The ICC involves a number of services, but a key one is health coaching in long-term conditions:
Health coaching involves ‘a regular series of phone calls between patient and health professional . . . to provide support and encouragement to the patient, and promote healthy behaviours such as treatment control, healthy diet, physical activity and mobility, rehabilitation, and good mental health’.
Table 9 shows key dimensions of health coaching interventions.83–86
| Populations | Identification of patients | Technology | Responsiveness | Model |
|---|---|---|---|---|
| Coaching can be preventative or target those with existing conditions. If the latter, this can involve those with a specific disorder, a range of conditions or multimorbidity. Other methods of targeting include a focus on high health-care users or those at a high risk of admission | Patients can come to health coaching through self-referral, identification through routine consultations (or post discharge) or the use of formal risk stratification models | Technology can involve conventional telephone and mobiles, or enhancements such as telemonitoring, videophone, automated telephone support, SMS or combinations of technologies | Coaching can recruit patients through referral from services, or proactively identify patients ‘at need’ or ‘at risk’. The delivery of the coaching itself can be more or less scripted | A variety of models of coaching can be used, based on counselling, CBT, self-management and self-efficacy, or motivational interviewing |
| Target | Practitioner | Intensity | Care context | |
| The targeted outcomes for coaching can include education and information, decision-making, motivation and self-efficacy, self-care behaviours, health-care utilisation, mental health and substance abuse | Coaching can be delivered by peers, trained non-clinical staff, clinicians, or may be automated to various degrees | The intensity of coaching may vary in terms of the length and number of calls per week, the overall duration of contacts and the use of ‘booster’ sessions | Coaching can be used as part of a ‘stand-alone’ intervention, or delivered as part of wider programme of care. Linkage to other services (such as primary care) may also vary |
What is the evidence for health coaching?
A number of reviews have tried to assess the overall evidence. A review87 of the effects of health coaching on adults with chronic disease found 13 studies in a broad range of populations and conditions. Only a minority used telephone health coaching. Benefits were reported for a variety of outcomes, with the most consistent results for weight, physical activity and health status. However, the studies included adults of a range of ages rather than older people. A second review85 found 30 studies of health coaching for long-term conditions and, again, reported evidence of positive effects on a range of outcomes (including self-efficacy, satisfaction and health status). An integrative review88 of qualitative and quantitative research found 15 studies and rated 40% as showing improvement in one or more health behaviours. A review84 specific to telephone coaching services for people with long-term conditions found 34 eligible studies, focused on diabetes mellitus or cardiovascular conditions. Most reported some outcomes in favour of health coaching, although reporting of cost outcomes was rare. The variation in the numbers of included studies in reviews highlights variable definitions in this area, but the overall evidence suggests an intervention that is promising but far from proven.
Other recent studies in the last 5 years also give a mixed picture. A quasi-experiment89 exploring the impact of telephone health coaching on care utilisation reported no impact on emergency admissions, but found savings of US$412 in total costs per person, largely through reduced outpatient and inpatient expenditures. Again, the sample was adults and only a small proportion were aged ≥ 65 years. A second quasi-experiment88 in an adult Medicaid population found the opposite: health coaching was not associated with changes in a range of utilisation measures and expenditures, but did reduce emergency department use. A recent evaluation90 of the Birmingham OwnHealth health coaching service in 2698 patients and matched controls explored impacts of a service targeted at people with heart failure, coronary heart disease, diabetes mellitus or COPD. The analysis found no reductions in utilisation with a nurse-led health coaching service, although other outcomes (such as empowerment and quality of life) were not measured. A large (n = 1535) study of health coaching in patients aged ≥ 45 years with one of three long-term conditions and unmet treatment goals found that blood pressure control improved in the intervention group, but found no other benefits on primary end points.91 A small (n = 43) study92 of health coaching for older patients with multimorbidity in nursing homes in Korea reported benefits in self-management, self-efficacy and health status. A trial93 of 232 patients with long-term conditions and depression found that coaching added only short-term benefits over access to a self-care intervention in an older population (mean age 55 years). A cluster trial94 of 473 patients receiving a practice nurse-based health coaching intervention found no benefits over usual care on glycated haemoglobin (HbA1c) outcomes. A trial of patients95 with coronary heart disease in university teaching hospitals receiving telephone coaching found a significant impact on total cholesterol outcomes at 6 months compared with controls. The PACCTS (Pro-active call centre treatment support) study96 randomised 591 patients with type 2 diabetes mellitus to telephone support from paraprofessionals and found significant changes in only a subgroup of those with poor glucose control at baseline. A trial97 in 436 older patients with chronic kidney disease (CKD) found that coaching by paraprofessionals and supported by a bespoke website led to improvements in health-related quality of life and blood pressure control, and was highly likely to be cost-effective.
