How to Work with FHIR REST APIs using Python

Learn to communicate with FHIR servers in under 20 lines of Python code.

If you’re looking to learn how to interact with a FHIR Server, this guide is for you. We’ll explore how to perform real-world operations like creating, reading, updating and deleting patient data using FHIR REST APIs and Python. If you’re new to FHIR, this comprehensive introduction to FHIR might be a good place to start. You can also explore this earlier tutorial on how to create a FHIR resource in Python.

In this guide, we will interact with the Medblocks Students FHIR server which is publicly accessible. We’ll also be using the requests library in Python to send HTTP requests. Let’s get started.

The FHIR RESTful API

A key component of the FHIR ecosystem are the FHIR RESTful APIs. Each FHIR resource can be managed using the same set of defined interactions. Servers can choose which interactions they support, and must provide a Capability Statement that specifies it.

Interactions are defined as follows,

VERB [base]/[type]/[id] {?_format=[mime-type]}        

Where,

• `base`: The FHIR server URL
• `type`: The resource name, e.g. “Patient”
• `id`: The logical identifier of a resource
• `mime-type`: The media type of a resource, e.g `application/fhir+xml`, `application/fhir+json` or `application/fhir+turtle`

We’ll be using the Medblocks students’ FHIR server at https://fhir-bootcamp.medblocks.com/fhir/.

Pre-requisites and set up

Installing the requests library

The `requests` library in Python allows you to run HTTP requests very easily. Let’s begin by installing it.

pip3 install requests        

Next we must import the library by adding the following line of code to the imports section,

import requests        

Declaring base_url

Let’s also declare a variable named `base_url` and set its value as the URL to the Medblocks FHIR server.

base_url = "https://fhir-bootcamp.medblocks.com/fhir/"        

Post data to the FHIR server

First, we must create a patient locally as we did in the previous tutorial, Create Your First FHIR Resource in Python. Once patient is created, you can then prepare to post it to the FHIR server.

# where 'patient' is the FHIR Patient resource created
patient_object = patient.model_dump()

response = requests.post(base_url + "/Patient", json=patient_object)

print("Status:" + str(response.status_code))
print(response.content)        

The output you get on running this code will look something like this,

Status: 201
b'{\n "resourceType": "Patient",\n "id": "1ebd1a89-c3c6–4d96–9e6e-43395d5a81a3",\n "meta": {\n "versionId": "1",\n "lastUpdated": "2025–04–30T12:43:23.819+00:00"\n },\n "text": {\n "status": "generated",\n "div": "<div xmlns=\\"http://www.w3.org/1999/xhtml\\"><div class=\\"hapiHeaderText\\">John <b>DOE </b></div><table class=\\"hapiPropertyTable\\"><tbody/></table></div>"\n },\n "name": [ {\n "family": "Doe",\n "given": [ "John" ]\n } ],\n "telecom": [ {\n "system": "email",\n "value": "example@abc.com"\n } ],\n "gender": "male"\n}'        

The id returned can be used to retrieve this patient’s details from the server, and that’s what we’ll do next.

Retrieve patient data from the FHIR server

To view the patient you just created on your browser, you can visit the following URL appended by the patient `id` returned above, https://fhir-bootcamp.medblocks.com/fhir/Patient/

To execute this in Python, you can use `requests.get()` as depicted in the following line of code. Also print the status and contents of the response.

response = requests.get(base_url + "/Patient/" + id) # where ID is the identifier assigned to the resource as a string

print("Status: " + str(response.status_code))
print(response.content)        

Using the ID of the example patient created in the previous section, this is the output generated on running the above code snippet.

