This page provides you with instructions on how to extract data from HIPAA and analyze it in Looker. (If the mechanics of extracting data from HIPAA seem too complex or difficult to maintain, check out Stitch, which can do all the heavy lifting for you in just a few clicks.)
What is HIPAA?
The Health Insurance Portability and Accountability Act (HIPAA) defines rules that American organizations must follow to securely handle and maintain Protected Health Information (PHI). To remain in compliance, organizations are required to have a signed Business Associate Agreement (BAA) from any partner organization that creates, receives, maintains, or transmits PHI. The partner must ensure that it will safeguard the PHI that passes through its systems. Businesses also have to meet a long checklist of compliance rules and practices.
What is Looker?
Looker is a powerful, modern business intelligence platform that has become the new standard for how modern enterprises analyze their data. From large corporations to agile startups, savvy companies can leverage Looker's analysis capabilities to monitor the health of their businesses and make more data-driven decisions.
Looker is differentiated from other BI and analysis platforms for a number of reasons. Most notable is the use of LookML, a proprietary language for describing dimensions, aggregates, calculations, and data relationships in a SQL database. LookML enables organizations to abstract the query logic behind their analyses from the content of their reports, making their analytics easy to manage, evolve, and scale.
Getting HIPAA data
You migrate PHI just as you would any other data, but you must stay cognizant of HIPAA regulations. No one but you and the data source can handle the data unless you have a BAA in place with them.
You can use any methods your data provider offers to extract data from their service. Many cloud-based data sources provide APIs that expose data to programmatic retrieval. Others allow you to set up webhooks to push event data to requesters. For data that lives in a database, you can use SELECT statements or a utility that does a mass dump of the data you specify.
Loading data into Looker
To perform its analyses, Looker connects to your company's database or data warehouse, where the data you want to analyze is stored. Some popular data warehouses include Amazon Redshift, Google BigQuery, and Snowflake.
Looker's documentation offers instructions on how to configure and connect your data warehouse. In most cases, it's simply a matter of creating and copying access credentials, which may include a username, password, and server information. You can then move data from your various data sources into your data warehouse for Looker to use.
Analyzing data in Looker
Once your data warehouse is connected to Looker, you can build constructs known as explores, each of which is a SQL view containing a specific set of data for analysis. An example might be "orders" or "customers."
Once you've selected any given explore, you can filter data based on any column available in the view, group data based on certain fields in the view (known as dimensions), calculate outputs such as sums and counts (known as measures), and pick a visualization type such as a bar chart, pie chart, map, or bubble chart.
Beyond this simple use case, Looker offers a broad universe of functionality that allows you to conduct analyses and share them with your organization. You can get started with this walkthrough in Looker's documentation.
Keeping HIPAA data up to date
Once you've set up your data pipeline to your HIPAA data source, you can relax – as long as nothing changes. You have to keep an eye on any modifications that your sources make to the data they deliver. You should also watch out for cases where your script doesn't recognize a new data type. And since you'll be responsible for maintaining your script, every time your users want slightly different information, you'll have to modify the script. Keep in mind that HIPAA is all about rules and compliance, so you'll also have to know what HIPAA permits and proscribes, as will anyone else who works on the script.
From HIPAA to your data warehouse: An easier solution
As mentioned earlier, the best practice for analyzing HIPAA data in Looker is to store that data inside a data warehousing platform alongside data from your other databases and third-party sources. You can find instructions for doing these extractions for leading warehouses on our sister sites HIPAA to Redshift, HIPAA to BigQuery, HIPAA to Azure SQL Data Warehouse, HIPAA to PostgreSQL, HIPAA to Panoply, and HIPAA to Snowflake.
Easier yet, however, is using a solution that does all that work for you. Products like Stitch were built to move data from HIPAA to Looker automatically. With just a few clicks, Stitch starts extracting your HIPAA data via the API, structuring it in a way that is optimized for analysis, and inserting that data into a data warehouse that can be easily accessed and analyzed by Looker.