Instructions for Use: Interpreting your results

Summary and suggested actions

We hope that you find this report useful and that it serves a starting point, to summarise the data from your site.

Your ADCQR site report is intended as a summary and may be helpful in actioning the results for your centre. We suggest the following actions:

Enter your site-specific data in the accompanying PowerPoint slide deck. Depending on the process (e.g., measurement rates) and clinical outcomes (e.g., average patient results) for your site, you may not need all slides. Please edit out any that you don't need.

Present these results to your team, to help disseminate the process and clinical information about your site. This is a great opportunity to share the knowledge that you have gained from taking part in ADCQR!

Determine if there are any areas for improvement.

Review the following Quality Improvement (QI) guide, to help determine how you might develop a QI initiative to help improve practice. You can view the before/after results in your next ADCQR report and see how you compare to centres of a similar size/kind.

It may be helpful to review the QI resources.

QI guide,

Primary purpose: To implement measurable improvement in a specific aspect of healthcare delivery.

Employ an iterative process of testing change ideas: Involves a continuous process of planning and testing changes, comparing the results to the predicted outcome, and adapting the process for further improvement.

Consistent use of agreed methodology: There are many different methodologies for QI. These include Model for Improvement (Plan-Do-Study-Act), Lean, Six Sigma. Research shows that the type of methodology doesn't matter as much as making sure you complete each part rigorously and then review your results to determine where to action change next.

Empowerment of front-line staff and service users: QI work should encourage both staff and consumers to contribute to improvement. This means that there needs to be support of frontline staff and service users so that they feel able to participate and take ownership of QI initiatives.

Use data to drive improvement: ADCQR data provides a unique ability to regularly measure key clinical processes and outcomes. You can use this to establish a baseline and to measure how effective QI programs have been at subsequent data collections

Scale-up and spread: best practice is to take successful QI interventions, adapt them to other issues with room for improvement. We advise sharing with colleagues from other centres. Tips about what worked (and didn't) can help everyone improve!

Source: adapted from Backhouse and Ogunlayi, Quality improvement into practice, BMJ, 2020

Using ADCQR with QI

Source: adapted from Backhouse and Ogunlayi, Quality improvement into practice, BMJ, 2020

QI program steps

Step 1: What is your overall goal for your centre?
You will need a clear goal to aim for

Is your goal realistic or achievable?

Step 2: Assess your ADCQR data
Key considerations:

Where do you sit compared to other centres?

What areas show room for improvement?

Why do you think your data shows that there is room for improvement?

Do you have ideas for actions to help improve these areas?

Step 3: Deciding on a QI project
Decide your criteria for QI projects. These could be projects designed to make improvements that:

will be of most benefit to patients

will have the biggest impact

are aligned with organisational strategic goals

are most likely to succeed when the barriers are considered

Step 4: Plan your project
Your project will need a clear plan, so that you can work out who needs to do what to facilitate improvement. This may involve a series of small steps to work towards a larger goal.

Are you measuring a process outcome (e.g. the proportion of people tested for HbA1c), or a clinical measure (e.g. HbA1c level)?

How will you measure this?

Who needs to do what action to help enable the goal?

What are your plans to review, test and refine your QI program?

Common problems with QI development and/or implementation

Having an unachievable goal.
For example, you might want all patients with suboptimal BMI to lose weight, but this outcome is difficult to achieve. It is likely to need multiple iterations of QI programs targeting nutritional advice, exercise programs and psychological support, as well as highly motivated consumers. A smaller goal, such as increasing consumer knowledge about nutrition may be a more appropriate starting point.

Not being clear about the path to change
You must be conceptually clear about what needs to change and who needs to do what to enable this change. You must be clear about how you will measure change and what you consider improvement.

Not having a local change champion
In busy environments, it can be difficult to leverage change without local change champions. Whatever your goal is, there will be some degree of behavioural change involved for staff. Your project is more likely to succeed if there is someone willing to be a local champion who can guide the process.

Not reviewing your results and plan
Research shows that some healthcare initiatives using the Plan-Do-Study-Act cycle aren't effective and that many don't complete all sections of the cycle. Key components missed include having a clear plan to start with, reviewing the results achieved and how they compare to predicted results, summarising the learning for clinical staff, and using the results to plan changes to implement in the next iteration of the project. QI is not a one-off project, it's a method to make continual improvements and needs to be reviewed and adapted to become more successful.

