Hi,
I have a question regarding which statistical test to use. I am comparing patients with and without a specific disease. I am investigating the impact of a specific intervention (which will make the patients with the disease experiencing more symptoms).
I am attaching a graph that show the proportion of patients with and without the disease and the proportion of these two groups experiencing any of 15 related symptoms. If 1/15 symptoms is reported the patient is classified as experiencing symptoms. The same applies if the patient reports 15/15 sympotoms. I.e. it does not matter how many of the 15 symptoms that are reported a specific month. One is enough to classify that patient as experiencing symptoms.
I want to compare these two groups:
1. Is the proportion of patients experiencing symptoms significantly different between the two groups at month -24, -23, -22 ... +1, +2 and so on ?
2. It is obvious from looking at the attached graph that patients with the disease have a higher proportion of experienced symptoms, both at baseline before intervention and after intervention. But, does the two groups increase their proportion in the same degree? I.e are the groups increasing by the same factor after intervention and returning to "baseline" at the same time after intervention?
3. Is there a overall difference over the whole time period between the two groups?
The attached graph is showing (somewhat exaggarated) differences between the two groups over time, just to make my question clearer (hopefully). Dotted line is time of intervention. Patients with the disease is experiencing higher proportions over all time periods.
Best regards,
Jesper Eriksson
Ps! How much I have tried to explain myself clear in this forum, I have so far never succeded. Looking forward to supplementary questions!
I have a question regarding which statistical test to use. I am comparing patients with and without a specific disease. I am investigating the impact of a specific intervention (which will make the patients with the disease experiencing more symptoms).
I am attaching a graph that show the proportion of patients with and without the disease and the proportion of these two groups experiencing any of 15 related symptoms. If 1/15 symptoms is reported the patient is classified as experiencing symptoms. The same applies if the patient reports 15/15 sympotoms. I.e. it does not matter how many of the 15 symptoms that are reported a specific month. One is enough to classify that patient as experiencing symptoms.
I want to compare these two groups:
1. Is the proportion of patients experiencing symptoms significantly different between the two groups at month -24, -23, -22 ... +1, +2 and so on ?
2. It is obvious from looking at the attached graph that patients with the disease have a higher proportion of experienced symptoms, both at baseline before intervention and after intervention. But, does the two groups increase their proportion in the same degree? I.e are the groups increasing by the same factor after intervention and returning to "baseline" at the same time after intervention?
3. Is there a overall difference over the whole time period between the two groups?
The attached graph is showing (somewhat exaggarated) differences between the two groups over time, just to make my question clearer (hopefully). Dotted line is time of intervention. Patients with the disease is experiencing higher proportions over all time periods.
Best regards,
Jesper Eriksson
Ps! How much I have tried to explain myself clear in this forum, I have so far never succeded. Looking forward to supplementary questions!
Attached Files
