School for Primary Care Research Increasing the evidence base for primary care practice The School for Primary Care Research is a partnership between the Universities of Birmingham, Bristol, Keele, Manchester, Nottingham, Oxford, Southampton and UCL and is part of the National Institute for Health Research. School for Primary Care Research Exploration of the relationship between multimorbidity and use of primary care resources in the UK Sam Brilleman1, Sandra Hollinghurst1, Sarah Purdy1, Hugh Gravelle3, Frank Windmeijer2, Chris Salisbury1 1 School of Social and Community Based Medicine, University of Bristol on behalf of the National School for Primary Care Research 2 Department of Economics, University of Bristol 3 University of York, and linked to Manchester via NPCRDC 

School for Primary Care Research Increasing the evidence base for primary care practice The School for Primary Care Research is a partnership between the Universities of Birmingham, Bristol, Keele, Manchester, Nottingham, Oxford, Southampton and UCL and is part of the National Institute for Health Research. School for Primary Care Research Background • Increasing number of individuals living with multimorbidity • Long-term chronic illness managed predominantly in primary care setting • May be useful in budget setting • Previous work: USA/Secondary care 

School for Primary Care Research Increasing the evidence base for primary care practice The School for Primary Care Research is a partnership between the Universities of Birmingham, Bristol, Keele, Manchester, Nottingham, Oxford, Southampton and UCL and is part of the National Institute for Health Research. School for Primary Care Research Aim • To explore the relationship between multimorbidity and the use of primary care resources in the UK • Objectives: – How do different measures compare when used to predict primary care costs? – What is the nature of the relationship between multimorbidity and cost? 

School for Primary Care Research Increasing the evidence base for primary care practice The School for Primary Care Research is a partnership between the Universities of Birmingham, Bristol, Keele, Manchester, Nottingham, Oxford, Southampton and UCL and is part of the National Institute for Health Research. School for Primary Care Research Data • General Practice Research Database (GPRD) – 86,100 patients from 174 practices – Diagnoses up until 31st March 2007 • Used to measure multimorbidity – Resource use data for one year beginning on 1st April 2007 • Used to estimate cost 

School for Primary Care Research Increasing the evidence base for primary care practice The School for Primary Care Research is a partnership between the Universities of Birmingham, Bristol, Keele, Manchester, Nottingham, Oxford, Southampton and UCL and is part of the National Institute for Health Research. School for Primary Care Research Diagnoses recorded by GP Diagnoses grouped into specific chronic diseases Simple count of 17 chronic diseases included in QOF 17 diseases weighted (by association with mortality) Sum of disease weights All diagnoses input to ACG software Diagnoses grouped based on disease area (EDCs) Simple count of 114 (of 264) disease groups (EDCs) defined as chronic Diagnoses grouped based on expected resource need (ADGs) Individuals allocated to mutually exclusive categories (ACGs) based on age, gender and combination of ADGs QOF disease count Charlson Index score EDC count Mutually exclusive ACG categories 

School for Primary Care Research Increasing the evidence base for primary care practice The School for Primary Care Research is a partnership between the Universities of Birmingham, Bristol, Keele, Manchester, Nottingham, Oxford, Southampton and UCL and is part of the National Institute for Health Research. School for Primary Care Research Diagnoses recorded by GP Diagnoses grouped into specific chronic diseases Simple count of 17 chronic diseases included in QOF 17 diseases weighted (by association with mortality) Sum of disease weights All diagnoses input to ACG software Diagnoses grouped based on disease area (EDCs) Simple count of 114 (of 264) disease groups (EDCs) defined as chronic Diagnoses grouped based on expected resource need (ADGs) Individuals allocated to mutually exclusive categories (ACGs) based on age, gender and combination of ADGs QOF disease count Charlson Index score EDC count Mutually exclusive ACG categories 