QI resources

NADC QI resources
Included in your full appendix report

Agency for Healthcare Research and Quality (AHRQ)

Plan-Do-Study-Act (PDSA) directions and examples
https://www.ahrq.gov/health-literacy/improve/precautions/tool2b.html

BMJ Open Quality Quality Improvement Project Guide (Word document download)
https://bmjopenquality.bmj.com/pages/wp-content/uploads/sites/13/2017/07/Qualityimprovementprojectguide.docx

References

1. Backhouse, A., & Ogunlayi, F. (2020). Quality improvement into practice. BMJ, 368.
2. Cheung, NW, JJ Conn, MC d’Emden, JE Gunton, AJ Jenkins, GP Ross, AK Sinha, S Andrikopoulos, S Codplyr::lagiuri, and SM Twigg (2009). Position statement of the Australian Diabetes Society: individualisation of glycated haemoglobin targets for adults with diabetes mellitus. Medical Journal of Australia 191(6), 339–344.
3. Deed, G, E Ackermann, I Arthur, J Barlow, S Jagadeesan, D Kawol, G Kilov, S Leow, JA Manski-Nankervis, R Rasalam, et al. (2016). General Practice Management of Type 2 Diabetes: 2016–18.
4. Lalor, E, A Boyden, D Cadilhac, S Codplyr::lagiur, J Doust, D Fraser, M Harris, N Huang, D Johnson, G Johnson, et al. (2012). Guidelines for the management of absolute cardiovascular disease risk.
5. Living Evidence for Diabetes Consortium. (2020). Living Evidence Guidelines in Diabetes. available from: https://diabetessociety.com.au/living-guidelines.asphttps://app.magicapp.org/#/guideline/E5AbPE
6. Royal Australian College of General Practitioners. (2019). Supporting smoking cessation: A guide for health professionals (Second Edition). Available from: https://www.racgp.org.au/clinical-resources/clinical-guidelines/key-racgp-guidelines/view-all-racgp-guidelines/supporting-smoking-cessation

Registry Methodology

Ethics approval

The ADCQR was established to improve the quality of care and outcomes for people living with diabetes and is considered to be in the public’s interest.

To function as a CQR, the ADCQR collects, stores, and uses identifiable, personal and sensitive health information about people with diabetes.

The opt-out approach to participation is used to recruit participants due to the scale and significance of the Registry. In accordance with the State and Federal privacy legislation of Australia, the Australian National Statement for Ethical Conduct in Research,7 the ADCQR has ethics approval under the National Mutual Acceptance (NMA) scheme from the Monash Health Human Research Ethics Committee. Additional ethics approvals are sought from participating sites that do not operate under the NMA.

As part of local research governance processes, it is a requirement for all sites registered to obtain ethics approval and local research governance authorisation prior to commencing data collection.

Governance

The Registry custodian is the School of Public Health and Preventive Medicine (SPHPM), Monash University, and the ADCQR is operated by the Project Executive at SPHPM, under the leadership of Professor Zoungas.

The ADCQR Scientific Advisory Committee provides strategic guidance to the Project Executive to ensure the objectives, outcomes and deliverables of the ADCQR, as specified by the Australian Government Department of Health and Aged Care are achieved (Figure 2). This committee consists of representatives of key stakeholder organisations including endocrinologists, general practitioners, consumer representatives, Aboriginal and Torres Strait Islander representatives, and representatives from national peak bodies, and is working to the agreed Terms of Reference with the ultimate vision of assisting the ADCQR to maintain high visibility, appropriate engagement and relevance for diabetes service delivery. The ADCQR Project Executive and Scientific Advisory Committee members are listed at the end of this report.

Governance Structure

Key Registry Milestones

Expressions of interest and local research and governance processes are ongoing. To reduce the burden on participating health services, the ADCQR collects data during the months of May to June (n.b. participating health services self-select a continuous four-week period during this sampling period to collect data on consecutive patients with diabetes who attend the service and meet the inclusion criteria). The data collection period may be extended until the censorship date of 31 August of each year, at which time the Registry freezes the data for data cleaning, analysis and reporting.

ADCQR Key Milestones

Recruitment

Site recruitment

Sites are recruited through the National Association of Diabetes Centres (NADC), a sub-division of the Australian Diabetes Society (ADS). The NADC, established in 1994, is a national collective of centres that are involved either directly or indirectly in diabetes services and care, brought together by a common desire to see improvement in the standard of diabetes care in Australia.8 The NADC takes a leadership role in developing, fostering and supporting networks between diabetes services (from primary care to tertiary (hospital) care settings), recognising that diabetes care requires a shared, multidisciplinary approach. The NADC facilitates and promotes improved standards of diabetes care by implementing evidence-based policies and procedures, including developing national standards and auditing and benchmarking activities.

