Patterns of Mental Health Service Utilisation: A Population-Based Linkage of Over 17 Years of Health Administrative Records.

A cross-sectoral partnership was formed in 2021 in support of the recommendations in an audit on access to state-funded mental health services. In this first paper, we aimed to describe the demographic and service utilisation of adults with a mental health diagnosis in the Western Australian state-funded health system from 2005 to 2021. Inpatient, emergency department, specialised (ambulatory) community mental health service, and death records were linked in individuals aged ≥ 18 years with a mental health diagnosis in Western Australia. Altogether, 392,238 individuals with at least one mental health service contact between 1st January 2005 and 31st December 2021 were included for analysis. Females, Aboriginal and/or Torres Strait Islander people, and those who lived outside major cities or in the most disadvantaged areas were more likely to access state-funded mental health services. While the number of individuals who accessed community mental health services increased over time (from 28,769 in 2005 to 50,690 in 2021), the percentage increase relative to 2005 was notably greater for emergency department attendances (127% for emergency department; 76% for community; and 63% for inpatient). Conditions that contributed to the increase for emergency department were mainly alcohol disorder, reaction to severe stress and adjustment disorders, and anxiety disorders. Sex differences were observed between conditions. The pattern of access increased for emergency department and the community plus emergency department combination. This study confirmed that the patterns of access of state-funded mental health services have changed markedly over time and the potential drivers underlying these changes warrant further investigation.

Early outcomes after post-cardiotomy extracorporeal membrane oxygenation in paediatric patients: a contemporary, binational cohort study.

OBJECTIVES: The aim of this study was to assess the early outcomes and risk factors of paediatric patients requiring extracorporeal membrane oxygenation after cardiac surgery (post-cardiotomy). METHODS: Retrospective binational cohort study from the Australia and New Zealand Congenital Outcomes Registry for Surgery database. All patients younger than 18 years of age who underwent a paediatric cardiac surgical procedure from 1 January 2013 to 31 December 2021 and required post-cardiotomy extracorporeal membrane oxygenation (PC-ECMO) in the same hospital admission were included in the study. RESULTS: Of the 12 290 patients included in the study, 376 patients required post-cardiotomy ECMO (3%). Amongst these patients, hospital mortality was 35.6% and two-thirds of patients experienced a major complication. Hypoplastic left heart syndrome was the most common diagnosis (17%). The Norwood procedure and modified Blalock-Taussig shunts had the highest incidence of requiring PC-ECMO (odds ratio of 10 and 6.8 respectively). Predictors of hospital mortality after PC-ECMO included single-ventricle physiology, intracranial haemorrhage and chylothorax. CONCLUSIONS: In the current era, one-third of patients who required PC-ECMO after paediatric cardiac surgery in Australia and New Zealand did not survive to hospital discharge. The Norwood procedure and isolated modified Blalock-Taussig shunt had the highest incidence of requiring PC-ECMO. Patients undergoing the Norwood procedure had the highest mortality (48%). Two-thirds of patients on PC-ECMO developed a major complication.

CardioCel® for repair of congenital heart defects: nationwide results of over 1000 implants.

OBJECTIVES: To assess the mid-term performance of CardioCel for the repair of congenital heart defects. METHODS: Data were retrospectively collected from databases and hospital records in 3 congenital cardiac surgery centres in Australia. Kaplan-Meier curves and log-rank tests were used to test for associations between patient age, gender, patch type and site of implantation. Multivariable Cox regression was used to test whether any specific implantation site was associated with reintervention risk, after adjusting for age group, gender and patch type. RESULTS: A total of 1184 CardioCel patches were implanted in 752 patients under the age of 18 years. Median age at implant was 12 months [interquartile range (IQR) 3.6-84]. Median follow-up was 2.1 years (IQR 0.6-4.6). Probability of freedom from CardioCel-related reintervention was 93% [95% confidence interval (CI) 91-95] at 1 year, 91% (95% CI 88-93) at 3 years and 88% (95% CI 85-91) at 5 years, respectively. On multivariable regression analysis, aortic valve repair had a higher incidence of reintervention [hazard ratio (HR) = 7.15, P = 0.008] compared to other sites. The probability of reintervention was higher in neonates (HR = 6.71, P = 0.0007), especially when used for augmentation of the pulmonary arteries (HR = 14.38, P = 0.029), as compared to other age groups. CONCLUSIONS: CardioCel can be used for the repair of a variety of congenital heart defects. In our study, in patients receiving a CardioCel implant, reinterventions were higher when CardioCel was used to augment the pulmonary arteries in neonates and for aortic valve repair as compared to other sites.