School for Primary Care Research Increasing the evidence base for primary care practice The School for Primary Care Research is a partnership between the Universities of Birmingham, Bristol, Keele, Manchester, Nottingham, Oxford, Southampton and UCL and is part of the National Institute for Health Research. School for Primary Care Research Diagnoses recorded by GP Diagnoses grouped into specific chronic diseases Simple count of 17 chronic diseases included in QOF 17 diseases weighted (by association with mortality) Sum of disease weights All diagnoses input to ACG software Diagnoses grouped based on disease area (EDCs) Simple count of 114 (of 264) disease groups (EDCs) defined as chronic Diagnoses grouped based on expected resource need (ADGs) Individuals allocated to mutually exclusive categories (ACGs) based on age, gender and combination of ADGs QOF disease count Charlson Index score EDC count Mutually exclusive ACG categories 

School for Primary Care Research Increasing the evidence base for primary care practice The School for Primary Care Research is a partnership between the Universities of Birmingham, Bristol, Keele, Manchester, Nottingham, Oxford, Southampton and UCL and is part of the National Institute for Health Research. School for Primary Care Research Diagnoses recorded by GP Diagnoses grouped into specific chronic diseases Simple count of 17 chronic diseases included in QOF 17 diseases weighted (by association with mortality) Sum of disease weights All diagnoses input to ACG software Diagnoses grouped based on disease area (EDCs) Simple count of 114 (of 264) disease groups (EDCs) defined as chronic Diagnoses grouped based on expected resource need (ADGs) Individuals allocated to mutually exclusive categories (ACGs) based on age, gender and combination of ADGs QOF disease count Charlson Index score EDC count Mutually exclusive ACG categories 

School for Primary Care Research Increasing the evidence base for primary care practice The School for Primary Care Research is a partnership between the Universities of Birmingham, Bristol, Keele, Manchester, Nottingham, Oxford, Southampton and UCL and is part of the National Institute for Health Research. School for Primary Care Research Diagnoses recorded by GP Diagnoses grouped into specific chronic diseases Simple count of 17 chronic diseases included in QOF 17 diseases weighted (by association with mortality) Sum of disease weights All diagnoses input to ACG software Diagnoses grouped based on disease area (EDCs) Simple count of 114 (of 264) disease groups (EDCs) defined as chronic Diagnoses grouped based on expected resource need (ADGs) Individuals allocated to mutually exclusive categories (ACGs) based on age, gender and combination of ADGs QOF disease count Charlson Index score EDC count Mutually exclusive ACG categories 

School for Primary Care Research Increasing the evidence base for primary care practice The School for Primary Care Research is a partnership between the Universities of Birmingham, Bristol, Keele, Manchester, Nottingham, Oxford, Southampton and UCL and is part of the National Institute for Health Research. School for Primary Care Research Diagnoses recorded by GP Diagnoses grouped into specific chronic diseases Simple count of 17 chronic diseases included in QOF 17 diseases weighted (by association with mortality) Sum of disease weights All diagnoses input to ACG software Diagnoses grouped based on disease area (EDCs) Simple count of 114 (of 264) disease groups (EDCs) defined as chronic Diagnoses grouped based on expected resource need (ADGs) Individuals allocated to mutually exclusive categories (ACGs) based on age, gender and combination of ADGs QOF disease count Charlson Index score EDC count Mutually exclusive ACG categories 

School for Primary Care Research Increasing the evidence base for primary care practice The School for Primary Care Research is a partnership between the Universities of Birmingham, Bristol, Keele, Manchester, Nottingham, Oxford, Southampton and UCL and is part of the National Institute for Health Research. School for Primary Care Research 0 10 20 30 40 50 60 0 1 2 3 4 5 6+ QOF chronic disease count 0 10 20 30 40 50 60 70 Percentage of sample (%) 0 1 2 3 4 5 6 7+ Charlson Index score 0 5 10 15 20 0 1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 1718+ Expanded Diagnostic Cluster (EDC) count 0 2 4 6 8 10 12 Percentage of sample (%) 1731 5110 5200 1000 600 1600 1100 500 400 300 1300 1711 2400 1712 700 2800 1800 4000 2100 3900 2200 1721 1732 1722 3400 2500 3200 1500 3300 1751 1752 4710 1742 4310 1741 1200 4320 3500 4720 800 900 4810 3700 4820 2300 1762 4100 4330 1761 4410 2600 1400 2700 3600 4830 4730 1772 1771 4420 4910 4430 5040 4920 4930 4940 5050 5060 5070 Adjusted Clinical Group (ACG) Distribution of each measure of multimorbidity 

