Testing for baseline balance in clinical trials
The DOI matches its Crossref record; 38 of the 39 references it cites were found in a registry or an index.
Abstract
Once the data from a clinical trial are available for analysis it is common practice to carry out ‘tests of baseline homogeneity’ on prognostic covariates before proceeding to analyse the effects of treatment on outcome variables. It is argued that this practice is philosophically unsound, of no practical value and potentially misleading. Instead it is recommended that prognostic variables be identified in the trial‐plan and fitted in an analysis of covariance regardless of their baseline distribution (statistical significance).
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
“Between-group differences on change scores of the outcomes were tested using a General Linear Model (GLM) while adjusting for theoretically-important covariates (i.e., age, stage of cancer, education)39 and cancer treatments received, which differed significantly between arms at baseline.”
From Acceptance and commitment therapy for breast cancer survivors with fear of cancer recurrence: A 3-arm pilot randomized controlled trial. (2020) · cites this paper as [39]
Citations
Cites
Regression Models and Life-Tables
Cox, D. R. (1972) · Journal of the Royal Statistical Society Series B: Statistical Methodology · doi:10.1111/j.2517-6161.1972.tb00899.x · Cited by 39,793
Regression Models and Life-Tables
Cox, D. R. (1992) · Springer Series in Statistics · doi:10.1007/978-1-4612-4380-9_37 · Cited by 26,297
Assignment to Treatment Group on the Basis of a Covariate
Rubin, D. B. (1977) · Journal of Educational Statistics · doi:10.3102/10769986002001001 · Cited by 1,155
Cross-Over Trials in Clinical Research
(1994) · Technometrics · doi:10.1080/00401706.1994.10485798 · Cited by 763
Self-citationClinical Trials: Design, Conduct and Analysis
Meinert, C. L. (1986) · DigitalGeorgetown (Georgetown University Library) · doi:10.1093/acprof:oso/9780195035681.001.0001 · Cited by 756
GAIL, M. H., WIEAND, S., & PIANTADOSI, S. (1984) · Biometrika · doi:10.1093/biomet/71.3.431 · Cited by 698
Senn, S., Frison, L., & Pocock, S. (1994) · Statistics in Medicine · doi:10.1002/sim.4780130210 · Cited by 529
Self-citationRandomisation and baseline comparisons in clinical trials
Altman, D. G., & Doré, C. J. (1990) · The Lancet · doi:10.1016/0140-6736(90)90014-v · Cited by 508
On the Principle of Total Evidence
GOOD, I. J. (1967) · The British Journal for the Philosophy of Science · doi:10.1093/bjps/17.4.319 · Cited by 329
Comparability of Randomised Groups
Altman, D. G. (1985) · The Statistician · doi:10.2307/2987510 · Cited by 328
Research works printed but not linked: 3
- Maxwell C., 1973, Clinical Research For All
- Lindley D. V., 1985, Making Decisions
- Miller D., 1991, On the Maximization of Expected Futility
Cited by
Items 201–300 of 503
Impact of improved treatment of sexually transmitted disease on HIV infection
Rygnestad, T., Småbrekke, L., Nesje, L. B., Talsethagen, H., Rogne, P., Myhr, K., Bentzen, I., Habbema, J. D. F., de Vlas, S. J., Foulkes, M. A., et al. (1995) · The Lancet · doi:10.1016/s0140-6736(95)91830-2 · Cited by 19
The adjustment of random baseline measurements in treatment effect estimation
Chen, X. (2005) · Journal of Statistical Planning and Inference · doi:10.1016/j.jspi.2005.08.046 · Cited by 18
Gossage-Worrall, R., Holt, R. I. G., Barnard-Kelly, K. D., Carey, M., Davies, M. J., Dickens, C. M., Doherty, Y., Edwardson, C. L., French, P., Gaughran, F., et al. (2016) · Trials · doi:10.1186/s13063-016-1572-1 · Cited by 18
“Demographics and baseline characteristics will be summarised and assessed for comparability between the intervention and control arms [38, 39].”
BackgroundBautmans, I., Lefeber, N., Lievens, P., Scheerlinck, T., Vaes, P., & Beckwée, D. (2017) · The Journal of Knee Surgery · doi:10.1055/s-0037-1602134 · Cited by 23
Evaluation study design – a pluralist approach to evidence
Fives, A., Canavan, J., & Dolan, P. (2017) · European Early Childhood Education Research Journal · doi:10.1080/1350293x.2016.1266227 · Cited by 25
“Although across all possible RCTs the two groups will be equal at baseline, in any one study, precisely because allocation is random, it may lead to two unequal groups at baseline (Altman, 1985; Senn, 1994; Fives et al. et al., 2013).”
BackgroundWan, F. (2018) · Statistical Methods in Medical Research · doi:10.1177/0962280218789972 · Cited by 17
“A comparison of equations (3) to (7) shows that both the ANCOVA and the cRM estimators are virtually the same when the variance and covariance parameters are known.”