Mid- and long-term outcomes after surgical correction of subaortic stenosis: a 27-year experience.

OBJECTIVES: We reviewed the mid- and long-term surgical outcomes of patients with subaortic stenosis (SAS). METHODS: Patients operated for SAS from April 1990 to August 2016 were reviewed retrospectively. Patients with major associations such as aortic arch obstruction were excluded. Time to reintervention and predictors of recurrence were assessed using Kaplan-Meier analysis, log-rank test and uni/multivariable Cox regression. RESULTS: 120 patients at a median age of 4.7 years (interquartile range 2.9, 8.1) underwent primary operation (median peak preoperative left ventricular outflow tract gradient 52.5 mmHg, interquartile range 40, 70) involving fibrous tissue excision (n = 120) with septal myectomy (93%; n = 112) as the procedure of choice.At median follow-up of 13 years (interquartile range 7, 18), freedom from reintervention at 1, 3, 5 and 10 years was 99% (95% confidence interval 94%, 99%), 94% (87%, 97%), 93% (86%, 96%) and 90% (82%, 94%), respectively. Recurrence occurred in 18% (n = 20) with 15 patients undergoing reinterventions, 13 of whom required radical reoperation. Multivariable analysis revealed higher preoperative peak left ventricular outflow tract gradient (hazard risk 1.06, confidence interval 1.03, 1.09, P < 0.001), and presence of bicuspid aortic valve (hazard risk 14.13, confidence interval 3.32, 60.1, P < 0.001) as predictors for reintervention. Mild/moderate aortic regurgitation occurred in 49% (n = 55) of patients at the most recent follow-up. CONCLUSIONS: Reintervention for recurrent SAS is common, predicted by higher preoperative peak left ventricular outflow tract gradient, and presence of bicuspid aortic valve, and frequently involves a radical procedure. Aortic regurgitation is a major consequence of SAS, but its severity usually remains low. CLINICAL REGISTRATION NUMBER: SCHN HREC reference number 2019/ETH02729, approved on 09 July 2019.

Utility of routine chest radiographs after chest drain removal in paediatric cardiac surgical patients-a retrospective analysis of 1076 patients.

OBJECTIVES: Chest drains are routinely placed in children following cardiac surgery. The purpose of this study was to determine the incidence of a clinically relevant pneumothorax and/or pleural effusion after drain removal and to ascertain if a chest radiograph can be safely avoided following chest drain removal. METHODS: This single-centre retrospective cohort study included all patients under 18 years of age who underwent cardiac surgery between January 2015 and December 2019 with the insertion of mediastinal and/or pleural drains. Exclusion criteria were chest drain/s in situ ≥14 days and mortality prior to removal of chest drain/s. A drain removal episode was defined as the removal of ≥1 drains during the same episode of analgesia ± sedation. All chest drains were removed using a standard protocol. Chest radiographs following chest drain removal were reviewed by 2 investigators. RESULTS: In all, 1076 patients were identified (median age: 292 days, median weight: 7.8 kg). There were 1587 drain removal episodes involving 2365 drains [mediastinal (n = 1347), right pleural (n = 598), left pleural (n = 420)]. Chest radiographs were performed after 1301 drain removal episodes [mediastinal (n = 1062); right pleural (n = 597); left pleural (n = 420)]. Chest radiographs were abnormal after 152 (12%) drain removal episodes [pneumothorax (n = 43), pleural effusion (n = 98), hydropneumothorax (n = 11)]. Symptoms/signs were present in 30 (2.3%) patients. Eleven (<1%) required medical management. One required reintubation and 2 required chest drain reinsertion. CONCLUSIONS: The incidence of clinically significant pneumothorax/pleural effusion following chest drain removal after paediatric cardiac surgery is low (<1%). Most patients did not require reinsertion of a chest drain. It is reasonable not to perform routine chest radiographs following chest drain removal in most paediatric cardiac surgical patients.

A machine learning approach to predicting 30-day mortality following paediatric cardiac surgery: findings from the Australia New Zealand Congenital Outcomes Registry for Surgery (ANZCORS).

