## [1] "C:/Google Drive/My Research/matassa_lipids/matassa_lipids/data/aa"

Intro

Results might change a little bit as we clean data more but should not change too much and I assume that we our statistical significance / insignificance will remain where it is now.

Figure 0

## [1] "2020-01-01"

Figure 1

For now it is in the format that we had in our manuscript, I will likely change it so the bars PRE and POST intervention are by the side with p values above them on the graph.

Table 1 (by incitation)

We see a nice statistical significance in ASCVD-risk only, which is that our intervention was targeted. Of course it is only association likely not causation (likely not caused by intervention), but this is a nice result and I think it is good enough for an abstrace / QA/QI manuscript.

Figure 2 (all)

Table 2 (all)

Big table pre VS post

This compares directly pre intervention to post intervention group so shows how well the randimization go, looks like preety well. And Giselle also mentioned something about how lipid testing was improved by intervention?

Big table only ASCVD

MVA on OLD patients (pre intervention)

There are tiny differences between this and the manuscript result, I believe that the reason is that in manuscript he had 464 and here I am getting 463 patients pre-intervension, I would just ignore it.

MVA on NEW patients (post intervention)

I had to remove CKD because or model gets wild secondary to relatively smaller n than before. As expected we have much wider confidence intervals 2/2 smaller n.

MVA on ALL patients (pre and post intervention)

Misings / outliers

To fix

Checks of exclusion

These are not yet excluded:

Already excluded (from manuscript times + 3 current suggested by Sana):

Compare with manucrtipt

For quality control