Page 11/32 · Research Design, Biostatistics & Literature Evaluation

Self-Assessment

Research Design, Biostatistics & Literature Evaluation

Julie E. Farrar ~2 min read Module 2 of 20

/ 32

Research Design, Biostatistics, and Literature Evaluation

Practice Case

You are working with a multicenter group to develop a cluster-crossover randomized trial. Patients will be

recruited at 12 large academic medical centers in eight metropolitan cities. The study will involve a compari-

son of intravenous haloperidol with intravenous droperidol for acute agitation in the emergency department.

Which of the following would NOT qualify as an essential practical consideration for this proposed study

design?

A.Informed consent - unclear whether waiver of informed consent would be appropriate because both

agents may be considered current standard of care or whether community consent might be more

appropriate.

B.Social determinants of health - are some patients more likely to be enrolled in study depending on local

perceptions of etiologies of agitation (e.g., drug use, psychological disorders)

C.Subject recruitment - important to ensure that homogeneous populations are enrolled despite institu-

tional and geographic differences that could significantly affect results.

D.Blinding - unnecessary bias is introduced into the study due to the inability to blind patients and pro-

viders to drug selection.

IV.STUDY DESIGN (FIGURES 1 AND 2)

A.Randomized Controlled Trial

Hallmark of clinical research. Study design seeks to minimize bias through randomization of subjects,

blinding of investigators and all participants, and analytic approach. This should leave at least two

groups that differ only in study treatments.

2Experimental design to test the effects of an intervention compared with either placebo or the established

standard of care (treatment or process of care); allows for description and causality

Meta-analysis

RCT

Cohort

Case-control

Cross-sectional

Case reports/series

Opinions and reviews

Figure 1. Hierarchy of clinical evidence.

RCT = randomized controlled trial.

Sevransky JE, Checkley W, Martin GS. Critical care trial design and

interpretation: a primer. Crit Care Med 2010;38:1882-9.

RCT and observational studies

demonstrate consistency

Meta-analysis:

RCT or observational studies

Case reports/series

Opinions and reviews

Figure 2. Proposed revision to hierarchy of clinical

evidence.

Gershon AS, Jafarzadeh S, Wilson KC, et al. Clinical knowledge from

observational studies. Am J Respir Crit Care Med 2018;198:859-67.

HD Video Explanation — Synchronized with PDF

Starts at: minute 10 Open on YouTube