Airport sentinel surveillance and entry quarantine for dengue infections following a fever screening program in Taiwan

Table 1 The airport screening program mediated by noncontact infrared thermometers (NCITs) followed by laboratory confirmation, 2007–2010: sensitivity, specificity and predictive values

Year	2007	2008	2009	2010
Total inbound passengers (sample size, n)	12,508,621	12,202,392	12,499,365	14,837,391
Inbound passengers from Southeast Asia	1,164,291	471,938	1,486,357	2,729,618
Confirmed febrile passengers¹	11,118	12,158	12,286	12,553
Fever prevalence among total passengers %	0.09	0.10	0.10	0.08
Specimens taken for diagnostics²	5,654	4,946	3,706	3,956
Dengue importations detected in the airport fever screening program ³	72	100	108	126
Dengue importations detected in harbor screenings⁴	0	1	2	3
Dengue importations reported from communities	107	125	95	175
Sensitivity of the airport fever screening program %	40.22	44.44	53.20	41.86
Specificity of the airport fever screening program %	99.96	99.96	99.97	99.97
PPV⁵ of the airport fever screening program %	1.28	2.03	2.90	3.22
NPV⁶ of the airport fever screening program %	>99.99	>99.99	>99.99	>99.99

1. Confirmed febrile passengers were detected by NCITs (37.5°C) followed by ear thermometers (more than 38°C) in 2 airports.
2. Confirmed febrile inbound passengers from dengue-endemic countries, mainly Southeast Asia and South Asia.
3. The cumulative results of the International Taoyuan Airport and the International Kaohsiung Airport.
4. Dengue importations reported from fever screening in 9 harbors with 10 NCITs were not enrolled as true positives in calculating the PPV and NPV of airport fever screening.
5. PPV: positive predictive value of the airport fever screening program.
6. NPV: negative predictive value of the airport fever screening program.

ISSN: 1471-2334