Skip to main content
. 2017 Nov 2;217(1):112–121. doi: 10.1093/infdis/jix555

Table 1.

Parameter Values Used to Define Upper and Lower Bounds of Sampling Ranges

Parameter Sampled Range References Notes
Death threshold
(log10 units)
7.0–10.0a [37–39] Burdens in animal models rarely measure greater than 109 units per lung.
Width of the window between symptom and death thresholds
(log10 units)b
4.0–7.0a [39–41] Burdens in animals with asymptomatic infection may be as high as 103 units.
Progression → Recovery transition rate 0.001–0.35 per weeka,c Derived Assume 5% transition per year, and no more than 75% transition per month.
Recovery → Progression transition rate 10−5–10−4 per weeka,d [42] Probability of reactivation in latent infections estimated to be 0.8% per year.
Mode of the progression phase growth rate 0.035–0.714 per weeke,f [39–41] Assume patients progress from onset to death in 3–60 months, assuming 104 unit symptom window.
Mode of the recovery phase growth rate (−0.462)−0.035 per weeke,g [43] Assume patients self-resolve at 1/4 the rate of chemotherapeutic recovery (as defined by time to sputum conversion).

aRange sampled uniformly on the log10 scale.

bThe symptom threshold in a given simulation is derived by value of the death threshold and the value of the symptom window width.

cA 0.35 weekly rate of transition is equivalent to a 75% monthly probability of transition.

dRates in this range are equivalent to 0.05%–5.0% probabilities of transition.

eRange sampled uniformly on the log-modulus scale.

fProgression rates in this range are equivalent to net population doubling times in the range of 6–137 days.

gProgression rates in this range are equivalent to the range of a population half-life of 11 days to a population doubling time of 137 days.