PHARMACOLOGY • MEDICATION SAFETY, CALCULATIONS & DECISION-MAKING

Contraindications & Risk-Benefit — Contraindications and risk-benefit reasoning in clinical scenarios

Understanding when not to prescribe is as critical as knowing what to prescribe.

Historical Context & Motivation

The history of pharmacology is punctuated by tragedies that arose when medications were administered without adequate assessment of patient-specific risks. Long before formal drug regulation existed, practitioners relied on empirical observation and anecdotal evidence to determine whether a remedy might cause more harm than good. The concept of contraindications — conditions or factors that make a particular treatment inadvisable — evolved gradually from these painful lessons. Each catastrophic drug event catalyzed new regulatory frameworks and clinical reasoning tools designed to protect patients from foreseeable harm.

1937
Sulfanilamide Disaster
Elixir Sulfanilamide, dissolved in toxic diethylene glycol, killed over 100 people in the United States. This tragedy exposed the absence of pre-market safety testing and led directly to the Federal Food, Drug, and Cosmetic Act of 1938, which mandated proof of safety before a drug could be marketed.
1961
Thalidomide Crisis
Thalidomide, prescribed as a sedative and anti-nausea agent for pregnant women, caused severe birth defects (phocomelia) in thousands of infants worldwide. This disaster established pregnancy as a critical contraindication category and spurred the 1962 Kefauver-Harris Amendment requiring proof of both safety and efficacy.
1998
FDA Black Box Warning System Expanded
The FDA strengthened its boxed warning requirements on drug labels, creating the most prominent alert mechanism for serious contraindications and life-threatening adverse effects. Black box warnings remain the strongest safety warning a prescription drug can carry short of withdrawal from the market.
2007
FDA Amendments Act (FDAAA)
Congress expanded the FDA's authority to require post-market safety studies and Risk Evaluation and Mitigation Strategies (REMS). This legislation formalized ongoing risk-benefit surveillance throughout a drug's lifecycle, not just at the point of initial approval.

These historical milestones underscore a central question that every clinician must answer before initiating therapy: does the anticipated therapeutic benefit of this medication outweigh the potential risks for this particular patient? This question is not merely academic — it is the foundation of evidence-based prescribing and medication safety. The remainder of this lesson builds the framework you need to answer it systematically.

Core Principles & Definitions

Before analyzing clinical scenarios, you must master the foundational vocabulary that underpins risk-benefit reasoning. Contraindications are not monolithic prohibitions; they exist on a spectrum of severity that directly affects clinical decision-making. Understanding the distinctions between categories — and how they interact with patient-specific variables — is the first step toward safe, rational pharmacotherapy.

1

Absolute Contraindication

A condition or factor under which a drug must never be administered because the risk of a life-threatening or irreversible adverse effect is unacceptably high regardless of potential benefit. Example: administering methotrexate to a patient who is pregnant.
2

Relative Contraindication

A condition where the drug poses elevated risk, but the benefit may still outweigh harm in specific clinical circumstances. Use requires careful justification, close monitoring, and informed consent. Example: using NSAIDs in a patient with mild renal impairment when no alternative analgesic is effective.
3

Risk-Benefit Analysis

A systematic comparison of the probability and magnitude of therapeutic benefit against the probability and severity of adverse outcomes. This analysis incorporates patient-specific factors such as age, comorbidities, genetic polymorphisms, concurrent medications, and patient preferences.
4

Number Needed to Treat (NNT) vs. Number Needed to Harm (NNH)

NNT quantifies the number of patients who must be treated for one additional patient to benefit, while NNH quantifies how many patients must be exposed before one additional patient experiences harm. Comparing NNT to NNH provides a quantitative foundation for risk-benefit decisions.
5

Black Box Warning

The FDA's most serious drug label warning, enclosed in a black border, indicating that the medication carries risks of serious or life-threatening adverse effects. These warnings often highlight specific contraindications or mandatory monitoring requirements that clinicians must heed.
KEY TAKEAWAY
Think of absolute and relative contraindications like traffic signals. An absolute contraindication is a solid red light — you stop completely, no exceptions. A relative contraindication is a flashing yellow — you may proceed, but only after slowing down, looking carefully in every direction, and deciding that the road ahead is safer than stopping. The clinical equivalent of 'looking carefully' is a thorough risk-benefit analysis documented in the patient's record.

Visual Explanation — The Risk-Benefit Decision Framework

This flowchart illustrates the stepwise decision process a clinician follows when evaluating a medication for a specific patient. The red node represents a firm stop for absolute contraindications, the amber node triggers a risk-benefit analysis for relative contraindications, and the green nodes represent safe progression with or without enhanced monitoring.

