Search

Objectives Computational algorithms for the interpretation of laboratory test results can support physicians and specialists in laboratory medicine. The aim of this study was to develop, implement and evaluate a machine learning algorithm that automatically assesses the risk of low body iron storage, reflected by low ferritin plasma levels, in anemic primary care patients using a minimal set of basic laboratory tests, namely complete blood count and C-reactive protein (CRP). Methods Laboratory measurements of anemic primary care patients were used to develop and validate a machine learning algorithm. The performance of the algorithm was compared to twelve specialists in laboratory medicine from three large teaching hospitals, who predicted if patients with anemia have low ferritin levels based on laboratory test reports (complete blood count and CRP). In a second round of assessments the algorithm outcome was provided to the specialists in laboratory medicine as a decision support tool. Results Two separate algorithms to predict low ferritin concentrations were developed based on two different chemistry analyzers, with an area under the curve of the ROC of 0.92 (Siemens) and 0.90 (Roche). The specialists in laboratory medicine were less accurate in predicting low ferritin concentrations compared to the algorithms, even when knowing the output of the algorithms as support tool. Implementation of the algorithm in the laboratory system resulted in one new iron deficiency diagnosis on average per day. Conclusions Low ferritin levels in anemic patients can be accurately predicted using a machine learning algorithm based on routine laboratory test results. Moreover, implementation of the algorithm in the laboratory system reduces the number of otherwise unrecognized iron deficiencies.

Additional file 2. Survey results.

Additional file 1. Interview & Survey.

Additional file 3. Details of the cost-minimisation analysis.

ObjectiveTo externally validate various prognostic models and scoring rules for predicting short term mortality in patients admitted to hospital for covid-19.DesignTwo stage individual participant data meta-analysis.SettingSecondary and tertiary care.Participants46 914 patients across 18 countries, admitted to a hospital with polymerase chain reaction confirmed covid-19 from November 2019 to April 2021.Data sourcesMultiple (clustered) cohorts in Brazil, Belgium, China, Czech Republic, Egypt, France, Iran, Israel, Italy, Mexico, Netherlands, Portugal, Russia, Saudi Arabia, Spain, Sweden, United Kingdom, and United States previously identified by a living systematic review of covid-19 prediction models published inThe BMJ, and through PROSPERO, reference checking, and expert knowledge.Model selection and eligibility criteriaPrognostic models identified by the living systematic review and through contacting experts. A priori models were excluded that had a high risk of bias in the participant domain of PROBAST (prediction model study risk of bias assessment tool) or for which the applicability was deemed poor.MethodsEight prognostic models with diverse predictors were identified and validated. A two stage individual participant data meta-analysis was performed of the estimated model concordance (C) statistic, calibration slope, calibration-in-the-large, and observed to expected ratio (O:E) across the included clusters.Main outcome measures30 day mortality or in-hospital mortality.ResultsDatasets included 27 clusters from 18 different countries and contained data on 46 914patients. The pooled estimates ranged from 0.67 to 0.80 (C statistic), 0.22 to 1.22 (calibration slope), and 0.18 to 2.59 (O:E ratio) and were prone to substantial between study heterogeneity. The 4C Mortality Score by Knight et al (pooled C statistic 0.80, 95% confidence interval 0.75 to 0.84, 95% prediction interval 0.72 to 0.86) and clinical model by Wang et al (0.77, 0.73 to 0.80, 0.63 to 0.87) had the highest discriminative ability. On average, 29% fewer deaths were observed than predicted by the 4C Mortality Score (pooled O:E 0.71, 95% confidence interval 0.45 to 1.11, 95% prediction interval 0.21 to 2.39), 35% fewer than predicted by the Wang clinical model (0.65, 0.52 to 0.82, 0.23 to 1.89), and 4% fewer than predicted by Xie et al’s model (0.96, 0.59 to 1.55, 0.21 to 4.28).ConclusionThe prognostic value of the included models varied greatly between the data sources. Although the Knight 4C Mortality Score and Wang clinical model appeared most promising, recalibration (intercept and slope updates) is needed before implementation in routine care.

