Table 3.

Summary Estimates of the Diagnostic Accuracy of Signs and Symptoms for Bordetella pertussis Infection^*

Sign or Symptom	Studies^*	Sensitivity (95% CI)		Specificity (95% CI)	LR+ (95% CI)^{^‡}	LR− (95% CI)^{^‡}	DOR (95% CI)^{^‡}	AUROCC^{^‡}
Sign or Symptom	Studies^*	Studies Reporting Sensitivity and Specificity	All Studies^{^†}	Specificity (95% CI)	LR+ (95% CI)^{^‡}	LR− (95% CI)^{^‡}	DOR (95% CI)^{^‡}	AUROCC^{^‡}
Overall clinical impression	4/5	0.47 (0.36–0.59)	0.48 (0.39–0.58)	0.85 (0.71–0.93)	3.3 (1.9–5.5)^{^§}	0.63 (0.56–0.70)^{^§}	5.3 (3.3–8.1)^{^§}	0.659
Whooping cough
All	12/17	0.31 (0.23–0.41)	0.31 (0.24–0.40)	0.84 (0.74–0.91)	1.9 (1.2–3.1)^{^§}	0.83 (0.73–0.95)^{^§}	2.5 (1.4–4.2)^{^§}	0.545
Children	5/6	0.41 (0.24–0.61)	0.43 (0.29–0.57)	0.84 (0.52–0.96)	2.9 (1.2–7.3)	0.72 (0.60–0.84)	4.1 (1.4–9.8)	0.593
Adults	7/8	0.21 (0.16–0.28)	0.17 (0.11–0.23)	0.82 (0.70–0.90)	1.2 (0.81–1.8)	0.97 (0.81–1.8)	1.3 (0.75–2.0)	0.349
Posttussive vomiting
All	14/19	0.42 (0.33–0.52)	0.45 (0.35–0.54)	0.75 (0.68–0.80)	1.7 (1.4–1.9)^{^§}	0.77 (0.68–0.86)^{^§}	2.2 (1.7–2.7)^{^§}	0.646
Children	5/7	0.56 (0.39–0.72)	0.55 (0.41–0.68)	0.66 (0.52–0.78)	1.7 (1.3–2.2)	0.67 (0.49–0.84)	2.6 (1.6–3.9)	0.650
Adults	7/8	0.34 (0.26–0.43)	0.35 (0.27–0.42)	0.80 (0.74–0.84)	1.7 (1.3–2.3)	0.83 (0.73–0.93)	2.0 (1.3–3.0)	0.602
Paroxysmal cough
All	13/16	0.80 (0.71–0.86)	0.79 (0.73–0.86)	0.35 (0.24–0.49)	1.2 (1.1–1.4)^{^§}	0.58 (0.49–0.68)^{^§}	2.2 (1.7–2.7)^{^§}	0.653
Children	5/7	0.72 (0.60–0.81)	0.75 (0.65–0.84)	0.47 (0.32–0.62)	1.4 (1.2–1.7)	0.61 (0.52–0.72)	2.3 (1.6–3.1)	0.644
Adults	6/6	0.88 (0.80–0.92)	=	0.26 (0.12–0.47)	1.2 (1.04–1.6)	0.50 (0.33–0.75)	2.6 (1.4–4.4)	0.774
Sputum production	4/4	0.79 (0.52–0.93)	—	0.34 (0.18–0.56)	1.2 (1.1–1.3)^{^§}	0.63 (0.39–0.89)^{^§}	2.0 (1.3–3.0)^{^§}	0.555
Wheezing absent	4/4	0.75 (0.34–0.95)	—	0.36 (0.17–0.62)	1.1 (0.90–1.3)	0.69 (0.31–1.1)	1.9 (0.83–3.8)	0.514
CDC clinical criteria	3/3	0.90 (0.82–0.95)	—	0.16 (0.13–0.19)	1.07 (0.97–1.1)	0.66 (0.33–1.1)	1.8 (0.86–3.4)	0.19
Headache absent	3/3	0.60 (0.38–0.85)	—	0.54 (0.24–0.81)	1.3 (1.05–1.8)^{^§}	0.76 (0.51–0.95)^{^§}	1.8 (1.1–2.7)^{^§}	0.588
Sleep disturbed by cough	5/6	0.55 (0.21–0.85)	0.56 (0.31–0.80)	0.54 (0.19–0.85)	1.2 (1.0–1.5)^{^§}	0.83 (0.69–0.95)^{^§}	1.5 (1.1–1.9)^{^§}	0.557
Dyspnea	4/4	0.21 (0.09–0.42)	—	0.84 (0.57–0.95)	1.3 (0.92–1.9)	0.96 (0.90–1.05)	1.4 (0.89–2.1)	0.467
Sneezing	4/4	0.78 (0.58–0.90)		0.23 (0.13–0.36)	1.00 (0.87–1.09)	0.97 (0.63–1.4	1.1 (0.65–1.7)	0.423
Apnea	5/6	0.07 (0.02–0.22)	0.10 (0.03–0.17)	0.93 (0.83–0.97)	1.0 (0.64–1.5)	0.99 (0.94–1.02)	1.0 (0.64–1.5)	0.606
Cyanosis	4/4	0.05 (0.01–0.24)	—	0.95 (0.79–0.99)	0.99 (0.66–1.4)	1.0 (0.96–1.02)	1.0 (0.67–1.4)	0.545
Chest pain	3/3	0.25 (0.16–0.37)		0.74 (0.67–0.80)	0.96 (0.72–1.2)	1.01 (0.92–1.1)	0.95 (0.67–1.3)	0.55
Fever	7/8	0.13 (0.05–0.29)	0.17 (0.05–0.29)	0.82 (0.71–0.90)	0.76 (0.45–1.2)	1.0 (0.95–1.1)	0.73 (0.41–1.2)	0.549

The estimates presented here include only signs and symptoms for which at least 3 studies reported both sensitivity and specificity. Entries are sorted by diagnostic odds ratios.
↵* Data are the number of studies with reported values for both sensitivity and specificity among the total studies found.
↵† When studies reported sensitivity only, sensitivity was calculated using a random effects model.
↵‡ These values were calculated using only data from studies that were able to calculate both sensitivity and specificity.
↵§ Value is significantly different from 1.0.
CDC, Centers for Disease Control and Prevention; CI, confidence interval; DOR, diagnostic odds ratio; LR+, positive likelihood ratio; LR−, negative likelihood ratio; AUROCC, area under the receiver operating characteristic curve (also known as the c-statistic).