Table 4.
Sensitivity analysis for variation in model parameters.
| Range of variationa |
Robustness to joint variationb | |||
|---|---|---|---|---|
| Multiplication rate, M (/asexual cycle) | Elimination constant, k (/asexual cycle) | Rate of PfHRP2 secretion, R (×10−15 pg/cycle) | ||
| Initial model value | 7.5 | 1.26 | 5.2 | |
| UM vs. SM | 51–149% (3.8–11.2) | 0–170% (0–2.14) | 75–180% (3.9–9.36) | 64% |
| UM vs. SNP | 25–116% (1.88–8.7) | 0–126% (0–1.59) | 89–365% (4.63–18.98) | 47% |
| UM vs. all LA | 35–136% (2.6–10.2) | 0–153% (0–1.93) | 80–256% (4.16–13.31) | 60% |
| Correlation between lactate and sequestered biomassc | 58–137% (4.34–10.3) − r + |
18–155% (0.23–1.95) − r + |
80–158% (3.71–8.22) + r − |
63% |
UM, uncomplicated malaria; SM, severe malaria; SNP, severe non-prostrated; LA, hyperlactataemia.
a
Range of variation (absolute values) within which a likelihood-ratio test does not reject the null hypothesis that the median sequestered-parasite biomass in the comparison groups is equal.
b
Robustness (% of simulations in which the null hypothesis is not rejected) to joint variation of the parameters within their individually determined ranges of variation.
c
Using Spearmann's rank correlation test, the sign of the correlation coefficient, r, is presented for the limits of the parameter range.