Performance of quantitative measures of multimorbidity: a population-based retrospective analysis.
Chronic disease
Health planning
Health resources
Multimorbidity
Risk assessment
Journal
BMC public health
ISSN: 1471-2458
Titre abrégé: BMC Public Health
Pays: England
ID NLM: 100968562
Informations de publication
Date de publication:
18 10 2021
18 10 2021
Historique:
received:
05
02
2021
accepted:
05
10
2021
entrez:
19
10
2021
pubmed:
20
10
2021
medline:
3
11
2021
Statut:
epublish
Résumé
Multimorbidity measures are useful for resource planning, patient selection and prioritization, and factor adjustment in clinical practice, research, and benchmarking. We aimed to compare the explanatory performance of the adjusted morbidity group (GMA) index in predicting relevant healthcare outcomes with that of other quantitative measures of multimorbidity. The performance of multimorbidity measures was retrospectively assessed on anonymized records of the entire adult population of Catalonia (North-East Spain). Five quantitative measures of multimorbidity were added to a baseline model based on age, gender, and socioeconomic status: the Charlson index score, the count of chronic diseases according to three different proposals (i.e., the QOF, HCUP, and Karolinska institute), and the multimorbidity index score of the GMA tool. Outcomes included all-cause death, total and non-scheduled hospitalization, primary care and ER visits, medication use, admission to a skilled nursing facility for intermediate care, and high expenditure (time frame 2017). The analysis was performed on 10 subpopulations: all adults (i.e., aged > 17 years), people aged > 64 years, people aged > 64 years and institutionalized in a nursing home for long-term care, and people with specific diagnoses (e.g., ischemic heart disease, cirrhosis, dementia, diabetes mellitus, heart failure, chronic kidney disease, and chronic obstructive pulmonary disease). The explanatory performance was assessed using the area under the receiving operating curves (AUC-ROC) (main analysis) and three additional statistics (secondary analysis). The adult population included 6,224,316 individuals. The addition of any of the multimorbidity measures to the baseline model increased the explanatory performance for all outcomes and subpopulations. All measurements performed better in the general adult population. The GMA index had higher performance and consistency across subpopulations than the rest of multimorbidity measures. The Charlson index stood out on explaining mortality, whereas measures based on exhaustive definitions of chronic diagnostic (e.g., HCUP and GMA) performed better than those using predefined lists of diagnostics (e.g., QOF or the Karolinska proposal). The addition of multimorbidity measures to models for explaining healthcare outcomes increase the performance. The GMA index has high performance in explaining relevant healthcare outcomes and may be useful for clinical practice, resource planning, and public health research.
Sections du résumé
BACKGROUND
Multimorbidity measures are useful for resource planning, patient selection and prioritization, and factor adjustment in clinical practice, research, and benchmarking. We aimed to compare the explanatory performance of the adjusted morbidity group (GMA) index in predicting relevant healthcare outcomes with that of other quantitative measures of multimorbidity.
METHODS
The performance of multimorbidity measures was retrospectively assessed on anonymized records of the entire adult population of Catalonia (North-East Spain). Five quantitative measures of multimorbidity were added to a baseline model based on age, gender, and socioeconomic status: the Charlson index score, the count of chronic diseases according to three different proposals (i.e., the QOF, HCUP, and Karolinska institute), and the multimorbidity index score of the GMA tool. Outcomes included all-cause death, total and non-scheduled hospitalization, primary care and ER visits, medication use, admission to a skilled nursing facility for intermediate care, and high expenditure (time frame 2017). The analysis was performed on 10 subpopulations: all adults (i.e., aged > 17 years), people aged > 64 years, people aged > 64 years and institutionalized in a nursing home for long-term care, and people with specific diagnoses (e.g., ischemic heart disease, cirrhosis, dementia, diabetes mellitus, heart failure, chronic kidney disease, and chronic obstructive pulmonary disease). The explanatory performance was assessed using the area under the receiving operating curves (AUC-ROC) (main analysis) and three additional statistics (secondary analysis).
