Introduction

Over the past two decades, critical care pharmacists (CCPs) have become essential members of interdisciplinary teams caring for critically ill patients.¹ CCP interventions have shown substantial clinical value, including cost avoidance, mitigating adverse drug events, addressing complex drug-drug interactions (DDIs), and reducing medication errors.^2,3,4 Furthermore, these interventions correlate with traditional patient outcomes, influencing factors such as hospital and intensive care unit (ICU) length of stay (LOS), duration of mechanical ventilation, and mortality.^5,6,7

Quantifying the complexities of medication management in critical care is crucial for effective CCP resource allocation. The Medication Regimen Complexity-ICU (MRC-ICU) score serves as a novel tool, consisting of a 39-line item of medications and devices that quantitatively describes medication regimen complexity by scoring and weighting each individual line item.⁸ This score considers the number and type of medications, administration routes, and devices. The MRC-ICU is the sum of all the points assigned to the medications. Effective management of complex medication regimens is essential for preventing adverse drug events, optimizing therapeutic outcomes, and ensuring the efficient use of healthcare resources.

Recent studies have emphasized the importance of the MRC-ICU, demonstrating its association with key clinical outcomes such as mortality, lengths of ICU and hospital stay as well as other outcomes including severity of illness scores, drug interactions, fluid overload, and pharmacist workload.^9,10,11,12 The MRC-ICU provides insights into the complex pharmacotherapeutic care in ICU settings, revealing important factors that influence patient outcomes. However, while the MRC-ICU is promising in forecasting patient outcomes, the possible link between the cost-saving impact of pharmacist interventions and MRC-ICU scores remains unexplored.

This study aimed to examine the association between medication regimen complexity, as measured by the MRC-ICU score, and total cost avoidance (TCA) attributable to pharmacist interventions in a Chilean ICU. Secondary objectives were to identify and quantify the main types of pharmacist interventions contributing to cost avoidance and to evaluate their impact on ICU length of stay and mortality. By integrating economic and clinical endpoints, this study seeks to characterize the overall value of pharmacist-led optimization of complex medication regimens in the ICU setting.

Methods

Study design and setting

This retrospective cohort study was approved by the Ethics Committee of the Servicio de Salud Valdivia (Chile) prior to data collection, and the requirement for individual informed consent was waived given the observational, retrospective, and de-identified nature of the analysis. The study was conducted in the adult ICU of Hospital Base Valdivia, a 500-bed tertiary teaching hospital in southern Chile. The unit has 18 beds and a part-time critical care pharmacist (CCP) integrated into the multidisciplinary team, providing clinical services from 08:00 to 13:00, Monday to Friday. The study period covered five consecutive months (January–May 2022).

Participants and data collection

A convenience sample of ICU admissions was used. Patients aged ≥15 years admitted to the adult ICU during the study period were eligible for inclusion if they had at least one clinical pharmacist intervention that was accepted and implemented by the treating team. The age cutoff of ≥15 years reflects the operational definition of an adult patient applied at our tertiary public hospital, where pediatric ICU services receive patients under 15 years of age. Exclusion criteria were ICU stay <24 hours prior to death and incomplete medical records. Each ICU admission was analyzed as a separate observation; readmissions of the same patient were included and treated as distinct admissions.

By design, ICU admissions without accepted and implemented clinical pharmacist interventions were not included in the analytic cohort and therefore did not contribute to the estimation of total cost avoidance. Consequently, the study population should be interpreted as a pharmacist-treated subset of ICU admissions rather than the entire ICU population. Accordingly, the analysis was intended to explore the association between early medication regimen complexity and estimated cost avoidance among admissions with documented pharmacist activity, and not to determine the predictive performance of the MRC-ICU score across all ICU admissions.

Clinical pharmacist interventions were recorded prospectively using a standardized format and documented in the local electronic record (Microsoft Excel, 2023 version), where each intervention was logged with date, intervention category, target medication, and acceptance status. Only accepted and implemented interventions were considered for the economic analysis. From the electronic medical record, we extracted age, sex, primary ICU admission diagnosis, the MRC-ICU score calculated from medications and devices prescribed within the first 24 hours after ICU admission, and the Sequential Organ Failure Assessment (SOFA) score at ICU admission. Outcomes included ICU length of stay (LOS, days) and ICU mortality.