The overall picture on the effectiveness of health coaching is complex. There are a number of positive evaluations, but the studies have included a very mixed group of patients and interventions. Clearly, further research is required to assess the impact of this promising intervention and its place in integrated care for long-term conditions, especially given the limited evidence base in multimorbidity,98 which is highly prevalent in patients aged ≥ 65 years.1
The CLASSIC Proactive Telephone Coaching and Tailored Support (PROTECTS) trial was a pragmatic, individual-level randomised trial to evaluate the clinical effectiveness and cost-effectiveness of telephone coaching.
Eligibility criteria
All patients were aged ≥ 65 years, had two or more existing long-term conditions and were assessed as needing some assistance with self-management. We included the following self-reported conditions: asthma, back pain, cancer, CKD, COPD, diabetes mellitus, heart disease, heart failure, irritable bowel syndrome (IBS), osteoarthritis, osteoporosis, rheumatic disease, rheumatoid arthritis, stroke and thyroid problems.
We assessed self-management with the PAM, and included those with PAM levels of 2 or 3 (Table 10).
| Level | Description |
|---|---|
| 1 | Patients do not feel in charge of their own health and care, with low confidence in their ability to manage health and few problem-solving skills or coping skills |
| 2 | May lack basic knowledge about their long term-condition(s) and have low confidence in their ability to manage health, with limited knowledge about appropriate treatments and self-management behaviours. Patients expect their health and social care professionals to be in charge in terms of making decisions |
| 3 | Patients have basic facts relating to their long term-condition(s) and appropriate treatments. Patients will have some experience and success in making changes to self-management behaviour, as well as some confidence in handling limited aspects of their health |
| 4 | Patients have made most of the necessary behaviour changes, although they may face difficulty in maintaining behaviours over time or during times of stress |
Interventions
The intervention was health coaching, as defined previously.82 We describe the intervention according to the TIDieR (Template for Intervention Description and Replication) guidelines99 in Table 11 and show a schematic of the process in Figure 4. The intervention was based on three mechanisms: (1) health coaching, (2) social prescribing and (3) low-intensity support for low mood. Social prescribing is linking patients and resources in the wider community.100,101 Low-intensity support for low mood includes the assessment of common mental health problems, simple lifestyle advice and behavioural techniques to manage mood, and appropriate risk protocols.102,103
| TIDieR category | Description of PROTECTS |
|---|---|
| What | Telephone health coaching: the core telephone and health coaching materials include telephone and associated patient tracking and management software, and scripts for lifestyle support around diet, exercise, smoking and alcohol Social prescribing: advisors had access to local resources in Salford through the Ways to Well-being site Support for low mood: around three core areas (assessment of symptoms, advice and behavioural activation, risk assessment) |
| Who | The intervention was delivered by a health advisor (Agenda for Change band 4 worker) with essential skills in working with information technology and communication, as well as experience of working with the general public, good time management and an ability to work flexibly and under time pressure. The health advisors were supported by specialist nurses and managers within the ICC, with additional advice around mental health and social prescribing from the academic team |
| How and where | The health coaching was delivered via telephone from a central facility |
| When and how much | Proactive, monthly calls of around 20 minutes were made for a period of 6 months, with the option for additional calls to deal with complex patients or issues of risk |
| Tailoring | Health coaching staff were trained to customise the pace and detail of the call to the social context of the individual patient. Provision of support for low mood and access to community resources was provided when appropriate |
| Modifications | There were no major changes to the delivery of the intervention through the study |
| How well | The fidelity of the intervention was assessed by qualitative work with patients and staff and ensured by ongoing clinical supervision |
FIGURE 4 Schematic of health coaching process. PLANS, Patient-Led Assessment for Network Support. (Figure not reproduced here; see the original article.)
