Development and validation of a clinical factor-based nomogram for predicting imaging-defined cardiopulmonary abnormality risk in an asymptomatic screening population

Table 1 Clinical baseline characteristics of 1035 asymptomatic health examination individuals aged 40 years and above, n (%)

Index	Total (n = 1035)	Low-risk group (n = 346)	High-risk group (n = 689)	t/χ2/z	P value
Age	52.632 ± 7.987	50.812 ± 7.361	53.546 ± 8.136	5.261	< 0.001
BMI	24.244 ± 3.040	23.135 ± 2.653	24.802 ± 3.070	8.612	< 0.001
Gender	-	-	-	11.872	< 0.001
Male	680 (65.700)	202 (58.382)	478 (69.376)	-	-
Female	355 (34.300)	144 (41.618)	211 (30.624)	-	-
Metabolic status	-	-	-	121.603	< 0.001
Abnormal	524 (50.628)	91 (26.301)	433 (62.845)	-	-
Normal	511 (49.372)	255 (73.699)	256 (37.155)	-	-
Smoking history	-	-	-	134.002	< 0.001
Non-smoker	465 (44.928)	242 (69.942)	223 (32.366)	-	-
Mild (smoking index ≤ 200)	134 (12.947)	32 (9.249)	102 (14.804)	-	-
Moderate (200 < smoking index < 400)	91 (8.792)	16 (4.624)	75 (10.885)	-	-
Heavy (smoking index ≥ 400)	345 (33.333)	56 (16.185)	289 (41.945)	-	-
Alcohol consumption history	-	-	-	3.191	0.526
Non-drinker	557 (53.816)	197 (56.936)	360 (52.250)	-	-
Rarely (< 1/month)	211 (20.386)	64 (18.497)	147 (21.335)	-	-
Light (2-4/month)	145 (14.010)	50 (14.451)	95 (13.788)	-	-
Moderate (2-3/week)	54 (5.217)	15 (4.335)	39 (5.660)	-	-
Heavy (≥ 4/week)	68 (6.570)	20 (5.780)	48 (6.967)	-	-
Work style	-	-	-	14.008	0.001
Sedentary behavior	368 (35.556)	148 (42.775)	220 (31.930)	-	-
Light physical activity	231 (22.319)	60 (17.341)	171 (24.819)	-	-
Moderate-to-high physical activity	436 (42.126)	138 (39.884)	298 (43.251)	-	-
High-oil high-salt diet	-	-	-	0.756	0.385
No	885 (85.507)	301 (86.994)	584 (84.761)	-	-
Yes	150 (14.493)	45 (13.006)	105 (15.239)	-	-
Sleep duration	-	-	-	1.043	0.594
< 6 hours	57 (5.507)	16 (4.624)	41 (5.951)	-	-
6-8 hours	546 (52.754)	188 (54.335)	358 (51.959)	-	-
> 8 hours	432 (41.739)	142 (41.040)	290 (42.090)	-	-
NYHA class	-	-	-	1.867	0.600
NYHA Class I	969 (93.623)	327 (94.509)	642 (93.179)	-	-
NYHA Class II	60 (5.797)	18 (5.202)	42 (6.096)	-	-
NYHA Class III	3 (0.290)	1 (0.289)	2 (0.290)	-	-
NYHA Class IV	3 (0.290)	0 (0.000)	3 (0.435)	-	-
Family history of heart disease	-	-	-	1.437	0.231
No	650 (62.802)	208 (60.116)	442 (64.151)	-	-
Yes	385 (37.198)	138 (39.884)	247 (35.849)	-	-
Occupational exposure history	-	-	-	0.019	0.891
No	984 (95.072)	328 (94.798)	656 (95.210)	-	-
Yes	51 (4.928)	18 (5.202)	33 (4.790)	-	-
Family history of lung cancer	-	-	-	0.000	1.000
No	978 (94.493)	327 (94.509)	651 (94.485)	-	-
Yes	57 (5.507)	19 (5.491)	38 (5.515)	-	-

Work style was self-reported and originally classified into five categories. Due to the very small sample size in the high-intensity physical activity category (< 10 cases), the five categories were merged into three: Sedentary behavior (reference), light physical activity, and moderate-to-high physical activity. Abnormal metabolic status was defined as any one of fasting blood glucose, blood lipids (total cholesterol or triglycerides), or blood pressure exceeding the normal reference range. Smoking history was categorized by smoking index (pack-years) into four levels. NYHA: New York Heart Association; BMI: Body mass index.

