Cited by 892Statistical Methods and Bayesian InferenceStatistical Methods in EpidemiologyAdvanced Causal Inference Techniques

The DOI matches its Crossref record.

A critical look at methods for handling missing covariates in epidemiologic regression analyses.

Greenland, Sander, Finkle, William D.·1995·American Journal of Epidemiology

DOI 10.1093/oxfordjournals.aje.a117592

Open this paper in CoChat — free account

Abstract

Epidemiologic studies often encounter missing covariate values. While simple methods such as stratification on missing-data status, conditional-mean imputation, and complete-subject analysis are commonly employed for handling this problem, several studies have shown that these methods can be biased under reasonable circumstances. The authors review these results in the context of logistic regression and present simulation experiments showing the limitations of the methods. The method based on missing-data indicators can exhibit severe bias even when the data are missing completely at random, and regression (conditional-mean) imputation can be inordinately sensitive to model misspecification. Even complete-subject analysis can outperform these methods. More sophisticated methods, such as maximum likelihood, multiple imputation, and weighted estimating equations, have been given extensive attention in the statistics literature. While these methods are superior to simple methods, they are not commonly used in epidemiology, no doubt due to their complexity and the lack of packaged software to apply these methods. The authors contrast the results of multiple imputation to simple methods in the analysis of a case-control study of endometrial cancer, and they find a meaningful difference in results for age at menarche. In general, the authors recommend that epidemiologists avoid using the missing-indicator method and use more sophisticated methods whenever a large proportion of data are missing.

Integrity signals

Registry

Crossref · 10.1093/oxfordjournals.aje.a117592

Cites

Statistics

Recomputing p-values needs the paper’s full text.

Claims the paper makes

This paper’s full text isn’t openly available, so its claims and self-check weren’t run. The registry and citation checks still apply.

Claims other papers make about this one

Each is checked against this paper’s own text: does it say what the citing paper says it does?

