Kushal Dahal
Current Organizations / Institutes
The organizations/institutes from the scholar's latest article
Kushal Dahal has a total of 23 co-authors. Here are the top co-authors by number of shared publications:
- Michael E. Klepser
Ferris State University
4 shared publications Minji Sohn
2 shared publications- Minji Sohn
College of Pharmacy, Ferris State University, Big Rapids, MI 49307, USA
2 shared publications - Benjamin Pontefract
Ferris State University
1 shared publication David Basali
Stoner Periodontics & Implant Specialists , Dublin, Ohio
1 shared publication- Debra A. Goff
The Ohio State University Wexner Medical Center, Department of Pharmacy , Columbus, Ohio
1 shared publication
Latest Publications
- Open Access
- 10 July 2026
- Published by Elsevier in JAPhA Practice Innovations
- Vol. 3 (4)
Antibiotic prescribing in rural Michigan: Federally Qualified Health Centers (FQHCs) vs. Non-FQHCs
- 10 July 2026
- Published by Elsevier in JAPhA Practice Innovations
- Vol. 3 (4)
Open AccessAntibiotic prescribing in rural Michigan: Federally Qualified Health Centers (FQHCs) vs. Non-FQHCs
- Open Access
- 12 May 2026
- Published by University of Minnesota in INNOVATIONS in pharmacy
- Vol. 17 (1)
Differences by patient race in guideline concordant outpatient antibiotic use in adult patients in Michigan
Purpose: To determine if there are significant differences in the rate of guideline concordant antibiotic use by race in adult patients in the state of Michigan. Methods: Using the Collaboration to Harmonize Antimicrobial Registry Measures (CHARM) database, outpatient prescriptions from 6 health systems throughout the state of Michigan in 2021 were reviewed, and the 5 infectious disease diagnoses with the most antibiotic prescriptions were identified. The 2 most common races amongst patients were identified, and the differences in rates of guideline concordance were compared between them. Results: A total of 62,070 outpatient prescriptions were identified for the 2 most common races, White and Black or African American. Of those, 43,519 outpatient prescriptions were identified for the top 5 infectious diagnoses, acute cystitis, acute sinusitis, acute pharyngitis, cellulitis, and otitis media. Antimicrobial choice concordance was higher in patients who are White (71.6% vs. 68.5%; p<0.01), while antibiotic dosing concordance was higher in patients who are Black or African American (65.9% vs 68.4%; p=0.02). Differences in antimicrobial choice concordance was primary driven by acute cystitis with more patients who are Black or African American receiving doxycycline (1.1 % vs 6.0%; p<0.01). Differences in antimicrobial dose concordance was primary driven by acute pharyngitis with fewer patients who are White receiving concordant doses of amoxicillin (45.5 % vs 64.6%; p<0.01). Conclusion: Differences in rates of antibiotic choice and dosing concordance were seen between patients who are White and Black or African American. Further studies investigating the clinical significance of these differences are warranted. IRB approval: This study was evaluated by the local Institutional Review Board and determined to be exempt because it did not involve human subjects research.
- 12 May 2026
- Published by University of Minnesota in INNOVATIONS in pharmacy
- Vol. 17 (1)
Open AccessDifferences by patient race in guideline concordant outpatient antibiotic use in adult patients in Michigan
Purpose: To determine if there are significant differences in the rate of guideline concordant antibiotic use by race in adult patients in the state of Michigan. Methods: Using the Collaboration to Harmonize Antimicrobial Registry Measures (CHARM) database, outpatient prescriptions from 6 health systems throughout the state of Michigan in 2021 were reviewed, and the 5 infectious disease diagnoses with the most antibiotic prescriptions were identified. The 2 most common races amongst patients were identified, and the differences in rates of guideline concordance were compared between them. Results: A total of 62,070 outpatient prescriptions were identified for the 2 most common races, White and Black or African American. Of those, 43,519 outpatient prescriptions were identified for the top 5 infectious diagnoses, acute cystitis, acute sinusitis, acute pharyngitis, cellulitis, and otitis media. Antimicrobial choice concordance was higher in patients who are White (71.6% vs. 68.5%; p<0.01), while antibiotic dosing concordance was higher in patients who are Black or African American (65.9% vs 68.4%; p=0.02). Differences in antimicrobial choice concordance was primary driven by acute cystitis with more patients who are Black or African American receiving doxycycline (1.1 % vs 6.0%; p<0.01). Differences in antimicrobial dose concordance was primary driven by acute pharyngitis with fewer patients who are White receiving concordant doses of amoxicillin (45.5 % vs 64.6%; p<0.01). Conclusion: Differences in rates of antibiotic choice and dosing concordance were seen between patients who are White and Black or African American. Further studies investigating the clinical significance of these differences are warranted. IRB approval: This study was evaluated by the local Institutional Review Board and determined to be exempt because it did not involve human subjects research.
