Survival Guide to Clinical Research (2016) - Dr. Fang

Survival Guide to Clinical Research

I. General Comments

Understanding the vocabulary of clinical research, including statistical terms, and also the pitfalls of research, is critical to the practice of medicine.

Clinical research is as complex as proverbial rocket science, as it combines science and the element of human behavior. People are inherently biased and try to skew results to their expectations. Thus, the best research is conducted when there is sufficient uncertainty in the outcomes by both study investigators and patients that bias does not skew the results too far in any direction. Such skew leads to poor generalizability of the results and wasted time and money in the pursuit of inappropriate interventions. Complicating matters further is the need for pivotal research to establish safety and efficacy for regulators. Keep in mind that in the USA, the approval standards for procedures and devices is far less stringent than for pharmaceuticals at this time.

II. Vocabulary

A. Phase
B. Prospective
C. Retrospective
D. Case series
E. Inclusion / Exclusion Criteria
F. Adverse Event
G. Adverse Drug Reaction
H. Humanitarian Device Exemption
I. Investigational New Drug
J. Intention-to-Treat Analysis
K. Meta-analysis
L. New Drug Application
M. Primary outcome measure
N. Dose-response curve
O. Non-inferiority study
P. Equipoise
Q. FDA (510k) guidance

III. Statistics

A. Parametric
B. Non-parametric
C. Normal distribution
D. Odds ratio (Hazard ratio)
E. Hosmer-Lemeshow goodness of fit test for Survey data
F. Jaccard Index - This is simply a ratio of the intersecting portion of two sets vs. the total of the sets
G. Correlation vs. Regression
H. Cronbach Alpha - A method of measuring internal consistency within a confined population. Basically, larger sample sizes and lower variance result in higher alpha. High alpha translates to greater internal consistency. Think of this as a coefficient of consistency within a sample.
I. Bootstrapping - A method of assessing sampling error or bias by resampling from the original sample. This technique can be used multiple times. For bootstrapping to be statistically valid, the samples must be independent.

IV. Pitfalls

A. Unqualified investigators
B. Bias
C. Poor study design

1. Irrelevant question
2. Too restrictive
3. Unrealistic expectations
4. Lack of adequate controls

D. Failure to account for natural history of the disease
E. Failure to account for placebo effects
F. Failure to account for comorbidities

1. Inadequate sample
2. Inadequate blinding (or masking), where applicable
3. Lack of independence between variables (eg. studying the effect of age on mortality)

G. Unexpected changes in disease status