We aimed to investigate the impact of rationalizing general surgery surgical tray sets at an elective surgical hub in England. This was an analysis of data collected prospectively for a clinical quality improvement project between 25th March and 20th June 2024. Baseline data on use of surgical instruments for four general surgery procedures were collected, and the items categorized as high, medium and low use. Meetings were then held with surgical and operational teams involved in delivering each procedure and a set of surgical instruments for each procedure agreed upon and a rationalized tray was formed. Data were collected for 39 rectal procedures, 42 open hernia repairs, 15 laparoscopic hernia repairs and 14 laparoscopic cholecystectomies. A tailored rectal procedure tray was formed with 16 items from the original53 items small basic tray; the two trays used for open hernia repair were rationalized from 53 to 43 items (small basic tray) and 75 to 60 items (large basic tray). The same laparoscopic tray was used for hernia repairs and cholecystectomies and was rationalized from 55 to 45 items. Estimated base-case annual financial savings were £16,863 and carbon savings 42.6 kgCO2e for these four procedures. Saving of up to £62,648 and 3612.9 kgCO2e annually may be realized if rationalization avoids opening a second tray to obtain items not present for use on a single tray. Rationalization of surgical trays can reduce financial costs and carbon emissions. It may also yield co-benefits in terms of enhancing theater efficiency.

26-A-13875-ACC CLINICAL INVESTIGATION AND QUALITY IMPROVEMENT RESOURCE GUIDE FOR RESIDENTS: RATIONALE AND DESIGN BY THE TIDE-IM RESIDENT INVESTIGATORS

Clinical audit, quality improvement and data quality

WCN26-9183 USE OF A GOOGLE-BASED ELECTRONIC DATA COLLECTION SYSTEM FOR MANAGING HAEMODIALYSIS PATIENTS IN A LOW-RESOURCE SETTING

Cat allergy is a common cause of allergic rhinitis and asthma and is associated with significant symptom burden and impaired quality of life. Complete avoidance of cat allergens is often impractical, and allergen-specific immunotherapy (AIT) represents a potential disease-modifying treatment option. However, real-life data regarding the effectiveness and safety of cat allergen immunotherapy remain limited. To evaluate the real-life clinical effectiveness and safety of subcutaneous cat allergen immunotherapy using a cat allergen extract in patients with cat-induced allergic rhinitis and/or asthma. This retrospective observational study included patients who received subcutaneous cat allergen immunotherapy for at least 6 months between January 2023 and December 2025. Sensitization to cat allergen was confirmed by both skin prick testing and serum cat-specific IgE (≥0.35kU/L). Clinical outcomes, including symptom scores, medication scores, Visual Analog Scale (VAS), and Rhinoconjunctivitis Quality of Life Questionnaire (RQLQ) scores, were assessed at baseline and during follow-up. Changes between baseline and the sixth month of treatment were analyzed using nonparametric statistical methods. A total of 12 patients were included, all of whom were female, with a mean age of 35.17 ± 8.68 years. Asthma was present in 25% of patients. At the sixth month of immunotherapy, significant improvements were observed across all evaluated clinical parameters. Median symptom scores decreased from 3.0 to 1.0 (P = 0.002), medication scores from 2.0 to 1.5 (P = 0.038), and VAS scores from 9.0 to 5.0 (P = 0.002). RQLQ scores also showed a significant reduction from 113.0 to 86.0 (P = 0.028). Improvements were associated with large effect sizes. Ten patients achieved a >30% reduction in VAS scores. Moderate correlations were observed between changes in symptom and medication scores and improvements in quality of life. Immunotherapy was generally well tolerated; moderate-to-severe systemic reactions occurred in a subset of patients, with no life-threatening reactions or permanent treatment discontinuation. In this real-life cohort, subcutaneous cat allergen immunotherapy was associated with significant improvements in clinical symptoms, medication use, and quality of life, with an acceptable safety profile. These findings support the role of cat allergen immunotherapy as an effective treatment option in selected patients with persistent symptoms despite medical therapy.

