. Author manuscript; available in PMC: 2009 Mar 9.

Published in final edited form as: Nat Med. 2004 Sep 19;10(10):1117–1121. doi: 10.1038/nm1110

Table 2.

Model fit and simulation results

a Epidemiology of tuberculosis before drug introduction
Measure^a	Observed	Model results
Proportion of the population with latent infection²³	>90% over age 11 (Vienna, 1909)	82%
	>80% of 13−14-year-olds (Prague)
	44−79% of 18-year-olds (Europe, 1948−1951)
Prevalence of smear positive TB²³	Average 200−300 cases/100,000	1,649/100,000^b
	Range 150−950/100,000 (Africa and Asia, late 1950s)
TB mortality²³	600/100,000 per year (London, 1800)	494/100,000
	650/100,000 per year (Alaska, 1928−1932)
	380/100,000 per year (Czechoslovakia, 1900)
	200/100,000 per year (The Netherlands, 1900)
Proportion of all deaths which are due to TB²³	35% (Alaska, 1930)	20%
	10% (Germany, 1901)
	5% (Germany, 1938)

b Current tuberculosis epidemiology from selected high-burden countries²⁴ and model simulation results
						Model simulations^c
	Cambodia	China	India	Russia	S. Africa	RF=80%	RF=100%	RF=120%
Proportion with latent infection (sensitive and MDR)	66%	31%	46%	22%	39%	65%	65%	65%
Proportion with latent MDR infection						0.2%	0.2%	0.2%
Detected smear positive TB prevalence/100,000	492	82	215	229	93	234^d	235^d	235^d
Prevalence of infectious TB/100,000						464	464	465
Proportion of detected TB that is MDR²⁵	Median for 28 countries and regions = 2.2% (95% CI 1.1−3.8)					2.1%	2.2%	2.4%
Proportion of total TB that is MDR						1.3%	1.4%	1.6%
TB deaths/100,000	131	19	40	165	24	104	104	104

We assume that all infections and disease are due to drug-sensitive strains prior to the introduction of antibiotics.

The model result is for both detected and undetected cases. Additionally, the model results include all infectious TB cases (not only those who are sputum smear-positive).

RF = relative fitness of the ‘fit’ MDR strain compared with the drug-sensitive strain.

Includes all infectious cases (not only those who are sputum smear-positive).