Training and supervision
The intervention was delivered by a health advisor (Agenda for Change band 4) who was already delivering health coaching to patients with diabetes mellitus or pre-diabetes mellitus. Training focused on the additional skills needed when dealing with a wider range of long-term conditions and dealing with low mood. A session on long-term conditions was run by a GP from the CLASSIC team, to help prepare for potential queries. Advisors were encouraged to refer people to the NHS Choices website.
Advisors were trained in detecting and working with participants with low mood over a 2-day training session with further updates over time. They were given the opportunity to role play low mood assessments and delivering interventions.They also received a comprehensive manual to aid them in delivery of the low mood component.
The advisors received clinical supervision (initially fortnightly, then monthly from applicant KL) by group for 1 hour. They had an opportunity to discuss challenging calls and risk issues. They were supported in making decisions about how to progress the intervention for each patient who identified as having low mood and in delivering structured low mood intervention such as behavioural activation. The advisors had contact details for the supervisors for immediate concerns.
Outcomes
The PROTECTS trial was a cmRCT,20 for which a large population cohort is recruited and followed systematically over time. Participants were followed up as detailed in Chapter 7. Outcomes used in the PROTECTS trial were prespecified in an analytic plan (see Appendix 1, Tables 52–61).
We set up an appropriate Study Steering Committee for CLASSIC, but as the study was based on modification of an existing service, for which the likely risks are minimal, a Data Monitoring Committee was not used.104
Procedures
Recruitment and assignment of interventions
Participants were enrolled through the CLASSIC cohort. In a standard trial, patients receive information and then provide informed consent to participate. At that point, they are randomised. A significant drawback is that patients are told about different treatments in the different arms (including any new treatment), but only half the patients are randomised to that new treatment. This can cause dissatisfaction.
In the cmRCT, patients eligible for the trial are identified from the cohort and randomly selected. Patients who are randomly selected for usual care continue to be followed up in the cohort and are not informed about the trial or the randomisation. Patients who are randomised to the new treatment are then contacted and offered the treatment. They still provide consent to the new treatment and can decide whether or not they wish to receive it. If patients agree to the new treatment, they are provided with the new treatment and continue to be followed up in the cohort. If patients decide that they do not wish to receive the new treatment, they continue to receive usual care and continue to be followed up in the cohort.
We piloted these procedures in 50 patients to test the rate of uptake of the new treatment.
After assessment of eligibility, we selected patients randomly for health coaching or usual care using appropriate central randomisation through a clinical trials unit to ensure concealment of allocation. In this pragmatic evaluation, there was no blinding of patients or providers. All outcomes were either self-reported or routine data.
Sample size and analysis
At the time of study development there were no bespoke methods for powering cmRCTs, and, following existing cmRCTs, we used conventional methods. We powered the study to have 80% power (α = 5%) to detect a standardised effect size of 0.25 on any continuous outcome measure. Allowing for 25% attrition among participants – and assuming that outcome measures at baseline correlate 0.5 with their follow-ups – 504 patients were needed (252 per arm).
The initial uptake rate was lower than anticipated; hence, we selected a further 252 patients to be offered the intervention. However, within the cmRCT framework all 504 patients offered treatment remained in the treatment group in analysis, including those who declined. In consequence, the effect size between arms detectable at 80% power was 0.39 among those consenting to treatment.
Analysis followed intention-to-treat principles and a prespecified analysis plan (see Appendix 1). In summary, we report the trial and analysis in accordance with the updated Consolidated Standards of Reporting Trials (CONSORT) standards and utilising the extension for pragmatic trials. The main test of the intervention was that the overall main effect of the intervention is zero. Condition group was used as a binary variable. Continuous outcomes were assessed using linear regression, controlling, where appropriate, for baseline values of the respective outcome. Outcomes measured using ordinal scales were treated as continuous variables. Results for non-normal variables (skew or kurtosis > 1.0) were confirmed using bootstrap analysis. Baseline values of outcomes and design factors were included in all analyses. Some additional covariates were prespecified.
