Development of Machine Learning Models to Validate a Medication Regimen Complexity Scoring Tool for Critically Ill Patients

Al-Mamun, Mohammad A.; Brothers, Todd; Newsome, Andrea Sikora

doi:10.1177/1060028020959042

Cited by 28 publications

(33 citation statements)

References 34 publications

Supporting

Mentioning

Contrasting

Order By: Relevance

“…Medication regimen complexity, as measured by the MRC-ICU, has been previously incorporated into machine learning prediction models along with other relevant patient characteristics and resulted in improved mortality prediction in a small cohort of patients. 42 In this study, medication regimen complexity was highest in Patient Clusters 2 and 4, which is in line with previous investigations of MRC-ICU that used traditional inferential statistics to demonstrate a relationship between increasing medication regimen complexity and increased mortality, length of stay, and uid overload as well as increased need for critical care pharmacist interventions to optimize the medication regimens. [43][44][45][46][47][48] Taken together, the methodologies in this study appear to be able to appropriately group degree of critical illness (i.e., acuity) with degree of intervention intensity (e.g., mechanical ventilation, medications) with patient outcomes (e.g., mortality).…”

Section: Discussionsupporting

confidence: 90%

Cluster analysis driven by unsupervised latent feature learning of intensive care unit medications to identify novel pharmaco-phenotypes of critically ill patients

Newsome¹,

Jeong

Yu³

et al. 2022

Preprint

View full text Add to dashboard Cite

Unsupervised clustering of intensive care unit (ICU) medications may identify unique medication clusters (i.e., pharmacophenotypes) in critically ill adults. We performed an unsupervised analysis with Restricted Boltzmann Machine of 991 medications profiles of patients managed in the ICU to explore pharmacophenotypes that correlated with ICU complications (e.g., mechanical ventilation) and patient-centered outcomes (e.g., length of stay, mortality). Six unique pharmacophenotypes were observed, with unique medication profiles and clinically relevant differences in ICU complications and patient-centered outcomes. While pharmacophenotypes 2 and 4 had no statistically significant difference in ICU length of stay, duration of mechanical ventilation, or duration of vasopressor use, their mortality differed significantly (9.0% vs. 21.9%, p < 0.0001). Pharmacophenotype 4 had a mortality rate of 21.9%, compared with the rest of the pharmacophenotypes ranging from 2.5-9%. Phenotyping approaches have shown promise in classifying the heterogenous syndromes of critical illness to predict treatment response and guide clinical decision support systems but have never included comprehensive medication information. This first-ever machine learning approach revealed differences among empirically-derived subgroups of ICU patients that are not typically revealed by traditional classifiers. Identification of pharmacophenotypes may enable enhanced decision making to optimize treatment decisions.

show abstract

Section: Discussionsupporting

confidence: 90%

Cluster analysis driven by unsupervised latent feature learning of intensive care unit medications to identify novel pharmaco-phenotypes of critically ill patients

Newsome¹,

Jeong

Yu³

et al. 2022

Preprint

View full text Add to dashboard Cite

show abstract

“…Secondly, our results clearly demonstrate that MRC scores at 24 hours were associated with mortality suggesting they should be incorporated into current practice ( Table 2 ). Moreover, our pervious proof-of-concept ML model demonstrated that MRCICU scores are associated with ICU mortality[26]. Moreover, MRC has been shown to be a better predictor of mortality compared to polypharmacy alone[32].…”

Section: Discussionmentioning

confidence: 99%

“…The score has undergone validation testing for both internal and external validity [27]. Recently, we developed a proof-of-concept model using machine learning (ML) methods to validate the MRCICU score for improving the prediction of inpatient mortality [26]. Although a MRCICU score can be determined at any time during ICU admission, historical evaluation daily at 24-hour intervals have been most commonly utilized and have been applied to the present study.…”

Section: Mrc Scoring Toolsmentioning

confidence: 99%

“…The modified MRCICU is the first validated quantitative weighted scoring tool intended to predict clinical outcomes (i.e., ICU mortality, length of stay (LOS), and need for mechanical ventilation (MV)) [24,25]. In a previous study, we have developed a novel proof-ofconcept method demonstrating improvement in the prediction of patient outcomes by incorporating the MRCICU score into the previously established Acute Physiology and Chronic Health Evaluation (APACHE II) scoring tool [26]. To date, no study has assessed the association between these two MRC scoring tools and clinical outcomes within the critical care setting.…”

Section: Introductionmentioning

confidence: 99%

See 1 more Smart Citation

Evaluating the Association and Predictability of Complex Medication Regimen Scores with Clinical Outcomes Among the Critically Ill