BACKGROUND: Recognising patients who need immediate hospital treatment for sepsis while simultaneously limiting unnecessary referrals is challenging for GPs. AIM: To develop and validate a sepsis prediction model for adult patients in primary care. DESIGN AND SETTING: This was a prospective cohort study in four out-of-hours primary care services in the Netherlands, conducted between June 2018 and March 2020. METHOD: Adult patients who were acutely ill and received home visits were included. A total of nine clinical variables were selected as candidate predictors, next to the biomarkers C-reactive protein, procalcitonin, and lactate. The primary endpoint was sepsis within 72 hours of inclusion, as established by an expert panel. Multivariable logistic regression with backwards selection was used to design an optimal model with continuous clinical variables. The added value of the biomarkers was evaluated. Subsequently, a simple model using single cut-off points of continuous variables was developed and externally validated in two emergency department populations. RESULTS: A total of 357 patients were included with a median age of 80 years (interquartile range 71-86), of which 151 (42%) were diagnosed with sepsis. A model based on a simple count of one point for each of six variables (aged >65 years; temperature >38°C; systolic blood pressure ≤110 mmHg; heart rate >110/min; saturation ≤95%; and altered mental status) had good discrimination and calibration (C-statistic of 0.80 [95% confidence interval = 0.75 to 0.84]; Brier score 0.175). Biomarkers did not improve the performance of the model and were therefore not included. The model was robust during external validation. CONCLUSION: Based on this study's GP out-of-hours population, a simple model can accurately predict sepsis in acutely ill adult patients using readily available clinical parameters. ispartof: BRITISH JOURNAL OF GENERAL PRACTICE vol:72 issue:719 pages:E437-E445 ispartof: location:England status: published

The objective of the paper is to estimate the number of patients presenting with chest pain suspected of acute coronary syndrome (ACS) in primary care and to calculate possible cost effects of a future clinical decision rule (CDR) incorporating a point-of-care test (PoCT) as compared with current practice. The annual incidence of chest pain, referrals and ACS in primary care was estimated based on a literature review and on a Dutch and Belgian registration study. A health economic model was developed to calculate the potential impact of a future CDR on costs and effects (ie, correct referral decisions), in several scenarios with varying correct referral decisions. One-way, two-way, and probabilistic sensitivity analyses were performed to test robustness of the model outcome to changes in input parameters. Annually, over one million patient contacts in primary care in the Netherlands concern chest pain. Currently, referral of eventual ACS negative patients (false positives, FPs) is estimated to cost €1,448 per FP patient, with total annual cost exceeding 165 million Euros in the Netherlands. Based on ‘international data’, at least a 29% reduction in FPs is required for the addition of a PoCT as part of a CDR to become cost-saving, and an additional €16 per chest pain patient (ie, 16.4 million Euros annually in the Netherlands) is saved for every further 10% relative decrease in FPs. Sensitivity analyses revealed that the model outcome was robust to changes in model inputs, with costs outcomes mainly driven by costs of FPs and costs of PoCT. If PoCT-aided triage of patients with chest pain in primary care could improve exclusion of ACS, this CDR could lead to a considerable reduction in annual healthcare costs as compared with current practice.

Additional file 1. This file contains the reference values of the 15 laboratory tests, extensive descriptions of the multinomial and the binomial logistic regression model, the assumptions that have been tested, as well as the detailed results of the multinomial logistic regression model, of the best subset selection and of the variance inflation factor.

An influenza epidemic can greatly increase the workload in primary care and the emergency department (ED) and can even disrupt the healthcare system. It is difficult to diagnose influenza by history taking and physical examination. A fast diagnosis usinginfluenza point-of-care tests (POCTs) could reduce unnecessary antibiotic prescriptions, diagnostic tests, consultations and hospital admissions. Moreover, length of stay on EDs and length of admission could be shortened. The analytical accuracy of antigen detection tests for influenza is relatively low compared to the well performing RT-PCR assays (sensitivity and specificity approximately 95%). Only 1 randomized controlled trial has shown the effect of a (combined) RT-PCR assay for influenza detection on clinically relevant outcome measures. Observational research suggests that introduction of RT-PCR assays for influenza detection reduces length of stay on the ED and decreased time from sample reception to result. For practical reasons, we should embrace the introduction of RT-PCR assays for influenza detection on EDs. Before POCTs can be implemented in primary care (family medicine) the analytical accuracy and time to receive results should be improved and effects of its clinical impact should be proven.