Abstract
A group of GPs and specialists developed the Gold Standards Framework (GSF) in 2000, in order to improve end of life care in the community. There were a number of reasons that led to the development of this standard, but one of the main factors was the variable level of care provided in England to patients with palliative care needs.
By 2005, the use of the GSF had become standard in general practice; it was incorporated into the GP contract through Quality Outcomes Framework (QOF) points. By 2008, 95% of practices were earning their palliative QOF points by keeping an end of life (or GSF) register and having regular meetings to discuss the patients on their registers.
In 2009 the GSF team performed a national audit of end of life care which found that:
Only 27% of patients who died were included on a general practice GSF register Only 25% of patients who were included on a register had non-cancer conditions Patients on a GSF register received better quality, coordinated care
After this audit the GSF team focused on increasing the number of patients on the end of life register, particularly those with non-cancer conditions. They encouraged advanced care planning in the community through staff training. The GSF does not set a standard, but rather advises that as many patients as possible should be on an end of life register at the point of death. The RCGP Curriculum module 3.09 emphasizes the work of the GSF and states that GPs must be able to identify patients in their last year(s) of life and agree personalised plans for their future care.
In North East Essex (NEE) the My Care Choices Register (MCCR) is a secure online locality GSF register, on to which GPs upload their patients' details and end of life choices. The MCCR fulfills the requirements of the GSF QOF for general practice. In addition it allows healthcare staff in all settings to access patient information. In a fragmented NHS, the MCCR aims to improve continuity of care for patients at the end of life and ensure they get the right care every time.
In November 2014, I asked for the practice GSF register and received a patient list that had not been updated in 6 months. The list had been compiled by administration staff and relied solely on the GPs’ coding. There were no end of life discussions documented or meetings arranged by the practice, and over half the patients on the list had already died. In addition to this the practice did not have one patient on the NEE MCCR.
Aim
To improve the identification, registration, and care of patients approaching the end of life at this GP surgery.
Objectives
Identify what proportion of patients are on a GSF register at the point of death Identify whether patients on and off the GSF register have the following documented markers of care: Do Not Attempt Cardiopulmonary Resuscitation (DNACPR) and Preferred Place of Care (PPC) Identify what proportion of patients are on the locality GSF register ‘MCCR’ Identify patient demographics, cause of death and place of death
Methods
Data was collected retrospectively from the practice records with the GP partners' consent. The initial data collection took place in January 2015 and included all patients who died between 1 January 2014 and 31 December 2014. Case notes were reviewed for demographic data, cause of death, place of death, DNACPRs and PPCs. Notes were also checked for any end of life coding that would have included the patient on the practice GSF register. Patient details were also cross-referenced with the MCCR.
Recommendations were made to the practice in January 2015. Data collection was repeated in March 2016 for all patients who died between 1 January 2015 and 31 December 2015.
Results and discussion
In this practice there were 58 deaths (32 females, 26 males) in 2014, and another 58 deaths (28 females, 30 males) in 2015. One in a hundred of the population is expected to die each year, and so in a practice population of ∼5500 this was as expected. Figure 1 demonstrates the age distribution of patients who died, which was similar in both years.
Recorded deaths in the practice by age 2014–15.
Figure 2 shows that the proportion of home deaths increased from 25% to 50% at the GP surgery over the time of this service review, with a corresponding reduction in hospital deaths. When comparing with figures for NEE as a whole, the improvement was stark. According to the Office for National Statistics (ONS) data for NEE, in 2014, 26% of deaths were at home and 44% of deaths were in hospital. In 2015, 26% of deaths were at home and 43% of deaths were in hospital. This suggests that our intervention significantly improved the number of patients able to die at home, compared with the national average.
Place of death 2014–15.
Patients on and off the GSF register a and markers of care.
Patients with any end of life code were considered to be on the GSF register.
The practice had no patients on the MCCR at the time of the first data collection. Eight patients who died in 2015 were on the MCCR, with a further seven flagged by the administration team for inclusion. These seven would have been included had one of the GPs seen the patient before they died. Encouragingly, the practice had 25 active patients on the MCCR at the time of the second data collection.
What changes were made at the practice as a result of the service review?
In January 2015 the initial findings were presented to the practice staff and I recommended that they start adding their patients to the NEE MCCR.
The following changes were implemented in early-2015:
A member of the practice administration team created a MCCR template, which was attached to any patient’s notes if identified as approaching end of life through coding or discharge letters GPs were tasked with completing this form when the patient was next seen and returning it to the administration team The administration team uploaded the information to the MCCR
Conclusion
Throughout the service review, patients on a GSF register were more likely to have DNACPRs and PPCs than patients not on a register. Patients with cancer were also more likely to be on a register than those with non-cancer conditions. These findings are consistent with the national GSF primary care audit.
Figure 2 demonstrates that there was a substantial increase in the number of deaths at home over the service review period, and a corresponding decrease in deaths in hospital. These changes are far greater than the background rate of change seen in NEE as a whole over the same period.
The practice has increased the number of patients it has on the MCCR, and this appears to be having a positive impact on the quality of end of life care at the practice. It is hoped that by establishing these new principles at the practice, the standard of care for patients with palliative care needs will continue to develop in the future.