General practice

Spending on general practice in England has grown in the past four years, following a slight decline. Estimates suggest that GPs’ workload has risen faster than spending and, therefore, that they have become more efficient. They have done this by pooling resources between practices and making more use of telephone consultations, among other ways. But lengthening waits for appointments suggest that these techniques have not been enough to keep up with demand.

General practice – the ‘front door’ of the health system – is responsible for treating common medical conditions. Most of the workload in GP practices is made up of patient consultations, but practices also run clinics, provide vaccinations and carry out simple operations. As of March 2018 there were 7,271 practices in England.[1]

The GPs and other health professionals within general practice work closely with wider community professionals (such as health visitors and midwives), social services and voluntary sector providers. Where necessary, patients are referred on to hospitals or specialist units for specialist or urgent treatment.

Spending on general practice has increased by nearly 17% since 2010/11

Change in current and capital spending in general practice in England (real terms), since 2010/11

Between 2010/11 and 2017/18, annual current and capital spending (‘investment’) in general practice increased by 16.6% in real terms overall. Although there was initially a 2% real-terms drop in spending up to 2012/13, it began to rise in 2013/14. This has been supported since 2016 by the General Practice Forward View, which, alongside a range of reforms, included a commitment to an additional £2.4 billion (bn) a year of public spending on general practice by 2020/21.[2] Simon Stevens, Chief Executive of NHS England, has emphasised the importance of this increase in spending, writing that “if general practice fails, the whole NHS fails”.[3] However in 2017/18 only 7.2% of NHS commissioning funds were allocated to general practice.[4]

Demand for general practice services has continued to grow

Demand for general practice services is likely to have risen faster than the population is growing. The population of England grew by 5.7% between 2010 and 2017.[5] However, there has been more rapid growth among groups that use GP services more intensively.[6] For example, the population over the age of 65 has increased by 17% over this same period.[7]

Figure 1.2 Change in estimated demand for general practice consultations, since 2010

We can model the changing demand for general practice consultations by applying an estimate of the consultation needs of men and women of different ages, produced by the Department of Health in 2011, to the changing demographic profile of England.[8] This suggests that population-level demand for consultations grew by 8% between 2010 and 2017.

However, this is likely to underestimate the actual increase in demand, due to increases in the rate of multi-morbidity – people who have more than one chronic illness. In 2012 the Government projected that the number of people with multi-morbidity (which is estimated to account for half of general practice consultations) would increase from 1.9 million (m) in 2008 to 2.9m in 2018.[9]

Input: GP numbers are declining, but the overall workforce is growing

Most general practice spending goes on staff. The overall number of staff working in practices rose by 6.6% between September 2010 and September 2014, and then by 2.6% between September 2015 and March 2018.

Figure 1.3 Number of general practitioners (excluding locums), nurses and other staff involved in direct patient care within general practice (full-time equivalent), since 2010

The number of full-time equivalent GPs (excluding locums) has been falling since 2015.* After growing by 6.2% between September 2010 and September 2014, the number subsequently declined by 4.9% between September 2015 and June 2018 (to 32,370).** While the number of locums has increased, it has not been enough to compensate for the loss of other GPs; furthermore, locum numbers have also begun to decline (from a high of 1,029 in September 2017). As a result, the total number of practitioners fell by 4.1% between September 2015 and March 2018.

This is despite the Government’s commitment to a net increase of 5,000 GPs by 2020 (compared with 2015 figures). As part of this plan, the Government committed itself to training 3,250 new GPs each year.[10] This target has not yet been met, although the proportion of places filled has been increasing year on year (with 97% filled in 2017/18, compared with 85% in 2015/16).*** The Chief Executive of Health Education England has also said that the organisation is on track to meet the target in 2018/19.[11]

However, Pulse (the British magazine on primary care) found in its annual vacancy survey this year that 15.3% of GP posts were vacant, compared with 11.7% in 2016.[12] Also this year, The King’s Fund (an independent health think tank) carried out a survey of 729 trainee GPs and found that only 22% intended to work full time in general practice one year after qualification (compared with 31% of the 318 trainees surveyed in 2016).[13]

Meanwhile, the morale of existing staff is low. Between 2010 and 2017, the GP Worklife Survey reported a rise in all the stress factors it surveyed and a 13% drop in overall job satisfaction. The factors cited as causing the most stress were increasing workloads, insufficient time to perform work, and paperwork.[14] ‘Hours of work’ was the area of work where GPs had the lowest mean satisfaction rating – 3.57 on a scale of 1 (lowest) to 7 (highest) – down from 4.39 in 2010. Satisfaction with pay was at 4.22 in 2017, down from 4.87 in 2010.[15] This suggests that workload is the key driver of GP recruitment and retention issues, with pay a close second.