Medication regimen complexity (MRC-ICU)

Medication regimen complexity was quantified with the MRC-ICU score, a weighted index that sums predefined medication classes and modalities prescribed within the first 24 hours of ICU admission; higher scores indicate greater complexity. Standard components (e.g., continuous opioids/sedatives, stress-ulcer prophylaxis, venous thromboembolism prophylaxis, antimicrobials, insulin, anticonvulsants) were scored according to the original instrument. Vasoactive infusions, although recorded as a marker of clinical complexity, are not part of the original MRC-ICU and were not included in the score calculation.

Economic evaluation and total cost avoidance

Total cost avoidance (TCA) attributable to CCP interventions was estimated using the structured framework described by Hammond et al.¹⁴ Each accepted and implemented intervention was assigned to one of the predefined intervention categories, such as antimicrobial optimization, dose adjustment, prevention of adverse drug events, or discontinuation of unnecessary therapy, and the corresponding literature-derived cost-avoidance value was applied. These values represent estimated direct hospital costs avoided, including drug acquisition, monitoring, preventable adverse events, and additional ICU or ward days, rather than patient out-of-pocket expenses.

Because TCA was calculated from accepted CCP interventions, it was directly determined by both the number of interventions performed during each ICU admission and the assigned cost-avoidance value of each intervention category. Interventions without a matching category in the Hammond framework, or those considered cost-neutral, were assigned a value of zero. For each ICU admission, TCA was calculated as the sum of the cost-avoidance values assigned to all accepted and implemented CCP interventions during that admission.

Original cost-avoidance estimates reported in 2018 US dollars were converted to Chilean pesos using historical exchange rates and then adjusted to 2022 values using national inflation indices. Final estimates are reported as 2022 US dollars using the average 2022 exchange rate. Accordingly, TCA should be interpreted as a standardized estimate derived from documented pharmacist activity, rather than as an outcome independent of intervention frequency or intervention type.

Statistical analysis

Continuous variables are presented as mean (standard deviation, SD) or median (interquartile range, IQR), and categorical variables as counts and percentages. Spearman’s rank correlation coefficients were calculated between TCA and MRC-ICU score, age, sex, SOFA score at ICU admission, ICU LOS, ICU mortality, and primary ICU diagnosis. To address potential confounding, we additionally fitted a multivariable linear regression model with log-transformed TCA as the dependent variable (justified by the right-skewed distribution of cost data, in line with established statistical practice in healthcare cost analysis¹⁹) and MRC-ICU score, ICU LOS, SOFA score, and age as independent predictors. Standardized beta coefficients, 95% confidence intervals, and adjusted R² were reported. Model assumptions were verified by inspection of residual normality (Shapiro–Wilk test), homoscedasticity (Breusch–Pagan test), absence of autocorrelation (Durbin–Watson statistic), and absence of multicollinearity (variance inflation factor <2 for all predictors). A pre-specified subgroup analysis was performed comparing COVID-19 versus non-COVID-19 admissions to assess the consistency of the MRC-ICU–TCA association across case-mix categories, given the high proportion of pandemic-related admissions during the study period. Two-sided p-values <0.05 were considered statistically significant, and correlation coefficients were interpreted primarily by their magnitude to avoid over-stating weak associations. All analyses were performed using R (R Foundation for Statistical Computing) with RStudio version 2023.03.0+386.

Results

Patient characteristics

A total of 124 ICU admissions corresponding to 121 unique patients were included; three patients experienced a readmission to the ICU during the study period and were analyzed as separate admissions. All outcomes are reported at the admission level. The mean age was 56.9 years (SD 16.3), and 42.7% of admissions involved female patients. The most frequent ICU admission diagnoses were COVID-19 pneumonia (29.0%) and septic shock (17.7%), followed by status epilepticus (7.3%) and other causes (43.6%). Illness severity at ICU admission, as reflected by the SOFA score, is summarized in Table 2.

Medication regimen complexity and pharmacist interventions

Medication regimens were highly complex in this CCP-treated cohort. The mean MRC-ICU score at 24 hours after ICU admission was 20.4 (SD 8.5) (Table 2). The most frequent contributors to MRC-ICU complexity were continuous infusions of opioids/sedatives (96.8% of admissions), stress-ulcer prophylaxis (95.2%), and deep-vein thrombosis prophylaxis (58.9%), along with antimicrobials (43.6%), subcutaneous insulin (26.6%), anticonvulsants (25.0%), restricted antimicrobials (24.2%), and other continuous infusions (16.9%) (Table 1). Vasoactive continuous infusions were present in 91.9% of admissions and are an important marker of clinical complexity, although they are not part of the original MRC-ICU scoring tool.

