The accuracy of a handheld “disposable pneumotachograph device” in the spirometric diagnosis of airway obstruction in a Chinese population

Figures & data

Figure 1 The handheld “disposable pneumotachograph” spirometer.

Table 1 Correlation coefficient for all indices measured (n=241)

Index	Correlation coefficient (95% CI)	P-value
FVC	0.970 (0.961–0.977)	<0.001
FEV1	0.988 (0.984–0.990)	<0.001
FEV1/FVC	0.972 (0.964–0.978)	<0.001
FEF25%–75%	0.959 (0.948–0.968)	<0.001
PEF	0.949 (0.935–0.961)	<0.001

Abbreviation: FEF_25%–75%, mean FEF calculated between 25% and 75% of FVC.

Figure 2 Correlation of key airflow metrics between devices (n=241).

Note: Line of identity (thin solid line), line of regression (thick solid line), and 95% CIs (dotted line).
Abbreviation: FEF_25%–75%, mean FEF calculated between 25% and 75% of FVC.

Table 2 The mean difference and 95% LoA for all parameters measured by the two spirometers (n=241)

Variable	$\overline{d}$ (bias)a	Accuracy	P-value	LoA (LL, UL)b
FVC	0.02L	0.61%	>0.05	(−0.38, 0.42)
FEV1	−0.04L	−1.67%	<0.05	(−0.29, 0.21)
FEV1/FVC	−0.02	−2.74%	<0.05	(−0.09, 0.06)
FEF25%–75%	−0.13 L/s	−6.73%	<0.05	(−0.83, 0.57)
PEF	−21.2 L/min	−6.00%	<0.05	(−99.6, 57.1)

Notes:

a The mean differences ($\overline{d}$) between the two spirometers were regarded as the estimated bias;

b The 95% limits of agreement (LoA) are expressed as $\overline{d}\pm 1.96$ SD; the upper limit of agreement (UL) is $\overline{d}+1.96$ SD, and the lower limit of agreement (LL) is $\overline{d}-1.96$ SD.

Abbreviation: FEF_25%–75%, mean FEF calculated between 25% and 75% of FVC.

Figure 3 Bland–Altman plot of mean differences against averages of two readings for key airflow metrics (n=241).

Notes: The 95% limits of agreement (LoA) are expressed as mean differences ($\overline{d}$)±1.96 SD; the upper limit of agreement (UL) is $\overline{d}+1.96$ SD, and the lower limit of agreement (LL) is $\overline{d}-1.96$ SD. Line of mean difference (solid line), lines of 95% LoA (dashed line), and zero line (dotted line).
Abbreviation: FEF_25%–75%, mean FEF calculated between 25% and 75% of FVC.

Table 3 Spirometric classification of severity of airway obstruction by two spirometers for all enrolled subjects (n=241)

IQspiro (FEV1/FVC<70%)	Elite DX (FEV1/FVC <70%)				Total
	Non-obstruction	Obstruction (Stage 1)	Obstruction (Stage 2)	Obstruction (Stage 3 and 4)
Non-obstruction	100	0	0	0	100
Obstruction (Stage 1)	17	57	4	0	78
Obstruction (Stage 2)	3	4	28	0	35
Obstruction (Stage 3 and 4)	0	0	3	25	28
Total	120	61	35	25	241

Table 4 Spirometric classification of airway obstruction by two spirometers for individuals able to exhale for at least 6 seconds (FEV₁/FEV₆ <74%) (n=159)

IQspiro (FEV1/FEV6<74%)	Elite DX (FEV1/FVC <70%)		Total
	Negative	Positive
Negative	61 (92.4%)	5 (5.4%)	66
Positive	5 (7.6%)	88 (94.6%)	93
Total	66 (100%)	93 (100%)	159

Figure 4 The receiver operating characteristic curve that discriminates the optimal cutoff value of FEV₁/FEV₆ measured using the IQspiro (n=159).

Notes: (A) The sensitivity and specificity when the FEV₁/FEV₆ was <74%. (B) The sensitivity and specificity when the FEV₁/FEV₆ was <70%.

Table 5 Spirometric classification of airway obstruction by two spirometers for individuals able to exhale for at least 6 seconds (FEV₁/FEV₆ <70%) (n=159)

IQspiro (FEV1/FEV6<70%)	Elite DX (FEV1/FVC <70%)		Total
	Negative	Positive
Negative	64 (97.0%)	20 (21.5%)	84
Positive	2 (3.0%)	73 (78.5%)	75
Total	66 (100%)	93 (100%)	159