Case-finding Algorithm Research Articles

Trends in the incidence and prevalence of cirrhosis in Manitoba, Canada: A population-based study (2010-2019)

Introduction and objectivesThe burden of chronic liver disease and cirrhosis continues to increase in North America. We sought to estimate the incidence and prevalence of cirrhosis in Manitoba, Canada over time and assess changes in trends between 2010-2019. Material and methodsWe performed a population-based study using Manitoba administrative health care data, and two validated case-finding algorithms. Annual incidence and prevalence rates were estimated using a generalized linear model with generalized estimating equations, adjusting for age and sex. Changes in estimates were tested using linear trend regression models. ResultsTwo algorithms estimated the number of prevalent cirrhosis to be 16,140 and 29,943 respectively. The age- and sex-adjusted incidence rates increased over the study (from 149 to 264 cases per 100,000 population in 2010, to 177 to 388 cases per 100,000 population in 2019). Cirrhosis incidence increased annually by 2-6 %, with the largest increase (6-8 % 95 % CI 7-9 %, p <0.0001) in those aged 18-44 years. Irrespective of the algorithm used, females consistently exhibited higher cirrhosis incidence and prevalence compared to males over time (P <0.0001). Prevalence demonstrated an upward trend among all age groups over time for both algorithms (P < 0.0001). ConclusionsThis population-based study highlights concerning temporal trends in cirrhosis, characterized by rising annual incidence and prevalence estimates, particularly among young adults and females. These findings underscore the urgent need for comprehensive strategies that encompass prevention, early detection, and the delivery of high-quality healthcare and public health initiatives to effectively tackle this escalating health burden.

Algorithmic Identification of Multiple Sclerosis in the SAIL databank

BackgroundMultiple Sclerosis (MS) is a challenging disease to identify within large repositories of healthcare data. Diagnosis can be protracted requiring many modalities that may not form part of the repository data collection. Accurate case finding would have multiple applications for studying epidemiology. AimDevelop a case finding algorithm within the Secure Anonymised Information Linkage (SAIL) Databank that can reliably elicit Multiple Sclerosis. MethodUtilising a cross sectional cohort study we used multiple datasets within SAIL; General Practice, Inpatient, Outpatient and Office of National Statistics to develop our case finding algorithm based on the coding nomenclatures and timing within these datasets. The results of this algorithm were then tested against two patient data sets: The UK MS Register (n=836), consisting of mostly self confirmed disease and a clinical cohort from South West Wales (n=713). ResultsFrom 4,757,428 records, the algorithm identified 6,194 cases of MS within Wales on 31st December 2020 (prevalence 221.65 [95%CI 216.17–227.24] per 100,000). Case finding sensitivity and specificity was 96.8% and 99.9% for the clinically validated population-based cohort and sensitivity was 96.7% from the self-declared registry population. ConclusionWe successfully identified MS cases within Wales and verified this within two independent data sets.

Using machine learning to improve anaphylaxis case identification in medical claims data.

Anaphylaxis is a severe life-threatening allergic reaction, and its accurate identification in healthcare databases can harness the potential of "Big Data" for healthcare or public health purposes. This study used claims data obtained between October 1, 2015 and February 28, 2019 from the CMS database to examine the utility of machine learning in identifying incident anaphylaxis cases. We created a feature selection pipeline to identify critical features between different datasets. Then a variety of unsupervised and supervised methods were used (eg, Sammon mapping and eXtreme Gradient Boosting) to train models on datasets of differing data quality, which reflects the varying availability and potential rarity of ground truth data in medical databases. Resulting machine learning model accuracies ranged from 47.7% to 94.4% when tested on ground truth data. Finally, we found new features to help experts enhance existing case-finding algorithms. Developing precise algorithms to detect medical outcomes in claims can be a laborious and expensive process, particularly for conditions presented and coded diversely. We found it beneficial to filter out highly potent codes used for data curation to identify underlying patterns and features. To improve rule-based algorithms where necessary, researchers could use model explainers to determine noteworthy features, which could then be shared with experts and included in the algorithm. Our work suggests machine learning models can perform at similar levels as a previously published expert case-finding algorithm, while also having the potential to improve performance or streamline algorithm construction processes by identifying new relevant features for algorithm construction.