BackgroundMethodRandomization in clinical trials: can we eliminate bias?
Sverdlov, O., & Rosenberger, W. F. (2012) · Clinical Investigation · doi:10.4155/cli.12.130 · Cited by 22
Hind, D., Scott, E. J., Copeland, R. J., Breckon, J. D., Crank, H., Walters, S. J., Brazier, J. E., Nicholl, J. P., Cooper, C. L., & Goyder, E. (2010) · BMC Public Health · doi:10.1186/1471-2458-10-3 · Cited by 19
Clinical Research in Obstetrics and Gynecology: A Baedeker for Busy Clinicians
Grimes, D. A., & Schulz, K. F. (2002) · Obstetrical & Gynecological Survey · doi:10.1097/00006254-200209003-00001 · Cited by 17
“This question is gratuitous for a randomized controlled trial, since all differences in baseline characteristics are known to be due to chance (the investigator assigned participants to treatment at random) (70).”
BackgroundHamm, J. D., Dotel, J., Tamura, S., Shechter, A., Herzog, M., Brunstrom, J. M., Albu, J., Pi-Sunyer, X., Laferrère, B., & Kissileff, H. R. (2020) · Physiology & Behavior · doi:10.1016/j.physbeh.2020.113001 · Cited by 20
Continuous covariate imbalance and conditional power for clinical trial interim analyses
Ciolino, J. D., Martin, R. H., Zhao, W., Jauch, E. C., Hill, M. D., & Palesch, Y. Y. (2014) · Contemporary Clinical Trials · doi:10.1016/j.cct.2014.02.007 · Cited by 16
“Failure to do somay result in inflated type I error rates [16,15], biased treatment effect estimation [22], or loss in statistical efficiency (i.”
BackgroundStatistical testing of baseline differences in sports medicine RCTs: a systematic evaluation
Peterson, R. L., Tran, M., Koffel, J. B., & Stovitz, S. D. (2017) · BMJ Open Sport & Exercise Medicine · doi:10.1136/bmjsem-2017-000228 · Cited by 16
“However, there are sound reasons to not present statistical tests of baseline differences.(5) The aim of statistical testing is to find the probability that the baseline differences would be due to chance if the groups were the same.”
BackgroundFives, A., Russell, D. W., Kearns, N., Lyons, R., Eaton, P., Canavan, J., Devaney, C., & O'Brien, A. (2013) · Journal of MultiDisciplinary Evaluation · doi:10.56645/jmde.v9i20.371 · Cited by 20
“…been criticised by the CONSORT statement, on the grounds that it is ‘inappropriate’ when comparing two study conditions at pretest (Altman, Moher, Schulz, Egger, Davidoff, Elbourne, Gotsche, & Lang, 2001; see Senn, 1994; Assman, Pocock, Enos, & Kasten, 2000; Pocock, Assman, Enos, & Kasten, 2002).”
BackgroundInfluentialSeeking and reporting apparent research misconduct: errors and integrity
Kharasch, E. D., & Houle, T. T. (2018) · Anaesthesia · doi:10.1111/anae.14147 · Cited by 15
Errors and Integrity in Seeking and Reporting Apparent Research Misconduct
Kharasch, E. D., & Houle, T. T. (2017) · Anesthesiology · doi:10.1097/aln.0000000000001875 · Cited by 15
The Table I Fallacy: P Values in Baseline Tables of Randomized Controlled Trials
Pijls, B. G. C. W. (2022) · Journal of Bone and Joint Surgery · doi:10.2106/jbjs.21.01166 · Cited by 15
O’Connor, M., O’Reilly, G., Murphy, E., Connaughton, L., Hoctor, E., & McHugh, L. (2022) · Behaviour Research and Therapy · doi:10.1016/j.brat.2022.104120 · Cited by 19
Dawson, A. J., Kilpatrick, E. S., Coady, A., Elshewehy, A. M. M., Dakroury, Y., Ahmed, L., Atkin, S. L., & Sathyapalan, T. (2017) · BMC Endocrine Disorders · doi:10.1186/s12902-017-0194-2 · Cited by 17
“Given that these were randomised clinical trials baseline values were not compared or adjusted for [18].”
BackgroundScore confidence intervals and sample sizes for stratified comparisons of binomial proportions
Tang, Y. (2020) · Statistics in Medicine · doi:10.1002/sim.8674 · Cited by 15
Ritz, C. (2020) · European Journal of Clinical Nutrition · doi:10.1038/s41430-020-00750-z · Cited by 17
“situation where we hope for no difference [25, 26].”
