The increasing use of minimally invasive surgery in acute general surgical conditions: A decade of results from a national data set.

BACKGROUND: To assess the rate of uptake of acute laparoscopic surgery for common general surgical conditions using national-level data. METHODS: The use of laparoscopic surgery in the acute management of appendicitis, cholecystitis, adhesive small bowel obstruction, and inguinal hernias was assessed between 2013 and 2022 at a national level in New Zealand. RESULTS: Laparoscopic appendicectomy increased from 83% to 95% (P = .0002). Laparoscopic cholecystectomy increased from 94% to 96% (P = .001). Laparoscopic adhesiolysis increased from 42% to 60% (P = .001). Laparoscopic inguinal hernia repair increased from 3% to 18% (P = .004). The rate of laparoscopic conversion demonstrated a decrease for appendicectomy (1.9% to 0.24%), cholecystectomy (0.77% to 0.39%), and adhesiolysis (9% to 2.4%) across this time. The laparoscopic cohorts were all associated with a shorter and less expensive length of stay compared to the open cohort. Maori and Pacific Island patients had largely equitable or superior rates of laparoscopic use compared to the rest of the population. No changes in laparoscopic use were detected during the COVID-19 pandemic. Rates of laparoscopic cholecystectomy and appendicectomy are similar throughout the regions. The largest difference in rates detected was for adhesiolysis, which was more common in the northern region. CONCLUSION: There has been a statistically significant rise in the use of acute laparoscopic surgery for acute general surgical procedures. This rise is likely clinically and economically significant, particularly in appendicectomy and adhesiolysis, with rises of 12% and 17% across the 10 years, with the known associated patient and health care system benefits.

Patient perspectives on the use of health information.

AIM: This survey aimed to investigate patient perspectives, including preferences, needs and concerns, on the use of, and access to, individual healthcare information. METHOD: A mixed-methods cross-sectional survey of adult patients (n=1,377) in Waitemata District Health Board inpatient and outpatient services during November-December 2020. The survey was online and on paper and available in 10 languages. RESULTS: Over 80% of participants were comfortable with their health information being used across the scenarios presented (range: 81-89%). Maori were significantly more likely than non-Maori to be comfortable with their health information being combined with the health information of others to better understand population needs (p=0.006). The level of comfort with the use of individual health information was related to assurances that its use was for public good, data were stored securely, individual privacy was maintained, the information was accurate and there was communication on how it was used. DISCUSSION: This study has shown that most healthcare consumers are comfortable with the health service using their de-identified health information beyond their care if it benefits others.

Training clinicians to lead clinical IT projects.

Across New Zealand, a huge programme of work is being initiated to improve the health information systems of our sector. The goals of this plan are to address major risks and issues such as cybersecurity and our inability to securely share health data across organisations for clinical care. To fulfil the promise of planned health IT initiatives, we must involve clinicians of all disciplines to help lead, design and implement projects. However, there is currently little pragmatic training available for clinicians to learn how to do so. In 2019, Waitemata District Health Board and the National Institute for Health Innovation developed and delivered a 'hands-on' Clinical Digital Academy training programme for multidisciplinary clinicians. This paper describes the programme, the initial cohort's evaluation feedback and recommendations for the future.

A machine learning model for predicting risk of hospital readmission within 30 days of discharge: validated with LACE index and patient at risk of hospital readmission (PARR) model.

The objective of this study was to design and develop a predictive model for 30-day risk of hospital readmission using machine learning techniques. The proposed predictive model was then validated with the two most commonly used risk of readmission models: LACE index and patient at risk of hospital readmission (PARR). The study cohort consisted of 180,118 admissions with 22,565 (12.5%) of actual readmissions within 30 days of hospital discharge, from 01 Jan 2015 to 31 Dec 2016 from two Auckland-region hospitals. We developed a machine learning model to predict 30-day readmissions using the model types XGBoost, Random Forests, and Adaboost with decision stumps as a base learner with different feature combinations and preprocessing procedures. The proposed model achieved the F1-score (0.386 ± 0.006), sensitivity (0.598 ± 0.013), positive predictive value (PPV) (0.285 ± 0.004), and negative predictive value (NPV) (0.932 ± 0.002). When compared with LACE and PARR(NZ) models, the proposed model achieved better F1-score by 12.7% compared with LACE and 23.2% compared with PARR(NZ). The mean sensitivity of the proposed model was 6.0% higher than LACE and 41% higher than PARR(NZ). The mean PPV was 15.9% and 14.6% higher than LACE and PARR(NZ) respectively. We presented an all-cause predictive model for 30-day risk of hospital readmission with an area under the receiver operating characteristics (AUROC) of 0.75 for the entire dataset. Graphical abstract.

