Handling of missing component information for common composite score outcomes used in axial spondyloarthritis research when complete-case analysis is unbiased
Abstract Background Observational data on composite scores often comes with missing component information. When a complete-case (CC) analysis of composite scores is unbiased, preferable approaches of dealing with missing component information should also be unbiased and provide a more precise estima...
Saved in:
| Main Authors: | , , , , , , , , , , , , , , , , , , , , , , , , , , , |
|---|---|
| Format: | Article |
| Language: | English |
| Published: |
BMC
2025-02-01
|
| Series: | BMC Medical Research Methodology |
| Subjects: | |
| Online Access: | https://doi.org/10.1186/s12874-025-02515-3 |
| Tags: |
Add Tag
No Tags, Be the first to tag this record!
|
| _version_ | 1850237996221071360 |
|---|---|
| author | Christos Polysopoulos Stylianos Georgiadis Lykke Midtbøll Ørnbjerg Almut Scherer Daniela Di Giuseppe Merete Lund Hetland Michael John Nissen Gareth T. Jones Bente Glintborg Anne Gitte Loft Johan Karlsson Wallman Karel Pavelka Jakub Závada Ayten Yazici Maria José Santos Adrian Ciurea Burkhard Möller Brigitte Michelsen Pawel Mielnik Johanna Huhtakangas Heikki Relas Katja Perdan Pirkmajer Ziga Rotar Ross MacDonald Bjorn Gudbjornsson Irene van der Horst-Bruinsma Marleen van de Sande Myriam Riek |
| author_facet | Christos Polysopoulos Stylianos Georgiadis Lykke Midtbøll Ørnbjerg Almut Scherer Daniela Di Giuseppe Merete Lund Hetland Michael John Nissen Gareth T. Jones Bente Glintborg Anne Gitte Loft Johan Karlsson Wallman Karel Pavelka Jakub Závada Ayten Yazici Maria José Santos Adrian Ciurea Burkhard Möller Brigitte Michelsen Pawel Mielnik Johanna Huhtakangas Heikki Relas Katja Perdan Pirkmajer Ziga Rotar Ross MacDonald Bjorn Gudbjornsson Irene van der Horst-Bruinsma Marleen van de Sande Myriam Riek |
| author_sort | Christos Polysopoulos |
| collection | DOAJ |
| description | Abstract Background Observational data on composite scores often comes with missing component information. When a complete-case (CC) analysis of composite scores is unbiased, preferable approaches of dealing with missing component information should also be unbiased and provide a more precise estimate. We assessed the performance of several methods compared to CC analysis in estimating the means of common composite scores used in axial spondyloarthritis research. Methods Individual mean imputation (IMI), the modified formula method (MF), overall mean imputation (OMI), and multiple imputation of missing component values (MI) were assessed either analytically or by means of simulations from available data collected across Europe. Their performance in estimating the means of the Bath Ankylosing Spondylitis Disease Activity Index (BASDAI), the Bath Ankylosing Spondylitis Functional Index (BASFI), and the Ankylosing Spondylitis Disease Activity Score based on C-reactive protein (ASDAS-CRP) in cases where component information was set missing completely at random was compared to the CC approach based on bias, variance, and coverage. Results Like the MF method, IMI uses a modified formula for observations with missing components resulting in modified composite scores. In the case of an unbiased CC approach, these two methods yielded representative samples of the distribution arising from a mixture of the original and modified composite scores, which, however, could not be considered the same as the distribution of the original score. The IMI and MF method are, thus, intrinsically biased. OMI provided an unbiased mean but displayed a complex dependence structure among observations that, if not accounted for, resulted in severe coverage issues. MI improved precision compared to CC and gave unbiased means and proper coverage as long as the extent of missingness was not too large. Conclusions MI of missing component values was the only method found successful in retaining CC’s unbiasedness and in providing increased precision for estimating the means of BASDAI, BASFI, and ASDAS-CRP. However, since MI is susceptible to incorrect implementation and its performance may become questionable with increasing missingness, we consider the implementation of an error-free CC approach a valid and valuable option. Trial registration Not applicable as study uses data from patient registries. |
