Table 5.
Comparison of Asian and White groups in Leicester city in 2011–2015, adjusted for age (results 2011–2017)
| Year | Odds ratio (95% CI) for ethnicity 1/odds ratio |
P-value for ethnicity | Odds ratio (95% CI) for age |
P-value for age |
Pseudo r2 |
|---|---|---|---|---|---|
| 2011 | 1.13 (0.80–1.60) 0.93 |
0.48 | 2.92 (2.57–3.32) | <0.001 | 0.170 |
| 2012 | 0.87 (0.62–1.22) 1.15 |
0.43 | 2.03 (1.83–2.24) | <0.001 | 0.096 |
| 2013 | 0.80 (0.63–1.03) 1.25 |
0.06 | 2.68 (2.46–2.92) | <0.001 | 0.171 |
| 2014 | 0.72 (0.60–0.86) 1.37 |
<0.001 | 1.74 (1.65–1.83) | <0.001 | 0.077 |
| 2015 | 0.51 (0.43–0.60) 1.92 |
<0.001 | 2.19 (2.07–2.31) | <0.001 | 0.138 |
| 2016 | 0.67 (0.56–0.80) 1.49 |
<0.001 | 1.80 (1.71–1.90) | <0.001 | 0.085 |
| 2017 | 0.61 (0.48–0.76) 1.64 |
0.001 | 1.77 (1.65–1.89) | <0.001 | 0.074 |
Reference category for ethnicity is Asian (coded as 0). Age is modelled as a linear and continuous variable, so for 2011, for every rise in age category (5-year bands), the odds of being referred increase by a factor of nearly 3. After Bonferroni correction for multiple comparisons (0.05/14), the adjusted significance level becomes 0.003, so all results for 2014–2017 are significant.