Development and validation of a nomogram to better predict hypertension based on a 10-year retrospective cohort study in China

  1. Xinna Deng
  2. Huiqing Hou
  3. Xiaoxi Wang
  4. Qingxia Li
  5. Xiuyuan Li
  6. Zhaohua Yang
  7. Haijiang Wu  Is a corresponding author
  1. Departments of Oncology & Immunotherapy, Hebei General Hospital, China
  2. Physical Examination Center, Hebei General Hospital, China
  3. Department of Foreign Language Teaching, Hebei Medical University, China
  4. Department of Pathology, Hebei Medical University, China
  5. Medical Practice-Education Coordination & Medical Education Research Center, Hebei Medical University, China
8 figures, 4 tables and 1 additional file

Figures

Texture feature selection using the least absolute shrinkage and selection operator (LASSO) binary logistic regression model.

(A) Identification of the optimal penalization coefficient lambda (λ) in the LASSO model with 10-fold cross-validation in Group140/90. (B) LASSO coefficient profiles of 21 features in Group140/90. The trajectory of each hypertension-related features’ coefficient was observed in the LASSO coefficient profiles with the changing of the lambda in LASSO algorithm. (C) Identification of the optimal penalization coefficient lambda (λ) in the LASSO model with 10-fold cross-validation in Group130/80. (D) LASSO coefficient profiles of 21 features in Group130/80. The trajectory of each hypertension-related features’ coefficient was observed in the LASSO coefficient profiles with the changing of the lambda in LASSO algorithm. SBP: systolic blood pressure; DBP: diastolic blood pressure; BMI: body mass index; WBC: white blood cell count; LYMPH: lymphocyte count; NEUT: neutrophil count; LYMPHP: lymphocyte percentage; NEUTP: neutrophil percentage; RBC: red blood cell count; MCHC: mean cell hemoglobin concentration; RDWCV: red blood cell distribution width-coefficient of variation; RDWSD: red blood cell distribution width standard deviation; PLT: platelet count; MPV: mean platelet volume; PCT: plateletcrit; PDW: platelet distribution width; ALT: alanine aminotransferase; AST: aspartate transaminase; TP: total protein; TBIL: total bilirubin; GLU: glucose; CHOL: cholesterol; TG: triglycerides; NLR: neutrophil-to-lymphocyte ratio.

Nomogram for the prediction of hypertension.

(A) Nomogram140/90 was constructed based on the data of Group140/90. (B) Nomogram130/80 was constructed based on the data of Group130/80. The points of each features were added to obtain the total points, and a vertical line was drawn on the total points to obtain the corresponding ‘risk of hypertension’. SBP: systolic blood pressure; DBP: diastolic blood pressure; BMI: body mass index; MCHC: mean cell hemoglobin concentration; MPV: mean platelet volume; TP: total protein; TBIL: total bilirubin; TG: triglycerides; RDWSD: red blood cell distribution width standard deviation.

Receiver operating characteristic (ROC) curves for the prediction of hypertension in the training set and validation set.

(A) ROC curves of the factors and nomogram140/90 in the training set of Group140/90. (B) ROC curves of the factors and nomogram130/80 in the training set of Group130/80. (C) ROC curves of the factors and nomogram140/90 in the validation set of Group140/90. (D) ROC curves of the factors and nomogram130/80 in the validation set of Group130/80. SBP: systolic blood pressure; DBP: diastolic blood pressure; BMI: body mass index; MCHC: mean cell hemoglobin concentration; MPV: mean platelet volume; TP: total protein; TBIL: total bilirubin; TG: triglycerides; RDWSD: red blood cell distribution width standard deviation.

Calibration curves of the nomogram prediction in the training set and validation set.

(A) Calibration curves of nomogram140/90 prediction in the training set of Group140/90. (B) Calibration curves of nomogram130/80 prediction in the training set of Group130/80. (C) Calibration curves of nomogram140/90 prediction in the validation set of Group140/90. (D) Calibration curves of nomogram130/80 prediction in the validation set of Group130/80.

Decision curve analysis (DCA) of the nomogram prediction in the training set and validation set.

(A) DCA of nomogram140/90 prediction in the training set of Group140/90. (B) DCA of nomogram130/80 prediction in the training set of Group130/80. (C) DCA of nomogram140/90 prediction in the validation set of Group140/90. (D) DCA of nomogram130/80 prediction in the validation set of Group130/80. SBP: systolic blood pressure; DBP: diastolic blood pressure; BMI: body mass index; MCHC: mean cell hemoglobin concentration; MPV: mean platelet volume; TP: total protein; TBIL: total bilirubin; TG: triglycerides; RDWSD: red blood cell distribution width standard deviation.

Clinical impact curves of the nomogram prediction in the training set and validation set.