BackgroundIncreasing Barrier Method Use Among Oral Contraceptive Users at Risk of STDs
Steiner, M. J., Glover, L. H., Bou-Saada, I., & Piedrahita, C. (1998) · Sexually Transmitted Diseases · doi:10.1097/00007435-199803000-00006 · Cited by 15
Plotkin, D. L., Rodas, M. A., Vigotsky, A. D., McIntosh, M. C., Breeze, E., Ubrik, R., Robitzsch, C., Agyin-Birikorang, A., Mattingly, M. L., Michel, J. M., et al. (2023) · Frontiers in Physiology · doi:10.3389/fphys.2023.1279170 · Cited by 17
“Notably, baseline and within-group inferential statistics were not calculated, as baseline signi fi cance testing is inconsequential (Senn, 1994) and within-group outcomes are not the subject of our research question (Bland and Altman, 2011).”
BackgroundStang, A., & Baethge, C. (2018) · Clinical Epidemiology · doi:10.2147/clep.s161508 · Cited by 14
Figueroa, C. A., Hernandez-Ramos, R., Boone, C. E., Pathak, L. E., Yip, V., Luo, T. C., de la Sierra, V. Q., Xu, J., Chakraborty, B., Darrow, S. M., et al. (2020) · JMIR Research Protocols · doi:10.2196/23592 · Cited by 16
“This practice could potentially be misleading because any differences would be due to chance [26-28].”
The Effect of Two Interventions to Increase Breast Cancer Screening in Rural Women
Champion, V. L., Monahan, P. O., Stump, T. E., Biederman, E. B., Vachon, E. A., Katz, M. L., Rawl, S. M., Baltic, R. D., Kettler, C. D., Zaborski, N. L., et al. (2022) · Cancers · doi:10.3390/cancers14184354 · Cited by 15
“Baseline characteristics were descriptively reported for the overall sample and separately for women in each of the three groups [34].”
BackgroundA study to assess the effect of nursing interventions at the weekend for people with stroke
Davidson, I., Hillier, V. F., Waters, K. R., Walton, T., & Booth, J. (2005) · Clinical Rehabilitation · doi:10.1191/0269215505cr841oa · Cited by 16
Exercise Selection Differentially Influences Lower Body Regional Muscle Development
Burke, R., Piñero, A., Mohan, A. E., Hermann, T. M., Sapuppo, M., Augustin, F., Coleman, M., Androulakis‐Korakakis, P., Wolf, M., Swinton, P. A., et al. (2024) · Journal of Science in Sport and Exercise · doi:10.1007/s42978-024-00299-4 · Cited by 17
“Inferences were not drawn on baseline nor within-group change, as baseline testing is inconsequential [30] and within-group outcomes were not the subject of our research question [6], although we descriptively presented within-group changes to help contextualize our findings.”
Austin, P. C. (2010) · The International Journal of Biostatistics · doi:10.2202/1557-4679.1179 · Cited by 13
Adjusting for Baseline Covariates in Net Benefit Regression: How You Adjust Matters
Isaranuwatchai, W., Markle‐Reid, M., & Hoch, J. S. (2015) · PharmacoEconomics · doi:10.1007/s40273-015-0287-6 · Cited by 14
“When analyzing person-level data, the use of regression analysis is recommended over a simple unadjusted analysis because regression analysis can estimate the treatment effect with increased precision due to the adjustment of confounders [4–8].”
BackgroundMethodBerg, S. K., Zwisler, A., Koch, M. B., Svendsen, J. H., Christensen, A. V., Pedersen, P. U., & Thygesen, L. C. (2015) · Journal of Rehabilitation Medicine · doi:10.2340/16501977-1920 · Cited by 14
“As recommended, no significant test for detecting baseline differences was performed (12, 13).”
BackgroundRandomisierte klinische Studien (RCTF)
Windeler, J., Antes, G., Behrens, J., Donner‐Banzhoff, N., & Lelgemann, M. (2008) · Zeitschrift für Evidenz Fortbildung und Qualität im Gesundheitswesen · doi:10.1016/j.zefq.2008.05.002 · Cited by 19
Austin, P. C. (2008) · Statistics in Medicine · doi:10.1002/sim.3243 · Cited by 17
When does adjusting covariate under randomization help? A comparative study on current practices
Gao, Y., Liu, Y., & Matsouaka, R. (2024) · BMC Medical Research Methodology · doi:10.1186/s12874-024-02375-3 · Cited by 19
“…this paper, it is worth considering some data-driven based methods for variable selection before adjusting for covariates (e.g., AIC and BIC [53–55], LASSO [56–58], adaptive LASSO [59], or comparing marginal correlations to the outcome [60, 61]), which can further provide practically useful tools.”
MethodMetwally, M., Chatters, R., Pye, C., Dimairo, M., White, D., Walters, S., Cohen, J., Young, T., Cheong, Y. C., Laird, S., et al. (2022) · Health Technology Assessment · doi:10.3310/jnzt9406 · Cited by 15
Lenhart, A. E., Castillo, C. E., Villegas, E., Alexander, N. D. E., Vanlerberghe, V., der Stuyft, P. V., & McCall, P. J. (2022) · PLoS neglected tropical diseases · doi:10.1371/journal.pntd.0010135 · Cited by 16
“Instead, as a secondary analysis, we follow the recommendation of Altman & Doré [34] and of Senn [35] to include baseline values as a covariate in the regression model.”