Healthcare pathway discovery and probabilistic machine learning.

BACKGROUND AND PURPOSE: Healthcare pathways define the execution sequence of clinical activities as patients move through a treatment process, and they are critical for maintaining quality of care. The aim of this study is to combine healthcare pathway discovery with predictive models of individualized recovery times. The pathway discovery has a particular emphasis on producing pathway models that are easy to interpret for clinicians without a sufficient background in process mining. The predictive model takes the stochastic volatility of pathway performance indicators into account. METHOD: This study utilizes the business process-mining software ProM to design a process mining pipeline for healthcare pathway discovery and enrichment using hospital records. The efficacy of combining learned healthcare pathways with probabilistic machine learning models is demonstrated via a case study that applies the proposed process mining pipeline to discover appendicitis pathways from hospital records. Machine learning methodologies based on probabilistic programming are utilized to explore pathway features that influence patient recovery time. RESULTS: The produced appendicitis pathway models are easy for clinical interpretation and provide an unbiased overview of patient movements through the treatment process. Analysis of the discovered pathway model enables reasons for longer than usual treatment times to be explored and deviations from standard treatment pathways to be identified. A probabilistic regression model that estimates patient recovery time based on the information extracted by the process mining pipeline is developed and has the potential to be very useful for hospital scheduling purposes. CONCLUSION: This study establishes the application of the business process modelling tool ProM for the improvement of healthcare pathway mining methods. The proposed pipeline for healthcare pathway discovery has the potential to support the development of probabilistic machine learning models to further relate healthcare pathways to performance indicators such as patient recovery time.

Predicting Patients at Risk of 30-Day Unplanned Hospital Readmission.

We developed a machine learning model to predict 30-day readmissions using the model types; XGBoost, Random Forests and Adaboost with decision stumps as a base learner with different feature combinations and preprocessing procedures. The proposed model achieved the F1-score (0.386 ± 0.006), sensitivity (0.598 ± 0.013), positive predictive value (PPV) (0.285 ± 0.004) and negative predictive value (NPV) (0.932 ± 0.002). When compared with LACE and PARR (NZ) models, the proposed model achieved better F1-score by 12.5% compared to LACE and 22.9% compared to PARR (NZ). The mean sensitivity of the proposed model was 6.0% higher than LACE and 42.4% higher than PARR (NZ). The mean PPV was 15.9% and 13.5% higher than LACE and PARR (NZ) respectively.

Machine Learning-based Risk of Hospital Readmissions: Predicting Acute Readmissions within 30 Days of Discharge.

The objective of this study was to design and develop a 30-day risk of hospital readmission predictive model using machine learning techniques. The proposed risk of readmission predictive model was then validated with the two most commonly used risk of readmission models - LACE index and patient at-risk of hospital readmission (PARR). The study cohort consisted of 180,118 admissions with 22565 (12.5%) of actual readmissions within 30-day of hospital discharge, from 01 Jan 2015 to 31 Dec 2016 from two Auckland-region hospitals. We developed a machine learning model to predict 30-day readmissions using the model types: XGBoost, Random Forests and Adaboost with decision stumps as a base learner with different feature combinations and preprocessing procedures. The proposed model achieved the F1-score (0.386 ± 0.006), sensitivity (0.598 ± 0.013), positive predictive value (PPV) (0.285 ± 0.004) and negative predictive value (NPV) (0.932 ± 0.002). When compared with LACE and PARR (NZ) models, the proposed model achieved better F1-score by 12.5% compared to LACE and 22.9% compared to PARR (NZ). The mean sensitivity of the proposed model was 6.0% higher than LACE and 42.4% higher than PARR (NZ). The mean PPV was 15.9% and 13.5% higher than LACE and PARR (NZ) respectively.

