Record Linkage Study Research Articles

Recurrent pain and work disability: a record linkage study

PurposeWe examined the associations between recurrent single- and multisite pain and incident sickness absence (SA) of different lengths and the risk of disability pension (DP).MethodsThe data were derived from the Finnish Helsinki Health Study. Pain measures were recorded for panel 1 in 2000/2 and 2007, and for panel 2 in 2007 and 2012 (altogether 3191 employees). SA data were obtained from the employer’s personnel register and DP events from the Finnish Centre for Pensions. Negative binomial regression models with generalized estimation equations were used to model the incidence of self-certified short- (1–3 days), and medically certified medium- (4–14 days) and long-term (more than 14 days) SA episodes. Cox regression models were fitted for the associations between pain and all-cause DP and competing risk models for DP by diagnostic groups. Social and health-related covariates were adjusted for.ResultsRecurrent pain was associated with short-, medium- and long-term SA. Additionally, recurrent single- and multisite pain increased the risk of long-term SA. Recurrent single or multisite pain was further associated with an increased risk of DP, while a single instance of pain did not increase the risk.ConclusionsThese results suggest that recurrent pain is a robust determinant of subsequent SA and DP risk. Improved understanding of determinants of recurrent pain is needed to inform the development of targeted measures to reduce SA and premature exit from employment.

The Unmet Need for Psychotropic Medication within the Migrant Population of Northern Ireland - A Record Linkage Study

BackgroundA large proportion of people with mental ill-health do not receive the care that they require - constituting unmet need. Migrant populations are particularly at risk of not receiving necessary care for a range of reasons including perceptual differences and multiple barriers to health service.
 Main AimThis record linkage study compares the likelihood of people with poor mental health receiving psychotropic medication between migrant and settled communities and amongst different migrant groups in Northern Ireland.
 MethodsA cohort of 78,267 people aged 16-64 who reported chronic poor mental health in the 2011 Census records was generated and followed for the next 12 months by linkage to a centralised prescribing dataset to determine rates of treatment. All individual and household attributes were derived from Census forms. Absence of a psychotropic medication prescription from the beginning of 2011 represented unmet need. Logistic regression analyses quantified the relationship between psychotropic medication and migrant status, while accounting for relevant demographic and socio-economic factors.
 ResultsOverall, 86% of the settled population with chronic poor mental health received at least one psychotropic medication during the study period, compared to 67% of the equivalent population of 1,736 migrants. Although migrants were 61% less likely than the settled community to report poor mental health (OR 0.39 (95% CI 0.37 - 0.41)), those reporting poor mental health were less than one-third as likely as to be on psychotropic medication in fully adjusted models. There also existed considerable variation between individual migrant groups, with uptake of medication lowest amongst the Lithuanian migrant group (OR 0.09 (95% CI 0.05 – 0.14)), and highest amongst German migrant group (OR 1.30 (95% CI 0.87 – 1.94)).
 ConclusionAlthough this study suggests substantial unmet need for treatment of poor mental health amongst the migrant population of Northern Ireland this may not represent the whole story. Differences between individual migrant groups suggest that other factors may also be at play including differing approaches to treatment, or alternative routes to treatment. The implications of the study will be discussed in the paper.

How complete, representative and accurate is recording of child BMI in electronic general practice records? A record linkage study

Background with rationaleGeneral practitioner (GP) electronic health records (EHRs) are a potentially valuable, but unevaluated, source of child BMI measurements to inform clinical practice and research.
 Main AimTo assess representativeness and accuracy of child GP-BMI records.
 MethodsWe linked school National Child Measurement Programme (NCMP) records from 29,839 five-year-olds (49.1% girls) and 26,660 11-year-olds (49.1% girls) in City & Hackney (2013-17), Newham (2014-17) and Tower Hamlets (2015-17) to GP EHRs using pseudonymised NHS numbers (94.9% and 95.1% linked, respectively) and identified GP-BMI measurements using Read code “22K..”. We estimated adjusted odds ratios (aOR) of at least one GP-BMI by: sex (reference category: male); ethnic background (White); area-level deprivation (most deprived Index of Multiple Deprivation quintile); weight status (healthy weight; clinical UK1990); and long-term condition (none). We estimated mean BMI difference (NCMP-BMI minus GP-BMI kg/m2) and 95% Limits of Agreement (LoA; Bland and Altman method).
 ResultsWe identified at least one GP-BMI in 10.5% (2,964/28,330) and 26.0% (6,598/25,365) of 5- and 11-year-olds respectively.
 Five-year-old children who were underweight (aOR; 95% CI: 1.70; 1.28,2.25) or obese (1.45; 1.28,1.65), from South Asian backgrounds (1.63; 1.45,1.80) and with long-term conditions (9.58; 8.13,11.28) were more likely, and girls (0.88; 0.81,0.95) and those from less deprived areas (Wald statistic; p-value: 40.06; <0.0001) less likely, to have at least one GP-BMI measurement recorded. Findings among 11-year-olds were similar.
 We identified GP-BMI measurements made within one month of NCMP-BMI in 5.4% (160/2,964) of 5-year-olds and 4.0% (263/6,598) of 11-year-olds. There was poor agreement between NCMP-BMI and GP-BMI: mean difference (95% LoA): +0.55 (-2.49,+3.58) and +0.16 (-2.85,+3.18) in five- and 11-year-olds respectively.
 ConclusionChild BMI is not comprehensively recorded in UK GP settings. Access to BMI school measurements in GP settings could support discussions about child weight status between children, their families and general practitioners.

