Dr Anju Gupta and Richard Clements explain how improvements in the quality of diabetes care were achieved, and variation reduced, in a socioeconomically deprived London borough

Health inequalities, which were previously background static, have been caught up in the zeitgeist of COVID-19 and brought into sharp focus. The backstory starts with the 10 years between the Marmot Review, published in 2010, and its subsequent reappraisal in 2020, which saw variability in healthcare and outcomes widening.1,2 These inequalities are especially evident in long-term conditions like diabetes. The prevalence of long-term conditions increases with levels of deprivation, yet the quality of care is seen to deteriorate with deprivation.3 Socioeconomic deprivation and ethnicity have long been linked to poorer health outcomes4,5 because of a complex interplay between the long-term impact of migration, poor engagement with healthcare, and cultural differences; however, there is also an indication that variation may be related to service organisation.6,7

The challenges for commissioners and CCGs are not only to identify appropriate clinical improvements, but to weave and apply changes in health policy and systems correctly and advantageously.8

From 2016, Barking and Dagenham CCG has focused on the contribution of primary care to the delivery of care for people with diabetes. This article highlights our experiences, including consideration of the operational impact of primary care networks.

Variations in diabetes care highlighted the need for local change

Barking and Dagenham is a borough in East London with a population of around 210,000 people. It is one of the most socioeconomically deprived boroughs in London, with population trends more akin to a developing country (i.e. higher population growth rate, younger population, and a lower life expectancy than the UK average). There has also been a rapid shift in the proportions of various ethnic groups: the 2011 Census showed a large decrease in the white British ethnic group and a large increase in the black African ethnic group.9 Smoking and obesity were more prevalent compared with the UK national average, leading to shorter life expectancy, and increased mortality rate from cardiovascular disease and cancer.10

The prevalence of diabetes in Barking and Dagenham is high—for the 2018/19 period, it was 8.3% compared with a UK national average of 6.9%.11 In addition, the population affected is significantly younger (8.2% are aged under 40 years compared with the UK average, which stands at 4.0%), and more minority ethnically diverse (53.8% versus the UK average of 21.0%).11

Before starting our work to improve diabetes care, we noted there was significant variation in the provision of the NICE-recommended eight care processes for diabetes12 and the attainment of three treatment targets: blood pressure, cholesterol, and glycated haemoglobin (HbA1c). Data from the 2015–16 Quality and Outcomes Framework showed that the percentage of patients with diabetes on the register, whose last blood pressure reading (measured in the preceding 12 months) was 140/80 mmHg or less, varied across local practices from around 54–93%.13 For non-diabetic hyperglycaemia, we had nothing in place to either identify or support individuals. We structured our diabetes improvement scheme to address these challenges.

Implementation of the diabetes improvement scheme in 2016

All 37 practices in Barking and Dagenham participated in the scheme, with initially 10,287 patients with type 2 diabetes involved. Our approach was to apply quality improvement (QI) techniques supported by the use of contemporaneous data. We set initial baselines and targets. Practices identified their lead clinicians and administrator, who were supported with training calls and online educational support. Local leaders empowered practices to take ownership of the project, and to create and innovate solutions to any challenges they faced. This was crucial for the success of the project. Our investment was aligned to the following targets:

  • 60% of patients receiving the eight care processes recommended by NICE
  • 50% of patients receiving the nine care processes (i.e. including retinal screening)
  • a 3% improvement in achieving HbA1c targets
  • blood pressure (BP) and cholesterol targets as recommended by NICE
  • at least 50% of patients offered structured education
  • establishment of a prediabetes register to identify patients at high risk of developing diabetes.

Impact of the initiative on diabetes care processes

Significant improvements were observed after 12 months, and over 24 months the percentage of patients who received all of the NICE eight care processes increased from 28.4% to 67.2%. The programme’s trajectory of key indictors is highlighted in Table 1, and improvements in the eight care processes and HbA1c are represented in Figures 1 and 2.

