Investigating linkage rates among probabilistically linked birth and hospitalization records

Development and Validation of txSim: A Model of Advanced Lung Cancer Treatment in Australia

BACKGROUND AND OBJECTIVE

Since 2016, new therapies have transformed the standard of care for lung cancer, creating a need for up-to-date evidence for health economic modelling. We developed a discrete event simulation of advanced lung cancer treatment to provide estimates of survival outcomes and healthcare costs in the Australian setting that can be updated as new therapies are introduced.

METHODS

Treatment for advanced lung cancer was modelled under a clinician-specified treatment algorithm for Australia in 2022. Prevalence of lung cancer subpopulations was extracted from cBioPortal and the Sax Institute's 45 and Up Study, a large prospective cohort linked to cancer registrations. All costs were from the health system perspective for the year 2020. Pharmaceutical and molecular diagnostic costs were obtained from public reimbursement fees, while other healthcare costs were obtained from health system costs in the 45 and Up Study. Treatment efficacy was obtained from clinical trials and observational study data. Costs and survival were modelled over a 10-year horizon. Uncertainty intervals were generated with probabilistic sensitivity analyses. Overall survival predictions were validated against real-world studies.

RESULTS

Under the 2022 treatment algorithm, estimated mean survival and costs for advanced lung cancer 10 years post-diagnosis were 16.4 months (95% uncertainty interval [UI]: 14.7-18.1) and AU$116,069 (95% UI: $107,378-$124,933). Survival and costs were higher assuming optimal treatment utilisation rates (20.5 months, 95% UI: 19.1-22.5; $154,299, 95% UI: $146,499-$161,591). The model performed well in validation, with good agreement between predicted and observed survival in real-world studies.

CONCLUSIONS

Survival improvements for advanced lung cancer have been accompanied by growing treatment costs. The estimates reported here can be used for budget planning and economic evaluations of interventions across the spectrum of cancer control.

National data linkage assessment of live births and deaths in Mexico: Estimating under-five mortality rate ratios for vulnerable newborns and trends from 2008 to 2019

BACKGROUND

Linked datasets that enable longitudinal assessments are scarce in low and middle-income countries.

OBJECTIVES

We aimed to assess the linkage of administrative databases of live births and under-five child deaths to explore mortality and trends for preterm, small (SGA) and large for gestational age (LGA) in Mexico.

METHODS

We linked individual-level datasets collected by National statistics from 2008 to 2019. Linkage was performed based on agreement on birthday, sex, residential address. We used the Centre for Data and Knowledge Integration for Health software to identify the best candidate pairs based on similarity. Accuracy was assessed by calculating the area under the receiver operating characteristic curve. We evaluated completeness by comparing the number of linked records with reported deaths. We described the percentage of linked records by baseline characteristics to identify potential bias. Using the linked dataset, we calculated mortality rate ratios (RR) in neonatal, infants, and children under-five according to gestational age, birthweight, and size.

RESULTS

For the period 2008-2019, a total of 24,955,172 live births and 321,165 under-five deaths were available for linkage. We excluded 1,539,046 records (6.2%) with missing or implausible values. We succesfully linked 231,765 deaths (72.2%: range 57.1% in 2009 and 84.3% in 2011). The rate of neonatal mortality was higher for preterm compared with term (RR 3.83, 95% confidence interval, CI 3.78, 3.88) and for SGA compared with appropriate for gestational age (AGA) (RR 1.22 95% CI, 1.19, 1.24). Births at <28 weeks had the highest mortality (RR 35.92, 95%CI, 34.97, 36.88). LGA had no additional risk vs AGA among children under five (RR 0.92, 95%CI, 0.90, 0.93).

CONCLUSIONS

We demonstrated the utility of linked data to understand neonatal vulnerability and child mortality. We created a linked dataset that would be a valuable resource for future population-based research.

Cutaneous melanoma, prostate-specific antigen testing and the subsequent risk of prostate cancer diagnosis: a prospective analysis of the 45 and Up Study

BACKGROUND

The association between cutaneous melanoma and subsequent risk of prostate cancer (PC) was examined in a large population-based cohort study.

METHODS

Male participants in the Sax Institute's 45 and Up Study (Australia) were recruited between 2006 and 2009. Questionnaire data and linked administrative health data from the Centre for Health Record Linkage and Services Australia identified melanomas diagnosed between 1/1/1994 and 12 months before Study recruitment (i.e., between 2005 and 2008), incident PCs, primary healthcare utilisation and prostate-specific antigen (PSA) tests. Men were excluded from the current analyses if they had a recorded PC or other cancer diagnosis other than melanoma and non-melanoma skin cancer prior to recruitment. Multivariable Cox regression was used to estimate hazard ratios (HRs) adjusting for PSA-testing frequency before PC diagnosis.

RESULTS

Of 96,548 eligible men, 1899 were diagnosed with melanoma during the melanoma diagnosis period and 3677 incident PC diagnosed during follow-up (latest date 31/12/2013). Men with melanoma diagnosis had increased risk of a subsequent PC diagnoses (vs. no melanoma; fully adjusted HR = 1.32; 95% CI: 1.09-1.60). There was weak evidence of higher risks of a subsequent PC diagnosis for men diagnosed with more than one melanoma compared to men diagnosed with only one melanoma (p = 0.077), and if first melanoma diagnosis was 10 to 15 years before Study recruitment (fully adjusted HR = 2.05; 95% CI [1.35, 3.12]).

CONCLUSION

Melanoma diagnosis was associated with increased risk of subsequent PC diagnosis, after adjusting for PSA testing and primary healthcare utilisation. While our ability to adjust for PC screening reduced risk of detection bias, we acknowledge that residual confounding from increased medical surveillance after melanoma diagnoses cannot be entirely ruled out.

Large-Scale Population-Based Surveys Linked to Administrative Health Databases as a Source of Data on Health Utilities in Australia

OBJECTIVES

Large-scale health surveys that contain quality-of-life instruments are a rich source of health utility data for health economic evaluations, especially when linked to routinely collected, administrative health databases. We derived health utility values for a wide range of health conditions using a large Australian cohort study linked to population-wide health databases.

METHODS

Short-Form 6-Dimension utility values were calculated for 56 094 adults, aged 47+ years, in the New South Wales 45 and Up Study who completed the Social, Economic, and Environmental Factors survey (2010-2011). Mean utilities were summarized for major health conditions identified through self-report, hospital records, primary cancer notifications, and claims for government-subsidized prescription medicines and medical services. To identify unique associations between health conditions and utilities, beta regression was performed. Utility values were analyzed by time to death using linked death records.

RESULTS

Mean Short-Form 6-Dimension utility was 0.810 (95% confidence interval [CI] 0.809-0.811), was age dependent, and was higher in men than women. Utilities for serious health conditions ranged from 0.685 (95% CI 0.652-0.718) for lung cancer to 0.800 (95% CI 0.787-0.812) for melanoma whereas disease-free respondents had a mean of 0.859 (95% CI 0.858-0.861). Most health conditions were independently associated with poorer quality of life. Utility values also declined by proximity to death where participants sampled 6 months before death had a mean score of 0.637 (95% CI 0.613-0.662).

CONCLUSIONS

Our data offer a snapshot of the health status of an older Australian population and show that record linkage can enable comprehensive ascertainment of utility values for use in health economic modeling.

Pregnancy outcomes in women with endometriosis and/or ART use: a population-based cohort study

STUDY QUESTION

What is the association between endometriosis and adverse pregnancy outcomes with ART use and non-use?

SUMMARY ANSWER

Endometriosis and ART use are both associated with increased risk of preterm birth, antepartum haemorrhage, placenta praevia and planned birth (caesarean delivery or induction of labour).

WHAT IS KNOWN ALREADY

There are contradictory findings on the association between endometriosis and adverse pregnancy outcomes, and many large studies have not considered the effect of ART use.

STUDY DESIGN, SIZE, DURATION

Population-based cohort study of 578 221 eligible pregnancies during 2006-2015, comparing pregnancy outcomes across four groups (No endo/no ART, No endo/ART, Endo/no ART and Endo/ART).

PARTICIPANTS/MATERIALS, SETTING, METHODS

All female residents of New South Wales, Australia aged 15-45 years and their index singleton pregnancy of at least 20 weeks gestation or 400 g birthweight. Linked hospital, pregnancy/birth and mortality data were used. Modified Poisson regression with robust error variances was used to estimate adjusted risk ratios (aRRs) and 99% CIs, adjusting for sociodemographic and pregnancy factors.

MAIN RESULTS AND THE ROLE OF CHANCE

Compared to women without endometriosis who had pregnancies without ART use, there was increased risk of preterm birth (<37 weeks) in all groups [No endo/ART (aRR 1.85, 99% CI 1.46-2.34), Endo/no ART (aRR 1.24, 99% CI 1.06-1.44), Endo/ART (aRR 1.93, 99% CI 1.11-3.35)] and antepartum haemorrhage [No endo/ART (aRR 1.99, 99% CI 1.39-2.85), Endo/no ART (aRR 1.31, 99% CI 1.03-1.67), Endo/ART (aRR 2.69, 99% CI 1.30-5.56)] among pregnancies affected by endometriosis or ART use, separately and together. There was increased risk of placenta praevia [No endo/ART (aRR 2.26, 99% CI 1.42-3.60), Endo/no ART (aRR 1.66, 99% CI 1.18-2.33)] and planned birth [No endo/ART (aRR 1.08, 99% CI 1.03-1.14), Endo/no ART (aRR 1.11, 99% CI 1.07-1.14)] among pregnancies with endometriosis or ART use, separately. There was increased risk of placental abruption [No endo/ART (aRR 2.36, 99% CI 1.12-4.98)], maternal morbidity [No endo/ART (aRR 1.67, 99% CI 1.07-2.62)] and low birthweight (<2500 g) [No endo/ART (aRR 1.45, 99% CI 1.09-1.93)] among pregnancies with ART use without endometriosis. There was decreased risk of having a large-for-gestational age infant [Endo/no ART (aRR 0.83, 99% CI 0.73-0.94)] among pregnancies with endometriosis without ART use.

