Global evidence meets local reality: Collaborative medication safety
Authors: Jehath Syed¹, Prathiba Pereira², Sri Harsha Chalasani³
1. School of Health Sciences and Technology, Dr. Vishwanath Karad MIT World Peace University, Pune, India
2. JSS Medical College & Hospital, JSS Academy of Higher Education and Research, Mysuru, India
3. JSS College of Pharmacy, JSS Academy of Higher Education and Research, Mysuru, India
Introduction: When I discovered almost every patient was at risk
During my doctoral research as a geriatric clinical pharmacist in an 1800-bed tertiary care hospital in India, I witnessed an alarming trend. Our outpatient department serves over 200 older adult patients daily and I noticed a pattern: many older adult patients were taking medications that, although commonly prescribed, could be harmful to older adults.
The numbers confirmed these observations. In our study of 597 hospitalised older adults, we found that almost all (over 99%) were prescribed medications that experts call ‘potentially inappropriate medications’ or PIMs. These are not necessarily wrong medications, but they are drugs that carry higher risks for older people as opposed to younger adults. Studies show that 21% to 42% of emergency department visits by older adults are linked to the use of these medications.
To help identify these PIMs, healthcare providers worldwide use criteria such as Beers Criteria, STOPP and FORTA. These are safety checklists developed by experts to flag medications that might be problematic for older adult patients. However, when I attempted to implement these criteria in our busy Indian hospital, the disparity between the recommendations and our practical limitations became evident. These were not just statistics; they represented real patients at real risk and we needed practical solutions that worked in our reality.
The problem: When guidelines clash with reality
Our research revealed a troubling contradiction: the three main guideline tools we used often disagreed with each other regarding which medications were problematic. While Beers and STOPP showed reasonable agreement, FORTA identified completely different medications.
This creates confusion for healthcare providers. How do we explain to patients and families that experts themselves cannot agree on what is safe? How do we communicate this uncertainty while still making treatment decisions?
The challenges were more profound. Several drugs that the 2023 Beers Criteria removed as ‘rarely used’ in Western countries, such as chlorpropamide (a diabetes medication), methyldopa (for blood pressure) and reserpine, are still commonly prescribed in India.
When I discussed alternatives with physicians, they asked practical questions that the guidelines could not answer: ‘Is the safer alternative affordable for this patient? Is it available at our pharmacy? Will our insurance cover it?’
There was also a time-related problem. Manually checking each patient's medications against Beers (five categories), STOPP (over 80 items) and FORTA (299 entries) is time-consuming and error-prone. In our busy clinics, where doctors have only 10 minutes per patient, this is simply impossible.
The search for answers
To address these barriers, we created De-Med, a web-based tool that instantly identifies PIMs across all three criteria.
- Tested on 231 patient cases, with 100% concordance with manual assessment.
- Identified 1,496 PIMs.
- System Usability Scale = 83.75 (‘Excellent’).
As a result, even rotating interns could conduct medication reviews at an expert level. However, technology alone is insufficient; it is equally crucial to ensure effective communication both across various disciplines and with patients.
Collaborative communication in action
A case study illustrates how collaborative knowledge exchange transforms evidence into actionable care.
The sound of silence: Collaborative knowledge communication in action
An 80-year-old woman, Mrs. K, presented to the emergency department after a fall at home. She did not lose consciousness but began vomiting and complained of shoulder pain and cough after the fall. She had a long history of heart disease requiring a pacemaker (implanted 15 years ago) and hypertension.
On day 2 of her hospital stay, she was started on intravenous furosemide (a diuretic or ‘water pill’) 20 mg twice daily to manage fluid in her lungs, which was detected on examination. On the same day, she began complaining of giddiness and vertigo, which was severe enough that the doctors initially suspected either Meniere's disease or a condition called Benign Paroxysmal Positional Vertigo (BPPV).
The subsequent events demonstrate how different forms of knowledge—clinical expertise, lived experience and family observations—must come together to create safe, personalised care.