Sorted by citations

Items 501–600 of 892

Surrogate marker analysis in cancer clinical trials through time-to-event mediation techniques
Vandenberghe, S., Duchateau, L., Slaets, L., Bogaerts, J., & Vansteelandt, S. (2017) · Statistical Methods in Medical Research · doi:10.1177/0962280217702179 · Cited by 19
Adverse lipid profile elevates risk for subarachnoid hemorrhage: A prospective population-based cohort study
Lindbohm, J. V., Korja, M., Jousilahti, P., Salomaa, V. V., & Kaprio, J. A. (2018) · Atherosclerosis · doi:10.1016/j.atherosclerosis.2018.05.011 · Cited by 19
Mediterranean Diet Adherence and Risk of Depressive Symptomatology in a French Population-Based Cohort of Older Adults
Bardinet, J., Chuy, V., Carrière, I., Galéra, C., Pouchieu, C., Samieri, C., Helmer, C., Cougnard‐Grégoire, A., & Féart, C. (2022) · Nutrients · doi:10.3390/nu14194121 · Cited by 18
“Some covariates had missing data: to keep the maximum of information, we ran a multiple imputation by chained equations using the mice packages [33], an elaborate strategy limiting the bias introduced by the missing data under certain conditions (assumption of missing at random data) [34].”
Method
Repeat Abortions in New York City, 2010
Toprani, A. (2015) · Journal of Urban Health · doi:10.1007/s11524-015-9946-3 · Cited by 18
“Additionally, we had a relatively high proportion of missing values for education, requiring the use of multiple imputation, which although shown to produce less bias and greater precision than complete case analysis, may still introduce bias.(35) Finally, we could not exclude therapeutic abortions, i.”
Background
Changes in residential greenness between pregnancies and birth outcomes: longitudinal evidence from Michigan births 1990--2012
Margerison, C. E., Pearson, A. L., Lin, Z., & Sanciangco, J. C. (2020) · International Journal of Epidemiology · doi:10.1093/ije/dyaa158 · Cited by 18
Supported Employment for Veterans With Traumatic Brain Injury: Patient Perspectives
Carlson, K. F., Pogoda, T. K., Gilbert, T. A., Resnick, S. G., Twamley, E. W., O'Neil, M. E., & Sayer, N. A. (2017) · Archives of Physical Medicine and Rehabilitation · doi:10.1016/j.apmr.2017.06.027 · Cited by 18
Discriminating Inhalational Anthrax From Community-Acquired Pneumonia Using Chest Radiograph Findings and a Clinical Algorithm
Kyriacou, D. N., Yarnold, P. R., Stein, A. C., et al. (2007) · CHEST Journal · doi:10.1378/chest.06-1687 · Cited by 18
Decision-making in temporal lobe epilepsy surgery: The contribution of basic non-invasive tests
Uijl, S. G., Leijten, F. S. S., Arends, J. B. A. M., Parra, J., van Huffelen, A. C., & Moons, K. G. M. (2008) · Seizure · doi:10.1016/j.seizure.2007.11.022 · Cited by 19
The impact of preschool wheezing patterns on health-related quality of life at age 4 years
Groen, E. H., Hindori‐Mohangoo, A. D., Landgraf, J. M., de Jongste, J. C., Duijts, L., Moll, H. A., Jaddoe, V. W. V., Albert Hofman, D., & Raat, H. (2012) · European Respiratory Journal · doi:10.1183/09031936.00015712 · Cited by 18
“A multiple imputation method was used to impute missing covariates (28).”
Method
Predictive Factors for Delivery within 7 Days after Successful 48-Hour Treatment of Threatened Preterm Labor
Schuit, E., Scheepers, H. C. J., Bloemenkamp, K. W. M., Bolte, A. C., Duvekot, H., van Eyck, J., Kok, J. H., Kwee, A., Merién, A., Opmeer, B. C., et al. (2015) · American Journal of Perinatology Reports · doi:10.1055/s-0035-1552930 · Cited by 19
Describing Discrete-Time Event Occurrence Data
Singer, J. D., & Willett, J. B. (2003) · doi:10.1093/acprof:oso/9780195152968.003.0010 · Cited by 18
Ethnic Background and Television Viewing Time Among 4-Year-Old Preschool Children
Wijtzes, A. I., Jansen, W., Jaddoe, V. W. V., Moll, H. A., Tiemeier, H. W., Verhulst, F. C., Albert Hofman, D., Mackenbach, J. P., & Raat, H. (2013) · Journal of Developmental & Behavioral Pediatrics · doi:10.1097/dbp.0b013e31827b163a · Cited by 18
Impact of extracardiac vascular disease on acute prognosis in patients who undergo percutaneous coronary interventions (data from the Blue Cross & Blue Shield of Michigan Cardiovascular Consortium [BMC2]).
Mukherjee, D., Eagle, K. A., Smith, D. E., Kline-Rogers, E., Chetcuti, S., Grossman, P., Nallamothu, B., O'Donnell, M. J., Defranco, A., Maxwell-Eward, A., et al. (2003) · American Journal of Cardiology · doi:10.1016/s0002-9149(03)00981-0 · Cited by 20
A pragmatic single-blind randomised controlled trial and economic evaluation of the use of leukotriene receptor antagonists in primary care at steps 2 and 3 of the national asthma guidelines (ELEVATE study).
Price, D. B., Musgrave, S., Wilson, E., Sims, E. J., Shepstone, L., Blyth, A., Murdoch, J., Mugford, M., Juniper, E. F., Ayres, J. G., et al. (2011) · Health Technology Assessment · doi:10.3310/hta15210 · Cited by 17
Mean serum-level of common organic pollutants is predictive of behavioral severity in children with autism spectrum disorders
Boggess, A., Faber, S., Kern, J. C., & Kingston, H. M. (2016) · Scientific Reports · doi:10.1038/srep26185 · Cited by 18
Preoperative Hyponatremia Is Associated with Reoperation and Prolonged Length of Hospital Stay following Total Knee Arthroplasty
Abola, M. V., Tanenbaum, J., Bomberger, T. T., Knapik, D. M., Fitzgerald, S., & Wera, G. D. (2018) · The Journal of Knee Surgery · doi:10.1055/s-0038-1641156 · Cited by 18
Overview of Federated Facility to Harmonize, Analyze and Management of Missing Data in Cohorts
