General practice across England: The number and type of appointments are related to patient satisfaction

General practice now delivers more appointments than it did before the pandemic

Staff in general practice carried out 367.4 million appointments in 2024 compared to 300 million in 2019, an increase of 22.5% in five years. The number of appointments delivered by each GP remained steady: 4,529 in 2019 and 4,499 in 2024. 

It is difficult to draw conclusions about the number of appointments delivered by each non-GP staff member due to the expansion of DPC staff since 2019. In all primary care, the number of appointments delivered by other staff members (nurses and DPC staff) rose from 136.8 million in 2019 to 200.0 million in 2024, an increase of 46.2%. But the workforce grew much more quickly than that, meaning that the number of appointments per other staff member fell from 4,791 in 2019 to 3,009 in 2024 (a drop of 37.2%).

This is likely because much of the activity of DPC staff is not captured in appointment data. For example, care co-ordinators – who make up 14.2% of the PCN DPC workforce – spend much of their time liaising between different parts of a multi-disciplinary team to ensure a patient receives the correct care. That activity does not count as an appointment. That means that we cannot draw conclusions about changes in the ‘productivity’ of those staff. 

The number of patients registered in general practice has also risen. There were 59.3 million registered patients in September 2018 and 63.7 million in December 2024 – an increase of 7.4%. 

Ultimately, though, what matters to patients is not the number of appointments that a staff member delivers, but rather the number of appointments that they themselves can access. In 2019, general practice delivered 5.0 appointments per patient. This rose to 5.8 in 2024 – an increase of 15.6%. The number of appointments per patient delivered by non-GP staff rose by 33.2%. The number of GP appointments per patient has not changed, remaining at 2.6. 

There could also be other factors driving this increase in appointments. There is evidence that some of the increase in remote appointments has been because of coding differences between systems that practices use to track and submit appointment data. ¹⁹ Wyatt S, ‘Long term trends in GP practice consultation rates’, Midlands Decision Support Network, 2024, p. 28, www.strategyunitwm.nhs.uk/sites/default/files/2024-09/1-Long-term-trends-in-GP-Practice-consultation-rates-MDSN-240220.pdf  For example, one of the appointment systems in this work – CPRD Aurum – treats electronic messages exchanged with patients as a ‘consultation’, while other systems do not. ²⁰ Wyatt S, ‘Long term trends in GP practice consultation rates’, Midlands Decision Support Network, 2024, p. 28, www.strategyunitwm.nhs.uk/sites/default/files/2024-09/1-Long-term-trends-in-GP-Practice-consultation-rates-MDSN-240220.pdf  This could artificially increase the number of appointments that general practice appears to conduct, and also reduce the apparent proportion of appointments that staff conducted face-to-face.

Patient satisfaction is higher in areas that deliver more appointments per patient 

There are lots of reasons why patients could value additional staff, including a greater range of care offered or that it is easier to speak to someone when calling their GP practice. But one reason could be that they provide more appointments. We can observe the number of appointments that were delivered in each practice in 2023/24 and whether a GP or another staff member delivered them (but cannot split those into DPC and nursing staff appointments).

From regression 1 we can see that only the number of GP appointments delivered in a practice has a stronglystatistically significant relationship with patient satisfaction. An additional 1,000 GP appointments in a practice is associated with patient satisfaction increasing by 0.14ppt. For context, the average number of GP appointments delivered per practice was 25,439 in 2023/24. 

Additional GP appointments are not as strongly associated with QOF achievement as they are with patient satisfaction 

The major difference between regression 1 and regression 6 is that, unlike patient satisfaction, there is only a weakly statistically significant relationship between QOF outcomes and the number of GP appointments that a practice delivers. There is, however, a far stronger relationship between the number of appointments delivered by other staff (nurses and DPC staff) and QOF outcomes. This is likely because other practice staff – particularly nurses – play a large role in managing patients’ ongoing health conditions ²⁵ Khan N and Peckham S, ‘Advanced Nurse Practitioner (ANPs) experiences of the Quality and Outcomes Framework (QOF) Scheme: a UK case study’, BMJ Open, 2024, vol. 14, https://bmjopen.bmj.com/content/14/11/e087492  and delivering clinics that are required to meet QOF targets. ²⁶ Roland M and Guthrie B, ‘Quality and Outcomes Framework: what have we learnt?’, British Medical Journal, 2016, vol. 354, https://pmc.ncbi.nlm.nih.gov/articles/PMC4975019  

The size of the effect is also stronger for appointments by other practice staff than GP appointments: 1,000 additional GP appointments increases QOF achievement by 0.02ppt while 1,000 additional appointments from other practice staff increases the same metric by 0.05ppt. This is a very small effect. Because of the way that achievement is grouped towards the upper end of the distribution, if the median practice increases its QOF achievement by 0.05ppt, it would move from the 50th percentile to the 50.1st percentile.

