The Australian Diabetes Foot Registry: From Conceptualisation to National Collaboration

ABSTRACT

The Australian Diabetes Foot Registry (ADFR) is a national prospective, longitudinal clinical quality registry established in 2020 to standardise data capture and reporting in Interdisciplinary Diabetes High Risk Foot Services. We describe the ADFR's methodology and early implementation outcomes. Stakeholder consultation informed development of the minimum dataset, governance framework and data‐capture workflow. Centrally supported ethics and jurisdictional approvals facilitated site onboarding. Clinical, management and outcome data were collected, de‐identified and amalgamated in a national repository. Implementation outcomes included recruitment metrics, case ascertainment, mandatory‐field missing data, reporting cycles, conference presentations and inaugural report feedback. Seventy‐four sites expressed interest and 49 sites completed implementation. Median time from application to commencement was 7 (IQR 4–12) months under a nationally recognised ethics approval. By mid‐2025, 10 919 participants were registered, comprising 13 150 care episodes and 22 383 ulcerations. In 2024–2025, mean service‐level case ascertainment was 80%, and the mandatory‐field missing data rate was 0%–7%. Four annual reporting cycles have been completed, all services strongly endorsed participation after the inaugural report, and data informed 29 conference presentations. The ADFR demonstrates a feasible, scalable national registry model, enabling timely audit and benchmarking and actionable reporting to support quality improvement, research and collaboration.

Key Points

• Australian registry established for interdisciplinary diabetes foot services.

• Pragmatic methods and accessible systems offer a generalisable model.

• Implemented across 49 sites and enrolling 10 919 participants within five years.

• Benchmarking, quality improvement and research support quality care delivery.

1. Introduction

One in 20 (1.5 million) Australians are living with diabetes, with the prevalence increasing [1], and one‐third of these individuals live with diabetes‐related foot disease (DFD) [2]. DFD encompasses peripheral neuropathy, peripheral artery disease, infection, ulceration, Charcot neuro‐osteoarthropathy, gangrene and amputation [3]. The impact of DFD on healthcare systems is substantial and has recently been reported as a leading cause of national and global disease burden [4]. This is primarily driven by the comparatively high rates of poor quality of life, hospitalisation and amputation experienced by people living with DFD [4]. In Australia, the national incidence of hospitalisation and minor amputation for DFD has also increased significantly over the last decade [5], with foot infection now the leading cause of excess bed days among diabetes‐related hospitalisations [6]. Furthermore, Aboriginal and Torres Strait Islander Peoples are up to six‐fold more likely to develop DFD than non‐Indigenous Australians [7, 8].

Increasing availability of specialised interdisciplinary diabetes foot services has been a key strategy to reduce the burden of DFD globally for decades [9]. More recently, interlinking data‐driven outcome monitoring with accreditation of these services has proven successful in improving DFD care and reducing its burden in European countries [9, 10]. In Australia, a specific accreditation programme for Interdisciplinary Diabetes High Risk Foot Services (iHRFS) was launched by the National Association of Diabetes Centres (NADC, a division of the Australian Diabetes Society) in 2018 [11], underpinned by eight widely endorsed standards covering staffing, governance, evidence‐based management, access, continuity of care, equipment, wound care and quality improvement [12, 13]. However, many iHRFS around Australia subsequently identified challenges in meeting the quality improvement standard (Standard 8) due to the need for ongoing local data collection, and it was in this context that the centralised Australian Diabetes Foot Registry (ADFR) was conceived.

High‐level epidemiological data are crucial for comparing national health outcomes with global benchmarks and for identifying variations between states and regions within Australia for any condition [14]. The Australian evidence‐based guidelines for DFD identified significant gaps in data availability across several DFD domains, which hindered comprehensive evaluation of clinical management and outcomes both locally and nationally [15]. In addition, available DFD data often focused specifically on hospital‐admitted care and outcomes, such as amputation rates [5, 16], despite most people with DFD being managed in outpatient settings with a growing emphasis on hospitalisation avoidance. It follows that improving outcomes in this cohort requires a better understanding of differences in the structure, resourcing and clinical practices of iHRFS. Notably, before the establishment of the ADFR, data collection in these services was largely institution‐dependent and therefore inconsistent or non‐existent.