MethodVollenweider, D. J., Boyd, C. M., & Puhan, M. A. (2011) · BMC Medicine · doi:10.1186/1741-7015-9-73 · Cited by 12
“This is not a particularly useful comparison, because it tests the null hypothesis that treatment groups are not different, even though we know that through randomization the null hypothesis is true [28-30].”
BackgroundWan, F. (2019) · Statistical Methods in Medical Research · doi:10.1177/0962280219827971 · Cited by 11
“For example, ANCOVA utilizes the correlation between the baseline and post-randomization scores and is more efficient than analysis of variance model (ANOVA).(1,2) Even though ANCOVA and repeated measures model (RM) differ in such a way that the baseline score is treated as a covariate by one but jointly with the post…”
MethodMetwally, M., Chatters, R., Pye, C., Dimairo, M., White, D., Walters, S., Cohen, J., Young, T., Cheong, Y., Laird, S., et al. (2022) · White Rose Research Online (University of Leeds, The University of Sheffield, University of York) · Cited by 11
Comparison of Four Methods of Allocation for Clinical Trials with Small Sample Sizes
Davidson, I., & Hillier, V. F. (2002) · Physiotherapy · doi:10.1016/s0031-9406(05)60715-8 · Cited by 14
Petrovic, Z., Clayton, O., Matthews, J. M., Wade, C., Tan, L., Meyer, D., Gates, A., Almendingen, A., & Cann, W. G. (2019) · BMC Medical Research Methodology · doi:10.1186/s12874-019-0846-1 · Cited by 12
“As a check for appropriate balance between the control and intervention groups and to provide an overview of the study population [24], both at service and individual levels, data for the control and intervention groups will be compared at baseline.”
BackgroundHa, D. A., Oanh, T. T. M., Nguyen, H. L., Chiriboga, G., Goldberg, R. J., Phan, V. H., Nguyen, C. T., Giang, N. H., Pham, H. V., Nguyen, T., et al. (2020) · Trials · doi:10.1186/s13063-020-04917-8 · Cited by 14
“In accordance with best practice, differences in baseline characteristics of the intervention and comparison groups will be established based on standardized differences, rather than on tests of statistical significance [63, 64].”
BackgroundTen Statistics Commandments That Almost Never Should Be Broken
Knapp, T. R., & Brown, J. K. (2014) · Research in Nursing & Health · doi:10.1002/nur.21605 · Cited by 11
“For more on this matter, see Senn (1994) and Assmann, Pocock, Enos, and Kasten (2000).”
BackgroundClinical trials in acute myocardial infarction
Zeymer, U., & Neuhaus, K. L. (1999) · Current Opinion in Cardiology · doi:10.1097/00001573-199909000-00007 · Cited by 12
A Novel Criterion for Selecting Covariates
Berger, V. W. (2005) · Drug Information Journal · doi:10.1177/009286150503900303 · Cited by 10
Østbye, T., Stroo, M., Brouwer, R. J. N., Peterson, B. L., Eisenstein, E. L., Fuemmeler, B. F., Joyner, J., Gulley, L., & Dement, J. M. (2013) · Contemporary Clinical Trials · doi:10.1016/j.cct.2013.04.007 · Cited by 16
McDaniel, A. M., Vickerman, K. A., Stump, T. E., Monahan, P. O., Fellows, J. L., Weaver, M. T., Carlini, B. H., Champion, V. L., & Zbikowski, S. M. (2015) · BMJ Open · doi:10.1136/bmjopen-2014-007260 · Cited by 12
“Logistic regression models were used to model intervention efficacy on the 6-month and 12-month outcomes of abstinence, while adjusting for theoretically important potentially confounding baseline covariates.(26) There were four main outcomes (7-day and 30-day point prevalence quit rates assessed at 6 and 12 months) a…”
The quality of reports of randomized trials in multiple sclerosis: a review
Signori, A., Baccino, A., & Sormani, M. P. (2012) · Multiple Sclerosis Journal · doi:10.1177/1352458512444327 · Cited by 10
Tang, Y. (2018) · Statistics in Biopharmaceutical Research · doi:10.1080/19466315.2018.1459312 · Cited by 10
“It has two main advantages over the analysis of variance (ANOVA) and the two-sample t test (Senn 1994; Pocock et al. 2002; Senn 2006; Borm, Fransen, and Lemmens 2007; Egbewale, Lewis, and Sim 2014).”
