Active ADR Monitoring in General Medicine Wards: Causality, Severity and Preventability Assessment with PvPI Reporting

This project is a six-month prospective observational study of adverse drug reactions (ADRs) in the General Medicine wards of a 750-bed tertiary-care teaching hospital. Pharm.D students attend daily ward rounds as part of their clerkship, which puts them in the best position to pick up ADRs that busy clinicians often notice but never document or report.

The team screens every in-patient admitted to the study units using an active surveillance approach: case-sheet review, nursing notes, laboratory trends and direct patient interviews. Each suspected ADR is documented on the CDSCO Suspected Adverse Drug Reaction Reporting Form, forwarded through the hospital's ADR Monitoring Centre (AMC) under the Pharmacovigilance Programme of India (PvPI), and then assessed with four validated tools: the Naranjo algorithm and WHO-UMC system for causality, the modified Hartwig & Siegel scale for severity and the modified Schumock–Thornton criteria for preventability. Suspected drugs are coded with the WHO ATC classification and reactions are grouped by organ system.

The outcome is a clear picture of incidence, pattern and preventability of ADRs in medicine wards, the drug classes most often responsible, and the patient factors (age, polypharmacy, renal function, comorbidity) associated with ADRs, analysed with chi-square and logistic regression in SPSS. The study fits the PCI Pharm.D Year V Project Work slot (six months, 20 hours per week) and produces a 40–50-page report defended in a 30-mark oral examination.

Abstract

A prospective observational study will be conducted over six months in the General Medicine wards of a tertiary-care teaching hospital to determine the incidence, pattern, causality, severity and preventability of ADRs. Suspected reactions will be identified by active surveillance, reported to PvPI through the hospital AMC, and assessed with the Naranjo scale, WHO-UMC criteria, modified Hartwig & Siegel scale and modified Schumock–Thornton criteria. Risk factors will be analysed by chi-square test and binary logistic regression.

Introduction

An adverse drug reaction is a response to a drug that is noxious and unintended and occurs at doses normally used in humans for prophylaxis, diagnosis or therapy. ADRs prolong hospital stay, increase cost and erode patient trust. India's PvPI, coordinated by the Indian Pharmacopoeia Commission as the National Coordination Centre, depends on AMCs in teaching hospitals, yet spontaneous reporting in India remains low and under-reporting is the rule rather than the exception.

Review and gap

Published Indian studies report ADR incidence in medicine wards ranging widely because of differing definitions and surveillance intensity. Many rely on spontaneous reports only, assess causality with a single scale, or omit preventability. Few student studies report agreement between Naranjo and WHO-UMC or link ADRs to modifiable risk factors such as polypharmacy.

Proposed work

• Active, daily surveillance by trained Pharm.D students under a clinical pharmacist preceptor.
• Dual causality assessment with Cohen's kappa for agreement.
• Severity and preventability grading for every confirmed ADR.
• Risk-factor analysis to identify patients who need closer monitoring.

Feasibility

• Setting: the hospital already hosts a PvPI AMC and a Department of Pharmacy Practice.
• Time: six months of data collection at 20 hours per week fits the PCI project allocation.
• Cost: scales and reporting forms are free; SPSS is available in the college computer lab.
• Ethics: Institutional Ethics Committee approval and written informed consent are obtained before enrolment.

Adverse drug reactions are a leading cause of preventable harm in hospitalised patients, particularly older adults with multiple comorbidities receiving many drugs. In Indian teaching hospitals, suspected ADRs are frequently managed at the bedside — the drug is stopped or an antidote is given — but the event is not documented as an ADR, not assessed for causality and not reported to PvPI. As a result, the hospital has no data on which drugs and patients are at highest risk, and national signal detection is weakened.

There is a need for a systematic, prospective ADR monitoring study in medicine wards that measures incidence, characterises reactions by organ system and drug class using ATC codes, assesses causality, severity and preventability with validated scales, identifies risk factors, and demonstrates how a clinical pharmacist-led surveillance model can increase reporting to PvPI.

In scope

• All in-patients aged ≥ 18 years admitted to General Medicine units during the study period who give consent.
• ADRs occurring during admission and ADRs that caused the admission.
• Causality, severity and preventability assessment; ATC coding; outcome documentation; PvPI reporting.
• Risk-factor analysis with descriptive and inferential statistics.

Out of scope

• Medication errors without harm, therapeutic failure and intentional overdose (recorded separately but not analysed as ADRs).
• Paediatric, obstetric and ICU patients.
• Rechallenge of any drug for the purpose of the study — rechallenge information is used only when it occurred clinically.
• Cost-of-ADR analysis (listed as future scope).

Study design: prospective, observational, active-surveillance study.

Setting and duration: General Medicine wards (four units) of a 750-bed tertiary-care teaching hospital with a PvPI ADR Monitoring Centre; six months of data collection within the Year V project period.

Sample size: all eligible admissions during the period (census sampling). Assuming an ADR incidence of 10% from Indian ward studies, absolute precision of 3% and 95% confidence, n = Z²pq/d² = (1.96² × 0.10 × 0.90)/0.03² ≈ 384 patients; the wards admit about 150 patients a month, so six months comfortably exceeds this.

Inclusion criteria: in-patients aged ≥ 18 years of either gender, admitted for ≥ 24 hours, receiving at least one drug, and giving written informed consent (or consent from a legally acceptable representative).

Exclusion criteria: patients admitted with intentional poisoning or overdose, pregnant women, and patients transferred to ICU within 24 hours of admission.

Instruments: specially designed patient-profile form, CDSCO Suspected ADR Reporting Form, Naranjo scale (definite ≥ 9, probable 5–8, possible 1–4, doubtful ≤ 0), WHO-UMC causality categories, modified Hartwig & Siegel severity scale (levels 1–7: mild, moderate, severe), modified Schumock–Thornton criteria (definitely, probably, not preventable), WHO ATC index, Micromedex/Lexicomp for known reaction profiles.

Ethics: approval from the Institutional Ethics Committee registered with CDSCO, conducted as per the ICMR National Ethical Guidelines (2017); written informed consent in English, Kannada or the patient's language; confidentiality protected by study codes — no names, hospital numbers or identity documents are entered in the analysis sheet.

Statistical analysis: SPSS. Descriptive statistics (frequency, percentage, mean ± SD); incidence with 95% CI; chi-square/Fisher's exact test for associations; binary logistic regression for independent risk factors (odds ratio, 95% CI); Cohen's kappa for Naranjo vs WHO-UMC agreement; p < 0.05 significant.