The limited availability of GPs has encouraged primary care teams to expand, so that patients see the professional most suitable for their needs. The General Practice Forward View committed more than £150m by 2020 to support this.[16] In 2017/18, The King’s Fund’s Quarterly Monitoring Report found that the most common strategy that practices used to increase capacity to manage demand was diversifying the skills mix of staff.[17]

The highest rate of growth in the GP workforce has been among other professionals who are neither doctors nor nurses, but still deliver patient care, such as health care assistants, physiotherapists and phlebotomists. In June 2018 there were 11,980 of these types of workers – up from 9,149 in September 2015 (a 31% increase).[18]

Nurses have had a much more longstanding presence in general practice, and their numbers have also grown in recent years: from 15,398 in September 2015 to 15,925 in June 2018 (a 3% increase). Administrative staff numbers have also grown in this period, from 63,728 to 64,945 (a 2% increase). Within this growth there has been a 6.3% decline in GP practice managers, likely driven by the trend towards fewer, larger practices (see below).[19]

Staff working hours have also changed. Although the GP Worklife Survey suggests that there has only been a limited increase in a GP’s average weekly hours (from 41.4 in 2010 to 41.8 in 2017), the Government’s commitment to a ‘seven-day NHS’ has meant that the timing of those hours have changed.[20] Between October 2016 and March 2018 the percentage of practices with ‘full extended access’ – meaning the offer of pre-bookable appointments outside of core contractual hours (such as in the evening or at weekends)**** – more than doubled, from 18% to 39%.[21]

Medicines make up another big chunk of GP practice spending, although we do not know the precise proportion because we only have listed prices, not the final agreed prices. When products do not satisfy their cost-effectiveness requirement, the NHS tries to negotiate discounts with manufacturers. Known as ‘patient schemes’, these can range from simple discounts to complex reimbursement mechanisms, and remain confidential due to their commercial sensitivity.[22]

The evidence suggests that GPs are prescribing more medicines, but also that they are getting them cheaper. The number of items prescribed in the community increased from 0.9bn to 1.1bn between 2010 and 2017. However, over this same period, the net ingredient cost (meaning the basic cost of the drug) of all community-prescribed drugs only increased from £8.8bn to £9.2bn (which in real terms is a decline of about 7%).[23]

*    All staff numbers in this chapter refer to full-time equivalent staff unless otherwise noted.

**    Workforce figures for June 2018 are currently provisional and will be finalised in November 2018.

***    These figures do not include first-year trainees.

****    The precise definition of ‘extended access’ is access to pre-bookable appointments on Saturdays and Sundays, and on each weekday for at least one-and-a-half hours before 8am or after 6.30pm.


Output: estimates suggest that the workload has risen faster than spending

The most common activity within general practice is consultations. We know that, before 2008, there had been a steady growth in the number of consultations: a large longitudinal study of trends in consultations between 1995 and 2008 found that the number had increased by 38%, from 217m to 300m. That means there was also an increase in the number of consultations per person, per year – from around 3.9 to 5.5.[24] However, there is no longer any nationally collected data on consultations, so we cannot say for sure what has happened since 2008.

Smaller studies suggest that this rate of growth has been maintained or has even accelerated in more recent years. For example, an analysis of 100m consultations between 2007 and 2014 estimated that general practice workload had increased by 16%.[25] Similarly, The King’s Fund has estimated that there was a 15% increase in consultations between 2010/11 and 2014/15.[26]

Putting this together, the workload of GPs appears to have increased more than the demographic pressures predicted by our model of population-level demand. One explanation for this is the increasing complexity of needs among patients. The Health Foundation (an independent health think tank) analysed one practice in 2017 and found that, of frequent GP attendees (meaning attending twice a week), 53% had attended Accident & Emergency (A&E) more than twice in the previous year, 69% had a long-term condition, 43% had been diagnosed with depression and 33% were over the age of 60.[27]

GPs’ workload might also have increased as a consequence of reduced activity in other parts of the health system that have seen deeper cuts. For example, The Health Foundation has estimated a real-terms decline of 17% spending on public health per person between 2014/15 and 2018/19.[28] Meanwhile, there was a 15% decline in the number of community nurses between September 2009 and May 2018, to 35,221.[29]

