Using administrative data to better treat chronic limb threatening ischemia

Chronic limb threatening ischemia (CLTI) is the most severe manifestation of peripheral arterial disease represents a particularly high-risk subgroup of patients. As such, efforts to better understand this complex patient population through well-designed clinical research studies are critical to improving CLTI care. Prospective randomized clinical trials (RCTs) remain the gold standard in clinical research, but these trials are resource-intensive and have highly selective patient populations, which limit their feasibility and generalizability. Alternatively, retrospective studies are less expensive than RCTs, have a larger sample size, and are more generalizable owing to a broader patient population. Health care administrative data provide rich sources of information that may be used for research purposes and are increasingly being used for the study of vascular surgery conditions, including CLTI. Although administrative data are collected for billing purposes, they may be leveraged to study a broad range of topics in vascular surgery including those related to health care delivery, epidemiology, health disparities, and outcomes. This review provides an overview of administrative data available for CLTI research, the strengths and limitations of these data sources, current areas of investigation, and future opportunities for further study with the goal of improving outcomes in this high-risk population.

Ambulance Services Attendance for Mental Health and Overdose Before and During COVID-19 in Canada and the United Kingdom: Interrupted Time Series Study

BACKGROUND

The COVID-19 pandemic impacted mental health and health care systems worldwide.

OBJECTIVE

This study examined the COVID-19 pandemic's impact on ambulance attendances for mental health and overdose, comparing similar regions in the United Kingdom and Canada that implemented different public health measures.

METHODS

An interrupted time series study of ambulance attendances was conducted for mental health and overdose in the United Kingdom (East Midlands region) and Canada (Hamilton and Niagara regions). Data were obtained from 182,497 ambulance attendance records for the study period of December 29, 2019, to August 1, 2020. Negative binomial regressions modeled the count of attendances per week per 100,000 population in the weeks leading up to the lockdown, the week the lockdown was initiated, and the weeks following the lockdown. Stratified analyses were conducted by sex and age.

RESULTS

Ambulance attendances for mental health and overdose had very small week-over-week increases prior to lockdown (United Kingdom: incidence rate ratio [IRR] 1.002, 95% CI 1.002-1.003 for mental health). However, substantial changes were observed at the time of lockdown; while there was a statistically significant drop in the rate of overdose attendances in the study regions of both countries (United Kingdom: IRR 0.573, 95% CI 0.518-0.635 and Canada: IRR 0.743, 95% CI 0.602-0.917), the rate of mental health attendances increased in the UK region only (United Kingdom: IRR 1.125, 95% CI 1.031-1.227 and Canada: IRR 0.922, 95% CI 0.794-1.071). Different trends were observed based on sex and age categories within and between study regions.

CONCLUSIONS

The observed changes in ambulance attendances for mental health and overdose at the time of lockdown differed between the UK and Canada study regions. These results may inform future pandemic planning and further research on the public health measures that may explain observed regional differences.

A scoping review of the methodological approaches used in retrospective chart reviews to validate adverse event rates in administrative data

Patient safety is a key quality issue for health systems. Healthcare acquired adverse events (AEs) compromise safety and quality; therefore, their reporting and monitoring is a patient safety priority. Although administrative datasets are potentially efficient tools for monitoring rates of AEs, concerns remain over the accuracy of their data. Chart review validation studies are required to explore the potential of administrative data to inform research and health policy. This review aims to present an overview of the methodological approaches and strategies used to validate rates of AEs in administrative data through chart review. This review was conducted in line with the Joanna Briggs Institute methodological framework for scoping reviews. Through database searches, 1054 sources were identified, imported into Covidence, and screened against the inclusion criteria. Articles that validated rates of AEs in administrative data through chart review were included. Data were extracted, exported to Microsoft Excel, arranged into a charting table, and presented in a tabular and descriptive format. Fifty-six studies were included. Most sources reported on surgical AEs; however, other medical specialties were also explored. Chart reviews were used in all studies; however, few agreed on terminology for the study design. Various methodological approaches and sampling strategies were used. Some studies used the Global Trigger Tool, a two-stage chart review method, whilst others used alternative single-, two-stage, or unclear approaches. The sources used samples of flagged charts (n = 24), flagged and random charts (n = 11), and random charts (n = 21). Most studies reported poor or moderate accuracy of AE rates. Some studies reported good accuracy of AE recording which highlights the potential of using administrative data for research purposes. This review highlights the potential for administrative data to provide information on AE rates and improve patient safety and healthcare quality. Nonetheless, further work is warranted to ensure that administrative data are accurate. The variation of methodological approaches taken, and sampling techniques used demonstrate a lack of consensus on best practice; therefore, further clarity and consensus are necessary to develop a more systematic approach to chart reviewing.

Validating methods used to identify non-adherence adverse drug events in Canadian administrative health data

AIMS

Medication non-adherence is a type of adverse drug event that can lead to untreated and exacerbated chronic illness, and that drives healthcare utilization. Research using medication claims data has attempted to identify instances of medication non-adherence using the proportion of days covered or by examining gaps between medication refills. We sought to validate these measures compared to a gold standard diagnosis of non-adherence made in hospital.

METHODS

This was a retrospective analysis of adverse drug events diagnosed during three prospective cohorts in British Columbia between 2008 and 2015 (n = 976). We linked prospectively identified adverse drug events to medication claims data to examine the sensitivity and specificity of typical non-adherence measures.

RESULTS

The sensitivity of the non-adherence measures ranged from 22.4% to 37.5%, with a proportion of days covered threshold of 95% performing the best; the non-persistence measures had sensitivities ranging from 10.4% to 58.3%. While a 7-day gap was most sensitive, it classified 61.2% of the sample as non-adherent, whereas only 19.6% were diagnosed as such in hospital.

CONCLUSIONS

The methods used to identify non-adherence in administrative databases are not accurate when compared to a gold standard diagnosis by healthcare providers. Research that has relied on administrative data to identify non-adherent patients both underestimates the magnitude of the problem and may label patients as non-adherent who were in fact adherent.

Basal insulin initiation in adults with type 2 diabetes mellitus: A retrospective cohort study using administrative health data in Alberta, Canada

BACKGROUND

Pharmacological treatment of type 2 diabetes mellitus (T2DM) follows a stepwise approach. Typically, metformin monotherapy is first-line treatment, followed by other non-insulin anti-hyperglycemic agents (NIAHAs) or progression to insulin if hemoglobin A1c (A1C) targets are not achieved. We aimed to describe real-world patterns of basal insulin initiation in people with T2DM and A1C not at target despite treatment with ≥2 NIAHAs.

METHODS

A retrospective cohort study was conducted using administrative health data from Alberta, Canada among adults with T2DM, indexed on the first test with 7.0%

RESULTS

The cohort included 14,083 individuals. The KM cumulative probability of initiating basal insulin was 7.7% (95% CI: 7.3-8.2%) at 1 year, increasing to 43.1% (95% CI: 42.1-44.1%) at 8 years of follow-up. Higher A1C levels were associated with greater proportions of basal insulin initiation. By year 8, proportions with NIAHA intensification and clinical inertia were 12.1% and 19.3%, respectively, relative to year 7.

CONCLUSION

Despite current clinical practice guidelines recommending achieving A1C targets within six months, less than half of individuals with T2DM and clear indications for basal insulin initiated treatment within 8 years. Efforts to reduce delays in basal insulin initiation are required.

Collapse

Key Words

• Type 2 diabetes
• administrative data
• basal insulin
• retrospective cohort

Collapse

MESH Headings Collapse

Grants Collapse

Affiliation(s)

Peter Senior Alberta Diabetes Institute - Faculty of Medicine & Dentistry, University of Alberta, Edmonton, Alberta, Canada.
Jina Hahn Novo Nordisk Canada Inc., Mississauga, Ontario, Canada.
Godfrey Mau Novo Nordisk Canada Inc., Mississauga, Ontario, Canada.
Phongsack Manivong Medlior Health Outcomes Research Ltd., Calgary, Alberta, Canada.
Eileen Shaw Medlior Health Outcomes Research Ltd., Calgary, Alberta, Canada.

Collapse

Validating use of diagnostic codes in Canadian administrative data for identification of adverse drug events

AIMS

While diagnostic codes from administrative health data might be a valuable source to identify adverse drug events (ADEs), their ability to identify unintended harms remains unclear. We validated claims-based diagnosis codes for ADEs based on events identified in a prospective cohort study and assessed whether key attributes predicted their documentation in administrative data.

METHODS

This was a retrospective analysis of 3 prospective cohorts in British Columbia, from 2008 to 2015 (n = 13 969). We linked prospectively identified ADEs to administrative insurance data to examine the sensitivity and specificity of different diagnostic code schemes. We used logistic regression to assess which key attributes (e.g., type of event, symptoms and culprit medications) were associated with better documentation of ADEs in administrative data.

RESULTS

Among 1178 diagnosed events, the sensitivity of the diagnostic codes in administrative data ranged from 3.4 to 52.6%, depending on the database and codes used. We found that documentation was worse for certain types of ADEs (dose-related: odds ratio [OR]: 0.32, 95% confidence interval [CI]: 0.15, 0.69; nonadherence events (OR: 0.35, 95% CI: 0.20, 0.62), and better for those experiencing arrhythmias (OR: 4.19, 95% CI: 0.96, 18.28).

CONCLUSION

ADEs were not well documented in administrative data. Alternative methods should be explored to capture ADEs for health research.

