Investigation report: Continuity of care - delayed diagnosis in GP practices

This report is intended for healthcare organisations, policymakers and the public to help improve patient safety in relation to continuity of care in GP practices. For readers less familiar with this area of healthcare, medical terms are explained in section 1.

This is a legacy investigation completed by the Health Services Safety Investigations Body (HSSIB) under the National Health Service Trust Development Authority (Healthcare Safety Investigation Branch) Directions 2016.

Patients who visit their GP practice with an ongoing health problem may see several different GPs about the same symptoms. To make sure they receive safe and efficient care, there needs to be a system in place to ensure continuity of care. In the context of this report, continuity of care is where a patient has an ongoing relationship with a specific doctor, or when information is managed in a way that allows any doctor to care for a patient.

While some GP practices in England operate a formalised system of continuity of care, many do not. This investigation explored the safety risk associated with the lack of a system of continuity of care within GP practices.

This investigation’s findings, safety recommendations and safety observations aim to prevent the delayed diagnosis of serious health conditions caused by a lack of continuity of care and to improve care for patients across the NHS.

As an example, which is referred to as ‘the reference event’, the investigation reviewed the care of Brian, who had had treatment for breast cancer and later developed severe back pain.

Brian had a history of breast cancer and had been discharged from the breast cancer service. Two years later he began to have back pain. Initially the pain was so severe that Brian visited his local emergency department (ED). He was discharged from the ED with pain relief and was advised to contact his GP practice.

A month later, Brian telephoned his GP practice and saw his named GP. The GP referred Brian to the GP practice’s physiotherapist and requested a blood test. Brian saw the physiotherapist, who gave him advice about exercises to help relieve the back pain. The exercises did not relieve Brian’s pain and over the following 8 months he saw two out-of-hours GPs and six practice GPs, a nurse and a physiotherapist at the GP practice.

Brian also had consultations with healthcare professionals during this time for other conditions not relating to his back pain. When Brian saw a GP at end of the 8-month period, the GP found a lump on his spine and advised Brian to go to the local ED.

At the ED, Brian had a computerised tomography (CT) scan. A lump was found on his spine which was later diagnosed as metastatic breast cancer (that is, breast cancer that had spread to his spine).

The investigation worked with Brian, the GP practice, local ED and oncology services and national bodies to identify the themes for the national investigation. The investigation focused on:

• How GP practices manage continuity of care. This includes how electronic record systems alert GPs to repeat attendances for symptoms that are not resolving and how information is shared across the healthcare system.
• Workload pressures that affect the ability of GP practices to deliver continuity of care.

• The GP contract, which sets out the mandatory requirements for GP services commissioned by the NHS, does not specifically require GP practices to adopt an approach that ensures continuity of care, but practices can do so voluntarily.
• Many GP practices do not operate a formalised system of continuity of care.
• There is no standard framework to deliver continuity of care in GP practices, so it is done differently across the country.
• Many GPs understand the benefits of continuity of care; however, some practices did not believe that it was possible to deliver such a system. Other practices were able to maintain continuity of care through systems developed by those practices.
• There is no requirement for GP IT systems to consider continuity of care or to ‘surface’ information (that is, identify and flag up relevant patient information) to GPs when they see a patient with unresolving symptoms.
• Patients told the investigation that they found it beneficial to see their named GP for long-term health conditions, including mental health conditions.
• GPs working in a practice with a system of continuity of care had more time to process information during consultations and to carry out any follow-on actions to ensure patients received the care they needed.
• GP practices that operated a system of continuity of care reported to have better staff welfare and retention, and fewer recruitment issues, than those that did not.

HSSIB makes the following safety recommendations

HSSIB makes the following safety observation

1.1.1 GP practices are individual businesses that are commissioned by the NHS to provide for the needs of the local community. The contractual requirements for GP practices are set out in the ‘Standard general medical services contract’ (NHS England, n.d.), also known as the ‘GP contract’.

1.1.2 Continuity of care is referred to in several documents that support the GP contract. These are:

• ‘Quality and Outcomes Framework guidance for 2023/24’ (NHS England, 2023b)
• ‘Network Contracted Directed Enhanced Service – guidance for 2023/2024 in England’ (NHS England, 2023c)
• ‘Network Contract Directed Enhanced Service – early cancer diagnosis support pack’ (NHS England, 2023d).

1.2.1 There are several ways in which GP practices can be organised and run. This investigation focused on GP practices that are formed of a group of GPs working in partnership (NHS, 2022b).

1.2.2 A GP practice can deliver a wide range of health services including medical advice, examinations, treatment, vaccinations, medication prescriptions and referral to other health and social services.

1.2.3 GPs are expert medical generalists who provide the first point of contact with the NHS for most people and their communities (Royal College of General Practitioners, 2021).

1.3.1 A GP practice that works closely with community, mental health, social care, pharmacy, hospital and voluntary services in their community as a group is known as a primary care network (PCN). Over 99% of general practices are part of a PCN.

1.3.2 PCNs are led by clinical directors. A clinical director may be a GP, general practice nurse, clinical pharmacist or other professional working in general practice.

1.3.3 PCNs build on existing primary care services to enable greater provision of proactive, personalised, co-ordinated and more integrated health and social care for people close to home (NHS, n.d.).