Rajula, H. S. R., Odintsova, V. V., Manchia, M., & Fanos, V. (2019) · Applied Sciences · doi:10.3390/app9194103 · Cited by 18
“The missing indicator method comprises an additional category for the analysis created for applicants with missing data [41].”
Method
Imputation of missing covariate in randomized controlled trials with a continuous outcome: Scoping review and new results
Kayembe, M. T., Jolani, S., Tan, F. E. S., & van Breukelen, G. (2020) · Pharmaceutical Statistics · doi:10.1002/pst.2041 · Cited by 16
Usefulness of QT-interval parameters for cardiovascular risk stratification in type 2 diabetic patients with arterial hypertension
Salles, G. F., Deccache, W., & Cardoso, C. R. L. (2005) · Journal of Human Hypertension · doi:10.1038/sj.jhh.1001815 · Cited by 17
Simultaneous SNP identification in association studies with missing data
Li, Z., Gopal, V., Li, X., et al. (2012) · The Annals of Applied Statistics · doi:10.1214/11-aoas516 · Cited by 17
“In the correlation comparison, the true non-zero SNPs in the model were (1),(2),(4),(5),(8),(9),(10),(11),(16),(17 ),(18),(22),(24) and (25).”
BackgroundMethodInfluential
Introducing the Multilevel Model for Change
Singer, J. D., & Willett, J. B. (2003) · doi:10.1093/acprof:oso/9780195152968.003.0003 · Cited by 19
A new imputation method for incomplete binary data
Subasi, M. M., Subasi, E., Anthony, M., et al. (2011) · Discrete Applied Mathematics · doi:10.1016/j.dam.2011.01.024 · Cited by 17
“Each data-set is analyzed separately using techniques designed for complete data, and the results are then combined in such a way that the variability due to imputation may be incorporated.”
Method
Ethnic differences in the growth of low‐birthweight infants
Seed, P. T., Ogundipe, & Wolfe, C. D. A. (2000) · Paediatric and Perinatal Epidemiology · doi:10.1046/j.1365-3016.2000.00227.x · Cited by 17
The “Why” behind including “Y” in your imputation model
McGowan, L. D., Lotspeich, S. C., & Hepler, S. A. (2024) · Statistical Methods in Medical Research · doi:10.1177/09622802241244608 · Cited by 16
Secondary Hyperparathyroidism and Anemia in Children Treated by Hemodialysis
Smith, L. B., Fadrowski, J. J., Howe, C. J., et al. (2010) · American Journal of Kidney Diseases · doi:10.1053/j.ajkd.2009.09.033 · Cited by 16
“This statistical approach completes missing values for a given variable based on other covariates, specifically the covariates shown to predict whether the variable to be imputed (iPTH in this case) is missing (19).”
Method
Adjustment for Missing Data in Complex Surveys Using Doubly Robust Estimation
Wirth, K. E., Tchetgen, E. J. T., & Murray, M. B. (2010) · Epidemiology · doi:10.1097/ede.0b013e3181f57571 · Cited by 16
“Numerous methods have been proposed to correct for nonresponse, including inverse probability weighting and outcome regression.(5) Inverse probability weighting adjusts for nonresponse by weighting the outcomes of participants with nonmissing information by the inverse of the probability of having complete data (obtai…”
BackgroundMethod
Thromboembolic stroke in patients with a HeartMate-II left ventricular assist device – the role of anticoagulation
van den Bergh, W. M., Lansink‐Hartgring, A. O., van Duijn, A. L., et al. (2015) · Journal of Cardiothoracic Surgery · doi:10.1186/s13019-015-0333-7 · Cited by 16
“Excluding subjects with missing values not only leads to loss of statistical power, but also to biased results [15–17].”
Background
Missing data treatments matter: an analysis of multiple imputation for anterior cervical discectomy and fusion procedures
Ondeck, N. T., Fu, M. C., Skrip, L., McLynn, R. P., Cui, J. J., Basques, B. A., Albert, T. J., & Grauer, J. N. (2018) · The Spine Journal · doi:10.1016/j.spinee.2018.04.001 · Cited by 16
An Improved Imputation Method for Accurate Prediction of Imputed Dataset Based Radon Time Series
Mir, A. A., Çelebi, F. V., Rafique, M., Hussain, L., Almasoud, A. S., Alajmi, M., Al‐Wesabi, F. N., & Hilal, A. M. (2022) · IEEE Access · doi:10.1109/access.2022.3151892 · Cited by 16
“The missingness of the data can be classified according to the means through which it is generated [37].”
BackgroundMethod
Interleukin-13 Genetic Variants, Household Carpet Use and Childhood Asthma
Tsai, C., Tung, K., Su, M., Chiang, B., Chew, F. T., Kuo, N.-W., & Lee, Y. L. (2013) · PLoS ONE · doi:10.1371/journal.pone.0051970 · Cited by 16
“Subjects with missing covariate information were included in the model using missing indicators [29].”
Method
Maternal cigarette smoking during pregnancy and pubertal development in sons. A follow‐up study of a birth cohort
Håkonsen, L. B., Olsen, J., Støvring, H., Ernst, A., Thulstrup, A. M., Zhu, J. L., Shrestha, A., & Ramlau‐Hansen, C. H. (2013) · Andrology · doi:10.1111/j.2047-2927.2012.00038.x · Cited by 16
“As complete case analysis may lead to biased estimates of the association studied (Greenland & Finkle, 1995; Sterne et al., 2009), we addressed the missing data problem by using multiple imputations, which yield unbiased and more precise estimates if data are missing at random (Rubin, 1987; Sterne…”
Missing portion sizes in FFQ – alternatives to use of standard portions
Køster‐Rasmussen, R., Siersma, V., Halldorsson, T. I., et al. (2014) · Public Health Nutrition · doi:10.1017/s1368980014002389 · Cited by 16
An Incremental Algorithm for Repairing Training Sets with Missing Values
van Stein, B., & Kowalczyk, W. (2016) · Communications in computer and information science · doi:10.1007/978-3-319-40581-0_15 · Cited by 16
