The DOI matches its Crossref record; 46 of the 46 references it cites were found in a registry or an index.
Multiple imputation of missing values was not necessary before performing a longitudinal mixed-model analysis
Abstract
Background and objectives As a result of the development of sophisticated techniques, such as multiple imputation, the interest in handling missing data in longitudinal studies has increased enormously in past years. Within the field of longitudinal data analysis, there is a current debate on whether it is necessary to use multiple imputations before performing a mixed-model analysis to analyze the longitudinal data. In the current study this necessity is evaluated.Study design and setting The results of mixed-model analyses with and without multiple imputation were compared with each other. Four data sets with missing values were created-one data set with missing completely at random, two data sets with missing at random, and one data set with missing not at random). In all data sets, the relationship between a continuous outcome variable and two different covariates were analyzed: a time-independent dichotomous covariate and a time-dependent continuous covariate.Results Although for all types of missing data, the results of the mixed-model analysis with or without multiple imputations were slightly different, they were not in favor of one of the two approaches. In addition, repeating the multiple imputations 100 times showed that the results of the mixed-model analysis with multiple imputation were quite unstable.Conclusion It is not necessary to handle missing data using multiple imputations before performing a mixed-model analysis on longitudinal data.
Overview
Integrity signals
Registry
Cites
Statistics
Verification
Claims the paper makes
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?
In the citing paper
“Multiple imputation when performing mixed models analyses can lead to unstable results [26]”
In the citing paper
“Linear mixed modelling has been demonstrated to control for missing data, and therefore, multiple imputation was not required (Peters et al., 2012;Twisk et al., 2013).”
From Exercise combined with Acceptance and Commitment Therapy for chronic pain: One-year follow-up from a randomized controlled trial (2024) · cites this paper as Twisk et al., 2013
In the citing paper
“Twisk et al 71 provided evidence that multiple imputation for missing data is not necessary before computing longitudinal mixed models.”
In the citing paper
“However, completely missing measurements will not be imputed as this is not necessary for multilevel analysis (63).”
From Timing of imagery rescripting during schema therapy for borderline personality disorder: the LUCY trial (2023) · cites this paper as [63]
In the citing paper
“Therefore, patterns of missingness were not found to be MNAR and the decision to continue the analysis with the available data was made [84,85].”
From Negative and positive affect regulation in a transdiagnostic internet-based protocol for emotional disorders: Randomized controlled trial (2021) · cites this paper as [84]
In the citing paper
“Mixed model analysis handles missing data; no imputation method was applied since there are reports that imputations before a mixed model analysis is not necessary and simulations indicated that imputations before mixed models could result in unstable results [30].”
From Is sleep disturbance in patients with chronic pain affected by physical exercise or ACT-based stress management? - A randomized controlled study (2018) · cites this paper as [30]
In the citing paper
“We also performed all analyses on the data for 100 imputations. Because there is a current debate whether it is necessary to perform multiple imputations in combination with mixed-model analyses in longitudinal studies [65], we also performed the LMM analyses without multiple imputations.”
From Long-Term Results of a Web-Based Guided Self-Help Intervention for Employees With Depressive Symptoms: Randomized Controlled Trial (2014) · cites this paper as [65]
In the citing paper
“Because of the current debate on the necessity of multiple imputations in combination with mixed-model analysis in longitudinal studies [65], we also performed mixed-model analysis without multiple imputations. The results were comparable, indicating that data were robust and multiple imputations may not have been needed.”
From Long-Term Results of a Web-Based Guided Self-Help Intervention for Employees With Depressive Symptoms: Randomized Controlled Trial (2014) · cites this paper as [65]
Citations
Cites
Multiple Imputation for Nonresponse in Surveys
Rubin, D. B. (1987) · Wiley Series in Probability and Statistics · doi:10.1002/9780470316696 · Cited by 20,701
Statistical Analysis With Missing Data
Lahiff, M., Little, R. J. A., & Rubin, D. B. (1989) · Journal of the American Statistical Association · doi:10.2307/2289883 · Cited by 17,385
(2010) · Cited by 7,034
Missing data analysis: making it work in the real world
Graham, J. W. (2009) · Annual Review of Psychology · doi:10.1146/annurev.psych.58.110405.085530 · Cited by 6,163
Analysis of Incomplete Multivariate Data
Schafer, J. L. (1997) · doi:10.1201/9781439821862 · Cited by 6,012
Analysis of Incomplete Multivariate Data
Booth, D. E., & Schafer, J. L. (2000) · Technometrics · doi:10.2307/1271466 · Cited by 5,627
Schafer, J. L. (1999) · Statistical Methods in Medical Research · doi:10.1177/096228029900800102 · Cited by 3,949
Statistical Analysis With Missing Data
Lazar, N. A. (2003) · Technometrics · doi:10.1198/tech.2003.s167 · Cited by 2,477
Multiple Imputation of Missing Values
Royston, P. (2004) · The Stata Journal: Promoting communications on statistics and Stata · doi:10.1177/1536867x0400400301 · Cited by 2,340 · cited 1×
“[28,29] discuss the impact of the number of imputations on the precision of the estimates and suggest ways of determining the required number of imputations by evaluating the sampling error of the multiple-imputation estimates.”
BackgroundMultiple Imputation for Nonresponse in Surveys.
