Journal List > Allergy Asthma Respir Dis > v.4(5) > 1059204

TOOLS
Yun, Rah, Choi, Kim, Oh, Kim, Chang, Yoo, and Sohn: The development of patient-tailored asthma prediction model for the alarm system
Original Article

Allergy, Asthma & Respiratory Disease 2016; 4(5): 328-339.

Published online: 30 September 2016

DOI: https://doi.org/10.4168/aard.2016.4.5.328

The development of patient-tailored asthma prediction model for the alarm system

Hey-Suk Yun1, Wee Jin Rah1, Young Jin Choi1, Joo-Hwa Kim1, Jae-Won Oh1, Hyun-Hee Kim2, Yoon-Seok Chang3, Kwang-Ha Yoo4, Keon-Tae Sohn5

1Department of Pediatrics, Hanyang University College of Medicine, Seoul, Korea.

2Department of Pediatrics, Uijeongbu St. Mary's Hospital, The Catholic University of Korea College of Medicine, Uijeongbu, Korea.

3Division of Allergy and Clinical Immunology, Department of Internal Medicine, Seoul National University Bundang Hospital, Seoul National University College of Medicine, Seoul, Korea.

4School of Medicine, Konkuk University, Seoul, Korea.

5Department of Statistics, Pusan National University, Busan, Korea.

Correspondence to: Correspondence to: Jae-Won Oh. Department of Pediatrics, Hanyang University Guri Hospital, Hanyang University College of Medicine, 153 Gyeongchun-ro, Guri 11923, Korea. Tel: +82-31-560-2254, Fax: +82-31-552-9493, jaewonoh@hanyang.ac.kr

Received 20 March 2016       Revised 27 July 2016       Accepted 29 July 2016

© 2016 The Korean Academy of Pediatric Allergy and Respiratory Disease; The Korean Academy of Asthma, Allergy and Clinical Immunology

(open-access):

This is an Open Access article distributed under the terms of the Creative Commons Attribution Non-Commercial License (http://creativecommons.org/licenses/by-nc/4.0/).

Abstract

Purpose

The increased incidence of asthma due to rising allergic diseases requires the prevention of worsening asthma. It is necessary to develop a patient-tailored asthma prediction model.

Methods

We developed causative factors for the asthma forecast system: infant and young children (0–2 years), preschool children (3–6 years), school children and adolescents (7–18 years), adults (19–64 years), old aged adult (>64 years). We used the Emergency Department code data which charged the short-acting bronchodilator (Salbutamol sulfate) from Health Insurance Review and Assessment Service for the development of asthma prediction models. Three kinds of statistical models (multiple regression models, logistic regression models, and decision tree models) were applied to 40 study groups (4 seasons, 2 sex, and 5 age groups) separately.

Results

The 3 kinds of models were compared based on model assessment measures. Estimated logistic regression models or decision tree models were recommended as binary forecast models. To improve the predictability, a threshold was used to generate binary forecasts.

Conclusion

We suggest the binary forecast models as a patient-tailored asthma prediction system for this category. It may be needed the extended study duration and long-term data analysis for asthmatic patients for the further improvement of asthma prediction models.

Keywords: Asthma, Asthma alarm system

Figures and Tables

Fig. 1

Box plot showing distribution of patients with asthma by sex and age group. M, man subject; W, woman subject. M1 and W1: 0–2 years old, M2 and W2: 3–6 years old, M3 and W3: 7–18 years old, M4 and W4: 19–64 years old, M5 and W5: 65 years old.

aard-4-328-g001
Fig. 2

Histogram showing distribution of patients with asthma by sex and age group. M, man subject; W, woman subject. M1 and W1: 0–2 years old, M2 and W2: 3–6 years old, M3 and W3: 7–18 years old, M4 and W4: 19–64 years old, M5 and W5: 65 years old.

