Modernizing the substance abuse and mental health data archive

The Substance Abuse and Mental Health Data Archive (SAMHDA) site provides access to decades of research conducted by the Substance Abuse and Mental Health Services Administration (SAMHSA), empowering external data scientists and researchers to directly analyze information about substance use and mental health issues throughout the United States.
RESULTS AT A GLANCE
35%
faster page load times
84%
increased user engagement (based on average site visit duration)

Our team completed a redesign of the SAMHDA site to address usability, accessibility, and issues related to low user engagement. Our purpose was to modernize the site and create clear pathways to the data.

Challenge

The SAMHDA site has nearly 2,000 files available to users. Our challenge was to fix the pathways to these data files and address usability issues. In the previous information architecture, data was scattered across 356 pages, and downloads were a minimum of six clicks from the homepage. Each page duplicated the body copy of its parent pages, and in several instances, users introduced inconsistencies as they copied and pasted between years. The content management and technical debts created by this sprawling site were substantial.

SAMHSA also tasked our team with migrating the site from Drupal 7 to Drupal 9 to enhance site security, performance, and compliance.

Solution

To understand the core challenges of the website, we met with SAMHSA’s Center for Behavioral Health Statistics and Quality (CBHSQ) through a series of listening sessions to confirm their vision for the website. We also conducted user testing with representative users (especially around the data files and bibliography pathways), including statisticians and SAMHSA regional administrators. Using a content model, we showed the interconnectivity of content and proposed new user paths that would lead more directly to the files required. We audited and consolidated the content while engaging subject matter experts to ensure that it remained accurate while we implemented plain language best practices.

Our human-centered design approach allowed us to align the user interface and information architecture to specific user needs. We incorporated data visualizations on the homepage to highlight the data, added a navigation wizard to help users identify which survey collection would meet their research needs, and developed a faceted search for users to access the bibliography. We also built a help section so users could do more to self-service their needs.

Our team migrated the SAMHDA site from Drupal 7 to Drupal 9, assessing which modules contained deprecated and procedural code and then refactored them to leverage Object-Oriented Programming. We performed a content inventory and schema analysis from which we created migration classes. These scripts moved the content from the old site to the new while carefully maintaining existing file structures and URLs, and assuring the integrity of the dataset. We used the U.S. Web Design System (USWDS) in coordination with Drupal 9 for compliance with the 21st Century Integrated Digital Experience Act (IDEA) while at the same time elevating the SAMHSA brand.

Results

The information architecture and navigation now have the data files two to three clicks away from everywhere on the site. We reduced the site’s 356 pages to 13, and content is no longer duplicated throughout the site. The streamlined content has been a vital part of our SEO strategy for the site, and the increased findability means users are spending less time searching and more time engaging with data.

Reduced from 5-6 clicks to 2-3 clicks for reaching data files on the site
Reduced 356 web pages to 13 after content consolidation and de-duplication
Automated and manual QA checks of migrated data files completed in under one day

Today, our team is working on developing data visualizations for additional data collections to surface insights about substance abuse and mental health issues and improve data access for the research community.

Talk to an expert today

Related work