Predictive model for a second hip fracture occurrence using natural language processing and machine learning on electronic health records.

Hip fractures (HFx) are associated with a higher morbidity and mortality rates, leading to a significant reduction in life quality and in limitation of patient´s mobility. The present study aimed to obtain real-world evidence on the clinical characteristics of patients with an initial and a second hip fracture (HFx) and develop a predictive model for second HFx using artificial intelligence. Electronic health records from one hospital centre in Spain from January 2011 to December 2019 were analysed using EHRead® technology, based on natural language processing and machine learning. A total of 1,960 patients with HFx were finally included during the study period after meeting all inclusion and exclusion criteria. From this total, 1835 (93.6%) patients were included in the HFx subgroup, while 124 (6.4%) were admitted to the second HFx (2HFx) subgroup. The mean age of the participants was 84 years and 75.5% were female. Most of comorbidities were more frequently identified in the HFx group, including hypertension (72.0% vs. 67.2%), cognitive impairment (33.0% vs. 31.2%), diabetes mellitus (28.7% vs. 24.8%), heart failure (27.6% vs. 22.4%) and chronic kidney disease (26.9% vs. 16.0%). Based on clinical criteria, 26 features were selected as potential prediction factors. From there, 16 demographics and clinical characteristics such as comorbidities, medications, measures of disabilities for ambulation and type of refracture were selected for development of a competitive risk model. Specifically, those predictors with different associated risk ratios, sorted from higher to lower risk relevance were visual deficit, malnutrition, walking assistance, hypothyroidism, female sex, osteoporosis treatment, pertrochanteric fracture, dementia, age at index, osteoporosis, renal failure, stroke, COPD, heart disease, anaemia, and asthma. This model showed good performance (dependent AUC: 0.69; apparent performance: 0.75) and could help the identification of patients with higher risk of developing a second HFx, allowing preventive measures. This study expands the current available information of HFx patients in Spain and identifies factors that exhibit potential in predicting a second HFx among older patients.

Harmonization of indirect reference intervals calculation by the Bhattacharya method.

OBJECTIVES: The aim of this study was to harmonize the criteria for the Bhattacharya indirect method Microsoft Excel Spreadsheet for reference intervals calculation to reduce between-user variability and use these criteria to calculate and evaluate reference intervals for eight analytes in two different years. METHODS: Anonymized laboratory test results from outpatients were extracted from January 1st 2018 to December 31st 2019. To assure data quality, we examined the monthly results from an external quality control program. Reference intervals were determined by the Bhattacharya method with the St Vincent's hospital Spreadsheet firstly using original criteria and then using additional harmonized criteria defined in this study. Consensus reference intervals using the additional harmonized criteria were calculated as the mean of four users' lower and upper reference interval results. To further test the operation criteria and robustness of the obtained reference intervals, an external user validated the Spreadsheet procedure. RESULTS: The extracted test results for all selected laboratory tests fulfilled the quality criteria and were included in the present study. Differences between users in calculated reference intervals were frequent when using the Spreadsheet. Therefore, additional criteria for the Spreadsheet were proposed and applied by independent users, such as: to set central bin as the mean of all the data, bin size as small as possible, at least three consecutive bins and a high proportion of bins within the curve. CONCLUSIONS: The proposed criteria contributed to the harmonization of reference interval calculation between users of the Bhattacharya indirect method Spreadsheet.

Within-subject biological variation estimates using an indirect data mining strategy. Spanish multicenter pilot study (BiVaBiDa).

OBJECTIVES: The estimates of biological variation (BV) have traditionally been determined using direct methods, which present limitations. In response to this issue, two papers have been published addressing these limitations by employing indirect methods. Here, we present a new procedure, based on indirect methods that analyses data collected within a multicenter pilot study. Using this method, we obtain CVI estimates and calculate confidence intervals (CI), using the EFLM-BVD CVI estimates as gold standard for comparison. METHODS: Data were collected over a 18-month period for 7 measurands, from 3 Spanish hospitals; inclusion criteria: patients 18-75 years with more than two determinations. For each measurand, four different strategies were carried out based on the coefficient of variation ratio (rCoeV) and based on the use of the bootstrap method (OS1, RS2 and RS3). RS2 and RS3 use symmetry reference change value (RCV) to clean database. RESULTS: RS2 and RS3 had the best correlation for the CVI estimates with respect to EFLM-BVD. RS2 used the symmetric RCV value without eliminating outliers, while RS3 combined RCV and outliers. When using the rCoeV and OS1 strategies, an overestimation of the CVI value was obtained. CONCLUSIONS: Our study presents a new strategy for obtaining robust CVI estimates using an indirect method together with the value of symmetric RCV to select the target population. The CVI estimates obtained show a good correlation with those published in the EFLM-BVD database. Furthermore, our strategy can resolve some of the limitations encountered when using direct methods such as calculating confidence intervals.

Indirect determination of biochemistry reference intervals using outpatient data.