Hidden Harms of Hypnotics: a population based record linkage study of psychotropic medication and suicide risk

Background/Main Aim Although psychotropic medication is a known risk factor for suicide, studies investigating the relationship between specific types of psychotropic medication and suicide in adults are scarce.
 Methods A cohort of 1,115,647 individuals (18-75years) was identified from Northern Ireland’s 2011 Census records and linked to prescription records and death registrations until 2015. Individual, household and area-based attributes were derived from census records and selected on the basis of known suicide risk. Five categories of psychotropic medication was identified; none, antidepressants, anxiolytics, hypnotics and antipsychotics. Logistic regression was used to model the association between specific types of psychotropic medication (singularly and in combinations) and death by suicide adjusting for sociodemographic and physical health characteristics.
 Results About 14% of the cohort (n=153,319) had received ≥2 prescriptions of any psychotropic medication in the five months around Census. Eight hundred and seventy persons (n=870) died by suicide during 2011 to 2015. Suicide was higher in those on psychotropic medication than in those without (0.24% versus 0.05% respectively). In the fully adjusted logistic regression model, association between death by suicide and any psychotropic medication record remained statistically significant (AOR=3.69, 95%CI: 3.13–4.35). However, individuals solely on hypnotics had the highest AOR of those on singular medication (AOR=4.24, 95%CI: 2.96–6.06 compared to AOR=2.22, 95%CI: 1.75–2.82 for those on antidepressant medication only). In terms of medication combinations, hypnotics both with anxiolytics, and in combination with antidepressants and anxiolytics were the strongest predictors of death by suicide (AOR=8.76, 95%CI: 5.93–12.94 and AOR=8.53, 95%CI: 5.62–12.94 respectively).
 Conclusion Risk of death by suicide varied by medication type with hypnotic medications making an important contribution to death by suicide. Although previous work has focused on antidepressant therapy and suicide risk, those on hypnotic medication are an at-risk group that warrant further attention.

Mortality of Canadian military personnel over 40 years

Background Military personnel may be exposed to hazardous substances or environments, making health surveillance critical. However, surveillance is frequently handicapped by long lag times between exposure(s) and outcomes, which often manifest post-military release and are often not recorded.
 Aims To describe the burden of mortality in still serving and released Canadian Armed Forces (CAF) personnel.
 MethodsThe Canadian Forces Cancer and Mortality Study II (CF CAMS II) is an interdepartmental record linkage study using CAF pay data and Statistics Canada cancer and mortality data. The cohort included all Regular Force and class C Reservist men and women who first enrolled in the CAF between 1976 and 2012, inclusive. The anonymized linked data included death data, including cause and location of death. All-cause mortality (ACM) and International Classification of Disease (ICD)-10 chapter-level mortality (CLM) were quantified using standardized mortality ratios (SMRs), with the Canadian general population (CGP) as the reference population.
 ResultsApproximately 6870 deaths occurred during over 5 million person-years of observations. For ACM, the CAF risk of death was significantly lower than the CGP for both sexes (females: n = 540, SMR = 0.76 [95% CI 0.69–0.82]; males: n = 6330, SMR = 0.79 [95% CI 0.77–0.81]). In the CLM analysis, SMRs were significantly lower than, or not statistically different from, 1.0 for all ICD chapters.
 ConclusionsMilitary service may have a protective effect that may be partly explained by the healthy soldier effect and the stringent selection process at enrolment.

Male factor infertility and risk of death: a nationwide record-linkage study.