The clinical puzzle emerges
By day 7, owing to persistent giddiness and vertigo, the medical team requested an audiology review. The audiologist diagnosed bilateral moderate sensorineural hearing loss (SNHL), which is hearing loss caused by damage to the inner ear or hearing nerve. This was unexpected.
The healthcare team faced a complex dilemma. As a pharmacist reviewing her medications, I recognised that furosemide could cause hearing loss, particularly when administered intravenously. The cardiologist knew that preventing fluid buildup was critical to avoid heart failure progression. However, continuing the medication risks permanent hearing loss.
The audiologist's professional assessment
The audiologist provided crucial expertise and hearing loss was consistent with medication-induced damage to the cochlea (inner ear). She recommended Hearing Assistive Technology Systems (hearing aids). Her input was essential; without audiology expertise, we might have dismissed the hearing loss as simply age-related, missing the medication connection entirely.
The family's observations: What charts don't capture
During the rounds, Mrs. K's daughter shared observations that transformed our understanding of the urgency of the situation. ‘She keeps asking us to repeat ourselves. She reports hearing loss and is concerned about it.’ This was not just a medical issue; it was affecting Mrs. K's quality of life. The family's lived experience brought emotional and social dimensions that clinical assessments alone would miss.
The collaborative solution
Rather than any single specialist making a unilateral decision, the team worked together to develop a balanced approach.
- Medication adjustment: Instead of abruptly stopping furosemide, the dose was gradually reduced. By day 6, the frequency was decreased to once daily. At discharge, she was transitioned to a different diuretic combination (torsemide plus spironolactone) with a different safety profile.
- Monitoring plan: The team established follow-up protocols to monitor both cardiac status and hearing function.
- Assistive technology: Hearing aids were recommended to address existing hearing loss.
The outcome
At the 30-day follow-up, the patient reported feeling both mentally and physically healthy and her hearing had improved. Hearing loss stabilized and partially reversed, demonstrating that the collaborative approach successfully balanced cardiac needs with hearing preservation.
No single guideline can prescribe a solution. The Beers Criteria flag furosemide as potentially inappropriate in older adults; however, it does not account for individual patient circumstances in which the benefits may outweigh the risks. The solution emerged from a dialogue between specialists, each contributing their expertise to co-create a plan that would work for this specific patient.
This type of collaborative decision-making is especially critical in resource-limited settings, where alternative medications may not be affordable or available, patients often have multiple complex conditions, treatment decisions require weighing trade-offs rather than following rigid protocols and communication between specialists, patients and families must bridge different perspectives and knowledge systems.
Challenges and lessons learned
Initial failures
- Long PIM lists triggered defensive responses from physicians
- Overly technical language excluded patients
- Western guidelines did not fit local realities
- Manual criteria checks were impractical in busy settings
Emerging solutions: De-Med as a conversation facilitator
Initially, when we presented physicians with long lists of ‘inappropriate medications,’ we encountered defensive responses. Doctors felt criticised rather than supported. We realised that technology alone was not the answer; we needed to fundamentally change how we communicated across different stakeholder groups.
How De-Med works: Making complexity visible and discussable
De-Med displays all three criteria (Beers, STOPP and FORTA) side-by-side for each medication, instantly revealing where they agree and where they diverge. For example, when a physician prescribes pantoprazole (a common stomach acid medication) for an older adult patient:
- Beer's Criteria flags it: ‘Avoid scheduled use for >8 weeks unless for high-risk patients.’
- STOPP flags it: ‘STOP for uncomplicated peptic ulcer disease at full therapeutic dosage after 1-2 months.’
- FORTA classifies it as: ‘Class B - Drugs with proven or obvious efficacy in the older adult, but limited extent of effect and/or safety concerns.’
Rather than creating confusion, this visible disagreement is the starting point for a meaningful dialogue. The physician can see that while all three criteria raise concerns about long-term use, they have different thresholds and reasoning. This nuance is important for clinical decision-making.
