Evidence-Based Prescribing: From Guidelines to Clinical Practice

Reviewed by MedNext Clinical Team

Evidence-based medicine (EBM) was defined by Sackett and colleagues as "the conscientious, explicit, and judicious use of current best evidence in making decisions about the care of individual patients" ^[1]. This definition is important: it is not the uncritical application of guidelines, nor is it the dismissal of clinical expertise in favour of trial data. Evidence-based prescribing represents the integration of the best available external evidence with individual clinical expertise and patient values — a synthesis that is far more challenging in practice than it appears in principle.

The Hierarchy of Evidence

Not all evidence is created equal. The hierarchy of evidence ranks study designs by their susceptibility to bias and their power to establish causal relationships ^[1]:

1. Systematic reviews and meta-analyses — synthesise data from multiple randomised controlled trials; the highest level of evidence for therapeutic decisions
2. Randomised controlled trials (RCTs) — the gold standard for establishing causation; random allocation eliminates selection bias
3. Cohort studies — observe outcomes in defined groups over time; useful for long-term outcomes and rare events not amenable to RCT
4. Case-control studies — compare patients with and without an outcome; efficient for rare conditions but susceptible to recall bias
5. Cross-sectional studies — observe prevalence at a single point in time; cannot establish temporality
6. Case reports and expert opinion — lowest level; hypothesis-generating rather than confirmatory

The GRADE framework, developed by Guyatt and colleagues, provides a structured approach to rating the quality of evidence and the strength of recommendations derived from it ^[2]. GRADE distinguishes between the quality of evidence (high, moderate, low, very low) and the strength of the recommendation (strong or conditional/weak), recognising that a strong recommendation can be made even from low-quality evidence if the benefits clearly outweigh the harms.

Interpreting Clinical Guidelines

Clinical guidelines synthesise the available evidence and translate it into actionable recommendations. Understanding how to read a guideline critically is as important as knowing its content ^[2].

The Limitations of Guidelines

Guidelines are developed to apply to defined populations, and several important limitations must be recognised ^[1,2]:

• Trial populations may not reflect your patient — landmark RCTs frequently exclude the elderly, patients with multiple comorbidities, pregnant women, and children. Applying trial-derived recommendations to excluded populations requires caution.
• Surrogate endpoints — many trials report surrogate outcomes (blood pressure reduction, HbA1c lowering) rather than hard clinical endpoints (cardiovascular events, mortality). A drug may improve a surrogate while failing to improve or even worsening clinical outcomes.
• Publication bias — trials with negative or null results are less likely to be published, inflating the apparent effectiveness of interventions in the published literature.
• Guideline lag — the time from evidence generation to guideline publication and dissemination can be years, meaning guidelines may not reflect the most current evidence.
• Conflict of interest — guideline panels with significant industry ties may overstate the evidence for particular treatments.

Absolute Versus Relative Risk Reduction

One of the most important skills in interpreting trial evidence is the ability to distinguish between absolute risk reduction (ARR) and relative risk reduction (RRR). A drug that reduces the relative risk of an event by 50% sounds impressive — but if the baseline event rate is 2%, this represents an absolute risk reduction of only 1%, meaning 100 patients must be treated to prevent one event (NNT = 100). This context is frequently absent from summaries of clinical trial results ^[1].

Individual Patient Factors

Even the highest-quality evidence applies at the population level. Translating this to an individual patient requires integrating several additional factors ^[2]:

• Baseline risk — a treatment that produces a given relative risk reduction confers greater absolute benefit in higher-risk patients
• Comorbidities and polypharmacy — renal impairment, hepatic disease, and co-administered medications alter drug handling and effect
• Patient values and preferences — the weight a patient places on different outcomes (avoiding a stroke versus tolerating daily medication burden) is subjective and must be incorporated into shared decision-making
• Frailty and prognosis — preventive therapies targeting events that occur over a five- to ten-year horizon may offer little benefit to patients with limited life expectancy or severe frailty

Shared Decision Making

Effective evidence-based prescribing is not a unilateral clinical act — it is a collaborative conversation between clinician and patient ^[1]. The clinician brings expertise in the evidence, the mechanism of the treatment, and its likely benefits and risks for this patient. The patient brings their own values, circumstances, and preferences. When both are integrated, treatment decisions are more likely to be adhered to and more likely to align with what the patient actually wants.

MedNext Formulary supports evidence-based prescribing by providing comprehensive drug monographs from the MedNext Audited Proprietary Dataset — including indications, contraindications, drug interactions, and dose adjustment guidance — enabling clinicians to apply the best available evidence to each individual patient encounter efficiently and accurately.

References

1. Sackett DL, Rosenberg WM, Gray JA, Haynes RB, Richardson WS. Evidence based medicine: what it is and what it isn't. BMJ. 1996;312(7023):71-72.
2. Guyatt GH, Oxman AD, Vist GE, et al. GRADE: an emerging consensus on rating quality of evidence and strength of recommendations. BMJ. 2008;336(7650):924-926. [GRADE guidelines: a new series of articles in the Journal of Clinical Epidemiology. J Clin Epidemiol. 2011;64(4):380-382.]