aard-4-328-g002
Table 1

Potential predictors of asthma

aard-4-328-i001
Predictor Interval Type M1 M2 M3 M4 M5 W1 W2 W3 W4 W5
A Daily Quantitative 7 7 7 7 7 7 7 7 7 7
T Daily Quantitative 1 1 1 1 1 1 1 1 1 1
DT Daily Quantitative 1 1 1 1 1 1 1 1 1 1
MH Daily Quantitative 1 1 1 1 1 1 1 1 1 1
PR Daily Quantitative 1 1 1 1 1 1 1 1 1 1
HS Daily Quantitative 1 1 1 1 1 1 1 1 1 1
OZ Daily Quantitative 4 1 1 1 1 3 1 1 2 1
PL Daily Quantitative 1 1 1 1 1 1 1 1 1 1
FL Weekly Quantitative 1 1 1 1 1 1 1 1 1 1
PM Daily Quantitative 1 1 6 6 1 1 1 7 1 1
YS Daily Binary 1 2 2 3 1 1 2 6 1 2
D1–D6 Daily Dummy - - - - - - - - - -

Numbers indicate the significant lag times (days) between environmental factors and the Health Insurance Review and Assessment Service data (occurrence of asthma symptoms).

A, autocorrelated factor; T, mean temperature; DT, daily range; MH, minimum humidity; PR, pressure; HS, hours of sunshine; OZ, concentration of ozone; PL, pollen; FL, proportion of flu patients; PM, concentration of yellow sand; YS, presence of yellow sand; D, day of the week; M, man; W, woman.

Table 2

Assessment of model validity and predictability

aard-4-328-i002
Actual case Forecasted category
Continuous management Attention Total
Continuous management A B A+B
Attention C D C+D
Total A+C B+D A+B+C+D

A, negative correction; B, false alarm; C, miss; D, hit.

Table 3

Asthma cases in the HIRA dataset according to sex, age group (1–5), and season

aard-4-328-i003
Group Spring Summer Autumn Winter
Mean SD Mean SD Mean SD Mean SD
M1 14.0 6.9 7.6 5.3 13.7 6.5 12.4 6.2
M2 5.5 3.3 3.2 2.6 7.5 4.5 4.5 3.0
M3 2.7 2.0 1.6 1.8 4.7 4.5 2.1 1.9
M4 10.2 3.8 9.0 3.0 9.9 3.4 9.7 3.5
M5 17.5 5.0 14.1 5.0 14.7 4.7 16.2 4.9
W1 9.1 5.1 4.8 3.6 9.1 4.8 7.8 4.2
W2 3.8 2.7 2.1 1.7 4.7 3.0 2.9 2.1
W3 1.6 1.6 0.9 1.2 2.6 2.2 1.2 1.4
W4 7.7 3.0 6.1 2.6 7.8 3.4 8.0 3.2
W5 12.3 4.0 9.5 3.5 10.4 3.6 12.2 4.4
Total 84.5 21.0 58.9 16.2 85.1 21.0 76.9 17.3

HIRA, Health Insurance Review and Assessment Service; SD, standard deviation; M, man; W, woman.

Table 4

Potential predictors

aard-4-328-i004
Predictors Spring Summer Autumn Winter
M1 M2 M3 M4 M5 W1 W2 W3 W4 W5 M1 M2 M3 M4 M5 W1 W2 W3 W4 W5 M1 M2 M3 M4 M5 W1 W2 W3 W4 W5 M1 M2 M3 M4 M5 W1 W2 W3 W4 W5
A O O O O O O O O O O O O O O O O O O O O O O O O O O O O O O O O O O O O O O
T O O O O O O O O O O O O O O O O O O O O O
DT O O O O O O
MH O O O O O O O O O O O O O O O O O O
PR O O O O O O O O O O O O O O O O
HS O O O O
OZ O O O O O O O O O O O O
YS O O O O O O
PM O O O O O O O O
PL O O O O O O O O O O O O O O
FL O O O O O O O O O O O O O O O O O O O O O O O O O O O O O O O

A circle means that the predictor is significantly correlated with the corresponding group.