- Open Access
- 28 April 2026
- Published by Oxford University Press (OUP) in Open Forum Infectious Diseases
- Vol. 13 (5) , ofag287
Sustained Improvement in Antibiotic Prescribing Among Private Practice Dentists Following Antibiotic Stewardship Education and Dashboard Implementation
Background: The lack of procedure-specific, evidence-based dental antibiotic guidelines increases unnecessary and inappropriate antibiotic prescribing. The primary objective was to evaluate changes in antibiotic prescribing before and after antibiotic stewardship (AS) education by infectious diseases (ID) AS experts in addition to implementation of self-monitored antibiotic dashboards. A secondary objective was to link antibiotic prescriptions to standardized Current Dental Terminology (CDT) procedure codes. Methods: We conducted a retrospective, multi-practice cohort study among 9 private practice periodontists who completed AS education by ID-AS experts. Antibiotic data, CDT codes, and medical histories were extracted from electronic dental records from 1 January 2021 through 30 June 2025. Individualized dashboards displayed longitudinal antibiotic prescribing and linkage to CDT procedure codes. Outcomes were compared across pre-education, post-education year 1, and post-education year 3 periods. Results: A total of 11 795 antibiotic prescriptions for 7331 unique patients were identified; 85.6% were therapeutic (≥3 days) and 14.4% were prophylactic. Prophylactic antibiotics for patients with prosthetic joint implants decreased by 89% by post-education year 3 (P < .05). Clindamycin and fluoroquinolone use declined by 85% (P < .001) and 31% (P < .15), respectively. Ten-day antibiotic durations decreased by 87% (P < .001) with a corresponding increase in courses of 3–5 days across multiple CDT procedure categories. Improvements were sustained 3 years after education without ongoing external audit and feedback. Conclusions: Dentists demonstrated sustained improvements in antibiotic use following ID-AS–led stewardship education and implementation of self-monitored antibiotic dashboards. Linking prescriptions to CDT codes supports future development of evidence-based dental antibiotic guidelines.
- 28 April 2026
- Published by Oxford University Press (OUP) in Open Forum Infectious Diseases
- Vol. 13 (5) , ofag287
Open AccessSustained Improvement in Antibiotic Prescribing Among Private Practice Dentists Following Antibiotic Stewardship Education and Dashboard Implementation
Background: The lack of procedure-specific, evidence-based dental antibiotic guidelines increases unnecessary and inappropriate antibiotic prescribing. The primary objective was to evaluate changes in antibiotic prescribing before and after antibiotic stewardship (AS) education by infectious diseases (ID) AS experts in addition to implementation of self-monitored antibiotic dashboards. A secondary objective was to link antibiotic prescriptions to standardized Current Dental Terminology (CDT) procedure codes. Methods: We conducted a retrospective, multi-practice cohort study among 9 private practice periodontists who completed AS education by ID-AS experts. Antibiotic data, CDT codes, and medical histories were extracted from electronic dental records from 1 January 2021 through 30 June 2025. Individualized dashboards displayed longitudinal antibiotic prescribing and linkage to CDT procedure codes. Outcomes were compared across pre-education, post-education year 1, and post-education year 3 periods. Results: A total of 11 795 antibiotic prescriptions for 7331 unique patients were identified; 85.6% were therapeutic (≥3 days) and 14.4% were prophylactic. Prophylactic antibiotics for patients with prosthetic joint implants decreased by 89% by post-education year 3 (P < .05). Clindamycin and fluoroquinolone use declined by 85% (P < .001) and 31% (P < .15), respectively. Ten-day antibiotic durations decreased by 87% (P < .001) with a corresponding increase in courses of 3–5 days across multiple CDT procedure categories. Improvements were sustained 3 years after education without ongoing external audit and feedback. Conclusions: Dentists demonstrated sustained improvements in antibiotic use following ID-AS–led stewardship education and implementation of self-monitored antibiotic dashboards. Linking prescriptions to CDT codes supports future development of evidence-based dental antibiotic guidelines.