(A) Clinical impact curves of nomogram140/90 prediction in the training set of Group140/90. (B) Clinical impact curves of nomogram130/80 prediction in the training set of Group130/80. (C) Clinical impact curves of nomogram140/90 prediction in the validation set of Group140/90. (D) Clinical impact curves of nomogram130/80 prediction in the validation set of Group130/80.

Flowchart of the procedure.

A total of 51,165 and 209,636 subjects who underwent physical examination in 2009 and 2019 were enrolled in this study, respectively. 8020 subjects who underwent medical examination both in 2009 and 2019 were finally enrolled. At a cut-off value of 140/90 mmHg, 6201 subjects who had normal blood pressure in 2009 were enrolled in Group140/90. At a cut-off value of 130/80 mmHg, 3771 subjects who had normal blood pressure in 2009 were enrolled in Group130/80. The data of Group140/90 and Group130/80 were used to construct the nomogram140/90 and nomogram130/80 for predicting hypertension, respectively.

Author response image 1

Tables

Table 1
Baseline characteristicsh of individuals in training set and validation set of Group140/90.
VariablesTraining set (N = 4,134)Validation set (N = 2,067)P values
Hypertension status in 2019, n (%)
No3,086 (74.65%)1,579 (76.39%)0.1342
Yes1,048 (25.35%)488 (23.61%)
Gender
Female1,899 (45.94%)968 (46.83%)0.5052
Male2,235 (54.06%)1,099 (53.17%)
Family history of hypertension
No3,029 (73.27%)1,491 (72.13%)0.3424
Yes1,105 (26.73%)576 (27.87%)
Smoking status
No3,846 (93.03%)1945 (94.10%)0.1118
Yes288 (6.97%)122 (5.90%)
Drinking status
No3,715 (89.86%)1,871 (90.52%)0.4173
Yes419 (10.14%)196 (9.48%)
Age, year45.00 (37.00,54.00)45.00 (37.00,54.00)0.5448
SBP,mmHg110.00 (100.00,120.00)110.00 (100.00,120.00)0.1163
DBP,mmHg70.00 (70.00,80.00)70.00 (70.00,80.00)0.2938
Height, cm167.00 (161.00,173.00)166.00 (161.00,173.00)0.3414
weight, kg66.00 (59.00,76.00)66.00 (59.00,75.00)0.8163
BMI, kg/m224.03 (21.89,26.22)24.07 (21.89,26.09)0.9343
WBC, 109/L5.50 (4.60,6.40)5.50 (4.60,6.30)0.6647
LYMPH, 109/L1.80 (1.50,2.10)1.80 (1.50,2.10)0.9866
NEUT, 109/L3.20 (2.60,3.90)3.20 (2.60,3.80)0.7023
LYMPHP, %33.20 (28.90,37.80)33.60 (29.20,37.90)0.2718
NEUTP, %58.50 (53.70,63.20)58.30 (53.70,63.10)0.6369
RBC, 1012/L4.25 (3.93,4.57)4.25 (3.93,4.56)0.9433
HGB, g/L130.00 (119.00,141.00)129.00 (119.00,141.00)0.6263
HCT, %39.20 (37.00,42.30)39.10 (37.00,42.10)0.5409
MCV, fL92.20 (89.70,94.60)92.20 (89.60,94.60)0.5249
MCH, pg30.60 (29.60,31.00)30.60 (29.60,31.00)0.5454
MCHC, g/L331.00 (325.00,337.00)331.00 (325.00,337.00)0.5947
RDWCV, %14.40 (14.00,14.50)14.40 (14.00,14.50)0.1879
RDWSD, fL48.70 (46.50,50.20)48.70 (46.50,50.20)0.8534
PLT, 109/L211.00 (182.00,243.00)211.00 (184.00,243.00)0.9817
MPV, fL8.80 (8.30,9.30)8.80 (8.30,9.30)0.3084
PCT, %0.18 (0.16,0.21)0.19 (0.16,0.21)0.5154
PDW, fL15.80 (15.70,16.00)15.80 (15.70,16.00)0.8825
MID, 109/L0.40 (0.30,0.50)0.40 (0.30,0.50)0.1560
MIDP, %8.20 (7.10,9.00)8.10 (7.10,9.00)0.1760
ALT, U/L17.90 (14.20,24.30)18.00 (14.30,23.90)0.8201
AST, U/L19.40 (17.00,22.90)19.40 (17.00,22.60)0.6713
TP, g/L71.40 (69.00,73.90)71.40 (69.00,73.90)0.8951
ALB, g/L42.38 (40.82,44.18)42.44 (40.83,44.00)0.4247
TBIL, μmol/L16.90 (13.80,20.80)16.80 (14.00,20.80)0.6959
DBIL, μmol/L2.00 (2.00,3.00)2.00 (2.00,3.00)0.9172
GLU, mmol/L5.49 (5.19,5.86)5.50 (5.17,5.87)0.7616
CHOL, mmol/L4.71 (4.17,5.32)4.72 (4.19,5.34)0.3311
TG, mmol/L1.20 (0.82,1.78)1.20 (0.82,1.78)0.9119
NLR, %1.76 (1.43,2.19)1.74 (1.42,2.17)0.2978
PLR, %116.75 (96.07,141.88)117.06 (97.50,140.67)0.7375
  1. Data are presented as median (25% percentile, 75% percentile) for continuous variables and count (percentage) for categorical variables.