A, autocorrelated factor; T, mean temperature; DT, daily range; MH, minimum humidity; PR, pressure; HS, hours of sunshine; OZ, concentration of ozone; YS, presence of yellow sand; PM, concentration of yellow sand; PL, pollen; FL, proportion of flu patients; M, man; W, woman.

Table 5

Comparison of binary forecasting models for the spring season

aard-4-328-i005
Group Model Threshold Skill scores (%) Final predictors
Whole data Validation
HR POD FAR HR POD FAR
M1 REG * 79.78 47.87 34.78 47.83 32.26 23.08 A T FL D1
LOGISTIC 0.25 74.24 73.40 49.64 58.70 54.84 22.73 A FL PN
TREE 0.50 81.16 54.26 32.90 42.39 22.58 26.32 A
M2 REG * 77.01 47.71 33.33 55.43 30.36 10.53 A PR FL D1 D5 D6
LOGISTIC 0.25 70.01 72.48 49.68 63.04 62.50 27.08 A PR FL D1
TREE 0.20 76.62 61.47 40.71 39.13 00.00 ** A PR FL D1
M3 REG * 77.29 26.92 17.65 59.78 30.23 35.00 PR FL PN D1 D4
LOGISTIC 0.25 70.11 72.38 51.59 67.39 72.09 36.73 PR FL PN D1
TREE 0.25 67.12 75.24 54.60 59.78 37.21 38.46 A T FL PN D1
M4 REG * 70.65 13.79 27.27 61.95 34.29 50.00 FL D2
LOGISTIC 0.40 69.02 28.45 48.44 65.22 22.86 38.46 FL D2
TREE 0.50 72.83 18.10 19.23 65.22 22.86 38.46 FL D2
M5 REG * 74.79 27.17 48.98 55.43 37.50 51.61 A FL PN D1 D2
LOGISTIC 0.30 71.75 53.26 54.21 50.00 72.09 36.73 A FL PN D2
TREE 0.50 78.67 41.30 37.70 46.74 32.50 62.86 A FL PN
W1 REG * 78.12 41.49 38.10 50.00 19.64 8.33 A HS FL D1
LOGISTIC 0.25 75.62 74.47 47.76 67.39 60.71 19.05 A HS
TREE 0.25 84.21 70.21 30.53 52.17 37.50 30.00 A T DT HS PM D6
W2 REG * 73.13 35.54 30.65 54.35 39.34 17.24 A OZ FL D1
LOGISTIC 0.25 62.60 74.38 53.61 75.00 85.25 21.21 A FL
TREE 0.50 79.50 45.45 12.70 56.52 36.07 4.35 A T DT PM FL D6
W3 REG * 66.30 40.24 28.26 64.13 54.10 13.16 PR FL D1 D2
LOGISTIC 0.35 58.42 74.39 47.64 67.39 77.05 25.40 PR FL D1 D2
TREE 0.40 72.02 70.81 32.14 61.96 86.89 33.75 A T MINH PR FL
W4 REG * 72.55 0.99 50.00 68.48 6.67 33.33 DT FL D1 D2
LOGISTIC 0.40 59.78 60.40 63.91 52.17 80.00 61.29 DT FL D2
TREE 0.25 71.74 72.28 51.01 61.96 60.00 56.10 DT MH FL PN D1 D2
W5 REG * 68.98 22.95 39.13 52.17 47.06 61.90 A OZ FL D2
LOGISTIC 0.35 67.87 49.18 47.37 60.87 38.24 53.57 A D2
TREE 0.30 73.68 46.72 34.48 63.04 0.00 0.00 A FL PN D2 D3

A threshold cannot be used to determine the category of symptoms in the multiple regression model.

HR, hit rate; POD, probability of detection; FAR, false alarm rate; A, autocorrelated factor; T, mean temperature; DT, daily range; MH, minimum humidity; PR, pressure; HS, hours of sunshine; OZ, concentration of ozone; PL, pollen; FL, proportion of flu patients; PM, concentration of yellow sand; YS, presence of yellow sand; D, day of the week; M, man; W, woman.

*No numerical value. **There is no significant predictive factor or dummy variable denoting day of week.