Validity of Using Japanese Administrative Data to Identify Inpatients With Acute Pulmonary Embolism: Referencing the COMMAND VTE Registry

BACKGROUND

Acute pulmonary embolism (PE) is a life-threatening in-hospital complication. Recently, several studies have reported the clinical characteristics of PE among Japanese patients using the diagnostic procedure combination (DPC)/per diem payment system database. However, the validity of PE identification algorithms for Japanese administrative data is not yet clear. The purpose of this study was to evaluate the validity of using DPC data to identify acute PE inpatients.

METHODS

The reference standard was symptomatic/asymptomatic PE patients included in the COntemporary ManageMent AND outcomes in patients with Venous ThromboEmbolism (COMMAND VTE) registry, which is a cohort study of acute symptomatic venous thromboembolism (VTE) patients in Japan. The validation cohort included all patients discharged from the six hospitals included in both the registry and DPC database. The identification algorithms comprised diagnosis, anticoagulation therapy, thrombolysis therapy, and inferior vena cava filter placement. Each algorithm's sensitivity, specificity, positive predictive value (PPV), and negative predictive value (NPV) were estimated.

RESULTS

A total of 43.4% of the validation cohort was female, with a mean age of 67.3 years. The diagnosis-based algorithm showed a sensitivity of 90.2% (222/246; 95% confidence interval [CI], 85.8-93.6%), a specificity of 99.8% (228,485/229,027; 95% CI, 99.7-99.8%), a PPV of 29.1% (222/764; 95% CI, 25.9-32.4%) and an NPV of 99.9% (228,485/229,509; 95% CI, 99.9-99.9%) for identifying symptomatic/asymptomatic PE. Additionally, 94.6% (159/168; 95% CI, 90.1-97.5%) of symptomatic PE patients were identified using the diagnosis-based algorithm.

CONCLUSION

The diagnosis-based algorithm may be a relatively sensitive method for identifying acute PE inpatients in the Japanese DPC database.

Using unsupervised clustering approaches to identify common mental health profiles and associated mental healthcare service use patterns in Ontario, Canada

Mental health is a complex, multidimensional concept that goes beyond clinical diagnoses, including psychological distress, life stress and well-being. This study aims to use unsupervised clustering approaches to identify multidimensional mental health profiles that exist in the population, and their associated service use patterns. The data source for this study is the 2012 Canadian Community Health Survey- Mental Health linked to administrative healthcare data holdings, included were all Ontario adult respondents. We used a Partioning Around Medoids clustering algorithm with Gower's proximity to identify groups with distinct combinations of mental health indicators and described them by their sociodemographic and service use characteristics. We identified four groups with distinct mental health profiles, including one group who met the clinical threshold for a depressive diagnosis, with the remaining three groups expressing differences in positive mental health, life stress and self-rated mental health. The four groups had different age, employment and income profiles and exhibited differential access to mental healthcare services. This study represents the first step in identifying complex profiles of mental health at the population level in Ontario, Canada. Further research is required to better understand the potential causes and consequences of belonging to each of the mental health profiles identified.

Prevalence of Adult Type 2 Diabetes Mellitus and Related Complications in Alberta, Canada: A Retrospective, Observational Study Using Administrative Data

OBJECTIVES

Type 2 diabetes mellitus (T2DM) is a prevalent chronic disease and a leading cause of morbidity/mortality in Canada. We evaluated the burden of T2DM in Alberta, Canada, by estimating the 5-year period prevalence of T2DM and rates of comorbidities and complications/conditions after T2DM.

METHODS

We conducted a population-based, retrospective study linking administrative health databases. Individuals with T2DM (≥18 years of age) were identified between 2008-2009 and 2018-2019 using a published algorithm, with follow-up data to March 2020. The 5-year period prevalence was estimated for 2014-2015 to 2018-2019. Individuals with newly identified T2DM, ascertained between 2010-2011 and 2017-2018 with a lookback period between 2008-2009 and 2009-2010 and a minimum 1 year of follow-up data, were evaluated for subsequent cardiovascular, diabetic, renal, and other complication/condition frequencies (%) and rates (per 100 person-years). Complications/conditions were stratified by atherosclerotic cardiovascular disease (ASCVD) status at index and age.

RESULTS

The 5-year period prevalence of T2DM was 11,051 per 100,000 persons, with the highest prevalence in men 65 to <75 years of age. There were 195,102 individuals included in the cohort (mean age 56.7±14.7 years). The most frequently reported complications/conditions (rates per 100 person-years) were acute infection (23.10, 95% confidence interval [CI] 23.00 to 23.30), hypertension (17.30, 95% CI 16.80 to 17.70), and dyslipidemia (12.20, 95% CI 11.90 to 12.40). Individuals who had an ASCVD event/procedure and those ≥75 years of age had higher rates of complications/conditions.

CONCLUSIONS

We found that over half of the individuals had hypertension or infection after T2DM. Also, those with ASCVD had higher rates of complications/conditions. Strategies to mitigate complications/conditions after T2DM are required to reduce the burden of this disease on individuals and health-care systems.

Longitudinal trajectories of substance use disorder treatment use: A latent class growth analysis using a national cohort in Chile

BACKGROUND AND AIMS

Longitudinal studies have revealed that substance use treatment use is often recurrent among patients; the longitudinal patterns and characteristics of those treatment trajectories have received less attention, particularly in the global south. This study aimed to disentangle heterogeneity in treatment use among adult patients in Chile by identifying distinct treatment trajectory groups and factors associated with them.

DESIGN

National-level registry-based retrospective cohort.

SETTING AND PARTICIPANTS

Adults admitted to publicly funded substance use disorder treatment programs in Chile from November 2009 to November 2010 and followed for 9 years (n = 6266).

MEASUREMENTS

Monthly treatment use; type of treatment; ownership of the treatment center; discharge status; primary substance used; sociodemographic.

FINDINGS

A seven-class treatment trajectory solution was chosen using latent class growth analysis. We identified three trajectory groups that did not recur and had different treatment lengths: Early discontinuation (32%), Less than a year in treatment (19.7%) and Year-long episode, without recurrence (12.3%). We also identified a mixed trajectory group that had a long first treatment or two treatment episodes with a brief time between treatments: Long first treatment, or immediate recurrence (6.3%), and three recurrent treatment trajectory groups: Recurrent and decreasing (14.2%), Early discontinuation with recurrence (9.9%) and Recurrent after long between treatments period (5.7%). Inpatient or outpatient high intensity (vs. outpatient low intensity) at first entry increased the odds of being in the longer one-episode groups compared with the Early discontinuation group. Women had increased odds of belonging to all the recurrent groups. Using cocaine paste (vs. alcohol) as a primary substance decreased the odds of belonging to long one-episode groups.

CONCLUSIONS

In Chile, people in publicly funded treatment for substance use disorder show seven distinct care trajectories: three groups with different treatment lengths and no recurring episodes, a mixed group with a long first treatment or two treatment episodes with a short between-treatment-episodes period and three recurrent treatment groups.

Utilization of broad- versus narrow-spectrum antibiotics for the treatment of outpatient community-acquired pneumonia among adults in the United States

PURPOSE

To characterize antibiotic utilization for outpatient community-acquired pneumonia (CAP) in the United States.

METHODS

We conducted a cohort study among adults 18-64 years diagnosed with outpatient CAP and a same-day guideline-recommended oral antibiotic fill in the MarketScan® Commercial Database (2008-2019). We excluded patients coded for chronic lung disease or immunosuppressive disease; recent hospitalization or frequent healthcare exposure (e.g., home wound care, patients with cancer); recent antibiotics; or recent infection. We characterized utilization of broad-spectrum antibiotics (respiratory fluoroquinolone, β-lactam + macrolide, β-lactam + doxycycline) versus narrow-spectrum antibiotics (macrolide, doxycycline) overall and by patient- and provider-level characteristics. Per 2007 IDSA/ATS guidelines, we stratified analyses by otherwise healthy patients and patients with comorbidities (coded for diabetes; chronic heart, liver, or renal disease; etc.).

RESULTS

Among 263 914 otherwise healthy CAP patients, 35% received broad-spectrum antibiotics (not recommended); among 37 161 CAP patients with comorbidities, 44% received broad-spectrum antibiotics (recommended). Ten-day antibiotic treatment durations were the most common for all antibiotic classes except macrolides. From 2008 to 2019, broad-spectrum antibiotic use substantially decreased from 45% to 19% in otherwise healthy patients (average annual percentage change [AAPC], -7.5% [95% CI -9.2%, -5.9%]), and from 55% to 29% in patients with comorbidities (AAPC, -5.8% [95% CI -8.8%, -2.6%]). In subgroup analyses, broad-spectrum antibiotic use varied by age, geographic region, provider specialty, and provider location.

CONCLUSIONS

Real-world use of broad-spectrum antibiotics for outpatient CAP declined over time but remained common, irrespective of comorbidity status. Prolonged duration of therapy was common. Antimicrobial stewardship is needed to aid selection according to comorbidity status and to promote shorter courses.

Defining a low-risk birth cohort: a cohort study comparing two perinatal data sets in Ontario, Canada

Introduction

There are two main data sources for perinatal data in Ontario, Canada: the BORN BIS and CIHI-DAD. Such databases are used for perinatal health surveillance and research, and to guide health care related decisions.

Objectives

Our primary objective was to examine the level of agreement between the BIS and CIHI-DAD. Our secondary objectives were to identify the differences between the data sources when identifying a low-risk birth (LRB) cohort and to understand their implications.

Methods

We conducted a population-based cohort study comparing characteristics and clinical outcomes of all linkable births in BIS and CIHI-DAD between 1st April 2012 and 31st March 2018. We excluded out-of-hospital births, those with invalid healthcare numbers, non-Ontario residents and gestational age <20 weeks. We compared the portion of the cohort that met the criteria of a provincial definition of LRB based on each data source and compared clinical outcomes between the groups.