School for Primary Care Research Increasing the evidence base for primary care practice The School for Primary Care Research is a partnership between the Universities of Birmingham, Bristol, Keele, Manchester, Nottingham, Oxford, Southampton and UCL and is part of the National Institute for Health Research. School for Primary Care Research Costing Resources Total annual £ per patient Consultations Prescription drugs Tests and investigations 

School for Primary Care Research Increasing the evidence base for primary care practice The School for Primary Care Research is a partnership between the Universities of Birmingham, Bristol, Keele, Manchester, Nottingham, Oxford, Southampton and UCL and is part of the National Institute for Health Research. School for Primary Care Research Costing Resources Consultations Combination of… -Consultation type (surgery consultation, home visit, clinic, out of hours, telephone) and… - Primary staff role (GP, practice nurse, district nurse, physiotherapist, counsellor, etc.) Additional cost for administrative activities performed by a receptionist, administrator, or secretary 

School for Primary Care Research Increasing the evidence base for primary care practice The School for Primary Care Research is a partnership between the Universities of Birmingham, Bristol, Keele, Manchester, Nottingham, Oxford, Southampton and UCL and is part of the National Institute for Health Research. School for Primary Care Research Costing Resources Prescription drugs - Matched to cost using drug name, strength and formulation - Cost based on total quantity prescribed (eg. number of tablets, number of millilitres for liquids) 

School for Primary Care Research Increasing the evidence base for primary care practice The School for Primary Care Research is a partnership between the Universities of Birmingham, Bristol, Keele, Manchester, Nottingham, Oxford, Southampton and UCL and is part of the National Institute for Health Research. School for Primary Care Research Costing Resources Tests and investigations In-practice (during consultation): cost for consumables only (e.g. pregnancy tests or urine dipstick) Laboratory tests: cost by pathology discipline (source: NHS Reference Costs) Hospital based investigations: costed individually (source: NHS Reference Costs) 

School for Primary Care Research Increasing the evidence base for primary care practice The School for Primary Care Research is a partnership between the Universities of Birmingham, Bristol, Keele, Manchester, Nottingham, Oxford, Southampton and UCL and is part of the National Institute for Health Research. School for Primary Care Research Costing Resources Total annual £ per patient Consultations Prescription drugs Tests and investigations 

School for Primary Care Research Increasing the evidence base for primary care practice The School for Primary Care Research is a partnership between the Universities of Birmingham, Bristol, Keele, Manchester, Nottingham, Oxford, Southampton and UCL and is part of the National Institute for Health Research. School for Primary Care Research Patient cost, by age category and deprivation decile 0 200 400 600 800 20-29 30-39 40-49 50-59 60-69 70-79 80-89 90+ Age category 0 200 400 600 800 Mean annual patient cost (£) 1 2 3 4 5 6 7 8 9 10 Deprivation decile (10 most deprived) 

School for Primary Care Research Increasing the evidence base for primary care practice The School for Primary Care Research is a partnership between the Universities of Birmingham, Bristol, Keele, Manchester, Nottingham, Oxford, Southampton and UCL and is part of the National Institute for Health Research. School for Primary Care Research How well does each multimorbidity measure predict patient cost? Model Deviance-based R-squared Age, gender, deprivation, and practice + Charlson Index score + QOF disease count + ACG categories + EDC count 

School for Primary Care Research Increasing the evidence base for primary care practice The School for Primary Care Research is a partnership between the Universities of Birmingham, Bristol, Keele, Manchester, Nottingham, Oxford, Southampton and UCL and is part of the National Institute for Health Research. School for Primary Care Research How well does each multimorbidity measure predict patient cost? Model Deviance-based R-squared Age, gender, deprivation, and practice 14% + Charlson Index score 21% + QOF disease count 27% + ACG categories 29% + EDC count 31% 

