I can see the difference, I just cant replicate the difference. Both exports are the same single record. The one on the left labels the variables as "type" and the record values as "value" but the one on the right does not include those labels. There are 20 unique variables on the left just as there are on the right, the one on the left just labels the individual variables and corresponding values. ... View more

Thank you so much Bill!! This worked PERFECT to get all the text formatted properly at the top.    I think I am doing a poor job articulating the main issue I was having with the "type" and "value" requirements so hoping this picture of the 2 JSONs side-by-side will help.    The one on the left is what the system admin is telling me they need to be able to read the JSON and the one on the right is what I am able to export from SAS (and other online JSON converters if I use a csv to JSON converter). I have never seen a JSON formatted the way they are saying it has to be formatted. Open to any advice or thoughts on how to accomplish that formatting.     ... View more

Apologies, I was just trying to give a sample. The SAS program is extremely nuanced and joins lots of different data sets together to create the final data set before the JSON export. Since the system that is absorbing the JSON file has very specific naming conventions for the variables, I used options validvarname=any; and wrote the variable names exactly as they need to be exported to the JSON. All the variables are character variables per the requirements of the system bringing in the JSON file. Not sure if it is helpful, but this is the last step before I do the proc json.   proc sql; create table CME_final as select a.Agency, a.Client, b.Event_Code as Type, a.StartDateTime as 'Start Date'n, a.EndDateTime as 'End Date'n, '0' as 'Duration (hh:mm)'n, a.DiagnosisTreated as 'Diagnosis Treated'n, a.NPI, c.id_number as 'Completed By'n, a.PlaceofServiceCode as 'Place of Service Code'n, case when a.TelehealthModifiera = 'GT' then '1' else '0' end as 'Is Telehealth'n, '' as 'Service Track'n, "ohrs" as ProgramProvidingService, '' as 'Program Enrollment'n, '25' as 'Facility Providing Service'n, b.Is_Initial_Contact as 'Is Initial Contact?'n, b.Is_Crisis as 'Is Crisis'n, b.Discharge_Planning as 'Discharge Planning'n, b.Activity_Type as 'Activity Type'n, a.UniqueRecordID as 'Portal ID'n from CME_b as a left join JSONMapping as b on a.ProcedureCode = b.ProcedureCode and a.ProcedureModifier = b.ProcedureModifier left join EvolvNPIMapping as c on a.NPI = c.ibhis_code; quit; The JSON export from SAS gives me for each record: "Variable1":"record1value", "Variable2":"record1value", ... "Variable18":"record1value" The requirements I am receiving from the system that will be reading the JSON say the format needs to be: { "type":"Variable1", "value":"record1value" }, { "type":"Variable2", "value":"record1value" }, ... { "type":"Variable18", "value":"record1value" } I am just having trouble figuring out how to put the word "type" before every variable and "value" before each record's value.    I played with the containers as suggested by ballardw but with no luck.    I am curious now about trying writing the JSON with a data step instead of the PROC JSON syntax. I will try and do some reading and research on that approach.   ... View more

Sample Data attached ... View more

Sorry, here is the JSON exports using the </> feature Current: { "SASJSONExport": "1.0 PRETTY", "SASTableData+CMEBILLING_C": [ { "Agency": "28", "Client": "999999", "Type": "ohr_initial", "Start Date": "07/29/2022 09:00:00 AM", "End Date": "07/29/2022 09:00:00 AM", "Duration (hh:mm)": "0", "Diagnosis Treated": "XXX", "NPI": "9999999999", "Completed By": "999999", "Place of Service Code": 11, "Is Telehealth": 1, "Service Track": "", "ProgramProvidingService": "ohrs", "Program Enrollment": "", "Facility Providing Service": "25", "Is Initial Contact?": 0, "Is Crisis": 0, "Discharge Planning": 0, "Activity Type": "Phone Call", "Portal ID": "de839c8c-1b61-4c68-907f-652c3c99b00a" } ] } Required: { "method":"startautomation", "methodParameters":[ { "type":"IS_API_CALL", "value":"1" } ], "automationKey":"oh_rise", "version":"2.0", "detail":[ { "id":null, "foreignKeyID":null, "action":"ADD", "keyValue":null, "data":[ { "type":"Agency", "value":"28" }, { "type":"Client", "value":"999999" }, { "type":"Type", "value":"ohr_initial" }, { "type":"Start Date", "value":"07/05/2022 09:00:00 AM" }, { "type":"End Date", "value":"07/05/2022 09:00:00 AM" }, { "type":"Duration (hh:mm)", "value":"0" }, { "type":"Diagnosis Treated", "value":"XXX" }, { "type":"Completed By", "value":"999999" }, { "type":"Place of Service Code", "value":"11" }, { "type":"Is Telehealth", "value":"1" }, { "type":"Service Track", "value":"" }, { "type":"Program Providing Service", "value":"ohrs" }, { "type":"Program Enrollment", "value":"" }, { "type":"Facility Providing Service", "value":"25" }, { "type":"Is Initial Contact?", "value":"0" }, { "type":"Is Crisis", "value":"0" }, { "type":"Discharge Planning", "value":"0" }, { "type":"Activity Type", "value":"Face to Face" }, { "type":"Portal ID", "value":"2a229325-e425-402c-a760-bfefa354b1cb" } ], "subData":[ ] } ] } ... View more

When actual_date_b is within 90 minutes of itself, so in the table, the 3rd and 4th record would end up combined ... View more