Lynn, P. J., & Rubin, D. B. (1988) · Journal of the Royal Statistical Society Series D (The Statistician) · doi:10.2307/2348774 · Cited by 2,330
Cited by
Sorted by citations
Items 301–400 of 568
Elhakeem, A., Hughes, R. A., Tilling, K. M., Cousminer, D. L., Jackowski, S. A., Cole, T. J., Kwong, A. S. F., Li, Z., Grant, S. F. A., Baxter-Jones, A. D. G., et al. (2021) · medRxiv · doi:10.1101/2021.05.26.21257519 · Cited by 7
Gathier, A. W., Verhoeven, J. E., van Oppen, P., Penninx, B. W. J. H., Merkx, M. J. M., Dingemanse, P., Stehouwer, K. M. K. S., van den Bulck, C. M. M., & Vinkers, C. H. (2023) · BMC Psychiatry · doi:10.1186/s12888-023-04518-0 · Cited by 8
McQueen, A., Nordheim, D. V., Caburnay, C. A., Li, L., Herrick, C., Grimes, L. M., Broussard, D. J., Smith, R. E., Lawson, D., Yan, Y., et al. (2024) · International Journal of Environmental Research and Public Health · doi:10.3390/ijerph21070936 · Cited by 8
“Research suggests that mixed models are effective at addressing missingness in repeated measures data, and this approach is recommended over multiple imputation [40].”
BackgroundMacêna, M. D. L., Pereira, M. R., Silva, D. R., Júnior, A. E. D. S., de Oliveira, A. D. S., dos Santos, J. V. L., Paula, D. T. D. C., Galdino-Silva, M. B., Almeida, K. M. M., Ferro, D. C., et al. (2025) · Nutrition Metabolism and Cardiovascular Diseases · doi:10.1016/j.numecd.2025.104426 · Cited by 5
STeroids Against Radiculopathy (STAR) trial: a statistical analysis plan
ter Meulen, B. C., van Dongen, J. M., van der Vegt, M. H., Weinstein, H. C., & Ostelo, R. (2021) · Trials · doi:10.1186/s13063-020-05018-2 · Cited by 6
van Vugt, V. A., Ngo, H., van der Wouden, J. C., Twisk, J. W. R., van der Horst, H. E., & Maarsingh, O. R. (2023) · British Journal of General Practice · doi:10.3399/bjgp.2022.0468 · Cited by 6
“Level of education Low 33 (34) 37 (36) 36 (30) 106 (33) Middle 25 (26) 31 (30) 30 (25) 86 (27) High 40 (41) 36 (35) 54 (45) 130 (40)”
Self-citationImpact of Individual and Job Characteristics on Nurses’ Scope of Practice in Spanish Hospital Units
Saralegui‐Gainza, A., Escalada‐Hernández, P., García‐Vivar, C., Martín‐Rodríguez, L. S., & Soto‐Ruiz, N. (2024) · Journal of Nursing Management · doi:10.1155/2024/4796716 · Cited by 6
“Furthermore, the literature suggests that the benefts of using imputation in cases of mixed-model analysis are not necessary [46].”
Hanff, A.-M. R., Krüger, R., McCrum, C. L., Ley, C., NCER-PD, Acharya, G., Aguayo, G., Alexandre, M., Ali, M., Ammerlann, W., et al. (2024) · BMC Medical Research Methodology · doi:10.1186/s12874-024-02301-7 · Cited by 6
Cody, R., Beck, J., Brand, S., Donath, L., Faude, O., Hatzinger, M., Imboden, C., Kreppke, J.-N., Lang, U. E., Ludyga, S., et al. (2023) · Frontiers in Psychiatry · doi:10.3389/fpsyt.2022.1045158 · Cited by 7
“Linear mixed models are known to be robust with regard to missing values (72).”
BackgroundPsychosocial factors of physical activity among people with disabilities: Prospective cohort study.
Brandenbarg, P., Krops, L. A., Seves, B. L., Hoekstra, T., Hettinga, F. J., Twisk, J. W. R., van der Woude, L. H. V., Dekker, R., & Hoekstra, F. (2023) · Rehabilitation Psychology · doi:10.1037/rep0000488 · Cited by 10
Self-citationDeguire, F., López‐Arango, G., Knoth, I. S., Côté, V., Agbogba, K., & Lippé, S. (2023) · Scientific Reports · doi:10.1038/s41598-023-34669-9 · Cited by 6
Karangwa, I., Kotze, D., & Blignaut, R. J. (2016) · Brazilian Journal of Probability and Statistics · doi:10.1214/15-bjps292 · Cited by 6
Elbers, S., Pool, J. J. M., Wittink, H. M., Köke, A. J. A., Scheffer, E., & Smeets, R. (2020) · JMIR Research Protocols · doi:10.2196/18632 · Cited by 6
“Following recommendations of Twisk and colleagues [69], we will perform the multilevel analysis on incomplete data, rather than using multiple imputation procedures.”
