Skip to main content

Table 1 Pertussis cough characteristic trends associated with initial medical visit (week of antibiotic prescription) and simulated findings

From: The impact of healthcare visit timing on reported pertussis cough duration: Selection bias and disease pattern from reported cases in Michigan, USA, 2000–2010

Week of antibiotic prescription

 

Observed Percentage of records

Observed days

Simulated Daysa

Nb

Coughing at final interview % (95 % CI)

Macrolide antibiotic % (95 % CI)

Cough attribute % (95 % CI) c

Cough duration mean (95 % CI)

Mean differenced

Cough duration mean (95 % CI)

Mean differenced

0e

65

80.0 (70.0–90.0)

65.6 (52.7–76.6)

81.5 (71.8–81.2)

28.7 (24.8–32.5)

N/A

N/A

N/A

1st

809

79.7 (76.9–82.5)

79.7 (76.0–81.7)

86.8 (84.4–89.1)

24.8 (23.4–26.1)

−3.9

26.1 (25.6–26.5)

N/A

2nd

1043

84.1 (81.9–86.3)

84.8 (81.9–86.3)

92.0 (90.3–93.6)

25.6 (24.6–26.5)

0.8

26.9 (26.4–27.3)

0.8

3rd

613

86.5 (83.7–89.2)

85.7 (82.0–87.7)

92.0 (89.9–94.2)

30.5 (29.2–31.9)

4.9

31.8 (31.3–32.3)

4.9

4th

341

83.9 (79.9–87.8)

89.6 (85.2–92.0)

92.7 (89.9–95.4)

34.8 (33.5–36.1)

4.3

38.8 (68.5–39.4)

7.0

5thf

494

76.1 (72.3–80.0)

89.2 (85.9–91.5)

92.7 (90.4–95.0)

58.3 (55.1–61.5)

23.5

46.1 (45.4–46.8)

7.3

  1. CI confidence interval, N number
  2. a Simulated cough duration (number of days) was calculated using a hypothetical population distribution and applying the observed Michigan pertussis case data from actual surveillance, then removing people (left-truncating) 1 week at a time
  3. b N refers only to the observed data. Some characteristics were not reported in all surveillance records, therefore N varies
  4. c Cough attributes include cases having one or more of the following (proportion positive within the dataset): paroxysms (85.1 %), post-tussive vomiting (51.2 %), or whooping (37.0 %)
  5. d The difference of the mean cough duration of cases with their first medical visit in the current week (same row) compared to the previous week (previous row)
  6. e Week 0 cases had their first medical visit prior to the onset of cough, and received antibiotics prophylactically
  7. f The 5th week also contains all first medical visits in the fifth and following weeks