Table 6 Results of Practice level regression modelling, final models^a, N = 208 practices with response rates at least 50% in staff survey

Mean practice TCI 10 item score

1 Worst

to

5 Best

4.3155857

- 0.0597451*(practice size in thousands) {−0.0867164, −0.0327739}

 + 0.0015787*(practice size in thousands)² {0.0005114, 0.0026460}

- 0.0000113*(practice size in thousands)³ {−0.0000208, −0.0000018}

0.145

Going from 25,000 to 24,000 patients→ Team Climate improvement of 0.003

Going from 20,000 to 19,000 patients → Team Climate improvement of 0.011

Going from 15,000 to 14,000 patients → Team Climate improvement of 0.021

Going from 10,000 to 9000 patients → Team Climate improvement of 0.033

Going from 5000 to 4000 patients → Team Climate improvement of 0.046

Median practice job satisfaction

1 Extremely dissatisfied

to

7 Extremely satisfied

2.558481

 + 0.838210*(mean practice 10 item TCI) {0.680785, 0.995636}

 + 0.021088*(practice size in thousands) {0.006326, 0.035849}

- 0.000271*(practice size in thousands)² {-0.000478, -0.000065}

- 0.008327*(clinical FTE % that is GP) {−0.012114, −0.004539}

 + 0.543252*(Total permanent GP FTE per 1000 patients) {0.181725, 0.904779}

0.380

Increasing the mean TCI score by 0.5 is associated with a 0.419 improvement in job satisfaction score

Going from 4000 to 5000 patients → Job satisfaction improvement of 0.019

Going from 9000 to 10,000 patients → Job satisfaction improvement of 0.016

Going from 14,000 to 15,000 patients → Job satisfaction improvement of 0.013

Going from 19,000 to 20,000 patients → Job satisfaction improvement of 0.011

Going from 24,000 to 25,000 patients → Job satisfaction improvement of 0.008

A 10% decrease in the percentage of clinical FTE that is GP is associated with a 0.08 increase in job satisfaction score

Adding an extra full time permanent GP in a practice with 10,000 patients is associated with a 0.054 improvement in job satisfaction score

Median practice intention to be working in practice in 2 years

1 Very unlikely

to 5 Very likely

1.203

 + 0.721*(mean practice 10 item TCI) {0.562, 0.880}

 + 0.179 if practice trains GPs {0.056, 0.302}

0.286

Increasing the mean practice TCI score by 0.5 points is associated with a 0.361 points improvement in the likelihood of working in the practice in 2 years score

Being a GP training practice improves likely retention score by 0.179 points

Median practice burnout

0 Never

to

6 Every day

6.604

– 0.982*(mean practice 10 item TCI) {−1.252, −0.712}

−0.004*(% of GP headcount qualified in UK) {−0.009, −0.001}

- 0.618*(Total permanent GP FTE per 1000 patients) {−1.222, −0.013}

0.234

Increasing the mean practice TCI score by 0.5 points is associated with a reduction in the burnout score by 0.491

A 10% increase in the percentage of GP head count qualified in the UK is associated with a 0.04 reduction in burnout

Adding an extra full time permanent GP in a practice with 10,000 patients is associated with a 0.06 reduction in burnout

Percentage of patients having a fairly or very good experience of the practice

N = 206 (No patient experience data were available for 2 practices)

100%

maximum

21.244

 + 7.835*(mean practice 10 item TCI) {4.074,11.595}

 + 0.328*(% of practice patients > = 65 years old) {0.110, 0.547}

 + 0.064*(% of GP FTE performed by female GPs) {0.005, 0.123}

 + 0.086*(% of GP headcount trained in UK) {0.027, 0.145}

 + 18.422*(Total permanent GP FTE per 1000 patients) {9.002, 27.842}

0.276

A 0.5 point improvement in mean practice TCI score is associated with a 3.92% increase in the proportion of patients reporting a fairly or very good experience of their practice

A 5% improvement in the proportion of patients reporting a positive experience is associated with:

- an increase of 0.271 permanent GPs / 1000 patients;

- an increase of 78.1% in the % of GP FTE performed by female GPs;

- an increase of 58.14% in the % of GP headcount qualified in the UK;

- an increase of 15.24% in practice patients aged 65 or over