Table 1: Key programme metrics 2015–16 to 2019–20
Barking and Dagenham2015–16[A]2016–172017–182018–192019–20[B]

Diabetes register (type 2 diabetes)

10,287

10,975

12,610

13,460

14,225

Number of eight care processes completed

2920

5315

8470

9225

10,173

Percentage of eight care processes completed

28.4%

48.4%

67.2%

68.5%

71.5%

Number with HbA1c <58 mmol/mol

5101

6120

7045

7785

7745

Percentage of HbA1c <58 mmol/mol

55.3%

60.7%

60.9%

63.0%

63.6%

Sample size

36

34

37

34

34

[A] 2015–16 is pre-intervention year

[B] For 2019–20 National Diabetes Audit with QMUL adjustment

NHS Digital. Series/collection. National Diabetes Audit. Available at: digital.nhs.uk/data-and-information/publications/statistical/national-diabetes-audit(accessed 15 December 2020).

Figure 1 - Improvement of eight-care-process delivery with rising diabetes register (type 2)

Figure 1: Improvement in delivery of eight care processes with rising diabetes register (type 2 diabetes)

Figure 2 - Figure 2- Improvement in blood glucose control- HbA1c 58 mmol-mol, 2015–2020

Figure 2: Improvement in blood glucose control—HbA1c 58 mmol/mol, 2015–2020

It is notable that the percentage gains achieved are alongside a rise of 3250 in the number of patients with type 2 diabetes on the register (29.6%, from 10,995 to 14,225). Focusing on diabetes has closed our prevalence gap and brought people into the orbit of the care they need, helping to mitigate health inequalities.

Another question we asked ourselves was ‘were we also reducing inter-practice variation?’ The eight care processes data showed a pattern of an increasing average with decreasing standard deviation, which was encouraging, indicating practice variation was decreasing as overall performance improved (see Table 2).

Table 2: Eight care process means and standard deviation
 MeanStandard deviation

2016–17

46.2%

22.8%

2017–18

64.4%

20.7%

2018–19

66.2%

18.5%

2019–20

71.8%

17.3%

What is the role for primary care networks?

Examination of the distribution of results as box plots (see Figure 3) suggested a significant change in the results for 2019–20 along with a continuing QI performance improvement. This raised another question: had something else changed?

Figure 3 - Eight care procesess - distribution of results

Figure 3: Eight care processes—distribution of results

In short, 2019–20 was the formation year for primary care networks (PCNs). Recognising that PCNs were in their infancy (not a fault—just the reality of a new start), there was discussion within the CCG and with the local medical committee on the advantages of rewarding joint practice performance at the PCN level. Given the diabetes QI project’s previous success, it was decided to take confidence from this and use 10% of the total investment to reward PCN-level attainment—specifically directed at reducing variation of the NICE-recommended eight care process scores. In this way, the commissioners sought to encourage mutual support with PCNs. This was the only major change introduced in 2019–20.

A statistical analysis of the variance returned a probability significance of p=0.104 (a t-test, two-tailed, was applied between matched-paired practice results for 2018–19 and 2019–20, with H0 =no statistically significant change in variance of the practice eight care process scores between 2018–19 and 2019–20). This suggests that the probability of the change between 2018–19 and 2019–20 data being random is 10% and there is a 90% chance that the change is a result of different circumstances, which we would argue were the introduction of PCNs in 2019–20 alongside our inclusion of investment directed at intra-PCN cooperation. There is, however, the possibility that the improvement could have been due to a further year of QI effort and discipline and it is not possible to unpick what factors of the introduction of PCNs have had an impact—was it leadership, peer review, peer pressure, specific training, or the PCN-related reward commissioners added to the scheme? Typical of organisational change, it is likely to be a combination of such factors, with specific questions about ‘what has impacted on what’ deserving multivariate analysis at a later time.

Conclusion

Our experience confirms the value of QI interventions supported by investment in primary care. In addition, the organisational innovation of PCNs looks potentially promising in reducing practice variation across populations—a fact reassuring for policy makers and potentially motivating for the new and forming voice of primary care.

Dr Anju Gupta

MRCGP, MSc, DRCOG, GPSI Diabetes: GP Principal and Clinical Director, Barking and Dagenham CCG

Richard Clements

MRes (Management and Organisational Behaviour), Primary Care Manager, Barking and Dagenham CCG

Acknowledgements

National Diabetes Audit, NHSE; Clinical Effectiveness Group, Queen Mary University of London; Harpreet Singh, Data Analyst, BHR CCGs.

Conflict of Interests

The authors have no conflicts of interest.

