We recently began a study at our laboratory to determine the patient value of new diagnostic markers, in particular natriuretic peptides, to the management of heart failure. We selected heart failure because it has reached epidemic levels in our population, with many requiring multiple admissions to the hospital to manage their condition.

Our research process started by mapping the current patient journey for heart failure patients. Here we worked with a healthcare consulting company to ensure that we identified and addressed clinical care gaps. This was the most crucial step to ensure the success of the project. We identified two main gaps:

• Early and accurate diagnosis is difficult to establish in heart failure. It requires complex testing. But if heart failure is identified early, the mortality and morbidity drops.
• Follow up care for heart failure patients is not personalised. Because of the interdisciplinary nature of the care team and the different departments involved, care should vary greatly.

To assess the value of early detection and centralised management to patients, we designed two interventions that would see every diagnosis confirmed within one week. To do that, we made B-type natriuretic peptide (BNP) and N-terminal pro-BNP (NT-proBNP) testing the first line diagnostic test, extending its use from the emergency room to primary care. We also set up a heart failure unit where suspected cases can be referred. The unit was also designed to manage follow-up care for these patients, personalising the frequency of visits based on their NT-proBNP results.

We then defined a standard set of outcomes that we would track. We used the Outcome Measures Hierarchy [1] developed by Michael Porter from the Harvard Business School, as well as guidelines developed by the International Consortium for Health Outcomes Measurement (ICHOM) for outcomes measurement in heart failure patients [2] (see Figure 1 below).

With our outcomes defined, it was time to develop the data collection process. Following the ICHOM guidelines, we added these data collection points into our system at different time points throughout the care cycle. We also involved patients in the design of questionnaires and built in multiple check points to ensure that data entry is correct.

Next, we developed a set of patient outcomes and cost benefit KPIs based on different scenarios. Before we started on this project, our readmission rate over 90 days was 37%. We projected that our intervention could reduce it to 28%, 19% or 9%, depending on whether it was minimally successful, moderately successful or very successful.

We have not yet completed the full one year analysis, but data so far is very promising. Fewer patients get re-admitted and the number of re-admissions are also dropping. Our data shows that 30-day readmission rates are down 88%. Readmissions are also down 70% over six months and 45% over one year so far. We have also been able to reduce the number of emergency room visits made by heart failure patients. Patient health has also improved, with fewer patients falling into the New York Heart Association’s Class III category [3], where even less-than-ordinary physical activity becomes difficult.

[1] Porter, E.M., 2010. What is Value in Health Care? The New England Journal of Medicine. 363, pp.2477-2481

[2] ICHOM Heart Failure Standard Set

[3] New York Heart Association Classification

This article is based on the presentation “Value-based innovation in heart failure patient management” at the Roche Efficiency Days (RED) 2018 REDefining perspective in Guangzhou, China.