How can I define the preferred dataset when two sources don't agree?

[DSThayer]

In some cases, I am finding pregnancies that have a birth date in the birth registry (a survey in the CDM), as well as a delivery procedure code in hospital data. The dates differ by a few days.

When this occurs, we need to pick the birth registry date as the correct date. However, it is selecting the procedure date. Is there a way to set parameters so that the birth registry will be considered the gold standard?

[GiorgioLimoncella]

It is possible to make a DAP-specific hierarchy, but this should not be the case because the birth registry should already prevail in the hierarchy.

here is the hierarchy:

I saw that in github the files with the parameters for your datasource are missing, could you please send them so I can check and add them?

They would be:

1. 00_parameters_pregnancy.R (in p_parameters_pregnancy)
2. 01_prompts_USWAN.R (in p_parameters_pregnancy/01_prompts)
3. 02_itemsets_USWAN.R (in p_parameters_pregnancy/02_itemsets)
4. 03_conceptsets_USWAN.R (in p_parameters_pregnancy/03_conceptsets)

[RosaMGini]

@GiorgioLimoncella this is the first time we run the algorithm on a data source having EUROCAT, is it possible that this triggered a bug that messed with the hierarchy?

[GiorgioLimoncella]

yes it is possible! Thanks for the suggestion, I will check it out

[clobart]

There was a bug in the creation of the sample, but the content is correct!