The flowchart above encapsulates the clinical reasoning pathway that pharmacists, physicians, and nurse practitioners employ each time they evaluate a medication order. Notice that the process is hierarchical: absolute contraindications are screened first because they represent non-negotiable barriers to therapy. Only when no absolute contraindication exists does the clinician proceed to evaluate relative contraindications — and even then, a structured risk-benefit analysis must be completed before the drug is administered. The final documentation step is not optional; it provides medicolegal protection and ensures continuity of care if the patient's clinical status changes.

Quantitative Risk-Benefit Framework

While much of clinical risk-benefit reasoning is qualitative — weighing severity, patient preferences, and disease trajectory — pharmacology also provides quantitative tools that anchor decisions in evidence. Two key metrics derived from clinical trial data allow clinicians to compare the therapeutic yield of a drug against its potential for harm.

NUMBER NEEDED TO TREAT
NNT = 1 / ARR = 1 / (CER − EER)
NNT = Number Needed to Treat; ARR = Absolute Risk Reduction; CER = Control Event Rate (rate of adverse outcome in the control/placebo group); EER = Experimental Event Rate (rate of adverse outcome in the treatment group). A lower NNT indicates a more effective treatment — fewer patients need to be treated for one to benefit.
NUMBER NEEDED TO HARM
NNH = 1 / ARI = 1 / (EER_harm − CER_harm)
NNH = Number Needed to Harm; ARI = Absolute Risk Increase (for adverse effects); EER_harm = rate of adverse effect in the treatment group; CER_harm = rate of adverse effect in the control group. A higher NNH is desirable — more patients can be treated before one is harmed.
LIKELIHOOD OF BEING HELPED OR HARMED (LHH)
LHH = NNH / NNT
The Likelihood of being Helped or Harmed integrates both metrics into a single ratio. An LHH > 1 indicates that a patient is more likely to benefit than to be harmed; an LHH < 1 signals that harm outweighs benefit for the population studied. This ratio is a powerful quantitative anchor for risk-benefit conversations with patients and interdisciplinary teams.
💡 Clinical Pearl
NNT and NNH are population-level estimates derived from clinical trials, but individual patients may differ significantly due to genetic variation, comorbidities, and concurrent medications. Always combine quantitative data with patient-specific clinical assessment. A drug with an NNT of 10 and an NNH of 50 (LHH = 5) is generally favorable, but if the harm in question is irreversible organ damage and the benefit is modest symptom relief, the qualitative weight of the harm may override the favorable ratio.

Classifying Contraindications in Clinical Practice

Contraindications can be classified by their underlying mechanism, allowing clinicians to anticipate risk categories even for unfamiliar medications. The major categories include pharmacodynamic contraindications (where the drug's mechanism of action would exacerbate a pre-existing condition), pharmacokinetic contraindications (where altered absorption, distribution, metabolism, or excretion leads to dangerous drug levels), allergy-based contraindications, and teratogenicity-based contraindications. The following diagram and table illustrate these categories with clinically relevant examples.

This diagram classifies the four major categories of contraindications — pharmacodynamic, pharmacokinetic, allergy/immune, and teratogenic — with clinical examples for each. The risk severity spectrum at the bottom highlights that contraindications exist along a continuum, ranging from manageable risks requiring monitoring to absolute prohibitions against drug use.
Selected contraindication examples classified by mechanism and severity
CategoryMechanismClinical ExampleType
PharmacodynamicDrug's action directly worsens patient's conditionNon-selective β-blocker (propranolol) in asthma → bronchospasmAbsolute
PharmacokineticImpaired drug clearance leads to toxic accumulationMetformin in eGFR < 30 mL/min → lactic acidosisAbsolute
Allergy / ImmuneImmune-mediated hypersensitivity reactionPenicillin in patient with documented anaphylaxisAbsolute
TeratogenicDrug disrupts fetal developmentIsotretinoin in pregnancy → craniofacial malformationAbsolute
PharmacodynamicMild exacerbation of existing condition, manageable with monitoringNSAID in mild CKD (eGFR 45–59) with adequate hydrationRelative

Worked Example — Clinical Scenario

Consider the following clinical scenario: a 68-year-old male with a history of atrial fibrillation (AF), stage 3a chronic kidney disease (eGFR 52 mL/min/1.73 m²), a prior episode of gastrointestinal bleeding two years ago, and a CHA₂DS₂-VASc score of 4. The physician is considering initiating apixaban (Eliquis) for stroke prevention. Is this drug appropriate, and what does the risk-benefit analysis reveal?