GP practice staffing

1.3.4 GP practices consist of many different healthcare professionals, and can include:

• GPs
• nurses
• paramedics
• physiotherapists
• mental health staff
• social prescribers (who interact with the social care system)
• pharmacists
• business managers
• administrative staff.

1.4.1 The Royal College of General Practitioners has stated that GPs providing patients with continuity of care is a ‘critical element’ of the service they deliver. Continuity of care is defined by the Royal College of General Practitioners as:

‘… the extent to which a person experiences an ongoing relationship with a clinical team or member of a clinical team and the coordinated clinical care that progresses smoothly as the patient moves between different parts of the health service.

It can consist of relational continuity – seeing the same people or team, management continuity – management and coordination of care and informational continuity – continuity of patient records and information.’ (Royal College of General Practitioners, 2021).

1.5.1 Breast cancer is often thought of as something that only affects women, but men can get it in rare cases. It grows in the small amount of breast tissue men have behind their nipples. It usually happens in men over 60, but can very occasionally affect younger men (NHS, 2020).

1.5.2 Symptoms of breast cancer in men include:

• a lump in the breast
• the nipple turning inward
• fluid oozing from the nipple
• a sore or rash around the nipple
• the nipple or surrounding skin becoming hard, red or swollen
• small bumps in the armpit.

1.5.3 If breast cancer is found early, it may be possible to cure it. However, if it has spread beyond the breast, a cure is less likely and treatment options are available to relieve the symptoms and increase the length of the patient’s life. If breast cancer spreads, it is likely to spread into the bones and spine.

1.6.1 Back pain can have many causes. It is not always obvious what causes it, and it often gets better on its own (NHS, 2022a).

1.6.2 A common cause of back pain is an injury like a pulled muscle. Very rarely, back pain can be a sign of a serious problem such as a broken bone, cancer or an infection (NHS, 2022a).

1.6.3 It is recommended that someone with back pain sees a GP if (NHS, 2022a):

• the pain does not improve after treating it at home for a few weeks
• the pain is preventing day-to-day activities happening
• the pain is severe or getting worse over time
• the pain creates concern or means that coping is difficult.

1.7.1 The spinal cord is a long bundle of nerves that stretches from a person’s brain to the lower part of their back. The nerves send messages from the brain to control movement and feeling in different parts of the body (Cancer Research UK, 2021).

1.7.2 Spinal cord compression happens when there is pressure on a person’s spinal cord. This pressure causes the nerves in the spinal cord to swell and slows down or blocks their blood supply, stopping the nerves working normally (see figure 1). The pressure can create pain in the spine and the muscles surrounding the spine. The pain can also be felt in other parts of the body.

Figure 1 Example of a tumour on the spine compressing the spinal cord (Cancer Research UK, 2021)

Source: Cancer Research UK, the world’s leading independent cancer charity dedicated to saving lives through research, influence and information. © Cancer Research UK 2023. All rights reserved.

1.7.3 Not all spinal cord compression is a result of cancer. Around 3 to 5 in 100 people (3 to 5%) with cancer develop spinal cord compression.

1.7.4 Almost any type of cancer can spread to the spine. People are at higher risk of developing spinal cord compression if they have cancer that:

• has already spread to the bones
• is at high risk of spreading to the bones, such as prostate, breast, lung or myeloma (a type of blood cancer)
• started in the spine.

Brian’s GP appointments

3.1.1 Brian’s family told the investigation that during GP appointments they described the discomfort that Brian was in. They did not feel that doctors took full account of their concerns. Brian’s family acknowledged that this would not have affected his diagnosis, but said that it may have led to earlier investigations.

3.1.2 The GPs at Brian’s practice described to the investigation that when they saw Brian they did not see the symptoms, such as extreme pain, that Brian or his family described when they had contacted the practice to make the appointment.

3.1.3 GPs at the practice told the investigation that when they examined Brian, he was able to move well, did not have any difficulty with walking and when his spine was examined there was no pain. GP3 did record that breast cancer was a consideration and recommended that Brian contact the GP practice in 3 to 4 weeks if the symptoms continued.

3.1.4 All the GPs who assessed Brian told the investigation that they considered his history of breast cancer. They explained that during consultations the back pain was recorded in Brian’s notes as intermittent, rather than continuous. GPs told the investigation that the intermittent pain and good mobility on examination did not raise an immediate concern that this was a recurrence of his breast cancer. However, during some of the consultations blood tests were requested to help identify whether there may be any other reason for the pain.

Suspicion of cancer

3.1.5 The investigation was told by GPs at Brian’s practice, and the consultant oncologist (cancer specialist) that Brian saw, that GPs would expect pain related to cancer to be “more likely be continuous rather than intermittent”. This expectation was also supported by guidance from the National Institute for Health and Care Excellence (NICE) (2022). Brian was advised at every appointment to contact the GP if his symptoms did not get any better. Brian and his family told the investigation that they raised the concern of “returning breast cancer” during every GP appointment relating to back, loin or flank pain.

3.1.6 GPs told the investigation that they “could only assess what is in front of them” while considering a patient’s previous consultations medical history. All the GPs from the practice that the investigation spoke with did not consider that a referral for a scan was required at the point that they saw Brian. Brian and his family told the investigation that they followed the clinical guidance given, but remained concerned that the back pain might be breast cancer.