The primary objective of the ADFR was therefore to develop a standardised Australian data repository for iHRFS, where people typically present with diabetes‐related foot ulceration, infection and/or Charcot neuro‐osteoarthropathy. The primary aim of this paper was to present the methodology developed for the ADFR and early implementation outcomes, including site onboarding, participant recruitment, case ascertainment, data completeness, reporting outputs, and site feedback.

2. Materials and Methods

2.1. Study Design

The ADFR is a prospective, longitudinal, national repository of demographic, service, clinical/biomedical, DFD assessment, management and outcome data collected for people with DFD attending Australian iHRFS.

2.2. Ethics and Governance

The ADFR was developed, governed and funded under the auspices of the Australian Diabetes Society (ADS) and the National Association of Diabetes Centres (NADC). The ADS is the peak national medical and scientific body in Australia for diabetes. The NADC is a division of the ADS, a national collective of organisations involved in diabetes care, and provides several accreditation programmes for these organisations. St Vincent's Hospital Sydney is the research sponsor and provides the data and research infrastructure to enable data management and security as per protocol.

Following establishment of the NADC Australian Diabetes Foot Network, which was involved in developing and launching the NADC Collaborative iHRFS Standards and accreditation programme in 2018, an ADFR Advisory Committee was formed. This Committee comprises members from the governing organisations (ADS, NADC and St Vincent's Hospital Sydney) plus national DFD clinical and research experts and provides strategic and scientific support to the ADFR.

External expert advice from existing national registries, including the Australian National Diabetes Audit and the Australian Diabetes Data Network, as well as the New South Wales Agency for Clinical Innovation, was sought to inform development of the study protocol. Data management and security processes were advised by the St Vincent's Hospital Sydney security and risk team, solutions architect, web developer, systems lead and REDCap administrator, as well as a research data expert.

A national ethics approval was sought centrally under the National Mutual Acceptance (NMA) research ethics scheme in July 2020 and granted by the lead Human Research Ethics Committee (HREC) at St Vincent's Hospital Sydney (approval number 2020/ETH01087). The NMA scheme is recognised by publicly‐funded health services across Australia with some exceptions [17]. Separate ethics review was required for the Northern Territory, approved by the Northern Territory Department of Health and Menzies School of Health Research HREC (approval number 2021‐4131) and Central Australian HREC (approval number CA‐21‐4051), and St John of God Midland Hospitals, approved by the St John of God Health HREC (approval number 2160).

Governance approval for the project at each participating site is sought through submission of a site‐specific assessment by the Site Principal Investigator to the local research office. Throughout this process, ADFR staff support sites via virtual in‐services, responsive assistance and shared resources, including centrally drafted applications and supporting documents. Participating sites also enter into a Data Agreement with the Australian Diabetes Society, which outlines the terms and conditions of data sharing.

The consent approach was determined by ethics review and state legislation, such that participating sites in Western Australian and the Northern Territory employ a waiver of consent and otherwise opt‐out consent is used. Opt‐out is facilitated by the display of standardised posters in prominent locations at each participating site, such as waiting areas and treatment rooms, and potential participants are encouraged to raise any questions or concerns about the project without risk of prejudice. The National Health and Medical Research Council of Australia has specific requirements that must be met to use an opt‐out or waiver of consent approach [18].

2.3. Registry Development and Data Fields

The ADFR was informed by existing national and international standards and guidelines. Data fields were initially drafted in alignment with quality improvement recommendations from the 2018 NADC Collaborative iHRFS Standards [13]. Fields were further refined using the Diabetes Feet Australia Australian Diabetic Foot Ulcer Minimum Dataset Dictionary [19], Australian Bureau of Statistics classifications [20], the United Kingdom National Diabetes Footcare Audit [21] and International Working Group on the Diabetic Foot guidelines [22]. Structured feedback was obtained from the NADC Australian Diabetes Foot Network, and the ADFR Advisory Committee subsequently scrutinised each field to balance data yield against collection burden. Pilot sites tested data collection for face validity, which led to minor revisions and finalisation of the ADFR and data dictionary prior to project commencement in 2020.

Subsequently, two major amendments to the data dictionary were implemented in 2021 and 2024 to address user feedback and to reflect updates in national and international guidelines. The current ADFR Data Dictionary is publicly available as a living document on the NADC website (https://nadc.net.au/hrfs‐data‐collection/) [23], and covers 74 data input fields across eight domains: demographic, service, clinical/biomedical, ulcer assessment, active Charcot foot assessment, management, ulcer outcome and active Charcot foot outcome (see Figure 1). The data dictionary defines each field according to compatible REDCap field settings, Australian Institute of Health and Welfare Metadata Online Registry (METeOR) items [24], specific definitions and the suggested utility for each field.