Identifying people living with cystic fibrosis in the Danish National Patient Registry: A validation study

BackgroundThe Danish National Patient Registry (DNPR) serves as a valuable resource for scientific research. However, to ensure accurate results in cystic fibrosis (CF) studies that rely on DNPR data, a robust case-identification algorithm is essential. This study aimed to develop and validate algorithms for the reliable identification of CF patients in the DNPR. MethodsUsing the Danish Cystic Fibrosis Registry (DCFR) as a reference, accuracy measures including sensitivity and positive predictive value (PPV) for case-finding algorithms deployed in the DNPR were calculated. Algorithms were based on minimum number of hospital contacts with CF as the main diagnosis and minimum number of days between first and last contact. ResultsAn algorithm requiring a minimum of one hospital contact with CF as the main diagnosis yielded a sensitivity of 96.1 % (95 % CI: 94.2 %; 97.4 %) and a PPV of 84.9 % (82.0 %; 87.4 %). The highest-performing algorithm required minimum 2 hospital visits and a minimum of 182 days between the first and the last contact and yielded a sensitivity of 95.9 % (95 % CI: 94.1 %; 97.2 %), PPV of 91.0 % (95 % CI: 88.6 %; 93.0 %) and a cohort entry delay of 3.2 months at the 75th percentile (95th percentile: 38.7 months). ConclusionsThe DNPR captures individuals with CF with high sensitivity and is a valuable resource for CF-research. PPV was improved at a minimal cost of sensitivity by increasing requirements of minimum number of hospital contacts and days between first and last contact. Cohort entry delay increased with number of required hospital contacts.

Development and Validation of Case-Finding Algorithms to Identify Pancreatic Cancer in the Veterans Health Administration.

Survival in pancreatic ductal adenocarcinoma (PDAC) remains poor due to late diagnosis. Electronic Health Records (EHRs) can be used to study this rare disease, but validated algorithms to identify PDAC in the United States EHRs do not currently exist. To develop and validate an algorithm using Veterans Health Administration (VHA) EHR data for the identification of patients with PDAC. We developed two algorithms to identify patients with PDAC in the VHA from 2002 to 2023. The algorithms required diagnosis of exocrine pancreatic cancer in either ≥ 1 or ≥ 2 of the following domains: (i) the VA national cancer registry, (ii) an inpatient encounter, or (iii) an outpatient encounter in an oncology setting. Among individuals identified with ≥ 1 of the above criteria, a random sample of 100 were reviewed by three gastroenterologists to adjudicate PDAC status. We also adjudicated fifty patients not qualifying for either algorithm. These patients died as inpatients and had alkaline phosphatase values within the interquartile range of patients who met ≥ 2 of the above criteria for PDAC. These expert adjudications allowed us to calculate the positive and negative predictive value of the algorithms. Of 10.8 million individuals, 25,533 met ≥ 1 criteria (PPV 83.0%, kappa statistic 0.93) and 13,693 individuals met ≥ 2 criteria (PPV 95.2%, kappa statistic 1.00). The NPV for PDAC was 100%. An algorithm incorporating readily available EHR data elements to identify patients with PDAC achieved excellent PPV and NPV. This algorithm is likely to enable future epidemiologic studies of PDAC.

Development and Validation of Case-Finding Algorithms for Digestive Cancer in the Spanish Healthcare Database BIFAP.

electronic health records (EHRs) are helpful tools in epidemiology despite not being primarily collected for research. In Spain, primary care physicians play a central role and manage patients even in specialized care. All of this introduces variability that may lead to diagnostic inconsistencies. Therefore, data validation studies are crucial, so we aimed to develop and validate case-finding algorithms for digestive cancer in the primary care database BIFAP. from 2001 to 2019, subjects aged 40-89 without a cancer history were included. Case-finding algorithms using diagnostic codes and text-mining were built. We randomly sampled, clustered, and manually reviewed 816 EHRs. Then, positive predictive values (PPVs) and 95% confidence intervals (95% CIs) for each cancer were computed. Age and sex standardized incidence rates (SIRs) were compared with those reported by the National Cancer Registry (REDECAN). we identified 95,672 potential cases. After validation, the PPV (95% CI) for hepato-biliary cancer was 87.6% (81.8-93.4), for esophageal cancer, it was 96.2% (93.1-99.2), for pancreatic cancer, it was 89.4% (84.5-94.3), for gastric cancer, it was 92.5% (88.3-96.6), and for colorectal cancer, it was 95.2% (92.1-98.4). The SIRs were comparable to those reported by the REDECAN. the case-finding algorithms demonstrated high performance, supporting BIFAP as a suitable source of information to conduct epidemiologic studies of digestive cancer.

