Gender differentials in tuberculosis case detection in a rural district in Kenya
Tuberculosis remains a major health problem and it is estimated that each year, between 50 and 100 million people are infected by the tubercle bacillus and of these, 10 - 20 million develop overt disease and 3 million die. These figures are expected to increase as a result of the rising numbers of tuberculosis cases among persons with HIV and AIDS. The tuberculosis situation in Kenya is further highlighted by lower incidence rates of the disease among women as compared to the men. The aim of this study was to determine some of the factors that determine the gender differences in TB case detection in Kitui District, Kenya. The study was conducted in Kitui district between 7th June and 5th July, 1993. The study consisted of interviews, BeG-scar inspection, Mantoux Test and review of TB case notification data from 2 registering health facilities. A total of 910 respondents were interviewed and 430 of them given a Mantoux Test. The sex distribution of the respondents was 596 (65.5%) females and 314 (34.5%) males. The study revealed higher literacy rates and occupation levels among the males than the females. TB awareness was generally high. Eight-hundred and four (92.4%) of the respondents knew the local name for TB and 205 (22.5%) mentioned TB as one of the five most common diseases in the area. Furthermore, 36% of the respondents felt that TB was a problem in their area. Knowledge of several aspects of TB namely: signs, symptoms, cause, mode of jI transmission and control was significantly higher among the males than the females. However, the level of knowledge of TB prevention was higher among the females (62.4%) than the males (60.1 %). i Attitudes towards TB patients were positive overall, with the males showing more 'liberal' attitudes towards social interactions with TB patients than the females. Both males and females preferred to seek treatment for TB, from a health facility (91.8%) ifthere was need. Most (68.6%) of those interviewed made the decision as to where they should go for treatment if they were unwell. However, a significantly higher proportion of the males (79.6%) made this decision as compared to the females (62.8%). BCG-scar inspection revealed that 62.8% of the respondents had a scar. The presence of a BCG scar was significantly higher (p = 0.0047) among the males (69.0%) than the females (59.5%). Only 430 (49.2%) of those offered the Mantoux test accepted the test and 403 (93.7%) of these were available for the test to be read after 48 - 72 hours. Mantoux sizes ranged from 0- 40 mm, with 42.4% of the respondents being Mantoux positive. There was no difference in Mantoux positivity between the males and females even after controlling for age and presence of a BCG-scar. Mantoux positivity was however, significantly higher (p = 0.009) among those with a BCG-scar (47.9%) as compared to those without (34.8%). A total of 408 cases were notified in 1992 in 2 major registration facilities in the district. The data showed a male:female ratio of 1.85 : 1. From this study, it was found that: 1. The prevalence of tuberculosis infection is the same for both males and females; 2. Women have poorer knowledge about TB than males and their attitudes towards TB patients are less 'liberal' than those of the males. 3. Women are less likely to decide on where they should go for treatment when unwell and more likely to seek money for transport to a health facility from their spouses than the males. 4. TB case detection does not accurately reflect the findings of this study which showed no difference in Mantoux positivity between males and females. The study recommends further studies to determine the true prevalence of TB infection, the differences in exposure to TB infection and the breakdown rates from infection to disease between males and females. Health education campaigns should be intensified. Instead of only passive hospital out-patient screening of suspects for TB, active casefinding methods should be strengthened. Community based TB control programmes should be in~it,iated.