Hypertension: Development of a prediction model to adjust self-reported hypertension prevalence at the community level

Table 3 Sensitivity, specificity and agreement indicators for each determinant investigated (age, gender, marital status, educational level, income level)

	Sensitivity	Specificity	Accuracy¹	Kappa
	P(CH = 1\|SR = 1)	P(CH = 0\|SR = 0)	(overall agreement)
Full sample	0.90	0.80	0.83	0.66
Age
25-34	0.71	0.88	0.85	0.64
35-49	0.86	0.81	0.82	0.57
50+	0.95	0.69	0.83	0.83
Gender
Females	0.88	0.83	0.85	0.66
Males	0.9	0.8	0.8	0.59
Marital Status
Currently Married	0.90	0.82	0.84	0.65
Not Married	0.89	0.80	0.83	0.65
Race/ethnicity
Latino	0.89	0.85	0.86	0.65
non-Hispanic White	0.90	0.77	0.81	0.65
non-Hispanic Black	0.90	0.79	0.83	0.66
Education
<12	0.92	0.79	0.83	0.65
12	0.79	0.82	0.81	0.65
12+	0.86	0.82	0.83	0.66
Annual Household Income
<$10,000	0.77	0.83	0.81	0.66
$10,000-19,999	0.92	0.83	0.86	0.65
$20,000-34,999	0.89	0.79	0.81	0.65
$35,000+	0.97	0.77	0.83	0.66

Accuracy is the proportion of true results (both true positive and true negative). Refer to Additional file 1.

ISSN: 1472-6963