Six imputation models including different interaction terms, post-diagnosis treatment, and survival information were assessed, and comparisons drawn with the a priori optimal model. Models were further tested by setting stage values to missing for some patients under one plausible mechanism, then comparing actual and imputed stage distributions for these patients. Finally, a pattern-mixture sensitivity analysis was conducted.ResultsData from 196,511 colorectal patients were analysed, with 39.2% missing stage. Inclusion of survival time increased the accuracy of imputation: the odds ratio for change in early-stage diagnosis during 2008–2013 was 1.7 (95% CI: 1.6, 1.7) with survival to 1 year included, compared to 1.9 (95% CI 1.9–2.0) with no survival information. Imputation estimates of stage were accurate in one plausible simulation. Pattern-mixture analyses indicated our previous analysis conclusions would only change materially if stage were misclassified for 20% of the patients who had it categorised as late.InterpretationMultiple imputation models can substantially reduce bias from missing stage, but data on patient’s one-year survival should be included for highest accuracy. |
| |
| Keywords: | Stage at diagnosis Early diagnosis Time trends Temporal changes Population-based Routine data Missing stage Missing data Multiple imputation Sensitivity analysis Pattern mixture Survival |
| 本文献已被 ScienceDirect 等数据库收录！ |
|