Background The International Classification of Diseases, 10th Revision (ICD-10), is widely used for clinical care, quality assurance, and stroke research. Its ubiquity across healthcare systems makes it an attractive foundation for digital health tools that can support stroke surveillance and population health monitoring. However, a major limitation is that stroke detection algorithms derived from ICD codes have been developed primarily in socially homogenous populations, raising concerns about generalizability and fairness across racially diverse populations. Methods We developed and validated an acute ischemic stroke (AIS) detection algorithm using Classification and Regression Tree (CART) supervised machine learning, using a diverse derivation cohort. Input variables consisted of diagnostic and procedural ICD-10 codes, stratified by position and presence on admission. The model was trained on 75% and tested on 25% of the derivation cohort and externally validated in a second tertiary institution serving patients living in predominantly underrepresented and socially vulnerable communities. Performance of the algorithm was measured by sensitivity, specificity, positive predictive value (PPV), and Cohen's κ. Subgroup analyses were conducted by sex and race/ethnicity. Results In the derivation cohort, the CART model achieved sensitivity of 96%, specificity of 90%, PPV of 99%, and κ=0.78. Applied to the independent validation cohort, the algorithm identified 1,050 AIS cases and 1,664 non-AIS cases, with sensitivity 89%, specificity 95%, PPV of 92%, and κ=0.84. Performance was comparable between women and men (κ=0.80 for both), and strong across Black (κ=0.81), Hispanic (κ=0.76), and White (κ=0.80) subgroups. Lower accuracy was observed in the Asian subgroup (κ=0.73, PPV=62%). Discussion Our findings demonstrate that CART-based algorithms can provide accurate and interpretable AIS detection using ICD-10 data while explicitly addressing social fairness. The algorithm's reproducibility across independent and diverse populations highlights its potential as a low-friction, scalable, and cost-efficient tool for clinical care, surveillance, and quality improvement. Importantly, subgroup analyses underscore the necessity of ongoing fairness evaluation, as performance varied by race/ethnicity, particularly in the Asian subgroup. Limitations include potential missed cases in the gold standard, lack of confidence intervals due to retrospective data, and dependence on local coding practices. Conclusions This study shows that ICD-10-based machine learning algorithms, specifically CART, can serve as a model for developing an accurate and equitable digital health platform for AIS surveillance.

BACKGROUNDIntensive care unit (ICU) capacity is underdeveloped in sub-Saharan Africa, and outcomes for critical care lag behind higher resource settings. It is essential to understand local case-mix, processes of care and clinical outcomes in African ICUs to close these gaps. To this end, ICU registries are valuable tools for clinical research, quality improvement and capacity building. The Critical Care in Asia and Africa (CCAA) Network has specially developed a novel ICU registry for resource-limited settings.AIMTo evaluate the feasibility, acceptability, and perceived sustainability of the CCAA ICU registry in the Ethiopian clinical context.METHODSEight months following the pilot implementation of the CCAA ICU registry at two academic medical centers in Addis Ababa, Ethiopia, we conducted a qualitative evaluation. We conducted key informant interviews and focus group discussions with members of the ICU registry team, medical and nursing staff, and leadership team. We coded and analyzed transcripts deductively using a thematic content approach.RESULTSEmergent themes related to feasibility included data collection, data quality and factors necessary for success. Those related to acceptability included utility, accessibility and comparison to other methods. Those related to perceived sustainability included institutional future, ownership and expansion. Overall, respondents felt that the CCAA ICU registry was feasible and acceptable in their ICUs. They identified important threats to perceived sustainability including multiple channels of communication and infrastructure and human resource limitations, and proposed adaptations to address these threats.CONCLUSIONThe CCAA ICU registry is a promising tool for research and quality improvement in ICUs in sub-Saharan Africa, but successful implementation requires a clear understanding of regional and institutional influencing factors.