Carrying out more remote appointments means practices deliver more total appointments

We have speculated in previous work that rising proportions of remote appointments means that practices can conduct more appointments overall. But until recently there hasn’t been the practice-level appointment data which would make further analysis possible.

From regression 9, we see that there is a strongly statistically significant relationship between the proportion of appointments that were carried out face-to-face and the average number of appointments carried out by each clinical staff member. According to our central estimate, for each ppt reduction in the proportion of face-to-face appointments, staff carried out an average of 11.5 more appointments in 2023/24.

Regression 10 shows the impact on GP appointments per GP. Once again, the proportion of face-to-face appointments is strongly significant and a 1ppt reduction of that appointment mode was associated with GPs delivering 18.2 more appointments, compared to an average of 4,416 GP appointments conducted per GP in 2024.

This indicates that much of the increase in GP appointments between 2019 and 2024 can be attributed to a shift towards more remote appointments. The decline in the proportion of total appointments conducted face-to-face between 2019 and 2024 was 14.5ppt. GPs delivered 10.6 million more appointments in 2024 than 2019. If GPs followed the same trend as most of general practice, then the shift towards remote appointments would account for more than 90% of that change.

An increase in appointments driven by more remote consultations doesn’t necessarily mean that GPs are able to see more patients, however. There is evidence that a digital-first approach in fact increases GP workload – unless the consultations are kept particularly short, or if they do not often result in a follow-up face-to-face appointment. ²⁷ Salisbury C, Murphy M and Duncan P, ‘The Impact of Digital-First Consultations on Workload in General Practice: Modeling Study’, Journal of Medical Internet Research, 2020, vol. 22, no. 6, https://pmc.ncbi.nlm.nih.gov/articles/PMC7327596

Different staff groups deliver different numbers of appointments

We also explored the effect of an additional staff member on the number of appointments that a practice delivers.*The figures in Table 2 below are the results of a regression of total appointments on staffing at a practice level. As such, they do not measure directly the appointments delivered by each staff member, but the average increase in total appointments if a practice adds one additional staff member, holding other staff categories constant. In both regressions, every staff group was strongly significant.

Table 2 The effect of one extra staff member on appointment numbers per year, by appointment type and staff group

_{Notes: This shows the central estimate for each category and is in terms of FTE.}

It might be surprising that GP partners are associated with the largest increase in GP appointments. Alongside their clinical work, partners also have responsibility for managing their practices, which might suggest they would have less time to carry out appointments. But it is also possible that their responsibility for their practices is the cause of this, too: GP partners often carry out much of their administrative work outside of usual working hours, likely because they are either personally liable for the practice or are at least strongly incentivised to make sure that the practice is performing well.

Our finding that GP partners generate more additional appointments among non-GP staff (than do salaried GPs) also suggests that an additional GP partner improves the management capacity of a practice, leading to improved productivity of their colleagues.

At the other end of the spectrum, an additional GP trainee is not associated with as many appointments as their fully qualified colleagues. This is perhaps not surprising: trainees require supervision, meaning that they are not able to provide as many appointments as GP partners or salaried GPs.

More surprisingly, however, an additional DPC staff member adds less than half as many appointments as a salaried GP or a nurse on average. This may be because there is a wide range of roles covered by the DPC staff group – some of which will not include patient appointments. For example, there is evidence that some of the activity carried out by pharmacists is not captured by current appointment coding practices. ²⁸ Karampatakis G, Ryan K, Patel N, Lau W-M and Stretch G, ‘How do pharmacists in English general practices identify their impact? An exploratory qualitative study of measurement problems’, BMC Health Services Research, 2019, vol. 19, no. 34, https://bmchealthservres.biomedcentral.com/articles/10.1186/s12913-018-3842-y

_{* Please see regressions 3 and 4 in Appendix 1. In regression 3, the dependent variable is the number of total appointments delivered in general practice in 2023/24, with the number of GP partners, salaried GPs, GP trainees, DPC staff and nurses as the independent variables (all in terms of FTE). Regression 4 uses the number of GP appointments in 2023/24 as the dependent variable and the number of GP partners, salaried GPs and GP trainees as the independent variables.}

Patients are more satisfied with practices that provide more of their appointments face-to-face 

Alongside changes in staff mix, there has also been a substantial shift in how general practice delivers appointments. In 2019, more than four in five of appointments were delivered face-to-face (80.7%). The pandemic precipitated a rapid and radical shift in how general practice staff delivered appointments. In 2020/21, barely more than half of appointments took place face-to-face (53%). 