Results

During the study period, 779,979 eligible births were linkable between the two data sources. After applying the LRB exclusions, there were 129,908 cases in the BIS and 136,184 cases in CIHI-DAD. Most exclusion criteria had almost perfect, substantial or moderate agreement. The agreement for non-cephalic presentation and BMI ≥ 40 kg/m2 (kappa coefficients 0.409 and 0.256, respectively) was fair. Comparison between the two LRB cohorts identified differences in the prevalence of cesarean (14.3% BIS versus 12.0% CIHI-DAD) and NICU admission (8.7% BIS versus 7.5% CIHI-DAD) and only 0.01% difference in the prevalence of ICU admission.

Conclusions

Overall, we found high levels of agreement between the BIS and CIHI-DAD. Identifying a LRB cohort in either database may be appropriate, with the caveat of appropriate understanding of the collection, coding and definition of certain outcomes. The decision for selecting a database may depend on which variables are most important in a particular analysis.

Applying machine learning to identify pediatric patients with newly diagnosed acute lymphoblastic leukemia using administrative data

Case identification in administrative databases is challenging as diagnosis codes alone are not adequate for case ascertainment. We utilized machine learning (ML) to efficiently identify pediatric patients with newly diagnosed acute lymphoblastic leukemia. We tested nine ML models and validated the best model internally and externally. The optimal model had 97% positive predictive value (PPV) and 99% sensitivity in internal validation; 94% PPV and 82% sensitivity in external validation. Our ML model identified a large cohort of 21,044 patients, demonstrating an efficient approach for cohort assembly and enhancing the usability of administrative data.

Deprivation effects on length of stay and death of hospitalised COVID-19 patients in Greater Manchester

Introduction

The World Health Organisation declared a global pandemic in March 2020. The impact of COVID-19 has not been felt equally by all regions and sections of society. The extent to which socio-demographic and deprivation factors have adversely impacted on outcomes is of concern to those looking to 'level-up' and decrease widening health inequalities.

Objectives

In this paper we investigate the impact of deprivation on the outcomes for hospitalised COVID-19 patients in Greater Manchester during the first wave of the pandemic in the UK (30/12/19-2/1/21), controlling for proven risk factors from elsewhere in the literature.

Methods

We fitted Negative Binomial and logistic regression models to NHS administrative data to investigate death from COVID in hospital and length of stay for surviving patients in a sample of adult patients admitted within Greater Manchester (N = 10,372, spell admission start dates from 30/12/2019 to 02/01/2021 inclusive).

Results

Deprivation was associated with death risk for hospitalised patients but not with length of stay. Male sex, co-morbidities and older age was associated with higher death risk. Male sex and co-morbidities were associated with increased length of stay. Black and other ethnicities stayed longer in hospital than White and Asian patients. Period effects were detected in both models with death risk reducing over time, but the length of stay increasing.

Conclusion

Deprivation is important for death risk; however, the picture is complex, and the results of this analysis suggest that the reported COVID related mortality and deprivation linked reductions in life expectancy, may have occurred in the community, rather than in acute settings.

Highlights

Older age and male sex are predictive of longer hospital stays and higher death risk for hospitalised cases in this analysis.Deprivation is associated with death risk but not length of stay for hospitalised patients.Ethnicity is associated with length of stay, but not with death risk.There is a social gradient in health, but these data would suggest that once in the care of an NHS hospital in an acute health episode, outcomes are more equal.

Cohort profile: the British Columbia COVID-19 Cohort (BCC19C)-a dynamic, linked population-based cohort

Purpose

The British Columbia COVID-19 Cohort (BCC19C) was developed from an innovative, dynamic surveillance platform and is accessed/analyzed through a cloud-based environment. The platform integrates recently developed provincial COVID-19 datasets (refreshed daily) with existing administrative holdings and provincial registries (refreshed weekly/monthly). The platform/cohort were established to inform the COVID-19 response in near "real-time" and to answer more in-depth epidemiologic questions.

Participants

The surveillance platform facilitates the creation of large, up-to-date analytic cohorts of people accessing COVID-19 related services and their linked medical histories. The program of work focused on creating/analyzing these cohorts is referred to as the BCC19C. The administrative/registry datasets integrated within the platform are not specific to COVID-19 and allow for selection of "control" individuals who have not accessed COVID-19 services.

Findings to date

The platform has vastly broadened the range of COVID-19 analyses possible, and outputs from BCC19C analyses have been used to create dashboards, support routine reporting and contribute to the peer-reviewed literature. Published manuscripts (total of 15 as of July, 2023) have appeared in high-profile publications, generated significant media attention and informed policy and programming. In this paper, we conducted an analysis to identify sociodemographic and health characteristics associated with receiving SARS-CoV-2 laboratory testing, testing positive, and being fully vaccinated. Other published analyses have compared the relative clinical severity of different variants of concern; quantified the high "real-world" effectiveness of vaccines in addition to the higher risk of myocarditis among younger males following a 2nd dose of an mRNA vaccine; developed and validated an algorithm for identifying long-COVID patients in administrative data; identified a higher rate of diabetes and healthcare utilization among people with long-COVID; and measured the impact of the pandemic on mental health, among other analyses.

Future plans

While the global COVID-19 health emergency has ended, our program of work remains robust. We plan to integrate additional datasets into the surveillance platform to further improve and expand covariate measurement and scope of analyses. Our analyses continue to focus on retrospective studies of various aspects of the COVID-19 pandemic, as well as prospective assessment of post-acute COVID-19 conditions and other impacts of the pandemic.

Variation in colorectal cancer treatment and outcomes in Scotland: real world evidence from national linked administrative health data

Background

Colorectal cancer (CRC) is the fourth most common type of cancer in the United Kingdom and the second leading cause of cancer death. Despite improvements in CRC survival over time, Scotland lags behind its UK and European counterparts. In this study, we carry out an exploratory analysis which aims to provide contemporary, population level evidence on CRC treatment and survival in Scotland.

Methods

We conducted a retrospective population-based analysis of adults with incident CRC registered on the Scottish Cancer Registry (Scottish Morbidity Record 06 (SMR06)) between January 2006 and December 2018. The CRC cohort was linked to hospital inpatient (SMR01) and National Records of Scotland (NRS) deaths records allowing a description of their demographic, diagnostic and treatment characteristics. Cox proportional hazards regression models were used to explore the demographic and clinical factors associated with all-cause mortality and CRC specific mortality after adjusting for patient and tumour characteristics among people identified as early-stage and treated with surgery.

Results

Overall, 32,691 (73%) and 12,184 (27%) patients had a diagnosis of colon and rectal cancer respectively, of whom 55% and 53% were early-stage and treated with surgery. Five year overall survival (CRC specific survival) within this cohort was 72% (82%) and 76% (84%) for patients with colon and rectal cancer respectively. Cox proportional hazards models revealed significant variation in mortality by sex, area-based deprivation and geographic location.

Conclusions

In a Scottish population of patients with early-stage CRC treated with surgery, there was significant variation in risk of death, even after accounting for clinical factors and patient characteristics.

A classification system for youth outpatient behavioral health services billed to medicaid

Rates of youth behavioral health concerns have been steadily rising. Administrative data can be used to study behavioral health service utilization among youth, but current methods that rely on identifying an associated behavioral health diagnosis or provider specialty are limited. We reviewed all procedure codes billed to Medicaid for youth in one U.S. county over a 10-year period. We identified 158 outpatient behavioral health procedure codes and classified them according to service type. This classification system can be used by health services researchers to better characterize youth behavioral health service utilization.

The use of linked administrative data in Australian randomised controlled trials: A scoping review

BACKGROUND/AIMS

The demand for simplified data collection within trials to increase efficiency and reduce costs has led to broader interest in repurposing routinely collected administrative data for use in clinical trials research. The aim of this scoping review is to describe how and why administrative data have been used in Australian randomised controlled trial conduct and analyses, specifically the advantages and limitations of their use as well as barriers and enablers to accessing administrative data for use alongside randomised controlled trials.

METHODS

Databases were searched to November 2022. Randomised controlled trials were included if they accessed one or more Australian administrative data sets, where some or all trial participants were enrolled in Australia, and where the article was published between January 2000 and November 2022. Titles and abstracts were independently screened by two reviewers, and the full texts of selected studies were assessed against the eligibility criteria by two independent reviewers. Data were extracted from included articles by two reviewers using a data extraction tool.

RESULTS

Forty-one articles from 36 randomised controlled trials were included. Trial characteristics, including the sample size, disease area, population, and intervention, were varied; however, randomised controlled trials most commonly linked to government reimbursed claims data sets, hospital admissions data sets and birth/death registries, and the most common reason for linkage was to ascertain disease outcomes or survival status, and to track health service use. The majority of randomised controlled trials were able to achieve linkage in over 90% of trial participants; however, consent and participant withdrawals were common limitations to participant linkage. Reported advantages were the reliability and accuracy of the data, the ease of long term follow-up, and the use of established data linkage units. Common reported limitations were locating participants who had moved outside the jurisdictional area, missing data where consent was not provided, and unavailability of certain healthcare data.

CONCLUSIONS

As linked administrative data are not intended for research purposes, detailed knowledge of the data sets is required by researchers, and the time delay in receiving the data is viewed as a barrier to its use. The lack of access to primary care data sets is viewed as a barrier to administrative data use; however, work to expand the number of healthcare data sets that can be linked has made it easier for researchers to access and use these data, which may have implications on how randomised controlled trials will be run in future.

