Tailoring childhood immunisation services in a socioeconomically disadvantaged community in New South Wales, Australia: a cost analysis

Keywords: community health: services, equity, health care: costs, immunisation programs, primary health care.

Immunisation protects children against potentially deadly or debilitating diseases, including measles, diphtheria, pertussis and polio, and strengthens community resilience through herd immunity (Brisson and Edmunds 2003). In the USA, the Centers for Disease Control and Prevention estimates that for each $1 spent on childhood immunisations, there are approximately $5 of savings in future direct medical costs and avoided productivity losses (Philipson et al. 2017). Reports from the Australian Immunisation Register (AIR) indicate that Australia is close to achieving its aspirational target of 95% immunisation coverage for 1–5-year-old children (Australian Government Department of Health 2020a). As of September 2021, 94.8% of all 1 year olds, 92.6% of all 2 year olds and 95.1% of all 5 year olds were fully immunised (Australian Government Department of Health 2020a). Despite these high rates, pockets of lower immunisation coverage persist. Australia’s National Immunisation Strategy (2019–2024) prioritises actions to improve immunisation rates in areas where coverage is low and to facilitate equitable access to immunisation for all Australians, regardless of financial or geographic barriers (Australian Government Department of Health 2018).

Reaching pockets of under-immunised children require tailored strategies that address the unique barriers experienced by communities. The World Health Organization developed the Tailoring Immunization Programmes guide (TIP) to assist public health services identify areas of low coverage, to gain an understanding of barriers to immunisation uptake, and to develop effective tailored strategies for improving coverage rates. Since 2017, TIP has been gradually rolled out in Maitland, NSW, a regional city of almost 80 000 people located 170 km north of Sydney, with relative socioeconomic disadvantage (Australian Bureau of Statistics 2018).

The program focused on children in the Australian Bureau of Statistics Statistical Area Level 2 (Australian Bureau of Statistics 2016) zones of Maitland East and Maitland West. AIR data uncovered relatively high numbers and rates of children at least 30 days overdue for at least one scheduled immunisation (15.4%) in 2016. One-year-old children accounted for 37.8% of these cases (Thomas et al. 2018).

Applying the TIP approach, Public Health Unit (PHU) staff from Hunter New England Local Health District (HNELHD) identified factors driving low immunisation rates in this area: socioeconomic hardship and access barriers to health services (Thomas et al. 2022). A tailored ‘Three Step Process’ for 1-year-old children was developed and implemented by PHU staff and Child and Family Health Nurses (CFHNs; Thomas et al. 2018).

PHU immunisation nurses monitor the AIR data for children in target suburbs who appear as overdue for routine immunisations, entering them into the program. In Step 1, the child’s AIR record is reviewed, with details updated where applicable. For those children with a regular GP, PHU nurses contact the practice to recommend the child be recalled for overdue immunisations and/or to update the AIR. Parents of children without a regular GP are sent personalised reminder letters, advising them to seek immunisation through a local GP or the council clinic. If Step 1 does not result in the child becoming up to date, children progress to Step 2, where they are offered an outreach clinic appointment. If the child remains overdue, they progress to Step 3, where they are offered home visits. Steps 2 and 3 are provided by CFHNs. In some instances, nurses made discretionary decisions on when to progress children between the steps, based on individual circumstances (i.e. lack of transport, large number of other children, poor mental health of the mother, substance misuse). TIP services are above and beyond usual care, which would usually involve generating lists of children who appear as overdue on AIR across HNELHD and sending standard government reminders. In Maitland, immunisations were available through general practice and a local council clinic. The CFHN service did not offer immunisations, but rather, checked children’s records, and provided advice and information.

Decisions on the continuation and/or expansion of the TIP program required information on the resources the program consumes. Such information allows decision-makers to assess whether the use of resources in each TIP step is favourable, given resources are limited. The health service providing TIP has not previously had information available to allow a comparison of costs against outcomes.

To assist pragmatic decision-making in an operational health service, the aim of this study was to analyse the economic costs of the TIP program, where ‘economic costs’ are the opportunity costs of the resources expended in the running of TIP, which could be used in the production of alternative health services. Results are of interest to decision-makers when considering program value and efficiency, and for informing potential scale-up.

To the authors’ best knowledge, this is the first economic study of a stepped immunisation program for 1-year-old children in Australia.

This study was undertaken from the cost perspective of the TIP provider, public health funding agency, HNELHD. The time horizon for the study was 1 year: 1 June 2020 to 31 May 2021.

As TIP was conducted within standard immunisation services, TIP costs were wholly incremental, representing the difference between running and not running TIP.

The study represents a 12-month cross-section of activity generated by TIP. It was aligned to the implementation of TIP in the Maitland area, permitting the prospective identification and measurement of the resources associated with the program.

Two datasets were created during the study period:

1. A custom-built database identifying and measuring the economic costs associated with TIP

2. A custom-built database of TIP cases for the study period, flagging their receipt of TIP services for any of the steps

The TIP cases dataset was constructed from historical records and included immunisation outcomes for 6 months beyond the study period (to 30 November 2021) to allow for a lagged effect from service provision (Fig. 1).