There are six membership/accreditation levels of the NADC. The ADCQR currently recruits sites from primary, secondary, tertiary and centre of excellence services:

1. Centres of Excellence
   Diabetes centres that have demonstrated excellence in education, research, service delivery, practice/policy development and national influence. These centres must be tertiary-level facilities.
2. Tertiary Care Diabetes Services
   These centres are hospitals with a full range of clinical diabetes service providers including endocrinologists, credentialed diabetes educators, dietitians and podiatrists on staff (full-time) and have demonstrated a high standard of care through service delivery and organisational capacity.
3. Secondary Care Diabetes Services
   These centres are typically community services with a range of full and/or part-time diabetes staff but often do not have an endocrinologist as part of their usual team.
4. Primary Care Diabetes Services
   These centres have part-time staff and work closely with local general practitioners to provide care for people with diabetes.
5. Pharmacy Diabetes Services
   These centres have staff that have received training and/or have expertise in diabetes and work closely with the local general practitioners (GPs) and allied health staff to provide additional care and services in the pharmacy context.
6. Network Members
   The NADC Network membership is offered to Primary Health Networks (PHNs) and State and Territory Government organisations who work directly with GPs, and other health care providers to facilitate improved outcomes for patients.

In 2023, there were 204 NADC member diabetes centres across Australia; these operate in a range of locations and facilities from major metropolitan adult and children’s hospitals to community-based services including general practices and pharmacies.

Participant recruitment

Participants are recruited through the participating site (health service) based on the following criteria:
Inclusion criteria
Attend a participating centre
Age ≥ 18 years
Patients with T1DM, T2DM or other (secondary) forms of diabetes
Have the capacity to make the decision to opt-out or be included in the Registry

Exclusion criteria
Age < 18 years
Female patients with a diagnosis of gestational diabetes mellitus (not known to have established diabetes)

As participants are recruited using the opt-out approach, prospective eligible participants are provided with the Registry participant information sheet by a staff member at their health service to inform them that their information will be shared with the Registry, and how to opt out if they change their mind and do not want their information shared. Moreover, their data are entered into the Registry ‘holding database’ for a two-week period where the data is stored and not used for any purpose, prior to their consent being assumed and their data included in the Registry.

Data Collection

What information do we collect?

The ADCQR has leveraged from the formative work undertaken as part of ANDA.

In brief, the ANDA dataset used an enhanced version of the National Diabetes Outcomes Quality Review Initiative (NDOQRIN) dataset, aimed at improving diabetes care through a structured approach to patient management.9 Initially, this was based on the NDOQRIN minimum dataset linked to the NSW Clinical Management Guidelines for Diabetes,10 with subsequent updates/enhancements to the dataset over the years based on feedback from participating health services, as well as the latest research and evidence in diabetes care and quality improvement.

The ADCQR minimum dataset is based on ANDA 2022. This has considerable concordance with similar international datasets throughout the United States of America and Europe.11-14 Compared to international registries, the ADCQR provides comprehensive reporting on multidisciplinary care, diabetes complications and psychological factors, but lacks benchmarking of structural measures, smoking counselling, conception/pregnancy counselling and contraceptive counselling.12 Overall, the high rates of agreement with international practice, supports the validity of the ADCQR in the benchmarking of key quality indicators regarding diabetes care within Australia.

The ADCQR captures clinical indicators as well as patient self-management outcomes. The clinical component is collected by the clinician or staff member at the participating site and may be collected during the patients’ clinical consult or via medical records. It includes information on demographics, blood glucose control and management methods, other risk factors/biomarkers, medication use, as well as complications and comorbidities. The patient reported component is self-reported by the patient either directly or with a health professional at their diabetes health service prior to (in the waiting room) or during their clinical consult. Participating sites decide how best to deliver the patient reported component for their patient and/or health service. Completion of the questionnaire directly by patients was intended to reduce the burden of data collection on participating sites. The patient reported component is one-page and includes information on smoking status, vaccination status, health professional attendances, medication use, patient self-care practices including nutrition/diet management and physical activity.

The data collection forms captured most fields using yes/no responses or other choice options to reduce the amount of written data required.

The ADCQR provides participating health services with a data definitions document, including the ADS Algorithm15 to assist in the collection of data on treatments (Appendix 2). The ADCQR Public Facing Data Dictionary provides details on the variables collected and is available on the ADCQR website: https://www.monash.edu/medicine/sphpm/adcqr

How do we collect this information?