Intergenerational (Dis)continuity of Child Maltreatment: Variation by Parents' Childhood Victimization Experiences and Sex

This longitudinal population-based study examines the association between maltreatment victimization experiences and the likelihood of intergenerational (dis)continuity of maltreatment. Our data include all individuals born in 1983/1984 in Queensland (QLD), Australia who are registered as parents via birth records and who experienced system contacts for maltreatment victimization in childhood (n = 2906). Child safety data on system contacts as a child victim and person responsible for harm to a child were obtained from the Department of Children, Youth Justice and Multicultural Affairs. Out-of-home care experiences and maltreatment frequency, timing, and type were examined. Results indicated that childhood maltreatment experiences significantly differed between parents who were not subsequently responsible for harm to a child (cycle breakers) and parents who were subsequently responsible for harm to a child (cycle maintainers). Different patterns of association were observed across sex. These findings highlight the importance of recognizing the heterogeneity of victim maltreatment experiences and associated risk of maltreatment for their children, and can inform effective and targeted interventions by tailoring these by sex and developmental period.

Using the Functional Comorbidity Index with administrative workers' compensation data: Utility, validity, and caveats

BACKGROUND

Chronic health conditions impact worker outcomes but are challenging to measure using administrative workers' compensation (WC) data. The Functional Comorbidity Index (FCI) was developed to predict functional outcomes in community-based adult populations, but has not been validated for WC settings. We assessed a WC-based FCI (additive index of 18 conditions) for identifying chronic conditions and predicting work outcomes.

METHODS

WC data were linked to a prospective survey in Ohio (N = 512) and Washington (N = 2,839). Workers were interviewed 6 weeks and 6 months after work-related injury. Observed prevalence and concordance were calculated; survey data provided the reference standard for WC data. Predictive validity and utility for control of confounding were assessed using 6-month work-related outcomes.

RESULTS

The WC-based FCI had high specificity but low sensitivity and was weakly associated with work-related outcomes. The survey-based FCI suggested more comorbidity in the Ohio sample (Ohio mean = 1.38; Washington mean = 1.14), whereas the WC-based FCI suggested more comorbidity in the Washington sample (Ohio mean = 0.10; Washington mean = 0.33). In the confounding assessment, adding the survey-based FCI to the base model moved the state effect estimates slightly toward null (<1% change). However, substituting the WC-based FCI moved the estimate away from null (8.95% change).

CONCLUSIONS

The WC-based FCI may be useful for identifying specific subsets of workers with chronic conditions, but less useful for chronic condition prevalence. Using the WC-based FCI cross-state appeared to introduce substantial confounding. We strongly advise caution-including state-specific analyses with a reliable reference standard-before using a WC-based FCI in studies involving multiple states.

Linking migration and hospital data in England: linkage process and evaluation of bias

Introduction

Difficulties ascertaining migrant status in national data sources such as hospital records have limited large-scale evaluation of migrant healthcare needs in many countries, including England. Linkage of immigration data for migrants and refugees, with National Health Service (NHS) hospital care data enables research into the relationship between migration and health for a large cohort of international migrants.

Objectives

We aimed to describe the linkage process and compare linkage rates between migrant sub-groups to evaluate for potential bias for data on non-EU migrants and resettled refugees linked to Hospital Episode Statistics (HES) in England.

Methods

We used stepwise deterministic linkage to match records from migrants and refugees to a unique healthcare identifier indicating interaction with the NHS (linkage stage 1 to NHS Personal Demographic Services, PDS), and then to hospital records (linkage stage 2 to HES). We calculated linkage rates and compared linked and unlinked migrant characteristics for each linkage stage.

Results

Of the 1,799,307 unique migrant records, 1,134,007 (63%) linked to PDS and 451,689 (25%) linked to at least one hospital record between 01/01/2005 and 23/03/2020. Individuals on work, student, or working holiday visas were less likely to link to a hospital record than those on settlement and dependent visas and refugees. Migrants from the Middle East and North Africa and South Asia were four times more likely to link to at least one hospital record, compared to those from East Asia and the Pacific. Differences in age, sex, visa type, and region of origin between linked and unlinked samples were small to moderate.

Conclusion

This linked dataset represents a unique opportunity to explore healthcare use in migrants. However, lower linkage rates disproportionately affected individuals on shorter-term visas so future studies of these groups may be more biased as a result. Increasing the quality and completeness of identifiers recorded in administrative data could improve data linkage quality.

Predicting preterm birth using auto-ML frameworks: a large observational study using electronic inpatient discharge data

Background

To develop and compare different AutoML frameworks and machine learning models to predict premature birth.

Methods

The study used a large electronic medical record database to include 715,962 participants who had the principal diagnosis code of childbirth. Three Automatic Machine Learning (AutoML) were used to construct machine learning models including tree-based models, ensembled models, and deep neural networks on the training sample (N = 536,971). The area under the curve (AUC) and training times were used to assess the performance of the prediction models, and feature importance was computed via permutation-shuffling.

Results

The H2O AutoML framework had the highest median AUC of 0.846, followed by AutoGluon (median AUC: 0.840) and Auto-sklearn (median AUC: 0.820), and the median training time was the lowest for H2O AutoML (0.14 min), followed by AutoGluon (0.16 min) and Auto-sklearn (4.33 min). Among different types of machine learning models, the Gradient Boosting Machines (GBM) or Extreme Gradient Boosting (XGBoost), stacked ensemble, and random forrest models had better predictive performance, with median AUC scores being 0.846, 0.846, and 0.842, respectively. Important features related to preterm birth included premature rupture of membrane (PROM), incompetent cervix, occupation, and preeclampsia.

Conclusions

Our study highlights the potential of machine learning models in predicting the risk of preterm birth using readily available electronic medical record data, which have significant implications for improving prenatal care and outcomes.

Sepsis in Latvia-Incidence, Outcomes, and Healthcare Utilization: A Retrospective, Observational Study

This study explores the incidence, outcomes, and healthcare resource utilization concerning sepsis in Latvia's adult population. Using a merged database from the National Health Service and the Latvian Centre for Disease Prevention and Control, sepsis-related hospitalizations were analyzed from 2015-2020. Findings revealed a 53.1% surge in sepsis cases from 2015-2018 with subsequent stabilization. This spike was more prominent among elderly patients. The age/sex adjusted case fatality rate rose from 34.7% in 2015 to 40.5% in 2020. Of the 7764 sepsis survivors, the one-year mortality rate was 12% compared to 2.2% in a reference group of 20,686 patients with infections but no further signs of sepsis. Sepsis survivors also incurred higher healthcare costs, driven by longer rehospitalizations and increased pharmaceutical needs, though they accessed outpatient services less frequently than the reference group. These findings underscore the growing detection of sepsis in Latvia, with survivors facing poorer outcomes and suggesting the need for enhanced post-sepsis outpatient care.

Examining the quality and population representativeness of linked survey and administrative data: guidance and illustration using linked 1958 National Child Development Study and Hospital Episode Statistics data

Introduction

Recent years have seen an increase in linkages between survey and administrative data. It is important to evaluate the quality of such data linkages to discern the likely reliability of ensuing research. Evaluation of linkage quality and bias can be conducted using different approaches, but many of these are not possible when there is a separation of processes for linkage and analysis to help preserve privacy, as is typically the case in the UK (and elsewhere).

Objectives

We aimed to describe a suite of generalisable methods to evaluate linkage quality and population representativeness of linked survey and administrative data which remain tractable when users of the linked data are not party to the linkage process itself. We emphasise issues particular to longitudinal survey data throughout.

Methods

Our proposed approaches cover several areas: i) Linkage rates, ii) Selection into response, linkage consent and successful linkage, iii) Linkage quality, and iv) Linked data population representativeness. We illustrate these methods using a recent linkage between the 1958 National Child Development Study (NCDS; a cohort following an initial 17,415 people born in Great Britain in a single week of 1958) and Hospital Episode Statistics (HES) databases (containing important information regarding admissions, accident and emergency attendances and outpatient appointments at NHS hospitals in England).

Results

Our illustrative analyses suggest that the linkage quality of the NCDS-HES data is high and that the linked sample maintains an excellent level of population representativeness with respect to the single dimension we assessed.

Conclusions

Through this work we hope to encourage providers and users of linked data resources to undertake and publish thorough evaluations. We further hope that providing illustrative analyses using linked NCDS-HES data will improve the quality and transparency of research using this particular linked data resource.

The incidences and clinical outcomes of cryptococcosis in Taiwan: A nationwide, population-based study, 2002-2015

Large-scale epidemiological data on cryptococcosis other than cryptococcal meningitis (CM), human immunodeficiency virus (HIV)- or solid organ transplantation (SOT)-associated cryptococcosis are limited. This study investigated the disease burden of cryptococcosis in Taiwan over 14 years. Incident episodes of cryptococcosis, comorbidities, treatment, and outcomes were captured from Taiwan's National Health Insurance Research Database and National Death Registry between 2002 and 2015. Of 6647 episodes analyzed, the crude incidence rate per 100 000 population increased from 1.48 in 2002 to 2.76 in 2015, which was driven by the growing trend in the non-CM group (0.86-2.12) but not in the CM group (0.62-0.64). The leading three comorbidities were diabetes mellitus (23.62%), malignancy (22.81%), and liver disease (17.42%). HIV accounted for 6.14% of all episodes and was associated with the highest disease-specific incidence rate (269/100 000 population), but the value dropped 16.20% biennially. Within 90 days prior to cohort entry, 30.22% of episodes had systemic corticosteroid use. The in-hospital mortality of all episodes was 10.80%, which varied from 32.64% for cirrhosis and 13.22% for HIV to 6.90% for SOT. CM was associated with a higher in-hospital mortality rate than non-CM (19.15% vs. 6.33%). At diagnosis, only 48.53% of CM episodes were prescribed an amphotericin-based regimen. The incidence rate of cryptococcosis was increasing, especially that other than meningitis and in the non-HIV population. A high index of clinical suspicion is paramount to promptly diagnose, treat, and improve cryptococcosis-related mortality in populations other than those with HIV infection or SOT.

