Evaluating health and wellbeing coaching interventions

Kate Henry (Know Your Own Health), Kay Leedham-Green (Faculty of Medicine, Imperial College London), and Austen El-Osta (Director of the Self-Care Academic Research Unit at Imperial College London) explore three evidence-based measurements for assessing the impact of health and wellbeing coaching interventions.

Why Evaluate?

Health and wellbeing coaching is now embedded in many parts of the NHS, supporting people to take a more active role in managing their health and wellbeing. But how do we know if it’s working? 

A paper published in July¹ by Know Your Own Health and Imperial College London used a variety of measures that collectively demonstrate the impacts of a health and wellbeing coaching approach on patients with long-term conditions or multimorbidity. This includes mental as well as physical health conditions. 

While this study focuses on one particular coaching approach (the ‘StACC’ model), these measures provide a useful framework that could be applied when evaluating any health and wellbeing Coaching intervention being used to support the NHS England and personalised care agenda.

Patient Activation: A key indicator

At the heart of health and wellbeing coaching in the NHS is patient activation – how confident, knowledgeable, and able a person feels to manage their health and wellbeing.

Why does this matter? Studies show that people with higher activation levels experience better health outcomes and use health services more effectively2,3.  Using an activation questionnaire before and after coaching can therefore give an indication of whether patients feel more confident to self-manage, the stated aim of the role4,5.

That said, activation scores should be used only as a proxy measure: improvements in individual scores do not necessarily in themselves translate into sustained improvements in patients’ ability to self-manage effectively. Evaluations that combine activation with other measures provide a more reliable picture.

Beyond Activation: Other Indicators

If increases in activation mean that patients are more confident and able to self-manage effectively, what other impacts would we expect to see?

Patient-reported outcomes and feedback

Is coaching improving people’s day-to-day health and wellbeing?

Formal clinical recording of measures such as weight, HbA1c, blood pressure may be the most reliable way to assess various clinical improvements, but these can be difficult to capture routinely before and after a coaching intervention. Instead, the Know Your Own Health/Imperial evaluation used self-reported outcomes, through Completion Reports and Feedback questionnaires, to collect these and a range of other impacts.

Self-reported outcomes can provide valuable insight into how patients feel and what changes they notice, e.g. confidence, mood, sense of control, ability to work, etc, and what actions they have taken to improve their own health and wellbeing outcomes. Although less reliable for objective clinical changes, which can additionally be recorded where feasible, patients are nevertheless often aware of changes in weight, blood pressure and even HbA1C, where relevant, so self-reported clinical changes are still valuable in this respect, even if read with this limitation in mind. 

Collecting consistent data across all patients referred to coaching (including completion rates and drop-outs) also helps to create a clearer picture of what works, for whom, and under what circumstances.

Health Service Usage

Another question is whether coaching helps to reduce unnecessary reliance on clinical services. 

Comparing health service usage in the 12 months before and after coaching (e.g. GP appointments, A&E visits, admissions) for a cohort who have completed the intervention can provide useful insights. 

While changes in health service usage may result from a range of individual or wider or external factors, a sustained decrease in service usage following coaching, especially when combined with an increase in activation levels across the cohort and positive patient-reported changes in outcomes, suggests an association between coaching, improved outcomes and reduced reliance on clinical services (as intended by the intervention). See graphs below for illustrations of changes in PAM levels and changes in health service usage.