Applying random forest in a health administrative data context: a conceptual guide

AbstractBackgroundDiscrepancies between population-based estimates of the incidence of psychotic disorder and the treated incidence reported by early psychosis intervention (EPI) programs suggest additional cases may be receiving services elsewhere in the health system. Our objective was to estimate the incidence of non-affective psychotic disorder in the catchment area of an EPI program, and compare this to EPI-treated incidence estimates.MethodsWe constructed a retrospective cohort (1997–2015) of incident cases of non-affective psychosis aged 16–50 years in an EPI program catchment using population-based linked health administrative data. Cases were identified by either one hospitalization or two outpatient physician billings within a 12-month period with a diagnosis of non-affective psychosis. We estimated the cumulative incidence and EPI-treated incidence of non-affective psychosis using denominator data from the census. We also estimated the incidence of first-episode psychosis (people who would meet the case definition for an EPI program) using a novel approach.ResultsOur case definition identified 3245 cases of incident non-affective psychosis over the 17-year period. We estimate that the incidence of first-episode non-affective psychosis in the program catchment area is 33.3 per 100 000 per year (95% CI 31.4–35.1), which is more than twice as high as the EPI-treated incidence of 18.8 per 100 000 per year (95% CI 17.4–20.3).ConclusionsCase ascertainment strategies limited to specialized psychiatric services may substantially underestimate the incidence of non-affective psychotic disorders, relative to population-based estimates. Accurate information on the epidemiology of first-episode psychosis will enable us to more effectively resource EPI services and evaluate their coverage.

Download Full-text

Repeat emergency department visits by nursing home residents: a cohort study using health administrative data

BMC Geriatrics ◽

10.1186/s12877-018-0854-8 ◽

2018 ◽

Vol 18 (1) ◽

Cited By ~ 2

Author(s):

Andrea Gruneir ◽

Candemir Cigsar ◽

Xuesong Wang ◽

Alice Newman ◽

Susan E. Bronskill ◽

...

Keyword(s):

Emergency Department ◽

Cohort Study ◽

Nursing Home ◽

Administrative Data ◽

Nursing Home Residents ◽

Emergency Department Visits ◽

Health Administrative Data

Download Full-text

Tu1937 Health Care Utilization Patterns Based on Health Administrative Data Are Moderately Effective at Predicting Disease Behaviour at Diagnosis in Ulcerative Colitis Patients

Gastroenterology ◽

10.1016/s0016-5085(16)33330-3 ◽

2016 ◽

Vol 150 (4) ◽

pp. S983

Author(s):

Sanjay K. Murthy ◽

Tushar Shukla ◽

Marc Andre Belair ◽

Tim Ramsay ◽

Geoffrey C. Nguyen ◽

...

Keyword(s):

Ulcerative Colitis ◽

Health Care ◽

Health Care Utilization ◽

Administrative Data ◽

Care Utilization ◽

Health Administrative Data ◽

Utilization Patterns

Download Full-text

Understanding the role of the family physician in early psychosis intervention

BJPsych Open ◽

10.1192/bjo.2018.67 ◽

2018 ◽

Vol 4 (6) ◽

pp. 447-453 ◽

Cited By ~ 1

Author(s):

Kelly K. Anderson ◽

Suzanne Archie ◽

Richard G. Booth ◽

Chiachen Cheng ◽

Daniel Lizotte ◽

...

Keyword(s):