OBJECTIVES: We aim to develop the first risk prediction model for 30-day mortality for the Australian and New Zealand patient populations and examine whether machine learning (ML) algorithms outperform traditional statistical approaches. METHODS: Data from the Australia New Zealand Congenital Outcomes Registry for Surgery, which contains information on every paediatric cardiac surgical encounter in Australian and New Zealand for patients aged <18 years between January 2013 and December 2021, were analysed (n = 14 343). The outcome was mortality within the 30-day period following a surgical encounter, with ∼30% of the observations randomly selected to be used for validation of the final model. Three different ML methods were used, all of which employed five-fold cross-validation to prevent overfitting, with model performance judged primarily by the area under the receiver operating curve (AUC). RESULTS: Among the 14 343 30-day periods, there were 188 deaths (1.3%). In the validation data, the gradient-boosted tree obtained the best performance [AUC = 0.87, 95% confidence interval = (0.82, 0.92); calibration = 0.97, 95% confidence interval = (0.72, 1.27)], outperforming penalized logistic regression and artificial neural networks (AUC of 0.82 and 0.81, respectively). The strongest predictors of mortality in the gradient boosting trees were patient weight, STAT score, age and gender. CONCLUSIONS: Our risk prediction model outperformed logistic regression and achieved a level of discrimination comparable to the PRAiS2 and Society of Thoracic Surgery Congenital Heart Surgery Database mortality risk models (both which obtained AUC = 0.86). Non-linear ML methods can be used to construct accurate clinical risk prediction tools.

Performance of CardioCel in Cardiac Surgery: A Systematic Review.

A systematic review was performed for evaluation of the performance of CardioCel® in cardiac surgery. The review included all studies published from January 2013 to December 2020. We conclude that CardioCel is a strong, flexible tissue substitute with good handling characteristics and a low incidence of thrombosis, aneurysm formation, infection, or structural degeneration. It can be used for a variety of intracardiac and extracardiac repairs of congenital heart defects in all age groups with good durability at mid-term follow-up. However, the use of CardioCel in certain positions requires caution. Information on the long-term performance of CardioCel is lacking.

Trends of perinatal mental health referrals and psychiatric admissions in Queensland.

BACKGROUND: Perinatal depression is often underdiagnosed; consequently, many women suffer perinatal depression without follow-up care. Screening for depressive symptoms during the perinatal period has been recommended in Australia to increase detection and follow-up of women suffering from depressive symptoms. Screening rates have gradually increased over the last decades in Australia. OBJECTIVE: To explore trends in referrals of women to community mental health services during the perinatal period, and prenatal and postnatal admissions to psychiatric units, among those who gave birth in Queensland between 2009 and 2015. METHOD: Retrospective analyses of data from three linked state-wide administrative data collections. Trend analyses using adjusted Poisson regression models examined 426,242 births. Outcome variables included referrals to specialised mental health services; women admitted with a mood disorder during the second half of their pregnancy and during the first 3 months of the postnatal period; and women admitted with non-affective psychosis disorders during the second half of their pregnancy and during the first 3 months of the postnatal period. RESULTS: We found an increase in mental health referrals during the perinatal period over time (adjusted incidence rate ratio, 1.07; 95% confidence interval, [1.06, 1.08]) and a decrease in admissions with mood disorders during the first 3 months of the postnatal period (adjusted incidence rate ratio, 0.95; 95% confidence interval, [0.94, 0.98]). We did not find any changes in rates of admission for other outcomes. CONCLUSION: Since the introduction of universal screening in Queensland, referrals for mental health care during the perinatal period have increased, while admissions for mood disorders in the first 3 months after delivery decreased.

Technique of Coronary Button Transfer Has no Impact on Neoaortic Root Size in Simple Transposition.

