Table 4

Results of the regression analysis for microscopy for the diagnosis of UTI

Univariate analysis

Multivariate model

Variable*

RDOR (95% CI)

p-value

RDOR (95% CI)

p-value

Microscopy for pyuria: weighted (n = 28 studies)

Selection criteria reported: yes vs no

2.4 (1.0, 5.9)

0.057

Dropped^$

Reference standard details reported: yes vs no

3.6 (0.8, 16.1)

0.089

1.3 (1.1, 1.6)

0.007

Test review bias avoided: yes vs unclear

4.8 (1.6, 14.7)

0.008

Dropped

Diagnostic review bias avoided: yes vs unclear

5.5 (1.8, 17.1)

0.005

Dropped

Uninterpretable results reported: no vs unclear

0.3 (0.0, 4.1)

0.364

Dropped

Uninterpretable results reported: yes vs unclear

2.9 (0.9, 9.1)

0.073

Dropped

Withdrawals accounted for: no vs unclear

0.2 (0.1, 0.7)

0.012

0.2 (0.1, 0.4)

<0.001

Withdrawals accounted for: yes vs unclear

1.9 (0.7, 5.3)

0.200

1.8 (1.0, 3.4)

0.056

Europe vs North America

0.3 (0.1, 1.6)

0.190

Dropped

Asia vs North America

0.6 (0.1, 6.6)

0.646

Dropped

Other areas vs North America

0.2 (0.0, 1.3)

0.090

Dropped

Age <5 years vs <2 years

0.1 (0.0, 0.4)

0.004

Dropped

Age <12 years vs <2 years

0.1 (0.0, 0.3)

0.004

Dropped

Age <18 years vs <2 years

0.2 (0.1, 0.5)

0.001

Dropped

Sample centrifuged: yes vs no

0.3 (0.1, 0.6)

0.005

0.2 (0.1, 0.3)

<0.001

Microscopy for pyuria: unweighted (n = 28 studies)

Age <5 years vs <2 years

0.3 (0.1, 1.1)

0.076

Dropped

Age <12 years vs <2 years

0.5 (0.1, 2.1)

0.337

Dropped

Age <18 years vs <2 years

0.3 (0.1, 1.0)

0.048

Dropped

Sample centrifuged

0.5 (0.2, 1.1)

0.080

0.5 (0.2, 1.1)

0.080

Microscopy for bacteriuria: weighted (n = 22 studies)

Incorporation bias avoided: no vs yes

24.5 (1.0, 604.4)

0.050

3.0 (1.6, 5.5)

0.001

Diagnostic review bias avoided: yes vs unclear

3.2 (0.8, 12.8)

0.092

Dropped

Gram stain used: yes vs no

3.6 (1.3, 10.4)

0.018

5.3 (2.3, 12.0)

0.001

Microscopy for bacteriuria: unweighted (n = 22 studies)

Disease progression bias: yes vs unclear

0.05 (0.0, 1.5)

0.083

Dropped

Incorporation bias avoided: no vs yes

32.5 (1.2, 895.0)

0.041

3.2 (1.6, 6.4)

0.003

Uninterpretable results reported: yes vs unclear

7.1 (1.1, 46.9)

0.042

Dropped

Withdrawals reported: no vs unclear

6.6 (1.0, 43.3)

0.049

Dropped

Withdrawals reported: yes vs unclear

2.2 (0.3, 18.7)

0.447

Dropped

Sample centrifuged: yes vs no

0.2 (0.0, 1.1)

0.058

Dropped

Gram stain used: yes vs no

3.9 (0.9, 16.3)

0.062

6.5 (2.0, 21.2)

0.004

*Only items that showed moderate evidence (p < 0.10) for an association with the DOR in the univariate analysis are included.

^$Items were dropped where there were too few studies in one category to allow a coefficient to be calculated.

Westwood et al. BMC Medical Research Methodology 2005 5:20   doi:10.1186/1471-2288-5-20