Pediatric and Young Adult Household Transmission of the Initial Waves of SARS-CoV-2 in the United States: Administrative Claims Study

BACKGROUND

The correlates responsible for the temporal changes of intrahousehold SARS-CoV-2 transmission in the United States have been understudied mainly due to a lack of available surveillance data. Specifically, early analyses of SARS-CoV-2 household secondary attack rates (SARs) were small in sample size and conducted cross-sectionally at single time points. From these limited data, it has been difficult to assess the role that different risk factors have had on intrahousehold disease transmission in different stages of the ongoing COVID-19 pandemic, particularly in children and youth.

OBJECTIVE

This study aimed to estimate the transmission dynamic and infectivity of SARS-CoV-2 among pediatric and young adult index cases (age 0 to 25 years) in the United States through the initial waves of the pandemic.

METHODS

Using administrative claims, we analyzed 19 million SARS-CoV-2 test records between January 2020 and February 2021. We identified 36,241 households with pediatric index cases and calculated household SARs utilizing complete case information. Using a retrospective cohort design, we estimated the household SARS-CoV-2 transmission between 4 index age groups (0 to 4 years, 5 to 11 years, 12 to 17 years, and 18 to 25 years) while adjusting for sex, family size, quarter of first SARS-CoV-2 positive record, and residential regions of the index cases.

RESULTS

After filtering all household records for greater than one member in a household and missing information, only 36,241 (0.85%) of 4,270,130 households with a pediatric case remained in the analysis. Index cases aged between 0 and 17 years were a minority of the total index cases (n=11,484, 11%). The overall SAR of SARS-CoV-2 was 23.04% (95% CI 21.88-24.19). As a comparison, the SAR for all ages (0 to 65+ years) was 32.4% (95% CI 32.1-32.8), higher than the SAR for the population between 0 and 25 years of age. The highest SAR of 38.3% was observed in April 2020 (95% CI 31.6-45), while the lowest SAR of 15.6% was observed in September 2020 (95% CI 13.9-17.3). It consistently decreased from 32% to 21.1% as the age of index groups increased. In a multiple logistic regression analysis, we found that the youngest pediatric age group (0 to 4 years) had 1.69 times (95% CI 1.42-2.00) the odds of SARS-CoV-2 transmission to any family members when compared with the oldest group (18 to 25 years). Family size was significantly associated with household viral transmission (odds ratio 2.66, 95% CI 2.58-2.74).

CONCLUSIONS

Using retrospective claims data, the pediatric index transmission of SARS-CoV-2 during the initial waves of the COVID-19 pandemic in the United States was associated with location and family characteristics. Pediatric SAR (0 to 25 years) was less than the SAR for all age other groups. Less than 1% (n=36,241) of all household data were retained in the retrospective study for complete case analysis, perhaps biasing our findings. We have provided measures of baseline household pediatric transmission for tracking and comparing the infectivity of later SARS-CoV-2 variants.

Algorithms Identifying Patients With Acute Exacerbation of Interstitial Pneumonia and Acute Interstitial Lung Diseases Developed Using Japanese Administrative Data

BACKGROUND

We aimed to develop algorithms to identify patients with acute exacerbation of interstitial pneumonia and acute interstitial lung diseases using Japanese administrative data.

METHODS

This single-center validation study examined diagnostic algorithm accuracies. We included patients >18 years old with at least one claim that was a candidate for acute exacerbation of interstitial pneumonia, acute interstitial lung diseases, and pulmonary alveolar hemorrhage who were admitted to our hospital between January 2016 and December 2021. Diagnoses of these conditions were confirmed by at least two respiratory physicians through a chart review. The positive predictive value was calculated for the created algorithms.

RESULTS

Of the 1,109 hospitalizations analyzed, 285 and 243 were for acute exacerbation of interstitial pneumonia and acute interstitial lung diseases, respectively. As there were only five cases of pulmonary alveolar hemorrhage, we decided not to develop an algorithm for it. For acute exacerbation of interstitial pneumonia, acute interstitial lung diseases, and acute exacerbation of interstitial pneumonia or acute interstitial lung diseases, algorithms with high positive predictive value (0.82, 95% confidence interval: 0.76-0.86; 0.82, 0.74-0.88; and 0.89, 0.85-0.92, respectively) and algorithms with slightly inferior positive predictive value but more true positives (0.81, 0.75-0.85; 0.77, 0.71-0.83; and 0.85, 0.82-0.88, respectively) were developed.

CONCLUSION

We developed algorithms with high positive predictive value for identifying patients with acute exacerbation of interstitial pneumonia and acute interstitial lung diseases, useful for future database studies on such patients using Japanese administrative data.

Suicides, drug poisonings, and alcohol-related deaths cluster with health and social disadvantage in 4.1 million citizens from two nations

BACKGROUND

Deaths from suicides, drug poisonings, and alcohol-related diseases ('deaths of despair') are well-documented among working-age Americans, and have been hypothesized to be largely specific to the U.S. However, support for this assertion-and associated policies to reduce premature mortality-requires tests concerning these deaths in other industrialized countries, with different institutional contexts. We tested whether the concentration and accumulation of health and social disadvantage forecasts deaths of despair, in New Zealand and Denmark.

METHODS

We used nationwide administrative data. Our observation period was 10 years (NZ = July 2006-June 2016, Denmark = January 2007-December 2016). We identified all NZ-born and Danish-born individuals aged 25-64 in the last observation year (NZ = 1 555 902, Denmark = 2 541 758). We ascertained measures of disadvantage (public-hospital stays for physical- and mental-health difficulties, social-welfare benefit-use, and criminal convictions) across the first nine years. We ascertained deaths from suicide, drugs, alcohol, and all other causes in the last year.

RESULTS

Deaths of despair clustered within a population segment that disproportionately experienced multiple disadvantages. In both countries, individuals in the top 5% of the population in multiple health- and social-service sectors were at elevated risk for deaths from suicide, drugs, and alcohol, and deaths from other causes. Associations were evident across sex and age.

CONCLUSIONS

Deaths of despair are a marker of inequalities in countries beyond the U.S. with robust social-safety nets, nationwide healthcare, and strong pharmaceutical regulations. These deaths cluster within a highly disadvantaged population segment identifiable within health- and social-service systems.

Estimating the completeness of physician billing claims for diabetes case ascertainment: a multiprovince investigation

INTRODUCTION

Previous research has suggested that how physicians are paid may affect the completeness of billing claims for estimating chronic disease. The purpose of this study is to estimate the completeness of physician billings for diabetes case ascertainment.

METHODS

We used administrative data from eight Canadian provinces covering the period 1 April 2014 to 31 March 2016. The patient cohort was stratified into two mutually exclusive groups based on their physician remuneration type: fee-for-service (FFS), for those paid only on that basis; and non-fee-for-service (NFFS). Using diabetes prescription drug data as our reference data source, we evaluated whether completeness of disease case ascertainment varied with payment type. Diabetes incidence rates were then adjusted for completeness of ascertainment.

RESULTS

The cohort comprised 86 110 patients. Overall, equal proportions received their diabetes medications from FFS and NFFS physicians. Overall, physician payment method had little impact upon the percentage of missed diabetes cases (FFS, 14.8%; NFFS, 12.2%). However, the difference in missed cases between FFS and NFFS varied widely by province, ranging from -1.0% in Nova Scotia to 29.9% in Newfoundland and Labrador. The difference between the observed and adjusted disease incidence rates also varied by province, ranging from 22% in Prince Edward Island to 4% in Nova Scotia.

CONCLUSION

The difference in the loss of cases by physician remuneration method varied across jurisdictions. This loss may contribute to an underestimation of disease incidence. The method we used could be applied to other chronic diseases for which drug therapy could serve as reference data source.

Permanent Supportive Housing Receipt and Health Care Use Among Adults With Disabilities

This study assessed whether permanent supportive housing (PSH) participation is associated with health service use among a population of adults with disabilities, including people transitioning into PSH from community and institutional settings. Our primary data sources were 2014 to 2018 secondary data from a PSH program in North Carolina linked to Medicaid claims. We used propensity score weighting to estimate the average treatment effect on the treated of PSH participation. All models were stratified by whether individuals were in institutional or community settings prior to PSH. In weighted analyses, among individuals who were institutionalized prior to PSH, PSH participation was associated with greater hospitalizations and emergency department (ED) visits and fewer primary care visits during the follow-up period, compared with similar individuals who largely remained institutionalized. Individuals who entered PSH from community settings did not have significantly different health service use from similar comparison group members during the 12-month follow-up period.

Accuracy of self-reported private health insurance coverage

Studies on health insurance coverage often rely on measures self-reported by respondents, but the accuracy of such measures has not been thoroughly validated. This paper is the first to use linked Australian National Health Survey and administrative population tax data to explore the accuracy of self-reported private health insurance (PHI) coverage in survey data. We find that 11.86% of individuals misreport their PHI coverage status, with 11.57% of true PHI holders reporting that they are uninsured and 12.37% of true non-insured persons self-identifying as insured. Our results show reporting errors are systematically correlated with individual and household characteristics. Our evidence on the determinants of errors is supportive of common reasons for misreporting. We directly investigate biases in the determinants of PHI enrollment using survey data. We find that, as compared to administrative data, survey data depict a quantitatively different picture of PHI enrollment determinants, especially those capturing age, gender, language proficiency, labor force status, disability status, number of children in the household, or household income. We also show that PHI coverage misreporting is subsequently associated with misreporting of reasons for purchasing PHI, type of cover and length of cover.