There has since been some reversion to in-person appointments, but the new equilibrium appears to be somewhat lower than in 2019: in 2024, staff delivered two thirds (66.2%) of appointments face-to-face and 30.0% either by telephone or online. GPs and other practice staff work in different ways. GPs conducted 59.9% of appointments face-to-face in 2024, compared to 73.9% for other practice staff. This is because it is harder for nurses or other practice staff to deliver remote appointments when their work may require things like taking blood. ³⁰ Institute for Government interview.

The shift has generated negative media attention. The Daily Mail even created a tool ranking practices on the proportion of their appointments that are face-to-face, labelling those that do fewest the ‘worst’ practices in the country.8 This is despite relatively widespread acceptance of the shift: in the Office for National Statistics’ 2020 ‘Opinions and lifestyle survey’ more than two thirds of respondents said they would be ‘comfortable’ or ‘very comfortable’ attending an online appointment (68%), though the number was lower for those aged 70 and over (61%). ³² Morris J, ‘The remote care revolution in the NHS: understanding impacts and attitudes’, blog, Nuffield Trust, 16 December 2020, retrieved 28 March 2025, www.nuffieldtrust.org.uk/resource/the-remote-care-revolution-in-the-nhs-understanding-impacts-and-attitudes

And, as discussed below, it also appears there is a trade-off between how many appointments are delivered and what proportion take place face-to-face, as practices that conduct more of their appointments remotely are able to deliver more appointments overall. The NHS has recently started to publish practice level data of appointment mode, meaning that we can now look at the associations between appointment type and patient satisfaction. 

Regression 1 shows that there is a strongly statistically significant relationship between the proportion of face-to-face appointments and patient satisfaction.

We found that patient satisfaction is higher in practices that deliver more of their appointments face-to-face, conditional on the total number of appointments provided. While the relationship is statistically significant, the strength of the relationship is only moderate: according to our central estimate, a 10ppt increase in the proportion of appointments delivered face-to-face is associated with a 0.9ppt increase in patient satisfaction. 

We also found (using regression 2, which examines the change in satisfaction between 2019 and 2024) that those practices with more remote appointments in 2024 had experienced a bigger fall in satisfaction since 2019 (as shown in Figure 29). Assuming that these are the practices that have increased remote appointments the most since 2019 (data for appointment mode is not available at practice level before 2022), this would indicate that appointment mode is a driver of satisfaction. 

We have also heard anecdotal evidence that it is predominantly older patients who prefer face-to-face appointments. Regression 8 shows that older patients on average prefer face-to-face appointments. This could be because those aged over 65 now are more likely to find technology difficult to navigate compared to younger patients, meaning the effect could lessen over time as more tech-savvy patients age.

The GP practices that deliver more appointments face-to-face achieve higher QOF scores 

Regression 6 also shows that there is a moderately statistically significant relationship between the proportion of appointments delivered face-to-face and QOF achievement. An additional ppt of appointments that are delivered face-to-face is associated with a 0.03ppt increase in QOF achievement – a relatively small effect. 

Patients prefer having more face-to-face appointments over more appointments delivered by GPs 

The shift towards delivering more remote appointments pulls patient satisfaction in two directions. On one hand, patients report lower satisfaction in practices that provide fewer face-to-face appointments. But conducting a greater proportion of appointments remotely means clinical staff can carry out more appointments overall – a factor that is associated with higher patient satisfaction. As shown earlier, a 1ppt reduction in the proportion of appointments that are delivered face-to-face results in each GP on average being able to deliver 18.2 more appointments per year. From regression 1, we know that, at a practice level, that increase in appointments is associated with an increase of 0.015ppt in patient satisfaction. At the same time, patients report a 0.09ppt reduction in satisfaction for each ppt decline in the proportion of appointments that are delivered face-to-face.* This analysis suggests that increasing the proportion of appointments delivered remotely leads to lower satisfaction overall on average, even though it results in more appointments.

_{* For more information on how we calculated this effect, please see the Methodology in the PDF version of this report.}

Patient satisfaction is not impacted by the timeliness of appointments 

Both the government and the press focus on patients’ ability to access same-day appointments as an important indicator of general practice quality. But regression 1 shows that there is no statistically significant relationship between patient satisfaction and the time between the appointment being booked and taking place, whether that’s same day, 1–7 days, 8–14 days or even 14+ days.