Implementation of Artificial Intelligence Applications in Australian Healthcare Organisations: Environmental Scan Findings.

Artificial Intelligence (AI) has great potential to improve healthcare, but implementation into routine practice remains a challenge. This study scoped the extent to which AI and Natural Language Processing (NLP) is being implemented into routine practice in Australian healthcare organisations. An environmental scan of publicly available data was undertaken to identify AI applications. Publicly available data consisted of news posts from Australian public healthcare organisations and conference proceedings from key research organisations. Two researchers reviewed and analysed posts related to AI applications to create a list of potential implementation case studies. The final list of AI applications was reviewed by a governance committee in order to identify any missing applications. One application was identified by the governance committee and subsequently added. The environmental scan identified eighteen AI applications, of which eleven met all eligibility criteria. Only one application included NLP. Twelve applications were included when the application identified by the governance committee was added to the list. Implementation of AI applications is spread across four broad categories of use: 1) Decision Support, 2) Monitoring Treatment Effectiveness, 3) Personalised Care and 4) Risk Prediction.

Seasonal variation in salivary cortisol but not symptoms of depression and trait anxiety in pregnant women undergoing an elective caesarean section.

OBJECTIVES: Seasonal changes in mood and behaviour are commonly reported in the general population but considerably less is known regarding seasonality and pregnancy. This study investigated the relationship between seasons and depression and anxiety symptoms, salivary cortisol concentrations, custom birthweight centiles (CBWC) and placenta weight for pregnant women living in South Wales. METHODS: This study utilised data from the longitudinal Grown in Wales (GiW) cohort. Women were recruited at the presurgical elective caesarean section (ELCS) appointment, when they provided saliva samples and completed the Edinburgh Postnatal Depression Scale (EPDS) and trait subscale of the State-Trait Anxiety Inventory (STAI). Data on birthweight and placental weight was extracted from medical notes. Seasonal data was available for 316 participants. RESULTS: No association was identified between seasons and EPDS (p = .178), STAI scores (p = .544), CBWC (p = .683) or placental weight (p = .857). Significance was identified between seasons and salivary cortisol concentration (p<.001), with highest levels in autumn and winter. Adjusted linear regression identified spring (B=-.05, p=.007, 95% CI -.09, -.01) and summer (B=-.06, p = .001, 95% CI -09, -.02) compared to autumn, and spring (B=-.05, p=.009, 95% CI -.09, -.01) and summer (B=-.06, p=.002, 95% CI -.10, -.02) compared to winter to be associated with decreased cortisol concentrations. CONCLUSION: This study found no association between season and maternally-reported mental health symptoms, birthweight by CBWC or placental weight but did between season and term salivary cortisol. This finding will have implications for studies that do not account for seasonality when using salivary cortisol as a biomarker.

The Grown in Wales Study: Examining dietary patterns, custom birthweight centiles and the risk of delivering a small-for-gestational age (SGA) infant.

OBJECTIVES: Maternal lifestyles, including diet, have been linked to infant birthweight. However, customised birthweight centiles (CBWC), which more accurately identify small babies that have increased fetal growth restriction and are at higher risk of newborn morbidity and later life health complications, are rarely considered when studying maternal diet. This study investigated maternal dietary patterns and their impact on infant CBWC within a cohort of women living in South Wales. METHODS: This study utilised cross-sectional data from the longitudinal Grown in Wales (GiW) cohort. Women 18-45 years old were recruited the morning prior to an elective caesarean section (ELCS). Women completed a food frequency questionnaire (FFQ). Additional data on pregnancy and birth outcomes was extracted from medical notes. Data from 303 participants was analysed. RESULTS: 'Western' and 'Health conscious dietary patterns were identified. The 'Health Conscious' dietary pattern was significantly associated with maternal BMI, age, education, income and exercise. Adjusted regression analyses indicated that greater adherence to a 'Health Conscious' dietary pattern was significantly associated with increased CBWC (AOR = 4.75 [95% CI: 1.17, 8.33] p = .010) and reduced risk of delivering a small-for-gestational age (SGA) infant (AOR = .51 [95% CI: .26, .99] p = .046). CONCLUSION: A healthier diet was significantly associated with higher birthweight using CBWC and a reduced risk of delivering an SGA infant suggesting that birthweight will be improved in areas of Wales by focused support encouraging healthier dietary habits.

Metformin and insulin treatment prevent placental telomere attrition in boys exposed to maternal diabetes.

Shortened leukocyte and placental telomeres associated with gestational diabetes mellitus (GDM) suggest this exposure triggers telomere attrition contributing to adverse outcomes. We applied high resolution Single Telomere Length Analysis (STELA) to placenta from GDM pregnancies with different treatment pathways to determine their effectiveness at preventing telomere attrition. Differences in telomere length between control (N = 69), GDM lifestyle intervention (n = 14) and GDM treated with metformin and/or insulin (n = 17) was tested by Analysis of Covariance (ANCOVA) followed by group comparisons using Fisher's least significant difference. For male placenta only, there were differences in mean telomere length (F(2,54) = 4.98, P = 0.01) and percentage of telomeres under 5 kb (F(2,54) = 4.65, P = 0.01). Telomeres were shorter in the GDM lifestyle intervention group compared to both controls (P = 0.02) and medically treated pregnancies (P = 0.003). There were more telomeres under 5 kb in the GDM lifestyle intervention group compared to the other two groups (P = 0.03 and P = 0.004). Although further work is necessary, we suggest that early adoption of targeted medical treatment of GDM pregnancies where the fetus is known to be male may be an effective strategy for ameliorating adverse outcomes for children.

Persistence of anxiety symptoms after elective caesarean delivery.

BACKGROUND: In the UK, 11.8% of expectant mothers undergo an elective caesarean section (ELCS) representing 92 000 births per annum. It is not known to what extent this procedure has an impact on mental well-being in the longer term. AIMS: To determine the prevalence and postpartum progression of anxiety and depression symptoms in women undergoing ELCS in Wales. METHOD: Prevalence of depression and anxiety were determined in women at University Hospital Wales (2015-16; n = 308) through completion of the Edinburgh Postnatal Depression Scale (EPDS; ≥13) and State-Trait Anxiety Inventory (STAI; ≥40) questionnaires 1 day prior to ELCS, and three postpartum time points for 1 year. Maternal characteristics were determined from questionnaires and, where possible, confirmed from National Health Service maternity records. RESULTS: Using these criteria the prevalence of reported depression symptoms was 14.3% (95% CI 10.9-18.3) 1 day prior to ELCS, 8.0% (95% CI 4.2-12.5) within 1 week, 8.7% (95% CI 4.2-13.8) at 10 weeks and 12.4% (95% CI 6.4-18.4) 1 year postpartum. Prevalence of reported anxiety symptoms was 27.3% (95% CI 22.5-32.4), 21.7% (95% CI 15.8-28.0), 25.3% (95% CI 18.5-32.7) and 35.1% (95% CI 26.3-44.2) at these same stages. Prenatal anxiety was not resolved after ELCS more than 1 year after delivery. CONCLUSIONS: Women undergoing ELCS experience prolonged anxiety postpartum that merits focused clinical attention. DECLARATION OF INTEREST: None.