We studied the effect of various coronary transfer techniques (CTT) on neo-aortic root size after an arterial switch operation (ASO) in simple transposition by excluding the impact of recognized predisposing factors. One hundred and seventy-eight patients with simple transposition were reviewed retrospectively (January 2004-December 2018) and grouped as Punch Hole (n = 83/178), Nonpunch Hole (n = 65/178; Trapdoor or Standard) and Mixed (n = 30/178). Factors predicting the neo-aortic root z-scores- annulus, mid-sinus, and sinotubular junction (STJ) were analyzed by uni/multivariable linear regression. Follow-up was 6 years, Interquartile range (IQR) 3.4,10.6. Preoperative aortic (7.4 mm, IQR 6.9,8) and pulmonary annulus (7.5 mm, IQR 6.8,8.1) sizes were identical (P = 0.831). The changes in preoperative, postoperative, and latest median z-scores for neo-aortic annulus (-0.2, IQR -1.2,0.9; 0.0, IQR -0.9,0.9; 0.9, IQR -0.4,2.6; P < 0.001), mid-sinus (1.1, IQR-0.1,2; 2.6, IQR 1.6,3.7; 2.9, IQR 1.8,4.3; P < 0.001) and STJ (-0.1, IQR -0.8,1.1; 2.1, IQR 0.7,3; 2.4, IQR 1,3.5; P < 0.001) were significant. On multivariate analysis, preoperative pulmonary annulus z-score predicted the latest neo-aortic annulus z-score [Beta estimate (BE) = 0.32, 95% confidence interval (CI) = 0.03,0.62; P = 0.03] and STJ z-score (BE= 0.45, 95% CI= 0.20,0.70; P < 0.0001). CTT did not predict any of the latest neo-aortic z-scores (all P > 0.05). Mild plus neo-aortic regurgitation (neoAR) was not significantly different across CTT groups [punch hole 20% (n = 15/74), mixed 37% (n = 11/30), nonpunch hole 21% (n = 13/62); Fisher-exact P = 0.186], one patient required valve replacement for severe neoAR. The neo-aortic root enlarges significantly over time at all 3 levels following an ASO in simple transposition, however, this is not significantly influenced by the CTT utilized.

Exploring the utility of synthetic data to extract more value from sensitive health data assets: A focused example in perinatal epidemiology.

BACKGROUND: Privacy, access and security concerns can hinder the availability of health data for research. The use of synthesised data in place of de-identified electronic health records (EHRs) presents an opportunity to conduct research while minimising privacy concerns. OBJECTIVES: To examine whether synthesised data can replicate two prenatal epidemiological associations: between prenatal smoking and lower birthweight, and between prenatal mood disorders and lower birthweight, using data synthesised from de-identified health administrative data collections. METHODS: We generated two synthetic datasets, using parametric and non-parametric data generating methods, and examined the synthetic data for evidence of privacy concerns. Next, univariable and multivariable logistic regression was utilised to estimate the associations in both synthetic datasets, with results then compared to the real data. RESULTS: Both synthesised datasets performed well in identifying the reduction in birthweight associated with prenatal smoking, while the non-parametric data underestimated the reduction in birthweight associated with prenatal mood disorders. Improbable relationships between some variables were identified in the parametric synthesised data, however, these can be addressed with simple rules during data synthesis. No duplicate rows (i.e., exact copies of de-identified data) were found in the parametric data, while only 0.6% of the rows in the non-parametric data were duplicated. CONCLUSIONS: Both synthesised datasets performed well in replicating the statistical properties of the original data while addressing privacy issues. Data synthesis methods provide an opportunity for researchers to utilise health data while managing privacy and security concerns.

Feeding Neonates and Infants Prior to Surgery for Congenital Heart Defects: Systematic Review and Meta-Analysis.

BACKGROUND: Necrotising enterocolitis (NEC) is a significant cause of mortality and morbidity in neonates requiring cardiac surgery. Feeding practices vary significantly across institutions and remain controversial. We conducted a systematic review of the literature and a meta-analysis to identify associations between feeding practices and necrotising enterocolitis. METHODS: This study was carried out in accordance with the PRISMA guidelines. A literature search was performed in November 2022 using the Cochrane Central Register, Embase, and Pubmed. Two investigators then independently retrieved eligible manuscripts considered suitable for inclusion. Data extracted included gestational age, birth weight, sex, nature of congenital heart lesion, type of operation performed, time on ventilator, ICU stay, hospital stay, post-operative feeding strategy, and complications. The methodological quality was assessed using the Downs and Black score for all randomised control trials and observational studies. RESULTS: The initial search yielded 92 studies. After removing duplicates, there were 85 abstracts remaining. After excluding ineligible studies, 8 studies were included for the meta-analysis. There was no significant risk of NEC associated with pre-operative feeding [OR = 1.22 (95% CI 0.77,1.92)] or umbilical artery catheter placement [OR = 0.91 (95% CI 0.44, 1.89)] and neither outcome exhibited heterogeneity [I2 = 8% and 0%, respectively]. There was a significant association between HLHS and NEC [OR = 2.56 (95% CI 1.56, 4.19)] as well as prematurity and NEC [OR 3.34 (95% CI 1.94, 5.75)] and neither outcome exhibited heterogeneity [I2 = 0% and 0%, respectively]. CONCLUSIONS: There was no association between NEC and pre-operative feeding status in neonates awaiting cardiac surgery. Pre-operative feeding status was not associated with prolonged hospital stay or need for tube assisted feeding at discharge. HLHS and prematurity were associated with increased incidence of NEC.