Risk-Benefit Analysis: Apixaban for Stroke Prevention in AF
1
Step 1 — Check for Absolute ContraindicationsReview the FDA prescribing information for apixaban. Absolute contraindications include: active pathological bleeding, severe hepatic disease with coagulopathy, and documented hypersensitivity to apixaban. The patient has none of these — his GI bleed was two years ago and has resolved, and his liver function is normal.
No absolute contraindications identified. Proceed to relative contraindication screening.
2
Step 2 — Identify Relative ContraindicationsTwo relative contraindications are present: (1) prior GI bleeding, which increases the risk of recurrent hemorrhage on anticoagulation, and (2) stage 3a CKD, which may affect drug clearance. However, apixaban is 75% hepatically cleared and only 27% renally eliminated, making it one of the safer DOACs in moderate renal impairment. The prior GI bleed is the more significant concern.
Two relative contraindications identified: prior GI bleed and CKD stage 3a.
3
Step 3 — Quantify Stroke Risk (Benefit of Treatment)A CHA₂DS₂-VASc score of 4 corresponds to an annual stroke risk of approximately 4.0% without anticoagulation. The ARISTOTLE trial demonstrated that apixaban reduces stroke/systemic embolism by 21% relative to warfarin, with an absolute risk reduction of approximately 0.33% per year vs. warfarin and a significantly greater absolute benefit vs. no anticoagulation.
Untreated annual stroke risk ≈ 4.0%. Apixaban provides substantial stroke risk reduction.
4
Step 4 — Quantify Bleeding Risk (Harm of Treatment)The HAS-BLED score helps estimate bleeding risk. This patient's score is approximately 3 (hypertension, abnormal renal function, prior bleed), placing him in the high-risk category for major bleeding (≈ 3.7% per year). Notably, the ARISTOTLE trial showed apixaban reduced major bleeding by 31% compared to warfarin, including a 47% reduction in intracranial hemorrhage. GI bleeding rates were similar between apixaban and warfarin.
Estimated annual major bleeding risk ≈ 3.7%. Apixaban has a favorable GI bleeding profile compared to other DOACs.
5
Step 5 — Calculate LHH and Render Clinical JudgmentUsing trial data: NNT for stroke prevention (vs. no treatment) ≈ 25 per year. NNH for major bleeding ≈ 27 per year. LHH = NNH / NNT = 27 / 25 = 1.08. While the LHH ratio is only slightly above 1, the qualitative weight tilts the scale further toward benefit: ischemic stroke carries a high mortality/disability rate (≈ 25% fatal), whereas GI bleeds, while serious, are more often manageable and reversible. Adding a proton pump inhibitor (PPI) for GI protection and scheduling renal function monitoring every 3–6 months further reduces harm. The clinician documents the risk-benefit reasoning and discusses options with the patient.
Decision: Initiate apixaban 5 mg BID with concurrent PPI, regular renal monitoring, and documented informed consent. LHH ≈ 1.08 — benefit marginally outweighs risk, with qualitative factors favoring treatment.

Strengths & Limitations of Risk-Benefit Frameworks

Comparative analysis of risk-benefit assessment tools
AspectStrengthsLimitations
NNT / NNH MetricsProvide objective, evidence-based quantification of benefit and harm; easily communicated to patients and interdisciplinary teamsDerived from population averages; may not reflect individual genetic, metabolic, or comorbidity-related variation
Clinical JudgmentIncorporates patient-specific nuance, patient preferences, and qualitative severity weighting that statistical metrics cannot captureSusceptible to cognitive biases (anchoring, availability heuristic, confirmation bias) and inter-clinician variability
FDA Label / Black Box WarningsStandardized, legally enforceable, and based on rigorous regulatory review of pre- and post-market dataLabels may lag behind emerging evidence; 'off-label' use may be evidence-supported but not label-reflected
Scoring Systems (CHA₂DS₂-VASc, HAS-BLED)Structured and reproducible; reduce reliance on memory; facilitate guideline-concordant careCannot capture every relevant patient variable; may oversimplify complex clinical scenarios
KEY TAKEAWAY
No single tool — neither a scoring system nor a quantitative metric — is sufficient in isolation for risk-benefit reasoning. Think of these tools as instruments in an orchestra: the NNT/NNH ratio provides the rhythm, the clinical scoring systems supply the melody, the FDA label sets the key signature, and clinical judgment is the conductor who integrates all the parts into a coherent performance. The best clinical decisions arise from triangulating multiple data sources with the patient's own values and goals of care.