LIMITATIONS, REASONS FOR CAUTION

Endometriosis is often under-diagnosed and women with a history of hospital diagnosis of endometriosis may represent those with more symptomatic or severe disease. If the effects of endometriosis on pregnancy are greater for those with more severe disease, our results may over-estimate the effect of endometriosis on adverse pregnancy outcomes at a population level. We were unable to assess the effect of endometriosis stage or typology on the study outcomes.

WIDER IMPLICATIONS OF THE FINDINGS

These results suggest that women with endometriosis including those who used ART to achieve pregnancy are a higher-risk obstetric group requiring appropriate surveillance and management during their pregnancy.

STUDY FUNDING/COMPETING INTEREST(S)

This study was supported by the Prevention Research Support Program, funded by the New South Wales Ministry of Health. The funder had no role in the design, data collection and analysis, interpretation of results, manuscript preparation or the decision to submit the manuscript for publication. The authors have no conflicts of interest.

TRIAL REGISTRATION NUMBER

N/A.

Lung cancer treatment patterns and factors relating to systemic therapy use in Australia

AIM

Systemic therapies for lung cancer are rapidly evolving. This study aimed to describe lung cancer treatment patterns in New South Wales, Australia, prior to the introduction of immunotherapy and latest-generation targeted therapies.

METHODS

Systemic therapy utilization and treatment-related factors were examined for participants in the New South Wales 45 and Up Study with incident lung cancer ascertained by record linkage to the New South Wales Cancer Registry (2006-2013). Systemic therapy receipt to June 2016 was determined using medical and pharmaceutical claims data from Services Australia, and in-patient hospital records. Factors related to treatment were identified using competing risks regressions.

RESULTS

A total of 1,116 lung cancer cases were identified with a mean age at diagnosis of 72 years and median survival of 10.6 months. Systemic therapy was received by 45% of cases. Among 400 cases with metastatic non-small cell lung cancer, 51% and 28% received first- and second-line systemic therapy, respectively. Among 112 diagnosed with small-cell lung cancer, 79% and 29% received first- and second-line systemic therapy. The incidence of systemic therapy was lower for participants with indicators of poor performance status, lower educational attainment, and those who lived in areas of socioeconomic disadvantage; and was higher for participants with small-cell lung cancer histology or higher body mass index.

CONCLUSION

This population-based Australian study identified patterns of systemic therapy use for lung cancer, particularly small-cell lung cancer. Despite a universal healthcare system, the analysis revealed socioeconomic disparities in health service utilization and relatively low utilization of systemic therapy overall.

Characteristics Associated with the Use of Diagnostic Prostate Biopsy and Biopsy Outcomes in Australian Men

BACKGROUND

Population characteristics associated with the use of prostate biopsy are poorly understood. We described the use of diagnostic prostate biopsy and subsequent biopsy outcomes in a population-based Australian cohort.

METHODS

A total of 91,764 men from the Sax Institute's 45 and Up Study (New South Wales, Australia) recruited during 2006 to 2009 were included. Self-completed baseline questionnaires and linked administrative health data were used. Study period was from the date of recruitment to December 2013. Cox regression and logistic regression identified factors associated with receipt of biopsy and subsequent prostate cancer diagnosis.

RESULTS

During the study period, 5,089 participants had a diagnostic prostate biopsy, and 2,805 men (55.1% of those biopsied) received a cancer diagnosis. Men with a family history of prostate cancer (HR 1.55; 95% confidence interval (CI), 1.43-1.68), severe lower urinary tract symptoms (HR 1.62; 95% CI, 1.41-1.86), or a record of medication for benign prostatic hyperplasia (HR 1.34; 95% CI, 1.23-1.47) had increased risks of receiving a biopsy. Men with a family history of prostate cancer had increased odds of a positive biopsy (OR 1.21; 95% CI, 1.01-1.43). High alcohol consumption (≥21 drinks per week compared with 1-6 drinks per week) was associated with decreased risk of biopsy (HR 0.88; 95% CI, 0.80-0.96) but increased odds of a positive biopsy (OR 1.63; 95% CI, 1.32-2.02).

CONCLUSIONS

Certain characteristics are associated with both undertaking diagnostic prostate biopsy and positive biopsy outcomes.

IMPACT

This highlights the need to improve management of specific groups of men, especially those with clinical symptoms that overlap with prostate cancer, in their investigation for prostate cancer.

Cancer incidence and cancer death in relation to tobacco smoking in a population-based Australian cohort study

Tobacco smoke is a known carcinogen, but the magnitude of smoking-related cancer risk depends on country-specific, generational smoking patterns. We quantified cancer risk in relation to smoking in a population-based cohort, the 45 and Up Study (2006-2009) in New South Wales, Australia. Cox proportional hazards regressions estimated adjusted hazard ratios (HR) by self-reported smoking history at baseline (2006-2009) for incident, primary cancers via linkage to cancer registry data to 2013 and cancer death data to 2015. Among 229 028 participants aged ≥45 years, 18 475 cancers and 5382 cancer deaths occurred. Current-smokers had increased risks of all cancers combined (HR = 1.42, 95% confidence interval [CI], 1.34-1.51), cancers of the lung (HR = 17.66, 95%CI, 14.65-21.29), larynx (HR = 11.29, 95%CI, 5.49-23.20), head-and-neck (HR = 2.53, 95%CI, 1.87-3.41), oesophagus (HR = 3.84, 95%CI, 2.33-6.35), liver (HR = 4.07, 95%CI, 2.55-6.51), bladder (HR = 3.08, 95%CI, 2.00-4.73), pancreas (HR = 2.68, 95%CI, 1.93-3.71), colorectum (HR = 1.31, 95%CI, 1.09-1.57) and unknown primary site (HR = 3.26, 95%CI, 2.19-4.84) versus never-smokers. Hazards increased with increasing smoking intensity; compared to never-smokers, lung cancer HR = 9.22 (95%CI, 5.14-16.55) for 1-5 cigarettes/day and 38.61 (95%CI, 25.65-58.13) for >35 cigarettes/day. Lung cancer risk was lower with quitting at any age but remained higher than never-smokers for quitters aged >25y. By age 80y, an estimated 48.3% of current-smokers (41.1% never-smokers) will develop cancer, and 14% will develop lung cancer, including 7.7% currently smoking 1-5 cigarettes/day and 26.4% for >35 cigarettes/day (1.0% never-smokers). Cancer risk for Australian smokers is significant, even for 'light' smokers. These contemporary estimates underpin the need for continued investment in strategies to prevent smoking uptake and facilitate cessation, which remain key to reducing cancer morbidity and mortality worldwide.

Outcomes of Subsequent Births After Placenta Accreta Spectrum

OBJECTIVE

To investigate subsequent birth rates, maternal and neonatal outcomes for women with a history of placenta accreta spectrum (placenta accreta, increta, and percreta).

METHODS

A population-based record linkage study of women who had a first, second, or third birth in New South Wales from 2003 to 2016 was conducted. Data were obtained from birth and hospital records and death registrations. Women with a history of placenta accreta spectrum were matched to women without, on propensity score and parity, to compare outcomes with women who had similar risk profiles. Modified Poisson regression models were used to calculate adjusted relative risk (aRR) for a range of maternal and neonatal outcomes.

RESULTS

We identified recurrent placenta accreta spectrum in 27/570 (4.7%, 95% CI 3.0-6.5%) of second and 9/119 (7.6%, 95% CI 2.8-12.3%) of third pregnancies after placenta accreta spectrum in the preceding birth, with an overall recurrence rate of 38/689 (5.5%, 95% CI 3.9-7.5%, compared with the population prevalence of 25.5/10,000 births (95% CI 24.6-26.4). Subsequent births after placenta accreta spectrum had higher risk of postpartum hemorrhage (aRR 1.51, 95% CI 1.19-1.92), transfusion (aRR 2.13, 95% CI 1.17-3.90), cesarean delivery (aRR 1.19, 95% CI 1.02-1.37), manual removal of placenta (aRR 6.92, 95% CI 3.81-12.55), and preterm birth (aRR 1.43, 95% CI 1.03-1.98), with lower risk of small for gestational age (aRR 0.64, 95% CI 0.43-0.96), compared with similar-risk births.

CONCLUSION

Women with a history of placenta accreta spectrum have increased risk of maternal morbidity, preterm birth, and placenta accreta spectrum in the subsequent pregnancy compared with similar-risk women with no previous placenta accreta spectrum, although the absolute risks are generally low. These findings may be used to inform counseling of women on the risks of future pregnancies.

Pregnancy outcomes for women with a history of stroke: A population-based record linkage study

BACKGROUND

Little is known about the pregnancy outcomes of women who have had a stroke prior to a first pregnancy.

AIM

To identify a cohort of primiparous women giving birth to a single baby and compare the pregnancy outcomes of those with a pre-pregnancy stroke hospitalisation record to those without a stroke hospitalisation record.