3.1.7 Brian’s consultation with his named GP (GP1) with back pain in January 2020 was 2 years after his discharge from the breast cancer clinic, and 12 years after his initial diagnosis. A consultant oncologist who specialised in breast cancer told the investigation that if breast cancer spreads it is known to spread to the spine. The pain may not only be experienced in the spine but may be felt around the back and in the legs. Once breast cancer has spread, the management and treatment options are limited. The consultant oncologist stated that the recurrence of breast cancer 10 years after initial treatment was less likely. This view was also shared by the GPs who saw Brian for his back pain. The consultant oncologist also said that if a patient has had cancer, then it is more likely that they will develop another cancer.

Process for blood test results

3.1.8 When in January 2020 GP1 requested that blood tests be carried out, Brian did not attend for these to be undertaken. The investigation was unable to establish why. A consultant oncologist said that breast cancer does not have a specific marker in the blood, but blood tests are used to give a general indication of overall health. For example, if a blood test indicated that a patient was anaemic (had a low red blood cell count), this could be the result of several conditions, including cancer, and further investigations would be needed.

3.1.9 GP1 said that the reason they wanted the blood tests done was to rule out other causes for the back pain. They said that if the tests came back with no significant findings, they would refer Brian for further investigations at the hospital as they did not have the test facilities in the practice. Due to the GP contracting COVID-19 and then other personal commitments, the follow-up of the blood test was not carried out.

3.1.10 Within the GP practice, if a blood test was requested but not undertaken there was no warning system to flag that it was outstanding. If a blood test had been undertaken, and results were late, there was a warning system to alert staff to follow up the results. Some GPs told the investigation that they had taken it upon themselves to set personal reminders, either electronically or using handwritten notes, to follow up blood test requests and results, among other investigations that they had requested. This system of safety relies on busy individuals remembering to check whether tests had been done or to create their own individual reminder system, which was therefore not centrally held. This individual system could not be checked by other GPs if the GP who had created it was on leave or unwell.

3.1.11 The GP practice systems in place for tracking whether blood tests had been carried out were either ad hoc or had no closed feedback loop (a means of checking that an action has been completed satisfactorily). This means they were unable to effectively alert all appropriate GP practice staff when blood tests had not been carried out.

History taking

3.2.1 The investigation was told that taking and reviewing a patient’s medical history is central to a GP’s ability to diagnose their condition and develop a treatment plan. History reviews and the way clinical notes are displayed are particularly important when a patient is unable to see their named GP. In Brian’s case, he saw his named GP once at the start of the 8-month period, a further six GPs at his practice and two out-of-hours GPs.

3.2.2 Brian’s named GP (GP1) stated that they had a detailed knowledge of Brian’s medical history and were less reliant on having to read back through Brian’s notes to remind themselves of previous appointments. This was because they had been Brian’s GP for many years and would always try to see Brian, if possible, when he made an appointment. GP1 became unwell with COVID-19, followed by a period of recuperation, and therefore was not able to be involved in Brian’s care after the initial consultation. Other GPs who saw Brian told the investigation that they did not know him and therefore relied on the short time available to them to read through the notes from the last few appointments to gain a brief understanding of his history.

3.2.3 The GPs told the investigation that 10 minutes were allocated for each consultation session. This was due to the volume of patients that they needed to see on a daily basis. Consultations include:

• reviewing the patient’s previous history via an electronic notes system
• taking a history of the patient’s current symptoms
• examining the patient if seeing them face-to-face (or, if a telephone appointment, assessing whether the patient needs to come in)
• considering a potential diagnosis
• discussing a treatment plan with the patient for that diagnosis, or investigations such as tests/scans, as necessary
• if there is time at the end of the appointment before the next one starts, writing any prescriptions as needed, referring the patient to another service or requesting further investigations.

This meant that when GPs saw Brian there were a number of tasks to fit into the 10-minute appointments. They told the investigation that this created time pressures and it was not always possible to complete all tasks to their full extent at the time of the consultations.

Reviewing patients' notes

3.2.4 The GPs told the investigation that during a consultation, there was not enough time set aside for reading a patient’s notes thoroughly to review their previous appointments and conditions. They said that the time they had to read a patient’s notes was the time that it took for the patient to walk to the consultation room having been called in. GPs explained that a patient’s notes were open during a consultation and if something specific was mentioned, they could look back through the notes if needed.

3.2.5 As soon as the allocated time for an appointment had been reached, GPs moved directly to the next patient and had to start the consultation process again. This meant that they did not have time to re-assess their own decisions or revisit a patient’s history before moving on to the next patient.

3.2.6 Brian also went to his GP practice for other reasons, for example his learning disability and mental health annual check-ups and reviews. In some cases, these appointments happened between back pain consultations, so in the limited time that GPs had available to them to review Brian’s history they would see these rather than just the back pain appointments.

3.2.7 GPs told the investigation that clear electronic patient records were essential, particularly when they did not know a patient well. They described their current system as displaying notes chronologically. However, in a case such as Brian’s, where his repeated visits for back pain were mixed in with appointments for his other conditions, this did not help them to identify a pattern, given the limited time to review previous consultations.

3.2.8 GPs explained that there were also many ways to code (a quick means of recording patients symptoms in the IT system) back pain within their electronic patient record system: these included lower back pain, flank pain, loin pain and musculoskeletal pain. This added to the difficulties in quickly reviewing a patient’s history using electronic records.

3.2.9 In conclusion, the way information is presented to GPs on the electronic patient record system contributed to Brian’s repeated attendances for back pain being missed.

