Samantha J. Lain

Quality of data in perinatal population health databases: a systematic review.

BackgroundAdministrative or population health datasets (PHDS) are increasingly being used for research related to maternal and infant health. However, the accuracy and completeness of the information in the PHDS is important to ensure validity of the results of this research. ObjectiveTo compile and review studies that validate the reporting of conditions and procedures related to pregnancy, childbirth, and newborns and provide a tool of reference for researchers. MethodsA systematic search was conducted of Medline and EMBASE databases to find studies that validated routinely collected datasets containing diagnoses and procedures related to pregnancy, childbirth, and newborns. To be included datasets had to be validated against a gold standard, such as review of medical records, maternal interview or survey, specialized register, or laboratory data. Sensitivity, specificity, positive predictive value (PPV), negative predictive value (NPV), and/or &kgr; statistic for each diagnosis or procedure code were calculated. ResultsForty-three validation studies were included. Under-enumeration was common, with the level of ascertainment increasing as time from diagnosis/procedure to birth decreased. Most conditions and procedures had high specificities indicating few false positives, and procedures were more accurately reported than diagnoses. Hospital discharge data were generally more accurate than birth data, however identifying cases from more than 1 dataset further increased ascertainment. ConclusionsThis comprehensive collection of validation studies summarizing the quality of perinatal population data will be an invaluable resource to all researchers working with PHDS.

How accurate is the reporting of obstetric haemorrhage in hospital discharge data? A validation study.

Background: Routinely collected datasets are frequently used for population‐based research but their accuracy needs to be assured.

The prevalence of maternal medical conditions during pregnancy and a validation of their reporting in hospital discharge data.

Population health datasets are a valuable resource for studying maternal and obstetric health outcomes. However, their validity has not been thoroughly examined. We compared medical records from a random selection of New South Wales (NSW) women who gave birth in a NSW hospital in 2002 with coded hospital discharge records. We estimated the population prevalence of maternal medical conditions during pregnancy and found a tendency towards underreporting although specificities were high, indicating that false positives were uncommon.

Is there an association between endometriosis and the risk of pre-eclampsia? A population based study

BACKGROUND An association between endometriosis and reduced risk of pre-eclampsia has recently been reported. Longitudinally-linked electronic hospital records are a valuable resource for investigating such findings in a large, population-based sample. Our aim was to determine whether women with a history of endometriosis were at modified risk for pregnancy hypertension or pre-eclampsia. METHODS A population-based, longitudinal study of all women in the Australian state of New South Wales, aged from 15 to 45 years of age with a singleton birth during the period 2000-2005. Endometriosis was identified using ICD-10 codes. Endometriosis subgroups were analysed based on: (i) site of endometriosis (ovary or peritoneum), (ii) multiple (i.e. two or more) sites affected and (iii) infertility. To investigate the association between pregnancy hypertension and endometriosis, number of weeks gestation at birth and maternal age, we used logistic regression. RESULTS In the 3239 (1.6%) women with endometriosis diagnosed before their first birth, 352 (10.9%) had a diagnosis of pregnancy hypertension compared with 23,186/205,640 (11.3%) in women with no endometriosis diagnosis (OR 0.96; 95% CI 0.9-1.3). The frequency of pregnancy hypertension and pre-eclampsia was not significantly different in women with more severe endometriosis or endometriosis in conjunction with infertility when compared with those with no endometriosis. After adjusting for maternal age and weeks gestation there was still no altered risk. CONCLUSIONS We have found no evidence for an association between endometriosis and subsequent risk of either pregnancy hypertension or pre-eclampsia in this large population-based dataset.

Regional block versus general anaesthesia for caesarean section and neonatal outcomes: a population-based study

BackgroundAnaesthesia guidelines recommend regional anaesthesia for most caesarean sections due to the risk of failed intubation and aspiration with general anaesthesia. However, general anaesthesia is considered to be safe for the foetus, based on limited evidence, and is still used for caesarean sections.MethodsCohorts of caesarean sections by indication (that is, planned repeat caesarean section, failure to progress, foetal distress) were selected from the period 1998 to 2004 (N = 50,806). Deliveries performed under general anaesthesia were compared with those performed under spinal or epidural, for the outcomes of neonatal intubation and 5-minute Apgar (Apgar5) <7.ResultsThe risk of adverse outcomes was increased for caesarean sections under general anaesthesia for all three indications and across all levels of hospital. The relative risks were largest for low-risk planned repeat caesarean deliveries: resuscitation with intubation relative risk was 12.8 (95% confidence interval 7.6, 21.7), and Apgar5 <7 relative risk was 13.4 (95% confidence interval 9.2, 19.4). The largest absolute increase in risk was for unplanned caesareans due to foetal distress: there were five extra intubations per 100 deliveries and six extra Apgar5 <7 per 100 deliveries.ConclusionThe infants most affected by general anaesthesia were those already compromised in utero, as evidenced by foetal distress. The increased rate of adverse neonatal outcomes should be weighed up when general anaesthesia is under consideration.

