Skip to main content

Table 2 Univariate and multivariate logistic regression analysis for consulting for a cough (when coughing for more than 2 weeks), 2010

From: Patient predictors of health-seeking behaviour for persons coughing for more than two weeks in high-burden tuberculosis communities: the case of the Western Cape, South Africa

Variables Univariate (n = 922)a Multivariate (n = 921)b
ORc (95% CId) P value OR (95% CI) P value
Stigma index 0.88 (0.76–1.03) 0.13 1.05 (0.89–1.24) 0.58
Agee 1.01 (1.01–1.02) < 0.01 1.02 (1.01–1.04) < 0.01
Men 0.76 (0.57–0.99) 0.04 0.88 (0.63–1.22) 0.44
Black ethnicity 2.42 (1.60–3.67) < 0.01 1.99 (1.28–3.12) < 0.01
Educationf 1.00 (0.97–1.04) 0.87 1.05 (1.00–1.10) 0.05
Employed 0.85 (0.63–1.15) 0.29 0.92 (0.67–1.27) 0.63
Asset index 0.87 (0.77–0.98) 0.03 0.81 (0.71–0.92) < 0.01
HIV positive 1.37 (0.98–1.91) 0.06 1.30 (0.91–1.85) 0.15
Drinks daily 1.13 (0.66–1.95) 0.65 1.43 (0.80–2.54) 0.23
Smoked 0.53 (0.40–0.69) < 0.01 0.63 (0.45–0.88) < 0.01
  1. an = 921 for age variable
  2. b Due to one missing age variable
  3. c OR = odds ratio
  4. dCI = confidence interval
  5. e per year
  6. fper year of education
  7. Reference categories for dichotomous variables: Women, Cape coloured ethnicity, Not employed during last year, HIV negative/did not disclose HIV status, Never/occasional/used to drink, Never smoked. All other independent variables are continuous. Source: Own calculations, ZAMSTAR (2010)
Back to article page

BMC Health Services Research

ISSN: 1472-6963

Contact us