Identification of repeat attendances

3.2.10 GP4 had adopted a “three strikes and refer the patient” approach, meaning that if they had seen a patient for the same condition three times and there was no improvement despite a management plan being in place, then they would refer the patient for further investigations. They explained that the reason for this was that they would have exhausted the primary care treatment options by this point.

3.2.11 After reviewing Brian’s incident, the GPs in the practice reflected as a group that the approach of GP4 was adopted formally across the practice. This was achieved by recording the number of attendances on the patient’s electronic notes (1 to 3) to indicate the number of visits with the same symptoms. GPs told the investigation that their practice’s adoption of this management system was to help GPs when they saw patients, they were unfamiliar with, because the practice recognised that they were unable to ensure that patients always saw the same GP every time.

3.3.1 At the time that Brian was visiting his GP practice for back pain there were other factors which meant that GPs had less time to review his history and therefore identify a pattern of repeat attendance for back pain. These factors are described below and range from long working days and supporting other clinical colleagues, to fatigue and high cognitive workload.

GP tasks

3.3.2 GPs told the investigation that they started work at 08:00 hours and finished at 18:00 hours – a 10-hour day. All GPs said that the working day in the practice was only “the tip of the iceberg”. There was hidden work outside standard practice times (evenings and days off) that was used to catch up on referrals, read discharge letters, do medication reviews, review blood test results and carry out other tasks.

3.3.3 During the workday GPs had scheduled breaks, but on many occasions these breaks were used to catch up on tasks resulting from previous appointments as described above.

3.3.4 GPs would also be made available to provide support to other practice clinicians (such as nurses and paramedics) if they were concerned about a patient they were seeing. GPs in the practice were allocated to call back patients who had telephoned with urgent concerns that day, while others were allocated to scheduled appointments.

3.3.5 One GP said that they regularly “re-lived” their day’s consultations in the evening and would regularly “wake at night” worrying about whether they had made the right decisions. They said that this was due to not having enough time to assess patients and the sheer volume of patients that they see daily. This resulted in the GP moving to another practice that did not have the primary care network structure (see 1.3). This allowed them to see a greater variety of patients, with both complex and straightforward needs.

Time between patient consultations

3.3.6 As there was no time between consultations to process what was heard, GPs told the investigation they found it difficult to “reset for the next patient”. They might have been talking about palliative care or giving bad news in one consultation, and in the next seeing a patient with severe chest pain or a swollen leg. Some GPs said that they reset themselves after a particularly complex consultation by trying to remove themselves from the consultation room, even if only for a few moments. Several GPs described that there could be transference of psychological trauma from patients to doctors, especially if the patient was of similar age or situation to the doctor. GPs worked to a tight schedule, meaning that respite to process and reflect on an appointment was not always possible.

3.3.7 One GP said that resetting was a “learned skill” and necessary to protect themselves and be able to continue practising. Several GPs at the reference event practice told the investigation that they had made a very difficult decision to reduce their working time to deal with the mental and physical fatigue that they found themselves dealing with. Several experienced GPs said that they had been practising for many years, but the situation was worse now than it had ever been. This had resulted in some GPs making a difficult decision to retire early to protect their wellbeing.

3.3.8 One GP reflected that patients only have a short amount of time with doctors and appreciated that they should have the best possible service for that time. The GPs told the investigation that doctors would feel more tired as the day progressed and more focused on the next break opportunity. Several also indicated that their performance would not be as good at 17:30 hours as it would be at 08:00 hours. This theme has been identified in previous HSIB investigations and will be explored in further work by HSSIB.

3.3.9 Brian’s GP practice had adopted a primary care network (PCN) model of working (see 1.3). GPs said that their patients saw this as beneficial, giving them quick access to services such as minor ailment advice, physiotherapy, mental health care and dieticians. The investigation was told that this had created an unintended consequence whereby GPs only saw patients with the most complex needs all the time. Several GPs referred to “practising at the top of our [their] licence all the time” and told the investigation that the unintended consequences included:

• less experienced GPs not having the opportunity to identify and treat minor ailments and track how they may progress into more complex conditions for their assigned patients.
• GPs being less likely to get to know individual patients meaning that time for history taking and review became more important.
• an increase in decision fatigue, where GPs were continually making complex and important decisions for patients.

3.3.10 Experienced GPs at the reference event practice said that they had years of practice to draw on to care for patients with complex needs, but that newer GPs did not have that experience. They said that newer GPs no longer had exposure to minor ailments such as colds and coughs, or other medical discussions such as family planning. This removed GPs’ ability to build a picture of a patient. A partner in the GP practice stated that they were “asking a lot of new colleagues”, acknowledging that they did not have experience but still needed to provide a safe service.

3.3.11 GPs told the investigation that one consequence of doing continual complex consultations was “cognitive fatigue”. This can be described as ‘failure to maintain and optimize performance over acute but sustained cognitive effort resulting in performance that is lower and more variable than the individual’s optimal ability’ (Holtzer et al, 2011). Put simply, their heavy and complex workload meant GPs were becoming mentally tired, which affected their performance. The GPs said that this meant that they, and other GPs, were working reduced hours as a coping mechanism. In addition, the investigation was told that many more experienced GPs were choosing to retire earlier than they had intended because they recognised the impact of the pressures in the healthcare system on their individual welfare. Furthermore, a GP partner told the investigation that sickness relating to workload among GPs in the practice had been rising, affecting the practice’s ability to maintain GP–patient continuity and deliver a service.