Primary Care ◽

Administrative Data ◽

Family Physician ◽

Psychiatric Treatment ◽

First Episode Psychosis ◽

First Episode ◽

Health Administrative Data ◽

Episode Psychosis ◽

The Family

BackgroundThe family physician is key to facilitating access to psychiatric treatment for young people with first-episode psychosis, and this involvement can reduce aversive events in pathways to care. Those who seek help from primary care tend to have longer intervals to psychiatric care, and some people receive ongoing psychiatric treatment from the family physician.AimsOur objective is to understand the role of the family physician in help-seeking, recognition and ongoing management of first-episode psychosis.MethodWe will use a mixed-methods approach, incorporating health administrative data, electronic medical records (EMRs) and qualitative methodologies to study the role of the family physician at three points on the pathway to care. First, help-seeking: we will use health administrative data to examine access to a family physician and patterns of primary care use preceding the first diagnosis of psychosis; second, recognition: we will identify first-onset cases of psychosis in health administrative data, and look back at linked EMRs from primary care to define a risk profile for undetected cases; and third, management: we will examine service provision to identified patients through EMR data, including patterns of contacts, prescriptions and referrals to specialised care. We will then conduct qualitative interviews and focus groups with key stakeholders to better understand the trends observed in the quantitative data.DiscussionThese findings will provide an in-depth description of first-episode psychosis in primary care, informing strategies to build linkages between family physicians and psychiatric services to improve transitions of care during the crucial early stages of psychosis.Declaration of interestNone.

Download Full-text

Measuring indicators of health system performance for palliative and end-of-life care using health administrative data: a scoping review

F1000Research ◽

10.12688/f1000research.27121.1 ◽

2020 ◽

Vol 9 ◽

pp. 1438

Author(s):

Suman Budhwani ◽

Ashlinder Gill ◽

Mary Scott ◽

Walter P. Wodchis ◽

JinHee Kim ◽

...

Keyword(s):

Performance Measurement ◽

End Of Life ◽

Administrative Data ◽

Scoping Review ◽

System Performance ◽

End Of Life Care ◽

Population Level ◽

Health System Performance ◽

Life Care ◽

Health Administrative Data

Background: A plethora of performance measurement indicators for palliative and end-of-life care currently exist in the literature. This often leads to confusion, inconsistency and redundancy in efforts by health systems to understand what should be measured and how. The objective of this study was to conduct a scoping review to provide an inventory of performance measurement indicators that can be measured using population-level health administrative data, and to summarize key concepts for measurement proposed in the literature. Methods: A scoping review using MEDLINE and EMBASE, as well as grey literature was conducted. Articles were included if they described performance or quality indicators of palliative and end-of-life care at the population-level using routinely-collected administrative data. Details on the indicator such as name, description, numerator, and denominator were charted. Results: A total of 339 indicators were extracted. These indicators were classified into nine health care sectors and one cross-sector category. Extracted indicators emphasized key measurement themes such as health utilization and cost and excessive, unnecessary, and aggressive care particularly close to the end-of-life. Many indicators were often measured using the same constructs, but with different specifications, such as varying time periods used to ascribe for end-of-life care, and varying patient populations. Conclusions: Future work is needed to achieve consensus ‘best’ definitions of these indicators as well as a universal performance measurement framework, similar to other ongoing efforts in population health. Efforts to monitor palliative and end-of-life care can use this inventory of indicators to select appropriate indicators to measure health system performance.

Download Full-text

Validating International Classification of Disease 10th revision algorithms for identifying influenza and respiratory syncytial virus hospitalizations

University of Toronto Journal of Public Health ◽

10.33137/utjph.v2i2.37000 ◽

2021 ◽

Vol 2 (2) ◽

Author(s):

Mackenzie A Hamilton ◽

Andrew Calzavara ◽

Scott D Emerson ◽

Jeffrey C Kwong

Keyword(s):

