Evidence and rhetoric about access to UK primary careBMJ 2015; 350 doi: https://doi.org/10.1136/bmj.h1513 (Published 31 March 2015) Cite this as: BMJ 2015;350:h1513
- Thomas E Cowling, National Institute for Health Research doctoral research fellow1,
- Matthew J Harris, Commonwealth Fund Harkness fellow in healthcare policy and practice12,
- Azeem Majeed, professor of primary care1
- 1Department of Primary Care and Public Health, Imperial College London, London W6 8RP, UK
- 2Department of Nutrition, Food Studies, and Public Health, New York University, New York, USA
- Correspondence to: T E Cowling
- Accepted 5 March 2015
Access to general practice is a prominent issue in national policy debate ahead of the 2015 UK general election in May. The two main parties, Conservative and Labour, have both made bold pledges on this topic. However, the problem with access has not been precisely defined; nor has the evidence behind pledges been made clear. Below we discuss the evidence on problems with access and whether the party promises are likely to be achievable.
The prime minister has declared that everyone in England will be able to see a general practitioner between 8 am and 8 pm, seven days a week, by 2020 if the Conservative Party is re-elected to government.1 A £400m (€538m; $610m) commitment to implement the policy nationally follows a £50m pilot in about 14% (1147) of general practices.1 2 Results from independent evaluation of the pilot are yet to be published. The government also plans for an extra 5000 GPs and spending an additional £2bn on the frontline of the English NHS next year (currently there are 40 000 GPs and spending is £115bn).3 4
The Labour Party aims to guarantee a GP appointment within 48 hours—a policy recycled from past Labour governments.5 A pledge to invest an extra £100m in primary care a year was followed by the promise of a £2.5bn fund for use across the NHS.5 6 Labour also intends to recruit 8000 more GPs.6 The Liberal Democrats also plan to improve access to general practice (partly through longer opening hours),7 but their pledges are more reserved.
With each party striving for support, politicians have promoted increases in the supply of general practice, particularly numbers of GPs. Each party sees a problem with access that they intend to solve, we presume. But this problem has not been well characterised.
How much of a problem is access?
In 2013-14, 88.8% of patients reported being able to get an appointment last time they tried, and the appointment was convenient for 91.9% of them; 55.8% of patients had a preferred GP, 61% of whom reported seeing or speaking to this GP always, almost always, or a lot of the time.
Each of these measures decreased slightly from 2011-12 to 2013-14 (all P<0.001), as did the percentage who thought their general practice was open at convenient times (83.2% to 79.9%). Among patients who found opening times inconvenient, more patients began to think that extended opening before 8 am, after 6.30 pm, and at weekends would make it easier to see or speak to someone (table⇑).
Although about 90% of patients were able to get an appointment, 10% were not. What does this mean in absolute terms? The last estimate of the annual number of general practice consultations in England, for 2008-09, was 303.9 million.10 If 10% of attempts to get an appointment fail this equates to around 33.8 million unsuccessful attempts each year (81 per general practice each week).
Whether these data suggest a large problem with access depends on how they are framed. International comparisons of access to healthcare typically rank the UK highly, including first place in a recent assessment of 11 high income countries.11 Still, the responsiveness of the NHS in meeting patient demand and expectations, although they may never be fully met, could be seen as one of its core values.
Aims of increased access
The aims and rationale of the policies outlined by the Conservative and Labour parties do not seem explicit. One aim is undoubtedly to win election votes. But the aims for the NHS and for patients are less clear.
Politicians may intend to improve satisfaction with primary care. The GP Patient Survey shows an association between satisfaction and patients getting timely appointments on their last attempts.12 But it also shows that interpersonal aspects of care are the most strongly associated with satisfaction.12 Discrete choice experiments suggest that patients are often willing to forgo quicker access in favour of other appointment characteristics (such as seeing a particular GP) that are valued highly.13 14 15 16
Both parties hypothesise that their policies will reduce attendances at emergency departments (used here to denote all department types, including consultant led services and urgent care, minor injury, and walk-in centres). Visits to such services increased from 17.8 million in 2004-05 to 21.8 million in 2013-14,17 alongside reports of emergency services being stretched by demand.18 Estimates based on the GP Patient Survey, extending the figures in the table⇑, suggest as many as a quarter of emergency department visits (5.8 million/21.7 million in 2012-13) follow unsuccessful attempts to obtain convenient general practice appointments.19 This does not include patients who visit emergency departments without first trying to get a general practice appointment because they think they will be unsuccessful.
As the number of emergency department visits (21.8 million in 2013-1417) is small compared with the number of general practice consultations (303.9 million in 2008-0910), any shift in demand from general practice to emergency and urgent care has a much larger effect on emergency service providers. Furthermore, as there are far fewer emergency care providers (255) than general practices (8000),8 17 small changes for individual practices could have large aggregate effects on emergency departments.
Does access reduce use of emergency departments?
Existing evidence on whether improving access to general practice reduces use of emergency departments is inadequate to inform national policy.
Our national, cross sectional study found an association between general practices with a greater percentage of patients able to see a GP within two weekdays and lower rates of emergency department visits (most v least accessible fifth, rate ratio 0.90, 95% confidence interval 0.85 to 0.95).20 Other studies, using similar designs, suggest that greater access to general practice is also associated with lower rates of emergency admissions for specific conditions such as heart failure and stroke.21 22 23 24 25 26 In the GP Patient Survey those who report worse in-hours access, defined using five measures, are more likely to report using out of hours primary care services (for example, unable v able to get an appointment, odds ratio 1.21, 1.17 to 1.27).27
These analyses demonstrate consistent associations using national data. However, they all use an observational, cross sectional design, which is particularly vulnerable to confounding. It is unknown whether improvements in access to general practice over time are associated with reduced use of other services, or indeed health outcomes and healthcare costs. The unintended effects, which may include supply induced demand, are also unclear.