The aim of this study was to determine reference intervals in an outpatient population from Vall d'Hebron laboratory using an indirect approach previously described in a Dutch population (NUMBER project). We used anonymized test results from individuals visiting general practitioners and analysed during 2018. Analytical quality was assured by EQA performance, daily average monitoring and by assessing longitudinal accuracy between 2018 and 2020 (using trueness verifiers from Dutch EQA). Per test, outliers by biochemically related tests were excluded, data were transformed to a normal distribution (if necessary) and means and standard deviations were calculated, stratified by age and sex. In addition, the reference limit estimator method was also used to calculate reference intervals using the same dataset. Finally, for standardized tests reference intervals obtained were compared with the published NUMBER results. Reference intervals were calculated using data from 509,408 clinical requests. For biochemical tests following a normal distribution, similar reference intervals were found between Vall d'Hebron and the Dutch study. For creatinine and urea, reference intervals increased with age in both populations. The upper limits of Gamma-glutamyl transferase were markedly higher in the Dutch study compared to Vall d'Hebron results. Creatine kinase and uric acid reference intervals were higher in both populations compared to conventional reference intervals. Medical test results following a normal distribution showed comparable and consistent reference intervals between studies. Therefore a simple indirect method is a feasible and cost-efficient approach for calculating reference intervals. Yet, for generating standardized calculated reference intervals that are traceable to higher order materials and methods, efforts should also focus on test standardization and bias assessment using commutable trueness verifiers.

Rapid and accurate method for quantifying busulfan in plasma samples by isocratic liquid chromatography-tandem mass spectrometry (LC-MS/MS).

Objectives: Administration of busulfan is extending rapidly as a part of a conditioning regimen in patients undergoing hematopoietic stem cell transplantation (HSCT). Monitoring blood plasma levels of busulfan is recommended for identifying the optimal dose in patients and for minimizing toxicity. The aim of this research was to validate a simple, rapid, and cost-effective analytical tool for measuring busulfan in human plasma that would be suitable for routine clinical use. This novel tool was based on liquid chromatography coupled to mass spectrometry. Methods: Human plasma samples were prepared using a one-step protein precipitation protocol. These samples were then resolved by isocratic elution in a C18 column. The mobile phase consisted 2 mM ammonium acetate and 0.1% formic acid dissolved in a 30:70 ratio of methanol/water. Busulfan-d8 was used as the internal standard. Results: The run time was optimized at 1.6 min. Standard curves were linear from 0.03 to 5 mg/L. The coefficient of variation (%CV) was less than 8%. The accuracy of this method had an acceptable bias that fell within 85-115% range. No interference between busulfan and the interfering compound hemoglobin, lipemia, or bilirubin not even at the highest concentrations of compound was tested. Neither carryover nor matrix effects were observed using this method. The area under the plasma drug concentration-time curves obtained for 15 pediatric patients who received busulfan therapy prior to HSCT were analyzed and correlated properly with the administered doses. Conclusions: This method was successfully validated and was found to be robust enough for therapeutic drug monitoring in a clinical setting.

Big data and reference intervals: rationale, current practices, harmonization and standardization prerequisites and future perspectives of indirect determination of reference intervals using routine data.

Reference intervals are commonly used as a decision-making tool. In this review, we provide an overview on "big data" and reference intervals, describing the rationale, current practices including statistical methods, essential prerequisites concerning data quality, including harmonization and standardization, and future perspectives of the indirect determination of reference intervals using routine laboratory data.

A descriptive and validation study of a predictive model of severity of SARS-COV-2 infection.

Objectives: The strain the SARS-COV-2 pandemic is putting on hospitals requires that predictive values are identified for a rapid triage and management of patients at a higher risk of developing severe COVID-19. We developed and validated a prognostic model of COVID-19 severity. Methods: A descriptive, comparative study of patients with positive vs. negative PCR-RT for SARS-COV-2 and of patients who developed moderate vs. severe COVID-19 was conducted. The model was built based on analytical and demographic data and comorbidities of patients seen in an Emergency Department with symptoms consistent with COVID-19. A logistic regression model was designed from data of the COVID-19-positive cohort. Results: The sample was composed of 410 COVID-positive patients (303 with moderate disease and 107 with severe disease) and 81 COVID-negative patients. The predictive variables identified included lactate dehydrogenase, C-reactive protein, total proteins, urea, and platelets. Internal calibration showed an area under the ROC curve (AUC) of 0.88 (CI 95%: 0.85-0.92), with a rate of correct classifications of 85.2% for a cut-off value of 0.5. External validation (100 patients) yielded an AUC of 0.79 (95% CI: 0.71-0.89), with a rate of correct classifications of 73%. Conclusions: The predictive model identifies patients at a higher risk of developing severe COVID-19 at Emergency Department, with a first blood test and common parameters used in a clinical laboratory. This model may be a valuable tool for clinical planning and decision-making.

Enlightening discriminative network functional modules behind Principal Component Analysis separation in differential-omic science studies.

Omic science is rapidly growing and one of the most employed techniques to explore differential patterns in omic datasets is principal component analysis (PCA). However, a method to enlighten the network of omic features that mostly contribute to the sample separation obtained by PCA is missing. An alternative is to build correlation networks between univariately-selected significant omic features, but this neglects the multivariate unsupervised feature compression responsible for the PCA sample segregation. Biologists and medical researchers often prefer effective methods that offer an immediate interpretation to complicated algorithms that in principle promise an improvement but in practice are difficult to be applied and interpreted. Here we present PC-corr: a simple algorithm that associates to any PCA segregation a discriminative network of features. Such network can be inspected in search of functional modules useful in the definition of combinatorial and multiscale biomarkers from multifaceted omic data in systems and precision biomedicine. We offer proofs of PC-corr efficacy on lipidomic, metagenomic, developmental genomic, population genetic, cancer promoteromic and cancer stem-cell mechanomic data. Finally, PC-corr is a general functional network inference approach that can be easily adopted for big data exploration in computer science and analysis of complex systems in physics.