Table 1.
Data for model input parameters.
| Variable | Estimate | Reference |
|---|---|---|
| Initial cohort | ||
| CD4 cell count [cells/μl (SD)] | 331 (268) | [20] |
| Initial HIV RNA (copies/ml) distribution (%) | [21] | |
| > 100 000 | 69.21 | |
| 30 001–100 000 | 22.27 | |
| 3001–10 000 | 7.06 | |
| 501–3000 | 1.35 | |
| 0–500 | 0.11 | |
| Mean monthly CD4 cell decline (cells/μl) by HIV RNA strata (copies/ml) | [22] | |
| > 30 000 | 6.375 | |
| 10 001–30 000 | 5.400 | |
| 3001–10 000 | 4.600 | |
| 501–3000 | 3.733 | |
| ≤ 500 | 3.025 | |
| Percentage monthly risk of severe opportunistic infectionsa | [23] | |
| Bacterial | 1.08–3.36 | |
| Fungal | 0.00–0.50 | |
| Malaria | 0.00–0.50 | |
| Tuberculosis | 0.50–1.16 | |
| Isosporiasis | 0.17–0.75 | |
| Toxoplasmosis | 0.00–0.50 | |
| Non-tuberculosis mycobacteriosis | 0.00–2.06 | |
| Other | 1.24–7.07 | |
| Percentage monthly risk of mild opportunistic infectionsa | [23] | |
| Bacterial | 1.57–8.23 | |
| Fungal | 1.08–3.68 | |
| Other mild infections | 1.16–3.52 | |
| Rates of virological suppression | ||
| Two sequential regimens of ART available (% suppressed at 48 weeks) | ||
| first line, 2NRTI + NNRTI; second line, 2NRTI + PI | ||
| NNRTI + 2NRTI (no resistance) | 75 | [21,24–29] |
| NNRTI + 2NRTI (with resistance) | 38 | [30] |
| PI + 2NRTI (no resistance) | 58 | [32] |
| first line, 2NRTI + PI; second line, 2NRTI + NNRTI | ||
| PI + 2NRTI (no resistance) | 75 | [31] |
| NNRTI + 2NRTI (no resistance) | 30 | [33] |
| NNRTI + 2NRTI (with resistance) | 10 | Assumption |
| One regimen of ART available | ||
| NNRTI + 2NRTI (no resistance) | 75 | [21,24–29] |
| NNRTI + 2NRTI (with resistance) | 38 | [30] |
| PI + 2NRTI (no resistance) | 75 | [31] |
| Efficacy of co-trimoxazole (% reduction in probability of infection) | [6,20] | |
| Mild bacterial infection | 48.4 | |
| Severe bacterial infection | 49.8 | |
| Severe malaria | 88.4 | |
| Isosporiasis | 81.8 | |
| Toxoplasmic encephalitis | 83.2 | |
| Acute unexplained fever | 59.1 | |
| Mild fungal infectionsb | −46.4 | |
| Annual costs of treatment (2005 US$) | ||
| 2NRTI + NNRTI | 427 | [34] |
| 2NRTI + PI | 580 | [34] |
| Co-trimoxazole prophylaxis | 17 | [6] |
| Test costs ($US) | [35] | |
| CD4 cell count (per test) | 5 | |
| Monthly costs of chronic care (2005 US$) for CD4 cell count strata (cells/μl) | [6,11] | |
| > 200 | 21.96 | |
| ≥ 200 | 16.04 | |
| Terminal care cost | 38.38 | |
| Costs of specific components of care (2005 US$) | [6] | |
| Clinic visit | 5.79 | |
| Hospital day | 17.23 | |
| Day care | 8.55 | |
ART, antiretroviral therapy; NRTI, nucleoside reverse transcriptase inhibitor; NNRTI, non-nucleoside reverse transcriptase inhibitor; PI, protease inhibitor.
a
Risk is stratified by CD4 cell count; in nearly all cases monthly risk increases with decreasing CD4 cell count. The following five severe opportunistic infections are considered late stage and are used to trigger initiation of ART: severe fungal infection, isosporiasis, cerebral toxoplasmosis, non-tuberculous mycobacteriosis, or other severe illness.
b
The risk of mild fungal infections is increased by 46.4% in the presence of co-trimoxazole [6].