Respiratory Syncytial Virus ◽

Administrative Data ◽

Predictive Value ◽

Hospitalized Patients ◽

International Classification ◽

Health Administrative Data ◽

International Classification Of Disease ◽

Icd 10 ◽

Syncytial Virus

Objective: Routinely collected health administrative data can be used to efficiently assess disease burden in large populations, but it is important to evaluate the validity of these data. The objective of this study was to develop and validate International Classification of Disease 10PthP revision (ICD -10) algorithms that identify laboratory-confirmed influenza or laboratory-confirmed respiratory syncytial virus (RSV) hospitalizations using population-based health administrative data from Ontario, Canada. Study Design and Setting: Influenza and RSV laboratory data from the 2014-15 through to 2017-18 respiratory virus seasons were obtained from the Ontario Laboratories Information System (OLIS) and were linked to hospital discharge abstract data to generate influenza and RSV reference cohorts. These reference cohorts were used to assess the sensitivity, specificity, positive predictive value (PPV) and negative predictive value (NPV) of the ICD-10 algorithms. To minimize misclassification in future studies, we prioritized specificity and PPV in selecting top-performing algorithms. Results: 83,638 and 61,117 hospitalized patients were included in the influenza and RSV reference cohorts, respectively. The best influenza algorithm had a sensitivity of 73% (95% CI 72% to 74%), specificity of 99% (95% CI 99% to 99%), PPV of 94% (95% CI 94% to 95%), and NPV of 94% (95% CI 94% to 95%). The best RSV algorithm had a sensitivity of 69% (95% CI 68% to 70%), specificity of 99% (95% CI 99% to 99%), PPV of 91% (95% CI 90% to 91%) and NPV of 97% (95% CI 97% to 97%). Conclusion: We identified two highly specific algorithms that best ascertain patients hospitalized with influenza or RSV. These algorithms may be applied to hospitalized patients if data on laboratory tests are not available, and will thereby improve the power of future epidemiologic studies of influenza, RSV, and potentially other severe acute respiratory infections.

Download Full-text

Case-Ascertainment Models to Identify Adults with Obstructive Sleep Apnea Using Health Administrative Data: Internal and External Validation

Clinical Epidemiology ◽

10.2147/clep.s308852 ◽

2021 ◽

Vol Volume 13 ◽

pp. 453-467

Author(s):

Tetyana Kendzerska ◽

Carl van Walraven ◽

Daniel I McIsaac ◽

Marcus Povitz ◽

Sunita Mulpuru ◽

...

Keyword(s):

Obstructive Sleep Apnea ◽

Sleep Apnea ◽

Administrative Data ◽

External Validation ◽

Health Administrative Data ◽

Case Ascertainment ◽

Obstructive Sleep

Download Full-text

Identifying Prenatal Opioid Exposure in Health Administrative Data for Public Health Surveillance and Epidemiologic Research

International Journal for Population Data Science ◽

10.23889/ijpds.v5i5.1453 ◽

2020 ◽

Vol 5 (5) ◽

Author(s):

Andi Camden ◽

Teresa To ◽

Joel G Ray ◽

Tara Gomes ◽

Li Bai ◽

...

Keyword(s):

Prescription Drug ◽

Administrative Data ◽

Population Based ◽

Emergency Department Visits ◽

Prescription Data ◽

Opioid Use ◽

Health Administrative Data ◽

Maternal Characteristics ◽

Data Source ◽

The Impact

IntroductionAccurate estimation of prenatal opioid exposure (POE) is needed for population-based surveillance & research but can be challenging with health administrative data due to varying definitions & methods. Prior research has relied primarily on infant records with a diagnosis of neonatal abstinence syndrome (NAS). Objectives and Approach1) Evaluate the impact of using different definitions of maternal opioid use in the estimation of POE; 2) Investigate whether maternal characteristics vary by the type of definition used. Population-based cross-sectional study of all hospital births (N= 454,746) from 2014-2017 in Ontario, Canada. Multiple linked population-based health administrative databases were used to identify opioid-related pre- & perinatal Emergency Department visits & hospitalizations & opioid prescriptions. We examined how pre-conception & in-pregnancy maternal characteristics varied by using different approaches to ascertain POE. ResultsThere were 9624 live/still births with POE. Ascertainment of POE was highest using maternal prescription drug data (79%) & infant hospital records with NAS (45%). Maternal characteristics varied by data source used for POE ascertainment. Opioid-related health care during pregnancy identified a high-risk phenotype, contrasted with those ascertained through prescription data, with respective rates of 64% vs. 54% for social assistance, 37% vs. 12% for polydrug use, 23% vs. 6% for alcohol use, 26% vs. 19% for 3+ live births, 13% vs. 5% for victim of violence, 12% vs. 6% for involvement in criminal justice system & 64% vs. 17% for mental health & addictions hospital care. Conclusion / ImplicationsPOE ascertainment differs by health administrative data source & ability to link both across maternal records and with infant. Prescription drug data identified the highest number of opioid-exposed births and, with linked healthcare records, is useful to identify illicit opioid use & additional risk factors. Clinically meaningful differences in maternal characteristics of opioid users exist by POE ascertainment method.

Download Full-text