Table 6

Comparison of the binary forecasting models for the summer season

aard-4-328-i006
Group Model Threshold Skill scores (%) Final predictors
Whole data Validation
HR POD FAR HR POD FAR
M1 REG * 78.75 53.17 33.00 63.04 51.52 5.56 A T L4OZ FL D1
LOGISTIC 0.25 72.29 71.43 48.28 67.39 78.79 23.53 A FL D1
TREE 0.25 72.52 72.22 48.00 42.39 19.70 00.00 A FL D1
M2 REG * 70.90 38.41 36.26 61.96 40.82 23.08 A PR D1
LOGISTIC 0.30 65.36 68.21 49.76 58.70 63.27 39.22 A PR D1
TREE 0.25 65.82 84.11 49.40 56.52 22.45 15.38 A FL PR T D1
M3 REG * 57.97 23.00 37.84 55.43 17.78 33.33 A T MH D1
LOGISTIC 0.40 52.96 78.43 50.46 57.61 64.44 44.23 T MH D1
TREE 0.50 56.14 18.14 41.27 48.91 11.11 58.33 D1
M4 REG * 69.09 00.00 ** 59.78 00.00 ** D2
LOGISTIC 0.20 30.91 0100.0 69.09 40.22 0100.0 59.78 D2
TREE 0.30 59.82 61.03 59.31 66.30 24.32 25.00 A PL D2
M5 REG * 73.67 24.81 34.69 44.57 14.04 20.00 A OZ FL D1 D2
LOGISTIC 0.25 65.36 74.42 54.93 60.87 56.14 25.58 A FL D2
TREE 0.35 68.13 73.64 52.26 45.65 17.54 23.08 A OZ FL D2
W1 REG * 75.75 44.44 38.46 36.96 10.77 0.00 A T OZ FL D1
LOGISTIC 0.25 69.98 73.02 51.06 41.30 23.08 21.05 A T FL D1
TREE 0.30 84.53 71.43 25.62 50.00 38.46 19.35 A T OZ FL D1 D3 D4
W2 REG * 69.52 16.43 39.47 46.47 5.88 25.00 A T PR FL D1
LOGISTIC 0.30 62.60 74.38 53.61 54.35 21.57 15.38 FL D1
TREE 0.30 66.74 77.86 50.90 47.83 41.18 46.15 A T PR FL D4 D5
W3 REG * 56.14 46.88 32.20 47.83 33.33 28.57 T MH PR D1 D2
LOGISTIC 0.45 60.82 92.97 39.29 66.30 95.00 32.94 PR D1 D2
TREE 0.50 69.05 92.94 32.86 68.47 91.67 30.38 A T MH PR D1 D2 D5
W4 REG * 73.67 5.13 33.33 69.57 0.00 0.00 OZ FL D2
LOGISTIC 0.25 63.10 54.62 62.43 50.00 89.29 63.24 FL D2
TREE 0.25 71.36 55.46 52.52 73.91 28.57 33.33 DT OZ FL D2 D4
W5 REG * 75.06 10.00 45.00 52.17 2.27 50.00 A FL D1 D2
LOGISTIC 0.25 62.59 60.00 64.13 48.91 31.82 54.84 A FL D2
TREE 0.30 73.21 61.82 52.11 55.43 18.18 38.46 A OZ FL D2

A threshold cannot be used to determine the category of symptoms in the multiple regression model.

HR, hit rate; POD, probability of detection; FAR, false alarm rate; A, autocorrelated factor; T, mean temperature; DT, daily range; MH, minimum humidity; PR, pressure; HS, hours of sunshine; OZ, concentration of ozone; PL, pollen; FL, proportion of flu patients; PM, concentration of yellow sand; YS, presence of yellow sand; D, day of the week; M, man; W, woman.

*No numerical value. **There is no significant predictive factor or dummy variable denoting day of week.

