Table 2.
Odds ratios and probabilities that findings of any bacteria in the urine1 are associated with symptoms.
| Statistical correlation between bacteriuria and presence of symptom | Probability (%) for positive/negative culture to rule in/rule out that symptom is associated with bacterial finding | |||
|---|---|---|---|---|
| Symptoms | Prevalence2 | Odds ratio (95% CI) | p-value | Positive/negative etiologic predictive value3 |
| Restlessness | 9.1% | 1.3 (0.76-2.4) | 0.31 | 31 (0-71)/98 (91-100) |
| Fatigue | 12% | 1.7 (1.0-2.7) | 0.046 | 40 (0-72)/97 (91-100) |
| Confusion | 7.5% | 1.9 (1.0-3.5) | 0.044 | 48 (0-80)/96 (86-100) |
| Aggressiveness | 4.3% | 1.7 (0.77-3.9) | 0.19 | 31 (0-78)/98 (87-100) |
| Not being herself/himself | 2.3% | 3.3 (1.1-9.9) | 0.030 | 74 (0-99)/88 (17-100) |
| Dysuria | 1.8% | 1.4 (0.43-4.6) | 0.57 | 37 (0-90)/97 (72-100) |
| Urgency | 5.5% | 1.7 (0.85-3.5) | 0.13 | 38 (0-79)/97 (87-100) |
| Fever | 0.31% | 3.2 (0.17-61) | 0.44 | 56 (0-100)/94 (0-100) |
1Any bacteria were found in the urine in 207 of 651 urine samples.
2Appearance or increase of symptom or signs within the last three months in the studied population. All patients with fever acquired this within the last week.
3Etiologic predictive value (EPV) is a statistical method used to evaluate the clinical usefulness of dichotomous diagnostic tests. It estimates predictive values when there is no proper gold standard for comparison. EPVs are presented in this table with 95% confidence interval within parenthesis. To calculate the EPV the sensitivity of a urinary culture to detect presence of bacteriuria was estimated at 90%.