| format | Article |
| id | doaj-art-aacff8dafeb84b9a89d5869aa867dae5 |
| institution | OA Journals |
| issn | 1471-2288 |
| language | English |
| publishDate | 2025-02-01 |
| publisher | BMC |
| record_format | Article |
| series | BMC Medical Research Methodology |
| spelling | doaj-art-aacff8dafeb84b9a89d5869aa867dae52025-08-20T02:01:35ZengBMCBMC Medical Research Methodology1471-22882025-02-0125111610.1186/s12874-025-02515-3Handling of missing component information for common composite score outcomes used in axial spondyloarthritis research when complete-case analysis is unbiasedChristos Polysopoulos0Stylianos Georgiadis1Lykke Midtbøll Ørnbjerg2Almut Scherer3Daniela Di Giuseppe4Merete Lund Hetland5Michael John Nissen6Gareth T. Jones7Bente Glintborg8Anne Gitte Loft9Johan Karlsson Wallman10Karel Pavelka11Jakub Závada12Ayten Yazici13Maria José Santos14Adrian Ciurea15Burkhard Möller16Brigitte Michelsen17Pawel Mielnik18Johanna Huhtakangas19Heikki Relas20Katja Perdan Pirkmajer21Ziga Rotar22Ross MacDonald23Bjorn Gudbjornsson24Irene van der Horst-Bruinsma25Marleen van de Sande26Myriam Riek27Statistics Group, Swiss Clinical Quality Management (SCQM) FoundationCopenhagen Centre for Arthritis Research (COPECARE), Centre for Rheumatology and Spine Diseases, Centre for Head and OrthopaedicsCopenhagen Centre for Arthritis Research (COPECARE), Centre for Rheumatology and Spine Diseases, Centre for Head and OrthopaedicsStatistics Group, Swiss Clinical Quality Management (SCQM) FoundationDepartment of Medicine Solna, Clinical Epidemiology Division, Karolinska InstitutetCopenhagen Centre for Arthritis Research (COPECARE) and the DANBIO Registry, Centre for Rheumatology and Spine Diseases, Centre for Head and OrthopaedicsDepartment of Rheumatology, Geneva University HospitalAberdeen Centre for Arthritis and Musculoskeletal Health (Epidemiology Group), University of AberdeenCopenhagen Centre for Arthritis Research (COPECARE) and the DANBIO Registry, Centre for Rheumatology and Spine Diseases, Centre for Head and OrthopaedicsDepartment of Rheumatology, Aarhus University HospitalDepartment of Clinical Sciences Lund, Rheumatology, Skåne University Hospital, Lund UniversityInstitute of RheumatologyInstitute of RheumatologyDivision of Rheumatology, Department of Internal Medicine, School of Medicine, Kocaeli UniversitySociedade Portuguesa de ReumatologiaDepartment of Rheumatology, University Hospital Zurich, University of ZurichDepartment Rheumatology and Immunology, Inselspital - University Hospital BernCenter for Treatment of Rheumatic and Musculoskeletal Diseases (REMEDY), Diakonhjemmet HospitalSection for Rheumatology, Department for Neurology, Rheumatology and Physical MedicineDepartment of Rheumatology, Kuopio University HospitalRheumatology, Inflammation Center, Helsinki University Hospital and University of HelsinkFaculty of Medicine, University of LjubljanaFaculty of Medicine, University of LjubljanaAberdeen Centre for Arthritis and Musculoskeletal Health (Epidemiology Group), University of AberdeenFaculty of Medicine, University of IcelandRheumatology, Radboud University Medical CentreAmsterdam UMC, Department of Rheumatology & Clinical Immunology and Department of Experimental Immunology, Amsterdam Institute for Infection & Immunity, University of AmsterdamStatistics Group, Swiss Clinical Quality Management (SCQM) FoundationAbstract Background Observational data on composite scores often comes with missing component information. When a complete-case (CC) analysis of composite scores is unbiased, preferable approaches of dealing with missing component information should also be unbiased and provide a more precise estimate. We assessed the performance of several methods compared to CC analysis in estimating the means of common composite scores used in axial spondyloarthritis research. Methods Individual mean imputation (IMI), the modified formula method (MF), overall mean imputation (OMI), and multiple imputation of missing component values (MI) were assessed