MethodDeterminants of Health Status After Stroke: A Cohort Study with Repeated Measurements
Larsen, L. P., Johnsen, S. P., Andersen, G., & Hjøllund, N. H. (2020) · Clinical Epidemiology · doi:10.2147/clep.s270249 · Cited by 8
Schnatschmidt, M., Lollies, F., & Schlarb, A. A. (2022) · BMC Pediatrics · doi:10.1186/s12887-022-03631-5 · Cited by 6
Mixed Model Analysis in Longitudinal Studies
Twisk, J. W. R. (2019) · Cambridge University Press eBooks · doi:10.1017/9781108635660.010 · Cited by 6
Self-citationMassage Impact on Pain in Opioid-dependent Patients in Substance Use Treatment
Wiest, P., Asphaug, M., Carr, K. E., Gowen, L., & Hartnett, M. (2015) · International Journal of Therapeutic Massage & Bodywork: Research, Education, & Practice · doi:10.3822/ijtmb.v8i1.257 · Cited by 6
Kastner, C. T. (2024) · Frontiers in Psychology · doi:10.3389/fpsyg.2023.1324329 · Cited by 5
“Linear mixed-effects models can estimate model coefficients without list-wise exclusion of cases due to missing data, such that performing an intention-to-treat analysis in accordance with CONSORT guidelines (Schulz et al., 2010) does not require multiple imputation beforehand (Twisk et al., 2013).”
MethodNorman, E. R. V., & Forcht, E. R. (2024) · Journal of School Psychology · doi:10.1016/j.jsp.2024.101319 · Cited by 5
Tang, V. F. Y., Jiang, D., Kwok, J. Y. Y., Yeung, D. Y.-L., Choi, N. G., Warner, L. M., Ho, R. T. H., & Chou, K. (2026) · JAMA Network Open · doi:10.1001/jamanetworkopen.2025.57170 · Cited by 6
Cheng, N., Lin, A. P., Bowden, S. C., Gao, C. X., Yung, A. R., Nelson, B., Thompson, A. D., Yuen, H. P., Brewer, W. J., Cagliarini, D., et al. (2022) · Schizophrenia Research · doi:10.1016/j.schres.2022.08.006 · Cited by 5
Heß, J., Karageorgos, P., Müller, B., Riedmann, A., Schäper, P., Lugrin, B., & Richter, T. (2024) · Journal of Computer Assisted Learning · doi:10.1111/jcal.13021 · Cited by 7
Robinson, E. G., Gyllensten, H., Granås, A. G., Halvorsen, K. H., & Garcia, B. H. (2024) · Quality of Life Research · doi:10.1007/s11136-024-03689-x · Cited by 5
“However, a previous study indicated that mixed model regression without previous multiple imputation could handle this level of missingness [23].”
Brunborg, G. S., Bang, L., Skogen, J. C., & Andreas, J. B. (2024) · Child and Adolescent Psychiatry and Mental Health · doi:10.1186/s13034-024-00824-x · Cited by 5
“All the available data were used for estimation, and missing outcome values were not imputed [59].”
BackgroundMethodWildekamp, M., Krops, L. A., Seves, B. L., Hettinga, F. J., Houdijk, H., Dekker, R., & Hoekstra, F. (2024) · European Journal of Adapted Physical Activity · doi:10.5507/euj.2024.008 · Cited by 5
Jensen, C., Sommersten, C. H., Laupsa‐Borge, J., Storås, I., Valeur, J., Mellgren, G., Dierkes, J., Dankel, S. N., & Lied, G. A. (2024) · Clinical Nutrition · doi:10.1016/j.clnu.2024.11.041 · Cited by 5
Kastner, C. T. (2025) · Mindfulness · doi:10.1007/s12671-024-02491-7 · Cited by 5
Analysis of RCT Data with More Than One Follow-Up Measurement
Twisk, J. W. R. (2021) · Springer eBooks · doi:10.1007/978-3-030-81865-4_3 · Cited by 5
Self-citationTipsmark, L. S., Obel, B., Andersson, T., & Søgaard, R. (2021) · BMC Emergency Medicine · doi:10.1186/s12873-021-00538-9 · Cited by 5
Laupsa‐Borge, J., Grytten, E., Bohov, P., Bjørndal, B., Strand, E., Skorve, J., Nordrehaug, J. E., Berge, R. K., Rostrup, E., Mellgren, G., et al. (2023) · Frontiers in Nutrition · doi:10.3389/fnut.2023.1020678 · Cited by 5
“Therefore, we did not pre-specify another strategy for dealing with potential intermittent missing repeated outcome measurements or missing data resulting from dropouts, such as multiple imputation before mixed modeling, as this has shown to add no obvious gain compared to a standard mixed model approach without imput…”
MethodLi, B., Blijd-Hoogewys, E. M. A., Stockmann, L., & Rieffe, C. J. (2023) · Development and Psychopathology · doi:10.1017/s0954579423000913 · Cited by 7
Kwok, J. Y. Y., Cheung, D. S. K., Zarit, S. H., Cheung, S. L. K., Lau, B. H. P., LOU, V. W. Q., Cheng, S., Gallagher-Thompson, D. E., & Chou, K. (2023) · Trials · doi:10.1186/s13063-023-07801-3 · Cited by 7
Beck, C. D., Peynenburg, V. A., Patterson, T., Titov, N., Dear, B. F., & Hadjistavropoulos, H. D. (2024) · Internet Interventions · doi:10.1016/j.invent.2024.100709 · Cited by 5
