West Green Surgery wanted to improve the identification and management of diabetes and hypertension in its patients. In November 2017, the practice had a National Diabetes Audit (NDA) achievement of 37% and 28% of patients had poorly-controlled hypertension. The practice had spent much clinical and administrative time recalling all diabetics/hypertensives through telephone/letter in a non-prioritised alphabetical approach, but many continued to miss appointments. Software was developed that recalled patients more efficiently while educating them about their most recent results. It analysed the disease registers and sent personalised messages to patients informing them of their latest health values, comparing them with previous, urging medication compliance, providing URL links to health information or directing them to make an appointment to see an appropriate clinician. The software used protocol-driven artificial intelligence to determine poor/good control, tailor advice and direct patients accordingly. Patients have become more empowered with their own health data and more actively engaged in their health management. The software also analysed the practice’s database, identifying patients at risk of diabetes (HbA1c 42-47).
Get the latest updates
The previous method of patient recall was wasting valuable time and resources. In addition, the chronic disease clinics, including diabetes, had the poorest non-attendance rates, at 14% compared to 5% for other appointments. A major reason for this was patients’ lack of understanding of their disease and the importance of clinical review and essential blood tests. The idea was to make targeted recall more efficient and educate patients about their blood test to empower self-management. This would both save money and improve patient outcomes, resulting in better health and more income for the practice. Evidence showed patients who knew their last HbA1c result were more accurate in assessing their diabetes control and better understood their condition. A study showed that mobile phone intervention led to statistically-significant improvement in glycaemic control and self-management in diabetes. A surgery team member created flow diagrams for the program to use to ensure that the right pathway was chosen depending on the patient’s BP result or HbA1c figures. Bespoke examples of the text message were drafted and discussed with a small group of patients with chronic diseases to select the best types of SMS text message.
Patients’ consent was sought before SMS messages were sent. Those who had not given consent were excluded. The new software performed four functions running alongside EMIS and using the existing NHS mail infrastructure. It allowed clinical template-driven SMS messages to be sent to patients with information about their diabetes and hypertension, giving them specific values and their HbA1c targets, plus health information. It sent password-protected blood test request forms to patients via email and texted the password for those patients where a repeat test was required. It also invited them to see the most appropriate clinician depending on HbA1c score. The app also interrogated the records and sent selective messages to patients at risk of diabetes depending on how many previous raised HbA1c tests they had had. Sending patients the blood test if they required a repeat meant they did not need to return to the surgery, improving uptake of repeat blood tests. The whole practice team was involved in the inception, development and testing. Programmers, designers and developers were hired to make the concept a reality.
The practice has over 620 patients with diabetes, requiring significant time and resources to recall patients for blood tests. Previously, patients had been invited to see the most senior healthcare professional, regardless of the status of their condition. This was not cost efficient or appropriate for over half of the patients. Sending patients personalised information and their actual blood test results, along with online resource URLs, improved their understanding of diabetes. Patients now attend knowing their HbA1c value, what it means, and their previous personalised targets. This has allowed more consultation time to advise on lifestyle measures or recommend changes to medication. Better understanding of their condition has improved the numbers of patients attending chronic disease clinics, with non-attendance rates dropping from 12% to 5%. At least 400 GP appointments per year have been freed up, as 40% of diabetics have good control and do not require a face-to-face follow-up. The remaining patients are advised to repeat the test or see the most appropriate clinician, depending on HbA1c values. An audit two months after starting the project revealed that, from 110 SMS messages sent to poorly-controlled diabetics (>HbA1c 59), 23 patients subsequently repeated the test and had an improvement in HbA1c of 1.6mmol/mol from 73.4 to 71.8. The six-month period the software was used was compared with the same period a year prior and showed an average reduction in HbA1c of 5.85. An audit over a similar time for poorly-controlled hypertensives (QOF >150/90, >140/80 in diabetics) showed 97 patient-specific SMS messages sent, with 47 patients rechecking their BP with the nurse/HCA or online, with 36 BP normalised thereafter and 11 remaining high. This led to an average drop of 11.9mmHg in systolic BP (from 149.9 to 137.9) and a drop of 7.6mmHg diastolic BP (from 90.1 to 82.5). The percentage of well-controlled hypertensive patients increased from 72% to 79%. The NDA audit increased to 45% from 37% in three months and the number of diabetics with HbA1c <59 increased from 283 to 347.
Sustainability and Spread
Haringey Clinical Commissioning Group (CCG) invited the practice to present its findings to fellow GPs. There has also been interest from Haringey QIST team and Islington Federation. In May 2019 a poster presentation was made at the event Celebrating Primary Care. The software works with Windows 7 and 10 and can easily input patient data from the EMIS software system. SMSs are sent in bulk or individually to the list, but with patient-specific information and recommendations set by the practice. The software is scalable and can be tailored to local pathways and needs. Since it uses NHS.NET it can also use emails rather than SMS.
Back to the top