Prevalence and Incidence Trends of Attention Deficit/Hyperactivity Disorder in Children and Youth Aged 1-24 Years in Ontario, Canada: A Validation Study of Health Administrative Data Algorithms: Tendances de la prévalence et de l'incidence du trouble de déficit de l'attention/hyperactivité chez les enfants et les jeunes âgés de 1 à 24 ans, en Ontario, Canada: une étude de validation des algorithmes

To estimate prevalence and incidence rates over time in children and youth with attention deficit/hyperactivity disorder from the validation of population-based administrative data algorithms using family physicians' electronic medical records as a reference standard. A retrospective cohort study was conducted in Ontario, Canada to identify attention deficit/hyperactivity disorder among children and youth aged 1-24 years in health administrative data derived from case-finding algorithms using family physicians' electronic medical records. Multiple administrative data algorithms identifying attention deficit/hyperactivity disorder cases were developed and tested from physician-diagnosis of attention deficit/hyperactivity disorder in the electronic medical record to determine their diagnostic accuracy. We calculated algorithm performance using sensitivity, specificity, and predictive values. The most optimal algorithm was used to estimate prevalence and incidence rates of attention deficit/hyperactivity disorder from 2014 to 2021 in Ontario. The optimal performing algorithm was "2 physician visits for attention deficit/hyperactivity disorder in 1 year or 1 attention deficit/hyperactivity disorder-specific prescription" with sensitivity: 83.2% (95% confidence interval [CI], 81.8% to 84.5%), specificity: 98.6% (95% CI, 98.5% to 98.7%), positive predictive value: 78.6% (95% CI, 77.1% to 80.0%) and negative predictive value: 98.9% (95% CI, 98.8% to 99.0%). From 2014, prevalence rates for attention deficit/hyperactivity disorder increased from 5.29 to 7.48 per 100 population in 2021 (N = 281,785). Males had higher prevalence rates (7.49 to 9.59 per 100 population, 1.3-fold increase) than females (2.96-5.26 per 100 population, 1.8-fold increase) from 2014 to 2021. Incidence rates increased from 2014 (0.53 per 100 population) until 2018, decreased in 2020 then rose steeply in 2021 (0.89 per 100 population, N = 34,013). Males also had higher incidence rates than females from 2014 to 2020 with females surpassing males in 2021 (0.70-0.81 per 100 male population,1.2-fold increase versus 0.36-0.97 per 100 female population, 2.7-fold increase). Attention deficit/hyperactivity disorder is increasing in prevalence. We developed an administrative data algorithm that can reliably identify children and youth with attention deficit/hyperactivity disorder with good diagnostic accuracy.

Using machine learning to improve anaphylaxis case identification in medical claims data.

Anaphylaxis is a severe life-threatening allergic reaction, and its accurate identification in healthcare databases can harness the potential of "Big Data" for healthcare or public health purposes. This study used claims data obtained between October 1, 2015 and February 28, 2019 from the CMS database to examine the utility of machine learning in identifying incident anaphylaxis cases. We created a feature selection pipeline to identify critical features between different datasets. Then a variety of unsupervised and supervised methods were used (eg, Sammon mapping and eXtreme Gradient Boosting) to train models on datasets of differing data quality, which reflects the varying availability and potential rarity of ground truth data in medical databases. Resulting machine learning model accuracies ranged between 47.7% and 94.4% when tested on ground truth data. Finally, we found new features to help experts enhance existing case-finding algorithms. Developing precise algorithms to detect medical outcomes in claims can be a laborious and expensive process, particularly for conditions presented and coded diversely. We found it beneficial to filter out highly potent codes used for data curation to identify underlying patterns and features. To improve rule-based algorithms where necessary, researchers could use model explainers to determine noteworthy features, which could then be shared with experts and included in the algorithm. Our work suggests machine learning models can perform at similar levels as a previously published expert case-finding algorithm, while also having the potential to improve performance or streamline algorithm construction processes by identifying new relevant features for algorithm construction.

