Missing rows

[drohm002]

Hi,

My main table has 75,000 rows, one for each patient. I work in healthcare.  Another table has about 72,000 rows, one for each patient that matches the criteria built for that specific table, its filtered down a little.  When I load these two tables into my app, I only get the 72,000 patients to show, but I want to show all of the 75,000 patients.  I feel like i am missing something small somewhere for this to occur, does anyone have any idea?

When I click "Load Data" i can see that the app is fetching 75,000 rows for the first table, then it goes and fetches 72,000 rows for the next table.  These two tables have to be built seperately like I have them.  I dont want rows to drop off from the first table just because a patient/record doesnt exist in the second smaller table. Please help! Thank you

[Pablo007]

As I see it

There are basically three tables, Let's put for example:

Table 1 has 100 Rows

Table 2 has 75 rows

Table 3 has 75 Rows

The three tables have a common element with which they are related in their Model: “Most Recent Patient Encounter” Qlik will only show you from those three tables the rows that share that common element, because otherwise it will not find data to show results if they are not in all three.

In example, if I have a patient in the Most recent Visit table (is it a date field?), who has data in Epic Prayor but does not have data in Epic Results abstract, that row will not appear.

Anyway, I also think that the data will not be correct, since the union field appears to be a date field, not a unique identification field.

If two EMPIs have the same Most recent Patient encounter, perhaps the association is not correct because how will it know which data corresponds to each EMPI.

The solution is simple, you have to try to unify these three data sources into one. Or through a previous script we generate the TEXT EMPI field in the Epic Prayor tables, taking as master data the table that contains the data, and when there are no values ​​in that table, generate a Row That is

TEXT EMPI – HEALTH PLAN FLAG – %HEALTHPLAN

XXXXXXXX -        NONE ------------ -          NONE

And the same with the EMPI Results Abstract table So that there is always the same number of TEXT EMPI.

Or we use a "mapping" in the MAIN table that joins the fields of the other two tables and if that data is not there, complete it as empty

Excuse me if There is smth not accurate but is difficult to imagine the whole picture,  but I believe that the problem is clear,  and here you have a possible solution,  maybe someone can offer one better.  Many people do not like mappings due to need resources,  But I am sure that you do not do the ETL more than once a day maybe,  so that a mapping will work fine.

English is not may mother language,  I hope I can make words to be understood