MATERIALS AND METHODS

Record linkage study of all primiparous women aged 15-44 years with singleton pregnancies birthing in New South Wales, Australia from 2003 to 2015. Stroke was identified from 2001 to 2015 hospital data using International Classification of Diseases tenth Edition - Australian Modification codes I60-64. Women whose first hospital record of stroke was during pregnancy or <42 days after birth were excluded. Outcomes included diabetes or hypertension during pregnancy, mode of delivery, haemorrhage, severe maternal morbidity (validated composite outcome indicator), gestational age at birth, Apgar score (1 min < 7), and small-for-gestational age.

RESULTS

Of 487 767 women with a first pregnancy, 124 (2.5/10 000) had a hospital record which included a pre-pregnancy stroke diagnosis. Women with a stroke history were more likely to have an early-term delivery (37-38 weeks; relative risk (RR) 1.49, 95% CI 1.17-1.90) and a pre-labour caesarean (RR 2.83, 95% CI 2.20-3.63). There were no significant differences in other maternal or neonatal outcomes.

CONCLUSION

This is the largest reported study of pregnancy and birth outcomes for women with a history of stroke. With the exception of pre-labour caesarean, there were no differences in pregnancy outcomes for women with a history of stroke compared with women with no history of stroke.

A study of pregnancy after endometrial ablation using linked population data

INTRODUCTION

Endometrial ablation encapsulates a range of procedures undertaken to destroy the endometrial lining of the uterus as a treatment for heavy menstrual bleeding in women who no longer wish to bear children. Pregnancy following ablation, while unlikely, can occur and may carry higher rates of complications. The aim of this study was to identify factors associated with post-endometrial ablation pregnancy and to describe pregnancy and birth outcomes for post-endometrial ablation pregnancies.

MATERIAL AND METHODS

This population-based data linkage study included all female residents of New South Wales, Australia, aged 15-50 years with a hospital admission between July 2001 to June 2014 who birthed between July 2001 and June 2015. Cox proportional hazard regression was used to estimate associations between women's characteristics and post-endometrial ablation pregnancy of at least 20 weeks' gestation. Descriptive statistics were used to characterize pregnancy and birth outcomes.

RESULTS

Of 18 559 women with an endometrial ablation, 575 (3.1%) had a post-ablation pregnancy of at least 20 weeks' gestation. Nulliparity (adjusted hazard ratio [aHR] 12.2, 95% confidence interval [CI] 9.1-16.2), older age (35-39 years: aHR 0.39, 95% CI 0.29-0.51; 40-44 years: aHR 0.06, 95% CI 0.04-0.11), marital status (single: aHR 0.67, 95% CI 0.55-0.83; widowed/divorced/separated: aHR 0.58, 95% CI 0.36-0.94) and a diagnosis of heavy menstrual bleeding (aHR 0.09, 95% CI 0.07-0.13) were associated with post-ablation pregnancy. There were high rates of cesarean delivery (43%), preterm birth (13%), twin or higher order pregnancies (9%) and stillbirth (13.3/1000 births) among these post-ablation pregnancies.

CONCLUSIONS

Nulliparity at the time of endometrial ablation is associated with increased risk of post-ablation pregnancy, highlighting the importance of careful discussion and consideration of treatment options for heavy menstrual bleeding.

Alcohol consumption, drinking patterns and cancer incidence in an Australian cohort of 226,162 participants aged 45 years and over

BACKGROUND

Although overall alcohol consumption is known to increase the risk of a number of cancers internationally, evidence for Australia and evidence regarding the pattern of drinking and cancer risk is limited.

METHODS

Adjusted hazard ratios (HR) and 95% confidence intervals (CI) for cancer risk in relation to overall alcohol consumption (drinks/week) and pattern of drinking were calculated using Cox proportional hazard regressions for 226,162 participants aged ≥45 years (2006-2009) in the 45 and Up Study, an Australian prospective cohort study. Incident primary cancer cases were ascertained by linkage to the New South Wales Cancer Registry to 2013 by the Centre for Health Record Linkage.

RESULTS

Over a median of 5.4 years, 17,332 cancers were diagnosed. Increasing levels of alcohol intake were associated with increased risk of cancers of the upper aerodigestive tract (1.19; 1.10-1.29), mouth and pharynx (1.18; 1.08-1.29), oesophagus (1.22; 1.04-1.43), colorectum (1.09; 1.04-1.15), colon (1.13; 1.06-1.20), liver (1.22; 1.04-1.44) and breast (1.11; 1.02-1.21). Breast cancer risk was marginally associated with drinking pattern, with higher risk when intake was concentrated on 1-3 days/week compared to the same amount spread over 4-7 days (P_interaction = 0.049).

CONCLUSIONS

Alcohol consumption confers a significant risk of cancer, and drinking pattern may be independently related to breast cancer risk.

Factors Associated With Radiotherapy Utilisation In New South Wales, Australia: Results From The 45 and Up Study

AIMS

Actual radiotherapy utilisation rates tend to be lower than the estimated optimal rates. Little is known about the factors contributing to this difference. Our aim was to identify factors associated with radiotherapy receipt for a cohort of cancer patients in New South Wales (NSW), Australia.

MATERIALS AND METHODS

In total, 267 153 participants in the NSW 45 and Up Study completed a questionnaire during 2006-2009 providing detailed health and socio-demographic information and consented to record linkage with administrative health datasets. Single primary cancers diagnosed after study enrolment were identified through linkage with the NSW Cancer Registry to December 2013. Radiotherapy receipt was determined from claims to the Medicare Benefits Schedule and/or records in the NSW Admitted Patient Data Collection (2006 to June 2016). Competing risks regression was used to examine associations between health and socio-demographic characteristics and radiotherapy treatment.

RESULTS

Of 17 873 patients with an incident cancer, 5414 (30.3%) received radiotherapy during follow-up (median 5.3 years). Patients less likely to receive radiotherapy were aged <60 or 80+ years, female, had a Charlson co-morbidity index of 1+, needed help with daily tasks or lived ≥100 km from the nearest radiotherapy centre.

CONCLUSION

Distinct subgroups of patients are less likely to receive radiotherapy. Advocacy and/or policy changes are needed to improve access.

Gestational age, morbidity and mortality among twin births in New South Wales, Australia 2003-2014: A cohort study

BACKGROUND

Evidence suggests that the trend toward early planned births observed among singletons may be evident among twin pregnancies.

AIMS

To describe trends in gestational age at birth, pregnancy characteristics, neonatal morbidity and mortality among twin pregnancies.

MATERIALS AND METHODS

Population-based data linkage study of twin births of ≥30 weeks of gestation without a major congenital anomaly born in 2003-2014 in New South Wales (NSW), Australia. Linked pregnancy and birth, hospital and mortality data were used. Generalised linear regression was used to assess linear trends. Risk difference (RD) and 95% confidence intervals were estimated.

RESULTS

Among 28 076 eligible twin births (14 038 pregnancies), 49% of births occurred prior to 37 weeks and 69% of births were planned (pre-labour caesarean or induction of labour). There were increases over time in the proportion of twin births at preterm gestations (30-34 weeks (RD 2.1, 95% CI 0.1, 4.0), 35-36 weeks (RD 7.5, 95% CI 5.4, 9.7)) and in the rates of planned births (pre-labour caesarean (RD 6.4, 95% CI 4.0, 8.8), induction (RD 4.6, 95% CI 2.6, 6.6)). There was no significant change in stillbirth or neonatal death rates, but there was an increase in neonatal morbidity over the study period. Concurrently, there were increases in the prevalence of gestational diabetes; and decreases in pregnancy hypertension, assisted reproductive technology use, small-for-gestational age and birthweight discordance.

CONCLUSIONS

Gestational age at birth among twin births is decreasing and birth intervention is increasing. There are increasing rates of neonatal morbidity, but no overall change in perinatal mortality.

A profile of the Centre for Health Record Linkage

Context

The Centre for Health Record Linkage (CHeReL) was established in 2006 as a dedicated health and human services data linkage facility for two Australian jurisdictions, New South Wales and the geographically-nested Australian Capital Territory. The two jurisdictions have their own Governments and separate Health and Human Service systems.

Purpose and Operations

The primary purpose of the CHeReL is to make linked administrative and routinely collected health data available to researchers and government within relevant regulatory and governance frameworks. The CHeReL's data governance and technical operations draw on international best practice and have been refined by learnings from other data linkage centres.

Outcomes

Over twelve years of operation, more than 2,320 unique investigators from 140 institutions have used the CHeReL, producing 615 publications in peer-reviewed literature. A robust pipeline of new development is expected to further amplify the use of linked data for cutting edge medical research and support a vision of data-informed policy and data-driven government services.

A profile of the Centre for Health Record Linkage

CONTEXT

The Centre for Health Record Linkage (CHeReL) was established in 2006 as a dedicated health and human services data linkage facility for two Australian jurisdictions, New South Wales and the geographically-nested Australian Capital Territory. The two jurisdictions have their own Governments and separate Health and Human Service systems.

PURPOSE AND OPERATIONS

The primary purpose of the CHeReL is to make linked administrative and routinely collected health data available to researchers and government within relevant regulatory and governance frameworks. The CHeReL's data governance and technical operations draw on international best practice and have been refined by learnings from other data linkage centres.

OUTCOMES

Over twelve years of operation, more than 2,320 unique investigators from 140 institutions have used the CHeReL, producing 615 publications in peer-reviewed literature. A robust pipeline of new development is expected to further amplify the use of linked data for cutting edge medical research and support a vision of data-informed policy and data-driven government services.