Table 7

Comparison of the binary forecasting models for the autumn season

aard-4-328-i007
Group Model Threshold Skill scores (%) Final predictors
Whole data Validation
HR POD FAR HR POD FAR
M1 REG * 75.13 35.90 36.36 67.90 32.35 21.43 A T L4OZ D1
LOGISTIC 0.30 75.61 65.98 47.11 49.38 11.76 73.33 A OZ PL D1
TREE 0.20 80.91 81.20 36.24 59.26 64.71 48.84 A T MH PR L4OZ PM D1
M2 REG * 80.25 43.75 25.76 77.78 34.62 10.00 A FL D1 D3~D6
LOGISTIC 0.20 72.59 67.86 49.67 70.37 65.38 46.88 FL D1
TREE 0.25 80.49 63.39 34.86 44.44 84.62 65.08 A MH FL D1
M3 REG * 82.47 45.79 20.97 70.37 20.00 66.67 YS FL D1 D2
LOGISTIC 0.20 74.07 77.57 49.39 51.85 65.00 71.11 FL D1
TREE 0.20 82.47 75.70 35.71 75.31 0.00 0.00 A MH FL D1
M4 REG * 70.86 8.26 41.18 67.90 0.00 100.00 PR OZ D1 D2
LOGISTIC 0.30 65.68 52.07 56.25 69.14 60.00 50.00 PR D1 D2
TREE 0.25 62.72 76.86 56.94 ** ** ** PR OZ D1 D2 D6
M5 REG * 73.12 1.73 41.67 43.21 0.00 0.00 A FL PL D1 D2
LOGISTIC 0.25 65.31 56.67 63.57 54.32 30.43 26.32 A PL D2
TREE 0.20 79.40 73.39 39.85 46.91 26.09 42.86 A T FL PL D1 D4
W1 REG * 79.65 41.38 21.31 71.60 36.67 26.67 A T OZ D1
LOGISTIC 0.30 74.37 70.69 45.33 70.37 63.33 40.63 A T OZ
TREE 0.30 83.17 71.55 29.06 65.43 40.00 45.45 A T MH PR OZ
W2 REG * 71.85 38.17 39.76 54.32 16.28 12.50 FL D1
LOGISTIC 0.25 54.81 81.68 59.77 45.68 65.12 50.88 A FL
TREE 0.30 66.42 83.21 51.12 49.38 34.88 46.43 A MH FL PL D1 D4 D5
W3 REG * 75.56 18.92 30.00 61.73 0.00 0.00 FL D1
LOGISTIC 0.25 74.81 57.66 46.22 64.20 22.58 41.67 FL D1 D3
TREE 0.20 77.28 50.45 39.78 64.20 22.58 41.67 A MH FL D1 D3
W4 REG * 72.10 8.62 41.18 69.14 3.85 0.00 PR FL D1 D2
LOGISTIC 0.25 64.69 61.21 57.99 72.84 61.54 42.86 PM FL D1 D2
TREE 0.25 63.70 79.31 57.21 67.90 0.00 0.00 A DT PR PM FL D1 D2
W5 REG * 73.33 2.83 62.50 46.91 0.00 0.00 FL D2 D3
LOGISTIC 0.35 70.37 48.11 56.03 54.32 20.93 25.00 D2 D3
TREE 0.35 76.79 52.83 44.00 51.85 13.95 25.00 A FL PL D2 D3

A threshold cannot be used to determine the category of symptoms in the multiple regression model.

HR, hit rate; POD, probability of detection; FAR, false alarm rate; A, autocorrelated factor; T, mean temperature; DT, daily range; MH, minimum humidity; PR, pressure; HS, hours of sunshine; OZ, concentration of ozone; PL, pollen; FL, proportion of flu patients; PM, concentration of yellow sand; YS, presence of yellow sand; D, day of the week; M, man; W, woman.

*No numerical value. **There is no significant predictive factor or dummy variable denoting day of week.

