Bio Metric - Response Ratio for Menstrual Cycle Survey

[HopkinsDC]

Please run a one-time metric for the response rate for the next menstrual cycle survey for those eligible by site, one for research (BioSpm_Setting_v1r0; 650516960=0) and one for clinical (BioSpm_Setting_v1r0; 650516960=1).

The denominator is all participants eligible for the menstrual cycle survey (SrvMC_MenstSrvElig_v1r0; 289750687=1). The numerator is participants with a menstrual cycle survey completion status of Submitted (SrvMC_BaseComplete_v1r0; 459098666=2).

See the mock-up of tables below:

• The table should have column percents for "Eligible", "Not Started", "Started", and "Submitted".
• The "Response Rate" column should be percent (# Submitted/# Eligible)
• Percents in the table should have 1 decimal place
• The response rate in the "Total" row should be the overall rate for all eligible at all sites collectively (Total # Submitted/Total # Eligible)

[KELSEYDOWLING7]

@HopkinsDC Since SrvMC_MenstSrvElig_v1r0 (d_289750687) is a new variable it'll probably make more sense for me to keep the original eligibility requirements of (SrvBLM_ResSrvCompl_v1r0 OR SrvBlU_BaseComplete_v1r0) submitted and SrvBlU_MENST60_v1r0=yes, right? Or did you specifically want to test this new derived variable?

[KELSEYDOWLING7]

@HopkinsDC The data is also currently showing that only research sites have eligible participants. Does that sound possible?

I filtered instead by site, (as d_650516960 is only in the bioSurvey table and there are only 13 Connect_IDs in that table) and according to the menstrual cycle survey only those in HP, Henry Ford, Marshfield, Sanford, and UC have become eligible

[HopkinsDC]

@KELSEYDOWLING7

Please the following for "Eligible for menstrual cycle survey" demoniator:

SrvBLM_ResSrvCompl_v1r0 (299215535.265193023) = submitted (231311385) AND SrvBlU_MENST60_v1r0 (112151599) = YES (353358909) OR SrvBlU_BaseComplete_v1r0 (826163434.253883960) = submitted (231311385) AND SrvBlU_MENST60_v1r0 (112151599) = YES

Let me know if you have the same problem with only seeing research participants.

[KELSEYDOWLING7]

@HopkinsDC That's the logic that I have in place now and I'm still seeing that problem

[HopkinsDC]

@KELSEYDOWLING7 Ok, weird. Can you just use SrvMC_BaseComplete_v1r0 (459098666) instead, and instead of the Eligible column, have a total column at the end?

[KELSEYDOWLING7]

@HopkinsDC So I was able to do the Research table the way you wanted. Does this still work?

[HopkinsDC]

@KELSEYDOWLING7 Yes, this looks like it'll work.

Please make the following changes:

• add the count to each cell along with the percentage
• change to row percents
• remove the Total row
• change the Eligibility column to a Total column and move it to be between Submitted and Response Rate.
• calculate the Response Rate as the row Submitted over the row Total.

[KELSEYDOWLING7]

@HopkinsDC For that case wouldn't the response rate and the submitted values be the same? Submitted per site over Total eligible?

[KELSEYDOWLING7]

Or should the total still be the same values as eligible before, and then response rate be submitted over all participants of that site, eligible or not?

[HopkinsDC]

Oh, you're right! Yes, the total should be "Total Eligible", so whatever it was before, but just the count for the site; exclude the percent from that row only.

[KELSEYDOWLING7]

@HopkinsDC

Got it! Here's the final outcome then

[HopkinsDC]

Thank you, Kelsey!

@brotzmanmj Michelle, I'm unfamiliar with the eligibility criteria for the menstrual cycle survey, so does the table Kelsey generated reflect what you would expect as far as the Total Eligible is concerned? If not, please advise on what variable(s) should be used for this metric. Thanks!

[brotzmanmj]

@KELSEYDOWLING7 @HopkinsDC I looked at the document with the Menstrual Cycle Survey requirements and confirmed the requirements are (1) completed the biospecimen survey (research or clinical) and (2) Had a period in the last 60 days [SrvBlU_MENST60_v1r0] = “Yes” . To know if these numbers make sense, I guess the most straightforward thing to do is run a frequency of variable [SrvBlU_MENST60_v1r0] on the research and clinical biospecimen surveys. Kelsey, can you run that?

[brotzmanmj]

(also there are a few spelling errors in the title of the table above, if you can fix those on the next run too, thanks)

[KELSEYDOWLING7]

@brotzmanmj Yes, just confirmed the figures are correct. And sorry about that, please see the updated table below

[brotzmanmj]

Thanks Kelsey. I think we want one table for Research and Clinical (there really isn't a Research Menstrual Cycle Survey and a Clinical Menstrual Cycle Survey, there is just a Menstrual Cycle Survey). If we drop 'Research' from the start of the title can we then add all the KP sites to this table? And for now, if they have no one eligible for the menstrual cycle survey yet then they would appear on the table with the Total Eligible = 0.

[KELSEYDOWLING7]

@brotzmanmj right now according to the data the only sites (d_827220437) that have data with the menstrual cycle survey requirements are HP, Henry Ford, Marshfield, Sanford Health, and UC. None of the KP Sites.

I'll try to manually add the KP Sites and then just have 0 across the board for now.

[brotzmanmj]

@KELSEYDOWLING7 are you saying that we're not actually triggering the menstrual cycle survey following clinical collections yet? Can you confirm with Davin and Tony on this (or have you already)? thanks

[KELSEYDOWLING7]

@brotzmanmj No I don't think that's quite it. I'm saying as of right not KPs have no one eligible for the menstrual cycle survey just yet

[brotzmanmj]

Got it, OK thanks!

[KELSEYDOWLING7]

@brotzmanmj No problem! Let me know how this looks.

[brotzmanmj]

Great, thank you! The only last thing I would change if it's possible to do so is to put the percentages in parentheses instead of separately but a comma, just because it's easier to read.

[KELSEYDOWLING7]

Sure thing!

[brotzmanmj]

Terrific, thank you! We should add this to the survey completion metrics that we monitor weekly. I could see this fitting in either on the Biospecimen Weekly Metrics or on the Weekly Module Metrics where we monitor completion of all the other survey modules.

[KELSEYDOWLING7]

No problem! Ok, just let me know which you prefer it on and I can add it. I agree that that it would fit nicely with the Weekly Module Metrics with all other surveys.

[brotzmanmj]

Sounds good, let's put it there

[KELSEYDOWLING7]

Ok, that's now number 8

[HopkinsDC]

Thank you, Kelsey!

I am closing this issue now that the table has been added to the modules metrics report.