Validation of an ICD-10 case-finding algorithm for endometrial cancer in US insurance claims.

To evaluate the positive predictive value (PPV) of an endometrial cancer case finding algorithm using International Classification of Disease 10th revision Clinical Modification (ICD-10-CM) diagnosis codes from US insurance claims for implementation in a planned post-marketing safety study. Two algorithm variants were evaluated. Provisional incident endometrial cancer cases were identified from 2016 through 2020 among women aged ≥50 years. One algorithm variant used diagnosis codes for malignant neoplasms of uterine sites (C54.x), excluding C54.2 (malignant neoplasm of myometrium); the other used only C54.1 (malignant neoplasm of endometrium). A random sample of medical records of recent incident provisional cases (2018-2020) was requested for adjudication. Confirmed cases showed biopsy evidence of endometrial cancer, documentation of cancer staging, or hysterectomy following diagnosis. We estimated the PPV of the variants with 95% confidence intervals (CI) excluding cases that had insufficient information. Of 294 provisional cases adjudicated, 85% were from outpatient settings (n = 249). Mean age at diagnosis was 69.3 years. Among the 294 adjudicated cases (identified with the broader algorithm variant), the same 223 were confirmed endometrial cancer cases by both algorithm variants. The PPV (95% CI) for the broader algorithm variant was 84.2% (79.2% and 88.3%), and for the variant using only C54.1 was 85.8% (80.9% and 89.8%). We developed and validated an algorithm using ICD-10-CM diagnosis codes to identify endometrial cancer cases in health insurance claims with a sufficiently high PPV to use in a planned post-marketing safety study.

Validation of the recording of idiopathic pulmonary fibrosis in routinely collected electronic healthcare records in England

BackgroundRoutinely-collected healthcare data provide a valuable resource for epidemiological research. Validation studies have shown that for most conditions, simple lists of clinical codes can reliably be used for case finding in primary care, however, studies exploring the robustness of this approach are lacking for diseases such as idiopathic pulmonary fibrosis (IPF) which are largely managed in secondary care.MethodUsing the UK’s Clinical Practice Research Datalink (CPRD) Aurum dataset, which comprises patient-level primary care records linked to national hospital admissions and cause-of-death data, we compared the positive predictive value (PPV) of eight diagnostic algorithms. Algorithms were developed based on the literature and IPF diagnostic guidelines using combinations of clinical codes in primary and secondary care (SNOMED-CT or ICD-10) with/without additional information. The positive predictive value (PPV) was estimated for each algorithm using the death record as the gold standard. Utilization of the reviewed codes across the study period was observed to evaluate any change in coding practices over time.ResultA total of 17,559 individuals had a least one record indicative of IPF in one or more of our three linked datasets between 2008 and 2018. The PPV of case-finding algorithms based on clinical codes alone ranged from 64.4% (95%CI:63.3–65.3) for a “broad” codeset to 74.9% (95%CI:72.8–76.9) for a “narrow” codeset comprising highly-specific codes. Adding confirmatory evidence, such as a CT scan, increased the PPV of our narrow code-based algorithm to 79.2% (95%CI:76.4–81.8) but reduced the sensitivity to under 10%. Adding evidence of hospitalisation to the standalone code-based algorithms also improved PPV, (PPV = 78.4 vs. 64.4%; sensitivity = 53.5% vs. 38.1%). IPF coding practices changed over time, with the increased use of specific IPF codes.ConclusionHigh diagnostic validity was achieved by using a restricted set of IPF codes. While adding confirmatory evidence increased diagnostic accuracy, the benefits of this approach need to be weighed against the inevitable loss of sample size and convenience. We would recommend use of an algorithm based on a broader IPF code set coupled with evidence of hospitalisation.