Concurrent assessment of urban environment and cardiometabolic risk over 10 years in a middle‐aged population‐based cohort
Daniel, M., Carroll, S. J., Niyonsenga, T., Piggott, E. J., Taylor, A. W., & Coffee, N. T. (2018) · Geographical Research · doi:10.1111/1745-5871.12318 · Cited by 17
Lipid Screening in HIV-Infected Veterans
Korthuis, P. T., Asch, S. M., Anaya, H. D., Morgenstern, H., Goetz, M. B., Yano, E. M., Rubenstein, L. V., Lee, M. L., & Bozzette, S. A. (2004) · JAIDS Journal of Acquired Immune Deficiency Syndromes · doi:10.1097/00126334-200403010-00005 · Cited by 15
Inequality of opportunity in selection procedures limits diversity in higher education: An intersectional study of Dutch selective higher education programs
Mulder, L., Akwiwu, E. U., Twisk, J. W. R., Koster, A. S., Ravesloot, J. H., Croiset, G., Kusurkar, R. A., & Wouters, A. (2023) · PLoS ONE · doi:10.1371/journal.pone.0292805 · Cited by 18
“However, there are methods for handling missing data, e.g. multiple data imputation [66].”
Method
Covariate adjustment in randomized clinical trials with missing covariate and outcome data
Chang, C., Song, Y., Li, F., & Wang, R. (2023) · Statistics in Medicine · doi:10.1002/sim.9840 · Cited by 20
“The use of missing indicator method in the context of observational studies has been investigated extensively (Jones, 1996; Greenland and Finkle, 1995; Song et al., 2021).”
BackgroundMethod
A novel multi-parameter scoring system for distinguishing sarcoidosis from sputum negative tuberculosis.
Li, Q. H., Li, H. P., Shen, Y., et al. (2012) · Sarcoidosis vasculitis and diffuse lung diseases · Cited by 15
Overweight in Infancy: Which Pre- and Perinatal Factors Determine Overweight Persistence or Reduction A Birth Cohort Followed for 11 Years
van Rossem, L., Wijga, A. H., Brunekreef, B., et al. (2014) · Annals of Nutrition and Metabolism · doi:10.1159/000360305 · Cited by 15
Neighborhood Ethnic Diversity and Behavioral and Emotional Problems in 3 Year Olds: Results from the Generation R Study
Flink, I. J. E., Prins, R., Mackenbach, J., Jaddoe, V. W. V., Albert Hofman, D., Verhulst, F. C., Tiemeier, H. W., & Raat, H. (2013) · PLoS ONE · doi:10.1371/journal.pone.0070070 · Cited by 15
“To handle missing data in the individual-level confounders, multiple imputation was applied [27].”
Method
Acute Kidney Injury and Long-term Prognosis After Acute Myocardial Infarction
Bouzas‐Mosquera, A., Vázquez‐Rodríguez, J. M., Peteiro, J., et al. (2009) · Archives of Internal Medicine · doi:10.1001/archinternmed.2008.554 · Cited by 15
Outcomes of basilar artery occlusion in patients aged 75 years or older in the Basilar Artery International Cooperation Study
Vergouwen, M. D. I., Compter, A., Tanné, D., et al. (2012) · Journal of Neurology · doi:10.1007/s00415-012-6498-2 · Cited by 15
“Missing baseline data (\5 % for each variable) were imputed with regression imputation [10].”
Method
PROP taster status, food preferences and consumption of high‐calorie snacks and sweet beverages among 6‐year‐old ethnically diverse children
Wijtzes, A. I., Jansen, W., Bouthoorn, S. H., Jong, J. C. K., Jansen, P. W., Franco, O. H., Jaddoe, V. W. V., Albert Hofman, D., & Raat, H. (2016) · Maternal and Child Nutrition · doi:10.1111/mcn.12240 · Cited by 15
Diagnostic accuracy of modified transoesophageal echocardiography for pre-incision assessment of aortic atherosclerosis in cardiac surgery patients
van Zaane, B., Nierich, A. P., Bruinsma, G. J. B. B., Rosseel, P. M. J., Ramjankhan, F. Z., de Waal, E. E. C., Bühre, W., & Moons, K. G. M. (2010) · British Journal of Anaesthesia · doi:10.1093/bja/aeq122 · Cited by 15
Fitting Basic Discrete-Time Hazard Models
Singer, J. D., & Willett, J. B. (2003) · doi:10.1093/acprof:oso/9780195152968.003.0011 · Cited by 16
Missing Data in Epidemiologic Studies
Vach, W., & Blettner, M. (2005) · Encyclopedia of Biostatistics · doi:10.1002/0470011815.b2a03085 · Cited by 15
Measuring Quality With Missing Data: The Invisible Threat to National Quality Initiatives
Newgard, C. D., & Haukoos, J. S. (2010) · Academic Emergency Medicine · doi:10.1111/j.1553-2712.2010.00883.x · Cited by 14
Comparing Training Approaches for Technological Skill Development in Introductory Statistics Courses
Baglin, J., & Costa, C. D. (2013) · Technology Innovations in Statistics Education · doi:10.5070/t571014007 · Cited by 14
Multi-drug resistant TB and HIV in Thailand: overlapping, but not independently associated risk factors.
Akksilp, S., Wattanaamornkiat, W., Kittikraisak, W., et al. (2009) · PubMed · Cited by 14
Using the UNOS/SRTR and PHTS Databases to Improve Quality in Pediatric Cardiac Transplantation
Davies, R. R., & Pizarro, C. (2012) · World Journal for Pediatric and Congenital Heart Surgery · doi:10.1177/2150135112443971 · Cited by 14
“In addition, both may result in severely biased estimates of effect and statistical significance.(40,41) Imputation as a general method is simply the replacement of a missing value with an estimate of the likely value.”
Background
Nutrient Pathways and Breast Cancer Risk: The Long Island Breast Cancer Study Project
Bradshaw, P. T., Khankari, N. K., Teitelbaum, S. L., Xu, X., Fink, B., Steck, S. E., Gaudet, M. M., Kabat, G. C., Wolff, M. S., Neugut, A. I., et al. (2013) · Nutrition and Cancer · doi:10.1080/01635581.2013.767364 · Cited by 14
“The remaining 977 individuals did not respond; effects, however, were not estimated separately for subjects missing supplement use because it is an inappropriate method for dealing with missing data (31).”
Method