3.3.12 All the GPs at the reference event practice who assisted the investigation became emotional when discussing the pressures under which they were trying to provide safe services to their patients. One GP described the current system as “brutal” and that they could “not remember a time when it was more difficult to work in the primary care system”. They all wanted to be able to deliver a safe service to patients but felt that it was becoming increasingly difficult.

3.3.13 A partner at the GP practice told the investigation that GPs leaving or reducing their working hours was having an impact on primary care, including:

• fewer GPs to deliver a safe service
• longer patient waiting times to see their nominated GP
• increased volume of work for the remaining GPs, resulting in increased fatigue
• less continuity in seeing a named GP
• greater reliance on reviewing patients’ histories but less time for GPs to do this.

3.3.14 In conclusion, the additional pressure described above provided a context for the working conditions that GPs were experiencing during Brian’s care. The short consultation period available to them, and cognitive fatigue caused by continually making difficult decisions and reviewing patients’ histories, meant that identifying patterns became increasingly difficult for the GPs.

3.4.1 GPs told the investigation that when there are recruitment and retention challenges the practice can become “under doctored”. This resulted in merging of a number of practices, meaning that there were increased patient numbers in that practice and in turn it became more difficult to provide continuity of care.

3.4.2 In the reference event, ways of working adopted during the COVID-19 pandemic, demands on the service and Brian’s request for urgent appointments meant he was assigned to see the next available GP rather than his named GP. This led to Brian and his family having to “re-tell” their story at every appointment. It was only when he had seen the same GP (GP4) in quick succession that it was identified that he needed further investigations.

3.4.3 A GP partner told the investigation that they had to balance continuity of care with timely access to primary care. The investigation was informed that this could create a conflict, leaving patients with a choice of seeing their named GP when next available or the next available GP if urgent.

3.4.4 The practice Brian attended was in an area of high social deprivation. This meant that there was a greater prevalence of health problems and long-term health conditions (The King’s Fund, 2022), meaning that the GP practice was caring for large numbers of high-complexity patients. One GP told the investigation they see high numbers of patients not only with complex health needs but also social issues, as patients feel that there is nowhere else for them to go with these issues.

3.4.5 As well as putting greater pressure on the GP practice from a healthcare delivery perspective, this also had an impact on GP retention and recruitment. A partner in the practice said that they can train a new GP only to lose them to a more affluent area where there are lower numbers of patients with complex needs.

Summary

3.4.6 Every GP practice visit that Brian had in relation to his back pain restarted the consideration of his symptom because he was seen by different doctors. In all the GPs’ opinions, the signs and symptoms they saw at the individual visits did not warrant referral for hospital diagnostics.

3.4.7 The GP practice told the investigation that they understood the challenges around a patient seeing several different GPs. These challenges meant that there was a need for a robust system to support continuity of care through the electronic patient record system, linking patients’ attendances for the same symptoms. They recognised that in Brian’s case there was no system in place to do this, so the way that information was displayed to GPs in the electronic patient record system became more important.

3.4.8 All GPs in the practice said there was not enough time available to them during consultations to complete all the necessary tasks required of them. This meant that these tasks had to be undertaken in their own time, which was having a negative impact on their wellbeing. The investigation heard that this affected their ability to deliver the service that they wanted for their patients.

Governmental level discussions

4.1.1 A recent meeting of the Health and Social Care Committee (2022) (a cross-party committee of MPs that scrutinises the work of the Department of Health and Social Care) discussed GP continuity of care and the future of GP practices. It recognised that there is currently a requirement for all registered patients to have a named GP. However, in many cases this is purely administrative and does not equate to a clinical overview of the individual patient.

4.1.2 The Committee discussed whether continuity of care would be considered in the GP contract to be updated in 2024. The NHS primary care directorate told the investigation that those negotiations were ongoing for the GP contract (see 4.1.8).

4.1.3 The NHS England Primary Care Group (a group that manages policy in primary care) told the investigation that setting up a system and maintaining continuity of care for patients is important, but that currently it is not embedded in any policies. The Director for Primary Care at NHS England said that they are working on a new patient ‘access model’. The ‘Delivery plan for recovering access to primary care’ (NHS England, 2023a) discusses the importance of ‘relational continuity’ and gives examples of how continuity of care can be achieved and maintained. It states that ‘Evidence shows that relational continuity yields significant benefits for patients, systems and staff, and is especially important for patients with multiple or complex conditions.’

Academic literature

4.1.4 The Health Foundation (Wiltshire, 2019) found that patients who see the same GP more of the time had fewer unplanned hospital attendances. It also reported that ‘for those with multiple complex conditions, this relational continuity can mean the difference between appropriate, timely care and a missed diagnosis’.

4.1.5 A research paper by The King’s Fund (2010) found that ‘Relationship continuity is generally highly valued by patients and staff, and there is convincing evidence of its association with better health outcomes’. However, it also recognised that there were risks to this approach, such as patients accepting inappropriate waits to see the same clinician. The research paper concluded that on balance continuity of care is desirable and of benefit to both patients and clinicians.

4.1.6 ‘Improving continuity: the clinical challenge’ (Pereira Gray et al, 2016) cites safer prescribing as one of the advantages of continuity of care; it is safer for a patient to have medications prescribed by a GP who knows them well, rather than by one who does not. It also states that ‘trust in their doctors by patients is associated with earlier diagnosis of cancer’.

