Jove
Visualize
Contact Us
JoVE
x logofacebook logolinkedin logoyoutube logo
ABOUT JoVE
OverviewLeadershipBlogJoVE Help Center
AUTHORS
Publishing ProcessEditorial BoardScope & PoliciesPeer ReviewFAQSubmit
LIBRARIANS
TestimonialsSubscriptionsAccessResourcesLibrary Advisory BoardFAQ
RESEARCH
JoVE JournalMethods CollectionsJoVE Encyclopedia of ExperimentsArchive
EDUCATION
JoVE CoreJoVE BusinessJoVE Science EducationJoVE Lab ManualFaculty Resource CenterFaculty Site
Terms & Conditions of Use
Privacy Policy
Policies

Related Experiment Videos

Predicting need for pharmacokinetic consultation follow-up using discriminant analysis.

D E Beck1, E L Bradley, J A Farringer

  • 1School of Pharmacy, Auburn University, AL.

Clinical Pharmacy
|September 1, 1988
PubMed
Summary

This study aimed to develop a model to predict whether patients receiving aminoglycoside therapy would need more than one pharmacokinetic consultation. Researchers used patient data including lab values and ICU status to build a statistical model. They found that factors like leukemia status, serum creatinine, and ICU location were significant predictors. However, the model had a high misclassification rate and was not accurate enough for clinical use. The findings suggest that while the model has statistical value, it needs improvement to be useful in real-world settings. Researchers recommend further study to refine the model's accuracy and practical application.

Related Concept Videos

You might also read

Related Articles

Articles linked to this work by shared authors, journal, and citation graph.

Sort by
Same author

Transcervical endometrial biopsy in the baboon: Effects on the menstrual cycle and relations to protein and dry matter content and histology.

American journal of primatology·2020
Same author

Fenton's reagent for the rapid and efficient isolation of microplastics from wastewater.

Chemical communications (Cambridge, England)·2016
Same author

Model studies of migration from paper and board into fruit and vegetables and into Tenax as a food simulant.

Food additives & contaminants. Part A, Chemistry, analysis, control, exposure & risk assessment·2014
Same author

Printing ink compounds in foods: UK survey results.

Food additives & contaminants. Part B, Surveillance·2014
Same author

Analytical screening studies on irradiated food packaging.

Food additives & contaminants. Part A, Chemistry, analysis, control, exposure & risk assessment·2013
Same author

Timing of colonoscopy after resection for colorectal cancer: are we looking too soon?

Diseases of the colon and rectum·2013

Area of Science:

  • Clinical pharmacology
  • Hospital pharmacy services
  • Pharmacokinetic modeling

Background:

Healthcare providers often struggle to determine which patients will need multiple pharmacokinetic interventions. Prior research has shown that aminoglycoside dosing requires careful monitoring due to variable patient responses. However, no clear method existed to predict which patients would need more than one intervention. This gap motivated the development of a predictive model based on patient characteristics and lab values. Existing knowledge includes the role of serum creatinine and ICU status in drug metabolism. Yet, no prior work had resolved how these factors interact to predict the need for repeated consultations. This study aimed to address this uncertainty by applying statistical modeling to real-world patient data. The goal was to improve resource allocation and patient care outcomes. Establishing a reliable prediction method could streamline clinical workflows and reduce unnecessary interventions.

Purpose Of The Study:

The primary aim was to develop a discriminant model to predict whether patients would require more than one pharmacokinetic consultation. This model could help prioritize patients needing repeated interventions. The study focused on aminoglycoside therapy, where dosing adjustments are common. Researchers examined factors like serum concentrations and ICU status to identify predictive patterns. They aimed to validate the model in a new patient cohort. The motivation was to improve clinical decision-making and reduce unnecessary monitoring. By identifying high-risk patients early, the model could enhance treatment efficiency. The study also sought to evaluate the model's practical usefulness in real clinical settings.

Keywords:
pharmacokinetic modelingaminoglycoside dosingclinical prediction modelshospital pharmacy services

Frequently Asked Questions

The function predicted whether patients would need more than one pharmacokinetic consultation. However, it had a 23% misclassification rate for patients requiring a regimen change.

Leukemia status was the most significant predictor, followed by serum creatinine concentration and ICU location.

Patients in the ICU often have altered drug metabolism and require closer monitoring, which may influence the need for multiple consultations.

Serum creatinine was a significant predictor because it reflects kidney function, which affects aminoglycoside clearance.

Related Experiment Videos

Main Methods:

The study used a two-phase design to develop and test a discriminant model. Phase 1 included 150 patients with aminoglycoside therapy. Researchers collected peak and trough serum concentrations for each patient. Patients were grouped based on whether they needed a regimen change. Forty-seven variables were analyzed using univariate methods. Stepwise discriminant analysis identified significant predictors. Phase 2 applied the model to 47 new patients for validation. Variables included leukemia status, serum creatinine, ICU location, and others. The model classified patients into groups for monitoring or intervention. This approach allowed researchers to assess the model's predictive accuracy and clinical relevance.

Main Results:

The discriminant function identified several significant predictors in Phase 1. Leukemia status was the most significant variable followed by serum creatinine levels. ICU location and male sex also contributed to the model. Actual volume of distribution and therapeutic trough concentration were additional factors. In Phase 2, the model misclassified 23% of patients needing a regimen change. Eighteen percent of patients classified for monitoring did not require adjustments. These findings suggest the model has statistical significance but limited clinical utility. The function's accuracy was insufficient for reliable clinical use. Researchers observed a high rate of false classifications in both groups. These results highlight the challenges of applying statistical models to complex clinical scenarios.

Conclusions:

The derived discriminant function showed statistical significance but poor clinical utility. The model's high misclassification rate limits its practical application. Researchers found that leukemia status and serum creatinine were key predictors. However, these variables did not reliably predict the need for a second intervention. The function's accuracy was insufficient for guiding clinical decisions. The study suggests that additional factors may be needed to improve prediction. The authors propose further research to refine the model's variables. These findings indicate the need for more robust predictive tools in pharmacokinetic consultations.

The model misclassified 23% of patients who needed a regimen change and 18% of those who did not.

The authors propose that additional research is needed to refine the model's variables and improve its clinical utility.