Unilateral Disease Is Common in Patients With Primary Aldosteronism Without Adrenal Nodules
Sam, D., Kline, G. A., So, B., Przybojewski, S. J., & Leung, A. A. C. (2020) · Canadian Journal of Cardiology · doi:10.1016/j.cjca.2020.05.013 · Cited by 14
Clinical Prediction Rules
Holmes, J. F. (2011) · Springer eBooks · doi:10.1007/978-1-4614-1599-2_16 · Cited by 14
Analysis of the Benefits of a Mediterranean Diet in the GISSI-Prevenzione Study: A Case Study in Imputation of Missing Values from Repeated Measurements
Barzi, F., Woodward, M., Marfisi, R. M., Tognoni, G., Marchioli, R., & on behalf of GISSI-Prevenzione Investigators (2006) · European Journal of Epidemiology · doi:10.1007/s10654-005-5086-5 · Cited by 14
Genetic influences on adult body mass index followed over 29 years and their effects on late-life mobility: a study of twin sisters
Ortega‐Alonso, A., Sipilä, S., Kujala, U. M., Kaprio, J. A., & Rantanen, T. (2009) · Journal of Epidemiology & Community Health · doi:10.1136/jech.2008.080622 · Cited by 15
24 The role of interacting determinants in the localization of genes
Gauderman, W. J., & Thomas, D. C. (2001) · Advances in genetics · doi:10.1016/s0065-2660(01)42033-5 · Cited by 14
Statistical Issues in the Estimation of the Causal Effects of Smoking Due to the Conduct of the Tobacco Industry
Rubin, D. B. (2000) · Springer eBooks · doi:10.1007/978-1-4612-1216-4_16 · Cited by 13
Efficacy of statistical algorithms in imputing missing data of streamflow discharge imparted with variegated variances and seasonalities
Gao, Y., Semiromi, M. T., & Merz, C. (2023) · Environmental Earth Sciences · doi:10.1007/s12665-023-11139-z · Cited by 13
“Regression‑based imputation replaces missing data with predicted values from a regression estimation (Greenland and Finkle 1995).”
Method
Handling complete short-term data logging failure in smart buildings: Machine learning based forecasting pipelines with sliding-window training scheme
Papadopoulos, D. N., Javan, F. D., Najafi, B., et al. (2023) · Energy and Buildings · doi:10.1016/j.enbuild.2023.113694 · Cited by 13
Associations of ambient air pollution and lifestyle with the risk of NAFLD: a population-based cohort study
Kong, X., Huang, R., Geng, R., Wu, J., Li, J., Wu, Y., Zhao, Y., You, D., Yu, H., Du, M., et al. (2024) · BMC Public Health · doi:10.1186/s12889-024-19761-7 · Cited by 13
“Multiple imputation by chained equations was performed to impute missing covariates data (Additional file 1: eMethods) [25].”
Method
Results Differ by Applying Distinctive Multiple Imputation Approaches on the Longitudinal Cardiovascular Health Study Data
Ning, Y. M., McAvay, G., Chaudhry, S. I., Arnold, A. M., & Allore, H. G. (2013) · Experimental Aging Research · doi:10.1080/0361073x.2013.741968 · Cited by 13
“In contrast to the time ordering and exclusion of subjects after death in the sequential approach, the simultaneous approach such as that used by Harel et al. (2007) imputes all missing data together, including those post death; then, before the analysis stage, all values imputed after a participant’s death are remove…”
BackgroundMethod
Patient-reported Outcomes as Predictors of Change in Disease Activity and Disability in Early Rheumatoid Arthritis: Results from the Yorkshire Early Arthritis Register
Twigg, S., Hensor, E. M. A., Emery, P., Tennant, A., & Morgan, A. W. (2017) · The Journal of Rheumatology · doi:10.3899/jrheum.161214 · Cited by 13
Examining environmental contaminant mixtures among adults with type 2 diabetes in the Cree First Nation communities of Eeyou Istchee, Canada
Zuk, A. M., Tsuji, L. J. S., Nieboer, E., et al. (2019) · Scientific Reports · doi:10.1038/s41598-019-52200-x · Cited by 13
Modern multiple imputation with functional data
Rao, A. R., & Reimherr, M. L. (2020) · Stat · doi:10.1002/sta4.331 · Cited by 13
“ast, if the observed data is highly predictive of the missing values, the imputations will be more consistent across the multiple imputed data sets, resulting in smaller and reliable standard errors (Greenland and Finkle, 1995). Longitudinal studies are amenable to Functional Data Analysis (FDA), often contain sparse…”
Background
Propensity Score Analysis with Partially Observed Baseline Covariates: A Practical Comparison of Methods for Handling Missing Data
Bottigliengo, D., Lorenzoni, G., Ocagli, H., Martinato, M., Berchialla, P., & Gregori, D. (2021) · International Journal of Environmental Research and Public Health · doi:10.3390/ijerph18136694 · Cited by 13
Ignoring Non-ignorable Missingness
Rabe‐Hesketh, S., & Skrondal, A. (2022) · Psychometrika · doi:10.1007/s11336-022-09895-1 · Cited by 13
“Pothoff et al. (2006) call this assumptionMAR+, Greenland and Finkle (1995) refer to it as “stratifiedMCAR,” and Mohan et al. (2013) define their MAR assumption this way.”
Background
Estimating Crude or Common Odds Ratios in Case-Control Studies with Informatively Missing Exposure Data
Lyles, R. H. (2002) · American Journal of Epidemiology · doi:10.1093/aje/155.3.274 · Cited by 13
“While this assumption often helps facilitate a formal analytical treatment, it can be invalid in epidemiologic settings (1).”
BackgroundMethod
Validating and updating a prediction rule for neurological sequelae after childhood bacterial meningitis
Biesheuvel, C. J., Koomen, I., Vergouwe, Y., van Furth, A. M., Oostenbrink, R., Moll, H. A., Grobbee, D. E., & Moons, K. G. M. (2005) · Scandinavian Journal of Infectious Diseases · doi:10.1080/00365540500372671 · Cited by 13
