‘Validation of ultrasound examinations performed by general practitioners’

Figures & data

Table 1. Point-of-care ultrasound examinations included in the study

• Gallstones• Ascites• Abdominal aorta >5 cm diameter• Intrauterine pregnancy• Gestational age^a

^a The study data sheet was noted with ‘yes’ for both examinations (general practice and control examination) if the estimated gestational ages were within three days of each other. Otherwise ‘no’ was noted for the control examination.

Table 2. Kappa value interpretation

Value of Kappa	Level of agreement
0–.20	None
.21–.39	Minimal
.40–.59	Weak
.60–.79	Moderate
.80–.90	Strong
>.90	Almost perfect

Figure 1. Flow diagram of sample selection for the study.

Table 3. Study results

	Distribution of ultrasound scans				Percentageagreement	Kappa value	95% confidenceinterval
All scans			Control scan					Sensitivity	0.98
			Yes	No	96	0.93	0.8712–0.9796	Specificity	0.95
	GP scan	Yes	87	5				PPV^b	0.95
		No	2	94				NPV^c	0.98
Gallstones			Control scan					Sensitivity	0.92
			Yes	No	92	0.84	0.6969–0.9737	Specificity	0.92
	GP scan	Yes	24	3				PPV^b	0.89
		No	2	33				NPV^c	0.94
Ascites			Control scan					Sensitivity	1
			Yes	No	100	1	1.00–1.00	Specificity	1
	GP scan	Yes	3	0				PPV^b	1
		No	0	31				NPV^c	1
Abdominal aorta >5 cm			Control scan					Sensitivity	1
			Yes	No	100	1	1.00–1.00	Specificity	1
	GP scan	Yes	1	0				PPV^b	1
		No	0	28				NPV^c	1
Intrauterine pregnancy			Control scan					Sensitivity	1
			Yes	No	100	1	1.00–1.00	Specificity	1
	GP scan	Yes	31	0				PPV^b	1
		No	0	2				NPV^c	1
Gestational age			Control scan					Sensitivity	1
			Yes	No	93	NA^a	–	Specificity	0^d
	GP scan	Yes	28	2				PPV^b	0.93
		No	0	0				NPV^c	NA^e

^a A Kappa value could not be calculated because the actual data values were not binary for this parameter.

^b Positive predictive value.

^c Negative predictive value.

^d Value of zero because of the way data were registered for this specific parameter (Table 1).

^e NA because there were no negative test results because of the way these parameters were registered (Table 1).