The Danish clinical quality registries contribute to clinical quality improvement by utilizing quality indicators and development targets. This study examines fulfillment of development targets across all registries for cancer and non-cancer diseases at national, regional, and hospital levels. Annual summarized data from Danish clinical quality registries (2019-2023) were accessed. The proportion of fulfilled development targets was presented overall and in strata according to registries covering cancer or non-cancer diseases, indicators describing processes or results, and established or supplemental indicators. Regional differences and changes over time were evaluated. The data evaluated 13,305,861 patient pathways and 697 indicators from 65 registries. Overall, 57% of hospital-level development targets were fulfilled (median 58%, IQR:44-76%). The 24 cancer registries fulfilled 63% of targets (median60%, IQR:49-76%), while the 41 non-cancer registries fulfilled 54% (median54%, IQR:38-75%). Fulfillment was higher for result indicators (68%) than process indicators (44%), and for established (56%) than supplemental indicators (44%). Regional fulfillment ranged from 50% (Region Zealand) to 64% (the Central Denmark Region). Target fulfillment increased by five percentagepoints during the study period. The Danish clinical quality registers varied in the approach to defining development targets, ranging from very conservative to very ambitious. The development targets were more ambitious for registries of non-cancer diseases than cancer diseases, more ambitious for process- than for result indicators, and more ambitious for supplemental compared to established indicators. Setting clinical quality improvement targets requires balancing ambition with realism. The ambitious target setting for supplemental (often new) indicators and process indicators is expected, as these have a high potential for further improvement. The data suggest that the Danish clinical quality registries in cancer diseases could benefit from raising some of the development targets.

The modified Bereiter trochleoplasty for severe trochlear dysplasia demonstrates significant clinical improvement in knee function and quality of life at mid-term follow-up.

ObjectiveTo estimate the prevalence of potential overtreatment of type 2 diabetes mellitus (T2DM) among older adults and to develop and compare predictive models to identify patient and physician characteristics associated with overtreatment.DesignPopulation-based retrospective cohort study with predictive modelling.SettingA province-wide, publicly funded healthcare system in British Columbia, Canada, using linked administrative health claims data from 2016 to 2023.ParticipantsResidents of long-term care facilities over age 65, and community-dwelling individuals over age 75, with a diagnosis of T2DM and a glycated haemoglobin (A1C) laboratory value ≤7.0%. Participants were required to have ≥365 days of continuous provincial health insurance coverage prior to their index A1C test. Patients receiving palliative care and those with missing physician information were excluded.Primary and secondary outcome measuresPotential overtreatment of T2DM, defined a priori as overlapping prescriptions for ≥2 glucose-lowering medications or ≥1 insulin or sulfonylurea dispensing within 90 days after the index A1C test.Model performance outcomes included discrimination (area under the curve (AUC), sensitivity, specificity, positive predictive value and negative predictive value). Performance metrics were calculated with 95% CIs using a 25% temporally distinct test dataset (2021–2023). No changes were made to outcome definitions after protocol development.ResultsAmong 133 773 patients with an A1C≤7.0%, 38 074 (28.5%) were classified as overtreated. These patients had a mean age of 79.6 years, were 47% female, and had a median A1C of 6.4%. The gradient boost model was the best performing model overall, using a combination of expert-selected variables and data-driven variables, achieving an AUC of 0.87, sensitivity of 0.81 and negative predictive value of 0.89. The top predictors of overtreatment included use of blood glucose test strips, A1C test volume, polypharmacy, specialist involvement and measures of diabetes severity.ConclusionsOvertreatment of T2DM was prevalent among older adults in our cohort. Machine learning algorithms that integrate clinical expertise with data-driven variable selection performed the best in predicting T2DM overtreatment. We identified several patient and physician characteristics as key contributors that may inform future clinical practice and quality improvement initiatives, although external validation is required before clinical implementation.

Comparison of Sirolimus- vs Paclitaxel-Coated Balloon Angioplasty for Femoropopliteal Artery Disease: The SIRONA Randomized Noninferiority Trial.