“Therefore, it is better to impute the missing values [12,18].”
Background
Impact of imputation of missing data on estimation of survival rates: an example in breast cancer.
Baneshi, M. R., & Talei, A. (2010) · Iranian Journal of Cancer Prevention · Cited by 14
Derivation and Validation of Predictive Factors for Clinical Deterioration after Admission in Emergency Department Patients Presenting with Abnormal Vital Signs Without Shock
Henning, D. J., Oedorf, K., Day, D. E., et al. (2015) · Western Journal of Emergency Medicine · doi:10.5811/westjem.2015.9.27348 · Cited by 13
Setting up a multivariable analysis
Katz, M. H. (2006) · Cambridge University Press eBooks · doi:10.1017/cbo9780511811692.008 · Cited by 13
Correcting bias due to missing stage data in the non-parametric estimation of stage-specific net survival for colorectal cancer using multiple imputation
Falcaro, M., & Carpenter, J. R. (2017) · Cancer Epidemiology · doi:10.1016/j.canep.2017.02.005 · Cited by 12
“In this setting, the ad-hoc and yet relatively popular approach of creating an extra category for the missing values has been shown to lead to severe bias [8,24].”
Background
Depressive symptoms prior to and after incident cardiovascular disease and long-term survival. A population-based study of older persons
Freak‐Poli, R., Ikram, M. A., Franco, O. H., Albert Hofman, D., & Tiemeier, H. W. (2017) · Depression and Anxiety · doi:10.1002/da.22689 · Cited by 13
“Volume 6 • Issue 2 • 1000270 J Depress Anxiety, an open access journal ISSN: 2167-1044 age, sex, education and prior repeated measures of the confounder of interest [22].”
Background
Regional Variation and Trends in the Timing of Lower Extremity Reconstruction: A 10-Year Review of the Nationwide Inpatient Sample
Sheckter, C. C., Pridgen, B. C., Li, A. H., et al. (2018) · Plastic & Reconstructive Surgery · doi:10.1097/prs.0000000000004885 · Cited by 13
On the Relation between Prediction and Imputation Accuracy under Missing Covariates
Ramosaj, B., Tulowietzki, J., & Pauly, M. (2022) · Entropy · doi:10.3390/e24030386 · Cited by 13
“It is already well-known that more naive methods such as list-wise deletion or mean imputation can lead to severe estimation bias, see e.g. [7, 8, 9, 10, 1].”
Background
Handling incomplete smoking history data in survival analysis
Furukawa, K., Preston, D. L., Misumi, M., & Cullings, H. M. (2014) · Statistical Methods in Medical Research · doi:10.1177/0962280214556794 · Cited by 13
Intraindividual association between shift work and risk of drinking problems: data from the Finnish Public Sector Cohort
Cheng, W., Härmä, M., Koskinen, A., Mika Kivimaki, B., Oksanen, T., & Huang, M. (2021) · Occupational and Environmental Medicine · doi:10.1136/oemed-2020-107057 · Cited by 16
“22 but may lead to bias when the data were missing at random.(37) Although the missingness for “having young children” was not random but related to survey years (i.”
BackgroundMethod
Real‐world efficacy and safety outcomes of imatinib treatment in patients with chronic myeloid leukemia: An Australian experience
Adattini, J. A., Gross, A. S., Doo, N. W., et al. (2022) · Pharmacology Research & Perspectives · doi:10.1002/prp2.1005 · Cited by 12
Recalibration methods to enhance information on prevalence rates from large mental health surveys
Taub, N. A., Morgan, Z., Brugha, T. S., Lambert, P. C., Bebbington, P., Jenkins, R., Kessler, R. C., Zaslavsky, A. M., & Hotz, T. L. (2005) · International Journal of Methods in Psychiatric Research · doi:10.1002/mpr.13 · Cited by 12
“However, many of the methods have limitations in terms of bias, precision and computational feasibility (Greenland and Finkle, 1995; Zhou et al., 2001).”
Method
Predictors of a normal chest x‐ray in respiratory syncytial virus infection
Kneyber, M. C. J., Moons, K. G. M., Groot, R. D., et al. (2001) · Pediatric Pulmonology · doi:10.1002/ppul.1039 · Cited by 12
“Missing values were imputed using the expectation and maximization method available in the standard SPSS software to increase statistical ef®ciency.(9,10) Such imputation of missing variables is based on the correlation between each variable with missing values from all other variables.”
A prediction model for left ventricular mass in patients at high cardiovascular risk
Meijs, M. F. L., Vergouwe, Y., Cramer, M. J., et al. (2010) · European Journal of Cardiovascular Prevention & Rehabilitation · doi:10.1097/hjr.0b013e328332d4bc · Cited by 12
MARKOV CHAIN MONTE CARLO METHOD FOR HANDLING MISSING DATA IN AIR QUALITY DATASETS
(2017) · Malaysian Journal of Analytical Science · doi:10.17576/mjas-2017-2103-05 · Cited by 12
“MI gives much better results to handle missing data [14].”
Result
Inference for case–control studies when exposure status is both informatively missing and misclassified
Lyles, R. H., Allen, A. S., Flanders, W. D., et al. (2006) · Statistics in Medicine · doi:10.1002/sim.2500 · Cited by 12
“Reference [7]), with the topic receiving limited attention in the epidemiological literature [8, 9].”
A Descriptive Study of Verocytotoxigenic Escherichia coli (VTEC) Cases Reported in Ontario, 1990–1994
Michel, P., Wilson, J. B., Martin, S. W., et al. (1998) · Canadian Journal of Public Health · doi:10.1007/bf03403929 · Cited by 12
Statistical analysis of genotype and gene expression data
Schwender, H. R. (2007) · Technische Universität Dortmund Eldorado (Technische Universität Dortmund) · doi:10.17877/de290r-8430 · Cited by 12
Proper Use of Multiple Imputation and Dealing with Missing Covariate Data
