Development of a modified Cambridge Multimorbidity Score for use with SNOMED CT: an observational English primary care sentinel network study

Abstract

Background: People with multiple health conditions are more likely to have poorer health outcomes and greater care and service needs; a reliable measure of multimorbidity would inform management strategies and resource allocation. This study aims to develop and validate a modified version of the Cambridge Multimorbidity Score in an extended age range, using clinical terms which are routinely used in electronic health records across the world (SNOMED CT). Methods and Findings: We curated new variables describing 37 health conditions and modelled the associations between these and 1-year mortality risk using the Cox proportional hazard model in a development dataset (n=300,000). We then developed two simplified models – a 20-condition model as per the original Cambridge Multimorbidity Score, and a variable reduction model using backward elimination with Akaike information criterion as the stopping criterion. The results were compared and validated for 1-year mortality in a synchronous validation dataset (n=150,000), and for 1-year and 5-year mortality in an asynchronous validation dataset (n=150,000). Our final variable reduction model retained 21 conditions, and the conditions mostly overlapped with those in the 20-condition model. The model performed similarly to the 37- and 20-condition models, showing high discrimination and good calibration following recalibration. Conclusions: This modified version of the Cambridge Multimorbidity Score allows reliable estimation using clinical terms which can be applied internationally across multiple healthcare settings.

Postpandemic Sentinel Surveillance of Respiratory Diseases in the Context of the World Health Organization Mosaic Framework: Protocol for a Development and Evaluation Study Involving the English Primary Care Network 2023-2024

“…We used these data to derive the Cambridge Multimorbidity Score, a single measure of multimorbidity for all adults, and the electronic frailty index (eFI). While the eFI can be used from the age of 50 years, we will use it in the coming season for people aged 65 years or older [ 52 , 53 ].…”

Implementation and external validation of the Cambridge Multimorbidity Score in the UK Biobank cohort

“…A recently published study by Tsang et al [ 16 ] developed a modified version of the CMS (developing their own score using the same 37 pre-existing conditions) using data from the English primary care sentinel surveillance network, but did not implement or validate the original CMS developed by Payne et al [ 9 ]. Their new model had excellent discrimination for the outcome of death for both short term (C-index of 0.92 at 1 year) and long term (C-index of 0.91 at 5 years) follow-up in internal validation (also using the sentinel surveillance network cohort), which is in line with the performance of the CMS in its development cohort, but higher than the results we have seen in this external validation of the CMS in UK Biobank.…”

“…As all the primary care data linked to the UK Biobank cohort is from before 2018, we have not developed or used codelists based on the SNOMED framework. However, SNOMED codelists for the 37 conditions used in the CMS have recently been published [ 16 ], providing a complimentary resource to (ii).…”

Determining the Impact of Opioid Policy on Substance Use and Mental Health–Related Harms: Protocol for a Data Linkage Study

“…A modified Cambridge Multimorbidity Score (CMS) will be used to categorize comorbidities using SNOMED codes in the POLAR data [ 39 ]. There have been a number of approaches to measuring comorbidity, with the modified CMS recently validated for use in primary care, offering an advantage over the Cambridge Comorbidity Index, which, while widely used, was developed, and valued for use in secondary hospital settings.…”

“…The Cox proportional hazards models will be used to compare time-to-event, adjusting for covariates. These covariates included demographics (eg, age, gender, and socioeconomic status), characteristics of opioids prescribed, chronic pain diagnoses recorded (eg, musculoskeletal, neuropathic, migraine, and headache), polypharmacy (defined as having 5 or more different medications daily [ 43 ]), comorbidities (as measured by the modified CMS [ 39 ]), and ED presentations during the trajectory measurement period.…”