Background Standardised early mobilisation within Enhanced Recovery After Surgery (ERAS) for endometrial cancer is lacking. This study developed and evaluated a protocol. Methods This retrospective study analysed outcomes of a clinical quality improvement initiative conducted at a tertiary care centre (February 2021–May 2022). Phase 1: Protocol developed via evidence/expert consensus. Phase 2: Retrospective analysis of 90 patients who received care in two units during the study period. Patients admitted to unit A (n = 45) received a structured early mobilisation protocol implemented as part of routine clinical practice improvement, whilst patients admitted to unit B (n = 45) received standard care. Ward assignment was determined by bed availability at admission without researcher involvement in allocation. Primary outcomes: Pulmonary function, physical capacity (ambulation time/day 5 walking distance) and gastrointestinal recovery. Secondary: Complications, LOS, emotional status (Brief Profile of Mood States, BPOMS). Statistical analyses included effect sizes and 95% confidence intervals. Results The intervention group was associated with significantly better first ambulation (6.8 ± 2.1 h vs. 18.5 ± 4.3 h, mean difference: 11.7 h, 95% CI: 10.3–13.1, p < 0.001), day 5 walking distance (385 ± 68 m vs. 142 ± 45 m, mean difference: 243 m, 95% CI: 220–266, p < 0.001), time to first flatus (28.5 ± 6.2 h vs. 48.3 ± 8.5 h, mean difference: 19.8 h, 95% CI: 16.9–22.7, p < 0.001), fewer pulmonary complications (8.9% vs. 24.4%, OR: 0.30, 95% CI: 0.09–0.96, p = 0.047), shorter LOS (4.2 ± 1.1 d vs. 6.8 ± 1.5 d, mean difference: 2.6 d, 95% CI: 2.0–3.2, p < 0.001) and improved BPOMS anxiety/depression scores. Conclusions This retrospective analysis suggests that the structured early mobilisation protocol, implemented as a clinical quality improvement initiative, was associated with improved recovery, reduced complications/LOS and better emotional wellbeing in endometrial cancer patients within ERAS.

Inappropriate coagulation testing contributes to inefficiency, cost, and environmental harm. Baseline audits on our acute medical unit (AMU) showed that one-third of coagulation screen requests lacked a clear clinical indication. To reduce unnecessary coagulation screens on acute medical services [AMU and Medical Same Day Emergency Care (mSDEC)] by 90% within 12months, aligned with National Health Service (NHS) Net Zero ambitions and our Trust's Green Plan. Using the Model for Improvement, we conducted three Plan-Do-Study-Act cycles. Our analogue and digital interventions included revised triage order sets, condition-specific electronic order sets, and a digital decision prompt requiring clinicians to confirm test indication. Data were collected at baseline (January-February 2024) and re-audited after interventions (December 2024). Outcome measures were the proportion of inappropriate tests, cost savings, staff time, and carbon reduction. At baseline, 34%-39% of coagulation screens were inappropriate across AMU areas. Postintervention, inappropriate testing reduced to 20% in mSDEC, 10% in AMU 1B, and 15% in AMU 1C. This equates to a projected annual reduction of 44 000 tests, saving £130 000, 367 staff hours, and 3.6 tonnes CO₂e (equivalent to a 9000-mile car journey). Laboratory workload and plastic waste also fell substantially. Embedding decision prompts within electronic order systems achieved rapid, sustained reductions in unnecessary testing. This scalable, low-cost intervention aligns clinical practice with sustainability goals and offers a model for reducing unwarranted diagnostics across the NHS. In the face of the climate crisis, aligning practice with environmental goals is both a professional responsibility and an opportunity to improve care, efficiency, and outcomes. Key messages What is already known on this topic Inappropriate coagulation testing is common across NHS acute care, with studies showing over one-third of tests lack a clinical indication.Excess diagnostic testing contributes to financial costs, staff workload, plastic waste, and carbon emissions.Previous quality improvement initiatives have focused mainly on education or guideline dissemination, with variable success. What this study adds Embedding a simple digital decision prompt into electronic order sets significantly reduced inappropriate coagulation screens across an Acute Medical Unit.The intervention was low-cost, rapidly implemented, and co-designed with frontline clinicians to improve uptake and sustainability.Projected impact includes avoidance of 44 000 tests annually, saving ~£130 000, 367 staff hours, and 3.6 tonnes CO₂e, supporting NHS Net Zero targets.This scalable model demonstrates how small digital changes can drive large improvements in clinical quality, efficiency, and sustainability.