either analytically or by means of simulations from available data collected across Europe. Their performance in estimating the means of the Bath Ankylosing Spondylitis Disease Activity Index (BASDAI), the Bath Ankylosing Spondylitis Functional Index (BASFI), and the Ankylosing Spondylitis Disease Activity Score based on C-reactive protein (ASDAS-CRP) in cases where component information was set missing completely at random was compared to the CC approach based on bias, variance, and coverage. Results Like the MF method, IMI uses a modified formula for observations with missing components resulting in modified composite scores. In the case of an unbiased CC approach, these two methods yielded representative samples of the distribution arising from a mixture of the original and modified composite scores, which, however, could not be considered the same as the distribution of the original score. The IMI and MF method are, thus, intrinsically biased. OMI provided an unbiased mean but displayed a complex dependence structure among observations that, if not accounted for, resulted in severe coverage issues. MI improved precision compared to CC and gave unbiased means and proper coverage as long as the extent of missingness was not too large. Conclusions MI of missing component values was the only method found successful in retaining CC’s unbiasedness and in providing increased precision for estimating the means of BASDAI, BASFI, and ASDAS-CRP. However, since MI is susceptible to incorrect implementation and its performance may become questionable with increasing missingness, we consider the implementation of an error-free CC approach a valid and valuable option. Trial registration Not applicable as study uses data from patient registries.https://doi.org/10.1186/s12874-025-02515-3Composite scoreMissing componentsComplete-case analysisMultiple imputationAxial spondyloarthritis |
| spellingShingle | Christos Polysopoulos Stylianos Georgiadis Lykke Midtbøll Ørnbjerg Almut Scherer Daniela Di Giuseppe Merete Lund Hetland Michael John Nissen Gareth T. Jones Bente Glintborg Anne Gitte Loft Johan Karlsson Wallman Karel Pavelka Jakub Závada Ayten Yazici Maria José Santos Adrian Ciurea Burkhard Möller Brigitte Michelsen Pawel Mielnik Johanna Huhtakangas Heikki Relas Katja Perdan Pirkmajer Ziga Rotar Ross MacDonald Bjorn Gudbjornsson Irene van der Horst-Bruinsma Marleen van de Sande Myriam Riek Handling of missing component information for common composite score outcomes used in axial spondyloarthritis research when complete-case analysis is unbiased BMC Medical Research Methodology Composite score Missing components Complete-case analysis Multiple imputation Axial spondyloarthritis |
| title | Handling of missing component information for common composite score outcomes used in axial spondyloarthritis research when complete-case analysis is unbiased |
| title_full | Handling of missing component information for common composite score outcomes used in axial spondyloarthritis research when complete-case analysis is unbiased |
| title_fullStr | Handling of missing component information for common composite score outcomes used in axial spondyloarthritis research when complete-case analysis is unbiased |
| title_full_unstemmed | Handling of missing component information for common composite score outcomes used in axial spondyloarthritis research when complete-case analysis is unbiased |
| title_short | Handling of missing component information for common composite score outcomes used in axial spondyloarthritis research when complete-case analysis is unbiased |
| title_sort | handling of missing component information for common composite score outcomes used in axial spondyloarthritis research when complete case analysis is unbiased |
| topic | Composite score Missing components Complete-case analysis Multiple imputation Axial spondyloarthritis |
| url | https://doi.org/10.1186/s12874-025-02515-3 |
