Evaluation of Factors that Affect Antiretroviral Adherence in Siavonga District, Zambia
MetadataShow full item record
Zambia is among the countries with high prevalence of HIV in the world, estimated at 16% among the adult population (ZDHS 2002). Noting the importance of adherence in HIV/AIDS treatment a case controlled study was conducted to understand factors that make people on Antiretroviral drugs (ARVs) not to adhere to treatment in two health facilities in Siavonga district. The study was conducted between May 2007 and May 2008. Both qualitative and quantitative research methods were used. These included the use of semi structured questionnaires and interviews with patients receiving ARV treatment and health care providers. Data coding, checking and cleaning was done before entry into the computer statistical package, EPI-INFO version 6.04 and SPSS. Qualitative data proportions comparisons were checked using the chi-square and a p-value of less than 0.05 was used to determine significance. The non adherent patients were identified by pill count, pill identification test and lack of improvement in clinical parameters. The clinical parameters used in this case were improvement in haematological cell lines including CD4 count. The average defaulter rate in Siavonga District was estimated at around 11.4% compared with provincial rate of 12.6%.Socioeconomic factors affecting adherence were not examined in this study because several studies which have been done on this have shown consistently that there is no association (Fredrick Mulenga 2003). The study concentrated on patient's factors affecting adherence. The study found an association between the physical state of the patient and adherence. Respondents who were symptomatic (Diarrhoea) at initiation of ARVs were 3.7 time more likely to be adhering to treatment as asymptomatic patients (p value 0.001). Another association was found between the number of pills taken and adherence. Non adherents were 95% more likely to be on Septrin prophylaxis as adherent patients (p value 0.001). The more pills the patient takes the less likely that they will adhere to treatment. The study found no association between adherence and the type of regimen the patient is on, the level of education, and side effects. On the type of the regime the study found that there was no significant difference in terms of adherence among the different types of ARVs combination (p value 0.024).On side effects and adherence, the results were not statistically significant (p value 0.3). An association was found between the distance to the health facility and adherence. Adherent patients were two times more likely to be staying near the clinic as non-adherent patients (odds ratio 2). Even though pill burden and physical state of the patient have been shown to affect adherence they can not be used on their own to predict patient's adherence or non-adherence.