4.1.7 Some GP practices operate a system of personal lists, where a single GP is responsible for all the clinical information management for an individual patient, even if they are unable to see that patient personally (Pereira Gray et al, 2021). For example, during an observation visit the investigation observed a patient who was on GP A’s personal list visiting their GP practice and seeing GP B. After the consultation, GP B requested further examinations (a blood test), but the investigation was told that when these results were received, GP A would be responsible for reviewing them. This means that GP A, while not physically seeing the patient, would maintain and build a clinical picture of the patient.

The GP contract

4.1.8 The GP contract sets out the essential requirements that GP practices must deliver; continuity of care is not among them. GP practices that the investigation spoke to said that there are many competing priorities and if there is no requirement to deliver continuity of care, this may “slip” down the priority list due to challenges such as financial constraints, staffing levels or staff availability.

4.1.9 The NHS England Quality and Outcomes Framework guidance is voluntarily adopted by GP practices, but they are not required to do so. The indicators in the guidance on ‘Reducing avoidable appointments’ include the following:

‘Referencing the Royal College of General Practitioner’s 6 steps to start to improve delivering continuity of care from their Continuity Toolkit for those who need it and adapting to suit the needs of the practice.’ (NHS England, 2023b)

4.1.10 The Network Contracted Directed Enhanced Service guidance (NHS England, 2023c) discusses continuity of care from a pharmacy perspective, but not from an overall GP practice perspective.

4.1.11 The Network Contract Directed Enhanced Service early cancer diagnosis support pack (NHS England, 2023d) discusses the need to have feedback loops between primary and secondary care to ensure that continuity of care is maintained, but not specifically within a GP practice.

4.1.12 The investigation was told by some GP practices that they wanted to deliver continuity of care but have competing priorities. Because there is no explicit requirement to deliver continuity of care, it may not be given priority when other challenges in the practice, such as staff vacancies, come to the fore.

4.1.13 In conclusion, the investigation’s review of national bodies’ publications and research literature supports the benefits of continuity of care to improve efficiency of GP services and patient safety.

4.1.14 GP practices can voluntarily adopt systems of continuity of care in line with the GP contract; however, when there are competing priorities continuity of care may be deprioritised.

4.2.1 The investigation visited GP practices to observe ways of working, and spoke with GPs, administrative staff, nurses, pharmacists and allied health professionals (such as physiotherapists) and patients. The investigation also spoke with representatives of the Royal College of General Practitioners and NHS England. From these visits and discussions, the investigation observed that continuity of care is managed differently at different GP practices. Some use ‘personal lists’ (see 4.1.7) alongside informational continuity (continuity of patient records and information – see 1.4.1) to deliver continuity of care, while others rely on informational continuity alone when seeing the same GP is not possible.

4.2.2 Some practices, as in the reference event, did not have a mature system in place to maintain continuity of care. Staff at Brian’s practice, and at other practices, said that it was not possible to deliver personal continuity of care for the following reasons:

• A higher reliance on part-time and locum staff at practices located in socially deprived or rural/difficult to commute to areas, where many GPs may not want to work.
• There are not enough GPs to be able to deliver continuity of care.
• Patients visiting GP practices not only with complex healthcare needs but also with social care needs, meaning that consultations take longer, putting pressure on the healthcare system.

4.2.3 The investigation found that GPs recognised the benefits of continuity of care but could not see how to deliver it in practice. However, the investigation found other practices that were able to maintain continuity of care through systems that they had developed themselves. Some practices were able to do this despite challenging and diverse social conditions in their geographical area. The investigation found that in these practices:

• retention and recruitment was less challenging meaning there was consistency in staff numbers and regular staff to see patients
• more efficient use of consultation time as GPs’ prior knowledge of their patients meant that appointments were more targeted
• more tasks could be completed during the consultation, with less work being completed out of working hours
• there were robust triage systems in place to ensure patients who telephoned in received access to the appropriate service in the practice (including pharmacy, nursing teams and physiotherapists, among others)
• it was necessary for the whole team to focus their efforts on maintaining the system of continuity of care, and administrative staff were pivotal in this.

4.2.4 Patients told the investigation that they found it beneficial to see their named GP for long-term conditions, including mental health conditions. They described being prepared to wait to see their regular GP rather than explaining their condition to a new GP. GPs told the investigation that this doctor–patient relationship was not only beneficial to the patient but also to the efficient running of the GP practice. For example, a GP who knew a patient and their ongoing treatment well was able to complete the consultation faster than a colleague who would first have to establish the patient’s history. One GP stated that they “would take 3 minutes to see a patient they knew well whereas a colleague unfamiliar with the patient could take 10 minutes”.

Personal lists

4.2.5 The investigation observed one GP practice which operated the personal list model. It told the investigation that maintaining personal lists requires a whole-practice co-ordinated effort (“personal list culture”) involving clinical and non-clinical staff (such as reception and records management staff).

4.2.6 A GP partner told the investigation that they had noted stronger relationships between doctors and patients as a result of the personal list system being used. For example, if a minor error had been made by a patient’s GP, they found that the patient was more forgiving and understanding as they had an “existing therapeutic relationship”.

4.2.7 GPs reported that they were able to have more “difficult conversations” with patients about long-term health, such as stopping smoking and losing weight. They also described to the investigation that patients start to take more responsibility for their own health as a result of the personal relationship.