Improving quality of clinical coding of post-partum haemorrhage: the process and its effects on reported incidence rates in a New Zealand hospital maternity service.

AIM: The Waitemata District Health Board (DHB) aimed to investigate and improve the accuracy of its reporting of post-partum haemorrhage (PPH), to understand its true incidence. METHOD: The quality improvement project included multidisciplinary collaboration between maternity clinicians and clinical coders, substantive redesign of the Waitemata DHB's birth documentation form, systematic auditing and follow-up of clinical documentation by a dedicated quality midwife, linking of maternity clinicians to a key designated senior coder and ongoing PPH incidence monitoring and staff education. RESULTS: The coded rate of PPH has risen dramatically and is now in line with expected Australasian incidence levels. A corresponding increase in the value of cost-weighted discharges (estimated at $544,000 for the 2015/16 financial year) was realised as a result of the more accurate reported incidence. CONCLUSION: This case illustrates the value of coding to a clinical service and the importance of clinical leadership and engagement in achieving successful and sustainable service redesign initiatives. It provides an example of how to evaluate and update coding and a process for changing the way clinicians and coders work that could benefit other services in Waitemata DHB as well as in other New Zealand district health boards.

Frequency and cost of acute surgical admissions in over-anticoagulated patients: over-anticoagulation in the community.

BACKGROUND: Anticoagulation treatment in the community is common. This investigation was undertaken to determine the frequency of patient surgical admission with conditions associated with over-anticoagulation in the community and the surgical resource required to effectively and safely manage these patients acutely. METHODS: Hospital discharge data on individual patients admitted to Waitemata District Health Board hospitals between December 2014 and November 2015 inclusive were reviewed. Data were extracted on individual patients with relevant ICD-10 codes (D683, Y442, Y443). Individual records for patients admitted to general surgery were then reviewed and costing information related to admissions extracted. RESULTS: A total of 551 patients were admitted in a 12-month period (4.8 admissions/1000 warfarin users and 2.4 admissions/1000 dabigatran users) for conditions associated with over-anticoagulation, with 35 admitted to the general surgery service, of whom 29 were taking warfarin and six taking dabigatran. A total of 21 patients were admitted with haemorrhagic conditions, and 14 over-anticoagulated patients were admitted with general surgical conditions requiring treatment. All patients were managed by withholding anticoagulant medication, 12 required formal reversal, three required red cell transfusion and four haemostatic procedures. The average hospital stay was 4 days, with inpatient costs of NZ$3500. CONCLUSION: Management of patients admitted with over-anticoagulation in the community is a significant surgical workload; however, acute management is usually straightforward.

From scratch: developing a hepatic resection service for metastatic colorectal cancer.

BACKGROUND: Waitemata District Health Board has New Zealand's largest catchment and busiest colorectal unit. The upper gastrointestinal unit was established in 2005, in part to provide a hepatic resection service for patients with colorectal carcinoma metastatic to the liver. The aim of this investigation was to report on quality indicators for the hepatic resection of colorectal carcinoma in the development of a regional resection service. METHODS: Prospectively collected data on patients undergoing hepatic resection for colorectal carcinoma between 2005 and 2014 was reviewed and correlated with costing data and national hepatic resection rates. RESULTS: A total of 123 patients underwent 138 hepatic resections for metastatic colorectal cancer with a median hospital stay of 8 days (range 4-37 days), a zero 30-day mortality and a median cost of NZ$21 374 for minor hepatectomy and NZ$43 133 for major hepatectomy. Actuarial 5-year disease-free survival was 44%, with 28 patients alive and disease free at 5 years post-resection. Median overall survival was not reached. Review of national hepatic resection rates indicate that Waitemata District Health Board performs one sixth of all hepatic resections in New Zealand and that this treatment modality may be underutilized in the management of patients with metastatic colorectal cancer. CONCLUSION: A regional hepatic resection centre for colorectal metastases can be established in areas of population need and can provide a high-quality, cost-effective service.