Within the first 24 hours of ICU admission, dexamethasone, hydrocortisone, and human albumin were also frequently administered, reflecting intensive early pharmacologic management. These agents are not included in the MRC-ICU instrument and were not incorporated into TCA calculations. Overall, CCPs performed 531 interventions across the 124 ICU admissions, corresponding to a mean of 4.2 interventions per admission (SD 3.6) (Table 2).

Total cost avoidance and primary outcome

Mean TCA per ICU admission was 1,785.8 USD (2022 dollars), corresponding to a total estimated cost avoidance of 221,444 USD over the five-month study period (Table 2). Because TCA was calculated as the sum of assigned cost-avoidance values for accepted and implemented CCP interventions, higher TCA reflects either a greater number of accepted interventions, higher-value intervention categories, or both. The primary analysis showed a weak but statistically significant positive association between MRC-ICU score and TCA (Spearman rs = 0.246, p = 0.006) (Table 3). After adjustment for ICU LOS, SOFA, and age in a multivariable linear regression of log-transformed TCA, ICU LOS (β = 0.052, 95% CI 0.030–0.073; p < 0.001) and SOFA (β = 0.079, 95% CI 0.016–0.141; p = 0.014) remained independently associated with cost avoidance, whereas the MRC-ICU score showed a non-significant trend (β = 0.025, 95% CI −0.002–0.053; p = 0.073) (Table 4). In a parsimonious model adjusting only for ICU LOS, MRC-ICU remained independently associated with log(TCA) (β = 0.037, 95% CI 0.011–0.064; p = 0.006; adjusted R² = 0.17). Therefore, ICU admissions with more complex early medication regimens tended to have higher estimated TCA from CCP interventions; however, the magnitude of this association was modest and should be interpreted as exploratory rather than as evidence of an independent predictive relationship.

Secondary outcomes and associations with TCA

Higher TCA was associated with longer ICU LOS in the correlation analysis (rs = 0.425, p < 0.001), whereas the association with ICU mortality did not reach statistical significance (rs = 0.153, p = 0.089) (Table 3). These findings suggest that admissions with greater estimated cost avoidance tended to correspond to patients with longer ICU stays, while no significant unadjusted association was observed with ICU mortality. However, because TCA is derived from accepted CCP interventions, these associations should be interpreted cautiously and may reflect greater clinical complexity, longer exposure time for pharmacist involvement, and/or a higher number or value of accepted interventions, rather than an independent relationship between TCA and clinical outcomes.

Age and sex were not significantly correlated with TCA (p = 0.516 and p = 0.185, respectively). The SOFA score at ICU admission showed a weak but statistically significant correlation with TCA (rs = 0.224, p = 0.012). Overall ICU outcomes in the cohort are shown in Table 2. The mean ICU LOS was 13.9 days (SD 9.1), and ICU mortality was 29.0% (36/124 admissions).

Subgroup analysis: COVID-19 versus non-COVID-19 admissions

To examine whether the case-mix during the COVID-19 pandemic period influenced the observed associations, we conducted a pre-specified subgroup analysis. Of the 124 included admissions, 36 (29.0%) had a primary diagnosis of COVID-19 pneumonia and 88 (71.0%) corresponded to non-COVID-19 ICU diagnoses. The two subgroups differed in baseline pharmacotherapeutic complexity: mean MRC-ICU was lower in COVID-19 admissions (22.4, SD 4.5) than in non-COVID-19 admissions (19.5, SD 8.1; Mann–Whitney p = 0.001), and median ICU LOS was longer in COVID-19 admissions (Mann–Whitney p = 0.002). Mean TCA was similar in both groups (1,850 vs. 1,759 USD; p = 0.40). The MRC-ICU–TCA correlation was stronger and statistically significant in non-COVID-19 admissions (rs = 0.315, p = 0.003), but was weak and non-significant in COVID-19 admissions (rs = 0.137, p = 0.426). These results suggest that the modest overall MRC-ICU–TCA association observed in the entire cohort was driven primarily by non-COVID-19 ICU admissions, indicating that the case-mix during the pandemic may have attenuated the predictive performance of MRC-ICU for cost avoidance in this specific subgroup.