4.2.8 Staff in GP practices told the investigation that continuity of care prevents retraumatising patients, because seeing the same GP means they do not have to retell their health “story”.

4.2.9 The investigation spoke to several GPs who had previously worked in practices that operate a personal list system and in practices that didn’t. They said that they would be anxious to return to a non-personal list practice. Some GPs cited their welfare as a source of anxiety in non-personal list practices; this is discussed further in section 4.4.

4.2.10 In conclusion, GP practices that operate the personal list system as a way of delivering continuity of care told the investigation that there is no funding explicitly to do so. However, they recognised that this system has safety benefits for patients and staff, and therefore decided to invest in it outside of their contracted services.

Patients’ perspective

4.2.11 As part of its observation visits, the investigation reviewed the results of the national GP satisfaction survey 2022 (NHS England, 2022). There was a marked difference in the results, with the practice that ran a personal list achieving higher than the national average for patient satisfaction. Those practices that did not operate a system of continuity of care scored significantly lower than the national average. This contrast was most clear in the responses to the question about seeing your ‘preferred GP’. The investigation visited practices at either end of the satisfaction scale, as described below.

4.2.12 The investigation spoke to patients registered with GP practices who saw their named GP most of the time, and those who saw the next available GP with no priority given to seeing the named GP.

Patients in a practice operating a system of continuity of care

4.2.13 The investigation held a focus group with patients at a practice which operates a system of continuity of care. The patients told the investigation:

• They had a relationship with their GP which allowed the GP to know them and their health and personal circumstances.
• They did not have to re-explain long-term conditions which they felt made the appointments more time efficient.
• Patients were willing to wait to see their own GP for chronic conditions, which could be between a few days and several weeks.
• For acute conditions they were happy to see the duty doctor, but often due to the operating model at the GP practice, their named GP would identify that they had called in and add them to their list for the day.
• The relationship with their GP meant that they were able to talk openly and maintain confidence, particularly if a mistake had been made.
• They felt able to “respectfully disagree” with a doctor due to the relationship they had.
• They had noted “stability in staffing”, with some patients having seen the “same GP for 19 years”.
• They were aware of a buddy system being used to provide cover for annual leave or sickness absence.

Patients in a practice that does not operate a system of continuity of care

4.2.14 The investigation held a focus group with patients at a GP practice which does not operate a defined system of continuity of care. The patients told the investigation:

• When they rang in for an appointment, they would be triaged to determine whether they needed to see a GP or another healthcare professional.
• If they needed to see a GP, they were allocated the next available GP.
• They rarely or never saw their named GP.
• They would “love” to see their named GP on a regular basis, like “it was 20 years ago”.
• They had to “re-tell” their history to new GPs which was frustrating and could be traumatising, particularly in relation to mental health conditions.
• GPs did not know them or their wider personal circumstances.
• They felt that they got used to not seeing a regular GP and therefore would not accept a wait to see a regular GP.
• They couldn’t see how the practice could put a system in place that would mean they saw the same GP if needed and were surprised that other practices were able to operate a system that did.

Vignette

Conclusion

4.2.15 Staff and patients told the investigation that continuity of care is a vital component in delivering safe GP services. The investigation found that there is no standard framework to deliver continuity of care so there is variation across the country. The investigation also found that while GPs understand the benefits of continuity of care, some practices did not believe that it was possible to deliver such a system. Furthermore, there is no requirement in national policy or guidance, or within the GP contract, to provide continuity of care.

HSIBB makes the following safety recommendation

4.3.1 Another way of maintaining continuity of care is through the use of patient records. Most GP practices use electronic patient record systems (known as GP IT systems). The investigation observed how GP IT systems present information to GPs about current and ongoing medical conditions and symptoms.

4.3.2 As in the reference event, the investigation observed at other GP practices that GPs have limited time to read notes before and during consultations. This means ’red flags’ (important information highlighted by the system which identifies serious conditions that need urgent treatment or attention) and consistent symptom coding are essential.

4.3.3 GP practices told the investigation that GP IT systems do not present information in a way that enables GPs to quickly see whether a patient is returning with unresolving symptoms, as in the reference event. NHS England described this to the investigation as “surfacing of information”, where information is presented in such a way that it comes to the fore without onerous or frustrating warnings or notifications.

4.3.4 The GP contract does not set out any requirements for GP IT systems that relate to continuity of care. In the reference event, while breast cancer was considered at each of Brian’s attendances, the lack of internal processes and the way in which GP IT systems presented this information meant the opportunity for information continuity was lost. GPs at other practices told the investigation that situations similar to the reference event could occur because of the way information is displayed.

4.3.5 NHS England told the investigation that each practice is an independent contractor and has the freedom to select its own GP IT systems. The IT system should comply with the principles set out in ‘The good practice guidelines for GP electronic records – quick reference guide’ (Department of Health and Social Care, 2011). However, the guidelines do not consider:

• continuity of care within a practice setting
• how information is displayed
• how information is entered to reduce ambiguity and variation in symptom coding
• when ‘red flags’ should be used to alert clinicians.

4.3.6 In conclusion, the way that information within GP IT systems is presented to GPs is especially important when personal continuity between a GP and patient is not possible.