Prevalence of Chronic Pain Among People with Dementia: A Nationwide Study Using French Administrative Data

OBJECTIVE

Alzheimer's disease or Related Dementia (ADRD) is known to disturb pain perception and reduce the ability to report it, resulting in underestimation by practitioners and sub-optimal medical management. The aim of this study was to estimate the prevalence of all types of CP among people with ADRD.

DESIGN

Nationwide cross-sectional study.

SETTINGS

French community-dwelling and nursing home residents.

PARTICIPANTS

People with ADRD, >40 years old, treated with cholinesterase inhibitors or memantine, or with a diagnosis/long-term illness of ADRD and matched with a comparison sample.

SETTINGS

French community-dwelling and nursing home residents.

PARTICIPANTS

People with ADRD, >40 years old, treated with cognitive stimulants (cholinesterase inhibitors and memantine) or with a diagnosis/long-term illness of ADRD and matched with a comparison sample (non-ADRD).

MEASUREMENTS

The capture-recapture method was performed to provide estimates of the prevalence of CP. People treated with analgesic drugs for ≥6 months consecutively or with a medical diagnosis of CP (ICD-10 codes) or referred to a pain center were considered as having CP.

RESULTS

A total of 48,288 individuals were included, of which 16,096 had ADRD and 32,192 without ADRD. The estimated prevalence of CP in people with ADRD was from 57.7% [52.9;63.3] to 57.9%[53.0;63.9], and slightly higher than the non-ADRD sample (from 49.9%[47.0;53.2] to 50.4%[47.3;53.9], p <0.001).

CONCLUSIONS

The prevalence of CP among people living with ADRD was at least the same as or better than individuals without ADRD. This result should alert practitioners' attention to the need for effective pain assessment and management in this population who has difficulties to express and feel pain.

Identifying Predictors of Nursing Home Admission by Using Electronic Health Records and Administrative Data: Scoping Review

Background

Among older adults, nursing home admissions (NHAs) are considered a significant adverse outcome and have been extensively studied. Although the volume and significance of electronic data sources are expanding, it is unclear what predictors of NHA have been systematically identified in the literature via electronic health records (EHRs) and administrative data.

Objective

This study synthesizes findings of recent literature on identifying predictors of NHA that are collected from administrative data or EHRs.

Methods

The PRISMA-ScR (Preferred Reporting Items for Systematic Reviews and Meta-Analyses extension for Scoping Reviews) guidelines were used for study selection. The PubMed and CINAHL databases were used to retrieve the studies. Articles published between January 1, 2012, and March 31, 2023, were included.

Results

A total of 34 papers were selected for final inclusion in this review. In addition to NHA, all-cause mortality, hospitalization, and rehospitalization were frequently used as outcome measures. The most frequently used models for predicting NHAs were Cox proportional hazards models (studies: n=12, 35%), logistic regression models (studies: n=9, 26%), and a combination of both (studies: n=6, 18%). Several predictors were used in the NHA prediction models, which were further categorized into sociodemographic, caregiver support, health status, health use, and social service use factors. Only 5 (15%) studies used a validated frailty measure in their NHA prediction models.

Conclusions

NHA prediction tools based on EHRs or administrative data may assist clinicians, patients, and policy makers in making informed decisions and allocating public health resources. More research is needed to assess the value of various predictors and data sources in predicting NHAs and validating NHA prediction models externally.

Optimal Look-Back Period to Identify True Incident Cases of Diabetes in Medical Insurance Data in the Chinese Population: Retrospective Analysis Study

BACKGROUND

Accurate estimation of incidence and prevalence is vital for preventing and controlling diabetes. Administrative data (including insurance data) could be a good source to estimate the incidence of diabetes. However, how to determine the look-back period (LP) to remove cases with preceding records remains a problem for administrative data. A short LP will cause overestimation of incidence, whereas a long LP will limit the usefulness of a database. Therefore, it is necessary to determine the optimal LP length for identifying incident cases in administrative data.

OBJECTIVE

This study aims to offer different methods to identify the optimal LP for diabetes by using medical insurance data from the Chinese population with reference to other diseases in the administrative data.

METHODS

Data from the insurance database of the city of Weifang, China from between January 2016 and December 2020 were used. To identify the incident cases in 2020, we removed prevalent patients with preceding records of diabetes between 2016 and 2019 (ie, a 4-year LP). Using this 4-year LP as a reference, consistency examination indexes (CEIs), including positive predictive values, the κ coefficient, and overestimation rate, were calculated to determine the level of agreement between different LPs and an LP of 4 years (the longest LP). Moreover, we constructed a retrograde survival function, in which survival (ie, incident cases) means not having a preceding record at the given time and the survival time is the difference between the date of the last record in 2020 and the most recent previous record in the LP. Based on the survival outcome and survival time, we established the survival function and survival hazard function. When the survival probability, S(t), remains stable, and survival hazard converges to zero, we obtain the optimal LP. Combined with the results of these two methods, we determined the optimal LP for Chinese diabetes patients.

RESULTS

The κ agreement was excellent (0.950), with a high positive predictive value (92.2%) and a low overestimation rate (8.4%) after a 2-year LP. As for the retrograde survival function, S(t) dropped rapidly during the first 1-year LP (from 1.00 to 0.11). At a 417-day LP, the hazard function reached approximately zero (ht=0.000459), S(t) remained at 0.10, and at 480 days, the frequency of S(t) did not increase. Combining the two methods, we found that the optimal LP is 2 years for Chinese diabetes patients.

CONCLUSIONS

The retrograde survival method and CEIs both showed effectiveness. A 2-year LP should be considered when identifying incident cases of diabetes using insurance data in the Chinese population.

Impact of the COVID-19 pandemic on self-harm and self-harm/suicide ideation: population-wide data linkage study and time series analysis

BACKGROUND

The COVID-19 pandemic and associated lockdowns were predicted to have a major impact on suicidal behaviour, including self-harm. However, current studies have produced contradictory findings with limited trend data.

AIMS

Nine years of linked individual-level administrative data were utilised to examine changes in hospital-presenting self-harm and ideation (thoughts of self-harm or suicide) before and during the pandemic.

METHOD

National self-harm registry data were linked to demographic and socioeconomic indicators from healthcare registration records (n = 1 899 437). Monthly presentations of self-harm or ideation were split (pre-COVID-19 restrictions: April 2012 to February 2020; and during restrictions: March to September 2020). Auto-regressive integrated moving average (ARIMA) models were trained in R taking into consideration trends and seasonal effects. Forecast ('expected') monthly values were compared with 'actual' values, stratified by demographic factors and method of harm.

RESULTS

The number of individuals presenting with self-harm or ideation dropped significantly at the beginning of the pandemic (March-May 2020), before returning mostly to expected trends from June 2020. Stratified analysis showed similar presentation trends across most demographic subgroups except for those aged over 65 years, living alone or in affluent areas, where presentations remained unaffected, and those aged under 16 years, where numbers presenting with self-harm or ideation increased above expected levels.

CONCLUSIONS

Although population trends show an overall drop in presentations before a return to 'normal' from June 2020, the demographic profile of those presenting with self-harm or ideation varied significantly, with increases in children under the age of 16 years. This highlights important potential target groups who may have been most negatively affected by the pandemic.

Multicenter evaluation of a method to identify delayed diagnosis of diabetic ketoacidosis and sepsis in administrative data

OBJECTIVES

To derive a method of automated identification of delayed diagnosis of two serious pediatric conditions seen in the emergency department (ED): new-onset diabetic ketoacidosis (DKA) and sepsis.

METHODS

Patients under 21 years old from five pediatric EDs were included if they had two encounters within 7 days, the second resulting in a diagnosis of DKA or sepsis. The main outcome was delayed diagnosis based on detailed health record review using a validated rubric. Using logistic regression, we derived a decision rule evaluating the likelihood of delayed diagnosis using only characteristics available in administrative data. Test characteristics at a maximal accuracy threshold were determined.

RESULTS

Delayed diagnosis was present in 41/46 (89 %) of DKA patients seen twice within 7 days. Because of the high rate of delayed diagnosis, no characteristic we tested added predictive power beyond the presence of a revisit. For sepsis, 109/646 (17 %) of patients were deemed to have a delay in diagnosis. Fewer days between ED encounters was the most important characteristic associated with delayed diagnosis. In sepsis, our final model had a sensitivity for delayed diagnosis of 83.5 % (95 % confidence interval 75.2-89.9) and specificity of 61.3 % (95 % confidence interval 56.0-65.4).

CONCLUSIONS

Children with delayed diagnosis of DKA can be identified by having a revisit within 7 days. Many children with delayed diagnosis of sepsis may be identified using this approach with low specificity, indicating the need for manual case review.

Prescribing of valproate and oral antiepileptics for women of childbearing age and during pregnancy in Germany between 2010 and 2020

PURPOSE

To examine prescriptions of valproate and oral antiepileptic drugs (OAED) in Germany irrespective of the indication in women in general and particularly in women of childbearing age (13-49 years) and during pregnancy between 2010 and 2020, that is, before, during and after the implementation of the EU risk minimization measures (RMMs).

METHODS

Analysis of claims data.

STUDY POPULATION

all women continuously insured with the AOK health insurance fund in the respective observation year (2010-2020) and the previous year. OAED were identified by ATC code N03. Period of pregnancy was calculated based on birth information in claims data.

MAIN OUTCOMES MEASURES

(i) prevalent use of valproate/OAED: number of women with at least one prescription of valproate/OAED per year divided by all women of the study population (rate per 1000 women); (ii) percentage of OAED recipients with at least one valproate prescription during pregnancy (13-49 years) in the respective observation year.