Participating sites have the option to choose from three methods of data collection outlined below.

Web-based data collection – Research electronic data capture (REDCap)
The web-based electronic data capture application, REDCap16 has been previously used in ANDA since 2019, providing familiarity to many of the health services participating in the ADCQR in 2023. Study data were collected and managed using REDCap electronic data capture tools hosted and managed by Helix (Monash University). REDCap is a secure, web-based application designed to support data capture for research studies,17,18 providing:

1. an intuitive interface for validated data entry
2. audit trails for tracking data manipulation and export procedures
3. automated export procedures for seamless data downloads to common statistical packages
4. procedures for importing data from external sources

Branching logic coding was used to skip irrelevant questions. Data validations were put in place to help prevent data entry errors and reduce data queries. Staff were granted access to patients from their sites only.

Paper-based data collection
The Teleform© sotware was utilised for the design of paper data collection forms. Once completed by sites and sent to the ADCQR coordinating centre, the forms were entered directly into REDCap. Any printed data collection forms are stored in a locked room at SPHPM, Monash University.

Data Extraction
No sites collected data via this method in 2023.

How do we store this information?

All Registry data are kept electronically in accordance with Monash University’s Information Technology Services Security Framework policy. Patient identifiable data are stored in a highly secure database that is separate to the clinical and patient reported data, to reduce the risk of data breaches. Patient clinical and patient reported data are stored within the Registry database. The data are entered into the Registry ‘holding database’ for a two-week period, prior to consent being assumed and data included in the Registry. The date of the patient consult (clinic visit) indicates the commencement of the opt-out window.

The ADCQR has established a Risk Register to continually assess and manage risk, including any risks associated with data collection and storing. This is a standing item on the agenda of all Project Executive and Scientific Advisory Committee meetings.

Data Verification And Validation

Near complete data capture is required to ensure the Registry’s reporting is accurate. Data validations and quality checks were performed for each site at the end of their data collection period. Reports were generated for each site, querying missing data, potential duplicate records and invalid or out-of-range values. Sites were encouraged to address data queries prior to resubmission to the Registry. Where duplicate records were identified (multiple case record entries for the same patient), only the first entry was retained. Data assumptions and manipulations were made according to a pre-defined list of criteria (see the Supplement to this Report). Corrected data items were updated in the database prior to final analysis.

Statistical Analyses

Descriptive statistics
Results are presented descriptively as percentages for categorical variables, and mean and standard deviation (SD) for continuous variables. Variables that were not normally distributed are presented as median and interquartile range (IQR, where IQR is represented by the first quartile (Q1 or 25th percentile) and third quartile (Q3 or 75th percentile)). Percentages are calculated from total respondents (and did not include missing data in the denominator). Percentages may not always add to 100% due to rounding. Where N is reported, it refers to the number of patients with available data (denominator). The number of people that answered 'Yes' (numerator or n) can be estimated by multiplying the reported percentages by the number of patients with available data (denominator).

Complications/comorbidities are reported as ever reported (percentage of patients with a diagnosis/detection either in the last 12 months or prior to the last 12 months) and reported in the last 12 months only (percentage of patients with a diagnosis/detection in the last 12 months.

Urinary albumin and urinary protein levels were used to determine albuminuria. Albuminuria was determined using the Kidney Disease: Improving Global Outcomes (KDIGO) guidelines.19 Where albumin measurement was missing, and proteinuria measurement was available, patients were categorised into albuminuria using the relevant thresholds outlined by KDIGO. Albuminuria was defined as:

Normal to mildly increased: AER <30 mg/24 hours, ACR <3 mg/mmol, PER <150 mg/24 hours, or PCR <15 mg/mmol

Moderately increased: AER 30-300 mg/24 hours, ACR 3-30 mg/mmol, PER 150-500 mg/24 hours, or PCR 15-50 mg/mmol

Severely increased: AER >300 mg/24 hours, ACR >30 mg/mmol, PER >500 mg/24 hours, or PCR >50 mg/mmol

In Australia, the units of measurement that urinary albumin and urinary protein levels are most commonly reported in terms of mg/L (AER and PER) and ratio (ACR and PCR using mg/mmol). To categorise patients into albuminuria using mg/L units of measurement, the following thresholds were employed:

Normal to mildly increased: AER <20 mg/L or PER <20 mg/L

Moderately increased: AER 20-200 mg/L or PER 20-200 mg/L

Severely increased: AER >200 mg/L or PER >200 mg/L

eGFR levels were used to determine chronic kidney disease (CKD). KDIGO guidelines define CKD as any abnormality of kidney structure or function that is present for >3 months, with implications for health.19