The Australia and New Zealand Congenital Outcomes Registry for Surgery (ANZCORS): methodology and preliminary results.

BACKGROUND: Analysis of multi-institutional data and benchmarking is an accepted accreditation standard in cardiac surgery. Such a database does not exist for congenital cardiac surgery in Australia and New Zealand (ANZ). To fill this gap, the ANZ Congenital Outcomes Registry for Surgery (ANZCORS) was established in 2017. METHODS: Inclusion criteria included all cardiothoracic and extracorporeal membrane oxygenation (ECMO) procedures performed at five participating centres. Data was collected by data managers, validated by the surgical team, and securely transmitted to a central repository. RESULTS: Between 2015 and 2019, 9723 procedures were performed in 7003 patients. Cardiopulmonary bypass was utilized for 59% and 9% were ECMO procedures. Fifty-seven percent (n = 5531) of the procedures were performed in children younger than 1 year of age. Twenty-four percent of procedures (n = 2365) were performed in neonates (≤28 days) and 33% (n = 3166) were performed in children aged 29 days to 1 year (infants). The 30-day mortality for cardiac cases (n = 6572) was 1.3% and there was no statistical difference between the participating centres (P = 0.491). Sixty-nine percent of cases had no major post-operative complications (5121/7456). For cardiopulmonary bypass procedures (n = 5774), median stay in intensive care and hospital was 2 days (IQR 1, 4) and 9 days (IQR 5, 18), respectively. CONCLUSION: ANZCORS has facilitated pooled data analysis for paediatric cardiac surgery across ANZ for the first time. Overall mortality was low. Non-risk-adjusted 30-day mortality for individual procedures was similar in all units. The continued evaluation of surgical outcomes through ANZCORS will drive quality assessment in paediatric cardiac surgery across ANZ.

Association between screening for antenatal depressive symptoms and delivery outcomes: The Born in Queensland Study.

BACKGROUND: Evidence shows that depressive symptoms during pregnancy increase the risk of an intervention during delivery (induction, the use of forceps or vacuum, and caesarean sections (CS)). Many women with depression during pregnancy are not identified and therefore will not receive appropriate follow up of their symptoms. We hypothesised that routine screening for depressive symptoms during pregnancy could reduce detrimental consequences of depressive symptoms on delivery outcomes. AIM: We explored the association between screening for depressive symptoms during pregnancy and delivery outcomes. MATERIALS AND METHODS: A cross-sectional analysis of state-wide administrative data sets. The population included all women who delivered a singleton in Queensland between the July and December of 2015. Logistic regression analyses were run in 27 501 women (93.1% of the total population) with information in all variables. The following were the main outcomes: onset of labour, CS, instrumental vaginal delivery, and all operative deliveries (including both CS and instrumental vaginal deliveries). RESULTS: Women who completed the screening had increased odds of a spontaneous onset of labour (adjusted odds ratio (aOR) 1.18; 95% CI 1.09-1.27) and decreased odds of an operative delivery (instrumental or CS) (aOR 0.88; 95% CI 0.81-0.96). Among women who had a vaginal delivery, those who completed the screening had decreased odds of having an instrumental delivery (aOR 0.84; 95% CI 0.74-0.97). Sensitivity analyses in women who did not have a formal diagnosis of depression showed similar results. CONCLUSION: Our findings suggest that screening may decrease interventions during delivery in women with depressive symptoms.

Comparison of homografts and bovine jugular vein conduits in the pulmonary position in patients <20 years of age.