Early Discharge of Infants and Risk of Readmission for Jaundice

OBJECTIVES: To examine the association between early discharge from hospital after birth and readmission to hospital for jaundice among term infants, and among infants discharged early, to investigate the perinatal risk factors for readmission for jaundice. METHODS: Birth data for 781 074 term live-born infants born in New South Wales, Australia from 2001 to 2010 were linked to hospital admission data. Logistic regression models were used to investigate the association between postnatal length of stay (LOS), gestational age (GA), and readmission for jaundice in the first 14 days of life. Other significant perinatal risk factors associated with readmission for jaundice were examined for infants discharged in the first 2 days after birth. RESULTS: Eight per 1000 term infants were readmitted for jaundice. Infants born at 37 weeks’ GA with an LOS at birth of 0 to 2 days were over 9 times (adjusted odds ratio [aOR] 9.43; 95% CI, 8.34–10.67) and at 38 weeks’ GA were 4 times (aOR 4.05; 95% CI, 3.62–4.54) more likely to be readmitted for jaundice compared with infants born at 39 weeks’ GA with an LOS of 3 to 4 days. Other significant risk factors for readmission for jaundice for infants discharged 0 to 2 days after birth included vaginal birth, born to mothers from an Asian country, born to first-time mothers, or being breastfed at discharge. CONCLUSIONS: This study can inform guidelines or policy about identifying infants at risk for readmission for jaundice and ensure that appropriate post-discharge follow-up is received.

Risk factors and costs of hospital admissions in first year of life: a population-based study.

OBJECTIVE To identify the maternal and infant risk factors associated with hospital admission in the first year and estimate the associated costs of infant hospitalization. STUDY DESIGN Data from the Perinatal Data Collection for 599753 liveborn infants born in New South Wales, Australia, 2001-2007 were linked to hospital admission data. Logistic regression models were used to investigate the association between maternal and infant characteristics and admission to hospital once, and more than once in the first year; and average costs for total hospital admissions were calculated. RESULTS Almost 15% of infants were admitted to hospital once and 4.6% had multiple admissions. Gestational age <37 weeks was most strongly associated with admission to hospital once, and severe neonatal morbidity was most strongly associated with multiple admissions (aOR 2.60; 95% CI 2.47-2.75). Infants born <39 weeks gestational age, to adolescent mothers, mothers who smoke, are not married, or had a planned delivery also have an increased risk of multiple admissions. Infants with severe neonatal morbidity contributed 27% of total infant hospital costs. With each increasing week of gestational age the mean annual cost decreased on average 10% and 27% for infants with and without neonatal morbidity respectively. CONCLUSIONS Infants born with severe neonatal morbidity have increased hospitalizations in the first year; however, the majority of burden on health system is by infants without severe neonatal morbidity. Hospitalizations, and associated costs, increased with decreasing gestational age, even for infants born at 37-38 weeks. Targeted public health strategies may reduce the burden of infant hospitalizations.

Record linkage to obtain birth outcomes for the evaluation of screening biomarkers in pregnancy: a feasibility study

BackgroundLinking population health data to pathology data is a new approach for the evaluation of predictive tests that is potentially more efficient, feasible and efficacious than current methods. Studies evaluating the use of first trimester maternal serum levels as predictors of complications in pregnancy have mostly relied on resource intensive methods such as prospective data collection or retrospective chart review. The aim of this pilot study is to demonstrate that record-linkage between a pathology database and routinely collected population health data sets provides follow-up on patient outcomes that is as effective as more traditional and resource-intensive methods. As a specific example, we evaluate maternal serum levels of PAPP-A and free β-hCG as predictors of adverse pregnancy outcomes, and compare our results with those of prospective studies.MethodsMaternal serum levels of PAPP-A and free β-hCG for 1882 women randomly selected from a pathology database in New South Wales (NSW) were linked to routinely collected birth and hospital databases. Crude relative risks were calculated to investigate the association between low levels (multiples of the median ≤ 5th percentile) of PAPP-A or free β-hCG and the outcomes of preterm delivery (<37 weeks), small for gestational age (<10th percentile), fetal loss and stillbirth.ResultsUsing only full name, sex and date of birth for record linkage, pregnancy outcomes were available for 1681 (89.3%) of women included in the study. Low levels of PAPP-A had a stronger association with adverse pregnancy outcomes than a low level of free β-hCG which is consistent with results in published studies. The relative risk of having a preterm birth with a low maternal serum PAPP-A level was 3.44 (95% CI 1.96–6.10) and a low free β-hCG level was 1.31 (95% CI 0.55–6.16).ConclusionThis study provides data to support the use of record linkage for outcome ascertainment in studies evaluating predictive tests. Linkage proportions are likely to increase if more personal identifiers are available. This method of follow-up is a cost-efficient technique and can now be applied to a larger cohort of women.