What is the risk of death among men with oligospermia, unspecified male factor and azoospermia in the years following fertility treatment? No significantly elevated risk was observed among men with oligospermia and unspecified male factor, while an increased risk was found among men with azoospermia. Previous studies have shown associations between male factor infertility and risk of death, but these studies have relied on internal reference groups and the risk of death according to type of male infertility is not well characterized. In this prospective record-linkage cohort study, we identified men who had undergone medically assisted reproduction (MAR) between 1994 and 2015. Data was linked to the Danish causes of death register and sociodemographic registers through personal identification numbers assigned to all Danish citizens at birth. Men that had undergone MAR in Denmark (MAR Cohort; n= 64 563) were identified from the Danish IVF register, which includes data on whether infertility was due to male factor. For each man in the MAR cohort, five age-matched men who became fathers without fertility treatment were selected from the general population (non-MAR fathers; n= 322 108). Men that could not adequately be tracked in the Danish CPR register (n= 1259) and those that were censored prior to study entry (n= 993) were excluded, leaving a final population of 384 419 men. Risk of death was calculated by Cox regression analysis with age as an underlying timeline and adjustments for educational attainment, civil status and year of study entry. The risk of death was compared among men with and without male factor infertility identified from the IVF register (internal comparisons) as well as to the non-MAR fathers (external comparison). The risk of death between the MAR cohort (all men, regardless of infertility) and the non-MAR fathers was comparable [hazard ratio (HR), 1.07; 95% CI, 0.98-1.15]. When the MAR cohort was limited to infertile men, these men were at increased risk of death [HR, 1.27; 95% CI, 1.12-1.44]. However, when stratified by type of male factor infertility, men with azoospermia had the highest risk of death, which persisted when in both the internal [HR, 2.30; 95% CI, 1.54-3.41] and external comparison [HR, 3.32; 95% CI, 2.02-5.40]. No significantly elevated risk of death was observed among men with oligospermia [HR, 1.14; 95% CI, 0.87-1.50] and unspecified male factor [HR, 1.10; 95% CI, 0.75-1.61] compared with the non-MAR fathers. The same trends were observed for the internal comparison. Duration of the follow-up was limited and there is limited generalizability to infertile men who do not seek fertility treatment. Using national health registers, we found an increased risk of death among azoospermic men while no increased risk was found among men with other types of infertility. For the azoospermic men, further insight into causal pathways is needed to identify options for monitoring and prevention. This study is part of the ReproUnion collaborative study, co-financed by the European Union, Interreg V ÖKS. C.G.'s research stay at Stanford was funded by grants from the University of Copenhagen, Kong Christian den Tiendes Fond, Torben og Alice Frimodt Fond and Julie Von Müllen Fond. M.E. is an advisor for Sandstone and Dadi. All other authors declare no conflict of interests. Not relevant.

Measuring the burden of mental illness and substance use and the level and impact of health care response in patients with spinal trauma: a record linkage study.

Background with rationale Traumatic spinal injuries (TSI) include column fractures, spinal cord injury, or both. They are among the most severe injuries with potential long-term physical, psychological, and social consequences. Primary causes of TSIs are falls and motor vehicle crashes, however, mental illness and substance use are known to significantly increase all injury risk. Injury is also known to increase risks of mental deterioration and physical complications including self-harm and self-neglect.
 Main Aim We aimed to identify comorbid mental illness and/or substance use at incident TSI, quantifying associated costs and health service management of these inequities.
 Methods NSW record-linkage administrative data analyses (2013-2016) will determine accurate prevalence of mental illness and/or substance use disorder among all patients who sustained acute TSI during the study period. Using recurrent event analyses, we will estimate the contributions of mental illness and/or substance use disorder on the impact on hospital acquired complications (HAC), length of stay and costs; assessing records for social work and/or psychologist consultation.
 Results 13,489 individuals were hospitalised with acute TSI; 21% had either mental health and/or substance use diagnoses; 8.7% had both. These patients were more likely to have experienced falls or intentional self-harm, be male and have multiple comorbidity. Acute care stay and costs were on average twice that of patients with TSI without mental health and/or substance use diagnoses; additionally they were more than twice as likely to experience HACs. Only 56% of TSI patients with these comorbid conditions in the context of TSI, had documented social work or psychologist consultation.
 Conclusion Patients with mental illness and/or substance use disorder, experience significant health disparities that require concerted health system attention that should begin early in acute care.

