Page 144 - Timeliness of Infectious Disease Notification & Response Systems - Corien Swaan

142 Chapter 6
Figure 3. Box plot distribution of serial intervals. The boxplox show Intervals I (delay flight arrival and date of diagnosis), interval II (delay date of diagnosis and CT request) and interval III (date of CT request and contact details identified), representing median, 25%-75% Quartile (IQR, in boxes), values between 1.5 IQR (lines) and outliers (* and ° with flight numbers)
For the 17 completed contact investigations, interval I was the largest interval in the contact tracing procedure (mean 2.6 days, range 1–6, 95% CI 1.6- 3.6, n = 13). The other intervals II and III were shorter, with a mean of 0.8 days and 0.6 days respectively, see Table 1. Figure 3 shows the medians of the described intervals. Since 15/17 index cases were already ill before, or during the day of arrival of the flight, the delay in interval I is mainly caused by delay in seeking medical advice and diagnostic procedure itself. After acceptance of the request for CT by the CIb, GGD Kennemerland needed on average 0.6 days (range 0–2, 95% CI 0.3–0.9 days) to collect the passenger list from the airlines and complete contact details (interval III). The total delay between flight arrival and identifica- tion of contact details was on average 3.9 days (range 2–7 days, 95% confidence interval 3.2-4.7 days), see Table 1. In only 3 out of 17 contact investigations (18%), contacts were identified within 2 days after arrival. In 2 out of these 3 contact investigations, PLC’s were available. Interval III of the 5 CT with PLC’s available was shorter (0.4 days, SD 0.5) than for 12 CT’s without PLC (0.7 days, SD 0.7), this was not significant however (p: 0.25). Overall delay in CT with PLC’s