FIGURE 1.

Overview of ADFR data domains, fields and timing of completion per participant. *Based on entering data into the centrally hosted REDCap server. Includes non‐mandatory fields and fields that may be hidden by branching logic. Excludes automatically filled, calculated and descriptive fields. ^{^}Recorded per foot complication (ulceration and/or active Charcot foot) with forms able to be repeated at any timepoint within each care episode.

2.4. Participants and Settings

Inclusion criteria for eligible participants comprised a diagnosis of diabetes mellitus (any type), one or more active foot ulcerations and/or active Charcot neuro‐osteoarthropathy and willingness to have data collected. Foot ulceration is defined as a break in the skin that involves as a minimum the epidermis and part of the dermis and is distal to the malleoli [3]. Active Charcot neuro‐osteoarthropathy is defined as the presence of a red, warm, swollen foot with characteristic osseous abnormalities on imaging in a person with neuropathy [3].

Eligible participating sites were any iHRFS in Australia. An iHRFS is defined as a service providing interdisciplinary management for people with limb‐threatening DFD, including foot ulcerations, active Charcot neuro‐osteoarthropathy and foot infections [25]. The NADC Collaborative iHRFS Standards recommend that minimum staffing comprise a diabetologist/endocrinologist/physician, senior podiatrist and credentialled diabetes educator working together at least one session per week [12, 13]. Although many services are NADC accredited as a Centre of Excellence or Core Centre, service accreditation by the NADC is not a prerequisite for participation in the ADFR. Onboarding of sites was stepwise and initiated through a formal expression of interest from each service.

2.5. Procedures and Data Management

Following review of site suitability by the Coordinating Principal Investigator, the nominated Site Principal Investigator is supported by ADFR staff in the local establishment of the project, including preparation of the governance application, database training prior to commencement, and data quality and audit procedures. Participating sites have the option of hosting a local project instance on their institutional REDCap server or directly contributing to the centrally hosted St Vincent's Hospital Sydney REDCap server (see Figure 2) [26, 27]. Permitting local instances of the project was favoured to maximise site participation, particularly among larger institutions, albeit with greater data management complexity. Local REDCap servers typically permit use of identifiers, simplifying longitudinal data entry, and additional custom fields of local interest can be accommodated. The central REDCap server does not permit identifiers and data access is site‐specific by designation of Data Access Groups. Where care is transferred to an external healthcare provider prior to care episode completion, a survey can be issued to that provider to enter outcome data without requiring a REDCap login.

FIGURE 2.

Data flow diagram for local and central REDCap server options. *Database features ticked where available and crossed where absent; access for backend support of local databases is server/institution dependant.

Data extraction and transfer from local REDCap servers to the central server occur annually. The data extract is pre‐set to remove identifiers, custom fields and any ineligible participants. Sensitive data files are either directly uploaded to the central REDCap server or transferred to the Sponsor via LiquidFiles with end‐to‐end encryption. Data files are stored on OneDrive. The Sponsor has assessed OneDrive and REDCap as providing adequate storage, backup, recovery, retention and security features for sensitive data, and data centres are in Australia.

2.6. Data Quality and Reporting

Several mechanisms are in place to optimise data quality, including use of in‐built REDCap field validation and the Data Quality application, which allows rapid identification of missing or invalid data, outliers and other uncommon errors. Each Site Principal Investigator is responsible for developing an internal procedure for routinely reviewing and optimising data quality. Additionally, data quality in‐services are delivered to participating sites early in recruitment to pre‐empt recurring issues, and then as part of the national Data Quality and Submission Meeting series held annually. These sessions include training on building custom REDCap Data Reports to identify additional data discrepancies and using the ADFR Audit Tool, through export to Microsoft Excel, to detect missing data and invalid dates.

Annual reports are centrally prepared for participating sites. These reports present local site data alongside aggregated national data maintaining anonymity of other sites. Benchmarking figures include trend indicators, target indicators, national averages and accreditation status, the latter to allow comparison to similarly resourced services. Key metrics, such as wait times and healing outcomes, are stratified by baseline ulceration severity using SINBAD scoring [28]. A quality improvement plan template and resources accompany the annual report, encouraging internal data‐driven service evaluation. By design, the annual reports and on‐demand ADFR Audit Tool output directly align with NADC iHRFS accreditation processes [12, 13], and enable quality improvement projects to occur between annual reporting cycles.

