Martijn Heymans



Research output per year

If you made any changes in Pure these will be visible here soon.

Personal profile

Research interests

My research interest is in applied methodological research in (merged) cohort data to better answer medical and epidemiological questions in the field of Cancer, Low back pain research, Diabetes and Aging. The scientific questions can be etiological or predictive in nature. These data sets often contain unique missing data situations as missing data due to death or variables that are missing ​​for an entire cohort in combination with missing values ​​within separate cohort. Applied clinical researchers often do not know how to reliably combine data from different cohorts or when to use a recommended missing data method as Multiple Imputation. With this latter procedure statistical techniques are performed in the multiple completed data sets and then results have to be pooled. However, not for all statistical techniques it is known which pooling procedures have to be used. My research line focuses on the development and application of methods to reliably merge cohort data in combination with (multiple) imputation, data analysis and pooling, with development of hands-on RStudio applications for applied researchers.

Network Recent external collaboration on country level. Dive into details by clicking on the dots.

Research Output

Kinesiophobia is not required to predict chronic low back pain in workers: A decision curve analysis

Panken, A. M., Staal, J. B. & Heymans, M. W., 12 Mar 2020, In : BMC Musculoskeletal Disorders. 21, 1, 163.

Research output: Contribution to journalArticleAcademicpeer-review

Open Access

Passive imputation and parcel summaries are both valid to handle missing items in studies with many multi-item scales

Eekhout, I., de Vet, H. C. W., de Boer, M. R., Twisk, J. W. R. & Heymans, M. W., 1 Apr 2018, In : Statistical Methods in Medical Research. 27, 4, p. 1128-1140 13 p.

Research output: Contribution to journalArticleAcademicpeer-review

Multiple imputation of missing values was not necessary before performing a longitudinal mixed-model analysis

Twisk, J. W. R., de Boer, M. R., de Vente, W. & Heijmans, M. W., 2013, In : Journal of Clinical Epidemiology. 66, 9, p. 1022-1028

Research output: Contribution to journalArticleAcademicpeer-review