School for Primary Care Research Increasing the evidence base for primary care practice The School for Primary Care Research is a partnership between the Universities of Birmingham, Bristol, Keele, Manchester, Nottingham, Oxford, Southampton and UCL and is part of the National Institute for Health Research. School for Primary Care Research How well does each multimorbidity measure predict patient cost? Model Deviance-based R-squared Age, gender, deprivation, and practice 14% + Charlson Index score 21% + QOF disease count 27% + ACG categories 29% + EDC count 31% 

School for Primary Care Research Increasing the evidence base for primary care practice The School for Primary Care Research is a partnership between the Universities of Birmingham, Bristol, Keele, Manchester, Nottingham, Oxford, Southampton and UCL and is part of the National Institute for Health Research. School for Primary Care Research Predicted mean annual patient cost + 0 500 1000 1500 2000 2500 0 1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 Predicted mean annual patient cost (£) EDC count 

School for Primary Care Research Increasing the evidence base for primary care practice The School for Primary Care Research is a partnership between the Universities of Birmingham, Bristol, Keele, Manchester, Nottingham, Oxford, Southampton and UCL and is part of the National Institute for Health Research. School for Primary Care Research Conclusions • Including measures of multimorbidity improved model fit • A count of Expanded Diagnostic Clusters (EDCs) performed best in a model predicting patient cost in primary care • Increases in patient cost were roughly proportional to the number of EDCs 

School for Primary Care Research Increasing the evidence base for primary care practice The School for Primary Care Research is a partnership between the Universities of Birmingham, Bristol, Keele, Manchester, Nottingham, Oxford, Southampton and UCL and is part of the National Institute for Health Research. School for Primary Care Research Thank you This project was funded by the NIHR School for Primary Care Research 

School for Primary Care Research Increasing the evidence base for primary care practice The School for Primary Care Research is a partnership between the Universities of Birmingham, Bristol, Keele, Manchester, Nottingham, Oxford, Southampton and UCL and is part of the National Institute for Health Research. School for Primary Care Research 

School for Primary Care Research Increasing the evidence base for primary care practice The School for Primary Care Research is a partnership between the Universities of Birmingham, Bristol, Keele, Manchester, Nottingham, Oxford, Southampton and UCL and is part of the National Institute for Health Research. School for Primary Care Research References (1) • QOF/EDC count “chronic” classifications: – Salisbury, C., Johnson, L., Purdy, S., Valderas, J. M., & Montgomery, A. A. 2011, "Epidemiology and impact of multimorbidity in primary care: a retrospective cohort study", British Journal of General Practice, vol. 61, no. 582, pp. 18-24. • Charlson Index score: – Khan, N. F., Perera, R., Harper, S., & Rose, P. W. 2010, "Adaptation and validation of the Charlson Index for Read/OXMIS coded databases", BMC Family Practice, vol. 11. – Deyo, R. A., Cherkin, D. C., & Ciol, M. A. 1992, "Adapting A Clinical Comorbidity Index for Use with Icd-9-Cm Administrative Databases", Journal of Clinical Epidemiology, vol. 45, no. 6, pp. 613-619. – Charlson, M. E., Pompei, P., Ales, K. L., & Mackenzie, C. R. 1987, "A New Method of Classifying Prognostic Co-Morbidity in Longitudinal- Studies - Development and Validation", Journal of Chronic Diseases, vol. 40, no. 5, pp. 373-383. 

School for Primary Care Research Increasing the evidence base for primary care practice The School for Primary Care Research is a partnership between the Universities of Birmingham, Bristol, Keele, Manchester, Nottingham, Oxford, Southampton and UCL and is part of the National Institute for Health Research. School for Primary Care Research References (2) • John Hopkins ACG System: – Johns Hopkins Bloomberg School of Public Health. The Johns Hopkins ACG® Case-mix System Version 8.2. Baltimore, 2008. • Deviance-based R-squared: – Cameron, A. C. & Windmeijer, F. A. G. 1997, "An R-squared measure of goodness of fit for some common nonlinear regression models", Journal of Econometrics, vol. 77, no. 2, pp. 329-342.