Poor mental health and uptake of disability benefits

Background with rationale Disability related social security benefits were introduced to help off-set the additional costs incurred by people with disability and to help them better integrate into society. However, there are concerns that not everyone with a disability is receiving the benefits to which they are entitled.
 Aim To undertake a record-linkage study to compare levels of benefit uptake Disability Living Allowance (DLA) according to disability type and severity as reported in the Northern Ireland 2011 Census.
 Data/methods The linked data sources were the 2011 census records and all DLA records for calendar year 2011. The main comparison was of DLA receipt amongst those reporting chronic mobility difficulties (MD) and those with chronic poor mental health (PMH) (both yes/no) with the addition of limitation of daily activities (none, a little, a lot), which was used as a proxy for severity and DLA eligibility.
 Results Most people with disability had multiple disabilities but here the analysis is primarily amongst people with solitary conditions to avoid potential confounding. Approximately 20,000 people in the population reported either only MD or only PMH, though 68% of MD had their daily activities limited a lot compared to 55% of those with PMH. Overall, 71.8% of people with MD and daily activities limited a lot were on DLA compared to only 57.5% of similarly affected people with PMH the relative odds ratio being (OR 0.53 95%CI 0.50, 0.56).
 Conclusion It appears that people with poor mental health are less likely to receive disability benefits compared to those with mobility problems. This may be because they are less able to prove their eligibility for these benefits under the current assessment processes.

Congenital viral infections in England over five decades: a population-based observational study

Congenital viral infections cause substantial long-term morbidity but population-based data about diagnosis rates are scarce. The aim of this study was to assess the long-term trends in congenital viral infections in England and to report on how the rates of these infections might have changed with improved methods for detection, the introduction of the two-dose measles-mumps-rubella (MMR) vaccine in 1996, and the implementation of the Newborn Hearing Screening Programme (NHSP) in 2006. For this population-based, observational cohort study, we used national and regional hospitalisation data from 1968 to 2016 in England (Hospital In-Patient Enquiry, Hospital Episode Statistics, and Oxford Record Linkage Study) to calculate annual rates of hospital discharges coded with-and individuals aged younger than 1 month diagnosed with-congenital cytomegalovirus, herpes simplex virus (HSV), varicella zoster virus (VZV), and rubella. We investigated associations of congenital cytomegalovirus, HSV, and VZV with perinatal and maternal factors (sex, mother's ethnicity, mode of delivery, gestational age, birthweight, mother's age, mother's index of multiple deprivation, and number of previous pregnancies). In 2016, discharge rates per 100 000 infant population were 22·3 (95% CI 18·8-26·1) for congenital cytomegalovirus, 17·6 (14·6-21·1) for HSV, 32·6 (28·4-37·2) for VZV, and 0·15 (0·0-0·8) for rubella. Compared with earlier years of the study, the discharge rate in 2016 was higher for congenital cytomegalovirus, HSV, and VZV, whereas it was lower for rubella. For congenital cytomegalovirus, there was a significant step-increase between 2006 and 2007 following implementation of the NHSP (rate ratio comparing the trend line post-NHSP with that pre-NHSP 1·55 [95% CI 1·12-2·14], p=0·0072). Congenital cytomegalovirus infection was associated with birthweight less than 1 kg, maternal age younger than 25 years, socioeconomically deprived households, casearean section, and mothers of black ethnicity. Congenital HSV infection was associated with maternal age younger than 20 years, gestational age less than 32 weeks, and vaginal and emergency caesarean section deliveries, while VZV infection was associated with increased parity and black and south Asian ethnicities. The increase in hospital discharges coded with congenital cytomegalovirus is most likely due to the introduction of sensitive diagnostic techniques and retrospective diagnoses made in infants after implementation of the NHSP. Public health strategies to improve prevention and treatment of congenital viral infections are urgently warranted. The decrease in discharges for rubella is most likely due to the MMR vaccine. None.

Time-Dependent Effects of Maternal Body Mass Index in Pregnancy on Mental Disorders in Adult Offspring: A Record Linkage Study in Aberdeen, Scotland

