Skip to main content

Table 2 Associations between migration status and risk of ASD

From: Migration modulates the prevalence of ASD and ADHD: a systematic review and meta-analysis

Groups

No. studies

No. participants

Pooled ORs (95% CIs)

P for Z test

Heterogeneity

P for interaction

I2 (%)

P

Overall migration

13

6,532,546

1.32 (1.07, 1.63)

0.010

87.6

< 0.001

 

Diagnostic subgroup

 Autism

7

6,220,489

1.26 (1.06, 1.50)

0.010

57.0

0.030

0.583

 Autism spectrum disorder

6

312,057

1.46 (0.89, 2.40)

0.133

91.4

< 0.001

 

Comorbid intellectual disability

      

0.006

 Yes

4

2,435,381

1.21 (1.09, 1.35)

0.001

4.6

0.370

 

 No

3

809,027

0.68 (0.46, 1.01)

0.057

84.0

0.002

 

Parental country of birth

      

0.514

 Foreign-born mother

11

6,334,601

1.49 (1.19, 1.87)

0.001

81.9

< 0.001

 

 Foreign-born father

2

10,850

1.19 (0.86, 1.65)

0.302

0.0

0.617

 

 Both parents foreign-born

2

49,610

1.24 (0.63, 2.43)

0.541

90.9

0.001

 

Migration type a

      

0.016

 Africa to Europe

4

50,005

2.72 (1.25, 5.89)

0.011

93.6

< 0.001

 

 Americas to Europe

4

50,005

4.13 (1.26, 13.50)

0.019

96.6

< 0.001

 

 Eastern Mediterranean to Europe

2

49,610

0.88 (0.61, 1.26)

0.489

0.0

0.867

 

 Europe to Europe

5

993,669

1.20 (0.97, 1.50)

0.097

41.7

0.143

 

 South-East Asia to Europe

3

44,358

1.80 (0.71, 4.56)

0.213

92.7

< 0.001

 

 Western Pacific to Europe

1

43,963

0.90 (0.60, 1.30)

–

–

–

 

NOS score

      

0.739

  < 7

1

 

1.40 (1.00, 1.90)

–

–

–

 

  ≥ 7

12

 

1.31 (1.05, 1.64)

0.017

88.0

< 0.001

 

Influence analyses b

 Minimal

–

6,463,332

1.24 (1.01, 1.51)

0.036

84.4

< 0.001

 

 Maximal

–

6,425,593

1.39 (1.11, 1.73)

0.004

88.3

< 0.001

 
  1. aThe number of participants was calculated based on the total population of included studies in each group
  2. bInfluence analysis was conducted by eliminating one study at a time; excluded study by Haglund et al. [47] for minimal pooled ORs, and excluded study by van der Ven et al. [22] for maximal pooled ORs
  3. Abbreviations: CI Confidence interval, NOS Newcastle-Ottawa scale, OR Odds ratio