Clinical Trial Design

Author: Mark Karadsheh

Topic updated on 11/24/12 12:03pm

Introduction

Clinical trial design impacts Level of Evidence
Clinical trials may be either observational or experimental
- observational
  - researchers observe patient groups without allocation of intervention
  - may be either prospective or retrospective
  - may be descriptive or analytic
    - descriptive
      - useful for obtaining background information for more advance studies
      - examples
        
        case reports
        
        case series
        
        cross-sectional studies
    - analytic
      - explores the association between a given outcome and a potentially related variable
      - examples
        
        case-control
        
        cohort
        
        meta-analysis
- experimental
  - researchers allocate treatment
  - allows the evaluation of efficacies of therapeutic interventions
  - examples
    - double-blinded, prospective, randomized clinical trial is the gold standard for evidence based medicine

Randomized controlled trial

Definition
- a study in which patients are randomly assigned to the treatment or control group and are followed prospectively
- provides the most compelling evidence that the study treatment causes the expected effect on human health
- randomization minimizes study bias
Crossover design
- administration of two or more therapies, one after the other, in a random order
- susceptible to bias if washout period is inadequate
Analysis
- intent-to-treat analysis
  - outcomes based on the group into which they were randomized, regardless of whether the patient actually received the planned intervention
  - minimizes non-responder bias
- per protocol
  - excludes patients who were not compliant with the protocol guidelines
Example
- you want to determine whether your new toothpaste prevents cavities better than your old toothpaste. You randomly assign a large number of patients to either an intervention group, which uses the new toothpaste, or to a control group, which uses the old toothpaste. You would then measure the amount of cavities between the groups over time.
Orthopaedic Literature Examples

Cohort study

Definition
- a study in which patient groups are separated non-randomly by exposure or treatment, with exposure occurring after (prospective), or before (retrospective), the initiation of the study
Evidence
- Level II or III evidence
Analysis
- results usually reported as relative-risk
Example
- you want to determine if smoking is a risk factor for the development of lung cancer. You identify a group of smokers and a group of non-smokers, and follow them over time measuring the desired outcome, in this case, lung cancer.
Orthopaedic Literature Examples

Case-control study

Definition
- a study in which patient groups are separated by the current presence (cases) or absence (controls) of disease and examined for the prior exposure of interest
Evidence
- Most are Level III evidence
Analysis
- usually reported as odds-ratio
Example
- you want to determine if smoking is a risk factor for the development of lung cancer. You compare the smoking history of individuals with lung cancer (cases) and those without (controls).
Orthopaedic Literature Examples

Meta-analysis

Definition
- a systematic review that summarizes results of other studies
Evidence
- may be used in increase the statistical power of several under-powered studies
Example
- you want to determine if wearing sunscreen results in fewer cases of melanoma. You pool the results of 9 randomized controlled studies and statistically analyze the data to determine the effect of the relationship.
Orthopaedic Literature Examples

Cross-sectional Study

Definition
- study group is analyzed at a given time ("snapshot") with no follow-up
Example
- you want to determine the prevalence of baseball injuries during the 2003 little-league season
Orthopaedic Literature Examples

Case Series

Definition
- a retrospective account of multiple patients with the same injury or treatment with no control or comparison group
  - useful for generating hypotheses for additional studies
Evidence
- level IV evidence
Example
- you have found that several of your patients who have used a new lipid lowering medication have developed hemorrhagic cysts. You want to alert other members of the community of this possible association.
Orthopaedic Literature Examples

Please Rate Educational Value!

3.0

Average 3.0 of 5 Ratings

Qbank (4 Questions)

TAG

(OBQ09.34) What is the primary purpose of utilizing a computer random number generator for patient randomization when designing a Level I therapeutic controlled trial? Topic

Review Topic

1. To increase treatment effect
2. To prevent selection bias
3. To generate enough power
4. To achieve statistical significance
5. To perform as-treated analysis

PREFERRED RESPONSE ▶

DISCUSSION: A randomized controlled trial is the study design that can provide the most compelling evidence that the study treatment causes the expected effect on human health. The advantages of proper randomization include:

It eliminates bias in treatment assignment, specifically selection bias and confounding.

It facilitates blinding (masking) of the identity of treatments from investigators, participants, and assessors.

It permits the use of probability theory to express the likelihood that any difference in outcome between treatment groups merely indicates chance.

The reference by Poolman et al reviewed all JBJS articles from 2003-2004 and found that 31% of the studies used modified outcome questionnaires, and frequently did not describe how they modified the questionnaires. They also recommend that operating surgeons do not collect the outcome data, in order to minimize bias.