Fig. 1.  Study period and data collection.

For (1), a ‘bottom-up’ costing method was employed to estimate the resources to deliver TIP. Program costs were collected for each month’s activity separately for cases in Step 1, Step 2 and Step 3 (Supplementary material S1). Costs were also collected for TIP ‘Clinical Coordination’ – management and administration. A subcategory of Step 1 costs was created for activity occurring when children flagged on AIR as overdue were found to be ‘up to date’ (Step 1A: AIR Review/Corrections). These activities involved PHU nurses reviewing AIR records, contacting parents and/or general practices, and, if applicable, updating or correcting the AIR. Children who were up to date then exited the program. Costs for this program segment were attributed to the cases that were exited. Children could exit the program at any time for other reasons, including parent’s declining immunisation or if the family moved outside the area. Step 1 costs related to reminders for children confirmed as overdue were placed in subcategory, ‘Step 1B: Personalised Reminders’.

Costed resources included: the amount of time spent implementing the intervention, travelling costs for meetings, materials used, personalised reminder letters and injection consumables. Staff involved with implementing TIP included: PHU Immunisation Clinical Nurse Consultant (CNC) and an Aboriginal Immunisation Health Worker, two CFHNs valued at the grade of clinical nurse specialists (CNS) and a PHU Data Analyst who extracted the AIR data used by the nurses for program monitoring.

A further cost was purchase of a dedicated vaccine refrigerator to safely store vaccines at the outreach clinic, amortised over 3 years. All resources were valued in 2021 Australian dollars (Australian Bureau of Statistics 2021). Vaccine costs for Steps 2 and 3 were omitted, as they are provided free of charge by the Commonwealth Government (Australian Government Department of Health 2020b) and are not a cost to HNELHD.

References for resource valuation included wage awards (Industrial Relations Commission of New South Wales 2021a, 2021b, 2021c); catalogues for medical materials, and service receipts for stationery and capital costs (vaccine fridge). Travel costs for meetings were based on the Australian Tax Office’s car travel expenses allowance of 72 cents per kilometre (Australian Tax Office 2022); phone calls were costed from the Telstra Business Plan (Telstra 2020). ‘On-costs’ (including holiday pay, superannuation) were estimated from a local published source (University of Western Sydney 2021) and added to labour costs. Overhead costs (including electricity, security) were included based on labour costs at a rate of 27.5% (Greater Metropolitan Health Services 2016). A full list of unit costs and parameters and their sources appears in Table 1.

Table 1.  Unit costs and parameters.

The custom-built database of TIP cases included all children who received a TIP step service during the study period, including those whose AIR records required updates. The database included children that had commenced TIP prior to the study period and received Step 2 or 3 services between 30 June 2020 and 31 May 2021. For cases still in TIP at the end of the study period, AIR data were collected to the time of writing (30 November 2021) to observe later positive outcomes.

Ethics approval for the study obtained from the Hunter New England Human Research Ethics Committee (2018/ETH00185: ID016220).

TIP services were provided for 139 cases; 130 entered during the study period, and nine commenced before 1 June 2020. Of the 130 cases that entered Step 1, 47 (36.2%) were excluded, being found to be up to date or declining immunisation. The other 83 received Step 1B: Personalised Reminders. Of these, 37 (44.9%) became up to date in Step 1 (26 in study period, 11 in post study period). Twenty-two cases progressed to Step 2 in the study period. Four others, that had entered Step 1 before 1 June 2020, also entered Step 2 in the study period. Among these 26 cases, 14 became up to date in Step 2 (12 in the study period; two in the post study period). Five cases entered Step 3 during the study period, all of which were already in Step 2 before the study period. At the end of the study period, all other cases remained in Steps 1 or 2, or had left the program (Fig. 2). In total, 56 cases (43 in study period, 13 in the post study period) had become up to date after receiving TIP services.

Fig. 2.  TIP cases (n) by step activity (1 June 2020 to 31 May 2021).

Table 2 shows the total economic cost for the program over the study period as $34 250 or $246 per case, and $612 per case becoming up to date. Respective total costs for Clinical Co-ordination, Step 1 A, Step 1B, Steps 2 and 3 were: $17 650, $5065, $3643, $4881 and $3011. Labour was the largest cost component, and was highest for Clinical Coordination and Step 1 A. Step 3 had the highest per case costs, given the greater per case resources with a home visit from two nurses (as per health policy).

Table 2.  TIP economic costs (A$) and outcomes (1 June 2020 to 31 May 2021).

Table 3 contains results of one-way sensitivity analyses. The first shows that if the labour of the immunisation CNC can be performed by a lower cost CNS, the total cost and cost per case becoming up to date would be reduced by 4.7%. The savings effect would be much larger for Step 1A: AIR Corrections/Exclusions (12.7%) and Step 1B: Personalised Reminders (9.8%). Holding monthly meetings by videoconference, thus eliminating travel costs, could reduce Clinical-Coordination resource use by 6.6% or overall, by 3.4%. This later change has already been implemented.

