Table 3.
Characteristics significantly associated with patient and provider delay in bivariate analysis among the 104 patients with a diagnosis of tuberculosis*
| Patient Delay | ||
| Characteristic, n (%) | Median Patient Delay (IQR) | p-value** |
| Gender | ||
| Male, 39 (37.5) | 30 (7–60) | 0.002 |
| Female, 65 (62.5) | 14 (7–21) | |
| First visit | ||
| Private general practitioner, 34 (33.0) | 10.5 (7–14) | 0.01 |
| Clinic, 57 (55.3) | 14 (7–30) | |
| Traditional healer, 10 (9.7) | 21 (7–60) | |
| Hospital, 2 (1.9) | 30 (30–30) | |
| Any cough reported | ||
| Yes, 81 (77.9) | 14 (7–30) | 0.003 |
| No, 23 (22.1) | 7 (4–14) | |
| Provider Delay | ||
| Characteristic, n (%) | Median Provider Delay (IQR) | p-value** |
| Employment status | ||
| Unemployed/grant/pension, 77 (74.0) | 39 (15–60) | 0.02 |
| Employed/casual work/student, 27 (26.0) | 19 (7–46) | |
*The following variables were assessed for their bivariate relationship with patient and provider delay: gender, age, marital status, dependents, employment status, educational level, working before illness, type of TB, HIV status, first symptom, symptoms during course of illness, cough as symptom and as first symptom, consulted a traditional healer, place of first visit, sputum in primary care, chest radiograph in primary care, and home remedy used. For those not presented in the table, the associations were found to be non-significant.
**p-values represent comparisons of medians across different categories within each variable.