Table 8

Comparison of the binary forecasting models for the winter season.

aard-4-328-i008
Group Model Threshold Skill scores (%) Final predictors
Whole data Validation
HR POD FAR HR POD FAR
M1 REG * 78.59 41.23 33.80 66.00 42.11 42.86 A FL D1 D3 D4 D5 D6
LOGISTIC 0.30 75.53 65.79 46.43 69.00 63.16 41.46 A FL D1
TREE 0.25 82.59 60.53 29.59 71.00 39.47 28.57 A T MH FL PL D1 D2
M2 REG * 71.29 28.57 41.54 62.00 37.50 60.00 A FL D1
LOGISTIC 0.30 68.52 61.31 49.70 57.00 56.25 67.27 FL D1
TREE 0.30 78.47 58.39 31.03 68.00 0.00 0.00 A T OZ FL PL D1 D5
M3 REG * 72.69 31.16 34.85 75.00 47.22 26.09 YS FL D1 D4
LOGISTIC 0.30 71.06 67.39 46.24 73.00 86.11 41.51 YS FL D1
TREE 0.30 76.62 71.01 38.36 76.00 69.44 34.21 A OZ YS FL D1 D4 D5
M4 REG * 72.92 6.56 27.27 68.00 5.88 0.00 PR OZ D2
LOGISTIC 0.25 50.00 72.13 67.41 70.00 61.76 44.74 PR D2
TREE 0.25 53.47 93.44 62.87 63.00 64.71 53.19 A T MH PR HS OZ D2 D3
M5 REG * 71.53 19.05 44.19 64.00 12.50 16.67 A HS D1 D2
LOGISTIC 0.25 60.71 79.37 58.51 71.00 70.00 37.78 A HS D1 D2
TREE 0.25 68.71 78.57 51.71 60.00 0.00 0.00 A T PR HS OZ D1 D2
W1 REG * 75.53 23.36 46.81 59.00 18.75 71.43 A T MH FL D1
LOGISTIC 0.30 74.59 57.01 50.41 69.00 71.88 48.89 A T D1
TREE 0.25 83.77 70.09 33.04 63.00 46.88 57.14 A T PR PM FL
W2 REG * 67.82 17.48 45.65 57.00 26.67 72.41 FL D1
LOGISTIC 0.30 67.59 61.54 49.13 51.00 70.00 65.57 FL D1
TREE 0.30 75.93 63.64 36.36 70.00 0.00 0.00 A MH FL D1 D2 D3 D6
W3 REG * 72.22 20.00 22.22 80.00 46.43 27.78 FL D4
LOGISTIC 0.35 69.91 50.00 46.15 52.00 75.00 66.13 FL
TREE 0.40 75.46 52.14 34.82 77.00 67.86 42.42 A FL D4
W4 REG * 71.99 0.84 75.00 66.00 8.70 86.67 FL D1 D2
LOGISTIC 0.35 66.20 40.34 60.98 59.00 21.74 82.14 D1 D2
TREE 0.25 71.53 63.87 51.28 77.00 0.00 0.00 A FL D1 D3 D4 D5
W5 REG * 73.41 3.54 50.00 73.00 3.70 50.00 A PR D1 D2
LOGISTIC 0.35 67.36 42.61 60.48 71.00 22.22 57.14 D2 D4
TREE 0.30 56.71 72.17 65.13 54.00 48.15 71.11 A MH PR D2

A threshold cannot be used to determine the category of symptoms in the multiple regression model.

HR, hit rate; POD, probability of detection; FAR, false alarm rate; A, autocorrelated factor; T, mean temperature; DT, daily range; MH, minimum humidity; PR, pressure; HS, hours of sunshine; OZ, concentration of ozone; PL, pollen; FL, proportion of flu patients; PM, concentration of yellow sand; YS, presence of yellow sand; D, day of the week; M, man; W, woman.

*No numerical value.