Sarcopenia and associated risk factors in oncology outpatients in specialized cancer centers in Saudi Arabia: a cross-sectional study.

To measure the prevalence of probable sarcopenia and the associated risk factors in patients with cancer. This is a cross-sectional study. A total of 324 cancer patients were screened for sarcopenia using a simple questionnaire to rapidly diagnose sarcopenia [SARC-F (strength, assistance with walking, rising from a chair, climbing stairs, and falls)] and a hand grip dynamometer tool. The study was conducted from 1 January 2021 till 28 February 2021, in the outpatient department of Princess Noorah Oncology Center. Among 324 cancer patients receiving active chemotherapy treatment, 28.4% screened positive for sarcopenia (SARC-F score ≥4). Moreover, 23.45% were identified as probable cases [SARC-F score ≥4 and a low hand grip strength (HGS)] of sarcopenia, according to the European Working Group on Sarcopenia in Older People consensus 2 (EWGSOP2) case-finding algorithm, which is sufficient to initiate a management plan. One-quarter of the cancer patients had probable sarcopenia at our institution. Sarcopenia risk was independently associated with patient age, and the risk of sarcopenia was low among patients with higher albumin concentrations. Screening cancer patients for sarcopenia using the SARC-F questionnaire and HGS may offer a useful strategy to mitigate the risk of unfavorable consequences that may occur during cancer treatment.

Zero Suicide Quality Improvement: Developmental and Pandemic-Related Patterns in Youth at Risk for Suicide Attempts

ABSTRACT The Zero Suicide (ZS) approach to health system quality improvement (QI) aspires to reduce/eliminate suicides through enhancing risk detection and suicide prevention services. This first report from our randomized trial evaluating a stepped care for suicide prevention intervention within a health system conducting ZS-QI describes (1) our screening and case identification process, (2) variation among adolescents versus young adults, and (3) pandemic-related patterns during the first COVID-19 pandemic year. Between April 2017 and January 2021, youths aged 12–24 years with elevated suicide risk were identified through an electronic health record (EHR) case-finding algorithm followed by direct assessment screening to confirm risk. Eligible/enrolled youth were evaluated for suicidality, self-harm, and risk/protective factors. Case finding, screening, and enrollment yielded 301 participants showing suicide risk indicators: 97% past-year suicidal ideation, 83% past suicidal behavior; and 90% past non-suicidal self-injury (NSSI). Compared to young adults, adolescents reported more past-year suicide attempts (47% vs. 21%, p < .001) and NSSI (past 6 months, 64% vs. 39%, p < .001); less depression, anxiety, posttraumatic stress, and substance use; and greater social connectedness. Pandemic onset was associated with lower participation of racial-ethnic minority youths (18% vs. 33%, p < .015) and lower past-month suicidal ideation and behavior. Results support the value of EHR case-finding algorithms for identifying youths with potentially elevated risk who could benefit from suicide prevention services, which merit adaptation for adolescents versus young adults. Lower racial-ethnic minority participation after the COVID-19 pandemic onset underscores challenges for services to enhance health equity during a period with restricted in-person health care, social distancing, school closures, and diverse stresses.

Epidemiology of multiple sclerosis in the Campania Region (Italy): Derivation and validation of an algorithm to calculate the 2015-2020 incidence.

We aim to validate an algorithm based on routinely-collected healthcare data to detect incidence of multiple sclerosis (MS) in the Campania Region (South Italy) and to explore its spatial and temporal variations. We included individuals resident in the Campania Region who had at least one MS record in administrative datasets (drug prescriptions, hospital discharge, outpatients), from 2015 to 2020. We merged administrative to the clinical datasets to ascertain the actual date of diagnosis, and validated the minimum interval from our study baseline (Jan 1, 2015) to first MS records in administrative datasets to detect incident cases. We used Bayesian approach to explore geographical distribution, also including deprivation index as a covariate in the estimation model. We used the capture-recapture method to estimate the proportion of undetected cases. The best performance was achieved by the 12-month interval algorithm, detecting 2,150 incident MS cases, with 74.4% sensitivity (95%CI=64.1%, 85.9%) and 95.3% specificity (95%CI=90.7%, 99.8%). The cumulative incidence was 36.68 (95%CI=35.15, 38.26) per 100,000 from 2016 to 2020. The mean annual incidence was 7.34 (95%CI=7.03, 7.65) per 100,000 people-year. The geographical distribution of MS relative risk shows a decreasing east-west incidence gradient. The number of expected MS cases was 11% higher than the detected cases. We validated a case-finding algorithm based on administrative data to estimate MS incidence, and its spatial/temporal variations. This algorithm provides up-to-date estimates of MS incidence, and will be used in future studies to evaluate changes in MS incidence in relation to different risk factors.

