Trend analysis

[j0hnj0hn]

In a period of 14 years we collected MRIs on different patients. For all the MRIs I have a variable on which day after diagnosis the MRI was made. I want to know if MRIs were made much faster at the end of the study than at the beginning --> so I want to know if there's a trend. How would one do this in SPSS?

 

[GretaGarbo]

Can you plot the curve? delay vs time point.

 

[j0hnj0hn]

something like this?

 

[hlsmith]

Yes, but I imagine you had more than 30ish MRIs performed over 14 year period? Are these the raw delay values for each MRI conducted? Also, you would want to confirm the high plot was really an outlier and not a calculation or data recording error.

Next you would look to plot a line to these values, which appear pretty stagnate per the naked eye. But watch out for the outlier's influence. What are the units on the y-axis (minutes, days, etc.)?

 

[j0hnj0hn]

Yes, but I imagine you had more than 30ish MRIs performed over 14 year period? Are these the raw delay values for each MRI conducted? Also, you would want to confirm the high plot was really an outlier and not a calculation or data recording error.

Next you would look to plot a line to these values, which appear pretty stagnate per the naked eye. But watch out for the outlier's influence. What are the units on the y-axis (minutes, days, etc.)?

It's a very rare disease with indeed 30 MRIs done (out of 50 patients). The outlier had CTscans performed first, MRI was done at a later stage.
These are indeed raw values. Units on y-axis are days.

 

[j0hnj0hn]

This is line I get when I choose the option on SPSS. Is it enough to assume based of off this that there is no trend? Or should I run additional tests?

 

[GretaGarbo]

Is it enough to assume based of off this that there is no trend?

No trend!

 

[hlsmith]

I want to know if MRIs were made much faster at the end of the study than at the beginning

What study, did you actually intervene some how or what would have driven clinicians to order MRIs faster? You may want to rerun the model without the outlier just to make sure the outlier isn't pulling the line up, but likely you won't see a difference.

 

[j0hnj0hn]

What study, did you actually intervene some how or what would have driven clinicians to order MRIs faster? You may want to rerun the model without the outlier just to make sure the outlier isn't pulling the line up, but likely you won't see a difference.

Our study was about radiological (ultrasound and MRI) abnormalities in GBS meningitis and its relation to neurodevelopmental development. Reasons to order MRIs faster would be change in protocol, more experienced radiologist (in reviewing MRIs) and reduced costs of making MRIs.

 

[hlsmith]

When was the protocol implemented? if it was within the about timeline you should add a vertical reference line, to distinguish the control and intervention periods.

Were there different variable to control for besides the passing of time that may influence the MRI readings/diagnosis that need to be controlled (e.g., provider, day of the week, etc.)?