Variation in adherence to recommendations for management of the third stage of labor

Severe bleeding is a major cause of maternal morbidity and mortality worldwide, and postpartum hemorrhage (PPH) rates are reportedly increasing [1,2]. Active management of labor reduces PPH and widespread implementation could reduce the burden of disease associated with obstetric hemorrhage [3,4]. A joint statement from the International Confederation of Midwives and the International Federation of Gynecology and Obstetrics (ICM/FIGO) recommends that “active management of the third stage of labor should be offered to women since it reduces the incidence of PPH due to atony” [3]. Active management is defined in the ICM/ FIGO statement as: (1) administration of uterotonic agents (10 units of oxytocin intramuscularly is recommended); (2) controlled cord traction; and (3) uterine massage after delivery of the placenta as appropriate [3]. Notably, Prendiville et al. [4] include early cord clamping and cutting as part of their active management protocol, rather than uterine massage. However, a recent systematic review did not find that early cord clamping reduced the risk of PPH [5]. We report a population-level assessment of the management of the third stage of labor, methods of PPH reporting, and procedures used to control PPH as documented in the medical records. We used data collected in a statewide review of the delivery admissions of 1200 women in Australia in 2002. Data were abstracted by 3 clinicians experienced in obstetric chart review. The methods have been described elsewhere [6], but probability sampling with over-representation of severe hemorrhage allowed us to estimate population rates of the extreme measures used to control PPH. Table 1 indicates the significant practice variation in the documented use and timing of prophylactic oxytocics, controlled cord traction, blood loss reporting (especially by mode of delivery), and PPH rates depending on the criteria used. Early cord clamping was not reported in any record. Data were not collected on the use of uterine massage because local health policy defines active management according to the Cochrane Systematic Review [4], and recommends 5–10 units of oxytocin [7]. Among the women who had PPH (using blood loss or diagnostic criteria, Table 1), 6.9% had a blood transfusion and 2.6% had manual removal of the placenta following vaginal birth. Population estimates (per 1000 PPH cases) of extreme measures used to control bleeding were: 32.7 for dilatation ⁎ Corresponding author. Department of Obstetrics and Gynecology, Level 4, Wallace Freeborn Building (26), Royal North Shore Hospital, St Leonards, 2065, Australia. Tel.: +61 2 9926 7013; fax: +61 2 9906 6742. E-mail address: clroberts@med.usyd.edu.au (C.L. Roberts). ava i l ab l e a t www.sc i enced i r ec t . com

Association between borderline neonatal thyroid-stimulating hormone concentrations and educational and developmental outcomes: a population-based record-linkage study

BACKGROUND Congenital hypothyroidism causes intellectual delay unless identified and effectively treated soon after birth. Newborn screening has almost eliminated intellectual disability associated with congenital hypothyroidism. However, clinical uncertainty remains about infants with thyroid-stimulating hormone (TSH) concentrations less than the newborn screening cutoffs. We assessed the association between neonatal TSH concentrations and educational and developmental outcomes. METHODS We did a population-based record-linkage study of all liveborn infants undergoing newborn screening from 1994 to 2008 in New South Wales, Australia, with assessments of childhood development or school performance. Very-low-birthweight babies (<1500 g) were excluded. Developmental and educational outcomes were obtained and these were linked to individual records by the New South Wales Centre for Health Record Linkage. The primary educational outcome was the proportion of students with National Assessment Program Literacy and Numeracy (NAPLAN) results lower than the national minimum standard in reading or numeracy measured at all ages, and the primary developmental outcome was the proportion of children who were classified as being developmentally high risk (vulnerable in two or more of the five developmental domains assessed by the Australian Early Development Census) at age 4-6 years. The proportions of infants with each outcome were calculated per percentile (0-100) of TSH concentration. Multivariable logistic regression was used to account for potential confounding by maternal and fetal variables known to affect neonatal TSH concentrations or neurodevelopmental outcomes. FINDINGS 503 706 infants had a neonatal TSH result that linked to a developmental or educational outcome. 149 569 infants born between 2002 and 2008 were linked to an Australian Early Development Census developmental outcome and 354 137 were linked to a NAPLAN educational outcome. Median follow-up for educational outcome was 10 years (IQR 8-12) and for developmental outcome was 5 years (5-6). 5·5% (14 137 of 257 752) of infants scored less than the national minimum standard for numeracy in percentiles lower than the 75th percentile and this increased with each increase of percentile group to 11·3% (15 of 133) of infants with a TSH concentration between the 99·90th and 99·95th percentile. Infants with a neonatal TSH concentration in the 99·95th percentile or higher (above newborn screening cutoff) and likely to have diagnosed and treated congenital hypothyroidism had similar results to infants with a TSH concentration lower than the 75th percentile for both educational and developmental outcomes. Infants with a neonatal TSH concentration between the 99·5th and 99·9th percentile were more likely to have special needs (adjusted odds ratio [aOR] 1·68, 95% CI 1·23-2·30), poor numeracy performance (aOR 1·57, 1·29-1·90), and developmentally high risk (aOR 1·52, 1·20-1·93). INTERPRETATION We found an association between neonatal TSH concentrations lower than the present newborn screening thresholds and poor educational and developmental outcomes. This association needs further investigation to assess whether assessment and treatment of these infants might improve their long-term cognitive outcomes. FUNDING Australian National Health and Medical Research.