Prospective evaluation of data-driven models to predict daily risk of Clostridioides difficile infection at 2 large academic health centers.
Ontology highlight
ABSTRACT:
SUBMITTER: Kamineni M
PROVIDER: S-EPMC10024639 | biostudies-literature | 2023 Jul
REPOSITORIES: biostudies-literature
Publications
Prospective evaluation of data-driven models to predict daily risk of <i>Clostridioides difficile</i> infection at 2 large academic health centers.
Kamineni Meghana M Ötleş Erkin E Oh Jeeheh J Rao Krishna K Young Vincent B VB Li Benjamin Y BY West Lauren R LR Hooper David C DC Shenoy Erica S ES Guttag John G JG Wiens Jenna J Makar Maggie M
Infection control and hospital epidemiology 20220919 7
Many data-driven patient risk stratification models have not been evaluated prospectively. We performed and compared the prospective and retrospective evaluations of 2 <i>Clostridioides difficile</i> infection (CDI) risk-prediction models at 2 large academic health centers, and we discuss the models' robustness to data-set shifts. ...[more]
Similar Datasets
Project description:OBJECTIVE An estimated 293,300 healthcare-associated cases of Clostridium difficile infection (CDI) occur annually in the United States. To date, research has focused on developing risk prediction models for CDI that work well across institutions. However, this one-size-fits-all approach ignores important hospital-specific factors. We focus on a generalizable method for building facility-specific models. We demonstrate the applicability of the approach using electronic health records (EHR) from the University of Michigan Hospitals (UM) and the Massachusetts General Hospital (MGH). METHODS We utilized EHR data from 191,014 adult admissions to UM and 65,718 adult admissions to MGH. We extracted patient demographics, admission details, patient history, and daily hospitalization details, resulting in 4,836 features from patients at UM and 1,837 from patients at MGH. We used L2 regularized logistic regression to learn the models, and we measured the discriminative performance of the models on held-out data from each hospital. RESULTS Using the UM and MGH test data, the models achieved area under the receiver operating characteristic curve (AUROC) values of 0.82 (95% confidence interval [CI], 0.80-0.84) and 0.75 ( 95% CI, 0.73-0.78), respectively. Some predictive factors were shared between the 2 models, but many of the top predictive factors differed between facilities. CONCLUSION A data-driven approach to building models for estimating daily patient risk for CDI was used to build institution-specific models at 2 large hospitals with different patient populations and EHR systems. In contrast to traditional approaches that focus on developing models that apply across hospitals, our generalizable approach yields risk-stratification models tailored to an institution. These hospital-specific models allow for earlier and more accurate identification of high-risk patients and better targeting of infection prevention strategies. Infect Control Hosp Epidemiol 2018;39:425-433.
| S-EPMC6421072 | biostudies-literature
Project description:BackgroundClostridioides difficile infection (CDI) is the most common hospital acquired infection in the USA, with recurrence rates > 15%. Although primary CDI has been extensively linked to gut microbial dysbiosis, less is known about the factors that promote or mitigate recurrence. Moreover, previous studies have not shown that microbial abundances in the gut measured by 16S rRNA amplicon sequencing alone can accurately predict CDI recurrence.ResultsWe conducted a prospective, longitudinal study of 53 non-immunocompromised participants with primary CDI. Stool sample collection began pre-CDI antibiotic treatment at the time of diagnosis, and continued up to 8 weeks post-antibiotic treatment, with weekly or twice weekly collections. Samples were analyzed using (1) 16S rRNA amplicon sequencing, (2) liquid chromatography/mass-spectrometry metabolomics measuring 1387 annotated metabolites, and (3) short-chain fatty acid profiling. The amplicon sequencing data showed significantly delayed recovery of microbial diversity in recurrent participants, and depletion of key anaerobic taxa at multiple time-points, including Clostridium cluster XIVa and IV taxa. The metabolomic data also showed delayed recovery in recurrent participants, and moreover mapped to pathways suggesting distinct functional abnormalities in the microbiome or host, such as decreased microbial deconjugation activity, lowered levels of endocannabinoids, and elevated markers of host cell damage. Further, using predictive statistical/machine learning models, we demonstrated that the metabolomic data, but not the other data sources, can accurately predict future recurrence at 1 week (AUC 0.77 [0.71, 0.86; 95% interval]) and 2 weeks (AUC 0.77 [0.69, 0.85; 95% interval]) post-treatment for primary CDI.ConclusionsThe prospective, longitudinal, and multi-omic nature of our CDI recurrence study allowed us to uncover previously unrecognized dynamics in the microbiome and host presaging recurrence, and, in particular, to elucidate changes in the understudied gut metabolome. Moreover, we demonstrated that a small set of metabolites can accurately predict future recurrence. Our findings have implications for development of diagnostic tests and treatments that could ultimately short-circuit the cycle of CDI recurrence, by providing candidate metabolic biomarkers for diagnostics development, as well as offering insights into the complex microbial and metabolic alterations that are protective or permissive for recurrence. Video Abstract.