Background: Maternal obesity in pregnancy predicts offspring psychopathology risk in childhood but evidence is scarce and conflicting on whether maternal obesity or underweight predict adulthood mental disorders. Methods: We examined longitudinally whether maternal body mass index (BMI) in pregnancy was associated with mental disorders in her offspring born 1950-1999, and whether the associations differed by child birth year, among 68571 mother-child dyads of Aberdeen Maternity and Neonatal Databank, Scotland. Maternal BMI was measured at a mean 15.7 gestational weeks and classified into underweight, normal weight, overweight, moderate obesity and severe obesity. Mental disorders were identified from nationwide health registers carrying diagnoses of all hospitalizations and deaths in Scotland in 1996-2017. Outcomes: Independently of maternal mental disorders and other covariates, maternal BMI was associated with offspring mental disorders in a time-dependent manner. In offspring born 1950-1974, maternal underweight predicted the hazard of mental disorders [Hazard Ratio(HR)=1.74, 95% Confidence Interval(CI)=1.01-3.00). In offspring born 1975-1999, maternal severe obesity predicted increased hazards of any mental disorder (HR=1.60, 95% CI=1.08-2.38), substance use (HR=1.91, 95% CI=1.03-3.57) and schizophrenia spectrum (HR=2.80, 95% CI=1.40-5.63) disorders. Interpretation: Our findings support the developmental origins of health and disease framework by showing time-specific associations between maternal BMI in pregnancy and mental disorders in adult offspring. Considering the marked rise in obesity prevalence, the findings carry important public health implications by underlining the possible effects of maternal nutrition on offspring mental disorders across the lifespan. Funding Statement: Academy of Finland, European Commission EraNet Neuron, University of Helsinki. We acknowledge the support of the British Heart Foundation (RE/18/5/34216) and Tommy’s. Declaration of Interests: None of the authors have any conflicts of interest to declare. Ethics Approval Statement: Ethics and Governance approvals were obtained from the NHS Scotland's Public Benefit and Privacy Panel for Health and Social Care (Reference 516-0594) and the AMND steering committee.

Beta-Blockers for the Secondary Prevention of Myocardial Infarction in People with Dementia: A Systematic Review.

Background:Cardiovascular disease remains the most common cause of death in industrialized countries. The use of beta-blockers is well established as a secondary prevention of myocardial infarction. However, little is known about the benefits of beta-blockers for people living with dementia.Objective:To evaluate the use of beta-blockers in people with dementia who have had a myocardial infarction, in order to identify associations between medication use, mortality, re-infarction and functional decline.Methods:We searched for all studies (randomized trials, observational cohorts) reporting beta-blocker use in populations with both dementia and previous myocardial infarction. Relevant keywords were used in Medline, Embase, and Web of Science up to October 2018. Titles and abstracts were independently screened by two reviewers. Quality of eligible studies was assessed using the Newcastle-Ottawa Scale. PRISMA recommendations were followed throughout.Results:Two observational studies were included, representing 10,992 individuals in a community setting and 129,092 individuals from a hospital record-linkage study. One showed use of beta-blockers reduced all-cause mortality (HR 0.74 (95% CI 0.64– 0.86) alongside evidence for an increased rate of functional decline in individuals aged≥65 with moderate to severe cognitive impairment (OR 1.34 (95% CI 1.11– 1.61)). The second study did not find an association between beta-blocker use and mortality in the population living with dementia.Conclusion:There is insufficient evidence to support use of beta-blockers to persons living with dementia. A single study provides limited evidence that beta-blockers improve survival rates but with associated detrimental effects on functional status in nursing home residents with cognitive impairment. Decisions to continue beta-blockers in persons living with dementia should be made on an individual basis.

Are volunteering and caregiving associated with suicide risk? A Census-based longitudinal study

BackgroundOpposing risks have been identified between different prosocial activities, with volunteering having been linked to better mental health while caregiving has been associated with higher prevalence and incidence of depression. This study explored suicide risk of people engaged in prosocial activities of caregiving and/or volunteering.MethodsA Census-based record linkage study of 1,018,000 people aged 25–74 years (130,816 caregivers; 110,467 volunteers; and 42,099 engaged in both) was undertaken. Caregiving (light: 1–19; intense: ≥20 h/week), volunteering and mental health status were derived from 2011 Census records. Suicide risk (45 months follow-up) was assessed using Cox models adjusted for baseline mental health.ResultsIntense caregiving was associated with worse mental health (ORadj = 1.15: 95%CI = 1.12, 1.18) and volunteering with better mental health (OR = 0.87: 95%CI = 0.84, 0.89). For those engaged in both activities, likelihood of poor mental health was determined by caregiving level. There were 528 suicides during follow-up, with those engaged in both activities having the lowest risk of suicide (HR = 0.34: 95%CI = 0.14, 0.84). Engaging in either volunteering or caregiving was associated with lower suicide risk for those with good mental health at baseline (HR = 0.66: 95%CI = 0.49, 0.88) but not for their peers with baseline poor mental health (HR = 1.02: 95%CI = 0.69, 1.51).ConclusionsAlthough an increased risk of poor mental health was identified amongst caregivers, there was no evidence of an increased risk of suicide.

