Downs and Black developed a checklist for evaluating the methodological quality of randomized and non-randomized studies of a health care intervention. The authors are from the London School of Hygiene and Tropical Medicine.
Number of items: 27
9 items answered Yes or No:
Is there a clear description of the study objective?
Are primary outcomes clearly described in the introduction and/or Methods section?
Are the characteristics of the patients clearly described ?
Are the interventions of interest clearly described?
Is there a clear description of the main findings?
Is there an estimate of the random variability in the data for the main outcome(s)?
Have all important adverse events associated with the intervention been reported?
Is there a clear description of the patients lost to follow-up?
Are any probability values for main outcomes between 0.001 and 0.049 reported?
|
Response
|
Points
|
|
no
|
0
|
|
yes
|
1
|
16 items answered Yes, No or unknown (unable to determine):
(1) subjects asked to participate representative
(2) subjects prepared are representative
(3) treatment facilities are representative
(4) subjects blinded
(5) observers blinded
(6) results from "data dredging" indicated
(7) adjustments for different folllow-ups
(8) statistical tests appropriate
(9) compliance with intervention reliable
(10) outcome measures accurate
(11) subjects in different groups from the same population
(12) subjects in different groups recruited over the same period of time
(13) subjects randomized to intervention groups
(14) randomized intervention assignment concealed from all until complete
(15) adequate adjustments made for confounding
(16) losses of patients on follow-up accounted for
|
Response
|
Points
|
|
no
|
0
|
|
yes
|
1
|
|
unknown
|
0
|
1 item answered Yes, partially and No
(1) The distribution of principal confounders in each group is clearly described.
|
Response
|
Points
|
|
no
|
0
|
|
partially
|
1
|
|
yes
|
2
|
1 item scored from 0 to 5
(1) The study has sufficient power to detect a clinically important effect where the probability value for a difference due to chance is less than 5%.
|
Size of Smallest Intervention Group
|
Points
|
|
less than n1
|
0
|
|
n1 - n2
|
1
|
|
n3 - n4
|
2
|
|
n5 - n6
|
3
|
|
n7 - n8
|
4
|
|
n8+
|
5
|
total score =
= SUM(points for all 27 items)
Interpretation:
• minimum score: 0
• maximum score: 32 (the paper states 31)
• The higher the score the better the study quality.
