Applying Research Evidence to the Individual

After identifying relevant studies, the clinician must think about their applicability to the individual patient, because even the best studies report estimates of effect in terms of an average. Study participants and real-world patients are likely to differ by degree, rather than grossly, in their response to treatment.15 Individualizing treatment decisions involves estimation of the balance of risks and benefits, combined with a consideration of patient values. The number needed to treat (NNT) is the number of patients that need to be treated to prevent one additional adverse event, and it is the inverse of the absolute risk reduction, as described by McAlister et al.,16 who provide a detailed guide to individualizing evidence from research. The number needed to harm (NNH), in contrast, is the number of patients treated who would be expected to experience one adverse event. NNT and NNH illustrate the balance between benefits and risks of a given intervention; in oncology, this is particularly relevant with regard to screening, which may result in harms from follow-up of false positives, and treatment, which often produces dose-dependent toxicity.

Clinicians should also consider the following levels of decision making when thinking about applying evidence to a particular patient, as conceptualized by Dr. Leon Gordis of Johns Hopkins University:

Level 1: "Would you have this done for yourself or for someone else in your immediate family?" Influenced by one's personal experience with the disease and capacity to deal with risk.

Level 2: "Would you make this recommendation for your own patients?" Also influenced by prior experience, but

TABLE 1.2. (continued)

Study design

Description

Nested case-control study

Cross-sectional study (also called prevalence study)

Ecologie studies

Case series or case reports

Preclinical studies: animal studies

Preclinical studies: laboratory studies below the whole-organism level

Analysis similar to case-control study, but subjects sampled from within cohort study.

Exposure and disease assessed at one point in time.

May be repeated with different population samples at set time intervals (e.g., annual surveys).

Population-wide disease incidence or prevalence is compared with population-wide exposure estimates.

Descriptions of disease manifestations or therapy outcomes in single or multiple individual subjects, without controls; data often collected retrospectively.

Experimental design using animals, typically genetically homogeneous.

Experimental design utilizing controlled conditions, often involving effects of toxins or drugs on immortalized cancer cell lines or other cells.

For prospective studies, combines efficiency of case-control design with ability to demonstrate that exposure precedes disease.

Useful for measuring prevalence of an exposure or disease.

Useful for generation of hypotheses and evaluating associations (as opposed to cause and effect).

Repeated sampling design allows evaluation of population trends.

Useful for generating hypotheses.

Useful for exposures that cannot be estimated on an individual level (e.g., ambient pollution).

May be the only feasible design for extremely rare diseases.

Large studies of animals (especially rodents) are useful for screening drugs and other chemicals for toxic or therapeutic effects.

Represent a level of control usually unattainable or unethical in humans.

Generate hypotheses for human studies.

Can elucidate biological mechanisms.

See prospective cohort study.

Exposure cannot be shown to precede disease. Inefficient for studying rare diseases.

Does not allow consideration of interindividual differences, which obscures confounding effects.

Lack control group.

Subject to selection bias.

Tend to lack information on confounders.

May be inappropriate to extrapolate from rodent and other species to humans.

Doses of chemicals used in toxicity studies may not be extrapolatable to doses likely to be experienced by humans.

Lack organismal context, and therefore difficult to extrapolate to whole humans.

the strength of the scientific evidence may play a greater role.

Level 3: "Would you make an across-the-board recommendation for a population?" Must be based even more on rigorous assessment of the scientific evidence.

Level 1 is the level at which we all operate when we are making our own personal decisions regarding a procedure; it rests heavily on our own personal value systems and tradeoffs. Nevertheless, it is important not to impose our own value systems on our patients. Level 2 is one in which clini-

TABLE 1.3. Levels of evidence.

Herbal Remedy Secret Uncovered

Herbal Remedy Secret Uncovered

Discover How To Use Herbal Medicine Effectively To Heal Away Disease amp illnesses That Most Of The Herbalist Do Not Want You To Know About. If You Have Never Know What Is All About Herbal Medicines amp The Correct Way Of Using Herbs To Build A Healthier Life, Then This Guide Is About To Reveal All Just That.

Get My Free Ebook


Post a comment