Prospective study of a case-finding algorithm to detect NAFLD with advanced fibrosis in primary care patients.

Primary care providers need strategies to identify NAFLD patients and select for specialty referral, but proposed algorithms have only been studied in established NAFLD patients. We implemented an algorithm for all adults with diabetes mellitus in a large primary care practice and excluded hepatitis B and C or alcohol use. Applying annual Fibrosis-4 Index and NAFLD Fibrosis Score for 5 years, we categorized patients as low-risk, indeterminate-risk, or high-risk for advanced fibrosis. We targeted all high-risk and messaged each primary care provider, recommending hepatology linkage. We collected final diagnosis and fibrosis (F0-4) outcomes. Using multivariable logistic regression, we assessed risk factors for advanced fibrosis stage (F3-4). Of 3028 patients, 1018 were low-risk, 577 indeterminate-risk, and 611 high-risk. There were 264 target patients; their 89 primary care providers received a message per patient suggesting hepatology referral. The majority (n=149) were referred; at triage, 118 were deemed likely NAFLD. Of these, 90 completed visits, 78/90 were diagnosed as NAFLD, and 69/78 underwent fibrosis staging, with F3 to 4 in 25/69. In multivariable analysis, hemoglobin A1c ≥8% (OR=7.02, 95% CI: 1.29-38.18) and Fibrosis-4 Index (OR=1.79, 95% CI: 1.07-2.99) were associated with increased risk of F3 to 4. This is the first prospective study testing a case-finding strategy in primary care and almost 1/3 of diabetes mellitus were high-risk for advanced fibrosis. When prompted, 73% of primary care providers placed referrals and 76% of patients completed visits, revealing 86% NAFLD and 36% F3 to 4. This study demonstrates the readiness for such a strategy in primary care; integrating hemoglobin A1c into this algorithm may further improve the performance of Fibrosis-4 Index in this setting.

Positive Predictive Value of Diagnosis Codes for Hidradenitis Suppurativa and Cutaneous Abscess

Background: Requisite to the application of clinical databases for observational research in hidradenitis suppurativa (HS) is the identification of an accurate case cohort. There is limited information on the validity of using diagnosis codes to identify HS and cutaneous abscess (CA) case cohorts. Objectives: The objective of the study was to assess the validity in utilizing diagnosis codes to establish the HS and CA cohorts from an integrated health system clinical database. Methods: In this retrospective study, we evaluated a case-finding algorithm for HS which included the presence of at least one diagnosis code for HS and no diagnosis codes for CA of the axilla, groin, perineum, or buttock at any time. We evaluated a case-finding algorithm for CA which included the presence of at least one diagnosis code for abscess in the axilla, groin, perineum, or buttock and no codes for HS at any time. Medical records for a random sample of 100 patients meeting algorithm criteria were reviewed and adjudicated by dermatologists for accuracy of diagnosis. The positive predictive values (PPVs) of the ICD code-based case definitions were calculated, using the dermatologist’s chart-based adjudication as the reference standard. Results: Among 100 HS patients, median age (interquartile range) in the HS cohort was 33 (24.75, 43.25) years. HS patients were predominantly female (82%) and white (70%). Median age among 100 CA patients was 31.5 (95% CI 6.75, 47.25) years. The majority of CA patients were female (56%) and white (69%). The PPVs of the ICD-based case-finding algorithms for HS and CA were 88% (95% CI 81.6–94.4%) and 75.0% (95% CI 66.5–83.5%), respectively. Conclusion: The case-finding algorithms employing at least one diagnosis code for HS and CA of the axilla, groin, perineum, or buttock, with the exclusion of the other’s diagnosis codes, has good to strong PPV and may provide balance in achieving accuracy and adequate power for cohorts identified within clinical databases.

