Timing of progression from Chlamydia trachomatisinfection to pelvic inflammatory disease: a mathematical modelling study

Table 1 Parameter values describing the natural history of chlamydia infection, PID development and the screening intervention

Parameters	Baseline values	Explanation	Sensitivity analysis			Source
			Distribution	Parameters
Model parameters
λ		Force of infection (per day), calculated using^*
1/r	365	Mean duration of infection (days) [6, 16]	N(μ,σ²)	μ=365	σ²=75²	Consensus
p	5.7%	Prevalence at baseline [11]	Bin(n,p)	n=2519	$p = \frac{143}{2519}$	[11]
α		Effective testing rate (per day), calculated using^†
c	22.2%	Coverage of testing uptake (per year) [11]	Bin(n,p)	n=2377	$p = \frac{527}{2377}$	[11]
δ	8.0%	Treatment failure [17]	U(a,b)	a=0%	b=50%	Consensus
f	estimated	Fraction of women becoming infected with chlamydia who will develop PID
Input parameter
x	30.0%	Proportion of PID cases due to chlamydia in control group [11]	Bin(n,p)	n=23	$p = \frac{7}{23}$	[11]

^*In the absence of the tria (α=0), to observe chlamydia prevalence p at steady state: $λ = \frac{p}{1 - p} r$ .
^†Reported uptake of chlamydia testing c during the follow-up period (outside of the trial) is reduced by the proportion with treatment failure δ, which results in the effective testing rate $α = \frac{- l n (1 - (1 - δ) c)}{365}$ per day [18].
N(μ,σ²), normal distribution (mean, variance); Bin(n,p), binomial distribution (size, probability); U(a,b), uniform distribution (minimum, maximum).

ISSN: 1471-2334