Development and validation of the Cambridge Multimorbidity Score.
Payne, Rupert A
Mendonca, Silvia C
Elliott, Marc N
CMAJ : Canadian Medical Association journal = journal de l'Association medicale canadienne
Canadian Medical Association/Association Medical Canadienne
MetadataShow full item record
Payne, R. A., Mendonca, S. C., Elliott, M. N., Saunders, C., Edwards, D., Marshall, M., & Roland, M. (2020). Development and validation of the Cambridge Multimorbidity Score.. CMAJ : Canadian Medical Association journal = journal de l'Association medicale canadienne, 192 (5), E107-E114. https://doi.org/10.1503/cmaj.190757
Background Health services have failed to respond to the pressures of multimorbidity. There is a need for improved measures of multimorbidity for research, planning services and resource allocation. Methods Development (N=300,000) and validation (N=150,000) data samples were extracted from the UK Clinical Practice Research Database. We modelled the association between 37 morbidities and key outcomes (primary care consultations, unplanned hospitalization, mortality) at one and five years using Cox and zero-inflated negative binomial regression. A general-outcome multimorbidity score was constructed by averaging the standardised weights of the separate outcome scores. Performance was compared with the Charlson co-morbidity index. Results Models including all 37 conditions were acceptable predictors of GP consultations, hospitalisation and mortality at one-year (C-indices 0.732 [95% confidence interval 0.731-0.734], 0.742 [0.737-0.747] and 0.912 [0.905-0.918] respectively, adjusted for age/gender). Reducing the models to the 20 conditions which had the greatest combined prevalence/weight made little difference to the predictive value of the models (C-indices 0.727 [0.725-0.728], 0.738 [0.732-0.743] and 0.910 [0.904-0.917] respectively). Prediction of outcomes at five years for the 20-condition model remained similar for consultations and mortality (C-indices 0.735 [0.734-0.736], 0.889 [0.885-0.892]) but performed less well for admissions (C-index 0.708 [0.705-0.712]). The general-outcome score performed similarly to the outcome-specific models. Models performed significantly better than those based on Charlson. Conclusions This analysis provides several robust, simple-to-use multimorbidity scores, both tailored and not tailored to specific health outcomes. The scores will be valuable to those planning clinical services, policymakers allocating resources, and researchers seeking to account for the effect of multimorbidity.
Humans, Patient Admission, Mortality, Proportional Hazards Models, Retrospective Studies, Predictive Value of Tests, Adult, Aged, Aged, 80 and over, Middle Aged, Referral and Consultation, Primary Health Care, Female, Male, Young Adult, Electronic Health Records, United Kingdom, Multimorbidity, Outcome and Process Assessment, Health Care
This paper presents independent research funded by the National Institute for Health Research School for Primary Care Research (NIHR SPCR, ref. FR10/283). The views expressed are those of the authors and not necessarily those of the NHS, the NIHR or the Department of Health.
National Institute for Health Research (NIHR) (via University of Oxford) (SPCR-2014-10043 grant 283)
External DOI: https://doi.org/10.1503/cmaj.190757
This record's URL: https://www.repository.cam.ac.uk/handle/1810/299632
All rights reserved