Samples import

[MateStrysewske]

We may need a way to import sample data into the system.

[Iheanacho2027]

As a suggestion, can we have a situation whereby one first exports SORMAS cases without 1) sample collection, sample shipment, recieved sample and sample result information (i.e completely without sample information 2) recieved sample and sample result information but has sample collection and shipment information present

After the export of these cases having case sn, the user can then enter the missing sample information in the exported .csv file and then proceed for import

This would be more accurate than a situation where one has to manually link with the case sn on SORMAS

@hzi-braunschweig @MateStrysewskeSym @bernardsilenou

[Iheanacho2027]

Hi @MateStrysewskeSym, can this ticket be prioritized. Its very important at the moment

[Iheanacho2027]

@MateStrysewskeSym @MartinWahnschaffeSymeda @bernardsilenou pls how can this be made a reality by the next release. We are in dire need of it in Nigeria

[MateStrysewske]

@Iheanacho2027 You might need to get in touch with HZI directly because priorities are currently coming exclusively from their side.

[Chinedar]

@bernardsilenou @hzi-old This ticket is a priority for us now. Can we bring it forward please?

[bernardsilenou]

@Chinedar Is Nigeria using bar code now? we need a unique identifier for a sample to implement this. Name, epid number, etc is not enough

[Iheanacho2027]

Can't we use the case ID for entering fresh sample information, then sample ID for entering results and other details of an already entered sample?

[bernardsilenou]

I think this is not enough since a case can have many samples

[Iheanacho2027]

Many samples can be linked to one case ID, just the same way many contacts can be linked to a single case ID as it is currently practiced in the contact import module

[ChristopherRiedel]

The problem is that there must be a way to check whether the sample is already in the system. For this there must be a unique identifier for each sample. Otherwise we will get many duplicates.

[bernardsilenou]

@Iheanacho2027 I strongly think Nigeria should use the bar code system, it will make it easier, otherwise there is no guarantee of not having duplicates when importing samples.

[Iheanacho2027]

@ChristopherRiedelSymeda can't the system use date variables as unique identifiers to check for duplicates?.

[MateStrysewske]

@Iheanacho2027 Dates can't be used as unique identifiers because they're not unique.

Also, the bar codes aren't enough to really make sure that duplicate samples won't be imported because it's not a mandatory feature and there might (and probably will) be countries who don't use it.

@bernardsilenou What about the lab sample ID? It can only be filled in when the sample has been received by the laboratory, but that might be a starting point.

[ChristopherRiedel]

@Iheanacho2027 No that's not possible. This way only one sample per day of calender could be entered. The absence of a unique identifier does not prevent us from implementing the sample import. It will just result in some duplicates. What I could also imagine is to use the field or lab sample id as identifier. You could at least check if a sample with the same id already exists.

[bernardsilenou]

Both field or lab sample ids can be used but not perfect because:

1. lab sample id is not uniq also since it is based on the specific lab. Lab sample id may uniquely identify the sample at the lab level but not at the country level.
2. The field sample id is not required field, thus will be missing for many samples. In addition, if this id is not registered by the scan system, we can not be sure it is unique also.

What about using field sample id (only those recorded by scanning the bar code). Samples with missing field sample id will not be imported. This is not also an optimal solution.

[Iheanacho2027]

@bernardsilenou @ChristopherRiedelSymeda i understand that checking for duplicates can be a problem for importing fresh samples. I believe import sample results for already created samples shouldnt be a problem. We can work around using the unique SORMAS generated Sample ID for importing the result. For a start, lets implement this

[bernardsilenou]

@Iheanacho2027 fro each sample registered in sormas, there is already a sample uuid. After Issue #1961 is implemented, do you think lab users will be able to document this uuid for each sample that they have?

[Iheanacho2027]

Yes they will, that's why I have been clamouring for #1961

On Sun, 9 Aug 2020, 14:24 Silelou Bernard Chawo, notifications@github.com wrote:

@Iheanacho2027 https://github.com/Iheanacho2027 fro each sample registered in sormas, there is already a sample uuid. After Issue #1961 https://github.com/hzi-braunschweig/SORMAS-Project/issues/1961 is implemented, do you think lab users will be able to document this uuid for each sample that they have?

— You are receiving this because you were mentioned. Reply to this email directly, view it on GitHub https://github.com/hzi-braunschweig/SORMAS-Project/issues/1154#issuecomment-671051869, or unsubscribe https://github.com/notifications/unsubscribe-auth/AMDDTGNIHRZPXDOA4QZZPJDR72PRJANCNFSM4HP4CV6Q .

[Iheanacho2027]

@bernardsilenou pls can this be prioritized?. Thanks

[Iheanacho2027]

Based on discussions with @bernardsilenou the import template should have the following variables (i) samples UUID (ii)Case ID (iii)Date of report (iv)District and region (v) First Name and Last name (iii)Samples information from collection to results stage (as found on the case import template)

Now if, (i) The case ID exists and samples information is entered on the import template without a samples UUID, the samples information should be imported as a new sample (ii) The samples UUID and case ID exists, and both IDs match as related on the system, sample information entered should be merged with already existing sample information or should over-ride already existing sample information

Date of report, Date of sample collection, First name, Last name, case ID and samples UUID should be used for validation(duplicate case and duplicate sample only)

[bernardsilenou]

@Iheanacho2027 I think information about pathogen test is missing in the specification + specification of duplicate detection of sample and pathogen test. We generally need information about case + sample + pathogen test

• duplicate detection begin with person + case+ sample+ pathogen test. If no duplicate case is found, then system create a new case + sample+test.
• if duplicate of any of the entities is found, then we merge

@Iheanacho2027 Please can you add more details infos?

[Iheanacho2027]

Based on discussions with @bernardsilenou the import template should have the following variables (i) samples UUID (ii)Samples Pathogen test UUID (iii)Case ID (iv)Date of report (v)District and region (vi) First Name and Last name (vii)Samples information from collection to results stage (as found on the case import template)

Now if, (i) The case ID exists and samples information is entered on the import template without a samples UUID, the samples information should be imported as a new sample (ii) The samples UUID and case ID exists, and both IDs match as related on the system, sample information entered should be merged with already existing sample information or should over-ride already existing sample information (iii) The samples pathogen test UUID, samples UUID and case ID exists, and all IDs match as related on the system, pathogen test information entered should be merged with already existing pathogen test information or should over-ride already existing pathogen information

Date of report, Date of sample collection, First name, Last name, case ID and samples UUID should be used for validation(duplicate person, duplicate case, duplicate sample and duplicate pathogen test result only)