Skip to main content

Table 3 Ten most frequently used methods to analyze each dataset

From: Developing a healthcare dataset information resource (DIR) based on Semantic Web

Dataset

Methods

    

NHANES

EM algorithm

Neural network model

Wilcoxon signed-rank test

Poisson regression

Chi-squared test

 

29.55%

19.63%

16.69%

15.02%

14.85%

 

Kruskal-Wallis test

Logistic regression

Log-rank test

Linear regression

T-test

 

14.32%

12.56%

12.17%

10.04%

8.51%

SEER-medicare

Chi-squared test

Logistic regression

Cox regression

Log-rank test

Survival analysis

 

54.52%

50.83%

39.64%

17.46%

14.87%

 

T-test

Regression model

Kaplan-Meier survival estimates

Linear regression

Propensity score matching

 

11.12%

10.45%

9.34%

8.85%

7.01%

Add health

Logistic regression

Chi-squared test

Linear regression

Regression model

Principal component analysis

 

50.00%

33.17%

13.13%

9.82%

8.07%

 

ANOVA

Poisson regression

T-test

Propensity score matching

Cox regression

 

7.49%

5.74%

5.06%

3.40%

3.40%

HCUP

Logistic regression

Chi-squared test

Linear regression

T-test

Regression model

 

57.91%

48.44%

20.24%

18.03%

15.61%

 

ANOVA

Poisson regression

Cox regression

Mann-Whitney U test

Bootstrap

 

9.87%

9.06%

7.45%

7.35%

4.23%

MDS

Logistic regression

Chi-squared test

Linear regression

Regression model

T-test

 

42.12%

39.73%

17.29%

14.90%

13.53%

 

ANOVA

Cox regression

Mann-Whitney U test

Bootstrap

Survival analysis

 

13.18%

9.93%

7.19%

4.11%

3.77%

CPRD

Logistic regression

Cox regression

Chi-squared test

Poisson regression

Propensity score matching

 

42.35%

31.03%

18.87%

12.37%

10.48%

 

Linear regression

Regression model

Survival analysis

T-test

Kaplan-Meier survival estimates

 

9.85%

8.60%

6.08%

5.66%

4.61%

MarketScan

Chi-squared test

Logistic regression

Cox regression

T-test

Poisson regression

 

[-2pt]47.88%

43.32%

19.22%

12.87%

12.21%

 

Propensity score matching

Linear regression

Regression model

ANOVA

Fisher’s exact test

 

10.91%

9.93%

9.77%

6.68%

5.86%

THIN

Logistic regression

Cox regression

Chi-squared test

Poisson regression

Regression model

 

37.33%

26.04%

23.27%

12.44%

9.91%

 

Inverse probability weighting

Linear regression

T-test

Survival analysis

Propensity score matching

 

8.99%

8.53%

8.06%

6.91%

6.68%

MIMIC

Logistic regression

Chi-squared test

T-test

Mann-Whitney U test

Regression model

 

45.39%

20.39%

17.76%

15.79%

14.47%

 

Support vector machine

Linear regression

Cox regression

Kolmogorov-Smirnov test

K-nearest neighbors

 

14.47%

11.84%

11.18%

9.87%

9.21%

Premier

Chi-squared test

K-means

Decision tree model

Logistic regression

Propensity score matching

 

41.05%

38.95%

27.37%

21.05%

14.74%

 

Kruskal-Wallis test

Linear discriminant analysis

Regression model

Linear regression

T-test

 

13.68%

11.58%

11.58%

8.42%

8.42%

Clinformatics

Linear regression

Bootstrap

Regression model

Kruskal-Wallis test

Chi-squared test

 

44.90%

28.57%

20.41%

14.29%

12.24%

 

F-test

Cox regression

Logistic regression

ANOVA

Survival analysis

 

12.24%

10.20%

10.20%

8.16%

6.12%

Humedica

Chi-squared test

Logistic regression

Bootstrap

Fisher’s exact test

Cox regression

 

33.33%

22.22%

22.22%

22.22%

11.11%

 

T-test

Linear regression

Propensity score matching

Survival analysis

Ensemble learning

 

11.11%

11.11%

11.11%

11.11%

11.11%