• Stage 1: eGFR ≥90 mL/min/1.73m² and evidence of kidney damage (albuminuria, urine sediment abnormalities, electrolyte and other abnormalities due to tubular disorders, structural abnormalities on histology or imaging and history of kidney transplantation)
• Stage 2: eGFR 60-89 mL/min/1.73m² and evidence of kidney damage
• Stage 3: eGFR 30-59 mL/min/1.73m²
• Stage 4: eGFR 15-29 mL/min/1.73m²
• Stage 5: eGFR <15 mL/min/1.73m²

End stage kidney disease was also captured through a separate question, where end stage kidney disease was defined as Stage 5 chronic kidney disease (eGFR <15 mL/min/1.73m²) and/or dialysis dependent (haemodialysis or peritoneal dialysis) and/or renal transplant recipient reported in the last 12 months.

Selected analyses are also reported for people with and without cardiovascular disease (CVD), where CVD includes myocardial infarction, CABG/angioplasty, stroke, congestive cardiac failure or peripheral vascular disease.

Subgroup analyses: Centre type
Given the different patient populations attending primary, secondary and tertiary health care settings, results by centre type are presented. For comparability, pooled patient data from CoEs & Tertiary care services were compared to pooled patient data from Secondary & Primary care services.

Additional analyses
The Supplement to this Report provides additional analyses including

Key findings reported in terms of patients meeting target

Frequency count data

Missing data

Descriptive reporting including bar charts to demonstrate the distribution of results across participating sites

Post data collection questionnaire results

Box and Whisker Plots
Box and whisker plots are used to show how data is distributed and any outliers. These have been included to show the distribution of HbA1c levels across patients at your site and compared to other similar sites. Current and historical data (if available) has been included for comparison.
The following is an example of how to interpret box and whisker plots:

The target level of HbA1c < 7% is shown by the horizontal green line.

Outliers are patients whose HbA1c is outside the upper and lower whisker.

The upper whisker is the upper boundary before patients are considered outliers. The upper whisker extends from the upper quartile to the largest value that is within 1.5 × the IQR from the upper quartile.

The lower whisker is the lower boundary before patients are considered outliers. The lower whisker extends from the lower quartile to the smallest value that is within 1.5 × the IQR from the lower quartile.
Risk Adjusted Funnel Plots
Funnel plots are graphical methods used to detect variation in clinical performance across participating sites by plotting the mean values or rates (proportion) of patients at each participating site for selected key clinical indicators. In this site report, four key clinical indicators are assessed: HbA1c, low-density lipoprotein cholesterol (LDL-C), systolic blood pressure (SBP) and rates of severe hypoglycaemia. Clinical performance was adjusted for statistically significant, non-care related patient factors prior to plotting such as age, sex, and duration of diabetes. This process allows comparability across sites, who may see very different patient populations. The factors that have been adjusted for are included at the bottom of each graph.
The following is an example of how to interpret the funnel plots:

The overall mean values or rates across all sites is shown by the horizontal red line.

The target level for each clinical indicator is shown by the horizonal green line.

‘Action’ control limits were set at 99.8% (3 standard deviations above the overall mean level) and ‘alert’ control limits set at 95.0% (2 standard deviations above the overall mean level) which are shown by the contour lines.

Sites positioned above the ‘action’ limit are considered outliers and should work towards implementing strategies to improve this outcome measure. Sites positioned above the ‘alert’ limit (but below the ‘action’ limit) may be at risk of outlier performance and should monitor performance.

Clinical performance includes patients who are attending the site for the first time (initial visit) and returning patients (non-initial visit).

Contact Information

Site reports are produced by the Australian Diabetes Clinical Quality Registry (ADCQR). Thank you to all diabetes centres and individuals who have generously participated in the ADCQR.
Suggested citation:

Australian Diabetes Clinical Quality Registry Site Report 2023. Monash University, School of Public Health and Preventive Medicine, December 2023.

Any enquiries should be directed to:

ADCQR Secretariat via email: adcqr@monash.edu

ADCQR via postal mail:

                    Australian Diabetes Clinical Quality Registry
                    Monash University
                    553 St Kilda Road
                    Melbourne
                    VIC 3004

Site Feedback Form

Thank you for using the ADCQR dashboard. Your feedback helps us continously improve the dashboard and reporting.
For each of the questions below, please select an answer. Submit