The effects of policies introduced by the Labour Party between 1997 and 2010 to improve access to general practice remain largely unknown because they were not evaluated.28 This includes the target of guaranteed access within 48 hours and financial incentives related to access and opening hours.
The 48 hour access target was far from being achieved. In the final year of the last Labour government (2009-10) 81.0% of patients were able to see a GP within two weekdays (GP Patient Survey data).9 The target may also have had some unintended effects, including practices reserving more appointments for same day requests.29 30 General practices have not been paid for performance against the target since 2011-12.31
Could primary care make progress on such a promise in the near future? The British Medical Association and Royal College of General Practitioners argue that more GPs are required if access is to be improved. Politicians’ promises of an extra 5000 to 8000 GPs are also likely to be challenging to achieve, however. The largest increase over any five year period—a parliamentary term—in the past 20 years was 5414 GPs (from 2004-05 to 2008-9; average five year increase was 3709 GPs).32 The challenge is compounded by recruitment and retention difficulties, such as too few GP training posts.33
The estimated number of full time equivalent GPs, excluding locums, increased by 26.4% from 2000 to 2009 (to 36 085); the number was similar in 2013 (36 294).34 35 The estimated number of general practice consultations increased by 34.9% from 2000-01 to 2008-09 (225.3 million to 303.9 million; 4.4 to 5.6 consultations per patient).10 More recent consultation data are unavailable, but the number is widely presumed to have increased considerably in contrast to GP supply. Greater administrative and managerial responsibilities may have further added to GP workload.
The public should question the promises of politicians. Pledges on the GP workforce are unlikely to be achieved within a parliamentary term. Guarantees of an appointment within 48 hours may prove unrealistic.
We suggest five points of action for policy and research if improvements in access are pursued:
Evaluate the extended opening hours pilot scheme—The government’s pilot scheme of extended general practice opening hours must be independently evaluated before a decision is taken about national implementation.
Make general practice consultation data routinely available—The Department of Health should provide data on the number, types, and content of consultations provided by each general practice each year. Existing data are limited to sample based national estimates up to 2009.
Discuss skill mix in general practice—The capacity of general practice could be increased by employing practice staff besides GPs more extensively. GPs could focus on complex care for the sickest patients, for whom continuity is important but is often not provided.36 Systematic reviews of randomised trials and quasi-experiments suggest that nurses can provide a quality of care similar to GPs37 38; further research should be a priority. Other staff could complete some of GPs’ non-clinical tasks (accounting for around 30% of salaried GPs’ time39). National research funders recently called for work on changes to skill mix in general practice, including extended roles for lay health workers.40 41 NHS England also plans to trial new staff roles.42
Evaluate new types of appointment—A systematic review on the effects of telephone consultations and triage concluded that further rigorous evaluation is needed.43 A recent trial of telephone triage for same day, face to face GP consultation requests redistributed workload at similar costs to usual care.44 Numbers of face to face GP contacts decreased but use of telephone consultations increased greatly.44 Different models of telephone triage and consultations are now being evaluated (such as email and internet consultations).45 46 47
Give due consideration to demand for general practice—The ability to sustain any improvements in access to general practice will be influenced by concurrent increases in demand. Alternatives to general practice (such as pharmacist led minor illness services) could be investigated. Consideration should also be given to the balance of healthcare provision and investment in public health interventions (such as antismoking legislation) that could have important long term effects on reducing disease burden.
Around 90% of attempts to get general practice appointments in England are successful
The likely effects of proposed changes in access to general practice are unclear
The public should question the pledges of politicians
Policies should be independently evaluated in pilot studies before wide implementation
Pull quotes: If 10% of attempts to get an appointment fail this equates to around 33.8 million unsuccessful attempts each year (81 per general practice each week)
It is unknown whether improvements in access to general practice over time are associated with reduced use of other services, or indeed health outcomes and healthcare costs
Cite this as: BMJ 2015;350:h1513
Contributors and sources: TEC is funded by the National Institute for Health Research to examine access to general practice in England and, in particular, its association with the use of acute hospital services. MJH and AM work closely with TEC in this research. MJH is a public health specialist whose works spans health services research, primary care, and global health. AM has researched the organisation and delivery of primary care for over 15 years. TEC conceived and designed the work, acquired and analysed the data, and wrote the article. MJH and AM revised initial article outlines and article drafts. All sources of information used in preparing the article are given in the text. TEC is the guarantor.
Competing interests: We have read and understood BMJ policy on declaration of interests and declare the following interests: this article is independent research supported by the National Institute for Health Research (TC, doctoral research fellowship); MJH is supported by a Harkness fellowship in health care policy and practice from the Commonwealth Fund; AM is a principal in a London general practice. The Department of Primary Care and Public Health at Imperial College London receives support from the Northwest London NIHR Collaboration for Leadership in Applied Health Research and Care, the Imperial NIHR Biomedical Research Centre, and the Imperial Centre for Patient Safety and Service Quality. The views expressed in this publication are those of the authors and not necessarily those of the NHS, NIHR, Department of Health, or Commonwealth Fund.
Provenance and peer review: Not commissioned; externally peer reviewed.