Is there an association between participation in extended hours and performance outcomes in English general practices?

Talk Code: 
Louis Levene
Richard Baker, John Bankart, Kamlesh Khunti, Nicola Walker, Christopher Williams
Author institutions: 
University of Leicester


The General Medical Services contract now includes a directed enhanced service for extended access. There is debate as to how effective this has been in improving health outcomes. Current evidence suggests that increased extended hours access has a limited association with three patient experience measures. Our research question tested the following null hypotheses: variations in levels of extended access do not predict variations in three performance-related outcomes (controlled blood pressure in patients with hypertension, satisfactory glycaemic control in patients with diabetes, and having and seeing a preferred GP), after adjusting for population and organisational characteristics.


Cross-sectional study of general practices in England in 2017-2018, with three dependent variables:1. % patients with hypertension with last blood pressure reading (measured in the preceding 12 months) <150/90 mmHg (QOF HYP001)2. % patients with diabetes with last IFCC-HbA1c is <59 mmol/mol (7.5%) in the preceding 12 months (QOF DM007)3. % patients who have and are able to see a preferred GP in January-March 2018 (GPPS Q8x9)The main independent variable was number of extended access days per week (0-7) in March 2018 (from NHS Digital). We included 10 independent confounder variables: IMD 2015 score, % 75+ years (2016- latest), % Black ethnicity (2018), % South Asian ethnicity (2018), % on practice hypertension register (2018), % on practice diabetes register (2018), Geographical region, list size (2018), FTE GPs/1,000 patients (2016) and FTE practice nurses/1,000 patients (2016).We undertook descriptive statistics, univariable analyses (correlations), and linear regressions.


6,892 practices in England.11.7% offered 0 days’ and 40.9% offered 7 days’ extended access. The distribution of extended access was skewed. Spearman’s rho values were weakly negative (all p<0.05) for extended access with controlled hypertension (-0.029), satisfactory glycaemic control (-0.040) and continuity of care (-0.038).In regressions, levels of extended access did not independently predict variations in achievement for any dependent variable.


Increased extended access was not associated with improvements in three important performance indicators. Is the additional investment in extended access (£258 million in 2018-9) cost-effective? Further research is needed to examine associations between extended access and other outcomes.

Submitted by: 
Louis Levene
Funding acknowledgement: 
This study was carried out within an existing programme of research. There was no additional funding.