Good evening everyone,
I am currently conducting a clinical trial where I am taking an intervention population (which is referred to a new service) and comparing hospital readmission rates to historical controls. Because the intervention group will be more likely to meet the primary outcome (they are referred because they are at higher risk of readmission than the average population), I have decided to match historical controls to the intervention group based on these three criteria (in order of preference):
My data is stored in an Excel file, and I have created a column to differentiate between intervention patients (1) and historical controls (0). From my limited knowledge on this, I am under the impression that I should attempt to select patients from the historical control pool (~7000 patients) in order to create similar averages and modes for each of the criteria listed above, and that I should select approximately the same number of patients out of the historical control pool as is present in the intervention group to achieve this.
My question is, how exactly would I go about doing this? I was thinking that I should first select patients with similar average/mode as the highest priority criteria (Comorbidity index), but what should I do if I find that, say, Comorbidity index is similar but Age is different among groups?
Any help with this matter would be greatly appreciated. Thanks in advance for your time,
cardgage
I am currently conducting a clinical trial where I am taking an intervention population (which is referred to a new service) and comparing hospital readmission rates to historical controls. Because the intervention group will be more likely to meet the primary outcome (they are referred because they are at higher risk of readmission than the average population), I have decided to match historical controls to the intervention group based on these three criteria (in order of preference):
- Comorbidity index
- Discharge season
- Age
My data is stored in an Excel file, and I have created a column to differentiate between intervention patients (1) and historical controls (0). From my limited knowledge on this, I am under the impression that I should attempt to select patients from the historical control pool (~7000 patients) in order to create similar averages and modes for each of the criteria listed above, and that I should select approximately the same number of patients out of the historical control pool as is present in the intervention group to achieve this.
My question is, how exactly would I go about doing this? I was thinking that I should first select patients with similar average/mode as the highest priority criteria (Comorbidity index), but what should I do if I find that, say, Comorbidity index is similar but Age is different among groups?
Any help with this matter would be greatly appreciated. Thanks in advance for your time,
cardgage