Table 5.
Multivariable regression examining association between reported delivery and aesthetic conditions, sewerage and measured water quality in a sample of households of three rural counties in Alabama 2012.
| Reported delivery and aesthetic conditions | Presence of total coliforms in kitchen samples OR (95%CI) | Presence of total coliforms in outside samples OR (95%CI) | Presence of free chlorine OR (95%CI) | Presence of total chlorine OR (95%CI) | Turbidity > 0.3NTU OR (95%CI) | Log-transformed Pressure Quintiles* OR (95% CI) |
|---|---|---|---|---|---|---|
| Intermittent Service | 0.99 (0.55–1.8) † | 0.82 (0.47–1.4)† | 2.2 (0.50–9.4) | 0.69 (0.14–3.3) † | 1.45 (0.92–2.0) | 0.98 (0.66–1.5) †, ‡ |
| Low Water Pressure | 0.78 (0.50–1.2) † | 0.73 (0.49–1.1) | 0.94 (0.44–2.1) | 2.5 (0.51–12) † | 0.76 (0.57–1.0) | 1.4 (1.1–1.8) |
| Objectionable Taste | 0.82 (0.48–1.4) | 0.80 (0.50–1.3) | 0.99 (0.41–2.4) | 0.29 (0.080–1.1) † | 0.94 (0.67–1.3) | 1.1 (0.82–1.5) |
| Objectionable Odor | 1.0 (0.59–1.8) | 1.17 (0.66–1.7) | 0.92 (0.37–2.3) | 0.21 (0.060–0.77) † | 0.98 (0.68–1.4) | 0.94 (0.67–1.3) |
| Objectionable Color | 1.6 (0.95–2.6) † | 1.36 (0.78–2.0) | 1.5 (0.50–4.4) | 0.93 (0.19–4.5) † | 0.77 (0.53–1.1) | 0.99 (0.70–1.4) |
Notes: * Ordinal logistic regression performed on quintiles of log-transformed pressure; † indicates multivariable model produced ≥10% change in effect size of exposure variable ‡ Logistic regression performed on binary log-pressure when proportional odds assumption was not met; Statistically significant results in bold.