Skip to main content
. 2016 May 26;11(5):e0156219. doi: 10.1371/journal.pone.0156219

Table 3. The logistic regression analysis to identify factors relevant to achieving at least SI at week 8*.

Variables Estimated Odds Ratio (95% CI) P value
PDT-hemoporfin vs. PDT-placebo 29.324(16.490,54.244) <0.001
Location of PWS (vs. Centrofacial)
    Non-centrofacial 1.107(0.603,2.052) 0.743
    Neck 4.030(1.475,12.629) 0.010
    Other 1.059(0.053,39.141) 0.973
Type of PWS (vs. Pink type)
    Purple type 1.140(0.617,2.092) 0.673
    Hypertrophic type 0.354(0.137,0.951) 0.035

Abbreviations: SI, some improvement; CI, confidence interval.

Denotations

* Candidate continuous variables considered were age, systolic blood pressure (mmHg), diastolic blood pressure (mmHg), heart rate (bpm), body-mass index, area of targeted site (cm2), ΔEI at baseline and power density of laser irradiation (mW/cm2). Candidate categorical variables considered were intervention group (PDT-hemoporfin vs. PDT-placebo), study site (other sites vs. site 1), sex (male vs. female), occupation (physical laborers vs. non-physical laborers), location of PWS (other locations vs. centrofacial) and type of PWS (other types vs. pink type). The variables with significance level of P < 0.2 in univariate analyses were included in the stepwise multivariate logistic regression; ethnicity and previous therapy was not included in the analyses because of the small number of cases in the subgroups. The logistic regression model fit was tested with the likelihood ratio test (P < 0.001).