The volume, cost and outcomes of pancreatic resection in a regional centre in New Zealand.

BACKGROUND: The frequency, costs and outcome of pancreatic resection (both pancreaticoduodenectomy and distal pancreatectomy) were reviewed in our own institution and correlated with regional population growth as well as national resection rates and locations. METHODS: Demographic, pathological and outcome data on pancreaticoduodenectomy and distal pancreatectomy were obtained from a prospectively maintained database for the years 2005-2009 and 2010-2014. During this period, the catchment population grew from 460 000 to 567 000. Costing information was obtained from the hospital-independent costing and coding committee, and the locations and rates of pancreatic resection were obtained by interrogating the national minimum dataset. RESULTS: A total of 41 pancreatectomies (29 pancreaticoduodenectomy, 12 distal pancreatectomy) were performed between 2005 and 2009, increasing to 84 pancreatectomies (55 pancreaticoduodenectomies, 27 distal pancreatectomies and two total pancreatectomies) between 2010 and 2014. This constituted one sixth of the national volume of pancreatic resections. There was no difference in patient demographics or indications for resection between the two time periods; however, portal vein resection was used more frequently in the second period. Margin positivity rate decreased (7 of 41 versus 8 of 84) and lymph node harvest increased (median 8 nodes versus median 15 nodes) between the two time periods. Overall 30-day mortality was 1.6%. CONCLUSION: In New Zealand, regional rates of pancreatic resection reflect regional population demands, and institutional growth is driven by local population requirements. Institutional growth can be achieved with the maintenance of internationally accepted outcomes and quality indicators.

Costs of bariatric surgery in a randomised control trial (RCT) comparing Roux en Y gastric bypass vs sleeve gastrectomy in morbidly obese diabetic patients.

AIM: To provide a longitudinal analysis of the direct healthcare costs of providing laparoscopic sleeve gastrectomy (LSG) and laparoscopic Roux-en-Y gastric bypass (LRYGB) surgery service in the context of a randomised control trial (RCT) of obese patients with type 2 diabetes in Waitemata District Health Board, Auckland, New Zealand. METHODS: The Waitemata District Health Board costing system was used to calculate costs in New Zealand Dollars (NZD) associated with all pre- and post-operative hospital clinic visits, peri-operative care, hospitalisations and medication costs up to one year after bariatric surgery. Healthcare costs of medications, laboratory investigations and hospital clinic visits for one year prior to enrolment into the RCT were also calculated. RESULTS: One hundred and fourteen patients were randomised to undergo laparoscopic sleeve gastrectomy (LSG, n=58) or laparoscopic Roux en Y gastric bypass (LRYGB, n=56). Total costs one year pre-enrolment was $203,926 for all patients (mean $1,789 per patient). Total cost of surgery was $1,208,005 (mean $9,131 per LSG patient and mean $12,456 per LRYGB patient). Total cost one year post-operatively was $542,656 (mean $4,760 per patient). The total medication cost reduced from $118,993.72(mean $1,044 per patient) to $31,304.93 (mean $274.60 per patient), p<0.005. The largest cost reduction was seen with annual diabetic medications reducing from $110,115.78(mean $965.93 per patient) to $7,237.85 (mean $63.48 per patient), p<0.005. CONCLUSIONS: Among patients with type 2 diabetes and morbid obesity undergoing LSG and LRYGB, health service costs were greater in the year after surgery than in the year before, although prescription costs were lower post-operatively. There was no significant difference in reduction in prescription cost by surgical procedure at 12 months. However, the LRYGB surgery was more expensive than LSG, primarily because of the longer operative time required.

Evaluation of a New Zealand program to improve transition of care for older high risk adults.