  2. SBP: systolic blood pressure; DBP: diastolic blood pressure; BMI: body mass index; WBC: white blood cell count; LYMPH: lymphocyte count; NEUT: neutrophil count; LYMPHP: lymphocyte percentage; NEUTP: neutrophil percentage; RBC: red blood cell count; HGB: hemoglobin; HCT: hematocrit; MCV: mean corpuscular volume; MCH: mean corpuscular hemoglobin; MCHC: mean cell hemoglobin concentration; RDWCV: red blood cell distribution width-coefficient of variation; RDWSD: red blood cell distribution width standard deviation; PLT: platelet count; MPV: mean platelet volume; PCT: plateletcrit; PDW: platelet distribution width; MID: middle cell count; MIDP: middle cell percentage; ALT: alanine aminotransferase; AST: aspartate transaminase; TP: total protein; ALB: albumin; TBIL: total bilirubin; DBIL: direct bilirubin; GLU: glucose; CHOL: cholesterol; TG: triglycerides; NLR: neutrophil-to-lymphocyte ratio; PLR: platelet-to-lymphocyte ratio.

Table 2
Baseline characteristics of individuals in training set and validation set of Group130/80.
VariablesTraining set (N = 2,514)Validation set (N = 1,257)P values
Hypertension status in 2019, n (%)
No1,574 (62.61%)767 (61.02%)0.3425
Yes940 (37.39%)490 (38.98%)
Gender
Female1,436 (57.12%)685 (54.49%)0.1255
Male1,078 (42.88%)572 (45.51%)
Family history of hypertension
No1,886 (75.02%)949 (75.50%)0.7491
Yes628 (24.98%)308 (24.50%)
Smoking status
No2,403 (95.58%)1,195 (95.07%)0.4743
Yes111 (4.42%)62 (4.93%)
Drinking status
No2,342 (93.16%)1,173 (93.32%)0.8547
Yes172 (6.84%)84 (6.68%)
Age, year43.00 (36.00,53.00)44.00 (37.00,53.00)0.2488
SBP, mmHg104.00 (100.00,110.00)104.00 (100.00,110.00)0.3309
DBP, mmHg70.00 (62.00,70.00)70.00 (64.00,70.00)0.1694
Height, cm165.00 (160.00,172.00)165.00 (160.00,172.00)0.8938
weight, kg63.00 (56.00,72.00)63.00 (56.00,72.00)0.3730
BMI, kg/m223.24 (21.30,25.39)23.24 (21.09,25.27)0.3171
WBC, 109/L5.30 (4.50,6.30)5.30 (4.50,6.30)0.7810
LYMPH, 109/L1.80 (1.50,2.10)1.70 (1.50,2.10)0.3251
NEUT, 109/L3.10 (2.50,3.80)3.10 (2.50,3.80)0.9247
LYMPHP, %33.40 (29.00,37.90)33.40 (29.20,38.20)0.5755
NEUTP, %58.30 (53.70,63.20)58.50 (53.25,63.20)0.6634
RBC, 1012/L4.15 (3.87,4.48)4.14 (3.86,4.47)0.4514
HGB, g/L126.00 (116.00,138.00)126.00 (117.00,138.00)0.5630
HCT, %38.10 (37.00,41.30)38.20 (37.00,41.40)0.6286
MCV, fL92.20 (89.50,94.60)92.30 (89.75,94.80)0.2552
MCH, pg30.50 (29.40,31.00)30.60 (29.55,31.00)0.0824
MCHC, g/L330.00 (324.00,336.00)331.00 (324.00,337.00)0.0561
RDWCV, %14.50 (14.00,14.50)14.50 (14.10,14.50)0.9732
RDWSD, fL48.70 (47.20,50.20)48.70 (47.20,50.20)0.1531
PLT, 109/L210.00 (181.00,241.00)210.00 (181.00,239.50)0.8556
MPV, fL8.80 (8.30,9.30)8.90 (8.40,9.30)0.3158
PCT, %0.18 (0.16,0.21)0.18 (0.16,0.21)0.9297
PDW, fL15.80 (15.68,16.00)15.80 (15.70,16.00)0.6276
MID, 109/L0.40 (0.30,0.50)0.40 (0.30,0.50)0.5171
MIDP, %8.10 (7.10,9.00)8.10 (6.90,9.00)0.5763
ALT, U/L16.60 (13.50,22.20)16.70 (13.50,22.30)0.5375
AST, U/L18.70 (16.50,22.20)18.90 (16.50,22.00)0.5839
TP, g/L71.30 (69.00,73.90)71.20 (68.80,73.60)0.3776
ALB, g/L42.21 (40.64,43.96)42.29 (40.64,43.96)0.4390
TBIL, μmol/L16.70 (13.70,20.50)16.70 (13.70,20.45)0.6017
DBIL, μmol/L2.00 (2.00,3.00)2.00 (2.00,3.00)0.8340
GLU, mmol/L5.43 (5.15,5.77)5.42 (5.13,5.73)0.3640
CHOL, mmol/L4.65 (4.10,5.24)4.69 (4.13,5.32)0.1182
TG, mmol/L1.05 (0.74,1.56)1.06 (0.75,1.60)0.4149
NLR, %1.75 (1.43,2.19)1.75 (1.41,2.17)0.5353
PLR, %119.23 (98.57,144.44)117.62 (96.10,143.43)0.1833
  1. Data were presented as median (the 25% percentile, the 75% percentile) for continuous variables and count (percentage) for categorical variables.