2.7. Outcomes of Interest

The implementation outcomes of interest for this paper were site onboarding and participant recruitment metrics, case ascertainment, mandatory‐field missing data, reporting cycles, related conference presentations and inaugural report feedback. Site onboarding was measured by the number of new sites commencing data collection, as well as the interval from ethics submission for new site addition to site commencement. Participant recruitment was measured by the numbers of individuals and their associated care episodes and discrete complications (i.e., foot ulcerations and/or active Charcot neuro‐osteoarthropathy).

Case ascertainment was calculated at the time of data submission as an estimate of the percentage of all people with diabetes and active foot complications attending a participating iHRFS who were captured in the database and hence in the ADFR. The target for case ascertainment was 80% or greater. To obtain these figures, sites used either manual audit of a representative (minimum 3‐month) period or an administrative data extract over the 12‐month audit period. While discouraged, consensus approximation by two senior staff was permitted in extenuating circumstances.

Mandatory‐field missing data was reported as proportions, noting that invalid date field data was also considered missing. A presentation log was maintained to capture invited talks and abstracts utilising ADFR data at major meetings and conferences. Site feedback was obtained following the inaugural report release, with all 19 participating services then completing anonymous surveys. Strength of agreement was recorded by respondents against four statements covering the structure and content of the report, quality improvement application and value of participation.

2.8. Statistical Analysis

Categorical data are presented as numbers with percentages, and continuous data as mean with standard deviation if normally distributed or median with interquartile range if not normally distributed. Data are reported by financial year, from 2020–2021 (first participant entered 24 November 2020) to latest completed period, 2024–2025. Case ascertainment was measured and is therefore reported from 2022–2023. Trends in missing data rates across years were assessed using chi‐square tests for linear trend, based on yearly counts of missing and non‐missing eligible field instances. All p values were two‐sided, with p < 0.05 considered statistically significant. Statistical analyses were performed using IBM SPSS Statistics for Mac, version 28.0.1 (IBM Corp., Armonk, NY, USA). Figure 3 was created using GraphPad Prism for Mac, version 10.6.1 (GraphPad Software, Boston, Massachusetts, USA).

FIGURE 3.

Cumulative number of participants and associated data by year. Data are n; *first participant entered 24 November 2020; active Charcot feet not shown due to comparatively small numbers.

3. Results

3.1. Site Recruitment

Since the establishment of the ADFR in July 2020, 74 sites have submitted expressions of interest to participate. Of these, 49 sites (operating under 37 iHRFS) have completed site implementation and commenced data collection, including 6 sites in 2020–2021, 17 in 2021–2022, 19 in 2022–2023, 1 in 2023–2024, and 6 in 2024–2025. Two sites permanently ceased data collection in 2025 due to the departure of key iHRFS staff, and a further five sites have suspended data collection due to staffing or technical challenges. Thus, 42 sites were actively collecting data at the end of June 2025. A map of participating sites can be accessed via the NADC website (https://nadc.net.au/hrfs‐data‐collection/) [23].

For the 44 (90%) sites participating in the NMA scheme, the median time from submission of an ethics amendment for site addition to commencement of data collection was 7 (IQR 4–12) months. This comprised 5 (3–11) days for ethics amendment approval, 71 (37–163) days for governance application preparation, 39 (18–68) days for governance review and approval, and 38 (16–87) days to data collection commencement.

3.2. Participant Recruitment

From ADFR establishment to end June 2025, 10 919 unique participants have been registered in the ADFR, with 13 150 care episodes, 22 383 ulcerations and 355 presentations of active Charcot foot recorded (see Figure 3). Participant numbers at individual sites ranged from 20 to 776.

Case ascertainment was typically estimated by manual audit or administrative data, with a minority of services using consensus approximation (see Table 1). Considering the total number of eligible people attending participating sites during each year, mean case ascertainment across services improved from 67% (SD 28) in 2022–2023 to 77% (SD 20) in 2023–2024 and 80% (SD 24) in 2024–2025. The target of 80% eligible participant inclusion was achieved by 20 (63%) services in 2024–2025.

TABLE 1.

Case ascertainment estimation methods and outcomes by year.