4.3.7 The investigation found that there is no requirement for GP IT systems to consider continuity of care or to ‘surface’ information to GPs when they see a patient with unresolving symptoms. In other settings outside healthcare, artificial intelligence is being used to surface information, and in a way that is usable (Sutcliffe, 2023: Thomson Reuters, 2020).

4.3.8 Surfacing of information would enable insight to a patient’s relevant clinical history and information, at the right time, to the right healthcare professional. This insight would provide a system that better supports decision making in GP practices where named GPs, or personal lists, are not embedded in routine practice.

HSIBB makes the following safety recommendation

Brian's case demonstrates the impact that a lack of continuity of care can have on patient care and the impact on patients and their families of a missed diagnosis. However, during the investigation GPs were increasingly reporting concerns about their own welfare as a result of the demands of GP work.

Due to challenges in accessing GP practices (see appendix), the investigation was only able to carry out observations at three GP practices. However, the experiences described below were supported by national bodies, in particular those that support staff welfare.

GP practices that do not operate a formalised system of continuity of care

4.4.1 The investigation spoke with GPs at two practices that did not operate a continuity of care model. These practices were in areas of high social deprivation and provided a service for patients from a combination of rural and urban settings. Across these sites there were similar issues around low staff morale and wellbeing.

4.4.2 The investigation observed that the challenges faced by GPs mirrored those in the reference event and included:

• Working at the “top end of [their] licence” – that is, seeing patients with the most complex needs – because other clinical staff in the practice dealt with minor ailments or specialty areas such as pharmacy and physiotherapy.
• No time between appointments to process emotionally difficult conversations, so GPs were not able to mentally deal with any emotional and traumatic issues that they may have heard before seeing the next patient.
• No time to process technically difficult consultations resulting in cognitive fatigue, making decision making harder as the day progressed.
• Feeling unable to do all the tasks required during a consultation (in this case 15 minutes per consultation) and taking work home and into days off and weekends so that they were unable to have a “normal” life outside work.

4.4.3 The investigation observed that the personal impact on GPs was significant, with many becoming upset when speaking to the investigation about their workload and work environment. GPs told the investigation that poor staff wellbeing resulted in:

• Many GPs planning to retire early (Tilley, 2022).
• GPs leaving practices to seek employment where patients had less complex needs and were fewer in number.
• GPs reducing their hours so that they could have a normal life outside work or leaving the profession to work in other areas of medicine or medical research. Despite reducing clinic hours many GPs reported working at least full-time hours (at home) to keep on top of administrative tasks.
• Retention and recruitment challenges at the practices.
• A negative impact on GPs’ overall wellbeing caused by the sheer volume and pace of work, and the cognitive impact of starting afresh with each patient (The Health Foundation, 2023).

GP practices that operate a formalised system of continuity of care

4.4.4 A subject matter advisor told the investigation that there are currently no studies into the effects of continuity of care on the welfare of staff in GP practices. However, one study, the ‘Increasing Continuity of Care in General Practice programme’ (Mott MacDonald, 2022) stated that ‘increased staff satisfaction was reported across the programme for both clinical and non-clinical staff’.

4.4.5 The investigation spoke to GPs and other staff at a GP practice that operates a model of continuity of care. Patient numbers at this practice were similar to the practice described above and it was in an area with high social deprivation; however, it was in an urban rather than rural setting. The investigation found that the GPs had more time to process information during consultations and carry out follow-on actions needed to ensure patients received the care they needed.

4.4.6 Staff at this practice were encouraged to meet daily for lunch in a staff room, and this was possible because of the systems of continuity of care in place. The time that was freed up by GPs seeing patients they knew made decision making more efficient. Staff told the investigation that this, along with the wider team involvement with the management of appointments, meant that nearly all consultations ran on time.

4.4.7 GPs told the investigation that they felt that their workload was manageable within the time of a consultation (15 minutes per consultation). They took some work home to complete, but not to the same extent as the GPs at the practice described above (see 4.4.2). The GP practice not operating a system of continuity of care allocated 15 minutes per consultation.

4.4.8 A partner at the GP practice told the investigation that running a model of continuity of care needed continual commitment and management focus, but that the benefits to the practice and staff outweighed that effort. Staff described some of the benefits to the investigation, as follows:

• staff retention was high
• staff were able to work reduced hours while maintaining continuity of care
• GPs, other clinicians and administrative staff wanted to come and work at that practice
• it was easier to integrate locum GPs when needed due to the presence of a buddy system
• sharing patient information across the practice was easy.

4.4.9 In conclusion, the investigation visited one GP practice delivering a system of continuity of care and two that did not have a formalised system of continuity of care. Although this is a small sample of GP practices in England, the investigation found a visible difference between the demeanour of staff in the practices that operate continuity of care and those that did not.

4.4.10 Some staff at the practices that did not operate continuity of care became visibly upset, some looked tired, and some others had resigned themselves to the challenges that they faced. In the practice that operated a system of continuity of care, the investigation observed staff who were primarily positive, engaged and had capacity and willingness to engage in the investigation.

4.4.11 The GP practice that operated a system of continuity of care reported better staff welfare and retention, and fewer recruitment issues, than those that did not. One administrative staff member told the investigation that the practice “feels as one team working together for patients”.

4.4.12 The wellbeing challenges facing GPs are similar to those identified in the third interim bulletin from HSIB’s investigation ‘Harms caused by delays in transferring patients to the right place of care’ (Healthcare Safety Investigation Branch, 2023).

HSSIB makes the following safety observation