BackgroundMethodThe design and analysis of vaccine trials for COVID ‐19 for the purpose of estimating efficacy
Senn, S. J. (2022) · Pharmaceutical Statistics · doi:10.1002/pst.2226 · Cited by 12
Self-citationColeman, M., Burke, R., Augustin, F., Piñero, A., Maldonado, J., Fisher, J. P., Israetel, M. A., Androulakis‐Korakakis, P., Swinton, P. A., Oberlin, D. J., et al. (2024) · PeerJ · doi:10.7717/peerj.16777 · Cited by 15
“Inferences were not drawn on baseline nor within-group change, as baseline testing is inconsequential ( Senn, 1994 ) and within-group outcomes are not the subject of our research question ( Bland & Altman, 2011 ), although we descriptively presented within-group changes to help contextualize our…”
BackgroundNicholas, K. S., Yeatts, S. D., Zhao, W., Ciolino, J. D., Borg, K. T., & Durkalski, V. (2015) · Statistics in Medicine · doi:10.1002/sim.6447 · Cited by 10
“In light of the statistical evidence in favor of adjusted analyses, it is clear that important prognostic covariates that are included in the randomization scheme should also be included in the final analysis in the form of a properly specified analysis of covariance [2,10,19], and in fact, this results in unbiased es…”
Backgroundon behalf of the HeLP Trial Management Group, Creanor, S., Lloyd, J., Hillsdon, M. M., Dean, S. G., Green, C., Taylor, R. S., Ryan, E. C., & Wyatt, K. M. (2016) · Trials · doi:10.1186/s13063-016-1737-y · Cited by 10
“sensitivity analyses may be performed, to allow for such variable(s), in addition to the prespecified variables for adjustment, to assess the robustness of the primary analysis [19, 22].”
BackgroundPreventive vaccines against bioterrorism: evaluation of efficacy and safety
Horne, A. D. (2004) · Vaccine · doi:10.1016/j.vaccine.2004.04.037 · Cited by 10
Sathyapalan, T., Javed, Z., Rigby, A. S., Kilpatrick, E. S., & Atkin, S. L. (2017) · Journal of the Endocrine Society · doi:10.1210/js.2016-1068 · Cited by 11
“Baseline parameters were not compared statistically because this was a randomized controlled study [20, 21].”
BackgroundHarvey, L. A., Glinsky, J. V., Bowden, J. L., & Arora, M. (2014) · Spinal Cord · doi:10.1038/sc.2014.152 · Cited by 9
A comparison of nonparametric and parametric methods to adjust for baseline measures
Carlsson, M. O., Zou, K. H., Yu, C., Liu, K., & Sun, F. (2014) · Contemporary Clinical Trials · doi:10.1016/j.cct.2014.01.002 · Cited by 9
“An exhaustive search for significant prognostic factors can bias the adjusted effect of treatment [19,20].”
BackgroundMofid, L. S., Casapía, M., Montresor, A., Rahme, E., Fraser, W. D., Marquis, G. S., Vercruysse, J., Allen, L. H., & Gyorkos, T. W. (2015) · BMJ Open · doi:10.1136/bmjopen-2015-008560 · Cited by 11
“In addition to estimation of crude intervention effects, multivariable linear regression analysis will be used to compare infant mean weight gain between intervention groups, while adjusting for baseline covariates that were a priori determined to be important confounders in the published literature.(57) These include…”
MethodEffectiveness of the Elos 2.0 Program, a Classroom Good Behavior Game Version in Brazil
Caetano, S. C., Mariano, M., da Silva, A. R., de Sousa, M. M. M., Ferreira, N. R., Cogo‐Moreira, H., Mari, J. D. J., & Sanchez, Z. M. (2024) · International Journal of Mental Health and Addiction · doi:10.1007/s11469-024-01256-6 · Cited by 11
“…statistical testing on the decision of selecting potential adjustments, we adopted a quantitative policy based on arguments against baseline testing of the total randomized group (Altman, 1985; Senn, 1994) and relevant literature on statistical testing (Molen-berghs et al., 2008; Rubin, 1976).”
Zhao, W., Yeatts, S. D., Broderick, J. P., Selim, M. H., Adeoye, O. M., Durkalski‐Mauldin, V. L., Meinzer, C., Martin, R., Dillon, C. R., Cassarly, C. N., et al. (2023) · Stroke · doi:10.1161/strokeaha.122.040743 · Cited by 8
Sen, S., Cherkerzian, S., Herlihy, M., Hacker, M. R., McElrath, T. F., Cantonwine, D. E., Fichorova, R. N., Oken, E., & Meydani, S. N. (2024) · International Journal of Obesity · doi:10.1038/s41366-024-01472-z · Cited by 9
“Analyses were performed using models both unadjusted and adjusted for imbalance in a priori baseline covariates prognostic of outcome (maternal educational attainment, and baseline serum measurement, infections, prenatal vitamins, and analgesic medication use) [26 – 29].”
Cognitive and Emotional Outcomes of Latino Threat Narratives in News Media: An Exploratory Study
Wei, K., Booth, J. M., & Fusco, R. A. (2019) · Journal of the Society for Social Work and Research · doi:10.1086/703265 · Cited by 11
“been recommended as amore reliable approach than only conducting tests of significance with baseline variables (Assmann, Pocock, Enos, & Kasten, 2000; Egbewale, 2015; Egbewale, Lewis, & Sim, 2014; Senn, 1994).”