| work_keys_str_mv | AT christospolysopoulos handlingofmissingcomponentinformationforcommoncompositescoreoutcomesusedinaxialspondyloarthritisresearchwhencompletecaseanalysisisunbiased AT stylianosgeorgiadis handlingofmissingcomponentinformationforcommoncompositescoreoutcomesusedinaxialspondyloarthritisresearchwhencompletecaseanalysisisunbiased AT lykkemidtbøllørnbjerg handlingofmissingcomponentinformationforcommoncompositescoreoutcomesusedinaxialspondyloarthritisresearchwhencompletecaseanalysisisunbiased AT almutscherer handlingofmissingcomponentinformationforcommoncompositescoreoutcomesusedinaxialspondyloarthritisresearchwhencompletecaseanalysisisunbiased AT danieladigiuseppe handlingofmissingcomponentinformationforcommoncompositescoreoutcomesusedinaxialspondyloarthritisresearchwhencompletecaseanalysisisunbiased AT meretelundhetland handlingofmissingcomponentinformationforcommoncompositescoreoutcomesusedinaxialspondyloarthritisresearchwhencompletecaseanalysisisunbiased AT michaeljohnnissen handlingofmissingcomponentinformationforcommoncompositescoreoutcomesusedinaxialspondyloarthritisresearchwhencompletecaseanalysisisunbiased AT garethtjones handlingofmissingcomponentinformationforcommoncompositescoreoutcomesusedinaxialspondyloarthritisresearchwhencompletecaseanalysisisunbiased AT benteglintborg handlingofmissingcomponentinformationforcommoncompositescoreoutcomesusedinaxialspondyloarthritisresearchwhencompletecaseanalysisisunbiased AT annegitteloft handlingofmissingcomponentinformationforcommoncompositescoreoutcomesusedinaxialspondyloarthritisresearchwhencompletecaseanalysisisunbiased AT johankarlssonwallman handlingofmissingcomponentinformationforcommoncompositescoreoutcomesusedinaxialspondyloarthritisresearchwhencompletecaseanalysisisunbiased AT karelpavelka handlingofmissingcomponentinformationforcommoncompositescoreoutcomesusedinaxialspondyloarthritisresearchwhencompletecaseanalysisisunbiased AT jakubzavada handlingofmissingcomponentinformationforcommoncompositescoreoutcomesusedinaxialspondyloarthritisresearchwhencompletecaseanalysisisunbiased AT aytenyazici handlingofmissingcomponentinformationforcommoncompositescoreoutcomesusedinaxialspondyloarthritisresearchwhencompletecaseanalysisisunbiased AT mariajosesantos handlingofmissingcomponentinformationforcommoncompositescoreoutcomesusedinaxialspondyloarthritisresearchwhencompletecaseanalysisisunbiased AT adrianciurea handlingofmissingcomponentinformationforcommoncompositescoreoutcomesusedinaxialspondyloarthritisresearchwhencompletecaseanalysisisunbiased AT burkhardmoller handlingofmissingcomponentinformationforcommoncompositescoreoutcomesusedinaxialspondyloarthritisresearchwhencompletecaseanalysisisunbiased AT brigittemichelsen handlingofmissingcomponentinformationforcommoncompositescoreoutcomesusedinaxialspondyloarthritisresearchwhencompletecaseanalysisisunbiased AT pawelmielnik handlingofmissingcomponentinformationforcommoncompositescoreoutcomesusedinaxialspondyloarthritisresearchwhencompletecaseanalysisisunbiased AT johannahuhtakangas handlingofmissingcomponentinformationforcommoncompositescoreoutcomesusedinaxialspondyloarthritisresearchwhencompletecaseanalysisisunbiased AT heikkirelas handlingofmissingcomponentinformationforcommoncompositescoreoutcomesusedinaxialspondyloarthritisresearchwhencompletecaseanalysisisunbiased AT katjaperdanpirkmajer handlingofmissingcomponentinformationforcommoncompositescoreoutcomesusedinaxialspondyloarthritisresearchwhencompletecaseanalysisisunbiased AT zigarotar handlingofmissingcomponentinformationforcommoncompositescoreoutcomesusedinaxialspondyloarthritisresearchwhencompletecaseanalysisisunbiased AT rossmacdonald handlingofmissingcomponentinformationforcommoncompositescoreoutcomesusedinaxialspondyloarthritisresearchwhencompletecaseanalysisisunbiased AT bjorngudbjornsson handlingofmissingcomponentinformationforcommoncompositescoreoutcomesusedinaxialspondyloarthritisresearchwhencompletecaseanalysisisunbiased AT irenevanderhorstbruinsma handlingofmissingcomponentinformationforcommoncompositescoreoutcomesusedinaxialspondyloarthritisresearchwhencompletecaseanalysisisunbiased AT marleenvandesande handlingofmissingcomponentinformationforcommoncompositescoreoutcomesusedinaxialspondyloarthritisresearchwhencompletecaseanalysisisunbiased AT myriamriek handlingofmissingcomponentinformationforcommoncompositescoreoutcomesusedinaxialspondyloarthritisresearchwhencompletecaseanalysisisunbiased |