RESULTS
The adult population included 6,224,316 individuals. The addition of any of the multimorbidity measures to the baseline model increased the explanatory performance for all outcomes and subpopulations. All measurements performed better in the general adult population. The GMA index had higher performance and consistency across subpopulations than the rest of multimorbidity measures. The Charlson index stood out on explaining mortality, whereas measures based on exhaustive definitions of chronic diagnostic (e.g., HCUP and GMA) performed better than those using predefined lists of diagnostics (e.g., QOF or the Karolinska proposal).
CONCLUSIONS
The addition of multimorbidity measures to models for explaining healthcare outcomes increase the performance. The GMA index has high performance in explaining relevant healthcare outcomes and may be useful for clinical practice, resource planning, and public health research.
Identifiants
pubmed: 34663289
doi: 10.1186/s12889-021-11922-2
pii: 10.1186/s12889-021-11922-2
pmc: PMC8524794
doi:
Types de publication
Journal Article
Langues
eng
Sous-ensembles de citation
IM
Pagination
1881Informations de copyright
© 2021. The Author(s).
Références
Maturitas. 2018 Mar;109:53-62
pubmed: 29452782
Lancet. 2012 Jul 7;380(9836):37-43
pubmed: 22579043
BMJ Open. 2016 Apr 15;6(4):e010301
pubmed: 27084274
Aten Primaria. 2020 Feb;52(2):96-103
pubmed: 30765102
Ann Fam Med. 2012 Mar-Apr;10(2):134-41
pubmed: 22412005
BMC Fam Pract. 2020 Feb 17;21(1):39
pubmed: 32066377
Med Care Res Rev. 2011 Aug;68(4):387-420
pubmed: 21813576
PLoS One. 2015 Mar 04;10(3):e0118432
pubmed: 25738806
J Health Organ Manag. 2006;20(2-3):150-62
pubmed: 16869351
BMJ Open. 2021 May 5;11(5):e041219
pubmed: 33952533
Gac Sanit. 2019 Jan - Feb;33(1):24-31
pubmed: 29129491
Med Care. 2004 Jan;42(1):81-90
pubmed: 14713742
Fam Pract. 2009 Apr;26(2):79-80
pubmed: 19287000
J Clin Epidemiol. 2008 Feb;61(2):151-9
pubmed: 18177788
Pediatrics. 1993 Apr;91(4):787-93
pubmed: 8464668
J Health Econ. 2014 May;35:109-22
pubmed: 24657375
JAMA. 2012 Jun 20;307(23):2493-4
pubmed: 22797447
Br J Gen Pract. 2011 Jan;61(582):e12-21
pubmed: 21401985
J Gerontol A Biol Sci Med Sci. 2017 Oct 1;72(10):1417-1423
pubmed: 28003375
Health Qual Life Outcomes. 2005 Nov 23;3:74
pubmed: 16305743
Aten Primaria. 2017 Apr;49(4):240-247
pubmed: 27592535
Rev Esp Salud Publica. 2020 Jul 3;94:
pubmed: 32618288
BMC Health Serv Res. 2013 Jul 09;13:269
pubmed: 23837560
J Clin Epidemiol. 2005 Jun;58(6):595-602
pubmed: 15878473
Fam Pract. 2013 Apr;30(2):172-8
pubmed: 23045354
J Chronic Dis. 1987;40(5):373-83
pubmed: 3558716
Clin Epidemiol. 2017 May 10;9:279-289
pubmed: 28546772
Aten Primaria. 2016 Dec;48(10):674-682
pubmed: 27495004
Clin Epidemiol. 2013 Jul 01;5:199-203
pubmed: 23861599
J Am Med Dir Assoc. 2019 Apr;20(4):456-461
pubmed: 30287263
Gac Sanit. 2019 Jan - Feb;33(1):60-65
pubmed: 28826908
Ann Fam Med. 2005 May-Jun;3(3):223-8
pubmed: 15928225
Aten Primaria. 2019 Mar;51(3):153-161
pubmed: 29433758