Inequalities in Access to Tuberculosis Services in South Africa: Does Gender Matter?


 Background: Tuberculosis (TB) remains the leading cause of death in South Africa. As an infectious respiratory disease, control of TB includes antibiotic treatment over a number of months in order to cure the disease and reduce transmission to others. The need to adhere to treatment over an extended period highlights the importance of reducing access barriers to TB services. While gender related access barriers have been identified as critical in the treatment and control of other conditions such as Human Immunodeficiency Virus (HIV), no quantitative analyses currently exist that assess the gender-related dimensions of access to TB services in South Africa. Methods: This study aims to assess the gender-based differences in access to TB services in South Africa, from the perspective of TB patients. Using a comprehensive framework where access is defined as the opportunity to use services, we interviewed 1,229 TB patients using services in four provinces of South Africa. Comparisons of access barriers and adherence between men and women were examined using multivariate linear and logistic regressions.Results: There was no significant association between levels of adherence and gender (all p-values>0.05). Among availability-related variables, men spent significantly less time at the clinic to fetch TB medication compared with women (coefficient, -7.06; 95% CI, [-13.5, -0.7]). Regarding affordability, men were significantly less likely to receive a disability grant (AOR, 0.48; 95% CI, [0.36, 0.63]). Concerning the acceptability dimension, men were less likely to report that queues to visit a healthcare provider were too long or the cleanliness of the facility to be sub-standard (AOR, 0.69; 95% CI, [0.52, 0.91], and AOR, 0.67; 95% CI, [0.46, 0.97], respectively). Conclusions: Our findings indicate that there is no association between the level of adherence to TB treatment and gender. Moreover, there was no evidence of systematic gender-based inequalities in access to TB services. However, the findings reveal concerns about the condition and cleanliness of health facilities that may impact the patients' adherence and be a barrier, specifically, in women's use of TB services.


100
Catastrophic healthcare expenditure was computed as expenditure on healthcare exceeding 10% 101 of per capita household expenditure. Finally, for the dimension of acceptability (or cultural access), participants were asked to report their perceptions of staff attitudes (proxied by questions regarding whether they felt respected by staff and whether the staff were too busy to answer their 104 questions), the length of queues, cleanliness of waiting areas and toilets, and stigma (proxied by whether the participant felt that people in the community judge them negatively for attending the different for individuals with distinct levels of social advantage, including with respect to their

120
In each sub-district, two-stage sampling was used: firstly, we selected a representative sample of 121 public primary healthcare facilities, and secondly we selected a representative sample of TB users 122 within chosen facilities [20]. Since most public primary healthcare facilities in the country provide 123 TB services, a minimum of five facilities was selected in each sub-district using the probability 124 proportional to size (PPS) method, based upon the total number of TB users in each facility.

125
Thereafter, a random sample of patients was interviewed until the proposed facility sample size 126 was met. Respondents were eligible if they were over the age of 18, were judged by clinical staff 127 to be sufficiently well to be interviewed and had been on TB treatment for at least eight weeks.

128
After obtaining informed consent from participants in the study, the interview was conducted by   Whitney U test (for medians) and Student's t-test (for means). A p-value less than 0.05 was 145 accepted as statistically significant. Multivariate logistic and linear regressions were run in order 146 to test for differences in access barriers by gender after controlling for age, socioeconomic status 147 (asset index), level of education, employment status and sub-district setting. In this way, we focus 148 on gender-related inequalities in access while holding the other measures of social 149 (dis)advantage constant. study, approximately half were living in an urban setting. As is shown, women reported a higher 160 number of years in school and more women reported receiving the government's disability 161 (chronic care) grant. In contrast, more men reported that they were employed, as well as married 162 or living with a partner. There was no significant difference in the asset index, with men and 163 women having similar asset-based socioeconomic status. Since the study sample is based on 164 patients who attended primary healthcare facilities and who were well enough to participate, those 165 with more severe illness were less likely to be included. As such, the majority of participants 166 reported that this was the first time they were treated for TB. Concerning the type of DOTS, nearly 167 one-third of respondents reported daily observed therapy at clinics (clinic DOTS) with the 168 remainder reporting weekly or even monthly clinic attendance. Self-reported adherence was high, 169 with 82% reporting that they had never missed TB visits or doses of medication, including 81.6% 170 of men and 82.4% of women.

299
Further approval for the analysis presented in this paper was obtained from the University of Cape

304
The data that support the findings of this study are available from Centre of Health Policy, School

305
of Public Health at University of Witwatersrand as REACH Database, but restrictions apply to the 306 availability of these data, which were used under license for the current study, and so are not 307 publicly available. Data are however available and the request could be addressed to Dr. Susan 308 Cleary who is one of co-investigators of REACH Project and the co-author of this research article.

313
The REACH project was carried out with support from the Global Health Research Initiative, a  Agency of Canada, Health Canada. We would like to thank the REACH team, the patients, as 329 well as the health workers who agreed to be involved in this project.
opportunities for achieving universal access to diagnosis and effective treatment. South African