Saffari, S. E., Volovici, V., Ong, M. E. H., Goldstein, B. A., VAUGHAN, R. D., Dammers, R., Steyerberg, E. W., & Liu, N. (2022) · World Neurosurgery · doi:10.1016/j.wneu.2021.10.090 · Cited by 11
Individuals with schizophrenia who act violently towards others profit unequally from inpatient treatment—Identifying subgroups by latent class analysis
Kirchebner, J., Lau, S., Kling, S., Sonnweber, M., & Günther, M. P. (2020) · International Journal of Methods in Psychiatric Research · doi:10.1002/mpr.1856 · Cited by 11
“Prior to applying any statistical technique, multiple imputation (Greenland & Finkle, 1995; Zhou, Eckert, & Tierney, 2001) by chained equations (MICE; Van Buuren, Boshuizen, & Knook, 1999) was applied to optimally address missing values.”
Method
Tramadol versus codeine and the short‐term risk of cardiovascular events in patients with non‐cancer pain: A population‐based cohort study
Ou, L. B., Azoulay, L., Reynier, P., Platt, R. W., Yoon, S., Grad, R. M., & Filion, K. B. (2021) · British Journal of Clinical Pharmacology · doi:10.1111/bcp.15099 · Cited by 10
“Finally, the use of an indicator variable for missing data could lead to bias if the variable with missing data is an important confounder.(38) Therefore, we used multiple imputation using the fully conditional specification approach to impute missing data for BMI, smoking, and blood pressure for the primary endpoint…”
Method
Imputation of missing values for cochlear implant candidate audiometric data and potential applications
Pavelchek, C., Michelson, A. P., Walia, A., Ortmann, A. J., Herzog, J. A., Buchman, C. A., & Shew, M. A. (2023) · PLoS ONE · doi:10.1371/journal.pone.0281337 · Cited by 12
“While simple, these methods can introduce significant bias [5,6,9,10].”
Method
Accuracy of symptoms, signs, and C-reactive protein for early chronic obstructive pulmonary disease
Broekhuizen, B. D., Sachs, A. P. E., Verheij, T. J., et al. (2012) · British Journal of General Practice · doi:10.3399/bjgp12x654605 · Cited by 11
Impact of diabetes mellitus on 5-year clinical outcomes following successful endovascular revascularization for peripheral artery disease
Lee, M. S., Choi, B., & Rha, S. (2020) · Vascular Medicine · doi:10.1177/1358863x19879751 · Cited by 6
Healthy Lifestyle Pattern is Protective Against 30-Yr Cancer Incidence in Men and Women: A Cohort Study
Orenstein, L., Chetrit, A., & Dankner, R. (2016) · Nutrition and Cancer · doi:10.1080/01635581.2016.1153673 · Cited by 11
Changes in adiposity and other factors in relation to age at natural menopause: analyses from the coronary artery risk development (CARDIA) study
Costanian, C., McCague, H., Edgell, H., Ardern, C. I., & Tamim, H. (2018) · Menopause The Journal of The North American Menopause Society · doi:10.1097/gme.0000000000001196 · Cited by 12
Short- and Long-term Reliability of Information on Previous Illness and Family History as Compared with that on Smoking and Drinking Habits in Questionnaire Surveys.
Zhu, S., Toyoshima, H., Kondo, T., et al. (2002) · Journal of Epidemiology · doi:10.2188/jea.12.120 · Cited by 11
Assessing alternative imputation strategies for infrequently missing items on multi-item scales
Rezvan, P. H., Comulada, W. S., Fernandez, M. I., & Belin, T. R. (2022) · Communications in Statistics Case Studies Data Analysis and Applications · doi:10.1080/23737484.2022.2115430 · Cited by 10
“Furthermore, complete-case analysis has the potential to induce bias in estimates of quantities of interest when individuals with incomplete observations differ systematically from those with complete data (Greenland and Finkle 1995; Horton and Kleinman 2007; Little 1988; Sterne et al. 2009).”
Background
The transition to grandparenthood: No consistent evidence for change in the Big Five personality traits and life satisfaction
Krämer, M. D., van Scheppingen, M. A., Chopik, W. J., & Richter, D. (2022) · European Journal of Personality · doi:10.1177/08902070221118443 · Cited by 10
“Therefore, we performed multiple imputations to address missingness in the covariates (Greenland & Finkle, 1995).”
Population-based cohort study of outpatients with pneumonia: rationale, design and baseline characteristics
Eurich, D. T., Majumdar, S. R., & Marrie, T. J. (2012) · BMC Infectious Diseases · doi:10.1186/1471-2334-12-135 · Cited by 10
“Potential approaches will include baseline or last value carried forward, missing indicator approaches, and single or multiple imputation [27,28].”
Method
The Influence of Perceived Discrimination on Health-Related Quality of Life in an East Coast American Indian Tribe
Johansson, P., Muller, C. J., Samos, M., & Goldberg, J. (2013) · Journal of Health Care for the Poor and Underserved · doi:10.1353/hpu.2013.0183 · Cited by 10
Prenatal diethylstilbestrol exposure and reproductive hormones in premenopausal women
Wise, L. A., Troisi, R. J., Hatch, E. E., et al. (2015) · Journal of Developmental Origins of Health and Disease · doi:10.1017/s2040174415000082 · Cited by 10
“Multiple imputation is considered the most valid method for dealing with missing data when data are missing at random conditional on measured characteristics.(22,37) Although the percentage of missingness for hormones measured late in follow-up was high, we imputed missing data using several covariates associated with…”
Method
Tratamiento estadístico de la falta de respuesta en estudios epidemiológicos transversales
Carracedo‐Martínez, E., & Figueiras, A. (2006) · Salud Pública de México · doi:10.1590/s0036-36342006000400009 · Cited by 10

