This systematic review aimed to identify, summarise, and synthesise existing evidence to estimate the costs associated with DFD management in Australia. Our findings revealed a paucity of contemporary evidence on the financial burden of DFD within Australia, particularly in relation to the frequency and variation of services and resources required. Furthermore, there is variation in the reported cost estimates within the Australian literature. Despite the limited evidence at present, this review found that by adhering to evidence-based guidelines, health outcomes can be improved and can bear significant cost savings for the Australian healthcare system.

The variation of reported DFD cost estimates within the Australian literature could be attributed to the differing methodological approaches seen within the included studies, such as the characteristics of study cohorts, the definitions of care/comparators used, the sources of the cost data, and the analyses performed. While the two cost-effectiveness analyses [22, 26] comparing ‘usual care’ to ‘guideline-based care’ used similar modelling methods, the first study [22] used a hypothetical cohort of patients with diabetes at high risk of developing DFUs, while the second study [26] used a large prospective real-world cohort of people with DFU attending Diabetic Foot Services within one state of Australia. As observed in the latter study [26], the method of using individual patient-level data (as opposed to aggregated data) may have better informed the model parameters for guideline-based care versus current practice (i.e. usual care) thus reducing the risk of sampling bias associated with trial-informed time-to-event parameterisation [22, 26]. In addition, the patient-level data allowed for use of a discrete event simulation model in this study, with the flexibility to add specific attributes to each person simulated in the model [26].

When comparing the reported costs for DFD in these cost-effectiveness studies, the first study [22] estimated a total five-year cost per patient (expressed in 2013 AUD) to be between $15,781 and $19,461 (~ $3,156 to $3,892 per year) for usual care, and between $6,681 and $7,066 (~ $1,336 to $1,413 per year) for guideline-based care. The second study [26] estimated a total three-year cost per patient (expressed in 2020 AUD) of $49,918 (~ $16,639 per year) for current practice (i.e. sub-optimal care) and $48,075 (~ $16,025 per year) for guideline-based care. Even despite the inflation of costs between 2013 and 2020, the largely differing cost estimates reported in these studies may also be explained by their respective definitions of ‘guideline-based care’. The first study [22] defined guideline-based care as per the optimal care program outlined in the 2011 National Evidence-based Guideline: Prevention, Identification and Management of Foot Complications in Diabetes [28], while the second study [26] defined guideline-based care as per the core principles of DFU care outlined in international guidelines [35] and included costs associated with healthcare consultations, consumables (i.e. dressings, pressure offloading devices, footwear) and antibiotics. This study [26] further defined healthcare consultations, whereby all visits must have been ≤ 21 days since the previous visit and ≥ 75% of visits must have documented ulcer classification, sharp debridement, appropriate wound dressings, knee-high offloading, and antibiotics prescribed (only if the wound was classified as infected) [26]. Based on the two studies’ definitions of guideline-based care, particularly in relation to the frequency of healthcare consultations, it is apparent that the latter study [26] was more comprehensive in its approach to DFU care than the former study [22]. Hence why the cost estimates may have been significantly larger in this study. It is also important to note that the earlier study [22] reported costs in 2013 AUD, while the other [26] reported costs in 2020 AUD. Therefore, the cost estimates of the earlier study [22] may not be as representative of the current economic burden of DFD. Importantly, both studies [22, 26] demonstrated overall cost savings and improved health outcomes associated with guideline-based care compared to usual care. Accordingly, it is likely that the increase in outpatient costs to perform guideline-based care in these studies were offset by the reduced costs associated with DFU-related hospitalisation and costs of minor and major amputation procedures [26].

