Home > Full Text Reviews > First-Generation Versus... > Risk of Bias Assessment for Randomized...

PubMed Health. A service of the National Library of Medicine, National Institutes of Health.

Abou-Setta AM, Mousavi SS, Spooner C, et al. First-Generation Versus Second-Generation Antipsychotics in Adults: Comparative Effectiveness [Internet]. Rockville (MD): Agency for Healthcare Research and Quality (US); 2012 Aug. (Comparative Effectiveness Reviews, No. 63.)

Appendix ERisk of Bias Assessment for Randomized Controlled Trials and Nonrandomized Controlled Trials

Guidelines and Decision Rules for Risk of Bias Assessments

Sequence generation

-

If computer-generated, random number list, flipping coins, randomly picking envelopes, etc. is specified → YES

-

If the description only includes “random,” “randomly generated,” “randomized,” etc, do not assume additional details → UNCLEAR

-

If the description is quasi-randomized (e.g. alternate randomization, day of the year, day of the month, birth date, birth month, beginning letter of last name, availability of investigator or specialist, etc) → NO

Allocation concealment

-

If the assignment is conducted by central telephone, pharmacy, etc → YES

-

If dark (or opaque), sealed, sequentially-numbered envelopes are used → YES

-

If the envelopes are not stated to dark and sealed, or sequentially-numbered → UNCLEAR

Note: sequential numbering of the envelopes is only required for adequate allocation concealment if the method of randomization was anything other than randomly picking envelopes (i.e., the envelopes were only used for allocation concealment and not as part of the randomization process).

Blinding

-

Describe who is blinded: patient, clinician, outcomes assessor, etc.

-

If the study was stated to be blinded (masked), and the blinding is considered to be possible and not likely to be broken → YES

-

If the study is only stated to be blinded, double-blinded, double-dummy, etc. without any further details → UNCLEAR

-

If the study states the use of a placebo (dummy), but with no further details → UNCLEAR

-

If no mention of blinding → UNCLEAR

Incomplete outcome data (longest time point)

-

Look for intention-to-treat (ITT) analysis (all randomized patients are analyzed) → YES

-

If all participants were accounted for (i.e., no dropouts or censored analysis conducted) → YES

-

If the numbers and reasons for withdrawal or dropouts were described and comparable across groups (and ≤ approximately 10 percent) → YES

-

If there is between 10–30 percent dropout and no ITT analysis → UNCLEAR

-

If there is greater 30 percent dropout and no ITT analysis → NO

Selective outcome reporting

-

If the study protocol is available (referenced in the manuscript), compare the outcomes reported in the publication to those specified in the protocol. If they match → YES

-

If the study protocol is available (referenced in the manuscript), compare the outcomes reported in the publication to those specified in the protocol. If they do not match, but there is reference to another publication with this information presented → YES

-

If the study protocol is not available, compare the outcomes reported in the methods and results sections. If they match → YES

-

If the study protocol is not available, compare the outcomes reported in the methods and results sections. If they do not match → NO

-

If the study protocol is not available, compare the outcomes reported in the methods and results sections. If they match but not in an extractable format (e.g., stating that there was no difference between the groups regarding the outcome) → UNCLEAR

Other sources of bias

-

Assess for baseline imbalances that could have biased the results (or were not accounted for)

-

Assess for appropriateness of crossover design (e.g., inadequate wash-out period).

-

Note any “other” sources of bias

Risk of Bias Assessments

Cover of First-Generation Versus Second-Generation Antipsychotics in Adults: Comparative Effectiveness
First-Generation Versus Second-Generation Antipsychotics in Adults: Comparative Effectiveness [Internet].
Comparative Effectiveness Reviews, No. 63.
Abou-Setta AM, Mousavi SS, Spooner C, et al.

Download

AHRQ (US Agency for Healthcare Research and Quality)

PubMed Health Blog...

read all...

Recent Activity

Your browsing activity is empty.

Activity recording is turned off.

Turn recording back on

See more...