Owing to implementation delays, no patient was offered treatment up to 6 months after the baseline assessment and for some the offer was not made until month 12 or later. This caused variations in the duration of time before start of the treatment, ranging from 259 to 513 days. Length of follow-up from the end of treatment to 20-month follow-up was similarly variable. Thus, the trial is considered to have run for > 20 months, with patients receiving treatment at any time within that period.
The cmRCT design provides an estimate of the mean effect in people offered treatment. Compared with a pragmatic trial, which provides an estimate of the mean effect in people agreeing to treatment, the effect is ‘diluted’ by the proportion of patients in the treatment arm who do not consent to treatment. An estimate of the effect size in patients consenting to treatment was obtained through application of a complier-average causal effect (CACE) analysis.105,106 CACE does not increase the power to detect an effect.
Economic analysis
The economic analysis aimed to assess the incremental cost-effectiveness of health coaching compared with usual care.
The primary outcome measure for the economic evaluation was health-related quality of life measured by the EQ-5D-5L,107 a new version developed as a result of concerns over the lack of sensitivity to change of the original. Published English general population preference weightings70 were used to convert responses to a single utility index for each time point.
This was combined with in-hospital mortality information taken from the secondary care utilisation data, applying a utility value of zero to all patients on death. Quality-adjusted life-years (QALYs) were calculated using the ‘area under the curve’ method, assuming linear extrapolation of utility between time points. QALYs experienced in the second year of the trial were discounted at an annual rate of 3.5%, as specified by the National Institute for Health and Care Excellence (NICE) in its reference case.108
Resource utilisation and costs
Resource utilisation and costs were calculated from the perspective of the UK NHS. Patient-level utilisation data were collected from two sources. Information on GP contacts in the previous 6 months was collected from cohort data at 6, 12 and 18 months. Hospital utilisation data were extracted from linked administrative patient records provided by the NHS, divided into emergency admissions (short stays, ≤ 5 days; long stays, > 5 days), elective admissions, elective day cases, outpatients and A&E attendances.
Utilisation data were combined with relevant unit cost data for the price year 2014/15 to calculate total costs. Unit costs not available for this price year were inflated to 2014/15 prices using the Consumer Price Index.109 Costs occurring in the second year were discounted at a rate of 3.5%.108
Unit cost figures were sourced from the Personal Social Services Research Unit’s (PSSRU’s) unit costs of health and social care (2015)110 and national NHS reference costs.111
Health coaching costs
Costs were estimated combining the cost of training and supervising staff, materials and delivery of the health coaching sessions. The intervention was offered to all participants randomly selected, although only 189 received at least one call and were used to estimate costs.
Missing data
Data required for QALY and cost calculation were missing in a small number of cases (n = 2), and were imputed. Missing information on age and sex was sourced from administrative data (sex, n = 6; age, n = 35) or imputed (missing age n = 30), to ensure independence from allocation.112
For missing EQ-5D-5L and resource use data, we used multiple imputation by chained equations to generate 50 imputed data sets, assuming that the data were missing at random. The independent variables specified in the imputation models were age, sex, treatment arm and baseline EQ-5D-5L. To account for non-normality, predictive mean matching was used to ensure values observed in the original data set. Multiple imputation was conducted using Stata’s ICE package and analysis using Stata’s MI package (version 14.2; StataCorp LP, College Station, TX, USA).
Cost–utility analysis
The economic analysis estimates the incremental cost-effectiveness of the offer of health coaching compared with usual care at standard UK willingness-to-pay thresholds.
The primary analysis was based on a comparison on the full sample with multiple imputation. A sensitivity analysis was performed using only the complete-case sample (health coaching n = 206, usual care n = 378). Analysis used Stata version 14.
The incremental cost-effectiveness ratio (ICER) was calculated, adjusting for age, sex and baseline EQ-5D-5L index score.113 To assess uncertainty surrounding the estimates and to account for the typically skewed nature of cost data, incremental costs and QALYs were bootstrapped using pairwise bootstrapping with replacement using 10,000 replications. Cost-effectiveness planes plot these 10,000 bootstrap replications of the ICER estimates to illustrate the uncertainty around the point estimate of the ICER in probabilistic terms. Finally, cost-effectiveness acceptability curves (CEACs) were plotted to represent graphically the probability of the intervention being cost-effective across a range of cost-effectiveness thresholds.