Compliance with evidence-based treatment guidelines, supported by quality-assured parasitological diagnosis, is the mainstay of malaria case-management in Nigeria. However, despite increased attention, the quality of inpatient paediatric and adult, test-and-treat malaria case-management, and routine accuracy of malaria microscopy, has rarely been examined in public and private hospitals. A cross-sectional assessment was undertaken at 18 public and private hospitals in September 2024 in Kano State, Nigeria. Data collection included hospital assessments, interviews with inpatient health workers, review of all paediatric and medical ward admission files for August 2024, and re-checking of routine malaria slides archived during the 3-month post-assessment period. Descriptive analyses included 18 hospitals, 72 health workers, 2,814 suspected malaria admissions, and 211 malaria slides. Nearly all hospitals (94.4%) provided parasitological diagnostic services (microscopy or RDT) and stocked recommended antimalarials (injectable artesunate and ACT). Most health workers had received training on severe malaria (73.6%), but only 16.7% received supportive supervision. The composite test-and-treat performance was 39.3%, higher for children than adults (45.7% vs 26.5%) and in public compared to private hospitals (39.8% vs 30.8%). Among suspected malaria patients, 73.7% were tested on admissions and 90.2% of those with severe malaria were treated with artesunate. Children, compared to adults, were more commonly tested (79.8% vs 61.7%) and treated with artesunate (93.5% vs 80.1%). Patients in private hospitals, compared to public, were more often tested (84.3% vs 73.0%) but less frequently treated with artesunate (73.1% vs 91.2%). Only 30.0% of artesunate-treated patients were prescribed ACT-more commonly among adults than children (48.3% vs 23.0%) and in private than in public hospitals (89.2% vs 26.9%). ACT use for admitted non-severe cases was rare (2.4%), whereas non-compliance with test negative results was high (75.8%). The sensitivity, specificity, positive and negative predictive values of routine microscopy compared to expert readings were 93.2%, 42.5%, 29.9% and 95.9%, respectively. Inpatient compliance with malaria test-and-treat guidelines varied between performance tasks, age groups, and hospital sectors. Clinicians can be confident in negative slides but should be cautious with positive results. Quality assurance of malaria diagnosis and continuous clinical and laboratory quality improvement interventions, with enhanced linkages, are needed.

The authors analyzed methods to increase patient reach and clinical adoption of an electronic patient-reported outcome (ePRO) intervention, identified barriers and facilitators to ePRO implementation, and applied iterative modifications in psychiatric urgent care. ePRO implementation was examined from August 2021 to December 2023 at two psychiatric urgent care clinics in Maryland. The authors used the learning evaluation and RE-AIM (reach, effectiveness, adoption, implementation, and maintenance) frameworks to guide and evaluate the implementation of PROs via an ePRO system. Stakeholder feedback informed rapid iteration cycles, driving the development and deployment of technical and procedural modifications. Quantitative data were organized via RE-AIM metrics and analyzed with descriptive statistics and regression analyses. Qualitative data, derived from stakeholder feedback, were analyzed through deductive and inductive coding and sentiment analysis. A total of 22,610 care episodes were analyzed. Monthly ePRO completion increased to 73% in 2023. ePRO completion varied across clinics and was lower among men, Black patients, and those with neurocognitive or substance use disorders and was higher for patients with anxiety diagnoses. ePRO adoption increased in 2023. Six modifications were iteratively implemented to assist with ePRO completion and documentation. Qualitative analyses revealed administrative, clinical, and technological factors associated with ePRO implementation and completion rates and with an overall positive sentiment toward ePROs. Representing the largest known ePRO implementation in psychiatric urgent care, the study's findings suggest that a stakeholder-driven, iterative process can successfully integrate ePROs in fast-paced clinical environments and inform future quality improvement efforts.