Benefits of not smoking during pregnancy for Australian Aboriginal and Torres Strait Islander women and their babies: a retrospective cohort study using linked data

OBJECTIVES

To provide evidence for targeted smoking cessation policy, the aim of this study was to compare pregnancy outcomes of Aboriginal mothers who reported not smoking during pregnancy with Aboriginal mothers who reported smoking during pregnancy.

DESIGN

Population based retrospective cohort study using linked data.

SETTING

New South Wales, the most populous Australian state.

POPULATION

18 154 singleton babies born to 13 477 Aboriginal mothers between 2010 and 2014 were identified from routinely collected New South Wales datasets. Aboriginality was determined from birth records and from four linked datasets through an Enhanced Reporting of Aboriginality algorithm.

EXPOSURE

Not smoking at any time during pregnancy.

MAIN OUTCOME MEASURES

Unadjusted and adjusted relative risks (aRR) and 95% CIs from modified Poisson regression were used to examine associations between not smoking during pregnancy and maternal and perinatal outcomes including severe morbidity, inter-hospital transfer, perinatal death, preterm birth and small-for-gestational age. Population attributable fractions (PAFs) were calculated using adjusted relative risks.

RESULTS

Compared with babies born to mothers who smoked during pregnancy, babies born to non-smoking mothers had a lower risk of all adverse perinatal outcomes including perinatal death (aRR=0.58, 95% CI 0.44 to 0.76), preterm birth (aRR=0.58, 95% CI 0.53 to 0.64) and small-for-gestational age (aRR=0.35, 95% CI 0.32 to 0.39). PAFs (%) were 27% for perinatal death, 26% for preterm birth and 48% for small-for-gestational-age. Compared with women who smoked during pregnancy (n=8919), those who did not smoke (n=9235) had a lower risk of being transferred to another hospital (aRR=0.76, 95% CI 0.66 to 0.89).

CONCLUSIONS

Babies born to women who did not smoke during pregnancy had a lower risk of adverse perinatal outcomes. Rates of adverse outcomes among Aboriginal non-smokers were similar to those among the general population. These results quantify the proportion of adverse perinatal outcomes due to smoking and highlight why effective smoking cessation programme are urgently required for this population.

The Population Health Research Network - Population Data Centre Profile

The Population Health Research Network (PHRN) is an Australian data linkage infrastructure capable of securely and safely linking and integrating data collections from a wide range of sources. It is an example of a national data linkage infrastructure in a country with a federated system of government. This population data centre profile describes Australia's unique approach to enabling access to linked data from single jurisdictions and from multiple jurisdictions. It covers the background to the establishment of the PHRN as well as information about how it operates today including operating models, governance, data, data linkage and data access. Some of the challenges of data linkage across jurisdictions are also discussed.

Contributions of prognostic factors to socioeconomic disparities in cancer survival: protocol for analysis of a cohort with linked data

INTRODUCTION

Socioeconomic disparities in cancer survival have been reported in many developed countries, including Australia. Although some international studies have investigated the determinants of these socioeconomic disparities, most previous Australian studies have been descriptive, as only limited relevant data are generally available. Here, we describe a protocol for a study to use data from a large-scale Australian cohort linked with several other health-related databases to investigate several groups of factors associated with socioeconomic disparities in cancer survival in New South Wales (NSW), Australia, and quantify their contributions to the survival disparities.

METHODS AND ANALYSIS

The Sax Institute's 45 and Up Study participants completed a baseline questionnaire during 2006-2009. Those who were subsequently diagnosed with cancer of the colon, rectum, lung or female breast will be included. This study sample will be identified by linkage with NSW Cancer Registry data for 2006-2013, and their vital status will be determined by linking with cause of death records up to 31 December 2015. The study cohort will be divided into four groups based on each of the individual education level and an area-based socioeconomic measure. The treatment received will be obtained through linking with hospital records and Medicare and pharmaceutical claims data. Cox proportional hazards models will be fitted sequentially to estimate the percentage contributions to overall socioeconomic survival disparities of patient factors, tumour and diagnosis factors, and treatment variables.

ETHICS AND DISSEMINATION

This research is covered by ethical approval from the NSW Population and Health Services Research Ethics Committee. Results of the study will be disseminated to different interest groups and organisations through scientific conferences, social media and peer-reviewed articles.

Creating the first national linked dataset on perinatal and maternal outcomes in Australia: Methods and challenges

BACKGROUND

Data linkage offers a powerful mechanism for examining healthcare outcomes across populations and can generate substantial robust datasets using routinely collected electronic data. However, it presents methodological challenges, especially in Australia where eight separate states and territories maintain health datasets. This study used linked data to investigate perinatal and maternal outcomes in relation to place of birth. It examined data from all eight jurisdictions regarding births planned in hospitals, birth centres and at home. Data linkage enabled the first Australia-wide dataset on birth outcomes. However, jurisdictional differences in data collection created challenges in obtaining comparable cohorts of women with similar low-risk pregnancies in all birth settings. The objective of this paper is to describe the techniques for managing previously linked data, and specifically for ensuring the resulting dataset contained only low-risk pregnancies.

METHODS

This paper indicates the procedures for preparing and merging linked perinatal, inpatient and mortality data from different sources, providing technical guidance to address challenges arising in linked data study designs.

RESULTS

We combined data from eight jurisdictions linking four collections of administrative healthcare and civil registration data. The merging process ensured that variables were consistent, compatible and relevant to study aims. To generate comparable cohorts for all three birth settings, we developed increasingly complex strategies to ensure that the dataset eliminated women with pregnancies at risk of complications during labour and birth. It was then possible to compare birth outcomes for comparable samples, enabling specific examination of the impact of birth setting on maternal and infant safety across Australia.

CONCLUSIONS

Data linkage is a valuable resource to enhance knowledge about birth outcomes from different settings, notwithstanding methodological challenges. Researchers can develop and share practical techniques to address these challenges. Study findings suggest that jurisdictions develop more consistent data collections to facilitate future data linkage.

Childhood health and education outcomes following early term induction for large-for-gestational age: A population-based record linkage study

INTRODUCTION

There is debate about optimal management of pregnancies with a large-for-gestational age baby. A recent randomized controlled trial reported that early term induction of labor reduced cesarean section rates and infant morbidity. However, long term childhood outcomes have not been assessed. The aim of this study was to assess maternal, neonatal and child health and education outcomes for large-for-gestational age babies induced at 37-38 weeks' gestation.

MATERIAL AND METHODS

Population-based record linkage study of term (37+ weeks), cephalic-presenting singleton pregnancies with a large-for-gestational age baby in New South Wales, Australia, 2002-2006. Linked birth, hospital, mortality and education data were used with at least 9 years follow up from birth. Exposure was induction of labor at 37-38 weeks, compared to expectant management (spontaneous birth at ≥37 weeks and planned births at ≥39 weeks). Relative risks and 95% confidence intervals were estimated using Modified Poisson regression with robust variance.

RESULTS

Among 10 174 eligible pregnancies, 412 (4.0%) had an induction at 37-38 weeks. Women in the induction group were less likely to have a cesarean section (RR: 0.65, 95% CI: 0.51-0.82). Infants had higher rates of: low Apgar scores, birth trauma, neonatal jaundice and phototherapy use, and admission to special care nursery or neonatal intensive care than their expectantly managed counterparts. As children, they had higher rates of hospital admission (RR: 1.16, 95% CI: 1.04-1.30) and special needs (RR: 1.98, 95% CI: 1.12-3.50). However, by age 8 there was no difference in overall literacy and numeracy achievement.

CONCLUSIONS

Although women who had an early term labor induction with large-for-gestational age were less likely to have a cesarean section, the increased risk of neonatal morbidities and additional healthcare utilization suggests the need for caution in early induction of large-for-gestational age babies before 39 weeks' gestation.

Processing of Electronic Medical Records for Health Services Research in an Academic Medical Center: Methods and Validation

Background

Electronic medical records (EMRs) contain a wealth of information that can support data-driven decision making in health care policy design and service planning. Although research using EMRs has become increasingly prevalent, challenges such as coding inconsistency, data validity, and lack of suitable measures in important domains still hinder the progress.

Objective

The objective of this study was to design a structured way to process records in administrative EMR systems for health services research and assess validity in selected areas.

Methods

On the basis of a local hospital EMR system in Singapore, we developed a structured framework for EMR data processing, including standardization and phenotyping of diagnosis codes, construction of cohort with multilevel views, and generation of variables and proxy measures to supplement primary data. Disease complexity was estimated by Charlson Comorbidity Index (CCI) and Polypharmacy Score (PPS), whereas socioeconomic status (SES) was estimated by housing type. Validity of modified diagnosis codes and derived measures were investigated.

Results

Visit-level (N=7,778,761) and patient-level records (n=549,109) were generated. The International Classification of Diseases, Tenth Revision, Australian Modification (ICD-10-AM) codes were standardized to the International Classification of Diseases, Ninth Revision, Clinical Modification (ICD-9-CM) with a mapping rate of 87.1%. In all, 97.4% of the ICD-9-CM codes were phenotyped successfully using Clinical Classification Software by Agency for Healthcare Research and Quality. Diagnosis codes that underwent modification (truncation or zero addition) in standardization and phenotyping procedures had the modification validated by physicians, with validity rates of more than 90%. Disease complexity measures (CCI and PPS) and SES were found to be valid and robust after a correlation analysis and a multivariate regression analysis. CCI and PPS were correlated with each other and positively correlated with health care utilization measures. Larger housing type was associated with lower government subsidies received, suggesting association with higher SES. Profile of constructed cohorts showed differences in disease prevalence, disease complexity, and health care utilization in those aged above 65 years and those aged 65 years or younger.