Facilitating conversations between clinicians
Before De-Med, a pharmacist reviewing this case might simply say: ‘This medication is inappropriate according to Beers criteria.’ The physician might respond defensively: ‘But the patient needs it for GERD.’
With De-Med, the conversation becomes more sophisticated.
Pharmacist: ‘De-Med shows all three criteria flagged pantoprazole for long-term use; the patient has been on it for 6 months. Beers suggests limiting to 8 weeks, STOPP to 1-2 months for uncomplicated cases. FORTA rated it as Class B, acknowledging its efficacy but noting safety concerns. Can we review whether continued use is still warranted?’
Physician: ‘I see. The patient originally had severe reflux, but the symptoms are now controlled. You are correct; we should consider tapering or discontinuing it. If symptoms recur, we could use it intermittently rather than continuously.’
The tool shifts the conversation from confrontation to collaboration. Instead of ‘You're wrong,’ it became ‘Let's look at what the evidence says and decide together.’
Supporting conversations with families
For family members involved in medication management, De-Med's visual comparison helps them to understand complexity:
Family member: ‘My mother is on 12 different medications. Which ones are the most concerning?’
Pharmacist (using De-Med): ‘Let me show you. When we run her medication list through De-Med, we can see which of the three criteria agree on which medications are concerning versus where there is disagreement. For example, diphenhydramine (a common sleep aid) is flagged as 'avoid' by all three, making it a clear priority to address. But her pantoprazole shows more nuance; it is about duration of use rather than the medication itself being always inappropriate.’
This visual evidence helps families prioritise discussions with physicians and understand that medication optimisation is not about stopping everything but about thoughtful, evidence-informed adjustments.
Enabling policy and system-level discussions
For hospital administrators and policymakers, De-Med’s ability to compare criteria across patient populations provides actionable data.
Quality improvement team: ‘When we run our older adult patient population data through De-Med, we see that 89% have at least one medication flagged by all three criteria, these are clear targets for intervention. This tells us we need clinical pharmacist resources to help physicians navigate these grey areas, not just automated alerts that say 'stop this medication.’
This data-driven approach helps secure resources for collaborative medication review programs rather than purely technological solutions.
Facilitating interdisciplinary team discussions
During interdisciplinary rounds, De-Med creates a shared visual reference.
Geriatrician: ‘Mrs. S was administered metoclopramide for nausea. What do our criteria show?’
Pharmacist (checking De-Med): ‘All three criteria flag it strongly; Beers says avoid due to extrapyramidal effects, STOPP says avoid in Parkinson's disease or those at risk, FORTA rates it Class C or D depending on the indication. The agreement is clear here.’
Nurse: ‘She does have some tremors that I have noticed. Could this be contributing?’
Geriatrician: ‘Possibly. Let's switch to an alternative and monitor.’
The tool enables quick, evidence-based discussions during time-constrained rounds, ensuring that all team members can contribute, regardless of their familiarity with each specific criterion.
The critical shift: From judgment to dialogue
What made De-Med effective was not the technology itself, but how it facilitated different types of communication:
- Between clinicians: From ‘You prescribed incorrectly’ to ‘Let's review the evidence together.’
- With patients and families: From technical jargon to visual, understandable comparisons.
- For policy makers: From anecdotes to data-driven resource allocation.
- Across disciplines: From siloed decisions to shared understanding.
By making the complexity visible, showing where evidence aligns and where it diverges, De-Med created space for nuanced, collaborative discussions that medication safety requires.
This is what we mean by technology as a conversation facilitator; not replacing human judgement and dialogue, but providing a shared foundation that enables more productive, evidence-informed and genuinely collaborative decisions across all stakeholder groups.
Lessons learned
- Identifying PIMs is crucial for preventing drug-related harm in older adults. In our hospital, we witnessed preventable falls, confusion and hospitalisations that could have been avoided with a better medication review.