Connection to Advanced Pharmacovigilance & Precision Medicine

The foundational framework of contraindication screening and risk-benefit analysis presented in this lesson represents the standard of care across all clinical settings. However, emerging technologies and paradigms are transforming how clinicians identify and weigh patient-specific risks. Pharmacogenomics allows clinicians to predict drug metabolism phenotypes before a medication is administered — for example, testing for HLA-B*5701 before prescribing abacavir to prevent a potentially fatal hypersensitivity reaction. Clinical decision support systems (CDSS) integrated into electronic health records automate real-time contraindication checks by cross-referencing patient allergies, diagnoses, renal function, and concurrent medications against drug databases. These systems reduce human error but require intelligent override protocols to avoid alert fatigue, in which clinicians become desensitized to frequent warnings and begin dismissing clinically significant alerts.

Evolution from population-level to individualized risk-benefit analysis
FeatureTraditional Risk-BenefitPrecision Medicine Approach
Data SourcePopulation-level clinical trials, FDA label, clinical scoring toolsIndividual genomic profile, pharmacokinetic modeling, AI-driven risk prediction
Contraindication IdentificationManual chart review, pharmacist verification, allergy checkingAutomated CDSS with pharmacogenomic integration and drug-drug interaction engines
Risk QuantificationNNT/NNH from trial populations; clinical scoring systemsPatient-specific predicted response curves; Bayesian risk modeling
Key ChallengeCognitive bias and incomplete patient informationAlert fatigue, data integration complexity, cost and access equity

As you advance in your clinical education, you will encounter increasingly sophisticated tools for risk stratification. However, the cognitive framework introduced in this lesson — checking absolute contraindications first, evaluating relative contraindications through systematic risk-benefit analysis, and documenting your reasoning — remains the bedrock upon which all advanced approaches are built. Master this framework, and the precision medicine tools of the future will amplify, not replace, your clinical judgment.

Practice Problems

PROBLEM 1CONCEPTUAL
Explain the difference between an absolute contraindication and a relative contraindication. Provide one clinical example of each, identifying the drug, the patient condition, and the reason the contraindication applies.
PROBLEM 2BASIC CALCULATION
In a clinical trial of Drug X for migraine prevention, the placebo group experienced migraines in 40% of patients over six months, while the treatment group experienced migraines in 28% of patients. Calculate the Absolute Risk Reduction (ARR) and the Number Needed to Treat (NNT).
PROBLEM 3INTERMEDIATE
Using the same Drug X trial data from Problem 2, the treatment group experienced hepatotoxicity in 5% of patients, while the placebo group experienced hepatotoxicity in 1% of patients. Calculate the NNH and the Likelihood of being Helped or Harmed (LHH). Interpret the clinical significance of the LHH value.
PROBLEM 4APPLIED
A 55-year-old woman with rheumatoid arthritis and a history of peptic ulcer disease (PUD) presents with a disease flare. Her rheumatologist is considering starting naproxen (an NSAID). She has stage 2 CKD (eGFR 72 mL/min), is on low-dose aspirin for cardiovascular prophylaxis, and takes no PPI. Identify all contraindications (absolute and relative), describe the steps of a risk-benefit analysis, and recommend a course of action.
PROBLEM 5CRITICAL THINKING
A clinical decision support system fires an alert warning that warfarin is contraindicated in a 32-year-old woman because her chart lists 'pregnancy' as an active diagnosis. The pharmacist discovers that the pregnancy was documented 8 months ago and the patient delivered 2 months ago; the diagnosis was never resolved in the EHR. She now has a mechanical heart valve requiring anticoagulation. Discuss the implications of this scenario for (a) alert fatigue and patient safety, (b) the limitations of automated contraindication screening, and (c) the role of clinical judgment in overriding electronic alerts.

Lesson Summary

This lesson established the framework for contraindication screening and risk-benefit reasoning in clinical pharmacology. We traced the historical tragedies — from the 1937 Sulfanilamide disaster to the thalidomide crisis — that catalyzed modern drug safety regulation. We defined absolute contraindications (never administer) and relative contraindications (proceed only after rigorous analysis) and classified them by mechanism — pharmacodynamic, pharmacokinetic, allergy-based, and teratogenic.

The quantitative tools for risk-benefit analysis — NNT, NNH, and LHH — provide an evidence-based foundation, while clinical scoring systems (e.g., CHA₂DS₂-VASc and HAS-BLED) structure the qualitative assessment. Ultimately, safe prescribing requires triangulating population-level data with patient-specific factors, documenting your reasoning, and incorporating the patient's values and preferences — a process that will only become more precise as pharmacogenomics and clinical decision support systems continue to evolve.

Varsity Tutors • Pharmacology • Contraindications & Risk-Benefit