Table 2 Allocation of training and validation sets among 1035 asymptomatic health examination individuals aged 40 years and above, n (%)

Index	Training set (n = 724)	Validation set (n = 311)	t/χ2/z	P value
Age	52.481 ± 7.874	52.984 ± 8.247	-0.929	0.353
BMI	24.255 ± 2.983	24.220 ± 3.173	0.168	0.866
Sex	-	-	1.716	0.190
Male	466 (64.365)	214 (68.810)	-	-
Female	258 (35.635)	97 (31.190)	-	-
Metabolic status	-	-	0.070	0.791
Abnormal	369 (50.967)	155 (49.839)	-	-
Normal	355 (49.033)	156 (50.161)	-	-
Smoking history	-	-	5.055	0.168
Non-smoker	335 (46.271)	130 (41.801)	-	-
Mild (smoking index ≤ 200)	85 (11.740)	49 (15.756)	-	-
Moderate (200 < smoking index < 400)	68 (9.392)	23 (7.395)	-	-
Heavy (smoking index ≥ 400)	236 (32.597)	109 (35.048)	-	-
Alcohol consumption history	-	-	3.908	0.419
Non-drinker	394 (54.420)	163 (52.412)	-	-
Rarely (< 1/month)	154 (21.271)	57 (18.328)	-	-
Light (2-4/month)	95 (13.122)	50 (16.077)	-	-
Moderate (2-3/week)	38 (5.249)	16 (5.145)	-	-
Heavy (≥ 4/week)	43 (5.939)	25 (8.039)	-	-
Work style	-	-	1.285	0.526
Sedentary behavior	254 (35.083)	114 (36.656)	-	-
Light physical activity	157 (21.685)	74 (23.794)	-	-
Moderate-to-high physical activity	313 (43.232)	123 (39.550)	-	-
High-oil high-salt diet	-	-	2.127	0.145
No	611 (84.392)	274 (88.103)	-	-
Yes	113 (15.608)	37 (11.897)	-	-
Sleep duration	-	-	1.514	0.469
< 6 hours	36 (4.972)	21 (6.752)	-	-
6-8 hours	381 (52.624)	165 (53.055)	-	-
> 8 hours	307 (42.403)	125 (40.193)	-	-
NYHA class	-	-	3.284	0.350
NYHA Class I	679 (93.785)	290 (93.248)	-	-
NYHA Class II	41 (5.663)	19 (6.109)	-	-
NYHA Class III	3 (0.414)	0 (0.000)	-	-
NYHA Class IV	1 (0.138)	2 (0.643)	-	-
Family history of heart disease	-	-	0.200	0.655
No	451 (62.293)	199 (63.987)	-	-
Yes	273 (37.707)	112 (36.013)	-	-
Occupational exposure history	-	-	0.136	0.713
No	690 (95.304)	294 (94.534)	-	-
Yes	34 (4.696)	17 (5.466)	-	-
Family history of lung cancer	-	-	0.035	0.852
No	683 (94.337)	295 (94.855)	-	-
Yes	41 (5.663)	16 (5.145)	-	-

No statistically significant differences were observed between the training and validation sets for any variable (all P > 0.05), indicating adequate balance of random allocation. NYHA: New York Heart Association; BMI: Body mass index.

Table 3 Variance inflation factor analysis of clinical characteristics with significant differences in the study population

Variable	VIF	Assessment (< 5)
Sex	1.397	Pass
Age	1.032	Pass
BMI	1.098	Pass
Metabolic status	1.111	Pass
Smoking history	1.334	Pass
Work style	1.101	Pass

All variance inflation factor values were below 5, indicating no significant multicollinearity among the clinical characteristics, confirming their suitability for inclusion in the multivariate regression model. BMI: Body mass index; VIF: Variance inflation factor.

Table 4 Univariate and multivariate binary logistic regression analysis of clinical characteristics

Variable	Univariate analysis			Multivariate (enter) regression
	OR	95%CI	P value	OR	95%CI	P value
Sex	1.483	1.078-2.039	0.015	0.427	0.276-0.660	< 0.001
Age	1.041	1.020-1.063	< 0.001	1.028	1.005-1.052	0.018
BMI	1.187	1.121-1.256	< 0.001	1.150	1.079-1.226	< 0.001
Metabolic status	4.015	2.877-5.602	< 0.001	3.266	2.259-4.723	< 0.001
Smoking history	1.819	1.587-2.085	< 0.001	1.968	1.665-2.325	< 0.001
Work style	1.199	1.006-1.429	0.043	1.020	0.829-1.255	0.851
Alcohol consumption history	1.058	0.927-1.208	0.403
High-oil high-salt diet	1.326	0.852-2.062	0.211
Sleep duration	0.938	0.718-1.227	0.641
NYHA class	1.363	0.752-2.470	0.307
Family history of heart disease	0.775	0.565-1.063	0.113
Occupational exposure history	0.802	0.395-1.631	0.543
Family history of lung cancer	1.227	0.615-2.449	0.562

The multivariate analysis employed the forced entry (enter) method, incorporating all six variables identified as significant in univariate screening. Work style did not achieve statistical significance in the multivariate model (P = 0.851). The final nomogram incorporated five independently associated factors: Gender, age, body mass index, metabolic status, and smoking history. NYHA: New York Heart Association; BMI: Body mass index; OR: Odds ratio; CI: Confidence interval.

Table 5 Model performance metrics at optimal cut point (Youden Index)

Metric	Training set (n = 724) (%)	Validation set (n = 311) (%)
Optimal threshold (probability)	-	0.724
Sensitivity	58.3	57.5
Specificity	81	83.7
Positive predictive value	85.9	87.5
Negative predictive value	49.4	49.7