Table 3.  TIP economic costs (A$): sensitivity scenarios (1 June 2020–31 May 2021: 139 cases).

This study reports the cost to an operational health service of implementing a tailored strategy to improve childhood immunisation rates in a geographic area of low coverage. It provides decision-makers with information on the resources required to deliver each aspect of TIP-information that was not previously available.

The highest proportion of costs were for labour, particularly in the Clinical Coordination and Step 1A: AIR Corrections/Exclusions. Most cases that became up to date reached this status immediately after Step 1B: Personalised Reminders.

The three step process has re-oriented existing services rather than building a new team with additional staff and vehicles. PHU and CFHN nurses refocused resources on a target group of disadvantaged children using personalised reminders, outreach and home visiting. In their systematic review, Ozawa et al. (2018) found reminders to be the most common interventions used to improve coverage rates in high-income countries. The authors found that strengthening health service delivery was the least costly of intervention strategies, suggesting changes to policy and practice at a system level can be effective in improving rates (Ozawa et al. 2018).

Interventions that immunise large numbers of children, as in community-based programs, are likely to be more efficient than those targeting small groups. However, if the aim is to maximise immunisation rates across a given age cohort, it may be necessary to deliver tailored interventions to improve equity of access for those families that require more flexible services that meet their individual needs and circumstances. In this case, tailored programs, such as TIP, are likely to be less efficient than community-based programs; however, their ability to lift immunisation rates in pockets of low coverage where structural and access barriers exist can help achieve national targets and strengthen community resilience. This is also important, because high coverage rates and herd immunity in one area will not necessarily provide herd immunity in nearby areas of low coverage, particularly if there is little mixing of the groups (Tuckerman et al. 2008).

TIP requires an incremental cost for health services. Rather than replace usual care, it is in addition to it. Although each step has a progressively higher ‘per case’ and ‘per case becoming up to date’ cost, the ‘gateway’ approach means that fewer cases require the intensity of Step 2 and Step 3. There are possibilities for reducing the cost of TIP with the substitution of CNS for CNC nurses and by using video-conferencing instead of travelling to face-to-face staff meetings. A TIP program could operate with greater efficiency if the CFHN service prioritised immunisation for all children as core business; and GPs and federal government agencies ensured the accuracy and timeliness of AIR data.

It is possible that other factors influenced childhood immunisation rates in the target group; for example, the ‘No Jab No Pay’ legislation. However, previous analysis of AIR data found little improvement in immunisation rates in our target group after 1 year of its implementation. This may be because parents were not opposed to immunisation, but rather, experienced significant social and structural barriers to accessing health services (Thomas et al. 2017).

The average time for children to progress from ‘overdue’ to ‘fully up to date’ was approximately 9 months. Although this lag may seem long, the average number of injections initially overdue among children serviced by TIP was 4.3 (range 1–9). An appropriate catch-up schedule can require several appointments over 6–12 months (Australian Government Department of Health 2022). Many families in the program were found to have conflicting priorities that required rescheduling of immunisation appointments, which also had an impact on the time required to bring their children up to date (Thomas et al. 2022).

As a pragmatic economic study of a real-world public health program, this study had limitations. Clear paths of causation between the TIP intervention and becoming up to date were not identified. Except for children in Steps 2 and 3, the study had no patient data confirming the influence of TIP on families’ decisions to seek immunisations. Future studies of services, such as TIP, should collect this information. Full immunisation rates for the TIP target group have increased in East and West Maitland from 62.3% in 2016 (just before the implementation of TIP) to 86.2% in 2020 (Thomas et al. 2022), but this alone is not evidence of the effectiveness of TIP.

This study measured the costs of implementing TIP that are additional to usual care. It was not designed to compare the cost or outcomes of TIP with alternative services that also target groups who do not respond to community-based immunisation programs.

The study did not measure downstream cost savings arising from increasing the rate of fully immunised children, nor did it measure potential secondary benefits from TIP, including improved access to primary health care with early detection of health problems, timely referrals to appropriate services and associated improvement in quality of life.

The strength of any causal relationship between TIP and rates of full immunisation cannot be estimated with certainty; however, coverage of the 1-year-old age group is increasing across East and West Maitland (Thomas et al. 2022). It is feasible that the implementation of TIP has contributed to improved equity of access to childhood immunisation by offering more flexible options to families who may be struggling with social and structural barriers. This study has provided information to the health service on the cost of implementing a tailored program for a target group, helping to ensure that scarce resources are allocated to service those most in need. It has also identified ways in which TIP could be made more cost-effective. Decision-makers can use this information to consider whether the investment in TIP is favourable.

Supplementary material is available online.

The patient data used in this publication is not available for distribution. Cost data can be made available on request.

The authors declare that they have no conflicts of interest.

This project was supported by the Australian Government’s Medical Research Future Fund (MRFF) as part of the Rapid Applied Research Translation program. ST and XD-G were funded by New South Wales Health through the Prevention Research Support Fellowship.