“Although mixed-model procedure with maximum likelihood method can handle missing data (Shin et al., 2017; Twisk et al., 2013), following the intention-to-treat principle, the missing data on primary and secondary outcome measures were imputed using the multiple imputation method, generating 30…”
MethodSanabria‐Mazo, J. P., D’Amico, F., Cardeñosa, E., Ferrer, M., Edo, S., Borràs, X., McCracken, L. M., Feliu‐Soler, A., Sanz, A., & Luciano, J. V. (2024) · Journal of Pain · doi:10.1016/j.jpain.2024.01.337 · Cited by 8
Groot, L., Schers, H., Burgers, J. S., Smalbrugge, M., Uijen, A. A., Hoogland, J., van der Horst, H. E., & Maarsingh, O. R. (2024) · BMJ Open · doi:10.1136/bmjopen-2023-078169 · Cited by 5
Frankhouser, D. E., DeWees, T., Snodgrass, I. F., Cole, R. M., Steck, S. A., Thomas, D., Kalu, C. M. K., Belury, M. A., Clinton, S. K., Newman, J. W. I., et al. (2024) · American Journal of Clinical Nutrition · doi:10.1016/j.ajcnut.2025.04.021 · Cited by 5
Methods to Handle Incomplete Data
Johny, V., Philip, M., & Augustine, S. (2020) · MAMC Journal of Medical Sciences · doi:10.4103/mamcjms.mamcjms_54_20 · Cited by 5
Holmberg, S. T., Fredheim, O. M. S., Salvesen, Ø. O., Gulati, A. M., Solberg, T. K., Gulati, S., Skurtveit, S. O., & Nygaard, O. P. (2022) · Brain and Spine · doi:10.1016/j.bas.2022.101132 · Cited by 5
Hosch, A., Swanson, B., Harris, J. L., Oleson, J., Hazeltine, E., & Petersen, I. T. (2024) · Research on Child and Adolescent Psychopathology · doi:10.1007/s10802-023-01162-w · Cited by 5
Visser, T. S., Brul, S. L., Deng, J. F., Bonsel, J. M., van Es, E. M., Eygendaal, D. E., & de Vos, R. (2024) · British Journal of Sports Medicine · doi:10.1136/bjsports-2023-107633 · Cited by 5
Warner, F. M., Tong, B., Jutzeler, C. R., Cragg, J. J., Scheuren, P. S., & Kramer, J. L. K. (2017) · Neurology · doi:10.1212/wnl.0000000000004241 · Cited by 5
“This is a common method of accounting for missing data, with varying outcomes.(5,6) Presented in conjunction with the complete case sensitivity analysis, the imputed data allow for a transparent interpretation of the results.”
Casey, M. B., Takemasa, S., O'Reilly, T., Leamy, M., Mc Kearney, E., Buckley, M., Smart, K. M., Segurado, R., Lowry, D., Flanagan, D., et al. (2024) · European Journal of Pain · doi:10.1002/ejp.2229 · Cited by 4
“Linear mixed modelling has been demonstrated to con-trol for missing data, and therefore, multiple imputation was not required (Peters et al., 2012; Twisk et al., 2013).”
MethodQuadriceps Morphology 6‐ to 60‐Months After Anterior Cruciate Ligament Tear
Losciale, J. M., Lin, B., Théret, M., Schweitzer, A. M., Fliss, M. D., Scarr, T., Yan, X., Prasad, N., Hunt, M. A., Koehle, M. S., et al. (2025) · Scandinavian Journal of Medicine and Science in Sports · doi:10.1111/sms.70109 · Cited by 4
Hilbert, A., Strömer, A., Staerk, C., Schreglmann, B., Mansfeld, T., Sander, J., Seyfried, F., Kaiser, S., Stroh, C., Dietrich, A., et al. (2025) · International Journal of Eating Disorders · doi:10.1002/eat.24527 · Cited by 4
Li, Z., Li, B., Tsou, Y., Wang, L., Liang, W., & Rieffe, C. J. (2024) · Development and Psychopathology · doi:10.1017/s0954579424001408 · Cited by 4
Peynenburg, V. A., Sapkota, R. P., Titov, N., Dear, B. F., & Hadjistavropoulos, H. D. (2024) · Internet Interventions · doi:10.1016/j.invent.2024.100796 · Cited by 4
“A longitudinal mixed-model analysis with maximum likelihood method of estimation can handle missing data, imputation prior to analysis was not necessary (Twisk et al., 2013).”
Vabø, K. B., Aadland, E., Haugland, E. S., Johannessen, K., Tjomsland, H. E., Glosvik, Ø., Ommundsen, Y., Tomporowski, P. D., Pesce, C., Howard, S. J., et al. (2025) · Journal of Sports Sciences · doi:10.1080/02640414.2025.2460886 · Cited by 4
“We did not use imputation for missing data since the mixed approach provides unbiased estimates in the case of missing data (Enders, 2011; Twisk et al., 2013) or when its assumptions are violated (Schielzeth et al., 2020).”
Sapkota, R. P., Gretton, A., Wilhelms, A., Bisby, M. A., Dear, B. F., Titov, N., & Hadjistavropoulos, H. D. (2025) · Internet Interventions · doi:10.1016/j.invent.2025.100852 · Cited by 4
“A longitudinal mixed-model analysis using the maximum likelihood estimation method, which can accommodate MAR data, was conducted using unimputed data (Twisk et al., 2013).”
