Level of Evidence
Four studies were rated with evidence levels of
low;8,21-23 seven studies with
moderate,24-30 and three studies with high (Tables
1,2).31-33 In terms of lack of methodologyrigor, institutional review board approval was not reported in three of
the studies.21,26,29 Power calculations for sample
size were not reported in 10 of the
studies.8,21-26,28-30 Randomization procedures were
not used in 12 studies.8,21-30,32 Six studies had
sample sizes of less than 50
participants.8,22,23,27-30 Two studies were limited by
selection bias (convenience sampling),21,31 and two
studies used outcome measures without discussion of
validity.26,29 Consistency was evident in seven
studies where reproducibility of measures and interventions was
reported.8,22,25,27,31-33 Directness prompts
for an evaluation of similarities between comparison groups and was
discussed in comparison studies of different types of
kyphosis21,22,24,25 and in one study in which there
was an unequal population comparison due to a predominance of
females.28 Effect size was most certain in the
three studies that rated as high level of evidence due to power
analysis, adequate sample size, narrow confidence intervals, or large
effect size.31-33 Importantly, all 3 studies with a
high level of evidence were identified in the Biologic Perspective
search (Table 1).