	2022–2023		2023–2024		2024–2025
Services submitting data, n	30		30		32
Estimation method
Manual audit	18	(60)	19	(63)	21	(66)
Administrative data	8	(27)	10	(33)	10	(31)
Consensus approximation	4	(13)	1	(3)	1	(3)
Percentage case ascertainment, mean (SD)	67	(28)	77	(20)	80	(24)
Services achieving ≥ 80% case ascertainment	14	(47)	14	(47)	20	(63)

Note: Data are n (%) unless otherwise specified. Services may operate across multiple sites.

3.3. Missing Data

Mandatory fields with the greatest rates of missing data were date of first interdisciplinary assessment, WIfI ischaemia grade [29], diabetes duration and cumulative length of stay (see Table 2). Missing data rates for all other mandatory fields ranged from 0% to 3%. Missingness for date of first interdisciplinary assessment and WIfI ischaemia grade decreased significantly across the years (13.5% to 0.5% and 21.6% to 5.3%, respectively; both p for trend < 0.001), whereas missingness for diabetes duration increased significantly (0.7% to 5.9%; p for trend < 0.001).

TABLE 2.

Mandatory fields with the greatest missing data rates by year.

Field name	2020–2021		2021–2022		2022–2023		2023–2024		2024–2025		p for trend b
Date of first interdisciplinary assessment	36/266	(13.5)	205/1447	(14.2)	442/3119	(14.2)	349/3349	(10.4)	16/3021	(0.5)	< 0.001
WIfI ischaemia grade a	114/528	(21.6)	545/3084	(17.7)	966/6270	(15.4)	868/6751	(12.9)	303/5750	(5.3)	< 0.001
Diabetes duration	2/282	(0.7)	31/1595	(1.9)	95/3538	(2.7)	192/3943	(4.9)	224/3792	(5.9)	< 0.001
Cumulative length of stay	2/81	(2.5)	9/480	(1.9)	5/904	(0.6)	9/948	(0.9)	55/834	(6.6)	NR

Note: Data are n missing/n eligible (%). Field instances hidden by branching logic were excluded from the denominator. NR, not reported, because 2024–2025 cumulative length of stay data may be incomplete due to ongoing care episodes at the time of extraction.

^a WIfI grading became mandatory from July 2024.

^b p‐value from chi‐square tests for linear trend across years.

3.4. Reporting

To date, participating sites have received individualised site‐level reports for 2020–2022, 2022–2023, 2023–2024 and 2024–2025. Publicly available national annual reports have also been released for 2023–2024 and 2024–2025 [30, 31]. Twenty‐nine presentations have been delivered at national and international scientific meetings that were either dedicated to the ADFR or utilised registry data. Fourteen of these abstracts were submitted by participating sites independent of the ADFR.

Feedback from all 19 participating services following the inaugural site‐level reports is summarised in Table 3. Most services agreed or strongly agreed that the reports were intuitive and relevant, all services agreed or strongly agreed that the reports were useful for supporting quality improvement activities, and all services strongly agreed that there was value in ongoing participation. Selected verbatim feedback illustrated key perceived strengths and implementation considerations:

I love that […] we can audit our service and ensure that we are providing a good service and one that can be constantly improved.

Junior staff […] see the importance of the data collection itself for service provision and quality improvement initiatives within the department.

[Site Investigator] has really appreciated being able to contact you directly and also the meetings we have had to get this going.

It's vital that staff have protected time to enter data. I do believe that sharing the data entry between team members is the way to go.

TABLE 3.

Categorical feedback following the inaugural site‐level reports.

Statement	Response
	Strongly disagree	Disagree	Neutral	Agree	Strongly agree
‘Structure—The structure of the report was intuitive and easy to navigate.’	—	—	2 (11)	7 (37)	10 (53)
‘Content—Data contained in the report was of interest and relevant to our Service.’	—	—	1 (5)	8 (42)	10 (53)
‘Quality Improvement—Our Service has used or plans to use the Report to aid Quality Improvement activities.’	—	—	—	8 (42)	11 (58)
‘Participation—Our Service sees value in ongoing data collection and contribution to this project.’	—	—	—	—	19 (100)

Note: Data are n (%).