The incidence of suicide in University students in England and Wales 2000/2001–2016/2017: Record linkage study

BackgroundThere are growing concerns about the mental health and risk of suicide amongst university students. AimTo investigate trends in the incidence and characteristics of university student suicides in England & Wales for the academic years 2000/01 to 2016/17. MethodRecord linkage between Office for National Statistics mortality data and Higher Education Statistics Agency data for England and Wales. Poisson regression and chi-squared tests were used to investigate secular trends and the characteristics of students dying by suicide. ResultsThere were 1330 student suicide deaths from 2000/01 to 2016/17; the annual incidence in 2015/16–2016/17 was 4.7 per 100,000 students. There was evidence of a rise in incidence since 2009/10 (incidence rate ratio per year 1.04 ((95%CI 1.00–1.07) p = 0.029). Incidence in 2012/13 to 2016/17 was less than half the rate in the general population of a similar age. Incidence was higher in males than females and amongst undergraduates vs. postgraduates. There was some evidence of a reduced risk amongst black compared to white students (RR 0.53 (95%CI 0.32–0.88). Incidence was highest in January and lowest during the summer holidays (July–September). LimitationsThere was no age/sex or sociodemographic breakdown of the overall student population for 2000/01 to 2011/12. ConclusionRates of suicide are considerably lower amongst students than the general population. In keeping with trends in young people in the wider population, the incidence of student suicide has increased since 2009/10. To inform prevention, research is needed to understand reasons for the rise in suicide in young people.

OC01.03: State‐wide performance of traditional and cell‐free DNA‐based prenatal testing pathways: the Victorian Perinatal Record Linkage (PeRL) study

OBJECTIVES: Women's choices of prenatal screening and diagnostic pathways have increased in complexity since the introduction of cell-free DNA (cfDNA) screening and chromosomal microarrays. We performed individual record-linkage of women undergoing screening with cfDNA, combined first trimester screening (CFTS), second trimester serum screening (STSS), and/or prenatal and postnatal cytogenetic testing to (i) obtain population-based estimates on women's utilization of screening and diagnosis, (ii) analyse the performance of different screening strategies, and (iii) report the residual risks of any major chromosome abnormality following a low risk aneuploidy screen. METHODS: Retrospective study of women resident in Victoria, Australia, undergoing screening or prenatal diagnosis in 2015. Patient-funded cfDNA referrals from multiple providers were merged with state-wide results for government-subsidized CFTS, STSS and invasive diagnostic procedures. Postnatal cytogenetic results from products of conception and infants up to 12 months of age were obtained to ascertain cases of false negative screening results and atypical chromosome abnormalities. Individual record-linkage was performed with LinkageWizTM and statistical analyses with STATA v14.0. RESULTS: There were 79,140 births during the study period; 66,166 women (83.4%) underwent at least one form of aneuploidy screening. Linkage data were complete for 92.4% of women undergoing screening (n=61,911) and of these, 73.1% (n=45,275) used CFTS alone, 20.2% (n=12,520) used cfDNA alone; 5.3% (n=3268) used STSS alone, 1.3% (n=813) used both CFTS and cfDNA, and < 0.1% (n=35) used both STSS and cfDNA. CFTS had a combined sensitivity for trisomies 21, 13 and 18 of 89.57% (95%CI 82.64-93.93) for a screen positive rate (SPR) of 2.94%. There were 12 false negative results in the CFTS pathway (10 trisomy 21, 1 trisomy 18 and 1 trisomy 13). CfDNA had a combined sensitivity of 100% (95% CI 95.0-100.0) for trisomies 21, 13 and 18 for a 1.21% screen positive rate. When high risk cfDNA results for any chromosome, including the sex chromosomes, and failed cfDNA tests were treated as screen positives, the cfDNA SPR increased to 2.42%. The risk of any major chromosome abnormality (including atypical abnormalities) detected on prenatal or postnatal diagnostic testing after a low risk screening result was 1 in 1188 for CFTS (n=37) and 1 in 762 for cfDNA (n=16) (p= 0.13). The range of chromosome abnormalities detected after low risk cfDNA included pathogenic CNVs (n=6), triploidy (n=3), rare autosomal trisomies (n=3) and monosomy X (n=2). CONCLUSIONS: Our state-wide linkage analysis delineated the utilization and clinical performance of the multitude of screening pathways available to pregnant women. The sensitivity of cfDNA for trisomies 21, 13 and 18 was clearly superior to CFTS. While there was no statistically significant difference in the residual risk of any major chromosome abnormality after low risk CFTS or cfDNA result, there were fewer live infants diagnosed with a major chromosome abnormality in the cfDNA cohort. These data provide valuable population-based evidence to inform practice recommendations and health policy. This article is protected by copyright. All rights reserved.