The third study [25] included in this review used probabilistic economic modelling to estimate and compare healthcare costs for chronic wounds (including DFUs) in hospital and residential care settings in Australia. In 2012, it was estimated that the total cost for DFU management in hospitals was more than USD$238 million (~ $18,591 per patient), while the total cost in residential care was close to USD$11 million (~ $21,315 per patient). Based on the average conversion rate in 2012 (i.e.1 AUD = 1.0358 USD) [36], this equates to a cost of ~ AUD$17,948 per patient in the hospital setting and ~ AUD$20,578 per patient in the residential care setting. While this study has shown that a large proportion of costs associated with DFD management are incurred in the hospital system (as opposed to residential care services), it is important to consider that these estimates may not be a true reflection of the cost burden in Australia; a large proportion of the inputs for the economic model (i.e. unit costs and incidence of DFU) were informed from international studies, rather than from Australian data. Furthermore, the reported cost estimates showed large standard deviations, which also adds to the uncertainty of the economic modelling in this study [25].

One limitation of all three studies is the reliance on international data to inform specific model inputs, due to the limited published Australian cost and resource utilisation data available. This lack of local data is likely due to fragmented DFD care provision often seen in Australia, but particularly for Aboriginal and/or Torres Strait Islander Peoples [37]. For example, DFD care is commonly shared in the community by general practitioners, podiatrists and nurses, and/or within hospital outpatient specialist clinics and high-risk foot services in the public health system [22, 38]. This lack of local data is further compounded by the limited rebateable services for DFD care, resulting in difficulties for relevant data to be collated and applied to economic evaluations. When comparing Australian and international economic evaluations, it is apparent that data collection methods in Australia may be lacking, particularly when compared to the US and the UK [39, 40]. A more comprehensive system to record item numbers and resources utilised for DFD care within the overall Australian healthcare system may enable a more realistic and representative cost estimate to be determined. With the inception of the National Association of Diabetes Centres (NADC) High Risk Foot Services database [41], and in combination with International Classification of Diseases codes from tertiary care, this could address some of these knowledge gaps, particularly surrounding service provision and resource utilisation within hospital outpatient services across Australia. However, a gap in resource utilisation still remains with DFD services accessed in the community.

Another finding of this review was the variation between the three studies in which Australian healthcare settings (i.e. cost data sources) were included. The first study [22] that used a hypothetical cohort sourced health system data from the Medicare Benefits Scheme (MBS) and the Pharmaceutical Benefits Scheme (PBS), which only partially covers costs of healthcare services and consumables. The second study [26] that used a large prospective real-world cohort presented two categories of care costs, including outpatient DFU services and event costs for hospitalisation and for minor and major amputations in an inpatient setting. This study sourced health system data from the Independent Hospital Pricing Authority and the PBS. For the hospital outpatient data (i.e. from Diabetic Foot Services) this was obtained from only one state in Australia. Therefore, the cost estimates may not be representative of outpatient DFU services throughout Australia. Finally, the third study [25] used total cost data limited to residential care facilities and hospital admissions, therefore, the estimated DFD care costs did not include costs of services and resources from the broader community or outpatient setting. Interestingly, none of the included studies were able to provide an overall cost estimate of DFD with respect to all aspects of care as they did not capture the multi-faceted level of community care throughout the entire patient journey between sectors and over time.

In reference to other Australian studies that were excluded from this review, two out of the three studies (excluded following full-text review) did not provide specific information or costings for DFD management, while the third [24] was a systematic review that included duplicate data from one of our already included studies [22]. Wilkie et al. [27] aimed to determine the actual cost of wound care using a survey to identify the number, type of wounds and their treatment costs including consumables and labour in Australian hospitals, residential aged care facilities, general practitioners, and community providers. Although the data collected on foot ulcers was categorised by the underlying aetiology (e.g. ischaemic, neuropathic, neuro-ischaemic), it was not clear which of these ulcers were directly attributed to DFD, and therefore, this study was excluded. Rather than an economic evaluation of the financial burden of DFD, Ahmed et al. [42] estimated the prevalence of DFD and the sociodemographic and health-related characteristics among people aged 45 years and over in New South Wales, Australia.