OBJECTIVE: To compare the performance of homografts and bovine jugular vein (BJV) conduits in the pulmonary position. METHODS: All patients with congenital heart disease up to age 20 years who underwent pulmonary valve replacement with homografts or BJV at 3 centers in Australia were evaluated. There were 674 conduits, with 305 (45%) pulmonary homografts (PHs), 303 (45%) BJV conduits, and 66 (10%) aortic homografts (AHs). Endpoints were freedom from reintervention, structural valve degeneration (SVD), and infective endocarditis (IE). Propensity score matching was used to balance the comparison of PH and BJV conduits. RESULTS: The median follow-up was 6.4 years (interquartile range, IQR, 3.1-10.7 years). Freedom from reintervention at 5 and 10 years was 92% and 80%, respectively, for PH, 74% and 37% for BJV, and 75% and 47% for AH. BJV conduits had a higher risk of reintervention (P < .001) and SVD (P < .001) compared with PHs. These findings were confirmed with propensity score matching valid for conduit size >15 mm. AHs >15 mm had a higher risk of reintervention (P < .001) and SVD (P < .001) compared with PHs >15 mm. The performance of AHs and BJV conduits was similar across all sizes (reintervention, P = .94; SVD, P = .72). The incidence of IE was 1% for PH, 10% for BJV, and 1.5% for AH. CONCLUSIONS: In patients age <20 years with a conduit >15 mm, PHs outperformed BJV conduits and AHs in the pulmonary position. The performance of AH and BJV was comparable. Small conduits (≤15 mm) had similar performance across all conduit types.

A Simplified Approach to Predicting Reintervention in the Arterial Switch Operation.

We investigated patients with transposition anatomy suitable for the arterial switch operation (ASO) to evaluate a simplified approach to prediction of reintervention. A retrospective review was performed of 180 consecutive patients who underwent ASO from 2009 to 2018. Patients were classified as Category I (n = 122) d-transposition of great arteries (dTGA) + intact ventricular septum, Category II (n = 28) dTGA + ventricular septal defect (VSD) and Category III (n = 30) dTGA + Aortic arch obstruction (AAO) +/- VSD or Taussig-Bing Anomaly (TBA) +/- AAO. Outcomes included reintervention-free survival (using Kaplan-Meier estimates) and predictors of reintervention. Median follow up was 3.3 (interquartile range 1.7-5.8) years with no difference between categories(P = 0.082). There were 3 mortalities- 2 early (one each in Category I and II) and one late (in Category I). Reintervention-free survival for the whole group at 1, 3, 5 and 8 years was 94%, 91%, 90% and 86% respectively. Conventional criteria predicting reintervention included the presence of TBA(P = 0.0054) and AAO(P = 0.027). Low birth weight did not predict reintervention(P = 0.2). When analyzed by category, multivariable analysis showed that patients in Category III carried a high risk of reintervention [Hazard risk (HR) = 7.43, 95% confidence interval (CI)=(2.39, 23.11), P < 0.001], but so did those in Category II [HR=6.90, 95% CI = (2.19, 21.75, P < 0.001] when compared to Category I. Conventional risk factors for technical difficulty may not be the best predictors of reintervention. A simplified approach highlights Category II patients (dTGA + VSD) as being at substantial risk of re-intervention, and not part of a low risk cohort.

Prenatal cannabis use disorders and offspring primary and secondary educational outcomes.

BACKGROUND AND AIMS: Cannabis use is increasing among women of reproductive age, warranting a greater understanding of the impact of prenatal cannabis use on offspring developmental outcomes. We tested for a potential relationship between prenatal cannabis use disorders (CUD) and offspring educational outcomes across primary and secondary school. DESIGN: Data were drawn from the New South Wales (NSW) Perinatal Data Collection, which included all live births in the Australian state of NSW between January 2003 and December 2005. These were linked with the NSW Admitted Patient Data collection for mothers and offspring, and the NSW National Assessment Program-Literacy and Numeracy (NAPLAN). SETTING: New South Wales, Australia. PARTICIPANTS: A total of 189 558 offspring who completed the NAPLAN in grades 3, 5 and 7 (resulting in 568 674 examination periods). MEASUREMENTS: The exposure variable was ICD-10 cannabis use disorders (CUD = F13.0-F13.9). The study included five outcome variables measured at three time-points as not meeting the minimum national standards for: (i) numeracy, (ii) reading, (iii) spelling, (iv) writing and (v) grammar and punctuation. FINDINGS: In unadjusted analyses, prenatal CUD was associated with an increased risk for not meeting the national minimum standard of all outcomes [odds ratios (OR) ranging from 3.42 (95% confidence interval (CI) = 2.94, 3.99) to 4.17 (95% CI = 3.55, 4.91)], with no evidence for an interaction across time. However, the associations attenuated greatly after exact matching by covariates, with reading and numeracy no longer associated with prenatal CUD, while the increased risk of the other outcomes ranged from OR = 1.31 (95% CI = 1.09, 1.57) to OR = 1.40 (95% CI = 1.17, 1.68). CONCLUSION: Socio-economic status appears to confound the association between prenatal cannabis use disorder and poor educational performance in offspring.