Al-Mamun

Strock

Sharker

et al. 2022

Preprint

Self Cite

View full text Add to dashboard Cite

Introduction Medication Regimen Complexity (MRC) refers to the combination of medication classes, dosages, and frequencies. The relationship between MRC and clinical outcomes in the intensive care unit (ICU) has not been examined. The objective of this study is to examine the association between MRC scoring tools and their utility in predicting clinical outcomes. Methods We conducted a retrospective cohort study that includes 322 adult patients admitted and stayed (>24 hours) to the ICU between February 1, 2020, and August 30, 2020 in a community-based, teaching hospital in Rhode Island .Medication complexity was assessed using two MRC scoring tools: MRC Index (MRCI) and the MRC in ICU (MRCICU). We used a multivariable logistic regression model to identify the association between MRC scores and clinical outcomes and to predict the clinical outcomes. RESULTS Among the 317 patients included in the study (55.2% men with a median age of 62 [IQR: 51-75] years). Higher MRC scores (i.e., > 63 MRCI or > 6 MRCICU) were associated with increased mortality (14% and 15%), longer ICU length of stay (LOS )(30% and 34%), and need for MV (24% and 28%). MRCICU scores at 24 hours were found to be a significant risk factor in all clinical outcomes (ICU mortality, LOS, and MV) with Odds Ratio (ORs) of 1.12 (95% CI: 1.06-1.19), 1.17 (1.1- 1.24), and 1.21 (1.14- 1.29), respectively. In the prediction models, both MRCI and MRCICU models performed similarly (AUC: 0.88 [0.75-0.97] and 0.88 [0.76-0.97] in predicting mortality. The Medication Model included 15 medication classes outperformed others (AUC: 0.82 [0.71-0.93] in predicting ICU LOS and the MRCICU model outperformed others (AUC: 0.87 [0.77-0.96]) in predicting the need for MV. CONCLUSION MRC scores are associated with poorer clinical outcomes and improves the prediction of poorer clinical outcomes which will support clinicians to prescribe safer therapies.

show abstract

“…The studies that chronicle the development and evaluation of the MRC-ICU are summarized in Supplemental Digital Content -Table 2 (http://links.lww.com/CCM/H141). The primary limitation of all MRC-ICU evaluations to date has been the small sample and one (or two) center designs that inherently lack the robust external validity necessary for widespread use (6)(7)(8)(9)(10)(11)(12)(13)(14)(15).…”

mentioning

confidence: 99%

Impact of Pharmacists to Improve Patient Care in the Critically Ill: A Large Multicenter Analysis Using Meaningful Metrics With the Medication Regimen Complexity-ICU (MRC-ICU) Score*

Sikora

Ayyala

Rech

et al. 2022

Critical Care Medicine

View full text Add to dashboard Cite

OBJECTIVES: Despite the established role of the critical care pharmacist on the ICU multiprofessional team, critical care pharmacist workloads are likely not optimized in the ICU. Medication regimen complexity (as measured by the Medication Regimen Complexity-ICU [MRC-ICU] scoring tool) has been proposed as a potential metric to optimize critical care pharmacist workload but has lacked robust external validation. The purpose of this study was to test the hypothesis that MRC-ICU is related to both patient outcomes and pharmacist interventions in a diverse ICU population. DESIGN: This was a multicenter, observational cohort study. SETTING: Twenty-eight ICUs in the United States. PATIENTS: Adult ICU patients. INTERVENTIONS: Critical care pharmacist interventions (quantity and type) on the medication regimens of critically ill patients over a 4-week period were prospectively captured. MRC-ICU and patient outcomes (i.e., mortality and length of stay [LOS]) were recorded retrospectively. MEASUREMENTS AND MAIN RESULTS:A total of 3,908 patients at 28 centers were included. Following analysis of variance, MRC-ICU was significantly associated with mortality (odds ratio, 1.09; 95% CI, 1.08-1.11; p < 0.01), ICU LOS (β coefficient, 0.41; 95% CI, 00.37-0.45; p < 0.01), total pharmacist interventions (β coefficient, 0.07; 95% CI, 0.04-0.09; p < 0.01), and a composite intensity score of pharmacist interventions (β coefficient, 0.19; 95% CI, 0.11-0.28; p < 0.01). In multivariable regression analysis, increased patient: pharmacist ratio (indicating more patients per clinician) was significantly associated with increased ICU LOS (β coefficient, 0.02; 0.00-0.04; p = 0.02) and reduced quantity (β coefficient, -0.03; 95% CI, -0.04 to -0.02; p < 0.01) and intensity of interventions (β coefficient, -0.05; 95% CI, -0.09 to -0.01). CONCLUSIONS:Increased medication regimen complexity, defined by the MRC-ICU, is associated with increased mortality, LOS, intervention quantity, and intervention intensity. Further, these results suggest that increased pharmacist workload is associated with decreased care provided and worsened patient outcomes, which warrants further exploration into staffing models and patient outcomes.

show abstract

Development of Machine Learning Models to Validate a Medication Regimen Complexity Scoring Tool for Critically Ill Patients

Cited by 28 publications

References 34 publications

Cluster analysis driven by unsupervised latent feature learning of intensive care unit medications to identify novel pharmaco-phenotypes of critically ill patients

Cluster analysis driven by unsupervised latent feature learning of intensive care unit medications to identify novel pharmaco-phenotypes of critically ill patients

Evaluating the Association and Predictability of Complex Medication Regimen Scores with Clinical Outcomes Among the Critically Ill

Impact of Pharmacists to Improve Patient Care in the Critically Ill: A Large Multicenter Analysis Using Meaningful Metrics With the Medication Regimen Complexity-ICU (MRC-ICU) Score*

Contact Info

Product

Resources

About