RESULTS

Prevalence rate/1000 women for valproate use decreased by -31.33% across all age groups (2010-2014: -7.48%; 2014-2018: -16.47%; 2018-2020: -11,17%) with a strong reduction in women 13-49 years between 2014 and 2018 (-28.74%). The rate for OAED across all age groups rose from 33.43/1000 women in 2010 to 41.03/1000 (+22,73%). Valproate use during pregnancy of women with OAED declined from 1.29% in 2010 to 0.59% in 2020 (-54,26%) (2010-2014: -5.14%; 2014-2018: -42.31%; 2018-2020: -16.69%).

CONCLUSION

Even if, due to the descriptive nature of the study, no causal relationship can be postulated between the RMMs and the strong decrease in valproate prescriptions, our results are compatible with the hypothesis that the measures have improved drug therapy safety.

Trajectory of health care resources among adults stopping or reducing treatment frequency of botulinum toxin for chronic migraine treatment in Alberta, Canada

OBJECTIVE

Understand health resource, medication use, and cost of adults with chronic migraine who received guideline-recommended onabotulinumtoxinA (botulinum toxin) treatment frequency and then continued or reduced/stopped.

BACKGROUND

Botulinum toxin may be a beneficial treatment for chronic migraine; the trajectory of health resources utilization among those with continued or reduced/stopped use is unclear.

METHODS

A retrospective population-based cohort study utilizing administrative data from Alberta, Canada (2012-2020), was performed. A cohort of adults who received ≥5 botulinum toxin treatment cycles for chronic migraine over 18 months (6-month run-in; 1-year pre-index period) were grouped into those who (1) continued use (≥3 treatments/year), or (2) stopped or reduced use (stopped for 6 months then received 0 or 1-2 treatments/year, respectively) over a 1-year post-index period. Health resources and medication use were described, and pre-post costs were assessed. A second cohort that received ≥3 treatments/year immediately followed by 1 year of stopped or reduced use was considered in sensitivity analysis.

RESULTS

Pre-post health resource, medication use, and costs were similar among those with continued use (n = 3336). Among those who stopped or reduced use (n = 1099; 756 stopped, 343 reduced), health resource, medication use, and costs were lower in the post- (total median per-person cost [IQR]: all-cause $4851 [$8090]; migraine-related $835 [$1915]) versus pre- (all-cause $6096 [$7207]; migraine-related $2995 [$1950]) index period (estimated cost ratios [95% CI]: total all-cause 0.86 [0.79, 0.95]; total migraine-related 0.44 [0.40, 0.48]). In the second cohort (n = 3763), return to continued use (≥3 treatments/year) occurred in up to 70.4% in those with reduced use.

CONCLUSIONS

Of adults treated with botulinum toxin for chronic migraine, 75.2% had continued use, stable health resource and medication use, and costs over a 2 year period. In those that stopped/reduced use, the observed lower health resource and migraine medication use may indicate improved symptom control, but the resumption of guideline-recommended treatment intervals after reduced use was common.

Validation of Electronic Health Record-Based Algorithms to Identify Specialist Palliative Care Within the Department of Veterans Affairs

BACKGROUND

The measurement of specialist palliative care (SPC) across Department of Veterans Affairs (VA) facilities relies on algorithms applied to administrative databases. However, the validity of these algorithms has not been systematically assessed.

MEASURES

In a cohort of people with heart failure identified by ICD 9/10 codes, we validated the performance of algorithms to identify SPC consultation in administrative data and differentiate outpatient from inpatient encounters.

INTERVENTION

We derived separate samples of people by receipt of SPC using combinations of stop codes signifying specific clinics, current procedural terminology (CPT), a variable representing encounter location, and ICD-9/ICD-10 codes for SPC. We calculated sensitivity, specificity, and positive and negative predictive values (PPV, NPV) for each algorithm using chart review as the reference standard.

OUTCOMES

Among 200 people who did and did not receive SPC (mean age = 73.9 years (standard deviation [SD] = 11.5), 98% male, 73% White), the validity of the stop code plus CPT algorithm to identify any SPC consultation was: Sensitivity = 0.89 (95% Confidence Interval [CI] 0.82-0.94), Specificity = 1.0 [0.96-1.0], PPV = 1.0 [0.96-1.0], NPV = 0.93 [0.86-0.97]. The addition of ICD codes increased sensitivity but decreased specificity. Among 200 people who received SPC (mean age = 74.2 years [SD = 11.8], 99% male, 71% White), algorithm performance in differentiating outpatient from inpatient encounters was: Sensitivity = 0.95 (0.88-0.99), Specificity = 0.81 (0.72-0.87), PPV = 0.38 (0.29-0.49), and NPV = 0.99 (0.95-1.0). Adding encounter location improved the sensitivity and specificity of this algorithm.

CONCLUSIONS

VA algorithms are highly sensitive and specific in identifying SPC and in differentiating outpatient from inpatient encounters. These algorithms can be used with confidence to measure SPC in quality improvement and research across the VA.

Asthma-related emergency admissions and associated healthcare resource use in Alberta, Canada

BACKGROUND

There is a lack of real-world research assessing asthma management following asthma-related emergency department (ED) discharges. The objective of this study was to characterise follow-up care, healthcare resource use (HCRU) and medical costs following ED admissions in Alberta, Canada.

METHODS

A retrospective cohort study was conducted on adults with asthma using longitudinal population-based administrative data from Alberta Health Services. Adult patients with asthma and ≥1 ED admission from 1 April 2015 to 31 March 2020 were included. ED admissions, outpatient visits, hospitalisations and asthma-specific medication use were measured in the 30 days before and up to 90 days after each asthma-related ED admission. Mean medical costs attributable to each type of HCRU were summarised. All outcomes were stratified by patient baseline disease severity.

RESULTS

Among 128 063 patients incurring a total of 20 142 asthma-related ED visits, a substantial rate of ED readmission was observed, with 10% resulting in readmissions within 7 days and 35% within 90 days. Rates increased with baseline asthma severity. Despite recommendations for patients to be followed up with an outpatient visit within 2-7 days of ED discharge, only 6% were followed up within 7 days. The mean total medical cost per patient was $C8143 in the 30 days prior to and $C5407 in the 30 days after an ED admission.

CONCLUSIONS

Despite recommendations regarding follow-up care for patients after asthma-related ED admissions, there are still low rates of outpatient follow-up visits and high ED readmission rates. New or improved multidimensional approaches must be integrated into follow-up care to optimise asthma control and prevent readmissions.

Use of Big Data and Ethical Issues for Populations With Substance Use Disorder

With expanding data availability and computing power, health research is increasingly relying on big data from a variety of sources. We describe a state-level effort to address aspects of the opioid epidemic through public health research, which has resulted in an expansive data resource combining dozens of administrative data sources in Massachusetts. The Massachusetts Public Health Data Warehouse is a public health innovation that serves as an example of how to address the complexities of balancing data privacy and access to data for public health and health services research. We discuss issues of data protection and data access, and provide recommendations for ethical data governance. Keeping these issues in mind, the use of this data resource has the potential to allow for transformative research on critical public health issues.

Fracture risk and opioid use in patients aged 17-64 years: An analysis of administrative claims data

STUDY OBJECTIVE

The objective of this work is to describe the risk of fracture in adults aged 17 to 64 years across categories of daily opioid dose.

DESIGN

A retrospective analysis of insurance claims data.

DATA SOURCE

Information from a private health insurance provider for the years 2003-2014 for enrolled individuals living in Iowa.

PATIENTS

Patients who were aged 17 to 64 years and prescribed an opioid were followed for the length of opioid treatment plus 30 days to determine whether a fracture occurred.

MEASUREMENTS

Average daily morphine milligram equivalent (MME) was determined for patients who received a prescription for opioids. The outcome of interest was incident fracture following opioid treatment initiation. Cox proportional hazard models were used to compare the risk of fracture across categorical groups of average daily MME. Inverse probability weighting was used to control for patient-level variables.

MAIN RESULTS

In total, 316,024 individual patients were identified in the administrative claims data as having an opioid prescription from 2003 to 2014, and 3038 fractures occurred during follow-up (9.6 fractures per 1000 opioid-prescribed patients). Relative to the lowest quartile of prescribed opioid (⟨22 MME), patients in each higher prescribed level were more likely to experience a fracture (22-36 MME Hazard Ratio (HR) = 1.35, 95% Confidence Interval (CI): 1.21-1.51; 37-50 MME HR = 1.38, 95% CI: 1.24-1.54; ⟩50 MME HR = 1.35, 95% CI: 1.20-1.52).

CONCLUSIONS

We observed an increased fracture risk in all three higher categories of opioid exposure relative to the lowest category (⟨22 MME). The risk of a fracture should be considered in opioid treatment decisions.