There are many challenges when reviewing economic health evaluations due to substantial variability in the standard of care across and within healthcare systems [43]. Economic health evaluations based on assumptions and probabilities of disease states have inherent limitations. They rely on making projections and estimated costs based on various assumptions. These limitations are due to unpredictable factors such as uncertainty of the actual course of the disease, assumptions of human behaviour such as adherence with preventative measures, or data reliability which may affect the accuracy and reliability of the evaluations. It is also acknowledged that systematic reviews of economic evaluations commonly have wide variations in population characteristics, study settings and healthcare systems, therefore reviews are unlikely to generate a one size fits all analysis regarding cost-effectiveness and their comparators [43,44,45].

While this systematic review was designed to be comprehensive in capturing contemporary data for the costs associated with DFD management within the Australian context, its findings should be considered in relation to several limitations. First, the quality of the evidence in this review is limited by the small number of included studies. While there were only three economic evaluations identified, two were rated as either ‘very good’ or ‘excellent’ on quality assessment, therefore the findings from these studies [22, 26] are likely to be valid. Second, studies eligible for inclusion in the review were exclusively from Australia, and therefore, cost comparisons with other countries were not extrapolated. Third, searches were limited to the last decade, as we wanted to ensure that cost estimates were representative of present day. Therefore, not all economic evaluations conducted in Australia may have been included in this review. Fourth, as only three studies met the eligibility criteria and there was heterogeneity of the data, pooling of cost data in meta-analyses was not possible. Fifth, we did not find any published Australian data concerning indirect costs (e.g. cessation or reduction of work productivity) associated with DFD management, therefore, only the direct costs have been presented.

That being said, there are several notable strengths of this review. A robust and comprehensive search strategy was employed, and pre-determined decision rules were followed throughout all stages [20]. For example, the study selection, data extraction and quality appraisal steps were conducted by two independent reviewers, with conflicts resolved through consultation with a third party. The reporting of data and the results underwent cross-checking by all authors, ensuring we were transparent and unbiased in our findings.

This systematic review has demonstrated the paucity of evidence regarding not only the costs of DFD management, but also the frequency and variations of services and resources required for management of this patient cohort. There is also a lack of data comparing Australians living in metropolitan, regional or remote communities, and in particular the First Nations population. Of the few studies conducted on DFD prevalence in Australia [8, 11, 42, 46], the data obtained has been predominantly from hospital-based diabetes populations and defined geographic areas, so they may not be reflective of the overall DFD burden within Australian communities [9].

Given the detrimental impact of DFD and its substantial financial burden on the Australian healthcare system, there is a pressing requirement for further economic evaluations utilising up-to-date Australian data. The recent publication of the 2021 evidenced-based Australian guidelines for diabetes-related foot disease [47,48,49,50,51,52] may inform future Australian economic evaluations concerning the cost-effectiveness of implementing these guidelines (versus usual care), which may prove invaluable for informing national service delivery and improving health and economic outcomes. Within future economic evaluations, it is also crucial to consider the diverse cultural backgrounds, geographic locations and socioeconomic disparities within Australia, as these factors play an important role in evaluating cost implications. Since the Australian DFD guidelines were adapted from international guidelines and tailored to the Australian context by incorporating considerations for First Nations peoples and those living in rural and remote regions, there is now an opportunity to conduct cost-effectiveness analyses for guideline-based care with the Australian context in mind.

To produce robust baseline data across Australia against which improvement to care, access, management and surveillance can be benchmarked, data capture using compatible systems that incorporate the entire patient journey between sectors and over time would be required. Such data could also be used to compare with other countries.

Ideally, the total cost of DFD should be established to understand the true financial impact of DFD to the Australian healthcare system. Data inputs for economic modelling should include the cost of the multidisciplinary team, equipment/consumables, diagnostic tests, medications, hospital and procedure costs, and labour costs. In the absence of alternative data sources to inform these parameters, an additional approach may be to obtain clinical expert opinion to generate estimates of resource utilisation. Multisource data is required to inform real-world resource utilisation and costs associated with care for patients with DFD. This information would enable policy makers the financial incentive to improve access to optimal care for DFD and ultimately reduce the financial burden to both the patient and the Australian healthcare system.