| S-EPMC9178838 | biostudies-literature
Project description:Effect on gene transcript levels upon up- or down-regulation of two component system WalRK (Wal-ON and Wal-OFF)
2022-04-09 | GSE200346 | GEO
Project description:BackgroundClostridioides difficile infection (CDI) is a major cause of severe diarrhea. In this retrospective study, we identified CDI risk factors by comparing demographic and clinical characteristics for Kaiser Permanente Northern California members ≥18 years old with and without laboratory-confirmed incident CDI.MethodsWe included these risk factors in logistic regression models to develop 2 risk scores that predict future CDI after an Index Date for Risk Score Assessment (IDRSA), marking the beginning of a period for which we estimated CDI risk.ResultsDuring May 2011 to July 2014, we included 9986 CDI cases and 2 230 354 members without CDI. The CDI cases tended to be older, female, white race, and have more hospitalizations, emergency department and office visits, skilled nursing facility stays, antibiotic and proton pump inhibitor use, and specific comorbidities. Using hospital discharge as the IDRSA, our risk score model yielded excellent performance in predicting the likelihood of developing CDI in the subsequent 31-365 days (C-statistic of 0.848). Using a random date as the IDRSA, our model also predicted CDI risk in the subsequent 31-365 days reasonably well (C-statistic 0.722).ConclusionsThese results can be used to identify high-risk populations for enrollment in C difficile vaccine trials and facilitate study feasibility regarding sample size and time to completion.
| S-EPMC7953654 | biostudies-literature
Project description:Background:The global epidemic of nosocomial diarrhea caused by Clostridioides (Clostridium) difficile started in 2000, with high mortality rates and emergence of a new hypervirulent strain NAP1/BI/027. The aim of this study was to assess the presence of ribotype 027 and other C. difficile ribotypes in a Serbian University Hospital, compare the temporal variability of ribotypes 3 years apart, as well as to compare clinical, demographic and laboratory characteristics and disease outcome among patients infected with 027 and non-027 ribotype. This was a prospective observational cohort study addressing 4-month intervals during 2014/2015 and 2017/2018. Results:Ribotyping was performed in 64 non-duplicate C. difficile strains. Ribotype 027 was the most prevalent, and was detected in 53 (82.8%) patients (43/45 and 10/19 patients in 2014-2015 and 2017/2018, respectively). Other detected ribotypes were 001/072 in 4 (6.3%), 002 in 4 (6.3%), 014/020 in 2 (3.1%) and 176 in 1 (1.5%) patient. The percentage of the patients infected with ribotype 027 significantly decreased during the 3-year period, from 95.6 to 52.6% (p?<?0.001). Ribotype 027 infection was associated with fluoroquinolone treatment more frequently than infection with other ribotypes [33 (62.3%) vs. 2 (18.2%), p?=?0.010)]. A severe C. difficile infection was diagnosed more often in patients with the detected ribotype 027 compared to those infected with non-027 ribotypes (p?=?0.006). No significant difference in the mortality and recurrence rates was found between the patients infected with ribotype 027 and those infected with other ribotypes [10/53 (18.8%) vs. 2/11 (18.2%), p?=?0.708, and 10/35 (28.6%) vs. 0/2 (0%), p?=?1.000, respectively]. Conclusion:Clostridium difficile ribotype 027 was the most prevalent ribotype among patients in a large Serbian hospital, but there is a clear decreasing trend.
| S-EPMC7243319 | biostudies-literature
Project description:Clostridioides difficile is one of the most common nosocomial pathogens and a global public health threat. Upon colonization of the gastrointestinal tract, C. difficile is exposed to a rapidly changing polymicrobial environment and a dynamic metabolic milieu. Despite the link between the gut microbiota and susceptibility to C. difficile, the impact of synergistic interactions between the microbiota and pathogens on the outcome of infection is largely unknown. Here, we show that microbial cooperation between C. difficile and Enterococcus has a profound impact on the growth, metabolism, and pathogenesis of C. difficile.. Through a process of nutrient restriction and metabolite cross-feeding, E. faecalis shapes the metabolic environment in the gut to enhance C. difficile fitness and increase toxin production. These findings demonstrate that members of the microbiota, such as Enterococcus, have a previously unappreciated impact on C. difficile behavior and virulence.