MethodWollum, A., Zuniga, C., Lezama, N., Grossman, D., & Grindlay, K. (2020) · Journal of Women s Health · doi:10.1089/jwh.2020.8706 · Cited by 8
“21 We also performed a secondary analysis controlling for the following clinician characteristics, which were identified a priori 22 as potentially affecting support of a OTC OC as documented in prior literature 15,16 or due to the research team’s hypotheses: clinician type (primary care physician,…”
Clayson, P. E., Molina, J. L., Joshi, Y. B., Thomas, M. L., Sprock, J., Nungaray, J. A., Swerdlow, N. R., & Light, G. A. (2021) · Psychiatry Research · doi:10.1016/j.psychres.2021.114239 · Cited by 8
Lauzon, S. D., Zhao, W., Nietert, P. J., Ciolino, J. D., Hill, M. D., & Ramakrishnan, V. (2021) · Statistical Methods in Medical Research · doi:10.1177/09622802211055856 · Cited by 8
“This is magnified in trials with small or medium sample sizes.(3,4,8) Clinical trials with substantial imbalances in important prognostic factors come under criticism, even though they may occur by chance.”
BackgroundWilk, A. S., Drewry, K. M., Zhang, R. V., Pastan, S. O., Thorsness, R., Trivedi, A. N., & Patzer, R. E. (2022) · JAMA Network Open · doi:10.1001/jamanetworkopen.2022.25516 · Cited by 9
Handling of baseline measurements in the analysis of crossover trials
Chen, X., Meng, Z., & Zhang, J. (2012) · Statistics in Medicine · doi:10.1002/sim.4466 · Cited by 8
Assessment of P values for demographic data in randomized controlled trials
Ahn, E. J., Kim, J., Kim, T. K., Park, J. H., Lee, D. K., Lee, S., In, J., & Kang, H. (2018) · Korean Journal of Anesthesiology · doi:10.4097/kja.d.18.00333 · Cited by 8
“In studies investigating baseline balance in clinical trials, Roberts and Torgerson [1], and Senn [2] suggested that significance tests to detect baseline differences are inappropriate.”
BackgroundInfluentialOn the variability of covariate adjustment
Lesaffre, E. M. E. H., Bogaerts, K., Li, X., & Bluhmki, E. (2002) · Controlled Clinical Trials · doi:10.1016/s0197-2456(01)00201-x · Cited by 8
Senn, S. J. (2011) · Wiley eBooks · doi:10.1002/9781119951445.ch3 · Cited by 8
Self-citationMartens, M., Zghoul, T., Watson, E., Rieger, S. W., Capitão, L. P., & Harmer, C. J. (2024) · Translational Psychiatry · doi:10.1038/s41398-024-02944-6 · Cited by 11
“There was also signi fi cant group by timepoint interaction for VAS ratings of alertness (F (2,56) = 6.67, p = 0.003, η 2 = 0.192) (see also Table 2).”
InfluentialSherry, A. D., Passy, A. H., McCaw, Z. R., Jaoude, J. A., Lin, T. A., Kouzy, R., Miller, A. M., Kupferman, G. S., Beck, E. J., Msaouel, P., et al. (2024) · JCO Clinical Cancer Informatics · doi:10.1200/cci.24.00102 · Cited by 12
Larkin, C., Wijesundara, J. G., Nguyen, H. L., Ha, D. A., Vuong, A., Nguyen, C. K., Amante, D. J., Chau, N. Q., Phan, P. T., Pham, Q. T. L., et al. (2021) · JMIR Research Protocols · doi:10.2196/30947 · Cited by 8
“In accordance with best practice, differences in baseline characteristics of the intervention and comparison groups will be established based on standardized differences rather than on tests of statistical significance [16,17].”
Eriksson, L. B., Gordh, T., Karlsten, R., LoMartire, R., Thor, A. L. I., & Tegelberg, Å. (2023) · British Journal of Pain · doi:10.1177/20494637231222327 · Cited by 8
“41 Nevertheless, we can state that the difference between the groups were minimal and non-signi fi cant regarding demographics as well as pain score data [Table 1].”
Dunbar, H., Crossland, M. D., Bunce, C., Egan, C. A., & Rubin, G. S. (2012) · Ophthalmic and Physiological Optics · doi:10.1111/j.1475-1313.2012.00914.x · Cited by 9
Zhang, Z., Tang, L., Liu, C., & Berger, V. W. (2018) · Clinical Trials · doi:10.1177/1740774518813120 · Cited by 10
Evaluation of dental trials comparing baseline differences usingpvalues
Wadgave, U., Khairnar, M., Kadu, T. S., & Wadgave, Y. (2019) · Acta Odontologica Scandinavica · doi:10.1080/00016357.2018.1522448 · Cited by 7
“However, some important covariates may not always be balanced between intervention groups following randomization [2,3].”