Authors: Greenland, S., & Finkle, W. D.
Affiliations: University of California, Los Angeles (US)
Corresponding author: Greenland, Sander
ORCID: Greenland, Sander
Venue: American Journal of Epidemiology
Publisher: Oxford University Press (OUP)
Published: 1995-12-15
Volume: 142
Issue: 12
Pages: 1255-1264
ISSN: 0002-9262 (ISSN-L)
1476-6256
DOI: 10.1093/oxfordjournals.aje.a117592
OpenAlex: W2166561686
Cited by: 891
References: 0
Citations by year: 2026: 13 · 2025: 18 · 2024: 13 · 2023: 24 · 2022: 26 · 2021: 21 · 2020: 33 · 2019: 29 · 2018: 34 · 2017: 32 · 2016: 27 · 2015: 40 · 2014: 57 · 2013: 53 · 2012: 67
FWCI: 6.57
Type: journal article
Language: en
Open access: closed
Locations: American Journal of Epidemiology (publishedVersion)
PubMed (publishedVersion)
American Journal of Epidemiology
Funders: National Institutes of Health (n01cp095680-001)
National Cancer Institute (N01 CP 95680)
Topics: Physical Sciences › Mathematics › Statistics and Probability › Statistical Methods and Bayesian Inference
Physical Sciences › Mathematics › Statistics and Probability › Statistical Methods in Epidemiology
Physical Sciences › Mathematics › Statistics and Probability › Advanced Causal Inference Techniques
Keywords: missing covariates, missing data, logistic regression, bias, model misspecification, multiple imputation, maximum likelihood estimation, weighted estimating equations, endometrial cancer, age at menarche, case-control study
MeSH terms: Epidemiologic Methods, Female, Humans, Menarche, Research Design, Logistic Models, Odds Ratio, Case-Control Studies, Endometrial Neoplasms

About these checks

Registry, retraction, Cites and statistics results are computed, not judged, with one exception: a printed reference that no index lists is read once by a model before it counts as a citation error. Claims are extracted by a model and kept only when their quote appears verbatim in the text; each cited source is then checked, with a model’s help, for whether it says what the claim relies on. Those verdicts can be wrong — read the claim’s quote and the cited source before relying on one.

Statistics and arithmetic are recomputed, not judged.

Confidence bands are shown only for fields calibrated on our annotated test set. None is calibrated yet, so no band is shown.

What counts as a claim ›How we measure confidence ›How we check this ›

Are you an author, or do you see a problem with this page? Request a review

A critical look at methods for handling missing covariates in epidemiologic regression analyses.

Abstract

Overview

Integrity signals

Verification

Claims the paper makes

Claims other papers make about this one

Citations

Cited by

Related

Metadata

Record

About these checks