Finally, these increases may be the result of supply-induced demand (where an increase in the services available leads to an increase in patients using those services, even if it is not necessary). There are concerns that better access to general practice services, such as through extended opening times, might prevent people from accessing expensive, unnecessary care, but it may also increase demand for GP services, as people seek assurance for complaints they may otherwise have dealt with without medical intervention.[30]

In addition to changing levels of activity in general practice, there is evidence that the nature of activity is changing as well. For example, the proportion of respondents to the GP Patient Survey who reported that they had had a telephone consultation has increased, from 5.5% in 2012 to 9.3% in 2018.*

Figure 1.4 Percentage of GP Patient Survey respondents who reported having an appointment over the phone with a member of a general practice surgery, since 2012

There is also evidence that GPs are increasingly connecting with patients online. Between 2015 and 2017 the proportion of respondents to the GP Patient Survey who had booked an appointment online in the previous six months increased from 6.8% to 8.9%. The 2018 survey (asking a slightly different question) found that 12.9% of respondents had booked an appointment online in the previous year.[31]

*    The GP Patient Survey was redesigned in 2018, with changes to the wording of this question. Results between 2012 and 2017 are the combined proportion of respondents who reported: ‘I got an appointment to speak to a GP on the phone’ and ‘I got an appointment to speak to a nurse on the phone’. The 2018 result is the proportion of respondents who reported: ‘I got an appointment to speak to someone on the phone.’


Output: patient satisfaction is slipping but care quality is still high

Historically, general practice has been the most popular element of the NHS. But satisfaction with the service has declined over recent years. In the 2017 British Social Attitudes survey, nearly two-thirds (65%) of respondents were satisfied with the service, compared with 77% in 2010, making it drop from the most popular part of the NHS, to the second most popular (behind outpatient treatment). Meanwhile, the rate of dissatisfaction has increased, from 14% in 2010 to an all-time high of 23% in 2017.[32]

In the GP Patient Survey, the percentage of respondents who described their overall experience of their doctor’s surgery as good or very good decreased slightly from 88% to 84% between 2012 and 2018, reflecting the trend of high but slipping patient satisfaction. Concerns about ease of access to general practice services appear to be driving dissatisfaction. People describing their experience making an appointment as good declined between 2012 and 2018, from 79% to 69%. However, patients continue to report high levels of trust and confidence in the care they have received, and 94.8% reported in 2018 that their needs were met to some extent during their last appointment.[33]

Figure 1.5 Percentage of GP patients rating their experience as ‘good’ or ‘very good’, since 2012

As of May 2017 the Care Quality Commission (CQC) rated 90% of practices as ‘good’ or ‘outstanding’ (although only 85% of practices achieved these ratings for ‘being safe’).[34] Out-of-hours GP services were also reported to be of high quality, with the CQC in March 2018 rating none as inadequate, and 89% as ‘good’ or ‘outstanding’.[35]

Indicator scores from the Quality and Outcomes Framework (QOF) also suggest that the quality of GP services has been maintained, or has slightly improved. The QOF is a voluntary annual reward and incentive programme for all GP surgeries, which provides additional funding for good practice in clinical and public health services. The requirements vary, from maintaining registers of patients in need, to regular reviews of patients’ conditions (for example, an asthma review every year).[36]

Has general practice become more efficient and can that be maintained?

The absence of national activity data makes it difficult to determine with any precision how much more efficient general practice has become since 2010, but the estimates we have suggest that GPs’ workload has increased more than spending.

General practices have made savings in spending on medicines. According to listed figures, there has been a 22% real-terms decline in the average net ingredient cost per item prescribed in the community, supported by the NHS Medicines Optimisation Programme.[37] These cost reductions are likely the result of efforts to increase the prescribing of generic drugs – driving both productivities (when generic brands are able to perform the same function as named brands) and economies (as named brands seek to compete). Listed figures may overestimate the price the NHS actually pays for drugs, which is confidential; it is likely that the NHS has negotiated lower prices, meaning greater actual efficiencies.[38]

The public sector pay cap has also been a major source of savings for the service overall.[39] However, the announcement of a consolidated 2% pay increase for GPs in July 2018 means that future efficiencies will have to come from elsewhere.[40] Some practices have also had to increase their use of more costly locum staff, which will have increased spending per input.

In delivery and organisation, general practice is beginning to adopt new ways of working to meet demand more productively. Practices are increasingly using telephone consultations, for example, as an alternative to the traditional and still- dominant consultation face to face with the patient (although the efficiencies of this approach are still debated, with concerns that it encourages patients to contact their general practice more).[41] Another change is the greater use of technology, such as the ability to book appointments or order repeat prescriptions online.