Fatt, S. J., Hay, P., George, E. S., JEACOCKE, N. A., Rogers, K. D., & Mitchison, D. (2025) · Sports Medicine - Open · doi:10.1186/s40798-025-00927-5 · Cited by 4
“The imputed datasets were used for the 6-month analyses but not the 12-month analyses, as multiple imputation may not be appropriate for multilevel mixed-effects analyses [35, 36].”
Sapkota, R. P., Peynenburg, V. A., Dear, B. F., Titov, N., & Hadjistavropoulos, H. D. (2025) · SSM - Mental Health · doi:10.1016/j.ssmmh.2025.100561 · Cited by 4
Lipoic Acid for Treatment of Progressive Multiple Sclerosis
Spain, R., Soldán, M. M. P., Freedman, M. S., Repovic, P., Solomon, A. J., Rinker, J. R., Wallin, M. T., Haselkorn, J. K., Stüve, O., Gross, R. A., et al. (2025) · Neurology · doi:10.1212/wnl.0000000000214454 · Cited by 5
Ingalls, A., Rebman, P., Martin, L., Kushman, E., Leonard, A., Cisler, A., Gschwind, I., Brayak, A., Amsler, A. M., & Haroz, E. E. (2022) · BMC Pregnancy and Childbirth · doi:10.1186/s12884-022-05057-4 · Cited by 5
“We did not impute missing data, as longitudinal mixed-effects models are robust to missing data and multiple imputation may result in unstable results [42].”
MethodBody Image and Weight Loss Outcome After Bariatric Metabolic Surgery: a Mixed Model Analysis
Makarawung, D. J. S., Dijkhorst, P. J., de Vries, C. E. E., Monpellier, V. M., Wiezer, M. J., van Veen, R. N., Geenen, R., & van der Molen, A. B. M. (2023) · Obesity Surgery · doi:10.1007/s11695-023-06690-4 · Cited by 4
“Additionally, all available data points are used without the need for imputation methods [47].”
MethodRobinson, E. G., Gyllensten, H., Johansen, J. S., Havnes, K., Granås, A. G., Bergmo, T. S., Småbrekke, L., Garcia, B. H., & Halvorsen, K. H. (2023) · Drugs & Aging · doi:10.1007/s40266-023-01077-7 · Cited by 4
Lauriola, M., Cicco, G. D., & Savadori, L. (2024) · Judgment and Decision Making · doi:10.1017/jdm.2024.2 · Cited by 4
“The mixed model itself is robust to missing data under this assumption, using all the available data to estimate model parameters, which is more efficient than deleting any cases with missing data (listwise deletion) or imputing missing data points (Twisk et al., 2013).”
Methodvan Doornik, R., der Oord, S. V., Luijckx, J., Groenman, A. P., Leijten, P. H. O., Luman, M., Hoekstra, P. J., van den Hoofdakker, B. J., & Dekkers, T. J. (2024) · BMC Psychiatry · doi:10.1186/s12888-024-05649-8 · Cited by 8
“To examine the effects of the intervention on the primary and secondary outcomes, we will conduct multilevel analyses (mixed modeling), taking missing data into account [75].”
MethodFiret, L., Teunissen, T. A. M., Kool, R. B., Akkermans, R. P., Lagro‐Janssen, A. L. M., van der Vaart, C. H., & Assendelft, W. J. (2024) · BMC Family Practice · doi:10.1186/s12875-024-02325-4 · Cited by 4
Kim-Knauss, Y., Degen, N. M., & Lang, F. R. (2024) · Gerontology · doi:10.1159/000539540 · Cited by 4
“…linear mixed model is in particular considered optimal for handling datasets like ours, characterized by fl uc-tuating sample sizes and non-equidistant data structures, as it can effectively address missing outcome values through likelihood solutions without additional missing value imputation [26].”
Ahmad, I., Chufal, K. S., Miller, A., Bajpai, R. C., Preetha, U., Sokhal, B. S., Bhatia, K., Pati, S., & Gairola, M. (2024) · Radiation Oncology · doi:10.1186/s13014-024-02512-8 · Cited by 4
Influence of street outreach work on the lives of marginalised people
Rauwerdink-Nijland, E., van den Dries, L., Metz, J. W., Verhoeff, A. P., & Wolf, J. (2024) · European Journal of Social Work · doi:10.1080/13691457.2024.2416679 · Cited by 4
“Secondly, the within-subject interpretation implies a change within one subject of 1-unit intermediate variable is associated with a change of β units in the primary outcome measure (Twisk, 2013).”
InfluentialMiller, M. B., Curtis, A. F., Chan, W. S., Deroche, C. B., & McCrae, C. S. (2020) · Journal of Clinical Sleep Medicine · doi:10.5664/jcsm.9002 · Cited by 5
“No imputation procedures were used because imputation does not improve the accuracy of multilevel modeling estimates.(24) Multilevel-model fixed effect and standard error estimates are also robust to violations of normality.”
