Sorkin, John
2025-Sep-18 11:08 UTC
[R] Matching when each subject has multiple records, but each subject should be used only once in the match
I have a file that contains longitudinal data for each subject. As a result, each subject can have multiple records. For example a given subject might have a record in Jan 2020, another in June 2020, another in Feb 2021, another in May 2021, another in Sept 2022, etc. At each time for which a subject has a record the subject is identified as a case or a control. Over the course of the longitudinal data, I want to match a given case to a given control. Once a subject is matched, I don't want the subject to be eligible for being matched again. If each subject had a single record, matching could easily be accomplished. How can I accomplish the match in my file having repeated measures for each subject? John David Sorkin M.D., Ph.D. Professor of Medicine, University of Maryland School of Medicine; Associate Director for Biostatistics and Informatics, Baltimore VA Medical Center Geriatrics Research, Education, and Clinical Center;? PI?Biostatistics and Informatics Core, University of Maryland School of Medicine Claude D. Pepper Older Americans Independence Center; Senior Statistician University of Maryland Center for Vascular Research; Division of Gerontology and Paliative Care, 10 North Greene Street GRECC (BT/18/GR) Baltimore, MD 21201-1524 Cell phone 443-418-5382
Sorkin, John
2025-Sep-18 11:37 UTC
[R] Matching when each subject has multiple records, but each subject should be used only once in the match
I should have added that I want to propensity score match the subjects, which normally could be accomplished with MatchIt, but MatchIt will not, as far as I can determine, work with longitudinal data if I want each subject to be matched exactly once. John David Sorkin M.D., Ph.D. Professor of Medicine, University of Maryland School of Medicine; Associate Director for Biostatistics and Informatics, Baltimore VA Medical Center Geriatrics Research, Education, and Clinical Center; PI Biostatistics and Informatics Core, University of Maryland School of Medicine Claude D. Pepper Older Americans Independence Center; Senior Statistician University of Maryland Center for Vascular Research; Division of Gerontology and Paliative Care, 10 North Greene Street GRECC (BT/18/GR) Baltimore, MD 21201-1524 Cell phone 443-418-5382 ________________________________________ From: R-help <r-help-bounces at r-project.org> on behalf of Sorkin, John <jsorkin at som.umaryland.edu> Sent: Thursday, September 18, 2025 7:08 AM To: Leo Mada via R-help <r-help at r-project.org> Subject: [R] Matching when each subject has multiple records, but each subject should be used only once in the match I have a file that contains longitudinal data for each subject. As a result, each subject can have multiple records. For example a given subject might have a record in Jan 2020, another in June 2020, another in Feb 2021, another in May 2021, another in Sept 2022, etc. At each time for which a subject has a record the subject is identified as a case or a control. Over the course of the longitudinal data, I want to match a given case to a given control. Once a subject is matched, I don't want the subject to be eligible for being matched again. If each subject had a single record, matching could easily be accomplished. How can I accomplish the match in my file having repeated measures for each subject? John David Sorkin M.D., Ph.D. Professor of Medicine, University of Maryland School of Medicine; Associate Director for Biostatistics and Informatics, Baltimore VA Medical Center Geriatrics Research, Education, and Clinical Center; PI Biostatistics and Informatics Core, University of Maryland School of Medicine Claude D. Pepper Older Americans Independence Center; Senior Statistician University of Maryland Center for Vascular Research; Division of Gerontology and Paliative Care, 10 North Greene Street GRECC (BT/18/GR) Baltimore, MD 21201-1524 Cell phone 443-418-5382 ______________________________________________ R-help at r-project.org mailing list -- To UNSUBSCRIBE and more, see https://stat.ethz.ch/mailman/listinfo/r-help PLEASE do read the posting guide https://www.r-project.org/posting-guide.html and provide commented, minimal, self-contained, reproducible code.
@vi@e@gross m@iii@g oii gm@ii@com
2025-Sep-18 14:05 UTC
[R] Matching when each subject has multiple records, but each subject should be used only once in the match
John, You may want to consider various ways to store and manipulate the data and even switch between views as needed. It sounds like one option is to have multiple tables. One table can be the one with longitudinal data and can include multiple records/rows with the same patient.id but another table can contain matches of patient.id and control.id with perhaps some being NA at times. When you want to, you can use various joins to make a combined table for various uses. Joining the above tables on patient.id will populate the result to include the same control.id for each relevant row if that is a helpful scenario. Obviously, this means some work to keep things synchronized when patients are added or removed. Other ways to deal with things can involve adding additional columns for special uses, such as to mark one row as being the primary for some purpose or as having already been matched or other techniques that help refine later analysis such as selecting a specific subset of rows. And, although you have said you would prefer to use base R, you can consider ways to do grouping in base R so that whatever actions you are doing will be done piecewise one group at a time. The dplyr package can be very useful in this regard as you can easily group things and then make changes or generate reports based on it. -----Original Message----- From: R-help <r-help-bounces at r-project.org> On Behalf Of Sorkin, John Sent: Thursday, September 18, 2025 7:08 AM To: Leo Mada via R-help <r-help at r-project.org> Subject: [R] Matching when each subject has multiple records, but each subject should be used only once in the match I have a file that contains longitudinal data for each subject. As a result, each subject can have multiple records. For example a given subject might have a record in Jan 2020, another in June 2020, another in Feb 2021, another in May 2021, another in Sept 2022, etc. At each time for which a subject has a record the subject is identified as a case or a control. Over the course of the longitudinal data, I want to match a given case to a given control. Once a subject is matched, I don't want the subject to be eligible for being matched again. If each subject had a single record, matching could easily be accomplished. How can I accomplish the match in my file having repeated measures for each subject? John David Sorkin M.D., Ph.D. Professor of Medicine, University of Maryland School of Medicine; Associate Director for Biostatistics and Informatics, Baltimore VA Medical Center Geriatrics Research, Education, and Clinical Center;? PI?Biostatistics and Informatics Core, University of Maryland School of Medicine Claude D. Pepper Older Americans Independence Center; Senior Statistician University of Maryland Center for Vascular Research; Division of Gerontology and Paliative Care, 10 North Greene Street GRECC (BT/18/GR) Baltimore, MD 21201-1524 Cell phone 443-418-5382 ______________________________________________ R-help at r-project.org mailing list -- To UNSUBSCRIBE and more, see https://stat.ethz.ch/mailman/listinfo/r-help PLEASE do read the posting guide https://www.R-project.org/posting-guide.html and provide commented, minimal, self-contained, reproducible code.