Conclusions

The framework proposed in this study would be useful for other researchers working with EMR data for health services research. Further analyses would be needed to better understand differences observed in the cohorts.

Impact of linkage quality on inferences drawn from analyses using data with high rates of linkage errors in rural Tanzania

BACKGROUND

Studies based on high-quality linked data in developed countries show that even minor linkage errors, which occur when records of two different individuals are erroneously linked or when records belonging to the same individual are not linked, can impact bias and precision of subsequent analyses. We evaluated the impact of linkage quality on inferences drawn from analyses using data with substantial linkage errors in rural Tanzania.

METHODS

Semi-automatic point-of-contact interactive record linkage was used to establish gold standard links between community-based HIV surveillance data and medical records at clinics serving the surveillance population. Automated probabilistic record linkage was used to create analytic datasets at minimum, low, medium, and high match score thresholds. Cox proportional hazards regression models were used to compare HIV care registration rates by testing modality (sero-survey vs. clinic) in each analytic dataset. We assessed linkage quality using three approaches: quantifying linkage errors, comparing characteristics between linked and unlinked data, and evaluating bias and precision of regression estimates.

RESULTS

Between 2014 and 2017, 405 individuals with gold standard links were newly diagnosed with HIV in sero-surveys (n = 263) and clinics (n = 142). Automated probabilistic linkage correctly identified 233 individuals (positive predictive value [PPV] = 65%) at the low threshold and 95 individuals (PPV = 90%) at the high threshold. Significant differences were found between linked and unlinked records in primary exposure and outcome variables and for adjusting covariates at every threshold. As expected, differences attenuated with increasing threshold. Testing modality was significantly associated with time to registration in the gold standard data (adjusted hazard ratio [HR] 4.98 for clinic-based testing, 95% confidence interval [CI] 3.34, 7.42). Increasing false matches weakened the association (HR 2.76 at minimum match score threshold, 95% CI 1.73, 4.41). Increasing missed matches (i.e., increasing match score threshold and positive predictive value of the linkage algorithm) was strongly correlated with a reduction in the precision of coefficient estimate (R2 = 0.97; p = 0.03).

CONCLUSIONS

Similar to studies with more negligible levels of linkage errors, false matches in this setting reduced the magnitude of the association; missed matches reduced precision. Adjusting for these biases could provide more robust results using data with considerable linkage errors.

Antecedents of Abnormally Invasive Placenta in Primiparous Women: Risk Associated With Gynecologic Procedures

OBJECTIVE

To evaluate the association between prior invasive gynecologic procedures and the risk of subsequent abnormally invasive placenta (ie, placenta accreta, increta, and percreta).

METHODS

We conducted a population-based data linkage study including all primiparous women who delivered in New South Wales, Australia, between 2003 and 2012. Data were obtained from linked birth and hospital admissions with a minimum lookback period of 2 years. Prior procedures invasive of the uterus were considered including gynecologic laparoscopy with instrumentation of the uterus; hysteroscopy, including operative hysteroscopy; curettage, including suction curettage and surgical termination; and endometrial ablation. Modified Poisson regression was used to determine the association between the number of prior gynecologic procedures and risk of abnormally invasive placenta.

RESULTS

Eight hundred fifty-four cases of abnormally invasive placenta were identified among 380,775 deliveries included in the study (22.4/10,000). In total, 33,296 primiparous women had at least one prior procedure (8.7%). Among women with abnormally invasive placenta, 152 (17.8%) had undergone at least one procedure compared with 33,144 (8.7%) among women without abnormally invasive placenta (P<.01). After adjustment, the relative risk was 1.5 for one procedure (99% CI 1.1-1.9), 2.7 for two procedures (99% CI 1.7-4.4), and 5.1 for three or more procedures (99% CI 2.7-9.6). Abnormally invasive placenta was also positively associated with maternal age, socioeconomic advantage, mother being Australia-born, placenta previa, hypertension, multiple births, use of assisted reproductive technology, and female fetal sex.

CONCLUSION

Women with a history of prior invasive gynecologic procedures were more likely to develop abnormally invasive placenta. These insights may be used to inform management of pregnancies in women with a history of gynecologic procedures.

Haemoglobin concentration following postpartum haemorrhage and the association between blood transfusion and breastfeeding: a retrospective cohort study

Objective

The aim of this study was to determine the association between red blood cell transfusion and breastfeeding among women who have suffered a postpartum haemorrhage at birth taking into account post-birth haemoglobin concentrations.

Results

Among 15,451 maternities with postpartum haemorrhage in New South Wales public hospitals between 2007 and 2010, 1828 (12%) received a red cell transfusion. Among transfused women, 686 (38%) had haemoglobin concentration pre-transfusion < 70 g/L, 792 (43%) had 70–90 g/L, and 350 (19%) had > 90 g/L. Rates and adjusted relative risks (aRR) for breastfeeding at hospital discharge were as follows: for women with haemoglobin concentrations < 70 g/L following birth and received a transfusion, 78.6% were breastfeeding and the aRR of breastfeeding compared to untransfused women was 0.90 (99% confidence interval (CI) 0.86–0.95); for women with haemoglobin concentrations 70–90 g/L, 81.3% were breastfeeding, aRR 0.94 (99% CI 0.90–0.98); and for women with haemoglobin concentrations > 90 g/L, 80.9% were breastfeeding, aRR 0.94 (99% CI 0.88–1.00).

The importance of including aliases in data linkage with vulnerable populations

BACKGROUND

Records pertaining to individuals whose identity cannot be verified with legal documentation may contain errors, or be incorrect by intention of the individual. Probabilistic data linkage, especially in vulnerable populations where the incidence of such records may be higher, must be considerate of the usage of these records.

METHODS

A data linkage was conducted between Queensland Youth Justice records and the Australian National Death Index. Links were assessed to determine how often they were made using the unverified (alias) records that would not have been made in their absence (i.e. links that were not also made using solely verified records). Anomalies in the linked records were investigated in order to make evaluations of the sensitivity and specificity of the linkage, compared to the links made using only verified records.

RESULTS

From links made using verified records only, 1309 deaths were identified (2.6% of individuals). Using alias records in addition, the number of links increased by 16%. Links made using alias records only were more common in females, and those born after 1985. Different records belonging to the same individual in the justice dataset did not link to different death records, however there were instances of the same death record linking to multiple cohort individuals.

CONCLUSIONS

The inclusion of aliases in data linkage in youths involved in the justice system increased mortality ascertainment without any discernible increase in false positive matches. We therefore conclude that alias records should be included in data linkage procedures in order to avoid biased attenuation of ascertainment in vulnerable populations, leading to the concealment of health inequality.

The impact of general anesthesia on child development and school performance: a population-based study

BACKGROUND

There has been considerable interest in the possible adverse neurocognitive effects of exposure to general anesthesia and surgery in early childhood.

AIMS

The aim of this data linkage study was to investigate developmental and school performance outcomes of children undergoing procedures requiring general anesthesia in early childhood.

METHODS

We included children born in New South Wales, Australia of 37+ weeks' gestation without major congenital anomalies or neurodevelopmental disability with either a school entry developmental assessment in 2009, 2012, or Grade-3 school test results in 2008-2014. We compared children exposed to general anesthesia aged <48 months to those without any hospitalization. Children with only 1 hospitalization with general anesthesia and no other hospitalization were assessed separately. Outcomes included being classified developmentally high risk at school entry and scoring below national minimum standard in school numeracy and reading tests.

RESULTS

Of 211 978 children included, 82 156 had developmental assessment and 153 025 had school test results, with 12 848 (15.7%) and 25 032 (16.4%) exposed to general anesthesia, respectively. Children exposed to general anesthesia had 17%, 34%, and 23% increased odds of being developmentally high risk (adjusted odds ratio [aOR]: 1.17; 95% CI: 1.07-1.29); or scoring below the national minimum standard in numeracy (aOR: 1.34; 95% CI: 1.21-1.48) and reading (aOR: 1.23; 95% CI: 1.12-1.36), respectively. Although the risk for being developmentally high risk and poor reading attenuated for children with only 1 hospitalization and exposure to general anesthesia, the association with poor numeracy results remained.

CONCLUSION

Children exposed to general anesthesia before 4 years have poorer development at school entry and school performance. While the association among children with 1 hospitalization with 1 general anesthesia and no other hospitalization was attenuated, poor numeracy outcome remained. Further investigation of the specific effects of general anesthesia and the impact of the underlying health conditions that prompt the need for surgery or diagnostic procedures is required, particularly among children exposed to long duration of general anesthesia or with repeated hospitalizations.

Pediatric admissions that include intensive care: a population-based study

Background

Pediatric admissions to intensive care outside children’s hospitals are generally excluded from registry-based studies. This study compares pediatric admission to specialist pediatric intensive care units (PICU) with pediatric admissions to intensive care units (ICU) in general hospitals in an Australian population.

Methods

We undertook a population-based record linkage cohort study utilizing longitudinally-linked hospital and death data for pediatric hospitalization from New South Wales, Australia, 2010–2013. The study population included all new pediatric, post-neonatal hospital admissions that included time in ICU (excluding neonatal ICU).

