globaldothealth/outbreak-schema

Occupation

Opened this issue · 7 comments

Occupation is currently a free text string field. There is a separate variable to indicate whether the person is a healthcare worker or not (responses Y | N | NA).
Consider categorising occupation into categories based upon risk factors for a particular disease. Can a library of categories be used that will cover relevant categories for most situations to limit the free text entry?
These are the categories used in ISARIC Core CRF for Dengue specifically;
1, Home-working or unemployed | 2, Indoors-office/health/education/hospitality/business/homes | 3, Indoors-factory | 4, Outdoors-agriculture/forestry/fisheries | 5, Outdoors-construction/industrial/mining | 6, Armed Forces | 7, Student | 88, Other

A standardised list of occupation types can be used as in the above link. However this is not focussed on any particular risks for a specific pathogen (as the ISARIC occupation type options are). To confirm if the standard options give sufficiently detailed data for analysis.

I do like the idea of categorizing occupations, especially from a data analysis side. I feel that it is important to leave the healthcare worker variable, since most occupations that are reported (especially later in the outbreak) focus on healthcare workers. Tracking disease spread in healthcare workers and health facilities is a huge area of focus and concern during outbreaks. For example, most of the occupations that were reported in the various reports for the MVD outbreak in EG focused specifically on how many healthcare/hospital workers had tested positive, how many healthcare workers had died due to MVD, and how many healthcare workers were being hospitalized at a time (see table 2 in MOH epi report). Rarely did we get information on occupations that were not related to healthcare.

I also feel that there should be some flexibility for the occupations variable, such as providing a comment option for more specifics outside of the base categorization. Not all occupations are viewed or defined equally across the globe. For example, during the MVD outbreak, we had cases that worked as healers (a spiritual approach to healthcare and medicine). Here in the U.S., we wouldn't think to categorize these healers as healthcare workers, which is reflected in our linelist (see EQ Guinea Outbreak, case IDs 16 and 17). In EG and other regions, healers may be viewed as healthcare workers or professionals.

Thanks Jacqui! Definitely agree on keeping a separate indicator for to capture data on whether the case is a healthcare workers as this is especially important.
The WHO definition for healthcare worker does include traditional healers, so perhaps this should be made clear on our definition. Looking at the WHO SMART implementation guide for immunizations, a list of healthcare occupations has been defined there as any snomed code that falls under the concept of healthcare professional. We will want to make sure that what we apply is consistent with other SMART guidelines so we may follow this. It would be nice if it were possible to auto-populate the healthcare worker field if any of the indicated occupations are selected to save having to classify this information during curation.

https://build.fhir.org/ig/WorldHealthOrganization/smart-immunizations/dictionary.html
https://build.fhir.org/valueset-practitioner-role.html.

However to note, in the WHO immunizations implementation, occupation is only collected where it is healthcare-related as is skipped where the case is not a health-care worker. Do we want occupation information for non-healthcare roles? In the clinical dataset currently, occupation is more based upon associated risk i.e.

1, Home-working or unemployed | 2, Indoors-office/health/education/hospitality/business/homes | 3, Indoors-factory | 4, Outdoors-animal contact (vet, animal farmer, abattoir worker) | 5, Outdoors-agriculture/forestry/fisheries | 6, Outdoors-DPS collector | 7, Outdoors-construction/industrial/mining | 8, Armed Forces | 9, Student

This could be an option for non-healthcare related roles.

What was the source of the current options in the dropdown on the curator portal?

Recommendation: healthcare professional roles may be of greater importance, and are the occupation most frequently reported upon. Ensure consistency with other WHO datasets, as above.