Modernization of a Large Spinal Cord Injuries and Disorders Registry: The Veterans Administration Experience.

Since the 1990s, Veterans Health Administration (VHA) has maintained a registry of Veterans with Spinal Cord Injuries and Disorders (SCI/Ds) to guide clinical care, policy, and research. Historically, methods for collecting and recording data for the VHA SCI/D Registry (VSR) have required significant time, cost, and staffing to maintain, were susceptible to missing data, and caused delays in aggregation and reporting. Each subsequent data collection method was aimed at improving these issues over the last several decades. This paper describes the development and validation of a case-finding and data-capture algorithm that uses primary clinical data, including diagnoses and utilization across 9 million VHA electronic medical records, to create a comprehensive registry of living and deceased Veterans seen for SCI/D services since 2012. A multi-step process was used to develop and validate a computer algorithm to create a comprehensive registry of Veterans with SCI/D whose records are maintained in the enterprise wide VHA Corporate Data Warehouse. Chart reviews and validity checks were used to validate the accuracy of cases that were identified using the new algorithm. An initial cohort of 28,202 living and deceased Veterans with SCI/D who were enrolled in VHA care from 10/1/2012 through 9/30/2017 was validated. Tables, reports, and charts using VSR data were developed to provide operational tools to study, predict, and improve targeted management and care for Veterans with SCI/Ds. The modernized VSR includes data on diagnoses, qualifying fiscal year, recent utilization, demographics, injury, and impairment for 38,022 Veterans as of 11/2/2022. This establishes the VSR as one of the largest ongoing longitudinal SCI/D datasets in North America and provides operational reports for VHA population health management and evidence-based rehabilitation. The VSR also comprises one of the only registries for individuals with non-traumatic SCI/Ds and holds potential to advance research and treatment for multiple sclerosis (MS), amyotrophic lateral sclerosis (ALS), and other motor neuron disorders with spinal cord involvement. Selected trends in VSR data indicate possible differences in the future lifelong care needs of Veterans with SCI/Ds. Future collaborative research using the VSR offers opportunities to contribute to knowledge and improve health care for people living with SCI/Ds.

Diagnostic accuracy of screening algorithms to identify persons with active pulmonary tuberculosis at prison entry: protocol of a systematic review and network meta-analysis.

Prison inmates are a high-risk group for tuberculosis (TB) infection and disease due to the increasing number of vulnerable fringe groups, risk factors (e.g., alcohol and drug addictions), contagious diseases (HIV, hepatitis), and their high-risk behavior. Compared to the general population, TB incidence and prevalence rates are significantly higher among prison inmates. Early identification of potentially infectious pulmonary TB (PTB) and targeted care of sick inmates are essential to effectively control TB within the prison system. The WHO recommends combining active and passive case-finding in prisons. No study has been published comparing the broad spectrum of screening tools using a diagnostic accuracy network meta-analysis (NMA). We aim to identify the most accurate TB case-finding algorithm at prison entry that is feasible in resource-limited prisons of high-burden TB countries and ensures continuous comprehensive TB detection services in such settings. Evidence generated by this NMA can provide important decision support in selecting the most (cost-) effective algorithms for screening methods for resource-limited settings in the short, medium, and long terms.

Development and validation of case-finding algorithms to identify acute pancreatitis in the Veterans Health Administration.

