Cost and cost-effectiveness of tuberculosis treatment shortening: a model-based analysis

Gomez, G. B.; Dowdy, D. W.; Bastos, M. L.; Zwerling, A.; Sweeney, S.; Foster, N.; Trajman, A.; Islam, M. A.; Kapiga, S.; Sinanovic, E.; Knight, G. M.; White, R. G.; Wells, W. A.; Cobelens, F. G.; Vassall, A.

doi:10.1186/s12879-016-2064-3

BMC Infectious Diseases

Table 1 Parameters

From: Cost and cost-effectiveness of tuberculosis treatment shortening: a model-based analysis

Parameter, value [range]	South Africa	Brazil	Bangladesh	Tanzania	Reference
Population distribution
Smear-positivity: HIV-negative, HIV-positive no ART, HIV-positive ART	0.69, 0.35, 0.45				[29] [30]
MDR prevalence, among new patients	1.8%	1.4%	1.4%	1.1%	[1, 31, 32]
Prevalence of HIV in TB patients	62%	17%	3%	37%	[1, 33, 34]
Diagnosis of TB
TB diagnosis, sensitivity, smear, HIV negative	0.72 [0.62–0.82]				[29, 35]
TB diagnosis, sensitivity, smear, HIV positive	0.47 [0.51–0.43]				[29, 35]
TB diagnosis, sensitivity, GeneXpert, smear positive	0.98 [0.97–0.99]				[36]
TB diagnosis, sensitivity, GeneXpert, smear negative	0.68 [0.59–0.75]				[36]
TB diagnosis, specificity, GeneXpert, all	0.98 [0.97–0.99]				[36]
TB diagnosis, specificity, smear, all	1				assumption
RIF-resistance diagnosis, sensitivity, GeneXpert, all	0.94 [0.87–0.97]				[36]
RIF-resistance diagnosis, specificity, GeneXpert, all	0.98 [0.97–0.99]				[36]
Outcomes (first line treatment)
Pr mortality:
Pan-sensitive, HIV negative	0.03 [0.02–0.03]				[37]
MDR, HIV negative	0.11 [0.08–0.13]				[37]
Pan-sensitive, HIV positive, ART	0.07 [0.05–0.09]				[38]
Pan-sensitive, HIV positive, no ART	0.33 [0.30–0.43]				[38]
MDR, HIV positive, ART	0.11 [0.10–0.21]				[38, 39]
MDR, HIV positive, no ART	0.85 [0.72–0.98]				[39]
Pr cure, if treatment completed:
Pan-sensitive	0.97 [0.95–0.98]				[13]
MDR	0.50 [0.40–0.55]				[37]
Pr cure, if less than 2 months treatment completed
HIV negative, smear negative	0.20 [0.15–0.25]				[40–42]
HIV negative, smear positive	0.30 [0.20–0.40]				[40–42]
HIV positive, smear neg/pos, no ART	0 [0–0.05]				[40–42]
HIV positive, smear negative, ART	0.10 [0.05–0.15]				[40–42]
HIV positive, smear positive, ART	0.05 [0–0.10]				[40–42]
Pr cure, if default at (standard 6mo regimen):
2–3 months, pan-sensitive	0.68 [0.50–0.80]				[43]
2–3 months, MDR	0.35 [0.21–0.45]				[43]
4–5 months, pan-sensitive	0.86 [0.70–0.89]				[44, 45]
4–5 months, MDR	0.48 [0.29–0.51]				[44, 45]
Pr cure, if default at (new 4mo regimen):
2–3 months, pan-sensitive	0.74 [0.57–0.83]				Assumption
2–3 months, MDR	0.38 [0.23–0.47]				Assumption
Outcomes (second round of treatment)
Pr patients returning to care after default	0.21 [0.10–0.70]				[46]
Pr patients staying in care after failure	0.60 [0.40–0.80]				Assumption
Pr mortality
HIV negative/positive ART, pan-sensitive	0.06 [0.04–0.07]				[37]
HIV negative/positive ART, MDR	0.15 [0.10–0.20]				[37]
HIV positive no ART, pan-sensitive	0.33 [0.30–0.43]				[38]
HIV positive no ART, MDR	0.85 [0.72–0.98]				[39]
MDR treatment and long term outcomes
Pr cure, MDR treatment (including default)	0.65–0.80				[47]
Mortality during MDR treatment	0.10				[48]
Pr long term mortality (chronic TB patient or default patient if no return to care)	0.75 [0.50–0.99]				[48]
Self-cure among chronic TB patients	0.01				[48]
DALYs averted (discounted at 0.03/year)
HIV negative, smear negative	12.5 [11.3–13.8]	19.3 [17.4–21.3]	15.3 [13.7–16.8]	14.2 [12.8–15.7]	Additional file 1
HIV negative, smear positive	15.2 [13.7–16.7]	22.0 [19.8–24.2]	17.9 [16.1–19.7]	16.9 [15.2–18.6]	Additional file 1
HIV positive, smear negative, no ART	1.8 [1.6–2.0]	1.8 [1.6–2.0]	1.8 [1.6–2.0]	1.8 [1.6–2.0]	Additional file 1
HIV positive, smear positive, no ART	2.0 [1.8–2.2]	2.0 [1.8–2.2]	2.0 [1.8–2.2]	2.0 [1.8–2.2]	Additional file 1
HIV positive, smear negative, ART	9.9 [8.9–10.9]	9.9 [8.9–10.9]	9.9 [8.9–10.9]	9.9 [8.9–10.9]	Additional file 1
HIV positive, smear positive, ART	10.1 [9.1–11.1]	10.1 [9.1–11.1]	10.1 [9.1–11.1]	10.1 [9.1–11.1]	Additional file 1

HIV human immunodeficiency virus, TB tuberculosis, ART antiretroviral treatment, MDR multidrug resistant, Pr probability
Parameters with one value are included in the model as point estimates; parameters with a value and a range were included in the model as triangular distributions; parameters with only a range of two values were included in the model as a uniform distribution

Back to article page

ISSN: 1471-2334

Contact us

Submission enquiries: bmcinfectiousdiseases@biomedcentral.com
General enquiries: ORSupport@springernature.com