Skip to main content
The British Journal of General Practice logoLink to The British Journal of General Practice
letter
. 2022 Sep 2;72(722):420–421. doi: 10.3399/bjgp22X720521

Bigger practices are associated with decreased patient satisfaction and perceptions of access

Peter J Edwards 1
PMCID: PMC9423067

All letters are subject to editing and may be shortened. General letters can be sent to bjgpdisc@rcgp.org.uk (please include your postal address for publication), and letters responding directly to BJGP articles can be submitted online via eLetters. We regret we cannot notify authors regarding publication.

For submission instructions visit: bjgp.org/letters

The number of GP practices continues to decline year-on-year, while the population grows.1 This has led to an increasing GP list size, and some practices have merged or been taken over by larger groups. In 2016, the Care Quality Commission chief inspector said the days of single-handed GPs are over.2

A colleague and I were recently discussing patient access and our personal experience as service users had become worse after our practices either merged or were taken over by a group. This made me question, ‘Is practice list size associated with patient satisfaction or perceptions of access?’

A seminal paper in 1995 concluded that patients preferred smaller practices,3 but was this still reflected in more recent data? I downloaded the 2022 GP Patient Survey (GPPS) results and the number of patients registered at practices.4,5 I excluded small practices (list size of <1000) and those where data were not available for the following questions: 1) ‘Overall, how would you describe your experience of making an appointment?’ and 2) ’Overall, how would you describe your experience of your GP practice?’

Data were analysed using Stata (version 17.0) and my code is publicly available.6 The final analysis included 6389 practices with a mean list size of 9551 (standard deviation 6394, range 1022 to 108 789).

Multilevel mixed-effects logistic regression was used to model the likelihood of patients reporting a ‘good’ (very good or fairly good) experience to the aforementioned GPPS questions against the number of patients registered at a practice. Adjustments were made for the number of patients per GP and factors reported to impact on patient satisfaction (Table 1).7

Table 1.

Patients’ overall experience of their practice and making an appointment by practice list size

List sizea Overall, how would you describe your experience of making an appointment?
Freq (n) Mean % reporting ‘good’b SD of % reporting ‘good’b Odds ratioc 95% CI P value
1000 to 9999 4019 63.3 15.7 1
10 000 to 19 999 2052 56.4 14.4 0.68 0.66 to 0.71 <0.001
20 000 to 29 999 238 53.0 14.1 0.59 0.54 to 0.64 <0.001
30 000 to 39 999 47 50.0 14.9 0.53 0.44 to 0.64 <0.001
40 000 to 49 999 17 49.2 15.3 0.48 0.35 to 0.66 <0.001
50 000 to 59 999 9 45.1 14.0 0.45 0.29 to 0.70 <0.001
≥60 000 7 48.9 14.9 0.64 0.39 to 1.05 0.08
Wald test across groups <0.001
List sizea Overall, how would you describe your experience of your GP practice?
Freq (n) Mean % reporting ‘good’b SD of % reporting ‘good’b Odds ratioc 95% CI P value
1000 to 9999 4019 77.7 12.1 1
10 000 to 19 999 2052 74.8 11.7 0.76 0.74 to 0.79 <0.001
20 000 to 29 999 238 72.1 11.4 0.65 0.60 to 0.71 <0.001
30 000 to 39 999 47 69.8 11.4 0.59 0.49 to 0.71 <0.001
40 000 to 49 999 17 68.9 15.3 0.54 0.40 to 0.74 <0.001
50 000 to 59 999 9 66.7 11.6 0.53 0.35 to 0.81 0.003
≥60 000 7 65.9 7.9 0.63 0.39 to 1.02 0.063
Wald test across groups <0.001
a

Practices with list size <1000, with missing data, or that reported the number of fully qualified full-time equivalent GPs as 0 were excluded (n = 118 from 2022 GPPS results).

b

’Good’ = very good or fairly good in 2022 GPPS.

c

Odds ratios were generated using multilevel mixed-effects logistic regression clustered by GP practice (n = 6389 practices). Adjusted on practice level for 2022 GPPS results for age (% aged ≥75 years), sex (% female), ethnicity (% white), presence of long-term condition (% yes), deprivation (using practice 2019 English Indices of Deprivation),8 and the number of patients per full-time equivalent fully qualified GP on 31 December 2021.9 CI = confidence interval. GPPS = GP Patient Survey. OR = odds ratio, SD = standard deviation.

Increasing practice list size was negatively associated with the likelihood of patients reporting a ‘good’ experience of making an appointment and overall with their practice (odds ratio [OR] 0.85, P<0.001 and OR 0.89, P<0.001 per increase of 5000 patients, respectively). Table 1 demonstrates this trend when practices were analysed categorically by list size.

Of course, association does not mean causation. Nevertheless, more research is needed to further assess the impacts of growing list size and practice mergers, as this analysis and my lived experience suggests bigger is not always better.

REFERENCES


Articles from The British Journal of General Practice are provided here courtesy of Royal College of General Practitioners

RESOURCES