There has also been a gradual shift away from traditional, smaller practices, with larger practices emerging to pool resources and finances. Between 2010 and 2018, the share of GP practices with only one GP fell from 15% to 11%, while the proportion with five or more GPs increased from 41% to 47%. Overall practice numbers also fell by 13%, to 7,271.[42] Beyond this, individual practices are also collaborating more – on training, for example, or sharing back-office functions. The Nuffield Trust (an independent health think tank) estimated in a survey in 2017 that 81% of practices were part of a collaboration (contractual or more informal), up from 73% in 2015. Of those not in collaborations, a third were considering it.[43] It is difficult to calculate what the implications of larger practices are for running costs but there are likely to have been some savings.

The increasing use of larger, mixed teams, with more patients per GP, has implications for the continuity of care – whether a patient can see the same GP. This has dropped substantially: in the GP Patient Survey, the percentage of patients with a preferred GP who saw them ‘always, or almost always, or a lot of the time’ fell from 65.3% to 50.2% between 2012 and 2018.[44]* The British Journal of General Practice has reported a 27.5% decline in the continuity of care between 2012 and 2017.[45] This is not necessarily a bad thing in all cases: there has also been a decline in the number of patients who say they have a preferred GP, indicating that some patient groups prioritise accessibility over continuity. However, there is evidence that continuity of care can contribute to improved outcomes, such as reduced emergency admissions for conditions that are better treated in the community (such as asthma, diabetes and hypertension).[46]

Any attempts to further increase productivity without comparable changes to working practices – thereby increasing GP workload – risk exacerbating existing recruitment and retention issues, and undermining this progress. In the 2017 GP Worklife Survey, 39% of GPs expressed an intention to leave direct patient care within five years, up from 22% in 2010.[47]

Furthermore, the Government’s extended access policy may put further pressures on GPs. The National Audit Office has said that the Government has “not fully considered the consequences and cost-effectiveness of their commitment to extend access”.[48]

*    The GP Patient Survey was redesigned in 2018, with changes to the wording of this question. Results between 2012 and 2017 are the proportion of respondents who reported ‘always, or almost always, or a lot of the time’ when asked: ‘How often do you see or speak to the GP you prefer?’ The 2018 result is the proportion of respondents who reported ‘always, or almost always, or a lot of the time’ when asked: ‘How often do you see or speak to your preferred GP when you would like to?’


Have efficiencies been enough to meet demand?

The data suggests that these efficiency measures have not been sufficient to allow general practice staff to hold consultations as quickly as they once did in the face of rising demand.

Figure 1.6 Patient waiting times for an appointment with a GP or general practice nurse, since 2012

In 2010, the Government scrapped a controversial target, which required all consultations to take place within 48 hours of an appointment request (which practices were already struggling to meet).[49] Since then the proportion of patients in the GP Patient Survey who report being seen a week or more after they have requested an appointment has increased, from 13% in 2012 to 20% in 2017.[50] The proportion of patients receiving an appointment on the same day as requested was initially broadly flat, but rose very slightly to 38% in 2017.[51] Given increases in demand, this implies that GPs were seeing more patients on the same day and prioritising those with the most urgent needs (the trade-off being that others would have to wait longer).

In the 2018 GP Patient Survey the reported figure for same-day appointments had dropped to 33%.[52] This may be due to changes in the way the question was worded in the 2018 survey.* But it may indicate an intensification of GPs’ struggles to keep up with demand. Due to the methodological change in the survey, we do not know.

Of course, if there is no clinical need for a patient to be seen in two days rather than eight, then we can interpret this increase in waiting times as an efficient use of resources, rather than a problem. However, this is at odds with the Government’s intention to increase – not limit – access to GPs.

Another sign of the struggles that GPs are facing in keeping up with demand is the apparent rise in the number of ‘list closures’. Under the GP contract, practices can apply to NHS England to temporarily refuse to accept new patients (referred to as ‘closing their lists’) with the permission of their clinical commissioning group. Pulse has estimated from Freedom of Information requests that the number of practices closing their lists, and the number of closures being refused, have increased. In 2012 there were 40 applications, with 20% refused.[53] In 2016/17 there were 230 applications, with 37% refused.[54]

*    The GP Patient Survey was redesigned in 2018, with changes to the wording of this question. Between 2012 and 2017 the question was: ‘How long after initially contacting the surgery did you actually see or speak to [the professional you had an appointment with]?’ In 2018 the question was: ‘How long after initially trying to book the appointment did the appointment take place?’