Anraad, C., van Empelen, P., Ruiter, R. A. C., & van Keulen, H. M. (2023) · Vaccine · doi:10.1016/j.vaccine.2023.10.068 · Cited by 4
Multivariate Mixed Model Analysis
Twisk, J. W. R. (2019) · Cambridge University Press eBooks · doi:10.1017/9781108635660.011 · Cited by 4
Self-citationClinical outcomes after surgery for cervical radiculopathy performed in public and private hospitals
Danielsen, E., Gulati, S., Salvesen, Ø. O., Ingebrigtsen, T., Nygaard, O. P., & Solberg, T. K. (2023) · The Bone & Joint Journal · doi:10.1302/0301-620x.105b1.bjj-2022-0591.r1 · Cited by 4
Pritlove, C., Capone, G., Ramasamy, M., Avery, L. M., Fierini, D., Ferguson, S. E., Han, K. C.-Z., & Jones, J. M. (2024) · Nutrients · doi:10.3390/nu16234227 · Cited by 4
“Mixed-effects modeling of longitudinal data allows all available data to be used and does not require imputation [63].”
Brabander, E. Y. D., Leibold, N. K., van Amelsvoort, T., van Westrhenen, R., Alizadeh, B. Z., Cahn, W., de Haan, L., van der Pluijm, M., Simons, C. J. P., van Os, J., et al. (2025) · Progress in Neuro-Psychopharmacology and Biological Psychiatry · doi:10.1016/j.pnpbp.2025.111363 · Cited by 3
“Furthermore, the Bayesian mixed models applied in the current study are able to handle missing data and the potential for systematic miss-ingness, accounting for potential impact of bias without imputation (Feng, 2021; Gadbury et al., 2003; Twisk et al., 2013).”
Blankers, M., van Beek, R. J. J., Spronk, D. B., den Hollander, W., Andree, R., Freeman, T. P., Grabski, M., Curran, H. V., Waldron, J., & van Laar, M. (2025) · Drug and Alcohol Dependence · doi:10.1016/j.drugalcdep.2025.112881 · Cited by 3
Sensitivity Analyses for Missing in Repeatedly Measured Outcome Data
Troendle, J. F., Sur, A., Leifer, E., & Powell‐Wiley, T. M. (2025) · Statistics in Medicine · doi:10.1002/sim.70282 · Cited by 3
Hadjistavropoulos, H. D., Dear, B. F., Titov, N., & Sapkota, R. P. (2025) · Internet Interventions · doi:10.1016/j.invent.2025.100894 · Cited by 3
“Missing data in longitudinal mixed-model analyses were handled using maximum likelihood (ML) estimation, which accommodates MAR without requiring imputation (Twisk et al., 2013).”
Health-related quality of life outcomes and recovery trajectories following burn injury
Tracy, L. M., Capell, E. L., Cleland, H. J., Edgar, D. W., Singer, Y. M., Teague, W. J., & Gabbe, P. B. (2026) · Burns · doi:10.1016/j.burns.2026.107891 · Cited by 3
Wei, X., Yu, H., Dai, W., Huang, L., Liu, Y., Cheng, L., Yang, D. C., Zhang, K., Liao, J. X., Wang, Y., et al. (2026) · Annals of Surgical Oncology · doi:10.1245/s10434-025-18834-0 · Cited by 3
Progression of GBA1 severe and risk variants: a longitudinal mixed model analysis
Hanff, A.-M. R., McCrum, C. L., Rauschenberger, A., Pachchek, S., Aguayo, G., Pauly, C., Jónsdóttir, S. R., Tsurkalenko, O., Pauly, L., Landoulsi, Z., et al. (2026) · Frontiers in Aging Neuroscience · doi:10.3389/fnagi.2026.1771789 · Cited by 4
“As individuals with PD were enrolled at different time points (Twisk, 2013) due to the dynamic cohort study design, the disease progression could be distinguished from cohort or period effects.”
Lowe, D. M., Zhang, S., Ellis, S., Buggiotti, L., Miller, C., Harris, V., Shields, A. M., Richter, A. G., Ashmore, E., Yu, L.-M., et al. (2026) · The Lancet Microbe · doi:10.1016/j.lanmic.2026.101360 · Cited by 3
Stilling, M. N., Lindberg, L., Tuck, L. D., Sørensen, O. H., Holtermann, A., & Rasmussen, C. D. N. (2025) · BMC Public Health · doi:10.1186/s12889-025-24463-9 · Cited by 3
“All analyses will be performed according to the intention-to-treat principle, including all eligible participants without imputations because mixed models inherently account for missing values [39].”
de Zwart, A. H., van der Leeden, M., Roorda, L. D., van der Esch, M., Twisk, J. W. R., Lems, W. F., & Dekker, J. (2020) · Osteoarthritis and Cartilage Open · doi:10.1016/j.ocarto.2020.100038 · Cited by 3
Self-citationExplaining Differences between Groups
Twisk, J. W. R. (2019) · Cambridge University Press eBooks · doi:10.1017/9781108635660.007 · Cited by 3
Self-citationFrailty measures in immuno-metabolic subtypes of late-life depression; A two-year prospective study
Kokkeler, K. J. E., Voshaar, R. C. O., Rhebergen, D., van den Brink, R. H. S., Spijker, J., & Marijnissen, R. M. (2021) · Archives of Gerontology and Geriatrics · doi:10.1016/j.archger.2021.104603 · Cited by 3
Yorozuya, K., Kubo, Y., Asaoka, Y., Hayashi, H., Fujita, T., & Hanaoka, H. (2024) · Psychogeriatrics · doi:10.1111/psyg.13186 · Cited by 3
“31,41 The explanatory variables were the baseline score on the indicator (the objective variable) and the presence or absence of the MNPI (MNPI = 1 vs. control = 0).”