4. Discussion

This paper reports the development, implementation and methodological outcomes of the ADFR, a national prospective longitudinal registry for iHRFS in Australia. We established a standardised dataset and practical data‐capture workflow, supported by a multi‐jurisdiction ethics and governance framework to facilitate participation across diverse iHRFS. Over the first 5 years, 49 sites operating under 37 iHRFS across all states and territories commenced data collection and collectively recruited 10 919 participants. In the latest year, mean case ascertainment reached 80%, while missing data rates were below 7% for all mandatory fields. Implementation also translated into reporting and engagement outputs, including four site‐level reporting cycles, two publicly available national annual reports, 29 conference presentations, and strong site agreement on the value of ongoing participation. These implementation outcomes demonstrate that the ADFR is feasible and scalable, and provide a foundation for interpreting patterns of care and outcomes and benchmarking performance nationally.

Outcome monitoring is an essential component of delivering quality and value‐based healthcare [32]. International diabetes‐related foot disease guidelines emphasise the importance of service auditing to ensure alignment with standards of care [22]. The National Diabetes Foot Care Audit in England and Wales is a leading example of how national audit can be used to identify variation in outcomes between services and inform targeted quality improvement [33]. Nonetheless, there remain few examples of dedicated national registries or audits for specialised interdisciplinary diabetes foot services

Data siloing remains a major barrier to standardised national healthcare data collection, with specialist datasets in Australia and internationally often relying on professional networks rather than integrated health record platforms, and typically capturing only a portion of the at‐risk population [34]. While imperfect, our experience suggests that waiting for widespread digitalisation of healthcare and harmonised data solutions often results in prolonged inaction. Although electronic medical records represent an important advance, they also introduce complexities, including dependence on information technology support, setup and maintenance costs, variability in systems and coding practices, barriers to updating data fields and accessing timely reports, and data quality issues. Provided these challenges can be overcome, it is of interest to align at least a subset of ADFR fields with electronic medical record data structures to enable semi‐automated data extraction.

In this context, initial enablers of the ADFR were primarily its accessibility and flexibility. Voluntary participation has been encouraged by freely available resources, infrastructure support, option of additional local data fields, responsiveness to feedback, timely assistance, and informative on‐demand outputs. Alignment with the NADC Collaborative iHRFS Standards also assists participating services in achieving NADC iHRFS accreditation [12, 13]. Importantly, the ADFR has fostered an investigative community of like‐minded and motivated clinicians. Collegial support and sharing of ideas are facilitated through virtual meetings and a web‐based project management system. For many investigators and junior clinicians involved in data collection, the project has served as an introduction to research, ethics and governance processes, and exemplifies the value of regular audit.

Growth of registry data has been steady, as is expected with incremental onboarding of new sites. Services entering their third or fourth annual reporting cycle generally achieve more complete capture of the eligible cohort, and as follow‐up accrues there is greater yield from outcome and longitudinal data. Data fields with higher rates of missing or erroneous data have informed optimisation of procedures. For example, completion of date of first interdisciplinary assessment improved after introduction of real‐time validation to ensure chronological consistency between date fields, and relocation of the field to better align with clinical workflow. To reduce missingness in WIfI ischaemia grading, the use of the WIfI classification alongside SINBAD scoring was mandated. In addition, early site‐specific data quality and audit training helped avoid recurrent issues.

Utilising an ethics framework offered several advantages, particularly the efficient onboarding of sites through the NMA scheme and the approval of an opt‐out or waiver of consent. This consent approach was taken to maximise participation, ensure equitable representation, and enhance potential benefits to the target population. Additionally, it minimises the burden on individuals already engaged in an intensive healthcare setting and reduces the administrative demand on clinicians. While centralisation of ethics processes decreased total workload, maintaining project oversight and timely support of governance processes has been challenging across a growing number of sites. Nonetheless, such support remains a critical component of the ADFR's success, as site investigators are often unfamiliar with governance systems and have competing clinical responsibilities.

The Australian Foot Health and Disease in Diabetes Strategy 2030 recognises progress made in addressing these challenges and advocates for commensurate funding of the ADFR so that its full potential can be realised [2]. Achievement of broader strategic goals, such as maintaining alignment with national and international evidence‐based DFD guidelines and best‐practice standards, and building collaborative research capability, relies on robust data collection. While clinical quality registries are recognised as fundamental in improving healthcare [32], their establishment in Australia has largely relied on the shared vision and commitment of clinicians. In this context, the solutions for developing the ADFR have often been frugal and pragmatic, while maintaining adaptability and future‐readiness.