Results

Of 498,466 pediatric hospitalizations, 7525 (1.5%) included time in an intensive care unit – 93.7% to PICU and 6.3% to ICU in a general (non-PICU) hospital. Non-PICU admissions were of older children, in rural areas, with shorter stays in ICU, more likely admitted for acute conditions such as asthma, injury or diabetes, and less likely to have chronic conditions, receive continuous ventilatory support, blood transfusion, parenteral nutrition or die.

Conclusions

A substantial proportion of children are admitted to ICUs in general hospitals. A comprehensive overview of pediatric ICU admissions includes these admissions and the context of the total hospitalization.

Childhood outcomes following preterm prelabor rupture of the membranes (PPROM): a population-based record linkage cohort study

OBJECTIVE

The objective of this study is to determine child health, development and educational outcomes for infants born following preterm prelabor rupture of the membrane (PPROM).

STUDY DESIGN

Population-based record linkage cohort study using data from NSW, Australia, 2001 to 2014.

RESULTS

Of 121 822 births at 20 to 37 weeks, 18 799 (15%) followed PPROM, 56 406 (46%) followed spontaneous labor and 46 617 (38%) were planned. Compared with infants of a similar gestational age born following spontaneous labor or planned delivery, exposure to PPROM did not increase the risk of childhood mortality, childhood hospitalization, developmentally vulnerable at school entry, low reading or numeracy scores. Median latency ranged from 12 days (interquartile range 3 to 37 days) at 25 weeks to 1 day (0 to 2 days) at 36 weeks. Longer latency and more advanced gestational age at birth were associated with better outcomes.

CONCLUSION

Infants born following PPROM are at no greater risk of adverse child health, development and education outcomes than those of similar gestational age born without PPROM.

Outcomes of Gallstone Disease during Pregnancy: a Population-based Data Linkage Study

BACKGROUND

Gallstone disease is a leading indication for non-obstetric abdominal surgery during pregnancy. There are limited whole population data on maternal and neonatal outcomes. This population-based study aims to describe the outcomes of gallstone disease during pregnancy in an Australian setting.

METHODS

Linked hospital, birth, and mortality data for all women with singleton pregnancies in New South Wales, Australia, 2001-2012 were analysed. Exposure of interest was gallstone disease (acute biliary pancreatitis, gallstones with/without cholecystitis). Outcomes including preterm birth (spontaneous and planned), readmission, morbidity and mortality (maternal and neonatal) were compared between pregnancies with and without gallstone disease. Adjusted risk ratios (aRRs) and 99% confidence intervals were estimated using modified Poisson regression and adjusted for maternal and pregnancy factors.

RESULTS

Among 1 064 089 pregnancies, 1882 (0.18%) had gallstone disease. Of these, 239 (12.7%) had an antepartum cholecystectomy and 1643 (87.3%) were managed conservatively. Of those managed conservatively, 319 (19.0%) had a postpartum cholecystectomy. Gallstone disease was associated with increased risk of preterm birth (aRR 1.3, 99% CI 1.1, 1.6), particularly planned preterm birth (aRR 1.6, 99% CI 1.2, 2.1), maternal morbidity (aRR 1.6, 99% CI 1.1, 2.3), maternal readmission (aRR 4.7, 99% CI 4.2, 5.3), and neonatal morbidity (aRR 1.4, 99% CI 1.1, 1.7). Surgery was associated with decreased risk of maternal readmission (aRR 0.4, 99% CI 0.2, 0.7).

CONCLUSIONS

Gallstone disease during pregnancy was associated with adverse maternal and neonatal outcomes. Most women with gallstone disease during pregnancy are managed conservatively. Surgical management was associated with decreased risk of readmission.

Contribution of maternal age, medical and obstetric history to maternal and perinatal morbidity/mortality for women aged 35 or older

BACKGROUND

As age is not modifiable, pregnancy risk information based on age alone is unhelpful for older women.

AIM

To determine severe morbidity/mortality rates for women aged ≥35 years according to maternal profile based on parity, pre-existing medical conditions and prior pregnancy complications, and to assess the independent contribution of age.

MATERIALS AND METHODS

Population-based record-linkage study using NSW hospitalisation and birth records 2006-2012. Maternal and perinatal mortality/morbidity were assessed for non-anomalous singleton births to women aged ≥35 years.

RESULTS

For 117 357 pregnancies among 99 375 women aged ≥35 years, the median age at delivery was 37 years (range 35-56 years), including: 35 652 (30.4%) multiparae without pre-existing medical or obstetric complications, 33,058 (28.2%) nulliparae without pre-existing medical conditions and 30 325 (25.8%) multiparae with prior pregnancy complications. Maternal and perinatal mortality/morbidity varied by maternal profile with ranges of 0.9-3.5% and 2.4-11.9%, respectively. For nulliparae, each five-year increase in age did not contribute significantly to maternal risk after controlling for medical conditions (adjustedodds ratio 1.08, 95% CI 0.93-1.25), but did confer perinatal risk (1.14; 1.05-1.25). For multiparae, each five-year increase in age beyond 35 years was independently associated with adverse maternal (1.23; 1.09-1.39) and perinatal outcomes (1.23; 1.09-1.39).

CONCLUSIONS

For women aged ≥35 years, presence of medical conditions conferred a greater risk for morbidity/mortality than age itself. For multiparous women, the effects of medical and obstetric history were additive. The contribution of maternal age to adverse outcomes in pregnancies without significant medical and obstetric history is modest.

Tobacco policy reform and population-wide antismoking activities in Australia: the impact on smoking during pregnancy

Introduction

This study examined the impact of antismoking activities targeting the general population and an advertising campaign targeting smoking during pregnancy on the prevalence of smoking during pregnancy in New South Wales (NSW), Australia.

Methods

Monthly prevalence of smoking during pregnancy was calculated using linked health records for all pregnancies resulting in a birth (800 619) in NSW from 2003 to 2011. Segmented regression of interrupted time series data assessed the effects of the extension of the ban on smoking in enclosed public places to include licensed premises (evaluated in combination with the mandating of graphic warnings on cigarette packs), television advertisements targeting smoking in the general population, print and online magazine advertisements targeting smoking during pregnancy and increased tobacco tax. Analyses were conducted for all pregnancies, and for the population stratified by maternal age, parity and socioeconomic status. Further analyses adjusted for the effect of the Baby Bonus maternity payment.

Results

Prevalence of smoking during pregnancy decreased from 2003 to 2011 overall (0.39% per month), and for all strata examined. For pregnancies overall, none of the evaluated initiatives was associated with a change in the trend of smoking during pregnancy. Significant changes associated with increased tobacco tax and the extension of the smoking ban (in combination with graphic warnings) were found in some strata.

Conclusions

The declining prevalence of smoking during pregnancy between 2003 and 2011, while encouraging, does not appear to be directly related to general population antismoking activities or a pregnancy-specific campaign undertaken in this period.

Data cleaning and management protocols for linked perinatal research data: a good practice example from the Smoking MUMS (Maternal Use of Medications and Safety) Study

Background

Data cleaning is an important quality assurance in data linkage research studies. This paper presents the data cleaning and preparation process for a large-scale cross-jurisdictional Australian study (the Smoking MUMS Study) to evaluate the utilisation and safety of smoking cessation pharmacotherapies during pregnancy.

Methods

Perinatal records for all deliveries (2003–2012) in the States of New South Wales (NSW) and Western Australia were linked to State-based data collections including hospital separation, emergency department and death data (mothers and babies) and congenital defect notifications (babies in NSW) by State-based data linkage units. A national data linkage unit linked pharmaceutical dispensing data for the mothers. All linkages were probabilistic. Twenty two steps assessed the uniqueness of records and consistency of items within and across data sources, resolved discrepancies in the linkages between units, and identified women having records in both States.

Results

State-based linkages yielded a cohort of 783,471 mothers and 1,232,440 babies. Likely false positive links relating to 3703 mothers were identified. Corrections of baby’s date of birth and age, and parity were made for 43,578 records while 1996 records were flagged as duplicates. Checks for the uniqueness of the matches between State and national linkages detected 3404 ID clusters, suggestive of missed links in the State linkages, and identified 1986 women who had records in both States.

Conclusions

Analysis of content data can identify inaccurate links that cannot be detected by data linkage units that have access to personal identifiers only. Perinatal researchers are encouraged to adopt the methods presented to ensure quality and consistency among studies using linked administrative data.

Electronic supplementary material

The online version of this article (doi:10.1186/s12874-017-0385-6) contains supplementary material, which is available to authorized users.

Neonatal exchange transfusions in tertiary and non-tertiary hospital settings, New South Wales, 2001-2012

AIM

To describe neonatal exchange transfusions in New South Wales (NSW) before and after release in January 2007 of a NSW Health guideline regarding exchange transfusions in tertiary and non-tertiary hospitals.

METHODS

The study population included neonates receiving exchange transfusion in NSW hospitals, 2001-2012. Linked birth and hospital data for mothers and babies were used to describe birth characteristics and maternal and neonatal conditions. Exchange transfusions were identified in hospital data and compared for 2001-2006 and 2007-2012. Maternal and neonatal characteristics were compared with χ² and Wilcoxon signed-rank tests.

RESULTS

Between 2001 and 2012, there were 286 exchange transfusions performed for 281 neonates in NSW hospitals. The number of exchange transfusions decreased from 187 in 184 neonates for 2001-2006 to 99 in 97 neonates 2007-2012 (P < 0.001). The percentage of exchange transfusions performed at tertiary hospitals increased from 85% in 2001-2006 to 91% in 2007-2012, although this was not statistically significant (P = 0.16). Most neonates requiring exchange transfusion were born in tertiary hospitals: 62% for 2001-2006 and 69% for 2007-2012. Among those born in a non-tertiary hospital, the percentage transferred or admitted to a tertiary hospital for exchange transfusion was 63% in 2001-2006 and 77% in 2007-2012.

