This economic evaluation has been conducted and reported in accordance with the Consolidated Health Economic Evaluation Reporting Standards (CHEERS) publication guidelines and good reporting practices (24).
Economic evaluation overview
Cost, cost-consequence and cost-effectiveness analyses will be undertaken comparing the intervention against usual practice from a public health service perspective. This perspective is justified because ongoing funding for this intervention, especially if it translates into routine practice, will fall on public health services. To further aid decision makers, budget impact analysis, including scale-up cost scenarios will be presented alongside the cost-effectiveness findings. Costs will be reported in 2019 Australian dollars ($AUD). The time horizon for inclusion of relevant costs and consequences will be the course of the trial (34 months). Costs and benefits occurring after 12 months will be discounted using an annual discount rate of 3% in the base-case. Annual discount rates of 0 and 5% will be applied in sensitivity analysis. The conduct, analysis and reporting of the economic analyses will adhere to cost and economic analysis guidelines (14, 15, 25) and Consolidated Health Economic Evaluation Reporting Standards guidelines (15).
Common to all forms of economic evaluation is the analysis of cost. In this study, costing and budget impact assessments will be conducted to quantify how much more it will cost to pursue implementation efforts to affect practice change. Budget impact assessment will translate the health economic findings into more meaningful and relevant results for healthcare decision makers and funders. In its simplest form, economic evaluation involves listing of all cost/benefit implications of the alternatives under consideration, as in cost-consequences analyses (26). Cost consequence analysis is employed in this analysis because it provides information for spending decisions when implementation strategies are complex and are expected to have outcomes that are too disparate to be combined meaningfully. In this trial, there are four primary outcomes (see section Identification and measurement of outcomes). Cost-consequence analyses permit value judgments without having to fully specify a relation between all the different measures of outcomes (11). Cost-effectiveness will depend on the effect of the intervention on care provider behaviour. The greater the difference in expected outcomes between usual practice and the new model of care, and the more widespread the implementation, the more likely a strategy is to be cost-effective. In this study, the likelihood of achieving an outcome difference will be maximised by using a staged process to both understand the barriers to guideline adoption and to develop the implementation strategies (3). All public antenatal services in the three sectors will receive the practice change intervention, including midwifery group practices, midwifery clinics, specialist medical services, Aboriginal Maternal Infant Health Services (AMIHS), and multi-disciplinary teams caring for women with complex pregnancies or identified vulnerabilities.
Trial-based economic evaluation and budget impact assessment
Identification and measurement of outcomes
It has been suggested that one of the ways to improve efficiency in conducting economic evaluations of implementation interventions is to confine studies to measures of the care process or intermediate outcomes (11); for example, change in professional guidance adherence or compliance (3). This approach is based on the premise that the guideline recommendations are cost-effective in and of themselves. In this study, the outcomes measures are confined to the care process for efficiency. The implementation trial has four primary outcomes. They are the proportion of all antenatal clinic appointments (at ‘booking in’, 27–28 weeks gestation and 35–36 weeks gestation) for which women report the following:
-
Being assessed for alcohol consumption and level of risk using the AUDIT-C.
-
Being provided with brief advice related to alcohol consumption during pregnancy.
-
Receiving, relative to their level of risk, the relevant elements of antenatal care for addressing alcohol consumption during pregnancy (advise and refer).
-
Being assessed for alcohol consumption and level of risk using the AUDIT-C and receiving, relative to their level of risk, the relevant elements of antenatal care for addressing alcohol consumption during pregnancy (advise and refer).
Receipt of care will be measured by participant report during a computer-assisted telephone survey conducted after an antenatal consultation, at each of the three time points (3).
A secondary outcome will also be included. For women attending antenatal appointments at ‘booking in’, 27–28 weeks gestation and 35–36 weeks gestation, alcohol consumption since pregnancy recognition will be collected. Outcome measurement will be based on self-report of women using the total AUDIT-C score. AUDIT-C is a validated tool for assessing risk of harm due to alcohol consumption (27).
Identification, Measurement and Valuation of Costs
Cost data pertaining to the development and implementation of the practice change intervention will be collected prospectively using a resource use capture tool in tangent with trial administrative records. The intervention program logic will be used to identify all the relevant costs directly and indirectly associated with the intervention. The cost-capture tool, developed in Microsoft Excel (2013), allows researchers to prospectively document the activity and materials consumed at different phases of the intervention (development, immediate execution and maintenance) from all relevant stakeholders. The cost capture tool includes the following categories 1) labour (health service and non-health service staff, including overheads to allow for additional costs of employment); 2) materials (non-labour cost items such as stationary, education materials, electronic hardware or software); 3) joint costs (incurred in connection with multiple projects, for example the maintenance costs of a website portal supporting different interventions; capital costs such as one-off investments such as the purchase of additional office buildings or motor vehicles); and 4) miscellaneous costs (which include costs not easily classified into the other categories, for example, venue hire, travel and overnight accommodation). To maintain a conservative approach to cost estimation, the non-capital implementation costs will not be amortised.