Table 9

Proposed models and thresholds for binary asthma forecasting

aard-4-328-i009
Group Spring Summer Autumn Winter
Model Threshold Model Threshold Model Threshold Model Threshold
M1 LOGISTIC 0.25 LOGISTIC 0.25 TREE 0.20 LOGISTIC 0.30
M2 LOGISTIC 0.25 LOGISTIC 0.30 LOGISTIC 0.20 LOGISTIC 0.30
M3 LOGISTIC 0.25 LOGISTIC 0.40 LOGISTIC 0.20 TREE 0.30
M4 TREE 0.40 TREE 0.30 LOGISTIC 0.30 TREE 0.25
M5 LOGISTIC 0.30 LOGISTIC 0.25 TREE 0.20 LOGISTIC 0.25
W1 LOGISTIC 0.25 TREE 0.30 LOGISTIC 0.30 TREE 0.25
W2 LOGISTIC 0.25 TREE 0.30 TREE 0.30 LOGISTIC 0.30
W3 TREE 0.40 TREE 0.50 TREE 0.20 TREE 0.40
W4 TREE 0.25 TREE 0.25 LOGISTIC 0.25 LOGISTIC 0.35
W5 LOGISTIC 0.35 TREE 0.30 TREE 0.35 LOGISTIC 0.35

M, man; W, woman.

Notes

This study was supported by the grant of Korean Centers for Disease Control and Prevention (2011E3302300).