CONCLUSION

Between 2001 and 2012, there was a decrease in neonatal exchange transfusions in NSW. After the 2007 guideline there was a non-significant increase in the proportion of exchange transfusions performed at tertiary hospitals. Although rare, exchange transfusions are still expected to occur occasionally in non-tertiary hospitals, requiring continuing support for this procedure in these settings.

Identifying incident colorectal and lung cancer cases in health service utilisation databases in Australia: a validation study

Background

Data from centralised, population-based statutory cancer registries are generally considered the 'gold standard' for confirming incident cases of cancer. When these are not available, or more current information is needed, hospital or other routinely collected population-level data may be feasible alternative sources. We aimed to determine the validity of various methods using routinely collected administrative health data for ascertaining incident cases of colorectal or lung cancer in participants from the 45 and Up Study in New South Wales (NSW), Australia.

Methods

For 266,844 participants in the 45 and Up Study (recruited 2006–2009) ascertainment of incident colorectal or lung cancers was assessed using diagnosis and treatment records in linked administrative health datasets (hospital, emergency department, Medicare and pharmaceutical claims, death records). This was compared with ascertainment via the NSW Cancer Registry (NSWCR, the 'gold standard') for a period for which both data sources were available for participants.

Results

A total of 2253 colorectal and 1019 lung cancers were recorded for study participants in the NSWCR over the period 2006–2010. A diagnosis of primary cancer recorded in the statewide Admitted Patient Data Collection identified the majority of NSWCR colorectal and lung cancers, with sensitivities and positive predictive values (PPV) of 95% and 91% for colorectal cancer and 81% and 85% for lung cancer, respectively. Using additional information on lung cancer deaths from death records increased sensitivity to 84% (PPV 83%) for lung cancer, but did not improve ascertainment of colorectal cancers. Hospital procedure codes for colorectal cancer surgery identified cases with sensitivity 81% and PPV 54%. No other individual indicator had sensitivity >50% or PPV >65% for either cancer type and no combination of indicators increased both the sensitivity and PPV above that achieved using the hospital cancer diagnosis data. All specificities were close to 100%; 95% confidence intervals for sensitivity and PPV were generally +/−2%.

Conclusions

In NSW, identifying new cases of colorectal and lung cancer from administrative health datasets, such as hospital records, is a feasible alternative when cancer registry data are not available. However, the strengths and limitations of the different data sources should be borne in mind.

Electronic supplementary material

The online version of this article (doi:10.1186/s12911-017-0417-5) contains supplementary material, which is available to authorized users.

Probabilistic Matching Approach to Link Deidentified Data from a Trauma Registry and a Traumatic Brain Injury Model System Center

OBJECTIVE

There is no civilian traumatic brain injury database that captures patients in all settings of the care continuum. The linkage of such databases would yield valuable insight into possible care interventions. Thus, the objective of this article is to describe the creation of an algorithm used to link the Traumatic Brain Injury Model System (TBIMS) to trauma data in state and national trauma databases.

DESIGN

The TBIMS data from a single center was randomly divided into two sets. One subset was used to generate a probabilistic linking algorithm to link the TBIMS data to the center's trauma registry. The other subset was used to validate the algorithm. Medical record numbers were obtained and used as unique identifiers to measure the quality of the linkage. Novel methods were used to maximize the positive predictive value.

RESULTS

The algorithm generation subset had 121 patients. It had a sensitivity of 88% and a positive predictive value of 99%. The validation subset consisted of 120 patients and had a sensitivity of 83% and a positive predictive value of 99%.

CONCLUSIONS

The probabilistic linkage algorithm can accurately link TBIMS data across systems of trauma care. Future studies can use this database to answer meaningful research questions regarding the long-term impact of the acute trauma complex on health care utilization and recovery across the care continuum in traumatic brain injury populations.

Early-onset preeclampsia appears to discourage subsequent pregnancy but the risks may be overestimated

BACKGROUND

Early-onset preeclampsia is associated with adverse maternal and perinatal outcomes. For women who consider another pregnancy after one complicated by early-onset preeclampsia, the likelihood of recurrence and the subsequent pregnancy outcome for themselves and their babies are pertinent considerations.

OBJECTIVES

The purpose of this study was to determine the subsequent pregnancy rate after a nulliparous pregnancy that was complicated by early-onset preeclampsia and among those who have a subsequent pregnancy, the risk of recurrence by gestational week, and adverse pregnancy outcomes.

STUDY DESIGN

This was a population-based record linkage cohort study. The study population included nulliparous women with a singleton pregnancy and early-onset preeclampsia (<34 weeks gestation) who gave birth in New South Wales Australia from 2001-2010 (the index birth), with follow-up data for a subsequent birth through 2012. Early-onset in the index birth was further categorized as <28 vs 28-33 weeks gestation. Subsequent pregnancy outcomes that were assessed included the pregnancy rate, preeclampsia recurrence, and maternal and perinatal morbidity and mortality rates. The risk of preeclampsia necessitating delivery at each gestational week for women who were at risk was plotted, and the net gain or loss of gestational age when comparing the index with the subsequent pregnancy was calculated.

RESULTS

Among 361,031 nulliparous women with singleton pregnancies, 1473 (0.4%) had early-onset preeclampsia. Women with early-onset preeclampsia in their first pregnancy had a lower subsequent pregnancy rate (59.7%) than women without preeclampsia (67.7%). Of the 758 women with a subsequent singleton birth, 256 (33.8%) experienced preeclampsia in the next pregnancy; 57 women (7.5%) with recurrent early-onset preeclampsia were included. Cumulative rates of preeclampsia in the subsequent pregnancy were higher at every gestation from 23 weeks gestation when the index birth was <28 weeks compared with 28-33 weeks gestation. The cumulative rate and gestation-specific risk of recurrent preeclampsia rose most steeply at 32-38 weeks gestation. Most women (94.6%) progressed to a later gestational age in their subsequent pregnancy. The median overall increase in gestational age at delivery was 6 weeks (interquartile range, 4-8); among women with recurrent preeclampsia, the median increase was 5 weeks (interquartile range, 2-7). Women with index birth <28 weeks gestation compared with 28-33 weeks gestation were more likely to deliver preterm (38.8% vs 28.7%; relative risk, 1.35; 95% confidence interval, 1.04-1.75) and have a perinatal death (4.3% vs 1.2%; relative risk, 3.46; 95% confidence interval, 1.15-10.39) at the subsequent birth, but live born infants had similar rates of severe morbidity (17.1% vs 15.0%; relative risk, 1.14; 95% confidence interval, 0.73-1.79).

CONCLUSION

Women with early-onset preeclampsia in a first pregnancy appear less likely than women without preeclampsia to have a subsequent pregnancy. Maternal and perinatal outcomes in the subsequent pregnancy are generally better than in the first; most women will not have recurrent preeclampsia, and those who do usually will give birth at a greater gestational age compared with their index birth.

Factors Associated with Recurrent Infant Feeding Practices in Subsequent Births

BACKGROUND

Previous breastfeeding experience has been associated with subsequent infant feeding practices. However, few longitudinal studies have investigated formula-only feeding patterns or the full range of potentially associated characteristics.

OBJECTIVE

This study aimed to determine the recurrence of infant feeding practices and maternal, birthing, and infant characteristics associated with recurrent formula-only feeding and changes between exclusive breastfeeding and formula-only feeding across subsequent births.

METHODS

We conducted a population-based record-linkage study of 317 027 mothers, with a term singleton live-birth in 2007-2011, New South Wales, Australia. Infant feeding patterns were described using sequential birth pairs. For mothers with a first and second birth, robust Poisson regression was used to investigate the association between maternal, birthing, and infant characteristics and infant feeding patterns. Combined relative risks (RRs) were calculated for selected maternal characteristics.

RESULTS

Across 69 994 sequential birth pairs, the recurrence rate of formula-only feeding was 71%, and 92% for exclusive breastfeeding. Maternal characteristics < 25 years old, being Australian born or single, smoking during pregnancy, and living in lower socioeconomic areas were most strongly associated with repeat formula-only feeding (RR, 22.1; 95% confidence interval [CI], 18.6-26.3), changing from exclusive breastfeeding to formula-only feeding (RR, 9.0; 95% CI, 7.4-10.7), and being less likely to change from formula-only feeding to exclusive breastfeeding (RR, 0.47; 95% CI, 0.38-0.59).

CONCLUSION

Infant feeding practices were strongly recurrent, highlighting the importance of successful breastfeeding for first-time mothers. Additional support for young mothers from disadvantaged backgrounds accounting for infant feeding history, experiences, and common barriers could improve recurrent exclusive breastfeeding and positively affect infant and maternal health.

Association of Gestational Age and Severe Neonatal Morbidity with Mortality in Early Childhood

BACKGROUND

Although infant and child mortality rates have decreased substantially worldwide over the past two decades, efforts continue in many nations to further these declines. The identification of pertinent perinatal factors that are associated with early childhood mortality would help with these efforts. We investigated the association of two crucial perinatal factors, gestational age and severe neonatal morbidity at birth, with mortality during infancy (29-364 days) and early childhood (1-5 years).