Discussion

This is, to our knowledge, the first study to examine the association between MRC-ICU and cost avoidance, and the first evaluation of the MRC-ICU in an ICU outside the United States. By applying this score in a Chilean ICU and linking it to a structured cost-avoidance framework, our work adds to the existing evidence on MRC-ICU and extends prior findings by incorporating the economic dimension of pharmacist involvement in the management of complex ICU pharmacotherapy.^9-13,15 In this context, MRC-ICU functions as a weighted index that summarizes the burden of high-risk and resource-intensive medications prescribed in the first 24 hours of ICU admission, with higher scores reflecting greater regimen complexity and a higher need for targeted pharmacist oversight.^9-13

In Chile, critical care pharmacists (CCPs) have traditionally focused on bedside pharmacotherapy monitoring and prevention of adverse drug events, particularly involving antimicrobials and sedatives.¹⁵ These medication classes are important contributors to pharmacotherapeutic complexity and have been associated with increased requirements for clinical intervention and adverse clinical outcomes, including prolonged ICU stay and mortality.^16,17 Within this practice model, the observed weak positive correlation between MRC-ICU and TCA, as well as the association of higher TCA with longer ICU length of stay and ICU mortality, is consistent with a pattern in which patients with greater pharmacotherapeutic and clinical complexity may generate more opportunities for pharmacist intervention and, consequently, higher estimated cost avoidance.^9–13,16,17

However, this association should be interpreted cautiously. The modest correlation coefficient between MRC-ICU and TCA, approximately 0.25, indicates that early medication regimen complexity accounts for only a limited proportion of the variability in estimated cost avoidance. Other clinical, organizational, and workflow-related factors are therefore likely to influence the occurrence, type, acceptance, and economic value of CCP interventions. To address potential confounding raised in editorial review, we extended the bivariate analysis with a multivariable linear regression of log-transformed TCA on MRC-ICU score, ICU LOS, SOFA, and age.¹⁹ In this model, ICU LOS and SOFA emerged as independent predictors of cost avoidance (p < 0.001 and p = 0.014, respectively), and MRC-ICU showed a borderline non-significant trend (p = 0.073) likely reflecting shared variance with SOFA (Pearson r = 0.33 between MRC-ICU and SOFA in our cohort), a correlation consistent with previously reported associations between MRC-ICU and severity-of-illness measures in critically ill adults.²⁰ In a parsimonious model adjusting only for ICU LOS, MRC-ICU remained independently associated with TCA (β = 0.037, p = 0.006). Taken together, these results suggest that early medication regimen complexity contributes to estimated cost avoidance independently of length of stay, but its incremental contribution beyond severity of illness as captured by SOFA is small. Because TCA was calculated as the sum of assigned cost-avoidance values for accepted and implemented CCP interventions, it remains intrinsically linked to pharmacist activity. Thus, patients with more complex medication regimens may require more intensive pharmacist review, generate a greater number of interventions or higher-value interventions, and consequently accumulate higher estimated TCA. This structural relationship should be considered when interpreting the association between MRC-ICU and cost avoidance.

ICU length of stay is particularly important in this context. Longer ICU stays provide more time for medication changes, clinical deterioration or recovery, pharmacist review, and intervention. In our multivariable analysis, ICU LOS was the strongest independent predictor of log-transformed TCA, consistent with the interpretation that exposure time materially shapes the magnitude of accumulated cost avoidance, in line with previous large multicenter cost-avoidance evaluations in critical care pharmacy.²¹ Importantly, MRC-ICU remained independently associated with TCA after adjustment for ICU LOS, indicating that early medication regimen complexity contributes to cost avoidance beyond the simple effect of longer pharmacist exposure. However, because ICU length of stay occurs after the initial MRC-ICU assessment and during the same period in which TCA accumulates, it should be regarded as a downstream variable that partially mediates the relationship between early complexity and cumulative pharmacist activity, rather than a strict baseline confounder; the multivariate estimate adjusted for LOS therefore represents a conservative lower bound of the independent association between MRC-ICU and TCA.