REFERENCES:

1. Einhorn TA, O’Keefe RJ, Buckwalter JA (eds): Orthopaedic Basic Science: Foundations of Clinical Practice, ed 3. Rosemont, IL, American Academy of Orthopaedic Surgeons, 2007, pp 87-101

2. Poolman RW, Struijs PA, Krips R, et al: Reporting of outcomes in orthopaedic randomized trials: Does blinding of outcome assessors matter? J Bone Joint Surg Am 2007;89:550- 558 PMID:17332104 (Link to Abstract)

Question Authors:

Ben Taylor MD, Patrick McCulloch MD, Ben Taylor MD,

Please Rate Educational Value!

2.0

Average 2.0 of 5 Ratings

TAG

(OBQ05.86) A clinical study for lateral epicondylitis allocates 1 group to receive physical therapy for 4 weeks and another group a new oral medication for 4 weeks. Then the 2 groups immediately switch therapies with one another for the next 4 weeks. The half-life of the medication used in the study is 2 weeks. Which of the following best describes the bias that is present in this study design? Topic

Review Topic

1. Recall bias
2. Verification bias
3. Washout period bias
4. Detection bias
5. Incorporation bias

PREFERRED RESPONSE ▶

DISCUSSION: In a crossover study design, the washout period is the time between therapies, receiving no therapy, so that the effect of the first therapy is allowed to wear off. In this scenario, the medication's long half-life may continue to have effects after the first group (receiving the medication) has switched to the physical therapy treatment. Verification bias occurs when results of a diagnostic test influence whether patients are assigned to a treatment group. Incorporation bias occurs when someone studies a diagnostic test that includes features of the target outcome. Recall bias occurs when patients who experience an adverse outcome have a different likelihood of recalling an exposure than others that don't have an adverse outcome. Detection bias occurs when one looks more carefully for an outcome in one specific group more than the other group(s). The review article by Kocher et al describes the basics of biostatistics for clinicians.

REFERENCES:

1. Buckwalter JA, Einhorn TA, Simon SR (eds): Orthopaedics Basic Science: Biology and Biomechanics of the Musculoskeletal System, Ed. 2. Rosemount, IL, American Academy of Orthopaedic Surgeons, 2000, pp 1-17.

2. Kocher MS, Zurankowski D: Clincal epidemiology and biostatics: A primer for orthopaedic surgeons. J Bone Joint Surg Am 2004;86:607-620 PMID:14996892 (Link to Abstract)

Question Authors:

Michael Hughes MD, Patrick McCulloch MD,

Please Rate Educational Value!

1.0

Average 1.0 of 9 Ratings

TAG

(OBQ04.149) A research study is initially designed as a high-quality randomized control trial. The study seeks to demonstrate a different clinical outcome with treatment A compared to treatment B for a specific disease. However, at the completeion of the study the investigators are only able to obtain 70% follow-up of patients. What level of evidence should be assigned to this study? Topic

Review Topic

1. 1
2. 2
3. 3
4. 4
5. 5

PREFERRED RESPONSE ▶

DISCUSSION: A randomized clinical trial (RCT) is a study design that is appropriate to demonstrate statistically that a clinical outcome in one treatment group is different from that in another treatment group. A properly designed RCT that meets certain criteria is considered Level 1 evidence.

AAOS Comprehensive Orthopaedic Review and the article by Wright et al states that randomized control trials that have "less than 80% follow-up, no blinding, or improper randomization" are lesser-quality studies and qualify as Level 2 evidence. This is also the criteria used by JBJS-Am for assigning levels of evidence for RCT's. The RCT in this question has a 70% follow up and therefore would be considered Level 2 evidence. Level 2 evidence also includes prospective cohort studies whereby patient groups are separated non-randomly by exposure or treatment, with exposure occurring after the initiation of the study. Level 3 evidence includes retrospective cohort studies and case-control studies whereby patient groups are separated by the current presence or absence of disease and examined for the prior exposure of interest. Level 4 evidence includes a case series of multiple patients with the same treatment, but no control group or comparison group. Level 5 evidence includes a case report, expert opinion, or anecdotal evidence.