Acute pancreatitis (AP) is a frequently encountered adverse drug reaction. However, the validity of diagnostic codes for AP is unknown. We aimed to determine the positive predictive value (PPV) of a diagnostic code-based algorithm for identifying patients with AP within the US Veterans Health Administration and evaluate the value of adding readily available structured laboratory information. We identified patients with possible AP events first based on the presence of a single hospital discharge ICD-9 or ICD-10 diagnosis of AP (Algorithm 1). We then expanded Algorithm 1 by including relevant laboratory test results (Algorithm 2). Specifically, we considered amylase or lipase serum values obtained between 2 days before admission and the end of the hospitalization. Medical records of a random sample of patients identified by the respective algorithms were reviewed by two separate gastroenterologists to adjudicate AP events. The PPV (95% confidence interval [CI]) for the algorithms were calculated. Algorithm 2, consisting of one ICD-9 or ICD-10 hospital discharge diagnosis of AP and the addition of lipase serum value ≥200 U/L, had a PPV 89.1% (95% CI 83.0%-95.2%), improving from the PPV of algorithm 1 (57.9% [95% CI 46.8-69.0]). An algorithm consisting of an ICD-9 or ICD-10 diagnosis of AP with a lipase value ≥200 U/L achieved high PPV. This simple algorithm can be readily implemented in any electronic health records (EHR) systems and could be useful for future pharmacoepidemiologic studies on AP.

Development and validation of a case-finding algorithm for the identification of non-small cell lung cancers in a region-wide Italian pathology registry.

PurposeTo develop and validate a case-finding algorithm for the identification of Non-Small Cell Lung Cancer (NSCLC) cases in a region-wide Italian pathology registry (PR).Materials and methodsData collected between 2009 and 2017 in the PR and the Pharmacy Database of the University Hospital of Siena and the PR of Tuscany region were used. A NSCLC-identification algorithm based on free-text keywords and SNOMED morphology and topography codes was designed and tested on data from Siena: indication for drug use (i.e. NSCLC) was the reference standard for sensitivity (SE); positive predictive value (PPV) was estimated through manual review. Algorithm modifications were then tested to improve algorithm performance: PPV was calculated against validated dataset from PR of Siena; a range of SE [min-max] was estimated in PR of Tuscany using analytical formulae that assumed NSCLC incidence equal either to 80% or 90% of overall lung cancer incidence recorded in Tuscany. The algorithm modification with the best performance was chosen as the final version of the algorithm. A random sample of 200 cases was extracted from the PR of Tuscany for manual review.ResultsThe first version of the algorithm showed a PPV of 74.7% and SE of 79% in PR of Siena. The final version of the algorithm had a SE in PR of Tuscany that grew with calendar time (2009 = [24.7%-28%]; 2017 = [57.9%-65.1%]) and a PPV of 93%.ConclusionsThe final NSCLC-finding algorithm showed with very high PPV. SE was in line with the expected contribution of PR to overall cases captured in the regional Cancer Registry, with a trend of increase over calendar time. Given the promising algorithm validity and the wide use of SNOMED terminology in electronic pathology records, the proposed algorithm is expected to be easily adapted to other electronic databases for (pharmaco)epidemiology purposes.

Multi-trajectory models of serum biomarkers among patients with monoclonal gammopathy of undetermined significance.

Understanding the progression of monoclonal gammopathy of undetermined significance (MGUS) to multiple myeloma (MM) is needed to identify patients who would benefit from closer clinical surveillance. Given that two of the defining criteria of MM are renal failure and anemia, we described the trajectories of creatinine (Cr) and hemoglobin (Hgb) over time in patients with a diagnosis of MGUS. Patients diagnosed with MGUS (n=424) were identified by a previously validated case-finding algorithm using health claims and electronic health record data (2007-2015) and followed through 2018. Group-based trajectory modeling identified patients with distinct laboratory value trajectories of Cr (mg/dl) and Hgb (g/dl). Most patients were non-Hispanic White (97.6%) with a mean age of 75years at MGUS diagnosis. Three multi-trajectory groups were identified: (1) Normal Cr/Hgb (n=225; 53.1%)-stable serum Cr levels and decreasing, normal Hgb levels; (2) Normal Cr/lower-normal Hgb group (n=188; 44.3%)-stable, slightly elevated levels of Cr and decreasing levels of Hgb; and (3) High Cr/borderline Hgb group (n=11; 2.6%)-increased Cr levels and stable low levels of Hgb. Patients with MGUS in Group 2 were older than patients in other groups, and patients in group 3 had more comorbidities than participants in all other groups. Few patients developed MM during the study period. We were able to identify distinct biomarker trajectories in patients with MGUS over time. Future research should investigate how these trajectories may be related to the risk of progression to MM, including M-protein levels.