Unemployment and benefit claims by UK veterans in the new millennium: results from a record linkage study

ObjectivesIn the first study of its kind in the UK, we linked pensions data on benefit claims with data from the King’s Military Cohort to assess uptake of unemployment and...

Early birth is a key factor in educational disadvantage of twins: A data linkage study.

To assess educational outcomes of twins and quantify the degree this is mediated by gestational age and other perinatal factors. We conducted a population-based record-linkage cohort study of all live births ≥24weeks gestation in New South Wales, Australia with a corresponding standardised school test result for grade 3 in 2008-2014. The primary outcome was whether a child met the National Minimum Standard (NMS) cut-off in literacy and numeracy domains. Robust multivariable Poisson models were used to obtain adjusted relative risks (aRRs), and mediation analysis conducted to assess contributing factors on the causal pathway. Of 351791 liveborn infants, 10365 (2.9%) were twins. After adjusting for maternal covariates and compared with singletons, twins had an increased risk of not meeting the NMS for all five literacy and numeracy domains (aRR 1.27-1.45, P<.001). Gestational age alone mediated up to 73% of aRRs and small for gestational age further attenuated these effects with only minimal risk remaining after adjusting for all mediators (aRR 0.94-1.07). Almost all of the educational disadvantage experienced by twins, compared with singletons, is attributable to the risk associated with shorter gestational age, and partly by poor foetal growth. These findings support efforts to prolong gestation of twin pregnancies.

Mortality among Italian male workers in the construction industry: a census-based cohort study.

Advances in technologies, occupational hygiene and increased surveillance have reduced the excess mortality previously found in the construction industry. This study is aimed to evaluate cause-specific mortality in a recent cohort of construction workers. We carried out a record-linkage cohort study based on the 2011 Italian census and the mortality archives (2012-2015), including 1 068 653 construction workers. We estimated mortality rate ratios (MRR) using Poisson regression models including terms for age and geographic area. Compared with non-manual workers, construction workers showed an excess mortality from all causes (MRR: 1.34), all neoplasms (MRR: 1.30), head and neck (MRR: 2.05), stomach (MRR: 1.56), liver (MRR: 1.62), lung (MRR: 1.80), prostate (MRR: 1.24) and bladder (MRR: 1.60) cancers, respiratory (MRR: 1.41) and liver (MRR: 1.79) diseases, all external causes (MRR: 1.87), falls (MRR: 2.87) and suicide (MRR: 1.58). Compared with manual workers in other industries, construction workers showed excess mortality from prostate (MRR: 1.27) and non-melanoma skin cancers (MRR: 1.95), all external causes (MRR: 1.14), falls (MRR: 1.94) and suicide (MRR: 1.18). Most of this excess mortality disappeared after adjusting for education, with the exception of prostate and non-melanoma skin cancers, all external causes, falls and suicide. Construction workers are at high risk of dying from external causes, while the excess mortality found for several cancers, liver and respiratory diseases may be at least partially due to the high prevalence of low education and unfavorable lifestyle factors. The excess mortality from prostate cancer requires further evaluations.

Impact of chronic health conditions and injury on school performance and health outcomes in New South Wales, Australia: a retrospective record linkage study protocol

IntroductionChildren who have sustained a serious injury or who have a chronic health condition, such as diabetes or epilepsy, may have their school performance adversely impacted by the condition, treatment...

Drug-related mortality after discharge from treatment: A record-linkage study of substance abuse clients in Texas, 2006–2012

