2.0 Form: Validate patient address with smarty

[KeepAustinWired]

User Story

As a SimpleReport operator, I want to use Smarty to validate patient addresses, so that I can ensure data quality.

Acceptance Criteria

• Integrate Smarty into the patient information section of the 2.0 form. Similar to the 1.0 new patient form
• Upon hitting the Next button on the patient information page, we hit smarty to validate the address
• If the address was invalid, we pop up a modal to show a suggested address. Similar to the 1.0 new patient form
• If the address was valid, we continue with the form submission.

Scenarios

For example, this could look like:

[valid address]

• User fills out all patient information, including a correct address
• User submits the form, continues to the next section

[invalid address with suggestion]

• User fills out all patient information, including an address with a misspelled street
• Modal pops, giving the user the option to Use address as entered or Use suggested address
• If the user picks one of the Use... options from the modal, we should update to use that address and not hit smarty again
• <- Go back to edit address and Save changes should still be options in the modal.

[invalid address without suggestion]

• User fills out all patient information, including an invalid address
• Modal pops, giving the user the option to Use address as entered. The box usually containing Use suggested address should instead say "No Suggested address found".
• There is an alert banner that reads "The address you entered could not be verified"
• If the user picks the Use... options from the modal, we should use that address and not hit smarty again
• <- Go back to edit address and Save changes should still be options in the modal.

Design Notes

Existing behavior from the 1.0 patient form:

Additional context

In my experience with Smarty in lowers, using an incorrect street number will not find a suggested address. Using a slightly misspelled street name will frequently come back with a suggested address.

When in doubt refer to the behavior of the 1.0 form.

A problem for the leads to wrestle with: How does cost/benefit change if we do this on every single test report where we used to do it on every single patient (and didn't change it once patient was added)?