Positive Predictive Value of International Classification of Diseases, Ninth Revision, Clinical Modification, and International Classification of Diseases, Tenth Revision, Clinical Modification, Codes for Identification of Congenital Heart Defects

Background Administrative data permit analysis of large cohorts but rely on International Classification of Diseases, Ninth Revision, Clinical Modification (ICD-9-CM), and International Classification of Diseases, Tenth Revision, Clinical Modification (ICD-10-CM) codes that may not reflect true congenital heart defects (CHDs). Methods and Results CHDs in 1497 cases with at least 1 encounter between January 1, 2010 and December 31, 2019 in 2 health care systems, identified by at least 1 of 87 ICD-9-CM/ICD-10-CM CHD codes were validated through medical record review for the presence of CHD and CHD native anatomy. Interobserver and intraobserver reliability averaged >95%. Positive predictive value (PPV) of ICD-9-CM/ICD-10-CM codes for CHD was 68.1% (1020/1497) overall, 94.6% (123/130) for cases identified in both health care systems, 95.8% (249/260) for severe codes, 52.6% (370/703) for shunt codes, 75.9% (243/320) for valve codes, 73.5% (119/162) for shunt and valve codes, and 75.0% (39/52) for "other CHD" (7 ICD-9-CM/ICD-10-CM codes). PPV for cases with >1 unique CHD code was 85.4% (503/589) versus 56.3% (498/884) for 1 CHD code. Of cases with secundum atrial septal defect ICD-9-CM/ICD-10-CM codes 745.5/Q21.1 in isolation, PPV was 30.9% (123/398). Patent foramen ovale was present in 66.2% (316/477) of false positives. True positives had younger mean age at first encounter with a CHD code than false positives (22.4 versus 26.3 years; P=0.0017). Conclusions CHD ICD-9-CM/ICD-10-CM codes have modest PPV and may not represent true CHD cases. PPV was improved by selecting certain features, but most true cases did not have these characteristics. The development of algorithms to improve accuracy may improve accuracy of electronic health records for CHD surveillance.

Why Are Child and Youth Welfare Support Services Initiated? A First-Time Analysis of Administrative Data on Child and Youth Welfare Services in Austria

Even if numerous children and young people are looked after by child and youth welfare, there are only a few scientific studies on the reasons for this support. The aim of this retrospective descriptive study was to examine the reasons why child and youth welfare was initiated. Therefore, administrative data, collected by the Lower Austrian Child and Youth Welfare Service, from the year 2021 will be presented. On the one hand, the frequencies of the different justifications provided by the social workers and, on the other hand, whether these are primarily based on problems of the parents/caregivers or the children are reported. In 2021, a total of 7760 clarifications of child welfare endangerments were initiated. The descriptive statistical analyses showed that the most frequent concerns were parental overload (49%), behavioral issues (10%), and difficult economic conditions (9%). Although a classification according to the caregiver or child level cannot always be clearly distinguished, there is a trend that in many cases (84% to 99% depending on the type of support) the problems lie at the caregiver level. Further studies are necessary so that the care of such vulnerable groups of people will be better supported by scientific findings.

Developing International Classification of Disease code definitions for the study of enteric infection sequelae in Canada

Background

Enteric infections and their chronic sequelae are a major cause of disability and death. Despite the increasing use of administrative health data in measuring the burden of chronic diseases in the population, there is a lack of validated International Classification of Disease (ICD) code-based case definitions, particularly in the Canadian context. Our objective was to validate ICD code definitions for sequelae of enteric infections in Canada: acute kidney injury (AKI); hemolytic uremic syndrome (HUS); thrombotic thrombocytopenic purpura (TTP); Guillain-Barré syndrome/Miller-Fisher syndrome (GBS/MFS); chronic inflammatory demyelinating polyneuropathy (CIDP); ankylosing spondylitis (AS); reactive arthritis; anterior uveitis; Crohn's disease, ulcerative colitis, celiac disease, erythema nodosum (EN); neonatal listeriosis (NL); and Graves' disease (GD).

Methods

We used a multi-step approach by conducting a literature review to identify existing validated definitions, a clinician assessment of the validated definitions, a chart review to verify proposed definitions and a final clinician review. We measured the sensitivity and positive predictive value (PPV) of proposed definitions.

Results

Forty studies met inclusion criteria. We identified validated definitions for 12 sequelae; clinicians developed three (EN, NL, GD). We reviewed 181 charts for 6 sequelae (AKI, HUS, TTP, GBS/MFS, CIDP, AS). Sensitivity (42.8%-100%) and PPV (63.6%-100%) of ICD code definitions varied. Six definitions were modified by clinicians following the chart review (AKI, TTP, GBS/MFS, CIDP, AS, reactive arthritis) to reflect coding practices, increase specificity or sensitivity, and address logistical constraints.

Conclusion

The multi-step design to derive ICD code definitions provided flexibility to identify existing definitions, to improve their sensitivity and PPV and adapt them to the Canadian context.

Linkage of whole genome sequencing and administrative health data in autism: A proof of concept study

Whether genetic testing in autism can help understand longitudinal health outcomes and health service needs is unclear. The objective of this study was to determine whether carrying an autism-associated rare genetic variant is associated with differences in health system utilization by autistic children and youth. This retrospective cohort study examined 415 autistic children/youth who underwent genome sequencing and data collection through a translational neuroscience program (Province of Ontario Neurodevelopmental Disorders Network). Participant data were linked to provincial health administrative databases to identify historical health service utilization, health care costs, and complex chronic medical conditions during a 3-year period. Health administrative data were compared between participants with and without a rare genetic variant in at least 1 of 74 genes associated with autism. Participants with a rare variant impacting an autism-associated gene (n = 83, 20%) were less likely to have received psychiatric care (at least one psychiatrist visit: 19.3% vs. 34.3%, p = 0.01; outpatient mental health visit: 66% vs. 77%, p = 0.04). Health care costs were similar between groups (median: $5589 vs. $4938, p = 0.4) and genetic status was not associated with odds of being a high-cost participant (top 20%) in this cohort. There were no differences in the proportion with complex chronic medical conditions between those with and without an autism-associated genetic variant. Our study highlights the feasibility and potential value of genomic and health system data linkage to understand health service needs, disparities, and health trajectories in individuals with neurodevelopmental conditions.

Comparison of Administrative Database-Derived and Hospital-Derived Data for Monitoring Blood Culture Use in the Pediatric Intensive Care Unit

BACKGROUND

Optimizing blood culture practices requires monitoring of culture use. Collecting culture data from electronic medical records can be resource intensive. Our objective was to determine whether administrative data could serve as a data source to measure blood culture use in pediatric intensive care units (PICUs).

METHODS

Using data from a national diagnostic stewardship collaborative to reduce blood culture use in PICUs, we compared the monthly number of blood cultures and patient-days collected from sites (site-derived) and the Pediatric Health Information System (PHIS, administrative-derived), an administrative data warehouse, for 11 participating sites. The collaborative's reduction in blood culture use was compared using administrative-derived and site-derived data.

RESULTS

Across all sites and months, the median of the monthly relative blood culture rate (ratio of administrative- to site-derived data) was 0.96 (Q1: 0.77, Q3: 1.24). The administrative-derived data produced an estimate of blood culture reduction over time that was attenuated toward the null compared with site-derived data.

CONCLUSIONS

Administrative data on blood culture use from the PHIS database correlates unpredictably with hospital-derived PICU data. The limitations of administrative billing data should be carefully considered before use for ICU-specific data.

Predictors and Consequences of Homelessness: Protocol for a Cohort Study Design Using Linked Routine Data

BACKGROUND

Homelessness is a global burden, estimated to impact more than 100 million people worldwide. Individuals and families experiencing homelessness are more likely to have poorer physical and mental health than the general population. Administrative data is being increasingly used in homelessness research.

OBJECTIVE

The objective of this study is to combine administrative health care data and social housing data to better understand the consequences and predictors associated with being homeless.

METHODS

We will be linking health and social care administrative databases from Northern Ireland, United Kingdom. We will conduct descriptive analyses to examine trends in homelessness and investigate risk factors for key outcomes.

RESULTS

The results of our analyses will be shared with stakeholders, reported at conferences and in academic journals, and summarized in policy briefing notes for policymakers.

CONCLUSIONS

This study will aim to identify predictors and consequences of homelessness in Northern Ireland using linked housing, health, and social care data. The findings of this study will examine trends and outcomes in this vulnerable population using routinely collected health and social care administrative data.

INTERNATIONAL REGISTERED REPORT IDENTIFIER (IRRID)

DERR1-10.2196/42404.

Trends in survival of children with severe congenital heart defects by gestational age at birth: A population-based study using administrative hospital data for England

BACKGROUND

Children with congenital heart defects (CHD) are twice as likely as their peers to be born preterm (<37 weeks' gestation), yet descriptions of recent trends in long-term survival by gestational age at birth (GA) are lacking.

OBJECTIVES

To quantify changes in survival to age 5 years of children in England with severe CHD by GA.

METHODS

We estimated changes in survival to age five of children with severe CHD and all other children born in England between April 2004 and March 2016, overall and by GA-group using linked hospital and mortality records.

RESULTS

Of 5,953,598 livebirths, 5.7% (339,080 of 5,953,598) were born preterm, 0.35% (20,648 of 5,953,598) died before age five and 3.6 per 1000 (21,291 of 5,953,598) had severe CHD. Adjusting for GA, under-five mortality rates fell at a similar rate between 2004-2008 and 2012-2016 for children with severe CHD (adjusted hazard ratio [HR] 0.79, 95% CI 0.71, 0.88) and all other children (HR 0.78, 95% CI 0.76, 0.81). For children with severe CHD, overall survival to age five increased from 87.5% (95% CI 86.6, 88.4) in 2004-2008 to 89.6% (95% CI 88.9, 90.3) in 2012-2016. There was strong evidence for better survival in the ≥39-week group (90.2%, 95% CI 89.1, 91.2 to 93%, 95% CI 92.4, 93.9), weaker evidence at 24-31 and 37-38 weeks and no evidence at 32-36 weeks. We estimate that 51 deaths (95% CI 24, 77) per year in children with severe CHD were averted in 2012-2016 compared to what would have been the case had 2004-2008 mortality rates persisted.

CONCLUSIONS

Nine out of 10 children with severe CHD in 2012-2016 survived to age five. The small improvement in survival over the study period was driven by increased survival in term children. Most children with severe CHD are reaching school age and may require additional support by schools and healthcare services.