BackgroundPatients have higher mortality immediately after substance abuse treatment discharge, but there are few data on post-discharge mortality differences across treatment modalities. MethodsA retrospective cohort study examined individuals discharged from substance abuse treatment during 2006–2012 and probabilistically matched treatment records to death records. Logistic regression examined associations between drug-related death (DRD) and demographics; route, frequency, and classes of drugs abused; and treatment. Primary outcome was DRD during post-discharge days 0–28; secondary outcomes examined DRD during days 29–90 and 91–365. ResultsWe examined 178,749 patients discharged from 254,814 treatment episodes. There were 97 DRD during days 0–28 (4.1/1000 person-years), 115 DRD during days 29–90 (2.6/1000 person-years; IRR 0.6 [95% CI 0.5–0.8]), and 293 DRD during days 91–365 (1.9/1000 person-years; IRR 0.5 [0.4–0.6]). Higher 28-day DRD was associated with abuse of opioids (aOR 2.5 [1.4–4.4]), depressants (aOR 2.0 [1.2–3.4]), or alcohol (aOR 1.7 [1.1–2.6]); and opioid injection (aOR 2.2 [1.3–3.7]). Lower DRD was associated with treatment completion (aOR 0.6 [0.4–0.9]), female sex (aOR 0.6 [0.4–0.8]), and employment (aOR 0.5 [0.3–0.9]). Among all patients, DRD rates were higher following residential (IRR 2.6, [1.6–4.2]) and detoxification (IRR 2.9, [1.7–4.9]) treatment compared to outpatient. Patients with prior opioid abuse had higher 28-day DRD after outpatient (6.7/1000 person-years; IRR 4.1 [1.8–9.1]), residential (13.6/1000 person-years; IRR 4.2 [2.2–8.2]), and detoxification (8.8/1000 person-years; IRR 3.2 [1.2, 8.5]) compared to those without. ConclusionsDrug-related mortality is highest during days 0–28 after discharge, especially following residential and detoxification treatment. Opioid abuse is strongly associated with early post-discharge mortality.

Effectiveness of pneumococcal conjugate vaccine against hospital admissions for pneumonia in Australian children: a retrospective, population-based, record-linked cohort study

Reductions in pneumonia hospitalisations following introduction of pneumococcal conjugate vaccines (PCVs) have been reported from high-incidence and low-incidence settings but long-term data comparing vaccinated with unvaccinated children are sparse. We did a retrospective, population-based, record-linkage cohort study in Australian children using administrative health data from the Western Australian Midwives' Notification System and New South Wales Perinatal Data Collection, and the birth and death registries in both states. PCV vaccination details, pneumonia-coded hospital admissions, and invasive pneumococcal disease notification records were individually linked for children born between 2001 and 2012. The primary outcome was defined as the first hospital admission for all-cause pneumonia. Cox models were used to calculate adjusted hazard ratios (HR) to estimate the effect of PCV doses on pneumonia-coded hospital admissions by Aboriginal status, birth period, remoteness, and pneumonia diagnostic category in children younger than 2 years. Person-time of follow-up time for each child started at birth and was censored at the earliest of first hospital admission for all-cause pneumonia, death, invalid PCV dose, when the child reached age 24 months, or the end date of the study period (Dec 31, 2013) FINDINGS: The study cohort comprised 1 365 893 children liveborn between Jan 1, 2001, and Dec 31, 2012, of whom 66 484 (4·9%) were identified as Aboriginal. The overall rate for all-cause pneumonia hospital admissions for children younger than 2 years over the entire study period was 17·6/1000 child-years in Aboriginal children and 5·5/1000 child-years in non-Aboriginal children. Compared with children born between 2001 and 2004 (ie, the pre-universal PCV period), the incidence of pneumonia-coded hospital admissions decreased in both vaccinated (6·5 vs 5·7 per 1000 child-years [12% reduction, 95% CI 3-21; p=0·01]) and unvaccinated non-Aboriginal children (6·8 vs 3·7 [45% reduction; 41-49]) born 2005-12 (the universal PCV period); among Aboriginal children, declines were significant only among those vaccinated (27·4 vs 14·1 [49% reduction, 40-55]). Among Aboriginal children born 2005-12, the risk of pneumonia-coded hospital admission after three doses of PCV was lower than those unvaccinated (adjusted HR 0·83, 95% CI 0·65-0·99) but, among non-Aboriginal children, the risk was similar (adjusted HR 1·09, 0·98-1·22). Overall, remote-born Aboriginal children had the highest incidence of hospital admission for pneumonia and among children born 2005-12, the adjusted risk was 37% lower (adjusted HR 0·63, 95% CI 0·42-0·96) among those fully vaccinated than those unvaccinated. Reductions in pneumonia-coded hospital admissions in unvaccinated children predominated in non-Aboriginal children with low incidence of pneumonia but were not significant in Aboriginal children with high incidence. These findings have potential implications for measuring PCV effect using a non-specific endpoint such as all-cause pneumonia in high-incidence populations. Commonwealth Government Collaborative Research Infrastructure Strategy and Education Investment Fund Super Science Initiative and the Australian National Health and Medical Research Council.