Table 4.
Sociodemographic characteristics as predictors for low PTHL
Univariate analyses | OR | 95% CI | p value |
---|---|---|---|
Agea | 1.035 | 1.014–1.058 | 0.001 |
Education | |||
(College/university/post-graduate) | 0.755 | 0.602–0.947 | 0.015 |
Monthly income per family membera | 0.619 | 0.420–0.912 | 0.015 |
Therapy for DMT2 (Insulin/Insulin and OHA) | 0.564 | 0.343–0.926 | 0.024 |
Active excercisea | 0.770 | 0.603–0.984 | 0.037 |
Smoking (Non smoker) | 0.770 | 0.603–0.984 | 0.037 |
Alcohol intakea | 1.429 | 1.072–1.906 | 0.037 |
Source of health informationa | |||
Pharmacists | 0.381 | 0.197–0.735 | 0.004 |
Internet | 1.848 | 0.569–6.004 | 0.307 |
Other | 2.464 | 1.364–4.453 | 0.003 |
Interest in healtha | 0.763 | 0.625–0.931 | 0.008 |
Self-estimation of health statusa | 1.236 | 1.007–1.518 | 0.043 |
Multivariate analyses | OR | 95% CI | p value |
Smoking (Non smoker) | 0.784 | 0.616–0.997 | 0.048 |
Interest in healtha | 0.439 | 0.255–0.757 | 0.004 |
Source of health informationa | |||
Pharmacists | 0.301 | 0.151–0.601 | 0.001 |
Internet | 0.760 | 0.258–2.242 | 0.619 |
Other | 1.471 | 0.862–2.510 | 0.157 |
Self-estimation of health statusa | 0.367 | 0.156–0.863 | 0.021 |
aAge – continuous variable, Monthly income per family member – ordered variable (< 40,000 RSD, 40,000–60,000 RSD, > 60,000 RSD 1 to 3), Active exercise – ordered variable (never, less than once a week, 1–2 times a week, 3 and more times a week were coded from 1 to 4), Smoking – ordered variable (smoker and non-smoker were ordered from 1–3), Alcohol intake – ordered variable (never, once a month, 2 or more times a month were ordered from 1–3), Source of health information – information received from doctors represented reference group, information received from Pharmacists, Internet and Other are coded 1,2 and 3 respectively, Interest in health – ordered variable (Not interested/Little, Medium and Much and very interested were ordered from 1–4), Self-estimation of health status (Bad, Good and very good were ordered from 1–3). Other Multivariate analyses were performed with significant predictors from univariate analyses