An additional consideration is the way cost avoidance was estimated. We used the intervention categories and literature-derived cost-avoidance values described by Hammond et al. and applied them to our Chilean ICU context, converting 2018 U.S. dollar estimates to local currency and updating them to 2022 before expressing results again in U.S. dollars for international comparability.¹⁴ Only accepted interventions that mapped to predefined categories received a non-zero value; cost-neutral or non-mappable interventions were assigned zero. Furthermore, our cohort included only ICU admissions with at least one accepted CCP intervention, so “negative” cases without pharmacist involvement—and with unmeasured MRC-ICU and TCA—were not captured. These methodological choices imply that our TCA estimates should be interpreted as standardized approximations of potential cost avoidance in a pharmacist-treated subset, rather than precise, generalizable savings across all ICU admissions.^14–17

The challenges faced by ICUs worldwide during the COVID-19 pandemic have emphasized the importance of effective and efficient medication management.^18,22 Many units, including ours, cared for a disproportionate number of critically ill patients with highly complex pharmacotherapy, often under resource constraints. The high proportion of COVID-19 admissions in our cohort (29.0%) raised concerns about the external validity of the overall findings to non-pandemic ICU practice. Our pre-specified subgroup analysis showed that the MRC-ICU–TCA association was statistically significant in non-COVID-19 admissions (rs = 0.315, p = 0.003) but not in the COVID-19 subgroup (rs = 0.137, p = 0.426). Notably, COVID-19 admissions had lower mean MRC-ICU scores (22.4 vs. 19.5) but longer ICU LOS than non-COVID-19 admissions, consistent with a relatively homogeneous early pharmacotherapy pattern (sedation, prophylaxis, and antimicrobial bundles) in pandemic patients despite their prolonged stays.²² This pattern may have compressed early-complexity heterogeneity in the COVID-19 subgroup and attenuated the MRC-ICU–TCA correlation, while the broader spectrum of non-COVID-19 diagnoses preserved the gradient between regimen complexity and pharmacist activity.²² These observations support the interpretation that our findings are likely applicable to non-pandemic ICU practice and that the overall correlation in the full cohort represents a conservative estimate of the relationship that would be expected in a contemporary, post-pandemic case-mix. In this setting, our findings align with prior work showing that ICU pharmacist participation is associated with both clinical and economic benefits,^9–13,15,21 and suggest that tools such as MRC-ICU may help to systematically identify patients in whom pharmacist time is likely to yield greater value.^9–13,20 However, differences in organization, financing, and drug pricing between a public Chilean tertiary hospital and U.S. centers limit direct extrapolation of absolute cost figures and underline the need for locally calibrated economic models.^14–18,21

This study has several limitations. First, its retrospective, single-center design and the use of a convenience sample limit external validity. In addition, the analytic cohort was restricted to ICU admissions with at least one accepted and implemented CCP intervention. This design introduces selection bias because ICU admissions without documented pharmacist interventions were excluded and, by definition, would have contributed a TCA of zero. Therefore, the cohort represents a pharmacist-treated subset of ICU admissions rather than the broader ICU population. Accordingly, the observed association between MRC-ICU and TCA should be interpreted as exploratory and limited to patients with documented pharmacist activity, rather than as evidence that MRC-ICU predicts economic impact across all ICU admissions.

Residual confounding is also likely. Although MRC-ICU captures early medication regimen complexity, it does not account for all clinical, organizational, or workflow-related factors that may influence the number, type, acceptance, or economic value of CCP interventions. Although we performed a multivariable regression to adjust for ICU LOS, SOFA, and age, we did not have detailed intervention-level information regarding intervention frequency or specific intervention category at the per-admission level, and we therefore could not directly evaluate whether MRC-ICU remained associated with TCA after accounting for these activity-related variables, nor did we assess the differential contribution of specific intervention types to cost avoidance or clinical outcomes.^{9–18,20–22} Finally, TCA estimates should be interpreted as standardized approximations derived from a published cost-avoidance framework, rather than as direct measurements of actual institutional savings.

Nevertheless, this study provides an initial framework for linking medication regimen complexity, pharmacist activity, and standardized cost avoidance in an ICU setting outside the U.S. Future prospective, multicenter studies should include all ICU admissions, incorporate locally derived cost data, and collect intervention-level information on timing, category, acceptance, clinical relevance, and economic value. Such studies could better determine whether MRC-ICU has value as a prioritization tool for CCP involvement across diverse health systems.^{9–18,20–22}

Conclusion

In this retrospective cohort of ICU admissions with at least one accepted and implemented clinical pharmacist intervention, higher early medication regimen complexity was weakly but independently associated with greater estimated total cost avoidance after adjustment for ICU length of stay. These findings suggest that the MRC-ICU score may help characterize pharmacotherapeutic complexity and potential economic impact within pharmacist-treated ICU patients. However, because admissions without pharmacist interventions were not included, the results should not be interpreted as evidence that MRC-ICU predicts cost avoidance across the broader ICU population. Prospective multicenter studies including all ICU admissions, locally derived cost estimates, and detailed characterization of pharmacist interventions are needed to define the role of MRC-ICU as a prioritization tool in critical care pharmacy practice.