REFERENCES:

1. Buckwalter JA, Einhorn TA, et al. (eds) Orthopaedic Basic Science: Biology and biomechanics of the musculoskeletal system, ed 2. Rosemont, IL AAOS, 2000 p 1-17

2. Huley SB, Cummins SR: Design in clinical research: An epidemiological approach. Baltimore, MD Williams & Wilkins, 1988

3. Wright JG, Swiontkowski MF, Heckman JD. Introducing levels of evidence to the journal. J Bone Joint Surg Am. 2003 Jan;85-A(1):1-3. PMID:12533564 (Link to Abstract)

Question Authors:

Michael Hughes MD, Patrick McCulloch MD,

Please Rate Educational Value!

3.0

Average 3.0 of 12 Ratings

Posts

Reporting of outcomes in orthopaedic randomized trials: does blinding of outcome assessors matter? Includes Poll

Poolman RW, Struijs PA, Krips R, Sierevelt IN, Marti RK, Farrokhyar F, Bhandari M
J Bone Joint Surg Am. 2007 Mar;89(3):550-8. PMID: 17332104 (Link to Pubmed)

Basic Science - Journal Club - Clinical Trial Design

3 weeks ago

12 responses

See More Posts

Evidence & References Show References

Level of Evidence 4 (Case Series)

Poolman RW, Struijs PA, Krips R, et al: Reporting of outcomes in orthopaedic randomized trials: Does blinding of outcome assessors matter? J Bone Joint Surg Am 2007;89:550- 558 PMID:17332104 (Link to Abstract)

Level of Evidence 5 and Other Journal Articles (includes Case Reports, Expert Opinions, Personal Observations, and Biomechanic Studies)

Wright JG, Swiontkowski MF, Heckman JD. Introducing levels of evidence to the journal. J Bone Joint Surg Am. 2003 Jan;85-A(1):1-3. PMID:12533564 (Link to Abstract)
Berger VW. A review of methods for ensuring the comparability of comparison groups in randomized clinical trials. Rev Recent Clin Trials. 2006 Jan;1(1):81-6. Review PMID: 18393784 (Link to Abstract)
Kocher MS, Zurankowski D: Clincal epidemiology and biostatics: A primer for orthopaedic surgeons. J Bone Joint Surg Am 2004;86:607-620 PMID:14996892 (Link to Abstract)

Textbooks

Review of Orthopaedics, 6th Edition, Mark D. Miller MD, Stephen R. Thompson MBBS MEd FRCSC, Jennifer Hart MPAS PA-C ATC, an imprint of Elsevier, Philadelphia, Copyright 2012
AAOS Comprehensive Orthopaedic Review, Jay R. Leiberman. Published by American Academy of Orthopaedic Surgeons, Rosemont IL. Copyright 2009
Orthopaedic Knowledge Update 10, John M Flyn. Published by American Academy of Orthopaedic Surgeons, Rosemont IL. Copyright 2011
Hoppenfeld SP. Surgical Exposures in Orthopaedics: The Anatomic Approach. Lipponcott, Williams, and Wilkins, Philadelphia, PA, Copyright 2009
Orthopaedic In-training Examination (OITE) Questions 2004-2012, American Academy of Orthopaedic Surgeons, Rosemont IL. Copyright 2004-2012
Self-Assessment Examination (SAE) Questions 2004-2012, American Academy of Orthopaedic Surgeons, Rosemont IL. Copyright 2004-2012
Bhandari M. Evidence-based orthopaedics: issues in research design, analysis, and critical appraisal. In: Einhorn TA, O'Keefe RJ, Buckwalter JA, eds. Orthopaedic Basic Science: Foundations of Clinical Practice. 3rd ed. Rosemont, IL: American Academy of Orthopaedic Surgeons; 2007:87-101.
Buckwalter JA, Einhorn TA, et al. (eds) Orthopaedic Basic Science: Biology and biomechanics of the musculoskeletal system, ed 2. Rosemont, IL AAOS, 2000 p 1-17
Buckwalter JA, Einhorn TA, Simon SR (eds): Orthopaedics Basic Science: Biology and Biomechanics of the Musculoskeletal System, Ed. 2. Rosemount, IL, American Academy of Orthopaedic Surgeons, 2000, pp 1-17.
Einhorn TA, O’Keefe RJ, Buckwalter JA (eds): Orthopaedic Basic Science: Foundations of Clinical Practice, ed 3. Rosemont, IL, American Academy of Orthopaedic Surgeons, 2007, pp 87-101
Huley SB, Cummins SR: Design in clinical research: An epidemiological approach. Baltimore, MD Williams & Wilkins, 1988

Topic Comments

Subscribe status: On Off

Recent on Bottom | Recent on Top

Page:1

Clinical Trial Design

Qbank (4 Questions)

Question Comments

Question Comments

Question Comments

Posts

Evidence & References Show References

Topic Comments