Table 2.
Logistic regression model.
| Variable | OR | 95%CI | p-value | |
|---|---|---|---|---|
| Lower | Upper | |||
| Age | ||||
| Years | 1.016 | 0.992 | 1.040 | 0.196 |
| Sex | ||||
| Female | 1 | |||
| Male | 1.455 | 0.856 | 2.472 | 0.166 |
| Ethnicity | ||||
| White | 1 | |||
| Asian | 0.956 | 0.542 | 1.687 | 0.877 |
| Black | 2.433 | 0.072 | 8.241 | 0.153 |
| Mixed | 1.850 | 0.411 | 8.330 | 0.423 |
| Other | 2.819 | 0.539 | 14.750 | 0.220 |
| Not stated | 2.948 | 0.563 | 15.436 | 0.201 |
This model analysed the relationship between age, sex and ethnicity on the likelihood of a participant being prescribed a drug for psychosis with CYP2D6 PGx association or not (outcome variable: (0) ‘CYP2D6-PGx antipsychotic’; (1) ‘non-CYP2D6-PGx antipsychotic’). Adjusted odds ratio (OR) has been reported which shows the strength of association between the variable and outcome: OR >1 = stronger association between variable and outcome; OR <1 = weaker association between variable and outcome. For age, the OR represents the change in the strength of association for every one age year gained. For sex and ethnicity, the OR shows the strength of association in comparison to the reference category, which for sex is ‘female’ and for ethnicity is ‘White’. The 95% confidence intervals (CI) show the range in estimates for the OR and because the range between lower and upper CI is above and below 1, the results are not statistically significant.