BackgroundInfluentialWhite, J., Holliday, J., Daniel, R., Campbell, R. M., & Moore, L. (2019) · Addiction · doi:10.1111/add.14862 · Cited by 7
“As all analyses adjusted for the baseline smoking status of resident(s), these imbalances do not preclude valid inferences being drawn about the intervention effect [15].”
Hernández, M., Lee, J. J., Yeap, B. Y., Ye, R., Foote, R. L., Busse, P. M., Patel, S. H., Dagan, R., Snider, J. W., III, Mohammed, N., et al. (2020) · Advances in Radiation Oncology · doi:10.1016/j.adro.2020.100635 · Cited by 9
“Sample sizes between treatment arms become imbalanced and the potential for confounding increases.(27-30) Thus, results from as-treated analyses should only be reported after using covariate adjustment, as described by Lee et al.”
BackgroundMethodA long-term follow-up evaluation of the Minnesota High Risk Revocation Reduction reentry program
McNeeley, S. (2018) · Journal of Experimental Criminology · doi:10.1007/s11292-018-9330-x · Cited by 12
“Coefficients are presented with standard errors in parentheses *p < .05 As recommended by Senn (1994) and de Boer et al. (2015), the main analyses are fully adjusted by controlling for all available prognostic variables.”
MethodMinimising statistical errors in the research domain
Dhulkhed, V., Tantry, T. P., & Kurdi, M. (2021) · Indian Journal of Anaesthesia · doi:10.4103/ija.ija_720_21 · Cited by 7
“, of the study groups is commonly observed and this practice is not advisable.[17] In case they have a confounding influence on the study findings,”
BackgroundAnalysis of Pretest-Posttest Data and the Regression to the Mean: A Review
学, 岩. (2002) · Kodo Keiryogaku (The Japanese Journal of Behaviormetrics) · doi:10.2333/jbhmk.29.247 · Cited by 8
CASSAI, A. D., Dost, B., Turunç, E., Turan, E. İ., Beldağlı, M., Yılmaz, M. A., Karapınar, Y. E., Boscolo, A., & Navalesi, P. (2025) · Anesthesiology · doi:10.1097/aln.0000000000005776 · Cited by 6
“…reporting of absolute difference or standardized difference could be more helpful, especially if accompanied by a critical discussion of the authors 17 Included Randomized Trials per Journal in case of identified, potentially relevant differences, the use of subgroups and sensitivity analysis…”
Frick, J., Gebert, P., Grittner, U., Letsch, A., Schindel, D., & Schenk, L. (2022) · BMC Cancer · doi:10.1186/s12885-022-09646-6 · Cited by 6
“Regardless of whether randomization is used or not, control of baseline distribution of covariates appears to be recommended [30, 31].”
BackgroundO’Connor, C. M., Mentz, R. J., & Whellan, D. J. (2011) · Heart Failure Clinics · doi:10.1016/j.hfc.2011.06.011 · Cited by 6
Oliveira, R. D. Á., Nakajima, E., Vasconcelos, V., Riera, R., & Baptista‐Silva, J. C. C. (2020) · Jornal Vascular Brasileiro · doi:10.1590/1677-5449.190086 · Cited by 6
Balancing after randomization in orthopedic trials: Are we even or even paying attention?
Landy, D. C., Joiner, J., Conley, C., Stone, A. V., Jacobs, C. A., & Stromberg, A. J. (2022) · Journal of Orthopaedic Research® · doi:10.1002/jor.25486 · Cited by 6
“While hypothesis testing for statistical significance is frequently used to assess differences in outcomes, this is not helpful to assess differences in baseline characteristics.(8,9) When groups are randomly selected from the same study population, we know by design that the groups do not represent different populati…”
MethodBaseline Adjustment in Longitudinal Studies
Senn, S. J. (2005) · Encyclopedia of Biostatistics · doi:10.1002/0470011815.b2a12007 · Cited by 6
Self-citationWu, J., & Zha, P. (2020) · Preprints.org · doi:10.20944/preprints202009.0572.v1 · Cited by 6
“Past studies by Altman, Senn, Zha and Burger [13-18] share several common errors: They treated health and disease properties of human beings as population.”
BackgroundResultSenn, S. J. (2005) · Biometrical Journal · doi:10.1002/bimj.200510109 · Cited by 6
Self-citationSharda, M., Tuerk, C., Chowdhury, R., Jamey, K., Foster, N. E. V., Custo-Blanch, M., Tan, M., Nadig, A., & Hyde, K. L. (2018) · PsyArXiv (OSF Preprints) · doi:10.31234/osf.io/2xnmv · Cited by 7
Less is More? First Impressions From COSMIC-313
Msaouel, P. (2022) · Cancer Investigation · doi:10.1080/07357907.2022.2136681 · Cited by 7
“It is a common error, known as the Table 1 fallacy, to scan for imbalances between the treatment groups that could explain the observed relative treatment effect estimates in RCTs such as COSMIC-313 (26).”