The greatest limitation of the ADFR is arguably the additional time required to collect data, although it should be noted that the alternative of periodic manual audit remains a far more time‐intensive solution. While there are a relatively large number of fields, several are non‐mandatory or conditional and not always required, and incomplete data does not preclude site contribution. Continued evaluation of the value of each field and rationalisation of the dataset are anticipated as the registry matures. Further, opportunities to integrate data collection with routine clinical workflow should continue to be explored. Another limitation is the ethics requirement to report site data anonymously, which reduces the transparency of publicly available reports. The voluntary nature of participation introduces the risk of ascertainment bias. Addressing these challenges will likely require government funding and data contribution mandate alongside a nationally unified effort to digitalise healthcare, as has been successfully implemented in the United Kingdom [21]. In Australia a precedent for this model has been established by the Queensland Government Department of Health, which financially incentivised use of the Queensland High Risk Foot Form Database across Queensland Health facilities. As a result, approximately 50% of the estimated population with diabetes‐related foot ulceration in Queensland was registered in that database [35, 36]. This existing system also likely accounts for the comparatively slower uptake of the ADFR within the state of Queensland.

Other limitations of the ADFR include that, although DFD is a broad term encompassing lower limb peripheral neuropathy and/or peripheral arterial disease, the ADFR is dedicated to active foot complications, specifically foot ulceration and active Charcot neuro‐osteoarthropathy, and is limited to those people being managed by iHRFS. Accordingly, this paper describes the development of a national registry in the clinical setting where hospitalisation and amputation risk are especially high. Reflecting this focus, registered participant numbers represent a minority of the approximately 51 000 Australians estimated to have active DFD each year [2], a gap only partly attributable to incomplete iHRFS onboarding. A separate approach will be required to capture the broader burden of DFD nationally, including more common presentations and preventive care delivered to people with DFD. This paper is also limited by its focus on the methodological development of the ADFR and a critical analysis of baseline characteristics and clinical outcomes will be reported separately.

Prioritised for future development of the ADFR is the inclusion of risk‐adjusted benchmarking in annual reports. This methodology aims to distinguish the confounding effect of case‐mix on service outcome variation from genuine opportunities for service improvement, as employed by the National Diabetes Foot Care Audit in England and Wales [33]. Lived experience should be brought to the forefront of care by incorporating patient reported outcome measures (PROMs) [2, 37], although work is needed to validate or develop PROMs appropriate for this setting [38]. As with any addition to the dataset, the introduction of PROMs would need to be balanced against data collection burden and potential impact on case ascertainment, and need not be universally mandated. There is broader interest in expanding data collection beyond mainstream iHRFS to ensure adequate representation of priority populations [2], such as Aboriginal and Torres Strait Islander Peoples, people who live in geographically remote areas without access to iHRFS, and people experiencing homelessness. This would require significant funding as well as stakeholder and community engagement. Finally, in the age of artificial intelligence, rapid advancement in automation of data collection from clinical records and in the generation of data insights from interactive dashboards appears inevitable.

In conclusion, the ADFR has achieved its primary objective of implementing a standardised national repository for specialised interdisciplinary diabetes foot services in Australia. Significant progress has been made in enabling service audit and outcome evaluation. These efforts drive care excellence and quality improvement. Clinicians and stakeholders have shown strong support, with nationwide collaboration underpinning the project's success. The registry is now well positioned to meaningfully improve interdisciplinary care and accelerate research, supporting efforts to reduce the substantial DFD burden in Australia and potentially informing similar initiatives elsewhere.

Funding

This work was supported by the Australian Diabetes Society and the National Association of Diabetes Centres.

Ethics Statement

St Vincent's Hospital Sydney Human Research Ethics Committee (HREC) (National Mutual Acceptance approval number 2020/ETH01087), Northern Territory Department of Health and Menzies School of Health Research HREC (approval number 2021‐4131), Central Australian HREC (approval number CA‐21‐4051), and St John of God Health HREC (approval number 2160).

Consent

Opt‐out consent was approved in the Australian Capital Territory, New South Wales, Queensland, South Australia, Tasmania and Victoria; a waiver of consent was approved in the Northern Territory and Western Australia.

Conflicts of Interest

J.W.J.L. is employed by the National Association of Diabetes Centres and St Vincent's Hospital Sydney. S.A. is employed by the Australian Diabetes Society. N.W. is employed by the National Association of Diabetes Centres. These relationships are declared in the interests of transparency. The authors otherwise declare no competing interests.

Data Availability Statement

The data that support the findings of this study are available on request from the corresponding author. The data are not publicly available due to privacy or ethical restrictions.