Skip to main content
. 2020 Dec 23;39(3):331–343. doi: 10.1007/s40273-020-00987-3

Table 1.

Multivariate-adjusted QALYs, costs (US$, year 2016 values) and cost-effectiveness (NAMSAL ANRS 12313 trial, n = 575)

EFV400-based regimen (n = 300) DTG-based regimen (n = 275) Differences (DTG vs EFV400) NHB Prob(DTG:CE)
Base-case analysis
 Total cost per patient 982.8 (899.8–1065.7) 955.0 (872.0–1037.9) −27.8 (−72.1 to 16.6) 0.056 (−0.037 to 0.153) 88%
 QALYs per patient 1.367 (1.248–1.486) 1.368 (1.249–1.486) 0.000 (−0.034 to 0.035)
Alternative scenarios
 Discount rate of 0%
  Total cost per patient 1016.8 (929.3–1104.3) 988.0 (900.5–1075.5) −28.8 (−74.4 to 16.9) 0.059 (−0.037 to 0.160) 89%
  QALYs per patient 1.436 (1.309–1.562) 1.437 (1.311–1.564) 0.002 (−0.034 to 0.038)
 Discount rate of 6%
  Total cost per patient 951.4 (872.5–1030.3) 924.6 (845.7–1003.5) −26.8 (−70.0 to 16.4) 0.053 (−0.037 to 0.148) 87%
  QALYs per patient 1.304 (1.192–1.416) 1.304 (1.192–1.416) −0.001 (−0.033 to 0.032)
 Intention-to-treat population (n = 613)
  Total cost per patient 984.4 (903.2–1065.6) 954.2 (873.0–1035.4) −30.2 (−72.0 to 11.7) 0.055 (−0.033 to 0.147) 89%
  QALYs per patient 1.368 (1.256–1.481) 1.363 (1.250–1.475) −0.005 (−0.038 to 0.028)
 Patients with > 100,000 copies/mL HIV RNA load only (n = 379)
  Total cost per patient 1005.8 (942.6–1,069.0) 951.7 (888.5–1014.9) −54.1 (−112.7 to 4.5) 0.092 (−0.031 to 0.221) 92%
  QALYs per patient 1.370 (1.238–1.502) 1.354 (1.222–1.486) −0.017 (−0.062 to 0.030)
Markov cohort simulation model
 Over 5 years
  Total cost per patient 2140.4 (2106.3–2171.6) 2041.3 (2010.2–2076.3) −99.1 (−142.3 to −51.8) 0.194 (0.123–0.261) 100%
  QALYs per patient 3.440 (3.397–3.482) 3.436 (3.395–3.474) −0.005 (−0.061 to 0.052)
 Over 10 years
  Total cost per patient 3831.1 (3710.3–3929.7) 3669.6 (3546.6–3764.9) −161.5 (−314.6 to −5.8) 0.484 (0.341–0.622) 100%
  QALYs per patient 6.270 (6.076–6.438) 6.431 (6.212–6.576) 0.161 (−0.097 to 0.404)

Data are presented as mean values (95% confidence interval). Covariates for the multivariate-adjusted models were gender, age, CD4 count, HIV RNA level, and utilities at baseline. 95% confidence intervals for the costs and QALYs differences, the NHB, and the Prob(DTG:CE) were computed based on 5000 bootstrap replicates. Cost-effectiveness threshold: $500 per QALY gained

DTG dolutegravir, EFV400 efavirenz 400 mg, HIV human immunodeficiency virus, NAMSAL New Antiretroviral and Monitoring Strategies in HIV-infected Adults in Low-Income Countries, NHB net health benefit, Prob(DTG:CE) probability of DTG being cost-effective at the $500 threshold, QALY quality-adjusted life-year