4.4.1 Costs
Consistent with the current policy on registration costs, most participants either reported paying 2 pula/P2 (about 40 US cents) or did not pay anything. The P2 charge is the normal administrative fee payable by anyone who visits a health facility. Twenty-nine per cent of the participants indicated that they had experienced a loss of income as a result of coming to the clinic and 57% indicated changes in general expenditures. However, there was no significant association between employment status and reported loss of income (χ2=1.526; p=0.217). The median cost of travelling to the facility was P10 (approx US$ 2.00) and 80% of the participants reported spending less than P15 (approx US$ 2.50) for transport. The mean cost of transport was not significantly different between the optimally adherent and sub-optimally adherent (t=0.0208; P=0.978).
4.4.2 Gender and employment
Using the visual analogue method, no association was observed between gender and adherence (χ2=0.743; P=0.389). However, there is a significant association between employment status and adherence (χ2=5.116; P=0.024), suggesting that people who are employed are more likely to adhere to treatment. A higher proportion of the employed (65%) had optimal levels of adherence compared to 55% among the unemployed.

Figure 4.2: Adherence rates (measured by visual analogue) by sex and occupation
4.4.3 Knowledge of HIV and ARVs
The knowledge about HIV and ARVs was rated using eight questions worth one point each. Fifty-eight per cent of the participants got a score of at least 75%. For the purpose of evaluating the impact of knowledge on adherence, a cut-off of 75% was used (>75% good knowledge). A significant correlation was observed between knowledge of HIV and ARVs and adherence level (optimal and sub-optimal) for pill count (χ2=13.558; P<0.0001) and visual analogue (χ2=3.890; 0.049). However, there was no correlation between knowledge of HIV and ARVs and adherence levels in the two-day recall measure (χ2=1.127; P=0.288).
Table 4.4: Association between knowledge of HIV and ARVs with level of adherence and measures used
Measure |
Level of Adherence |
Good |
Poor |
χ2 value |
P-value |
Pill count* (N = 322) |
Sub-optimal |
69 |
11 |
13.558 |
<0.0001 |
| |
Optimal |
156 |
86 |
|
|
Two-day recall (N=496) |
Sub-optimal |
9 |
13 |
1.127 |
0.288 |
| |
Optimal |
255 |
231 |
|
|
Visual analogue (N = 508) |
Sub-optimal |
95 |
106 |
3.890 |
0.049 |
| |
Optimal |
166 |
129 |
|
|
* Serowe data excluded from the analysis.
4.4.4 Education
There was no association between educational levels and adherence rates (χ2 =3.44; p=0.751).
4.4.5 Quality of health care services
The majority of the participants (92%) were satisfied with the quality of health care services.
4.4.6 Side-effects
Of the 58% of participants who reported having experienced side-effects, 8% reported having skipped their medication as a result.
4.4.7 Treatment supporters/reminders and appointments
Most of the participants (74%) said they had someone to remind them to take their medication. Twenty per cent reported having missed some appointments.