Skip to main content
. 2011 May 10;5(5):e1029. doi: 10.1371/journal.pntd.0001029

Table 3. Results of univariate and multivariate logistic regression analysis with a backwards elimination procedure.

Univariate Multivariate
Variable Control N (%) Case N (%) Crude Odds Ratio (95% CI) p-value Odds ratio (95% CI) p-value
Wealth quintile (asset index) 1 40 (20.1%) 25 (27.8%) 1.00
2 40 (20.1%) 20 (22.2%) 0.80 (0.38–1.67)
3 40 (20.1%) 16 (17.8%) 0.64 (0.30–1.38)
4 40 (20.1%) 17 (18.9%) 0.68 (0.32–1.45)
5 39 (19.6%) 12 (13.3%) 0.49 (0.22–1.12)
Assuming a linear trend 0.85 (0.71–1.02) p = 0.083
Income level (BDT) Mean 4108 4853
Std. Dev. 3978 3991 1.00 (1.00–1.00) p = 0.148
Educational level* High 113 (56.8%) 49 (54.4%) 1.00
Low 86 (43.2%) 41 (45.6%) 1.10 (0.67–1.81) p = 0.711
Household size Mean 5.28 4.96
Std. Dev. 2.30 1.97 0.93 (0.82–1.05) p = 0.248
Crowding No 129 (64.8%) 55 (61.1%) 1.00
Yes 70 (35.2%) 35 (38.9%) 1.17 (0.70–1.96) p = 0.544
Ever food shortage No 76 (38.2%) 30 (33.3%) 1.00
Yes 123 (61.8%) 60 (66.7%) 1.24 (0.73–2.09) p = 0.428
Food shortage in the last year No 128 (64.3%) 47 (52.2%) 1.00 1.00
Yes 71 (35.7%) 43 (47.8%) 1.65 (1.00–2.74) p = 0.052 1.79 (1.06–3.02) p = 0.030
Sex Female 116 (58.3%) 41 (45.6%) 1.00
Male 83 (41.7%) 49 (54.4%) 1.67 (1.01–2.76) p = 0.045
Age (years) <10 22 (11.1%) 6 (6.7%) 1.00 1.00
10–19 65 (32.7%) 20 (22.2%) 1.13 (0.42–3.17) p = 0.819 1.17 (0.41–3.32) p = 0.762
20–29 27 (13.6%) 21 (23.3%) 2.85 (0.98–8.30) p = 0.054 3.22 (1.09–9.51) p = 0.034
30–39 32 (16.1%) 15 (16.7%) 1.72 (0.58–5.12) p = 0.331 1.84 (0.61–5.55) p = 0.277
40–49 35 (17.6%) 11 (12.2%) 1.15 0.37–3.56) p = 0.805 1.28 (0.38–3.67) p = 0.781
50+ 18 (9.1%) 17 (18.9%) 3.46 (1.13–10.61) p = 0.030 3.56 (1.15–11.02) p = 0.028
Total 199 (100%) 90 (100%)

*Educational level: Low: highest educated person in the household had 0–5 years of schooling; High: highest educated person had more than 5 years of schooling.

‡Crowding: for this study defined as more than three people per sleeping room (average).