- Current tools are too complex for routine use in resource-constrained settings. When physicians see 40 to 50 patients in a morning shift, as is common in Indian public hospitals, manually checking three different criteria for each patient is impossible. Technology must reduce the workload, not add to it.
- Easy-to-use and accessible tools are essential for real-world impact. However, tools such as De-Med work best when they facilitate collaborative communication between clinicians, patients and families, rather than replacing clinical judgement.
- Country-specific criteria must consider the availability, affordability and regulation of medicines. Several drugs removed from Western guidelines as ‘rarely used’ remain frontline treatments in India because newer alternatives cost 5 to 10 times more or are not available. Guidelines must reflect local realities to be truly evidence based.
Next steps
Future work should focus on:
- Developing India-specific PIM criteria through a collaborative process: Engaging clinicians, pharmacists, patients, researchers and policy makers in co-creating guidelines adapted to local practice realities.
- Training healthcare providers in collaborative communication: Building skills for dialogue across professional boundaries, translation of technical concepts for different audiences and genuine partnership with patients and families.
- Expanding digital tools like De-Med across healthcare systems: Ensuring these tools facilitate rather than replace human communication and clinical judgement.
- Embedding PIM screening into national medication safety policies: Working with policy makers to create systems-level change that supports collaborative practice.
Key take-home messages
1. Technology enables evidence translation: Tools such as De-Med can bridge the gap between global guidelines and real-world practice, but only when they facilitate collaborative communication rather than imposing rigid judgements.
2. Local context and global evidence: Evidence-based care must balance scientific rigour with affordability, availability, cultural context and patient values. This requires integrating different forms of knowledge; clinical expertise, lived experience, family perspectives, economic realities and policy constraints.
3. Collaboration transforms care through communication: Effective medication safety is not about having the right answer; it is about enabling dialogue across different stakeholder groups, tailoring communication to different audiences and genuinely co-creating solutions that work in local contexts. Collaborative knowledge communication means valuing and integrating professional expertise, patient experience, family observations and community wisdom equally in decision-making.
References
Manias E, Kabir MZ, Maier AB. Inappropriate medications and physical function: a systematic review. Ther Adv Drug Saf. 2021;12. doi:10.1177/20420986211030371
American Geriatrics Society. American Geriatrics Society 2019 Updated AGS Beers Criteria® for Potentially Inappropriate Medication Use in Older Adults. J Am Geriatr Soc. 2019;67(4):674-94. https://doi.org/10.1111/jgs.15767
O'Mahony D, O'Sullivan D, Byrne S, O'Connor MN, Ryan C, Gallagher P. STOPP/START criteria for potentially inappropriate prescribing in older people: version 2. Age Ageing. 2015;44(2):213-8. https://doi.org/10.1093/ageing/afu145
Syed J, Pereira P, Tejeswini CJ, et al. Multi-Criteria Assessment of Potentially Inappropriate Medications in Hospitalized Indian Older Adults: Comparing Beers, STOPP and FORTA. J Appl Gerontol. 2025;0(0):1-12. https://doi.org/10.1177/07334648251348704
Syed J, Verma A, Chopra V, Khemchandani V, Pereira P, Patil V, et al. De-Med—a tool to identify and classify potentially inappropriate medication according to Beers criteria, STOPP and FORTA: a cross-sectional study, J Pharmaceut Health Serv Res. 2024;15(4). https://doi.org/10.1093/jphsr/rmae026
Syed J, Chalasani SH. The 2023 American Geriatrics Society Updated Beers Criteria® Application in low- and middle-income countries: A walk-through. Sr Care Pharm. 2024;39(8):286-90. doi: 10.4140/TCP.n.2024.286. PMID: 39080870.
To link to this article - DOI: https://doi.org/10.70253/EVHR6181
Links to additional resources
Disclaimer
The views expressed in this World EBHC Day Blog, as well as any errors or omissions, are the sole responsibility of the author and do not represent the views of the World EBHC Day Steering Committee, Official Partners or Sponsors; nor does it imply endorsement by the aforementioned parties.