References

1. Fanta CH. Asthma. N Engl J Med. 2009; 360:1002–1014.
crossref
2. Braman SS. The global burden of asthma. Chest. 2006; 130:1 Suppl. 4S–12S.
crossref
3. Kroegel C. Global Initiative for Asthma (GINA) guidelines: 15 years of application. Expert Rev Clin Immunol. 2009; 5:239–249.
crossref
4. Masoli M, Fabian D, Holt S, Beasley R;. GINA) Program. The global burden of asthma: executive summary of the GINA Dissemination Committee report. Allergy. 2004; 59:469–478.
crossref
5. Choi JH, Cha YM. The prevalence of allergic diseases among Korean children. Korean J Pediatr. 1964; 7:39–41.
6. Lee HR, Hong DS, Sohn KC. Survey of allergic diseases in children. J Korean Med Assoc. 1983; 26:254–262.
7. Shin TS, Lee GJ, Yoon HS. A survey of the distribution of allergic diseases in primary school children. Korean J Asthma Allergy Clin Immunol. 1990; 10:201–212.
8. Kim YK, Kim SH, Tak YJ, Jee YK, Lee BJ, Kim SH, et al. High prevalence of current asthma and active smoking effect among the elderly. Clin Exp Allergy. 2002; 32:1706–1712.
crossref
9. The Global Initiative for Asthma. The Global Asthma Report 2014. Global burden of disease due to asthma [Internet]. The Global Initiative for Asthma;2015 Sep 15. Available from: http://www.globalasthmareport.org/burden/burden.php.
10. Beasley R. The burden of asthma with specific reference to the United States. J Allergy Clin Immunol. 2002; 109:5 Suppl. S482–S489.
crossref
11. Kim CY, Park HW, Ko SK, Chang SI, Moon HB, Kim YY, et al. The financial burden of asthma: a nationwide comprehensive survey conducted in the republic of Korea. Allergy Asthma Immunol Res. 2011; 3:34–38.
crossref
12. A fresh perspective on asthma. Nat Me. 2012; 18:631.
13. Pedersen SE, Hurd SS, Lemanske RF Jr, Becker A, Zar HJ, Sly PD, et al. Global strategy for the diagnosis and management of asthma in children 5 years and younger. Pediatr Pulmonol. 2011; 46:1–17.
crossref
14. O'Connor GT. Allergen avoidance in asthma: what do we do now. J Allergy Clin Immunol. 2005; 116:26–30.
15. Soyiri IN, Reidpath DD. Evolving forecasting classifications and applications in health forecasting. Int J Gen Med. 2012; 5:381–389.
crossref
16. Tobías A, Sáez M, Galán I, Campbell MJ. Sensitivity analysis of common statistical models used to study the short-term effects of air pollution on health. Int J Biometeorol. 2003; 47:227–229.
crossref
17. Ivey MA, Simeon DT, Monteil MA. Climatic variables are associated with seasonal acute asthma admissions to accident and emergency room facilities in Trinidad, West Indies. Clin Exp Allergy. 2003; 33:1526–1530.
crossref
18. Chavarría JF. Short report: Asthma admissions and weather conditions in Costa Rica. Arch Dis Child. 2001; 84:514–515.
crossref
19. Carey MJ, Cordon I. Asthma and climatic conditions: experience from Bermuda, an isolated island community. Br Med J (Clin Res Ed). 1986; 293:843–844.
crossref
20. Khot A, Burn R, Evans N, Lenney W, Storr J. Biometeorological triggers in childhood asthma. Clin Allergy. 1988; 18:351–358.
crossref
21. Garty BZ, Kosman E, Ganor E, Berger V, Garty L, Wietzen T, et al. Emergency room visits of asthmatic children, relation to air pollution, weather, and airborne allergens. Ann Allergy Asthma Immunol. 1998; 81:563–570.
crossref
22. Rosas I, McCartney HA, Payne RW, Calderón C, Lacey J, Chapela R, et al. Analysis of the relationships between environmental factors (aeroallergens, air pollution, and weather) and asthma emergency admissions to a hospital in Mexico City. Allergy. 1998; 53:394–401.
crossref
23. Koh YI, Choi IS. Seasonal difference in the occurrence of exercise-induced bronchospasm in asthmatics: dependence on humidity. Respiration. 2002; 69:38–45.
crossref
24. Gioulekas D, Balafoutis C, Damialis A, Papakosta D, Gioulekas G, Patakas D. Fifteen years' record of airborne allergenic pollen and meteorological parameters in Thessaloniki, Greece. Int J Biometeorol. 2004; 48:128–136.
crossref
25. Jalaludin BB, O'Toole BI, Leeder SR. Acute effects of urban ambient air pollution on respiratory symptoms, asthma medication use, and doctor visits for asthma in a cohort of Australian children. Environ Res. 2004; 95:32–42.
crossref
26. Holmén A, Blomqvist J, Frindberg H, Johnelius Y, Eriksson NE, Henricson KA, et al. Frequency of patients with acute asthma in relation to ozone, nitrogen dioxide, other pollutants of ambient air and meteorological observations. Int Arch Occup Environ Health. 1997; 69:317–322.
crossref
27. Anderson HR, Ponce de Leon A, Bland JM, Bower JS, Emberlin J, Strachan DP. Air pollution, pollens, and daily admissions for asthma in London 1987-92. Thorax. 1998; 53:842–848.
crossref
28. Johnston SL, Pattemore PK, Sanderson G, Smith S, Campbell MJ, Josephs LK, et al. The relationship between upper respiratory infections and hospital admissions for asthma: a time-trend analysis. Am J Respir Crit Care Med. 1996; 154(3 Pt 1):654–660.
crossref
29. Hersoug LG. Viruses as the causative agent related to 'dampness' and the missing link between allergen exposure and onset of allergic disease. Indoor Air. 2005; 15:363–366.
crossref
30. Maheswaran R, Pearson T, Hoysal N, Campbell MJ. Evaluation of the impact of a health forecast alert service on admissions for chronic obstructive pulmonary disease in Bradford and Airedale. J Public Health (Oxf). 2010; 32:97–102.
crossref
31. Bakerly ND, Roberts JA, Thomson AR, Dyer M. The effect of COPD health forecasting on hospitalisation and health care utilisation in patients with mild-to-moderate COPD. Chron Respir Dis. 2011; 8:5–9.
crossref
32. Soyiri IN, Reidpath DD, Sarran C. Forecasting peak asthma admissions in London: an application of quantile regression models. Int J Biometeorol. 2013; 57:569–578.
crossref
33. Moustris KP, Douros K, Nastos PT, Larissi IK, Anthracopoulos MB, Paliatsos AG, et al. Seven-days-ahead forecasting of childhood asthma admissions using artificial neural networks in Athens, Greece. Int J Environ Health Res. 2012; 22:93–104.
crossref
TOOLS
ORCID iDs

Jae-Won Oh
https://orcid.org/http://orcid.org/0000-0003-2714-0065

Similar articles