2022-09-17 | GSE165751 | GEO
Project description:The response to iron limitation of several bacteria is regulated by the ferric uptake regulator (Fur). The Fur-regulated transcriptional, translational and metabolic networks of the Gram-positive, pathogen Clostridioides difficile were investigated by a combined RNA sequencing, proteomic, metabolomic and electron microscopy approach. At high iron conditions (760 g/l) the C. difficile fur mutant displayed a growth deficiency compared to wild type C. difficile cells. Several iron transporters were found induced by Fur regulation during low iron (11 g/l) conditions. The major adaptation to low iron conditions was observed for the central energy metabolism. Most ferredoxin-dependent amino acid fermentations were found significantly down regulated (had, etf, acd, grd, trx, bdc, hbd). The substrates of these pathways phenylalanine, leucine, glycine and some initial intermediates (phenylpyruvate, oxo-isocaproate, 3-hydroxy-butanoyl-CoA, crotonyl-CoA) were found accumulated, while some end product like isocaproate and butanoate were found reduced. Flavodoxin (fldX) formation and riboflavin biosynthesis (rib) were found enhanced, most likely to replace the missing ferredoxins. Proline reductase (prd), the corresponding ion pumping RNF complex (rnf) and the reaction product 5-aminovalerate were significantly enhanced. An ATP forming ATPase (atpCDGAHFEB,) of the F0F1-type was found induced while the formation of a V-type, mostly proton-pumping, ATP-consuming ATPase (atpDBAFCEKI, was decreased. The [Fe-S] enzyme-dependent pyruvate formate lyase (pfl), formate dehydrogenase (fdh) and hydrogenase (hyd) branch of glucose utilization and glycogen biosynthesis (glg) were significantly reduced, leading to an accumulation of glucose and pyruvate. The formation of [Fe-S] enzyme carbon monoxide dehydrogenase (coo) was inhibited. An intensive remodeling of the cell wall was observed most likely to increase antibiotic resistance. Polyamine biosynthesis (spe) was found induced leading to an accumulation of spermine, spermidine and putrescine. The fur mutant lost most of its flagella. Finally, the CRISPR/Cas and a prophage encoding operon were downregulated. Fur binding sites were found upstream of 20 of the regulated genes. Overall, adaptation to low iron conditions in C. difficile focused on an increase of iron import, significant decrease in metabolic iron utilization and protection during the complex transition.
2019-01-15 | GSE120189 | GEO
Project description:The intestines house a diverse microbiota that must compete for nutrients to survive, but the specific limiting nutrients that control pathogen colonization are not clearly defined. Clostridioides difficile colonization typically requires prior disruption of the microbiota, suggesting that outcompeting commensals for resources is key in establishing C. difficile infection (CDI). The immune protein calprotectin (CP) is released into the gut lumen during CDI to chelate zinc (Zn) and other essential nutrient metals. Yet, the impact of Zn limitation on C. difficile colonization is unknown. To define C. difficile responses to Zn limitation, we performed RNA sequencing on C. difficile exposed to CP. In media with CP, C. difficile upregulated genes involved in metal homeostasis and amino acid metabolism.
2019-10-24 | GSE135912 | GEO
Project description:We present predictive models for comprehensive systems analysis of Clostridioides difficile, the etiology of pseudomembranous colitis. By leveraging 151 published transcriptomes, we generated an EGRIN model that organizes 90% of C. difficile genes into a transcriptional regulatory network of 297 co-regulated modules, implicating genes in sporulation, carbohydrate transport, and metabolism. By advancing a metabolic model through addition and curation of metabolic reactions including nutrient uptake, we discovered 14 amino acids, diverse carbohydrates, and 10 metabolic genes as essential for C. difficile growth in the intestinal environment. Finally, we developed a PRIME model to uncover how EGRIN-inferred combinatorial gene regulation by transcription factors, such as CcpA and CodY, modulates essential metabolic processes to enable C. difficile growth relative to commensal colonization. The C. difficile interactive web portal provides access to these model resources to support collaborative systems-level studies of context-specific virulence mechanisms in C. difficile.
| S-EPMC8595754 | biostudies-literature
Project description:Nowadays, the role of a tutor is more important than ever to prevent students dropout and improve their academic performance. This work proposes a data-driven system to extract relevant information hidden in the student academic data and, thus, help tutors to offer their pupils a more proactive personal guidance. In particular, our system, based on machine learning techniques, makes predictions of dropout intention and courses grades of students, as well as personalized course recommendations. Moreover, we present different visualizations which help in the interpretation of the results. In the experimental validation, we show that the system obtains promising results with data from the degree studies in Law, Computer Science and Mathematics of the Universitat de Barcelona.
| S-EPMC5308611 | biostudies-literature