Funding

This research did not receive any specific grant from funding agencies in the public, commercial, or not-for-profit sectors.

Author contributions

Magdalena Latorre-Harcha: Data collection and final revision

Daniel Muñoz-Pichuante: Manuscript redaction

Lorenzo Villa-Zapata: Data analysis and conclusions

Andrea Sikora: Manuscript redaction and final revision

Susan Smith: Manuscript redaction and final revision

Conflicts of interest

The authors declare no conflict of interest.

Data availability

The data underlying this article will be shared on reasonable request to the corresponding author.

References

1. Donovan AL, Aldrich JM, Gross AK, Barchas DM, Thornton KC, Schell-Chaple HM, Gropper MA, Lipshutz AKM, University of California, San Francisco Critical Care Innovations Group. Interprofessional Care and Teamwork in the ICU. Crit Care Med. 2018;46(6):980-990. Available from: https://doi.org/10.1097/CCM.0000000000003067
2. Bakker T, Abu-Hanna A, Dongelmans DA, Vermeijden WJ, Bosman RJ, de Lange DW, Klopotowska JE, de Keizer NF, SIMPLIFY study group, Hendriks S, Ten Cate J, Schutte PF, van Balen D, Duyvendak M, Karakus A, Sigtermans M, Kuck EM, Hunfeld NGM, van der Sijs H, de Feiter PW, … Wesselink E. Clinically relevant potential drug-drug interactions in intensive care patients: A large retrospective observational multicenter study. J Crit Care. 2021;62:124-130. Available from: https://doi.org/10.1016/j.jcrc.2020.11.020
3. Leape LL, Cullen DJ, Clapp MD, Burdick E, Demonaco HJ, Erickson JI, Bates DW. Pharmacist participation on physician rounds and adverse drug events in the intensive care unit. JAMA. 1999;282(3):267-270. Available from: https://doi.org/10.1001/jama.282.3.267
4. Wang T, Benedict N, Olsen KM, Luan R, Zhu X, Zhou N, Tang H, Yan Y, Peng Y, Shi L. Effect of critical care pharmacist’s intervention on medication errors: A systematic review and meta-analysis of observational studies. J Crit Care. 2015;30(5):1101-1106. Available from: https://doi.org/10.1016/j.jcrc.2015.06.018
5. Zampieri FG, Salluh JIF, Azevedo LCP, Kahn JM, Damiani LP, Borges LP, Viana WN, Costa R, Corrêa TD, Araya DES, Maia MO, Ferez MA, Carvalho AGR, Knibel MF, Melo UO, Santino MS, Lisboa T, Caser EB, Besen BAMP, Bozza FA, … ORCHESTRA Study Investigators. ICU staffing feature phenotypes and their relationship with patients’ outcomes: an unsupervised machine learning analysis. Intensive Care Med. 2019;45(11):1599-1607. Available from: https://doi.org/10.1007/s00134-019-05790-z
6. Lee H, Ryu K, Sohn Y, Kim J, Suh GY, Kim E. Impact on Patient Outcomes of Pharmacist Participation in Multidisciplinary Critical Care Teams: A Systematic Review and Meta-Analysis. Crit Care Med. 2019;47(9):1243-1250. Available from: https://doi.org/10.1097/CCM.0000000000003830
7. Saadah LM, Khan AH, Syed Sulaiman SA, Bashiti IA. Maximizing acceptance of clinical pharmacy recommendations to reduce length of hospital stay in a private hospital from Amman, Jordan. BMC Health Serv Res. 2021;21(1):937. Available from: https://doi.org/10.1186/s12913-021-06966-4
8. Gwynn ME, Poisson MO, Waller JL, Newsome AS. Development and validation of a medication regimen complexity scoring tool for critically ill patients. Am J Health-Syst Pharm. 2019;76(Supplement_2):S34-S40. Available from: https://doi.org/10.1093/ajhp/zxy054
9. Olney WJ, Chase AM, Hannah SA, Smith SE, Newsome AS. Medication Regimen Complexity Score as an Indicator of Fluid Balance in Critically Ill Patients. J Pharm Pract. 2022;35(4):573-579. Available from: https://doi.org/10.1177/0897190021999792