Analysis of Longitudinal Outcome Data with Missing Values in Total Knee Arthroplasty
Kang, Y. G., Lee, J. T., Kang, J. Y., Kim, G. H., & Kim, T. K. (2015) · The Journal of Arthroplasty · doi:10.1016/j.arth.2015.06.067 · Cited by 3
“A previous study found that multiple imputation was not necessary when analyzing data with mixed model [15].”
BackgroundMethodNagler, B., Gleiss, A., Füreder, L., Buchtele, N., Hermann, A., Bojic, A., Schellongowski, P., Staudinger, T., & Robak, O. (2022) · ASAIO Journal · doi:10.1097/mat.0000000000001675 · Cited by 3
Scott, I. C., Bajpai, R. C., Hider, S., Helliwell, T., Mallen, C. D., & Müller, S. (2024) · Rheumatology Advances in Practice · doi:10.1093/rap/rkae081 · Cited by 3
“As multiple imputation is recommended in longitudinal data analysis with missing covariates [22], sensitivity analyses imputed missing baseline covariates and missing outcomes at any follow-up time using multiple imputation techniques with 20 imputations (based on the missingness pattern) following…”
BackgroundMethodGenerative Adversarial Networks Imputation for High Rate Missing Values
Wang, H., Chen, Y., Shen, B., Wu, D., & Ban, X. (2018) · IEEE International Conference on Internet of Things (iThings) and IEEE Green Computing and Communications (GreenCom) and IEEE Cyber, Physical and Social Computing (CPSCom) and IEEE Smart Data (SmartData) · doi:10.1109/cybermatics_2018.2018.00121 · Cited by 3
Danielsen, E., Ingebrigtsen, T., Gulati, S., Salvesen, Ø. O., Johansen, T. O., Nygaard, O. P., & Solberg, T. K. (2024) · Neurosurgery · doi:10.1227/neu.0000000000002852 · Cited by 2
Ayres, J., Chen, Y., Forbush, K. T., Bottera, A. R., & Christensen, K. A. (2025) · Appetite · doi:10.1016/j.appet.2025.108052 · Cited by 2
Belimbegovski, W., Schaap, L. S., Linn, A. J., Medlock, S. K., van Weert, J. C. M., van der Velde, N., & van Schoor, N. M. (2025) · Archives of Gerontology and Geriatrics · doi:10.1016/j.archger.2025.105936 · Cited by 2
Liu, Y., Sommer, I. E., Begemann, M., Gangadin, S. S., Perry, B. I., Scheurink, T. A. W., van Beveren, N. J. M., Sommer, I. E., de Haan, L., Veling, W., et al. (2025) · Schizophrenia Bulletin · doi:10.1093/schbul/sbaf116 · Cited by 2
“59 Further more, when we compared the baseline characteristics of participants with and without missing weight and movement disorder measurements, no differences were found in age, sex, smoking status, or olanzapine equivalent dose between the 2 groups.”
Arts, J., Altenburg, T. M., Lettink, A., Verhoeff, A. P., Gubbels, J. S., & Chinapaw, M. J. M. (2025) · Journal of Activity Sedentary and Sleep Behaviors · doi:10.1186/s44167-025-00085-9 · Cited by 2
“Linear mixed models use maximum likelihood estimation to handle missing data, eliminating the need for data imputation [40, 41].”
Liu, R. Y., & Liu, E. (2025) · Nutrients · doi:10.3390/nu17233660 · Cited by 2
“No specific imputation procedures were applied, as linear mixed models can appropriately handle missing data under the MAR assumption [24,25].”
Murtazin, K., Shvets, O., Karu, K.-E., Puusep, V., Vendelin, J., Meeter, M., & Piho, G. (2025) · Frontiers in Education · doi:10.3389/feduc.2025.1621644 · Cited by 2
Houtvast, D. C. J., Hendriks, F. K., Weijzen, M. E. G., Zorenc, A. H. G., Smeets, J. S. J., Emans, P. J., Wodzig, W. K. W. H., Larsen, M. S., Mikkelsen, U. R., Schotanus, M. G. M., et al. (2026) · American Journal of Clinical Nutrition · doi:10.1016/j.ajcnut.2026.101349 · Cited by 2
“Missing data were not imputed, because linear mixed models use maximum likelihood estimations to use all the available data under the assumption that data are missing at random [31,32].”
Vangen-Lønne, V., Madsbu, M. A., Salvesen, Ø. O., Nygaard, O. P., Solberg, T. K., & Gulati, S. (2022) · Brain and Spine · doi:10.1016/j.bas.2022.100894 · Cited by 2
“Our study is strengthened by prospective data collection, high external validity, and widely applied and validated outcome measures (Fairbank et al., 1980; Grotle et al., 2003; Solberg et al., 2005; Ostelo et al., 1976; Twisk et al., 2013) Although this is a single center study, it is the largest to date with prospect…”
MethodKleijburg, A., Wijnen, B. F. M., Evers, S., Kroon, H., & Lokkerbol, J. (2022) · BMC Psychiatry · doi:10.1186/s12888-022-04346-8 · Cited by 2
“Because longitudinal mixed-modelling performs internal HoNOS ∼ condition + time + condition ∶ time + site + (1|id) imputation, no other procedures for missing data are required [24].”
