DETERMINANTS OF TREATMENT ADHERENCE AMONG TUBERCULOSIS-INFECTED HIV PATIENTS IN CROSS RIVER STATE, NIGERIA

Abstract

Concurrent tuberculosis and HIV treatment is a standard practice in co-infected patients. However, adherence to combined treatment is challenging because of multiplicity of drugs involved. Although studies exist on adherence to either HIV or tuberculosis treatment, negligible attention has been paid to adherence among patients on combined HIV and tuberculosis treatments. This study therefore examined the factors influencing treatment adherence among tuberculosis-infected HIV patients in Cross River State (CRS), where prevalence is higher than the national average. Parsonian Social Action theory, Social Cognitive theory and Health Belief Model were employed as theoretical framework. The study adopted a cross-sectional design using 333 patients. A three-stage purposive sampling technique was used to select respondents. This involved identification of treatment facilities across CRS, selection of facilities that had up to 10 co-infected patients and selection of patients who had received concurrent treatment for three months or more prior to the study. A semi-structured questionnaire was used to generate data. Adherence was assessed with a 14-item scale categorised as low (>7), moderate (2-7) and high (0-1). Knowledge was measured using a 17-item instrument categorised as low (≤8) and high (>8). Four Focus Group Discussion sessions comprising seven discussants each were conducted and four case studies were undertaken with patients. Quantitative data were analysed using descriptive statistics, Chi-square and ordinal regression test at 0.05 level of significance while qualitative data were content analysed. The mean age of respondents was 34.5±9.6 and 61.9% were female. Fifty-seven percent of respondents had high knowledge of treatment and 48.7% did not link poor adherence to poor treatment outcomes. Level of adherence was high (38.1%), moderate (29.4%) and low (32.4%). Adherence to tuberculosis treatment was significantly higher than to HIV treatment. Respondents’ reasons for missing drugs included not being at home (64.7%), not having eaten (45.5%), being busy (44.9%) and avoiding status disclosure (25.1%). Having good knowledge of treatment was significantly related to low level of adherence. The likelihood of adherence was significantly high among males (OR: 1.8; 95% CI: 0.4-2.4), those with a minimum of secondary education (OR: 2.7; 95% CI: 1.2-3.4) and those not living in the same community as the location of their treatment facility (OR: 1.7; 95% CI: 1.0-3.5). Patients who received adequate social support showed the likelihood of better adherence relative to those who received little or no support (OR: 3.0; 95% CI: 1.3-4.7). Patients reported that when in the midst of other people, they did not want to be seen using drugs to forestall stigmatisation. Respondents demonstrated enthusiasm at the start of the treatment but adherence reduced when difficulties were encountered. Respondents whose spouses or regular sexual partners were not infected encountered more difficulties with adherence because they were believed to be under spiritual attacks not HIV. Patients benefitted much from counselling and good care-provider/patient relationship. Treatment adherence among tuberculosis-infected HIV patients was influenced by personal characteristics and health facility location. Training on how to overcome the stigma, initiation of patient-selected treatment facility options and policies that emphasise sustained patient counselling could improve adherence.