Transition interventions aim to improve care and reduce hospital readmissions but evaluations of these interventions have reported inconsistent results. We report on the evaluation of an intervention implemented in Auckland, New Zealand. Participants were people over the age of 65 who had an acute medical admission and were at high risk of readmission. The intervention included an improved discharge process and nurse telephone follow-up soon after discharge. Outcomes were 28 day readmission rates and emergency attendances. The study is observational, using both interrupted times series and regression discontinuity designs. 5239 patients were treated over a one year period. There was no change in readmission rates or ED attendances or secondary outcomes. Not all patients received all components of the intervention. This transition intervention was not successful. Possible reasons for this and implications are discussed. Although non-experimental methods were used, we believe the results are robust.

Do gerontology nurse specialists make a difference in hospitalization of long-term care residents? Results of a randomized comparison trial.

Residents of long-term care facilities have highly complex care needs and quality of care is of international concern. Maintaining resident wellness through proactive assessment and early intervention is key to decreasing the need for acute hospitalization. The Residential Aged Care Integration Program (RACIP) is a quality improvement intervention to support residential aged care staff and includes on-site support, education, clinical coaching, and care coordination provided by gerontology nurse specialists (GNSs) employed by a large district health board. The effect of the outreach program was evaluated through a randomized comparison of hospitalization 1 year before and after program implementation. The sample included 29 intervention facilities (1,425 residents) and 25 comparison facilities (1,128 residents) receiving usual care. Acute hospitalization rate unexpectedly increased for both groups after program implementation, although the rate of increase was significantly less for the intervention facilities. The hospitalization rate after the intervention increased 59% for the comparison group and 16% for the intervention group (rate ratio (RR) = 0.73, 95% confidence interval (CI) = 0.61-0.86, P < .001). Subgroup analysis showed a significantly lower rate change for those admitted for medical reasons for the intervention group (13% increase) than the comparison group (69% increase) (RR = 0.67, 95% CI = 0.56-0.82, P < .001). Conversely, there was no significant difference in the RR for surgical admissions between the intervention and comparison groups (RR = 1.0, 95% CI = 0.68-1.46, P = .99). The integration of GNS expertise through the RACIP intervention may be one approach to support staff to provide optimal care and potentially improve resident health.

Timely delivery of hip fracture care: a Middlemore Hospital audit.

INTRODUCTION: New Zealand (NZ) hospitals lack a centralised audit process to evaluate hip fracture care whereas UK hospitals audit hip fracture care in relation to best practice guidelines. This study sought to evaluate multiple factors in hip fracture care at Middlemore Hospital (MMH). Comparisons were made with an audit from MMH in 2008 and a multicentre UK audit. METHOD: A retrospective audit of patients with hip fractures was carried out at MMH between January and June 2012. RESULTS: 120 patient charts were reviewed. In 2012, 14.2% of patients were admitted from ED within the guideline recommended period of four hours compared to 5.6% of patients in 2008. 72.5% received operative management within the guideline suggested period of 48 hours in comparison to 51% in 2008. Lack of available theatre space accounted for 51% of delays in 2008. CONCLUSION: There have been considerable improvements to timely delivery of hip fracture care at MMH between 2008 and 2012. However, there are ongoing delays to ward admission and operative management at our institution resulting in care that falls beyond the times recommended by international guidelines. The lack of available theatre space remains a major cause of delayed surgery. We advocate the development of a multicentre audit in NZ hospitals.

Improving the productivity of elective surgery through a new 'package of care'.

In response to a need to improve the productivity and throughput of elective surgical services, one district health board (DHB) in New Zealand has introduced a 'package of care' (POC) in which incentive-based, risk-sharing contracts were developed collaboratively between DHB managers, surgeons and anaesthetists. The POC includes throughput targets and facilitates consistent surgical teams and the cohorting of patients. Whilst many staff are very supportive of the POC, some are of the view that it conflicts with the ideals and principles of working in a public health system, and creates inequities amongst the hospital staff. Analysis indicates that, after controlling for age, casemix and complexity the POC has resulted in shorter theatre times, shorter lengths of stay and lower average inpatient event costs compared with standard care at the public hospital. An unintended consequence could be that the POC may encourage throughput of less complex cases at the expense of more complex cases. The average complexity and range of cases performed publicly should be carefully monitored to ensure this does not occur.