Slide 1
Everyone must be a skeptic about clinical information. I can tell you what information was used by the panel convened by WHO, but you should decide for yourself.
Slide 2
Individually skeptical to change this workshop. All studies are cited on the page.
Slide 3
Capacity and capability
Slide 4
Research has gaps, some of the evidence is lacking.
Information is not available
Things get in the way of us practicing the best way
Slide 6
1a: Reviewers search for all randomized controlled trials on a subject, review the quality of the studies, and combine results that are of the best quality available and give a pooled result and recommendations for practice and for further research. Systematic reviews of RCTs are in Cochrane Library.
1b: Individual RCT compares two randomly assigned groups with similar characteristics. One group gets the intervention being studied and the other gets a placebo or whatever the routine treatment is. Outcomes are evaluated to see which treatment is better.
2a: A cohort study studies a group of people over time to see if their characteristics are associated with the development of certain outcomes. In a review, several cohort studies on the same subject are evaluated for their quality, results are pooled and a general recommendation is made.
2b: Individual cohort study- results from only one researcher whose group is followed over time
3a: A case-control study compares two similar groups, one with and one without a disease, and looks in their pasts to see if certain risk factors were present. A review compiles the results of several studies and makes conclusions.
3b: Individual case-control study compares two groups (one with and one without disease to see about previous risk factors.
4: Case series describes all the patients with a specific problem.
5: A person gives her/his opinion for practice, without having evidence to support it.
Slide 8
The quality of a test is the ability of the test to say the disease is there or not there.
The usefulness of a result if the ability of a person to believe a positive or negative test result.
Sensitivity = (a/a+c), the likelihood that if someone has HIV that the blood test will be positive.
Specificity = (d/b+d), the likelihood that if someone does not have HIV that the blood test will be negative
PPV = (a/a+b), likelihood that if someone’s HIV test is positive that they have HIV
NPV = (d/c+d), the likelihood that if someone’s HIV test is negative that they really do not have HIV
No test is 100% sensitive of specific.
Slide 11
Increases the power of a group of studies by pooling results.
Slide 12
The important point to learn is that if there is no difference in the outcome between the two groups, the RR (or OR) = 1 (ratio of 1:1 = 1). It is equally likely to happen in both groups. If the outcome is less likely to happen in one group, the RR (or OR) for them is <1.
Slide 14
All RR or OR are not exactly correct, but have range within which they can fall. The range is the confidence interval. It describes how accurate and believable the results are. If the confidence interval overlaps one, the result may simply be due to chance. If the interval range is very wide, the results are less likely to be accurate or meaningful.
Slide 17
It is important to note that results can be statistically significant, but may not be clinically significant (and may not be meaningful or warrant a change in practice.)
