AI has the potential to revolutionize local health, from streamlining administrative tasks to predicting and preventing outbreaks. Yet, for many, its benefits remain out of reach. A 2024 survey highlights a concerning gap: while 24% of large local health departments are actively using AI or planning to adopt it within the next year, only 5% of small and 7% of medium-sized departments report the same.
The good news is there’s interest in AI – about 40% of local health departments express interest in exploring AI. But without proactive efforts to bridge this divide, the consequences could be significant. Health workers will likely remain overburdened with repetitive administrative tasks, and communities relying on local health departments could face reduced access to essential, cost-effective health services.
That’s why it’s important that local health departments learn how to overcome the common roadblocks to AI implementation with simple, actionable steps. By prioritizing staff education, starting small with practical applications, and establishing clear guidelines, local health departments can lay the foundation for sustainable AI adoption. This approach ultimately enhances efficiency, improves health outcomes, and ensures cost-effective care for the communities they serve.
Defining AI for effective staff implementation
Misconceptions about AI pose a major barrier to implementation, often fueled by the overwhelming buzz surrounding the technology. Some workers are worried that widespread use of AI could lead to job elimination. Others fear that the hallucinations or misinformation provided sometimes by AI might decrease the quality of their work. To overcome these fears, local health leaders must take the lead in defining and modeling AI adoption. Clear, targeted training can ensure that AI is seen as a tool tto enhance the work of public health professionals, rather than replace it.
Effective staff education starts with demystifying AI and setting realistic expectations. Training should focus on practical applications relevant to local health work, highlighting its potential to naturally fit into ongoing projects. Additionally, hands-on workshops, case studies, and ongoing support can help employees build confidence in AI tools. By investing in education, local health departments can empower their workforce to embrace AI, ultimately improving efficiency and community impact.
Starting small for big impact
Critical to implementing AI is knowing where to start. When identifying first use cases for AI, small changes targeted at reducing time-consuming administrative work can actually be the most impactful. Among the 5% of local health departments currently using AI, the most common applications are generating communication materials and plans, drafting emails, and writing letter responses – routine tasks that, while necessary, consume valuable time.
Automating repetitive administrative duties allows public health workers to shift their attention toward higher-impact responsibilities. Instead of being bogged down by endless paperwork, employees can spend more time engaging directly with communities and patients. Whether providing care, conducting outreach, or responding to public health concerns in real time, AI-driven efficiency ensures that staff can focus on their core mission – delivering essential preventative health serves to guard communities from illness and injury.
Clear guidelines facilitate safe use
Another common barrier to AI implementation is the concern that data security and privacy might be compromised. Oftentimes, local health departments looking to implement AI may be met with hesitancy or push back from employees or policymakers who worry about the potential for disclosure of confidential information due to this. In fact, 78% of local health departments cite this as their most common AI-related fear.
To address this concern, local health departments must establish clear guidelines for AI use, particularly with regard to data privacy. Successful guidelines should institute strong data governance practices that protect sensitive datasets, de-identify patient or personal information data, and implement comprehensive protection throughout the entire data lifecycle.
Another key strategy is to utilize publicly available data where possible rather than data drawn from one’s own client or patient data. For example, using AI to help scan and summarize public comments related to health in open forums such as reddit, Twitter, or podcasts can help officials better understand current thinking regarding real-world health practices. In addition, leveraging AI to help users more rapidly access accurate information from official public documents and trainings can help improve the reliability and usefulness of otherwise undiscoverable information.
By setting clear policies and focusing on responsible AI implementation, local health departments can overcome security concerns and begin using AI in ways that enhance local health without introducing unnecessary risks.
Getting ahead of the AI curve
While AI already holds tremendous potential, the reality is that its power and accessibility will only continue to grow over time. And as the AI learning curve steepens, those who don’t start integrating AI now risk falling even further behind. Taking small steps today, from streamlining routine tasks to establishing clear guidelines and prioritizing staff education, will ensure that local health departments are prepared to leverage AI effectively and stay ahead of the curve.
Photo: erhui1979, Getty Images
John Auerbach is ICF’s primary federal health expert and thought leader within the company’s public sector business. John’s thought leadership advances ICF’s combination of proven domain and scientific expertise with leading-edge analytics and technology solutions to drive improved health outcomes for clients.
Eddie Kirkland is a Director of Data Science at ICF and a leading statistics and AI expert with more than 20 years of experience in data and organizational leadership. He specializes in guiding data-rich projects from concept to delivery, working directly with clients to identify areas of need, developing custom solutions in an agile framework, and delivering clear and meaningful results. As a data scientist, Eddie supports federal clients including the Centers for Disease Control and Prevention (CDC) in various research areas by delivering full-stack DataOps solutions. He created an automated surveillance system for tracking nationwide school closures and leveraged generative artificial intelligence from a self-hosted large language model. Eddie also architected and engineered a natural language processing engine—which helps distill public health trends from raw social media data—and a Robotic Process Automation (RPA) platform, which reduced months of painstaking research to a fully automated natural language processing-based system.
This post appears through the MedCity Influencers program. Anyone can publish their perspective on business and innovation in healthcare on MedCity News through MedCity Influencers. Click here to find out how.