Resource use valuation will be based on the concept of opportunity cost, that is, the value of the benefit forgone in not employing a resource for a different use. Market prices will be used as a proxy for the ‘value of benefit’ forgone (28).
Costing study
Appendix 3: Table 2 summarises the costs expected to be included in the study. Cost analysis will use measures of arithmetic means, between-group differences and variability of differences (29, 30). Costs will be calculated individually for each sector in the trial, as well as aggregated across all sites. Intervention component costs will be disaggregated to provide insight into the cost of individual practice change intervention strategies.
Table 2
Description of resource use data for inclusion in the economic evaluation
Intervention component | Resource use details | Data collection method |
Intervention strategy development | • Labour time: Health district project/implementation support officer time • Materials | • Resource use capture template |
Leadership and management | • Labour time: Health district project/implementation support officer time; health service clinical staff (management from antenatal services). | • Resource use capture template |
Local clinical practice guidelines | • Materials: Guideline and procedure document development and provision. • Electronic dissemination. | • Resource use capture template |
Electronic prompt and reminder system | • Materials: Online/computer based intervention component. • Electronic dissemination. | • Resource use capture template |
Local opinion leaders/champions | • Labour time: Change champion, clinicians and trainers. | • Resource use capture template |
Educational meetings and materials | • Labour time: Health district project/implementation support officer time; health service clinical staff. • Materials: educational tools and resources | • Resource use capture template |
Academic detailing | • Labour time: Project support officer, clinical service staff time. | • Project administrative records • Resource use capture template • REDCap self-report survey |
Monitoring and accountability | • Antenatal service managers will report, interpret and monitor performance measures for the model of care. • These results will be disseminated to antenatal service staff through team meetings, emails and other usual communication mechanisms. • Performance measures will be built into the existing monitoring and accountability frameworks for antenatal services. | • Resource use capture template |
<< Appendix 3: Table 2. Description of resource use data for inclusion in the economic evaluation >>
Cost-consequence, cost-effectiveness and equity
As outlined above, the range of outcomes measured in the implementation trial is diverse, which lends well to a cost-consequence analysis. The analysis will adopt a score-card approach to show a comparison of the costs and benefits associated with the intervention and usual practice. An economic summary measure is not calculated. A program logic model will be developed to describe all possible inputs (costs) and impacts (consequences) associated with the intervention and usual care (Appendix 1: Fig. 1).
Cost-effectiveness analysis will be conducted subject to assessment of intervention efficacy. The economic summary measure will be an incremental cost-effectiveness ratio (ICER). The ICER represents the additional cost required to achieve an additional unit of benefit (14, 29). For this study, the ICER will be calculated as the incremental cost per percent change in the proportion of participants reporting that receipt of ‘antenatal care for maternal alcohol consumption consistent with guideline recommendations’ was provided to them during their antenatal consultation.
Distributional cost-effectiveness analysis (DCEA) is a framework for incorporating health inequality concerns into the economic evaluation of health sector interventions. Full DCEA requires the distribution of direct health benefits to be estimated from a decision analytic model or trial-based analysis using parameter estimates specific to socioeconomic groups. However, a simplified version providing healthcare decision makers and stakeholders with an evidence-based technique for evaluating whether new interventions can help to achieve the objective of health inequality reduction, can be used when conducting a full DCEA is not practical or feasible (31). To assess the equity implications of the intervention, the use of distributional cost-effectiveness will be explored in the scenario examining scale-up subject to the availability of requisite data (31, 32).
Budget impact assessment
Economic evaluations and budget impact analyses share many of the same data elements and methodological requirements and should be viewed as complementary. However, there are important differences in their methods and use cases (13). The budget impact assessment will translate the results of the economic costing study into financial consequences relevant to decision makers and fund holders within the health districts.
A model will be developed to describe the financial resources associated with usual practice over the course of health districts’ budgeting cycles. This will represent the base case or ‘reference case’. The comparative scenario will model the required changes in health service resourcing that are expected to result from adoption of this alternate model of care, including indirect and downstream impacts on other parts of the health service. Resource use data will be sourced from the implementation trial and costing analysis. All model assumptions and data inputs will be described in full. Justification for the inclusion or exclusion of relevant model parameters will be provided.
The budget impact assessment will adhere to relevant local and international guidelines, as well as recommended formats for presenting the results so they are most useful to decision makers (33) (13).
Sensitivity and uncertainty analyses
All analyses will be subject to one-way and probabilistic sensitivity analysis. These analyses test the impacts of plausible variation in data parameters on the cost outcomes and economic summary measure and provide an understanding of which values are associated with the greatest amount of uncertainty. Differences in costs or outcomes that can be explained by variations between subgroups of patients with different baseline characteristics or other observed variability in effects that are not reducible by more information, will be reported.
In addition, a scenario analyses will be undertaken to explore the efficiency and budget impact of state-wide implementation of the practice change model of care in maternity services across the whole state of NSW.