Methods to increase reporting of childhood sexual abuse in surveys: the sensitivity and specificity of face-to-face interviews versus a sealed envelope method in Ugandan primary school children

Table 4 Sensitivity and specificity of the FTFI method compared to the SEM method for identifying cases of forced sex

	SEM^b
FTFI^a	Affirmative		Negative		Total
Affirmative	35	(0.9%)	7	(0.2%)	42	(1.1%)
Negative	233	(6.1%)	3568	(92.8%)	3801	(98.9%)
Total	268	(7.0%)	3575	(93.0%)	3,843
Sensitivity and Specificity Analysis:
			95% CI
Prevalence		7.0%	(6.2%, 7.8%)
Sensitivity		13.1%	(9.3%, 17.7%)
Specificity		99.8%	(99.6%, 99.9%)
Positive Likelihood Ratio		66.7	(29.9, 149.0)
Negative Likelihood Ratio		0.87	(0.83, 0.91)

^a FTFI face-to-face interview method
^b SEM sealed envelope method, assumed ‘gold standard’ mode of data collection

ISSN: 1472-698X