  2. SBP, systolic blood pressure; DBP, diastolic blood pressure; BMI, body mass index; WBC, white blood cell count; LYMPH, lymphocyte count; NEUT, neutrophil count; LYMPHP, lymphocyte percentage; NEUTP, neutrophil percentage; RBC, red blood cell count; HGB, hemoglobin; HCT, hematocrit; MCV, mean corpuscular volume; MCH, mean corpuscular hemoglobin; MCHC, mean cell hemoglobin concentration; RDWCV, red blood cell distribution width-coefficient of variation; RDWSD, red blood cell distribution width standard deviation; PLT, platelet count; MPV, mean platelet volume; PCT, plateletcrit; PDW, platelet distribution width; MID, middle cell count; MIDP, middle cell percentage; ALT, alanine aminotransferase; AST, aspartate transaminase; TP, total protein; ALB, albumin; TBIL, total bilirubin; DBIL, direct bilirubin; GLU, glucose; CHOL, cholesterol; TG, triglycerides; NLR, neutrophil to lymphocyte ratio; PLR, platelet-to-lymphocyte ratio.

Table 3
Risk factors for hypertension in the training set of Group140/90.
VariableModel
β-CoefficientOdds ratio (95%CI)p value
Family history of hypertension
NoReference
Yes0.4831.621 (1.372–1.913)<0.001
Age0.0361.037 (1.028–1.045)<0.001
SBP0.0411.041 (1.032–1.051)<0.001
DBP0.0311.031 (1.017–1.046)<0.001
BMI0.0391.040 (1.011–1.069)0.006
MCHC–0.0150.985 (0.975–0.995)0.003
MPV0.1611.175 (1.036–1.333)0.012
TP0.0251.025 (1.003–1.049)0.028
TBIL0.0151.015 (1.002–1.028)0.023
TG0.0981.102 (0.962–1.132)0.012
  1. SBP: systolic blood pressure; DBP: diastolic blood pressure; BMI: body mass index; MCHC: mean cell hemoglobinconcentration; MPV: mean platelet volume; TP: total protein; TBIL: total bilirubin; TG: triglycerides.

Table 4
Risk factors for hypertension in the training set of Group130/80.
VariableModel
β-CoefficientOdds ratio (95%CI)p value
Family history of hypertension
NoReference
Yes0.2821.326 (1.084–1.620)0.006
Age0.0321.032 (1.022–1.042)<0.001
SBP0.0401.041 (1.029–1.053)<0.001
DBP0.0331.034 (1.012–1.065)0.002
RDWSD–0.0470.954 (0.916–0.994)0.024
TBIL0.0151.016 (1.001–1.031)0.041
  1. SBP: systolic blood pressure; DBP: diastolic blood pressure; RDWSD: red blood celldistribution width standard deviation; TBIL: total bilirubin.

Additional files

Download links

A two-part list of links to download the article, or parts of the article, in various formats.

Downloads (link to download the article as PDF)

Open citations (links to open the citations from this article in various online reference manager services)

Cite this article (links to download the citations from this article in formats compatible with various reference manager tools)

  1. Xinna Deng
  2. Huiqing Hou
  3. Xiaoxi Wang
  4. Qingxia Li
  5. Xiuyuan Li
  6. Zhaohua Yang
  7. Haijiang Wu
(2021)
Development and validation of a nomogram to better predict hypertension based on a 10-year retrospective cohort study in China
eLife 10:e66419.
https://doi.org/10.7554/eLife.66419