Exercise: Caesarean section and obesity, 2

Question 2: Is there any evidence that previous vaginal delivery is related to the risk of a caesarean?

Suggested answer

Previous vaginal delivery is a qualitative, dichotomous variable so we can do a chi-squared test or Fisher's exact test.

We get the following output:

Previous vaginal delivery * Caesarian section Crosstabulation
  Caesarian section Total
No Yes
Previous vaginal delivery No 427 122 549
Yes 274 14 288
Total 701 136 837

Chi-Square Tests
  Value df Asymp. Sig.
Exact Sig.
Exact Sig.
Pearson Chi-Square 41.840a 1 .000
Continuity Correctionb 40.574   1 .000
Likelihood Ratio 49.272   1 .000
Fisher's Exact Test       .000 .000
Linear-by-Linear Association 41.790   1 .000
N of Valid Cases 837  
a. 0 cells (.0%) have expected count less than 5. The minimum expected count is 46.80.
b. Computed only for a 2x2 table

The chi-squared test is valid, because the expected frequencies are all greater than five. The relationship is highly significant (chi-squared = 41.84, d.f. = 1, P < 0.001). Hence there is very strong evidence that caesarean section is related to previous vaginal delivery.

We can also find the odds ratio by choosing Risk from Statistics in Crosstabs. We get:

Risk Estimate
    Value   95% Confidence
Lower Upper
Odds Ratio for Previous
vaginal delivery (No / Yes)
.179 .101 .317
For cohort Caesarian
section = No
.818 .776 .861
For cohort Caesarian
section = Yes
4.571 2.679 7.800
N of Valid Cases 837

Hence the odds ratio is 0.18 with 95% confidence interval 0.10 to 0.32.

Back to Exercise: Caesarean section and obesity.

To Applied Biostatistics index.

To Martin Bland's M.Sc. index.

To Martin Bland's home page.

This page maintained by Martin Bland.
Last updated: 21 February, 2012.

Back to top.