Key points

  • Barking and Dagenham is one of the most socioeconomically deprived boroughs in London
  • Prevalence of diabetes in Barking and Dagenham is higher and the population affected is significantly younger than the national average
  • There was be significant variation in the provision of the NICE recommended eight care processes for diabetes and the attainment of three treatment targets: blood pressure, cholesterol, and HbA1c
  • A diabetes-care quality improvement scheme was initiated in 2016 and aligned to a range of targets, including:
    • 60% of patients receiving the eight care processes recommended by NICE
    • a 3% improvement in achieving HbA1c targets
  • Significant improvements were observed after only 12 months, with the percentage of the NICE eight care processes completed increasing from 28% to 48%, alongside a reduction in inter-practice variation; by 2019–20, around 72% of the eight care processes were completed
  • Experience at Barking and Dagenham has confirmed the value of QI interventions when supported by investment in primary care
  • A further change occurred in 2019–20 with the introduction of PCNs—following the initial success of the QI project, additional investment was directed at encouraging intra-PCN cooperation and rewarding PCN-level attainment
  • Although PCNs are in their infancy, the organisational innovation that they bring may further support with a reduction in practice variation across populations.

References

  1. Institute of Health Equity. Fair Society, Healthy Lives (The Marmot Review). Institute of Health Equity, 2010. Available at: www.instituteofhealthequity.org/resources-reports/fair-society-healthy-lives-the-marmot-review (accessed 15 December 2020).
  2. Institute of Health Equity. Marmot review 10 years on. Institute of Health Equity, 2020. Available at: www.instituteofhealthequity.org/resources-reports/marmot-review-10-years-on (accessed 15 December 2020).
  3. Mercer S, Watt, G. The inverse care law: clinical primary care encounters in deprived and affluent areas of Scotland. TheAnnals of Family Medicine 2007; 5 (6): 503–510. Available at: doi.org/10.1370/afm.778
  4. Public Health England. Health profile for England: 2017. Chapter 5: inequality in health. London: PHE, 2017. Available at: www.gov.uk/government/publications/health-profile-for-england/chapter-5-inequality-in-health (accessed 15 December 2020).
  5. Public Health England. Public health outcomes framework: health equity report—focus on ethnicity. London: PHE, 2017. Available at: assets.publishing.service.gov.uk/government/uploads/system/uploads/attachment_data/file/733093/PHOF_Health_Equity_Report.pdf
  6. Public Health England. The NHS atlas of variation in healthcare—reducing unwarranted variation to increase value and improve quality. London: PHE, 2011. Available at: fingertips.phe.org.uk/profile/atlas-of-variation
  7. Fryer K, Antony J, Ogden S. Performance management in the public sector. International Journal of Public Sector Management 2009; 22 (6): 478–498.
  8. Glouberman S, Mintzberg H. Managing the care of health and the cure of disease—Part II: integration. Health Care Manage Rev 2001 (winter); 26 (1): 70–84.
  9. London Borough of Barking and Dagenham website. Population and demographic data. www.lbbd.gov.uk/population-and-demographic-data (accessed 10 October 2020).
  10. London Borough of Barking and Dagenham website. Health Profile 2011: Barking and Dagenham. www.lbbd.gov.uk/sites/default/files/attachments/BD-health-profile-2011.pdf (accessed 15 December 2020).
  11. Public Health England. Public Health Profiles. Diabetes. fingertips.phe.org.uk/profile/diabetes-ft/data#page/1/gid/1938133138/pat/44/par/E40000003/ati/154/are/E38000004/cid/4/page-options/ovw-do-0 (accessed 15 December 2020).
  12. NICE. NICEimpact diabetes. NICE, 2018. Available at: www.nice.org.uk/Media/Default/About/what-we-do/Into-practice/measuring-uptake/impact-diabetes.pdf
  13. NHS Digital. Quality and Outcomes Framework (QOF)—2015–16. QOF 2015–16: Prevalence, achievements and exceptions, high dependency and other long term conditions group, at GP practice level v2. Available at: digital.nhs.uk/data-and-information/publications/statistical/quality-and-outcomes-framework-achievement-prevalence-and-exceptions-data/quality-and-outcomes-framework-qof-2015-16 (accessed 15 December 2020).