- Open Access
- 27 February 2026
- Published by MDPI AG in Applied Sciences
- Vol. 16 (5) , 2300
Feasibility of Using Large Language Models for Structured Medication Extraction from Clinical Text: A Comparative Analysis of Zero-Shot and Few-Shot Paradigms
The digitization of healthcare has been accompanied by a rapid expansion of electronic health records (EHRs); however, a significant proportion of critical patient data, specifically medication regimens, remains entrapped within unstructured clinical narratives. The inability to seamlessly compute this data hinders advancements in pharmacovigilance, clinical decision support, and population health management. This study presents a comprehensive, rigorous evaluation of the feasibility of deploying Large Language Models (LLMs) to automate the extraction of structured dosage information (Dose, Daily Frequency, Duration) from outpatient antimicrobial clinical notes sourced from the Collaboration to Harmonize Antimicrobial Registry Measures (CHARM) registry. We scrutinized the performance of five distinct open-weight architectures, namely GPT-OSS:20B, Gemma 2:9B, Mistral 7B, Qwen3:14B and Llama 3.2, across both Zero-Shot and Retrieval Augmented Generation (RAG)-based Few-Shot prompting paradigms. Our analysis reveals a fundamental architectural trade-off: the reasoning-optimized GPT-OSS:20B dominates the zero-shot landscape (F1 > 0.90) by leveraging abstract schema understanding, whereas the instruction-tuned Gemma 2:9B excels in the few-shot setting (F1 ~ 0.99), effectively utilizing examples as guardrails to surpass larger models. Conversely, smaller models (Mistral, Llama) exhibit a prohibitive “hallucination barrier,” rendering them unsafe for unsupervised clinical application. Furthermore, we identify “Inconsistent Unit Handling” and “Complex Temporal Logic” as persistent failure modes that resist simple scaling laws. This report provides a definitive framework for selecting model architectures based on the availability of few-shot examples and highlights the necessity of dynamic RAG strategies to achieve production-grade reliability in medical informatics.
- 27 February 2026
- Published by MDPI AG in Applied Sciences
- Vol. 16 (5) , 2300
Open AccessFeasibility of Using Large Language Models for Structured Medication Extraction from Clinical Text: A Comparative Analysis of Zero-Shot and Few-Shot Paradigms
The digitization of healthcare has been accompanied by a rapid expansion of electronic health records (EHRs); however, a significant proportion of critical patient data, specifically medication regimens, remains entrapped within unstructured clinical narratives. The inability to seamlessly compute this data hinders advancements in pharmacovigilance, clinical decision support, and population health management. This study presents a comprehensive, rigorous evaluation of the feasibility of deploying Large Language Models (LLMs) to automate the extraction of structured dosage information (Dose, Daily Frequency, Duration) from outpatient antimicrobial clinical notes sourced from the Collaboration to Harmonize Antimicrobial Registry Measures (CHARM) registry. We scrutinized the performance of five distinct open-weight architectures, namely GPT-OSS:20B, Gemma 2:9B, Mistral 7B, Qwen3:14B and Llama 3.2, across both Zero-Shot and Retrieval Augmented Generation (RAG)-based Few-Shot prompting paradigms. Our analysis reveals a fundamental architectural trade-off: the reasoning-optimized GPT-OSS:20B dominates the zero-shot landscape (F1 > 0.90) by leveraging abstract schema understanding, whereas the instruction-tuned Gemma 2:9B excels in the few-shot setting (F1 ~ 0.99), effectively utilizing examples as guardrails to surpass larger models. Conversely, smaller models (Mistral, Llama) exhibit a prohibitive “hallucination barrier,” rendering them unsafe for unsupervised clinical application. Furthermore, we identify “Inconsistent Unit Handling” and “Complex Temporal Logic” as persistent failure modes that resist simple scaling laws. This report provides a definitive framework for selecting model architectures based on the availability of few-shot examples and highlights the necessity of dynamic RAG strategies to achieve production-grade reliability in medical informatics.