Methods of using research findings at the level of individual patients

After the study has been critically appraised for its validity, the clinician needs to determine what the results are, and their importance for the patient. Patients in research studies are always different from those in real-life clinical practice in ways that may be difficult to assess. Therefore the use of results from research studies in clinical practice should be cautious and always requires a degree of extrapolation. The contribution of clinical epidemiology is in developing methods of applying research results to individual patients that are biologically and statistically robust and are explicit about any assumptions made. One of the most useful questions to ask when applying the results of a research study is: 'Is my patient so different from those in the study that the results cannot be used?'. The next step is to try to interpret the study results for a particular patient, in terms of his or her clinical characteristics and treatment preferences.

Odds ratios are often used in systematic reviews and meta-analyses because they have useful statistical properties. In particular, they allow treatment effects to be combined across trials with different event rates and of different durations. However, they are of limited clinical use because they are difficult to interpret clinically. (26)

A more clinically useful measure is the number needed to treat (NNT).(27) The NNT is an estimate of the number of patients that would need to be treated with the intervention of interest, compared with the alternative, in order to achieve one good outcome or to avoid one harmful outcome. The NNT is calculated by taking the reciprocal of the difference between the rates of the outcome of interest in the experimental and control groups.

Example (continued)

It may be considered that a selective serotonin uptake inhibitor (SSRI) is more appropriate then a tricyclic for patients with dysthymia because, in the review, acceptability (as measured by the overall drop-out rate from the clinical trials) was better for SSRIs than for tricyclics. A psychiatrist may want to know how effective an SSRI is likely to be in producing a full remission of symptoms rather than merely a response to treatment. The Cochrane review identifies two studies that directly answer this question. One of these(24) was a multicentre trial performed in 17 university-affiliated research clinics in the United States including 410 outpatients (mean age 42 years, 65 per cent women) who met DSM-IIIR criteria for primary dysthymia with early onset (occurring before age 21 years). One group of 134 patients were allocated to sertraline, initially 50 mg, which was increased at the physician's discretion in 50-mg increments after 3 weeks at weeks 4, 6, and 7 to a maximum dose of 200 mg/day. A second group of 136 patients were allocated to imipramine, increased weekly in 50-mg increments from 50 to 300 mg/day unless a therapeutic response was achieved or adverse effects were dose limiting. The remaining 140 patients were allocated to matching placebo.

Beat Depression Now

Beat Depression Now

Let me be up front. My intention is to sell you something. Normally, it's not wise to come out and say that. However, I can do so because I have such an incredible deal for you that you'd be crazy to pass on it.

Get My Free Ebook

Post a comment