MethodStapleton, J. R., Ard, J. D., Beavers, D. P., Cogdill, L. S., Fernandez, A., Howard, M. J., Justice, J. N., Lynch, S. D., Newman, J. J., Weaver, A. A., et al. (2023) · Contemporary Clinical Trials Communications · doi:10.1016/j.conctc.2023.101181 · Cited by 2
“All models will use completers only, noting that mixed models with baseline covariate adjustments are fairly robust to missing data; [57,58] however, sensitivity analyses will be conducted to ensure results are not biased due to differential missing data.”
MethodBehavioral and Psychological Factors in Buffering Diabetes-related Disability Development
Tsai, Y., Chuang, L.-L., Lee, Y., & Chiu, C. (2023) · Annals of Behavioral Medicine · doi:10.1093/abm/kaad038 · Cited by 2
Eriksen, B., Fasiolo, M., Mathisen, U., Jenssen, T., Stefansson, V., & Melsom, T. (2023) · Scientific Reports · doi:10.1038/s41598-023-41181-7 · Cited by 1
“All study participants were included in the GAMM analyses regardless of whether they were examined at follow-up because mixed models allow for missing observations at one or more points in time 27,28 .”
Hung, S., Tierney, C., Klassen, T. D., Schneeberg, A., Bayley, M. T., Dukelow, S. P., Hill, M. D., Krassioukov, A. V., Pooyania, S., Poulin, M. J., et al. (2023) · Frontiers in Neurology · doi:10.3389/fneur.2023.1245881 · Cited by 3
“The package lme4: Linear Mixed-Effects Models used “Eigen” and S4 (R software), which utilizes all available data and has been demonstrated to be valid in the presence of data missing at random (28).”
Schlegl, S., Maier, J., Dieffenbacher, A., & Voderholzer, U. (2023) · European Eating Disorders Review · doi:10.1002/erv.3047 · Cited by 2
Badia-Aguarón, T., Royuela-Colomer, E., Pera-Guardiola, V., Balasch, P. V., Cebolla, A., Luciano, J. V., Soler, J., Feliu‐Soler, A., & Huguet, A. (2024) · Frontiers in Psychology · doi:10.3389/fpsyg.2024.1291198 · Cited by 2
“No imputation of missing data will be performed since it has been reported that multiple imputation is unnecessary for certain types of missing data (missing completely at random, missing at random, and missing not at random) when conducting longitudinal mixed model analysis (Twisk et al., 2013).”
MethodChrist, C., van Schaik, A., Kikkert, M., de Waal, M. M., Dozeman, E., Hulstijn, H. L., Koomen, L. E. M., Krah, I. M., Schut, D. M., Beekman, A. T. F., et al. (2024) · Journal of Affective Disorders · doi:10.1016/j.jad.2024.03.028 · Cited by 2
Zink, J., Weniger, M., Porst, P. T., Siegmund, C. B., McDonald, M., Rückert, F., Roessner, V., Knappe, S., & Beesdo‐Baum, K. (2024) · Research on Child and Adolescent Psychopathology · doi:10.1007/s10802-024-01221-w · Cited by 2
“Mixed model analyses have been shown to produce robust results even without prior multiple imputation (MI) (Twisk et al., 2013), but baseline values are needed to determine whether initial symptom severity moderates the effectiveness of the prevention program.”
Metadata
- Authors
- Twisk, J., de Boer, M., de Vente, W., & Heymans, M.
- Affiliations
- Vrije Universiteit Amsterdam (NL)
- Corresponding author
- Twisk, Jos
- Venue
- Journal of Clinical Epidemiology
- Publisher
- Elsevier BV
- Published
- 2013-09
- Volume
- 66
- Issue
- 9
- Pages
- 1022-1028
- ISSN
- 0895-4356 (ISSN-L)
- OpenAlex
- W1974992958
- Cited by
- 568
- References
- 43
- Citations by year
- 2026: 58 · 2025: 68 · 2024: 74 · 2023: 57 · 2022: 61 · 2021: 52 · 2020: 46 · 2019: 45 · 2018: 27 · 2017: 26 · 2016: 21 · 2015: 19 · 2014: 10 · 2013: 4
- FWCI
- 15.69
- Type
- journal article
- Language
- en
- Open access
- closed
- Locations
- Journal of Clinical Epidemiology (publishedVersion)
- PubMed (publishedVersion)
- UvA-DARE (University of Amsterdam) (publishedVersion)
- Pure Amsterdam UMC (publishedVersion)
- Amsterdam UMC (VU Amsterdam) - Institutional Repository (publishedVersion)
- VU Research Portal (publishedVersion)
- Data Archiving and Networked Services (DANS) (submittedVersion)
- Topics
- Physical Sciences › Mathematics › Statistics and Probability › Statistical Methods and Bayesian Inference
- Physical Sciences › Mathematics › Statistics and Probability › Statistical Methods in Clinical Trials
- Physical Sciences › Mathematics › Statistics and Probability › Statistical Methods in Epidemiology
- Keywords
- multiple imputation, missing data, longitudinal analysis, mixed models, missing completely at random, missing at random, missing not at random, time-varying covariates
- MeSH terms
- Data Interpretation, Statistical, Humans, Longitudinal Studies, Patient Dropouts, Research Design, Models, Statistical
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