Elective Caesarean Delivery Associated with Infant Hospitalisation for Intestinal But Not Respiratory Infection.

OBJECTIVES: We assessed the impact of caesarean delivery on offspring respiratory and intestinal infection. METHODS: Data were extracted from all live births (n = 429,058) occurring in the Australian state of Queensland between January 2009 and December 2015, and followed for 12 months. Births were categorised as either non-medically indicated caesarean or vaginal delivery and each offspring had a record (present/absent) of respiratory and intestinal infection hospitalisation for each month from birth to 12 months. RESULTS: Infants delivered by non-medically indicated caesarean were more likely to experience respiratory infection [OR = 1.51 (1.15, 1.99)] and intestinal infection [OR = 1.74 (1.19, 2.55)] than those born by vaginal delivery. In the propensity score weighted analyses the estimate for respiratory infection was similar but non-significant [OR = 1.52 (0.99, 2.31)], while the association with intestinal infection strengthened [OR = 2.21 (1.25, 3.89)]. CONCLUSIONS FOR PRATICE: Our findings provide strong evidence for a specific and clinically meaningful link between non-medically indicated caesarean delivery and infant intestinal infection.

Gender-Specific Drinking Contexts Are Associated With Social Harms Resulting From Drinking Among Australian Young Adults at 30 Years.

AIMS: Understanding contexts in which social harms from drinking occur can help develop context-based harm reduction efforts. However, there is little knowledge of specific drinking contexts where such harms occur and whether these are gender or age specific. We aimed to investigate associations of a range of drinking contexts and social harms from drinking among young adults at age 30. METHODS: We used data from 2187 30-year-old adults. Latent constructs of gender-specific drinking contexts were created using factor analysis. We performed multivariate logistic regression between drinking contexts and a range of social harms from drinking, separately by gender. RESULTS: After accounting for social roles and binge drinking, gender-specific solitary drinking contexts ('home drinking' for men and 'daytime drinking' for women) were positively associated with marital problems and problems with other family members and friends. Conversely, 'social drinking' was not. Work-related drinking among men was associated with marital/intimate relationship problems and friendship problems. After accounting for mental health symptoms, women's home drinking was associated with marital problems. CONCLUSION: We found that experiences of social harms from drinking at 30 years differ depending on the drinker's gender and context. Our findings suggest that risky contexts and associated harms are still significant among 30-year-old adults, indicating that a range of gender-specific drinking contexts should be represented in harm reduction campaigns. The current findings also highlight the need to consider gender to inform context-based harm reduction measures and to widen the age target for these beyond emerging adults.

Predicting neonatal respiratory distress syndrome and hypoglycaemia prior to discharge: Leveraging health administrative data and machine learning.

OBJECTIVES: A major challenge for hospitals and clinicians is the early identification of neonates at risk of developing adverse conditions. We develop a model based on routinely collected administrative data, which accurately predicts two common disorders among early term and preterm (<39 weeks) neonates prior to discharge. STUDY DESIGN: The data included all inpatient live births born prior to 39 weeks (n = 154,755) occurring in the Australian state of Queensland between January 2009 and December 2015. Predictor variables included all maternal data captured in administrative records from the beginning of gestation up to, and including, the delivery, as well as neonatal data recorded at the delivery. Gradient boosted trees were used to predict neonatal respiratory distress syndrome and hypoglycaemia prior to discharge, with model performance benchmarked against a logistic regression models. RESULTS: The gradient boosted trees model achieved very high discrimination for respiratory distress syndrome [AUC = 0.923, 95% CI (0.917, 0.928)] and good discrimination for hypoglycaemia [AUC = 0.832, 95% CI (0.827, 0.837)] in the validation data, as well as outperforming the logistic regression models. CONCLUSION: Our study suggests that routinely collected health data have the potential to play an important role in assisting clinicians to identify neonates at risk of developing selected disorders shortly after birth. Despite achieving high levels of discrimination, many issues remain before such models can be implemented in practice, which we discuss in relation to our findings.