Alzharani, A. A. Y., Alshami, A. M., Abualait, T., Azman, H. A., Alshamrani, F., Alzahrani, Y. A., & Althobaiti, Y. A. (2024) · Journal of Clinical Medicine · doi:10.3390/jcm13237060 · Cited by 5
“These participants experienced significant improvements in erectile function [46–48].”
Schmidt, C., Zapf, A., Ozga, A., Canbay, A. E., Denzer, U. W., Toni, E. N. D., Lohse, A. W., Schulze, K., Rösch, T., Stein, A., et al. (2024) · BMC Cancer · doi:10.1186/s12885-024-12693-w · Cited by 5
“Tests of statistical significance will not be undertaken for baseline characteristics; rather the clinical importance of any imbalance will be noted [22].”
BackgroundBalancing tests in stratified randomized controlled trials: A cautionary note
FIRPO, S. P., Foguel, M. N., & Jales, H. (2019) · Economics Letters · doi:10.1016/j.econlet.2019.108771 · Cited by 6
An approach to explore for a sweet spot in randomized trials
Redelmeier, D. A., & Tibshirani, R. J. (2019) · Journal of Clinical Epidemiology · doi:10.1016/j.jclinepi.2019.12.012 · Cited by 5
“A further weakness is that the statistical analysis will not identify (or correct) a failure of randomization or other fundamental flaw [31,32].”
BackgroundLi, Z., Luan, Y., Wang, M., Li, Y., Shen, X., Fu, G., & Zhang, W. (2021) · Trials · doi:10.1186/s13063-021-05099-7 · Cited by 5
Powers, J. H. (2020) · Clinical Infectious Diseases · doi:10.1093/cid/ciaa1795 · Cited by 6
A Note on Postrandomization Adjustment of Covariates
Chen, X., Liu, M., & Zhang, J. (2005) · Drug Information Journal · doi:10.1177/009286150503900405 · Cited by 5
Wu, X., & Lai, D. (2014) · Communications in Statistics - Simulation and Computation · doi:10.1080/03610918.2013.775295 · Cited by 5
“Senn (1994) recommended using ANCOVA regardless of the balance of the prognostic variables.”
MethodBaseline comparisons in surgical trials
Hall, J. C., & Hall, J. L. (2002) · ANZ Journal of Surgery · doi:10.1046/j.1445-2197.2002.02476.x · Cited by 5
“British Journal of Surgery 45 (22) Annals of Surgery 32 (16) Archives of Surgery 28 (14) Lancet 27 (13) American Journal of Surgery 19 (9) New England Journal of Medicine 14 (7) Surgery 12 (6) Journal of the American College of Surgeons 11 (5) World Journal of Surgery 11 (5) British Medical Journal 7 (3) Table 3.”
Background
Metadata
- Authors
- Senn, S.
- Corresponding author
- Senn, Stephen
- ORCID
- Senn, Stephen
- Venue
- Statistics in Medicine
- Publisher
- Wiley
- Published
- 1994-09-15
- Volume
- 13
- Issue
- 17
- Pages
- 1715-1726
- ISSN
- 0277-6715 (ISSN-L)
- 1097-0258
- OpenAlex
- W2121392364
- Cited by
- 503
- References
- 39
- Citations by year
- 2026: 18 · 2025: 18 · 2024: 25 · 2023: 17 · 2022: 36 · 2021: 25 · 2020: 47 · 2019: 19 · 2018: 43 · 2017: 24 · 2016: 15 · 2015: 28 · 2014: 21 · 2013: 22 · 2012: 19
- FWCI
- 3.08
- Type
- journal article
- Language
- en
- Open access
- closed
- Locations
- Statistics in Medicine (publishedVersion)
- PubMed (publishedVersion)
- Funders
- Royal Society
- Topics
- Physical Sciences › Mathematics › Statistics and Probability › Statistical Methods in Clinical Trials
- Physical Sciences › Mathematics › Statistics and Probability › Advanced Causal Inference Techniques
- Physical Sciences › Mathematics › Statistics and Probability › Statistical Methods and Bayesian Inference
- Keywords
- prognostic factors, covariance analysis, treatment effect estimation, clinical trial design, covariate selection
- MeSH terms
- Algorithms, Analysis of Variance, Data Interpretation, Statistical, Humans, Research Design, Selection Bias, Multivariate Analysis, Randomized Controlled Trials as Topic, Treatment Outcome, Sample Size
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
Data sources
Bibliographic data: Semantic Scholar, OpenAlex, Crossref.
Contains information from the Semantic Scholar Academic Graph Datasets, made available under the ODC Attribution License (ODC-BY 1.0).
Semantic Scholar’s data is described in: Kinney, R., et al. (2023). The Semantic Scholar Open Data Platform. arXiv:2301.10140.