Health economics

Introduction

This chapter reports on the health economic component of STRETCHED, addressing part of the third study objective, namely, the evaluation of costs and consequences of case management for the care of people who call the emergency ambulance service frequently. The evaluation is based on further emergency contacts (emergency ambulance service calls, ED attendances, emergency admissions to hospital) and costs of intervention and usual care. This health economic evaluation is reported using key sections of the Consolidated Health Economic Evaluation Reporting Standards 202267 checklist applicable to the reporting of CCAs. The evaluation of costs and consequences was undertaken in accordance with the study’s SHEAP (see Report Supplementary Material 4); no deviation from this plan occurred.

Deviation from originally planned analysis

In the original study design and published protocol paper, the use of a cost–benefit framework was initially mooted, with the intention of comparing the implementation costs of case management to the potential benefits of ‘avoiding expenditure’ in subsequent healthcare resource utilisation, and calculation of net present value and internal rate of return. However, the STRETCHED RMG agreed to widen the scope of the analysis to include a broader range of outcomes as part of a more comprehensive CCA that provided a less aggregated overview of all costs and consequences.

Intervention implementation and costs

Responses received to the initial resource questionnaire (see Report Supplementary Material 1) demonstrated that much of the data required were not recorded or available and services struggled to find the time to provide the detail required.

The following unit costs were identified and used in the costing:

• Band 7 mean annual basic pay per FTE for nursing, midwifery and health visiting staff by Agenda for Change band, NHS England is £39,994, corresponding to £3333 per month.
• Band 7 mean annual basic pay equivalent pay in Wales is £37,244, corresponding to £3104 per month.
• The electronic staff records system shows that the mean basic salary for a qualified community-based nurse (Band 5) is £32,949, corresponding to £2746 per month.
• The mean annual basic pay per FTE for administrative and clerical staff is £24,391, corresponding to £2033 per month.

While the aim and overall actions taken as part of case management are the same across sites (see Table 1), the ways these actions are delivered differ from site to site. Based on documents reviewed, interviews and information received from sites, the ambulance service frequent caller case management models differ in both staffing and case management strategies. Table 9 shows the estimated staffing costs (for ambulance services only) per patient case-managed per month for the four intervention sites. While the number of patients who are case-managed per month is approximate depending on resources available and complexity of the support required, the number of cases appears consistent over the 12-month period, with some patients requiring support over many months and some patient cases resolved within the month. Therefore, many patients will be case-managed over several months with resolved cases closed and new patients added to the caseload each month. Data have not been collected measuring staff time taken or other resources required by various staff to case manage each individual patient. Therefore, annual pay rates and contracted hours dedicated to the frequent caller enhanced intervention are used in relation to number of patients supported by the intervention.

Partner agencies supporting case management following referral such as NHS services, social services, voluntary sector and police service staff contributing to MDTs engaged to manage such complex patients did not receive additional funding to facilitate such support. This work was absorbed as part of their workload and no data were available to estimate these opportunity costs.

Multidisciplinary team case management meetings

It was noted during the qualitative interviews that NHS partners contacted when a frequent caller is identified carry out their role for the care management plan as an add-on to their current workload/role, with no specific funding provided for the specific care of identified patients. While we acknowledge this to be an area of potential opportunity cost, no data are available to estimate these costs. Table 10 shows the various NHS, social care, voluntary and police services that could potentially be engaged as part of a care management plan and attend MDT meetings, depending on the individual needs of each patient. It is also noted that, at MDT meetings, frequent users of other services (ED, out-of-hours GPs, police) can also be discussed in some areas, not just frequent caller patients referred by the ambulance service.

Total intervention implementation cost

Intervention implementation cost varies between sites and is difficult to separate from usual care. Based on the information available, it is estimated that case management costs between £25 and £333 per patient per month. This does not include cost to other services and opportunity costs, which remain unquantified.

Healthcare resource utilisation and costs

Cost of resource use included in the primary outcome

A summary of total and mean per patient healthcare resource use and cost associated with healthcare contacts included in the primary outcome can be found in Table 11, with adjusted comparisons.

When adjusted for baseline total number of healthcare service contacts and total baseline cost, together with age, gender, deprivation quintile and ethnicity (collapsed into white/non-white), patients in the case management intervention sites had marginally lower mean per patient healthcare resource use and costs included in the primary outcome compared to patients in the usual care sites. Overall, patients in the intervention sites had −0.13 (95% CI −5.38 to 5.11; p = 0.960) more emergency healthcare contacts (including non-elective hospital admissions, ED attendances, emergency ambulance service calls) compared with control site patients over the 6-month follow-up period, at a slightly reduced mean cost of £243.57 (95% CI −£1972.93 to £1485.79; p = 0.782) per patient. However, there was no statistically significant difference between usual care and intervention groups at the 5% level for mean per patient healthcare service contacts or costs for any of the service use components included in the primary outcome.

Cost of resource use included in the secondary outcome

A summary of total and adjusted mean per patient healthcare resource use and costs associated with healthcare contacts included as secondary outcomes can be found in Table 12.

When adjusted for baseline total number of healthcare service contacts and total baseline cost, together with age, gender, deprivation quintile and ethnicity (collapsed into white/non-white), total healthcare resource use and costs included as secondary outcomes were higher for patients in the intervention sites with an overall mean cost increase of £781.87 (95% CI −£198.40 to £1762.14; p = 0.118) per patient, equating to 2.20 more healthcare contacts (95% CI −0.35 to 4.77; p = 0.091) over the 6-month follow-up period. While the difference in total resource use and cost was not statistically significant, the increases in outpatient appointment numbers and costs as well as number of elective inpatient admissions and ambulance callouts were statistically significant, albeit small overall.

Total healthcare costs

Table 13 summarises overall resource use and cost. On combining all healthcare resource use within primary and secondary outcomes, patients identified as frequent callers in the intervention sites had, on average, five healthcare contacts more than patients in the control sites, at an estimated additional cost of £538.30 (95% CI −£1641.53 to £2718.13; p = 0.628). This difference was not statistically significant at the 5% level.

Healthcare costs by ambulance service

A summary of costs of healthcare contacts by ambulance service can be found in Table 22 in Appendix 4.

Cost–consequences analysis

The results of the CCA are presented in Table 14.

Adjusted and unadjusted comparisons

Given some imbalances observed at baseline (see Chapter 4), we provide here, for completeness, a summary of the differences in unadjusted and adjusted comparisons between study groups.

Without adjustment for baseline total number of service contacts and total baseline cost, age, deprivation quintile, ethnicity (white/non-white) and gender, patients in intervention sites, with an average of five additional healthcare contacts, incurred an average additional cost of £1804.39 (95% CI £107.56 to £3501.22; p = 0.037); this difference would be regarded as statistically significant at the 5% level. The difference in the number of contacts (weighted equally) would also be regarded as statistically significant (p = 0.031).

With adjustment, the estimated additional cost per patient in the intervention becomes £538.30 (95% CI −£1641.53 to £2718.13; p = 0.628).

Similar changes due to adjustment are seen in costs associated with primary and secondary outcomes (Table 15).

To the best of our knowledge, this is the first estimation of costs and consequences of case management for patients identified as frequent callers by ambulance services in the UK. We have transparently reported our analyses, but, as outlined in the discussion chapter, various limitations apply and the overall level of uncertainty in our findings remains high.