Page 2 of 29
PH10.{1-2,13} | PH10.{1-2,13} | Drug Information Appraisal and Promotional Literature Critique — SDL Guide — SDL Guide (Part 2)
Evaluating Promotional Literature and Drug Representative Interactions
Pharmaceutical representatives play a real role in the information ecosystem — they can provide timely updates on new drugs, supply samples for trial, and communicate safety alerts. However, their primary role is commercial promotion, and the evidence shows that even brief interactions influence prescribing towards promoted products, often without the prescriber being conscious of the effect. Recognising this dynamic is not cynicism — it is a professional safeguard.
Reading a promotional brochure — red flags:
- Claims based on surrogate endpoints (e.g. 'improves HbA1c') without data on clinical outcomes (mortality, hospitalisation)
- Comparison only with placebo, not standard-of-care
- Use of RRR without stating ARR or NNT
- Forest plots or graphs that truncate axes to exaggerate effect sizes
- Reference section citing only company-funded studies or conference abstracts (lower reliability)
- No disclosure of funding source, or conflicts buried in small print
Reading a package insert (prescribing information):
The package insert is a regulatory document and provides authoritative data on: approved indications and dosing (including renal/hepatic dose adjustments), contraindications and precautions, drug interactions, adverse effect frequency (from registration trials), and storage. It does NOT compare the drug favourably against competitors — that is the promotional brochure's job.
Pharmaceutical representative interaction — professional standards:
The NMC code and WHO guidelines recommend that:
1. Meetings should be in a professional, not social, setting
2. The prescriber should come with specific pre-formulated questions, not open-ended 'tell me about your drug'
3. Any claims should be verified against independent sources before influencing prescribing
4. Gifts, meals, and entertainment from industry create a documented reciprocity bias — Indian Medical Council Regulations 2002 (amended) explicitly prohibit accepting gifts above a nominal educational value
5. All information provided should be cross-checked against the package insert, standard references, and independent guidelines
| Characteristic | Promotional Brochure | Package Insert | Independent Guideline |
|---|---|---|---|
| Author | Manufacturer | Manufacturer (regulatory-approved) | Expert panel, independent society |
| Bias risk | High | Moderate (regulatory-constrained) | Lower (if process transparent) |
| Comparator | Often placebo | Approved indications only | Active comparators |
| Applicability to India | May not be validated | Based on global registration trials | May have local guideline variants |
SELF-CHECK
A promotional brochure for a new antihypertensive states: 'Drug X reduces cardiovascular events by 40% compared to placebo.' The absolute event rate in the placebo arm was 5%, and in the Drug X arm 3%. What is the Number Needed to Treat (NNT)?
A. A) 40
B. B) 50
C. C) 25
D. D) 20
Reveal Answer
Answer: B. B) 50
B is correct. ARR = 5% − 3% = 2%. NNT = 1/ARR = 1/0.02 = 50. The promotional claim of '40% reduction' is the relative risk reduction (2/5 = 40%), which sounds impressive but translates to treating 50 patients to prevent one cardiovascular event. Always convert promotional RRR claims to NNT/ARR before assessing clinical significance.
Appraising Drug Claims: Worked Case Studies
Applying the appraisal framework to real-world scenarios builds the habit that makes critical evaluation automatic in clinical practice. Consider two worked examples that illustrate the most common promotional distortions encountered in practice.
Case Study 1 — 'Superior to Standard Therapy':
A brochure for a new fluoroquinolone states it 'achieved bacteriological eradication in 94% of patients vs 87% for standard therapy — a statistically significant improvement (p=0.03).' Applying the appraisal framework:
- PICO: Was the comparator the actual local standard-of-care, or an older/suboptimal regimen? Was the population representative of the community-acquired infections you treat?
- Bias: Was the study company-funded? Was blinding adequate?
- Statistics: The ARR = 7 percentage points, NNT = ~14. Is this clinically meaningful given the higher cost and broader-spectrum resistance implications of the new fluoroquinolone?
- Applicability: Indian local resistance patterns for the organisms studied — are they comparable to the trial population?
- Verdict: Before prescribing, verify comparator choice and check current local antibiogram. 'Statistically significant' does not mean clinically necessary.
Case Study 2 — Package Insert vs Promotional Claim:
A representative claims their lipid-lowering drug 'reduces LDL by 50% and is safe in renal impairment at standard doses.' The package insert under 'Dosage in Renal Impairment' states 'dose reduction required for eGFR <30; not recommended in end-stage renal disease.' Cross-referencing with the package insert immediately exposes an overstatement in the verbal promotional claim.
General appraisal checklist for any drug claim:
1. Who funded the study? (Disclose conflicts)
2. What was the comparator? (Active vs placebo?)
3. What were the outcomes? (Clinical vs surrogate?)
4. Convert RRR → ARR → NNT
5. What is the NNH for the most important adverse effect?
6. Is the trial population applicable to your patients?
7. Have independent bodies (national guideline group, WHO EML review) accepted this evidence?
CLINICAL PEARL
Relative risk reduction always sounds better than it is. When evaluating any drug claim — from a brochure or a journal — always ask for the absolute risk reduction and calculate the NNT. A drug that reduces mortality from 0.4% to 0.2% has a 50% RRR but an NNT of 500. The WHO Essential Medicines List selection criteria require that evidence of efficacy be based on clinical outcomes, not surrogate markers — apply the same standard to every new drug you encounter.
Self-Assessment: Drug Information Competency Check
Before completing this module, consolidate your understanding by working through the following structured self-check questions. These test your ability to classify sources, apply the appraisal framework, and manage professional boundaries with pharmaceutical representatives.
Self-check 1: A trial reports that Drug A reduced stroke risk from 8% to 6% over 5 years (p=0.02, 95% CI 0.01–0.04). Calculate the ARR, NNT, and state whether the result has clinical significance beyond statistical significance.
Self-check 2: A pharmaceutical representative offers to provide 'educational materials' and take you to a dinner seminar to discuss their new insulin analogue. List three steps you should take before and during this interaction consistent with NMC guidelines.
Self-check 3: You are searching for information on drug interactions of a new antifungal. Rank the following in order of reliability: (a) the drug's promotional brochure, (b) a 2021 systematic review from Cochrane, (c) the package insert, (d) a pharmacology textbook chapter.
Expected answers:
- Self-check 1: ARR = 2%, NNT = 50. Statistical significance is confirmed (p=0.02, CI excludes 0); clinical significance depends on the cost, adverse-effect profile, and patient preference — treating 50 patients to prevent one stroke over 5 years may or may not be cost-effective depending on the drug and context.
- Self-check 2: Clarify the setting (professional, not social); prepare specific clinical questions beforehand; attend the seminar but independently verify all claims against the package insert and standard references; decline gifts exceeding nominal educational value; do not alter prescribing solely based on the dinner interaction.
- Self-check 3: (b) Cochrane systematic review > (c) package insert > (d) textbook > (a) promotional brochure. For drug interactions specifically, the Cochrane systematic review or a curated interaction database (BNF/Micromedex) is most reliable; the package insert lists interactions identified during regulatory trials but may miss post-marketing signals.