Long term condition morbidity in English general practice: a cross-sectional study using three composite morbidity measures

TY - JOUR

T1 - Long term condition morbidity in English general practice

T2 - a cross-sectional study using three composite morbidity measures

AU - Weston, Charlotte

AU - Gilkes, Alexander

AU - Durbaba, Stevo

AU - Schofield, Peter

AU - White, Patrick

AU - Ashworth, Mark

PY - 2016/11/29

Y1 - 2016/11/29

N2 - Background: The burden of morbidity represented by patients with long term conditions (LTCs) varies substantially between general practices. This study aimed to determine the characteristics of general practices with high morbidity burden. Method: Retrospective cross-sectional study; general practices in England, 2014/15. Three composite morbidity measures (MMs) were constructed to quantify LTC morbidity at practice level: a count of LTCs derived from the 20 LTCs included in the UK Quality and Outcomes Framework (QOF) disease registers, expressed as 'number of QOF LTCs per 100 registered patients'; the % of patients with one or more QOF LTCs; the % of patients with one or more of 15 broadly defined LTCs included in the GP Patient Survey (GPPS). Determinants of MM scores were analysed using multi-level regression models. Analysis was based on a national dataset of English general practices (n = 7779 practices); GPPS responses (n = 903,357); general practice characteristics (e.g. list size, list size per full time GP); patient demographic characteristics (age, deprivation status); secondary care utilisation (out-patient, emergency department, emergency admission rates). Results: Mean MM scores (95% CIs) were: 57.7 (±22.3) QOF LTCs per 100 registered patients; 22.8% (±8.2) patients with a QOF LTC; 63.5% (±11.7) patients with a GPPS LTC. The proportion of elderly patients and social deprivation scores were the strongest predictors of each MM score; scores were largely independent of practice characteristics. MM scores were positive predictors of secondary care utilization and negative predictors' access, continuity of care and overall satisfaction. Conclusions: Wide variation in LTC morbidity burden was observed across English general practice. Variation was determined by demographic factors rather than practice characteristics. Higher rates of secondary care utilisation in practices with higher morbidity burden have implications for resource allocation and commissioning budgets; lower reported satisfaction in these practices suggests that practices may struggle with increased workload. There is a need for a readily available metric to define the burden of morbidity and multimorbidity in general practice.

AB - Background: The burden of morbidity represented by patients with long term conditions (LTCs) varies substantially between general practices. This study aimed to determine the characteristics of general practices with high morbidity burden. Method: Retrospective cross-sectional study; general practices in England, 2014/15. Three composite morbidity measures (MMs) were constructed to quantify LTC morbidity at practice level: a count of LTCs derived from the 20 LTCs included in the UK Quality and Outcomes Framework (QOF) disease registers, expressed as 'number of QOF LTCs per 100 registered patients'; the % of patients with one or more QOF LTCs; the % of patients with one or more of 15 broadly defined LTCs included in the GP Patient Survey (GPPS). Determinants of MM scores were analysed using multi-level regression models. Analysis was based on a national dataset of English general practices (n = 7779 practices); GPPS responses (n = 903,357); general practice characteristics (e.g. list size, list size per full time GP); patient demographic characteristics (age, deprivation status); secondary care utilisation (out-patient, emergency department, emergency admission rates). Results: Mean MM scores (95% CIs) were: 57.7 (±22.3) QOF LTCs per 100 registered patients; 22.8% (±8.2) patients with a QOF LTC; 63.5% (±11.7) patients with a GPPS LTC. The proportion of elderly patients and social deprivation scores were the strongest predictors of each MM score; scores were largely independent of practice characteristics. MM scores were positive predictors of secondary care utilization and negative predictors' access, continuity of care and overall satisfaction. Conclusions: Wide variation in LTC morbidity burden was observed across English general practice. Variation was determined by demographic factors rather than practice characteristics. Higher rates of secondary care utilisation in practices with higher morbidity burden have implications for resource allocation and commissioning budgets; lower reported satisfaction in these practices suggests that practices may struggle with increased workload. There is a need for a readily available metric to define the burden of morbidity and multimorbidity in general practice.

KW - Long term conditions

KW - Multimorbidity

KW - Patient experience

KW - Primary care

KW - Secondary care utilisation

UR - http://www.scopus.com/inward/record.url?scp=84999156683&partnerID=8YFLogxK

U2 - 10.1186/s12875-016-0563-3

DO - 10.1186/s12875-016-0563-3

M3 - Article

AN - SCOPUS:84999156683

SN - 1471-2296

VL - 17

SP - 1

EP - 7

JO - BMC FAMILY PRACTICE

JF - BMC FAMILY PRACTICE

IS - 1

ER -