A novel easy-to-use index to predict institutionalization and death in older population – a 10-year population-based follow-up study

BMC Geriatrics

Research Article

Minna Löppönen¹  Laura Viikari²  Tero Vahlberg³  Kerttu Irjala⁴  Elisa Heikkilä⁵  Taina Katajamäki⁵  Marika Salminen⁶  Raimo Isoaho⁷  Sirkka-Liisa Kivelä⁸  Matti Viitanen⁹  Kari Pulkki^1,10  Marja-Kaisa Koivula^1,10  Anna Viljanen^1,11  Mikko S. Venäläinen^1,12  Laura L. Elo^1,13 

[1] City of Raisio, Social and Health Care for Elderly, 21200, Raisio, Finland;City of Turku, Welfare Division/Turku City Hospital, Kunnallissairaalantie 20, 20700, Turku, Finland;Faculty of Medicine, Department of Geriatrics, Turku City Hospital, University of Turku, Kunnallissairaalantie 20, 20700, Turku, Finland;Faculty of Medicine, Department of Clinical Medicine, Unit of Biostatistics, University of Turku, Turku, Finland;Faculty of Medicine, Department of Clinical Medicine, Unit of Clinical Chemistry, Turku University, FI-20014 University of Turku, 20521, Turku, Finland;Faculty of Medicine, Department of Clinical Medicine, Unit of Clinical Chemistry, Turku University, FI-20014 University of Turku, 20521, Turku, Finland;Tykslab, Laboratory Division, Turku University Hospital, Hospital District of Southwest Finland, Turku, Finland;Faculty of Medicine, Department of Clinical Medicine, Unit of Family Medicine, University of Turku and Turku University Hospital, 20014, Turku, Finland;City of Turku, Welfare Division/Turku City Hospital, Kunnallissairaalantie 20, 20700, Turku, Finland;Faculty of Medicine, Department of Clinical Medicine, Unit of Family Medicine, University of Turku and Turku University Hospital, 20014, Turku, Finland;City of Vaasa, Social and Health Care, 65101, Vaasa, Finland;Faculty of Medicine, Department of Clinical Medicine, Unit of Family Medicine, University of Turku and Turku University Hospital, 20014, Turku, Finland;Faculty of Pharmacy, Division of Social Pharmacy, University of Helsinki, 00014, Helsinki, Finland;Faculty of Medicine, Department of Geriatrics, Turku City Hospital, University of Turku, Kunnallissairaalantie 20, 20700, Turku, Finland;HUS Diagnostic Center, Helsinki University Hospital, Hospital District of Helsinki and Uusimaa (HUS), Helsinki, Finland;Diagnostic Center, Clinical Chemistry and Hematology, Helsinki University Hospital and University of Helsinki, Helsinki, Finland;Municipality of Lieto, Health Care Center, 21420, Lieto, Finland;Faculty of Medicine, Department of Geriatrics, Turku City Hospital, University of Turku, Kunnallissairaalantie 20, 20700, Turku, Finland;Turku Bioscience Centre, University of Turku and Åbo Akademi University, 20520, Turku, Finland;Turku Bioscience Centre, University of Turku and Åbo Akademi University, 20520, Turku, Finland;Institute of Biomedicine, University of Turku, Turku, Finland;

关键词: Institutionalization;    Mortality;    Aged;    Index;   

DOI  :  10.1186/s12877-023-03760-1

 received in 2022-01-19, accepted in 2023-01-13,  发布年份 2023

来源: Springer

【 摘 要 】

BackgroundVarious indexes have been developed to estimate the risk for mortality, institutionalization, and other adverse outcomes for older people. Most indexes are based on a large number of clinical or laboratory parameters. An index based on only a few parameters would be more practical to use in every-day clinical practice. Our aim was to create an index to predict the risk for mortality and institutionalization with as few parameters as possible without compromising their predictive ability.MethodsA prospective study with a 10-year follow-up period. Thirty-six clinical and fourteen laboratory parameters were combined to form an index. Cox regression model was used to analyze the association of the index with institutionalization and mortality. A backward statistical method was used to reduce the number of parameters to form an easy-to-use index for predicting institutionalization and mortality.ResultsThe mean age of the participants (n = 1172) was 73.1 (SD 6.6, range 64‒97) years. Altogether, 149 (14%) subjects were institutionalized, and 413 (35%) subjects deceased during the follow-up. Institutionalization and mortality rates increased as index scores increased both for the large 50-parameter combined index and for the reduced indexes. After a backward variable selection in the Cox regression model, three clinical parameters remained in the index to predict institutionalization and six clinical and three laboratory parameters in the index to predict mortality. The reduced indexes showed a slightly better predictive value for both institutionalization and mortality compared to the full index.ConclusionsA large index with fifty parameters included many unimportant parameters that did not increase its predictive value, and therefore could be replaced with a reduced index with only a few carefully chosen parameters, that were individually associated with institutionalization or death.

【 授权许可】

CC BY   
© The Author(s) 2023

【 图 表 】

Fig. 3

Fig. 1

Fig. 4