Table 1 Inputs in computer simulation
Variable | Base case | Plausible range in sensitivity analysis | Source |
|---|---|---|---|
Characteristics of simulated cohort | Â | Â | Â |
Age | 39 (SD 9) | NA | AMPATH |
CD4 count (cells/mm3) | 126 (SD 127) | NA | AMPATH |
Viral load (Log 10 units) | 4.5 (SD 1) | NA | AMPATH |
% Male | 38% | NA | AMPATH |
Initial cART regimen | Nevirapine + either ziduvidine or stavudine + other NRTI | NA | AMPATH |
Second cART regimen | Boosted PI + two other NRTIs other than those in initial regimen | NA | AMPATH |
Probabilities and rates | Â | Â | Â |
Compliance with cART (proportion of doses taken as directed) | 0.85 | 0.75-0.95 | Imputed from calibration |
Probability that mutation potentially causing resistance, results in resistance, NRTI or PI | 0.50 | Varied jointly from 0.5X to 1.5X, bounded by 0 and 1 | Johnson et al [21] |
Probability mutation potentially causing resistance, results in resistance, NNRTI | 0.90 | Varied jointly from 0.5X to 1.5X, bounded by 0 and 1 | Johnson et al [21] |
Probability of cross-resistance to other NRTI, given NRTI mutation conferring resistance (ziduvidine or stavudine) | 1.0 | Varied jointly from 0.5X to 1.5X, bounded by 0 and 1 | Johnson et al [21] |
Probability of cross-resistance to other NRTI, given NRTI mutation conferring resistance (other) | 0.48 | Varied jointly from 0.5X to 1.5X, bounded by 0 and 1 | Johnson et al [21] |
Probability of cross resistance to other PI, given PI mutation causing resistance | 0.24 | Varied jointly from 0.5X to 1.5X, bounded by 0 and 1 | Johnson et al [21] |
Probability of cross resistance to other NNRTI, given NNRTI mutation causing resistance | 0.88 | Varied jointly from 0.5X to 1.5X, bounded by 0 and 1 | Johnson et al [21] |
Rate of accumulating resistance mutations, per year | 0.18 | 0.014-0.018 | Braithwaite et al [14] |
Viral load decrement with cART consisting of 2 NRTIs + efavirenz (100% adherence) | 3.09 | Varied jointly from -1 to +1 | Braithwaite et al [22] |
Viral load decrement with cART consisting of 2 NRTIs + nevirapine (100% adherence) | 2.22 | Varied jointly from -1 to +1 | Braithwaite et al [22] |
Viral load decrement with cART consisting of boosted PI (100% adherence) | 2.68 | Varied jointly from -1 to +1 | Braithwaite et al [22] |
Augmentation in HIV-related mortality, multiplicative | 1 | Varied from 0.5X to 1.5X | Assumption |
Augmentation in non-HIV-related mortality, multiplicative | 1 | Varied from 0.5X to 1.5X | Assumption |
Utilities | Â | Â | Â |
Decrease in utility with cART | 0.053 | Varied jointly from -0.05 to +0.05 | Braithwaite et al 2007 [23] |
Utility with CD4 < 100 cells/mm3 | 0.81 | Varied jointly from -0.05 to +0.05 | Freedberg et al 1998 [24] |
Utility with CD4 between 100 cells/mm3 and 199 cells/mm3 | 0.87 | Varied jointly from -0.05 to +0.05 | Freedberg et al 1998 [24] |
Utility with CD4 200 cells/mm3 and above | 0.94 | Varied jointly from -0.05 to +0.05 | Freedberg et al 1998 [24] |
Costs (2008 US$) | Â | Â | Â |
Cost of outpatient care, annually, without cART ($/month) | $288 | Varied from 0.5X to 1.5X | AMPATH |
Cost of care per hospitalization | $390 | Varied from 0.5X to 1.5X | AMPATH |
Cost of cART, annually, first regimen | $189 | Varied from 0.5X to 1.5X | AMPATH |
Cost of cART annually, second regimen | $1361 | Varied from 0.5X to 1.5X | AMPATH |
Cost of cART annually, third regimen | $3067 | $1361 - $12,269 | AMPATH, Red Book [25] |
Cost of viral load test | $70.00 | Varied from 0.5X to 1.5X | AMPATH |
Cost of CD4 test | $11.20 | Varied from 0.5X to 1.5X | AMPATH |