10. Newsome AS, Smith SE, Olney WJ, et al. Medication regimen complexity is associated with pharmacist interventions and drug-drug interactions: A use of the novel MRC-ICU scoring tool. J Am Coll Clin Pharm. 2020;3:47-56. Available from: https://doi.org/10.1002/jac5.1146
11. Sikora A, Ayyala D, Rech MA, Blackwell SB, Campbell J, Caylor MM, Condeni MS, DePriest A, Dzierba AL, Flannery AH, Hamilton LA, Heavner MS, Horng M, Lam J, Liang E, Montero J, Murphy D, Plewa-Rusiecki AM, Sacco AJ, Sacha GL, … MRC-ICU Investigator Team. 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. Crit Care Med. 2022;50(9):1318-1328. Available from: https://doi.org/10.1097/CCM.0000000000005585
12. Smith SE, Shelley R, Sikora A. Medication regimen complexity vs patient acuity for predicting critical care pharmacist interventions. Am J Health-Syst Pharm. 2022;79(8):651-655. Available from: https://doi.org/10.1093/ajhp/zxab460
13. Al-Mamun MA, Brothers T, Newsome AS. Development of Machine Learning Models to Validate a Medication Regimen Complexity Scoring Tool for Critically Ill Patients. Ann Pharmacother. 2021;55(4):421-429. Available from: https://doi.org/10.1177/1060028020959042
14. Hammond DA, Gurnani PK, Flannery AH, Smetana KS, Westrick JC, Lat I, Rech MA. Scoping Review of Interventions Associated with Cost Avoidance Able to Be Performed in the Intensive Care Unit and Emergency Department. Pharmacotherapy. 2019;39(3):215-231. Available from: https://doi.org/10.1002/phar.2224
15. Muñoz-Pichuante D, Villa-Zapata L. Benefit of Incorporating Clinical Pharmacists in an Adult Intensive Care Unit: A Cost-saving Study. J Clin Pharm Ther. 2020;45(5):1127-1133. Available from: https://doi.org/10.1111/jcpt.13195
16. Ceric A, Holgersson J, May T, Skrifvars MB, Hästbacka J, Saxena M, et al. Level of sedation in critically ill adult patients: a protocol for a systematic review with meta-analysis and trial sequential analysis. BMJ Open. 2022;12(9):e061806. Available from: https://doi.org/10.1136/bmjopen-2022-061806
17. Garin N, Sole N, Lucas B, Matas L, Moras D, Rodrigo-Troyano A, et al. Drug related problems in clinical practice: a cross-sectional study on their prevalence, risk factors and associated pharmaceutical interventions. Sci Rep. 2021;11(1):883. Available from: https://doi.org/10.1038/s41598-020-80560-2
18. Erstad BL. Caring for the COVID Patient: A Clinical Pharmacist’s Perspective. Ann Pharmacother. 2021;55(3):413-414. Available from: https://doi.org/10.1177/1060028020954224
19. Manning WG, Mullahy J. Estimating log models: to transform or not to transform? J Health Econ. 2001;20(4):461-494. Available from: https://doi.org/10.1016/s0167-6296(01)00086-8
20. Webb AJ, Carver B, Rowe S, Sikora A. The use of electronic health record embedded MRC-ICU as a metric for critical care pharmacist workload. JAMIA Open. 2023;6(4):ooad101. Available from: https://doi.org/10.1093/jamiaopen/ooad101
21. Rech MA, Gurnani PK, Peppard WJ, Smetana KS, Van Berkel MA, Hammond DA, Flannery AH. PHarmacist Avoidance or Reductions in Medical Costs in CRITically Ill Adults: PHARM-CRIT Study. Crit Care Explor. 2021;3(12):e0594. Available from: https://doi.org/10.1097/CCE.000000000000059
22. Azimi H, Johnson L, Loudermilk C, Chase A, Forehand CC, Sikora A. Medication regimen complexity (MRC-ICU) for in-hospital mortality prediction in COVID-19 patients. Hosp Pharm. 2023;58(6):564-568. Available from: https://doi.org/10.1177/00185787231169460