Embedded, interventional health systems research is increasingly promoted to better understand and strengthen the performance of health systems. However, for these forms of research, boundaries between clinical care, quality improvement, and public health can be blurred, and ethical implications and frameworks to draw upon are unclear. While there is evolving ethical guidance, few health systems studies have documented ethical dilemmas experienced post ethics approval, and the value of support processes introduced to manage arising dilemmas. In this discussion paper, we share our approach to handling the ethical dilemmas that arose while conducting embedded interventional health systems research in public-sector newborn units in Kenya. Building on our past research, and literature on debriefs, reflective learning, ethics reflection groups and moral case deliberations, we evolved an approach to holding regular structured ethics debriefs to discuss and agree upon how to handle ethical issues experienced during 'fieldwork'. The research team maintained a 'living log' of all discussions, detailing all emerging ethical issues and any agreed actions. To prepare this paper, we conducted a thematic analysis of the living log and associated meeting minutes/recordings, and held a series of wider team meetings to reflect upon our learning. Numerous dilemmas were shared by research staff in our debrief fora. We grouped ethical issues encountered into 1) 'bystander' issues (defined here as background issues impacting health system functioning, facility staff, patients or families that were not caused or exacerbated by our research activities), 2) issues for those groups that were 'research imposed' and 3) issues related to the 'comfort and well-being of research team members'. Most dilemmas raised related to feeling like bystanders in highly constrained health systems, complicated by our positionalities as 'outsiders-within', whereby as health researchers spending time in facilities we were neither fully 'outsiders' nor 'insiders' to the health system. There was constant moral labour involved in considering our responsibilities for action, which ranged from immediate action from a safety perspective, through rethinking how the research was conducted, to various forms of engagement and feedback across a web of stakeholders. The approach we developed offers a framework to assist research team members with the significant ethical dilemmas and challenges that arise over the course of conducting studies. We suggest activities to support working prospectively through emerging ethical dilemmas in future studies.

Responsive caregiving supports infant and toddler wellbeing. Yet, based on nursing observational data, a significant number of one institution's inpatient infant and toddler patients with cancer-who are uniquely vulnerable due to the developmental risks associated with their illness and treatment-were not spoken to or held by their caregiver at any time when nursing was present over the course of day shifts. This clinical quality improvement project aimed to increase caregiver engagement in responsive interactions during inpatient stays. The Model for Improvement framework was used. Implementation, evaluation, and reporting followed the SQUIRE 2.0 framework. Root causes were analyzed with fishbone and key driver diagrams. Outcomes were tracked with control charts and percentage of nursing shifts during which responsive care was not observed. Statistical process control was used to study interventions. Two intervention cycles were completed and resulted in significant and meaningful (>1 sigma) reductions in nursing shifts during which infants and toddlers were not spoken to or held. Caregiver psychoeducation interventions increased responsive care of infants and toddlers in our oncology inpatient setting. This low-cost intervention may be adaptable across inpatient settings.