Status: 200
b'{\n "resourceType": "Patient",\n "id": "1ebd1a89-c3c6–4d96–9e6e-43395d5a81a3",\n "meta": {\n "versionId": "1",\n "lastUpdated": "2025–04–30T12:43:23.819+00:00",\n "source": "#a2f8b6f6-f6df-46"\n },\n "text": {\n "status": "generated",\n "div": "<div xmlns=\\"http://www.w3.org/1999/xhtml\\"><div class=\\"hapiHeaderText\\">John <b>DOE </b></div><table class=\\"hapiPropertyTable\\"><tbody/></table></div>"\n },\n "name": [ {\n "family": "Doe",\n "given": [ "John" ]\n } ],\n "telecom": [ {\n "system": "email",\n "value": "example@abc.com"\n } ],\n "gender": "male"\n}'        

You can also assign the parsed patient to a variable and use it for future operations,

parsed_patient = Patient.model_validate_json(response.content)
print(parsed_patient.id)        

Retrieve all patients from the FHIR server

If instead of a particular patient, you want to retrieve all patients on the FHIR server, you can use the below line of code.

response = requests.get(base_url + "/Patient")        

This is equivalent to visiting the URL https://fhir-bootcamp.medblocks.com/fhir/Patient/.

Update patient data on the FHIR server

To update a patient on the FHIR server, the HTTP verb PUT is used. The code statement would look like this,

response = requests.put(base_url + "/Patient/" + id, json = updated_patient_object)        

In the example patient used above, let’s update the given name to ‘James’ from ‘John’. This is the output generated,

Status: 200
b'{\n "resourceType": "Patient",\n "id": "1ebd1a89-c3c6–4d96–9e6e-43395d5a81a3",\n "meta": {\n "versionId": "2",\n "lastUpdated": "2025–04–30T13:03:19.409+00:00"\n },\n "text": {\n "status": "generated",\n "div": "<div xmlns=\\"http://www.w3.org/1999/xhtml\\"><div class=\\"hapiHeaderText\\">James <b>DOE </b></div><table class=\\"hapiPropertyTable\\"><tbody/></table></div>"\n },\n "name": [ {\n "family": "Doe",\n "given": [ "James" ]\n } ],\n "telecom": [ {\n "system": "email",\n "value": "example@abc.com"\n } ],\n "gender": "male"\n}'        

You’ll notice that the given name has changed to James, and the `versionId` is now updated to ‘2’.

This is how the updated patient looks on accessing it through the browser (which is equivalent to a simple GET request).

Delete data from the FHIR server

To delete a patient from the FHIR server, you can use the requests.delete() function.

response = requests.delete(base_url + "/Patient/" + id)
print("Status: " + response.status_code)
print(response.content)        

And on deleting the test patient we created, this response is generated,

Status: 200
b'{\n "resourceType": "OperationOutcome",\n "text": {\n "status": "generated",\n "div": "<div xmlns=\\"http://www.w3.org/1999/xhtml\\"><h1>Operation Outcome</h1><table border=\\"0\\"><tr><td style=\\"font-weight: bold;\\">INFORMATION</td><td>[]</td><td><pre>Successfully deleted 1 resource(s) in 7ms</pre></td>\\n\\t\\t\\t</tr>\\n\\t\\t</table>\\n\\t</div>"\n },\n "issue": [ {\n "severity": "information",\n "code": "informational",\n "diagnostics": "Successfully deleted 1 resource(s) in 7ms"\n } ]\n}'        

On attempting to access the same patient via browser, this is the output received,

Summary

The best way we’ve found to truly learn FHIR is to experiment with it hands-on. This and previous tutorials can help you create your own FHIR resources using Python and learn as you do. In the last tutorial, we learnt how to locally create a FHIR-compliant Patient resource. This time, we learnt how to send this information to a FHIR server, and perform basic CRUD operations (Create, Read, Update, Delete). The requests library in python is a simple HTTP library that allows you to perform HTTP requests very easily. You can use the Medblocks Students Test FHIR server to test out these new commands.

If you’re looking for a jumpstart to your FHIR learning, we recommend learning from experts. Attending our expert masterclass by Dr. Sidharth Ramesh is a great way to understand how FHIR is changing the health tech landscape, register now to reserve your seat!