THINK-HF v1.0

• Research type

  Research Study

• Full title

  Feasibility Trial of electronic Health record INtelligent tooling in primary care for quicKer diagnosis of Heart Failure (THINK-HF)

• IRAS ID

  332534

• Contact name

  Claire Lawson

• Contact email

  cl417@leicester.ac.uk

• Sponsor organisation

  University of Leicester

• Duration of Study in the UK

  1 years, 9 months, 31 days

• Research summary

  Research Summary
  Background:
  In 2018, the government issued best practice guidelines on how best to diagnose and treat heart failure (HF), with the goal of helping patients live better and longer. However, there is evidence that many patients face long waits from when they first report symptoms to their GP, to when they get diagnosed and treated. Most of these patients are only diagnosed when they end up in hospital, and they don't do as well as those diagnosed earlier by their regular doctors. People with other illnesses, women, and those from less wealthy backgrounds often face longer waits.

  We want to an electronic 'clinical decision support (CDS) system to a patient's digital health file. The CDS system will give doctors alerts about patients who are at high risk of HF and suggest specific tests, so that patients could be diagnosed earlier. This can help them receive treatment earlier and live better and longer lives.

  Project aim:
  We want to create and try out an electronic CDS system to help primary care doctors detect HF earlier and make sure all patients get equal care.

  What we will do:
  Even though there's good evidence that an electronic CDS system can help manage long-term illnesses better, there are still too many unanswered questions to test our idea fully. Therefore, in this project we want to:
  1. Work with doctors and patients using questionnaires and a meeting to make an electronic CDS system for doctors' offices.
  2. Observe some general practice consultations to understand the experiences of clinicians and patients and how the CDS system might be used in routine care.
  3. Try this CDS system in four doctor's offices. We'll see if doctors use it and get thoughts from staff and patients. We also want to know if the reminder helps doctors decide which tests to do, for which patients, and when

  Summary of Results
  Why this study was needed
  Heart failure is a serious condition where the heart cannot pump blood around the body properly. Many people are not diagnosed until they become very unwell and are admitted to hospital. Earlier diagnosis in general practice can help people start treatment sooner, improve quality of life and reduce emergency hospital admissions.

  However, heart failure can be difficult to spot. Symptoms such as breathlessness, tiredness and swollen ankles are common and may be mistaken for ageing, being overweight or other long-term conditions such as asthma, lung disease or depression. Women, people with several health problems, people from poorer backgrounds and those who find it difficult to access care may face longer delays.

  We wanted to develop THINK-HF, a digital tool for general practice computer systems that could help identify people who may have undiagnosed heart failure and prompt clinicians to arrange the right tests and referrals.

  How we developed the tool
  We worked with patients, GPs, nurses, pharmacists and heart failure specialists to decide which warning signs were most important.

  First, we carried out a series of questionnaires with patients and clinicians. Patients shared their experiences of delayed diagnosis and clinicians identified factors linked to missed heart failure. Together they agreed that delayed diagnosis was more likely in people with other long-term conditions, breathing problems, depression, obesity and poor access to care. They also highlighted important warning signs such as raised heart failure blood test results without follow-up, use of loop diuretic medicines (‘water tablets’), and a combination of heart and lung problems.

  We then asked a national group of heart failure experts to review the possible warning signs using a Red-Amber-Green (RAG) rating process. This helped us decide which signs were most useful, realistic and safe to include in the tool.

  The final version of THINK-HF was designed to look for:
  • Specific heart failure symptoms, such as breathlessness when lying flat, waking at night short of breath and swollen ankles
  • Patterns of illnesses linked to heart failure, including atrial fibrillation, kidney disease, diabetes and sleep apnoea
  • Raised blood test results that had not been followed up
  • Medicines that may suggest underlying heart failure, especially loop diuretics such as furosemide
  • Missing or incomplete heart failure diagnoses in the medical record

  Understanding how GP consultations work
  Before testing THINK-HF, we observed 21 GP consultations across three practices to understand how clinicians use patient records and alerts during routine appointments.

  We found that most clinicians checked patient notes before the consultation, including medications, test results and past history. Symptoms such as breathlessness were often typed into the record by the GP rather than coded in a standard way, which meant some symptoms could be difficult for electronic systems to detect.

  Clinicians told us that existing computer alerts can be frustrating if there are too many of them or if they interrupt the consultation. They said that alerts work best when they are short, relevant and provide clear advice.

  This helped us design THINK-HF so that it would show concise information, suggest appropriate tests such as blood tests and echocardiograms, and avoid creating too many unnecessary alerts.

  Testing THINK-HF in general practice
  We tested THINK-HF in six general practices in Leicester, Leicestershire and Rutland over four months. Together, these practices cared for almost 79,000 patients and served communities with different levels of deprivation and ethnic diversity.

  The tool searched patient records for signs that heart failure might have been missed. It could identify people who had combinations of symptoms and risk factors, people with abnormal blood test results that had not been followed up, and people taking medicines linked to heart failure without a clear diagnosis.

  During the study, 31 new cases of heart failure were diagnosed. Records also became more accurate, with more patients having the correct type of heart failure recorded in their notes. There were fewer people with abnormal test results that had not been acted on, and fewer people taking loop diuretics without further investigation.

  Clinicians reported that THINK-HF was helpful and fitted well into practice after changes were made to reduce the number of alerts. Patients told us that symptoms are often overlooked or mistaken for other causes, reinforcing the need for earlier recognition.

  What happens next
  THINK-HF appeared safe, practical and acceptable to use in routine care. The next step is to test it in a larger NHS study to see whether it can reduce delays, improve earlier diagnosis and reduce the number of people who are first diagnosed with heart failure in hospital.

  Patients and carers will continue to be involved in improving the tool and making sure it works well for different groups of people.

• REC name

  HSC REC A

• REC reference

  23/NI/0156

• Date of REC Opinion

  15 Nov 2023

• REC opinion

  Favourable Opinion