Although cases of squamous cell carcinoma of the cervix are declining, the increase in adenocarcinoma and in situ adenocarcinoma raises new concerns and requires improvements in diagnostic and therapeutic strategies. This study investigates the relationship between advanced CIN3 and concurrent cervical glandular diseases, emphasising the importance of high-risk HPV genotypes in the progression of these conditions. The clinical data of 104 patients who underwent conization for CIN3 at our centre in the last 10 years were reviewed. Cytological and histological diagnoses were performed according to the 2014 Bethesda system and WHO classification criteria, respectively. The analysis included patient age, HPV genotype of the surgical specimen, length of the cone, resection margin status, and follow-up details. Of the 104 patients, 92 (88.46%) were diagnosed with single CIN3 lesions (group 1), while 12 patients (11.54%) had coexisting glandular CIN3 lesions (group 2). All women tested were found to be positive for high-risk HPV (HPV-HR). The HPV18 genotype was found to be statistically more prevalent in group 2 (41.7%) than in group 1 (13%) (p = 0.011). At the first follow-up, HPV persistence was observed in 14/92 (15.2%) patients in group 1 and 10/12 (83.3%) patients in group 2 (p= 0.00001). Finally, the recurrence rate was 6.5% (6/92) in group 1 vs 30% (3/10) in group 2 (p= 0.012). In the management of cervical lesions, glandular pathology may go undetected at initial diagnosis, as it is often not visible on colposcopy or cytology. It is only through excisional treatment, which allows a more detailed analysis of the tissue, that these lesions can be identified. The presence of an associated glandular lesion radically changes the clinical picture and therapeutic management, requiring a more radical approach than that expected for an isolated CIN3 lesion. The results of the study not only highlight the complexity of CIN3 associated with AIS/AC lesions, but also the importance of an integrated strategy that includes early diagnosis, appropriate treatment, and rigorous follow-up. The ultimate goal is the improvement of clinical outcomes and quality of life for cervical cancer patients.

Spine care delivery in Canada is in a state of crisis and clinician advocacy to hospital leadership and government is essential to successfully implement positive change. While clinical research and quality improvement work are both well-established domains familiar to surgeons, business techniques such as granular financial quantification and operational modeling are not commonly used but can play a crucial role in advocacy efforts. This study describes 2 instances that business tools were used from October 2023 to October 2024 to successfully advocate for major progress in increasing the capacity of a provincial spine program. The first example of the use of a business technique was granular financial quantification to compare costs of an existing program to export spine surgeries to the USA against the costs of building capacity within the province instead - via capital investment to expand the spine program within the province. The second use case was quantification of the existing waitlist via an operational model in order to accurately ascertain the exact OR resources required to address a surgical backlog. The initiative led to the successful cancellation of surgeries being sent to the United States, yielding net savings of $30 949 per case and $4.1M per annum. The capital procurement for the secondary site expansion ($2.2M) along with an operating budget ($959 000) was approved due to the demonstration of a return on investment (ROI) of 27%. (2) The precise quantification of the waitlist demonstrated a 4.3 year waitlist for spine surgery, on track to reach 5.7 years by 2027. This data provided a strong case for additional OR time, which is being incrementally added, after a fifteen year status quo. Advocacy for additional resources was unsuccessful until the introduction of business tools resulting in accurate quantification (in this case, net profit/ROI values and precise wait times). We hope these techniques become more widely implemented nationally as our experience highlights the potential benefits of strategic data-driven decision-making.

The recent rise in the use of artificial intelligence (AI) in medicine has generated considerable enthusiasm. However, the emphasis on advanced tools, such as large language models, obscures the challenge of incomplete digital transformation in everyday clinical practice. In nephrology, as in other specialties, workflows remain heavily reliant on manual tasks and are often fragmented and non-interoperable. This situation not only adds a significant bureaucratic burden to clinicians but also hampers the development of high-quality and structured data, necessary to power AI models. In this review, we argue that a human-centered approach to digitalization is essential for unlocking AI's full potential in healthcare. By applying service design principles and prioritizing the needs and workflows of end-users, it becomes more probable that valuable digital health tools will be developed. This is best achieved through the active co-creation of these platforms with both healthcare professionals and patients, establishing true interoperability via common data standards, and designing systems that enable the capture of structured information for primary and secondary purposes, such as clinical research and quality improvement. We view nephrology as uniquely positioned to serve as a"living lab"for this digital transformation. The specialty manages a diverse patient population, including those with chronic kidney disease, transplant recipients, and individuals with rare diseases, all requiring complex and long-term care. By initially establishing a robust digital infrastructure to address urgent clinical challenges, we can lay the groundwork for AI to be deployed safely and effectively, thereby enhancing patient care.