METHODS

The study population included all singleton livebirths, ≥32 weeks' gestation in New South Wales, Australia in 2001-11. Birth data were linked to hospitalisation morbidity data and deaths data (linked birth cohort n = 871 916), and multivariable Cox regression models were used to assess mortality.

RESULTS

The median follow-up time per child was 4.95 years (range 0.00-5.92 years; 3 614 738 total person-years), with 984 deaths observed. Gestational age was associated with increased mortality, and specifically from deaths attributable to infections, respiratory conditions, and injuries during infancy, but not during early childhood. Severe neonatal morbidity strongly mediated the effects of gestational age during infancy, but not during early childhood, and was associated with increased mortality from circulatory, nervous, and respiratory system causes.

CONCLUSIONS

The direct effects of gestational age on mortality extended up to 1 year of age, whereas severe neonatal morbidity remained associated with heightened mortality into early childhood. Efforts to maximise the health and well-being of vulnerable infants, with emphasis on preventing infections and injuries, may help further reduce early childhood mortality.

Describing the linkages of the immigration, refugees and citizenship Canada permanent resident data and vital statistics death registry to Ontario's administrative health database

BACKGROUND

Ontario, the most populous province in Canada, has a universal healthcare system that routinely collects health administrative data on its 13 million legal residents that is used for health research. Record linkage has become a vital tool for this research by enriching this data with the Immigration, Refugees and Citizenship Canada Permanent Resident (IRCC-PR) database and the Office of the Registrar General's Vital Statistics-Death (ORG-VSD) registry. Our objectives were to estimate linkage rates and compare characteristics of individuals in the linked versus unlinked files.

METHODS

We used both deterministic and probabilistic linkage methods to link the IRCC-PR database (1985-2012) and ORG-VSD registry (1990-2012) to the Ontario's Registered Persons Database. Linkage rates were estimated and standardized differences were used to assess differences in socio-demographic and other characteristics between the linked and unlinked records.

RESULTS

The overall linkage rates for the IRCC-PR database and ORG-VSD registry were 86.4 and 96.2 %, respectively. The majority (68.2 %) of the record linkages in IRCC-PR were achieved after three deterministic passes, 18.2 % were linked probabilistically, and 13.6 % were unlinked. Similarly the majority (79.8 %) of the record linkages in the ORG-VSD were linked using deterministic record linkage, 16.3 % were linked after probabilistic and manual review, and 3.9 % were unlinked. Unlinked and linked files were similar for most characteristics, such as age and marital status for IRCC-PR and sex and most causes of death for ORG-VSD. However, lower linkage rates were observed among people born in East Asia (78 %) in the IRCC-PR database and certain causes of death in the ORG-VSD registry, namely perinatal conditions (61.3 %) and congenital anomalies (81.3 %).

CONCLUSIONS

The linkages of immigration and vital statistics data to existing population-based healthcare data in Ontario, Canada will enable many novel cross-sectional and longitudinal studies to be conducted. Analytic techniques to account for sub-optimal linkage rates may be required in studies of certain ethnic groups or certain causes of death among children and infants.

Association between interpregnancy interval and the risk of recurrent loss after a midtrimester loss

STUDY QUESTION

After an initial midtrimester loss, is the interval to the next conception associated with the risk of a recurrent loss?

SUMMARY ANSWER

Among women who had a pregnancy loss at 14-19 weeks gestation, conception at least 3 months after this initial loss was associated with a reduced risk of a recurrent loss.

WHAT IS KNOWN ALREADY

A short interpregnancy interval (IPI) has been thought to increase risk but recent studies of pregnancy after a loss have found no effect; however, these studies have been based almost entirely on an initial first trimester (<14 weeks) loss.

STUDY DESIGN, SIZE, DURATION

A retrospective cohort study drawing on over 997 000 linked birth and hospital records from New South Wales, Australia for 2003-2011. Index pregnancies were those of women who had a first recorded pregnancy loss of 14-23 weeks gestation (miscarriage, termination and perinatal death). The study population was 4290 women who conceived again within 2 years.

PARTICIPANTS/MATERIALS, SETTING, METHODS

The index loss was categorized by subgroups: 14-19 weeks gestation versus 20-23 weeks, and by whether spontaneous or a termination. The primary outcome was any loss or perinatal death before 24 weeks in the subsequent pregnancy.

MAIN RESULTS AND THE ROLE OF CHANCE

After a 14-19 weeks index loss, an IPI of ≤3 months had an increased rate of recurrent loss compared with an IPI of >9-12 months: 21.9% versus 11.3% (adjusted relative risk (aRR) = 2.02, 95% CI 1.44-2.83). For women who had a spontaneous index loss of 20-23 weeks, there was no evidence that a short IPI increased or decreased the risk of recurrent loss. For any gestational age group of index losses, an IPI of >18-24 months increased the risk of a recurrent loss; the risk was highest after a 20-23 weeks index loss (aRR = 2.15, 95% CI 1.18-3.91).

LIMITATIONS, REASONS FOR CAUTION

We do not know how many cycles were required to achieve conception. Pregnancies resulting in early first trimester losses are unlikely to have resulted in hospitalization so would not have been identified.

WIDER IMPLICATIONS OF THE FINDINGS

The risk of recurrent loss after an initial midtrimester loss may differ from the risk after an initial first trimester loss.

STUDY FUNDING/COMPETING INTERESTS

This work was supported by an Australian National Health and Medical Research Council (NHMRC) Centre for Research Excellence Grant (1001066). C.L.R. is supported by an NHMRC Senior Research Fellowship (#APP1021025). J.B.F. is supported by an ARC Future Fellowship (#120100069). The authors have no conflicts of interest to declare.

TRIAL REGISTRATION NUMBER

N/A.

Preparing linked population data for research: cohort study of prisoner perinatal health outcomes

BACKGROUND

A study of pregnancy outcomes related to pregnancy in prison in New South Wales, Australia, designed a two stage linkage to add maternal history of incarceration and serious mental health morbidity, neonatal hospital admission and infant congenital anomaly diagnosis to birth data. Linkage was performed by a dedicated state-wide data linkage authority. This paper describes use of the linked data to determine pregnancy prison exposure pregnancy for a representative population of mothers.

METHODS

Researchers assessed the quality of linked records; resolved multiple-matched identities; transformed event-based incarceration records into person-based prisoner records and birth records into maternity records. Inconsistent or incomplete records were censored. Interrogation of the temporal relationships of all incarceration periods from the prisoner record with pregnancies from birth records identified prisoner maternities. Interrogation of maternities for each mother distinguished prisoner mothers who were incarcerated during pregnancy, from prisoner control mothers with pregnancies wholly in the community and a subset of prisoner mothers with maternities both types of maternity. Standard descriptive statistics are used to provide population prevalence of exposures and compare data quality across study populations stratified by mental health morbidity.

RESULTS

Women incarcerated between 1998 and 2006 accounted for less than 1 % of the 404,000 women who gave birth in NSW between 2000 and 2006, while women with serious mental health morbidity accounted for 7 % overall and 68 % of prisoners. Rates of false positive linkage were within the predicted limits set by the linkage authority for non-prisoners, but were tenfold higher among prisoners (RR 9.9; 95%CI 8.2, 11.9) and twice as high for women with serious mental health morbidity (RR 2.2; 95%CI 1.9, 2.6). This case series of 597 maternities for 558 prisoners pregnant while in prison (of whom 128 gave birth in prison); and 2,031 contemporaneous prisoner control mothers is one of the largest available.

CONCLUSIONS

Record linkage, properly applied, offers the opportunity to extend knowledge about vulnerable populations not amenable to standard ascertainment. Dedicated linkage authorities now provide linked data for research. The data are not research ready. Perinatal exposures are time-critical and require expert processing to prepare the data for research.

Gestational age, mode of birth and breastmilk feeding all influence acute early childhood gastroenteritis: a record-linkage cohort study

Background

Acute gastroenteritis (AGE) is a leading cause of infectious morbidity in childhood. Clinical studies have implicated caesarean section, early birth and formula feeding in modifying normal gut microbiota development and immune system homeostasis in early life. Rates of early birth and cesarean delivery are also increasing worldwide. This study aimed to investigate the independent and combined associations of the mode and timing of birth and breastmilk feeding with AGE hospitalisations in early childhood.

Methods

Population-based record-linkage study of 893,360 singleton livebirths of at least 33 weeks gestation without major congenital conditions born in hospital, New South Wales, Australia, 2001–2011. Using age at first AGE hospital admission, Cox-regression was used to estimate the associations for gestational age, vaginal birth or caesarean delivery by labour onset and formula-only feeding while adjusting for confounders.

Results

There were 41,274 (4.6 %) children admitted to hospital at least once for AGE and the median age at first admission was 1.4 years. Risk of AGE admission increased with decreasing gestational age (37–38 weeks: 15 % increased risk, 33–36 weeks: 25 %), caesarean section (20 %), planned birth (17 %) and formula-only feeding (18 %). The rate of AGE admission was highest for children who were born preterm by modes of birth other than vaginal birth following the spontaneous onset of labour and who received formula-only at discharge from birth care (62–78 %).

Conclusions

Vaginal birth following spontaneous onset of labour at 39+ weeks gestation with any breastfeeding minimised the risk of gastroenteritis hospitalisation in early childhood. Given increasing trends in early planned birth and caesarean section worldwide, these results provide important information about the impact obstetric interventions may have on the development of the infant gut microbiota and immunity.

Electronic supplementary material

The online version of this article (doi:10.1186/s12887-016-0591-0) contains supplementary material, which is available to authorized users.