Will AI kill QI/PI?

Artificial Intelligence (AI) has the potential to significantly impact the field of Quality Improvement (QI)/Performance Improvement (PI), but it's unlikely to "kill" it. Instead, AI is more likely to firstly change the need for QI/PI, and then secondly transform and enhance how we do QI/PI work. 

How AI will change the need for QI/PI

The healthcare industry is on the brink of a significant transformation, driven by the rapid advancement of AI. As AI continues to evolve, its impact on healthcare improvement will be profound, reshaping everything from patient care to operational efficiency.

Reviewing the literature and increasing volume of content and commentary on the healthcare applications of AI reveals some generally agreed areas of benefit:

1. Enhancing Diagnostics

AI algorithms can analyse medical images, lab results, and patient data with remarkable accuracy, leading to faster and more precise diagnoses. This will allow healthcare providers to identify conditions earlier, improving patient outcomes and reducing the burden on healthcare systems.

Whilst it is still early days for AI being used to make diagnoses, Harvard’s School of Public Health estimates it could improve health outcomes by 40% whilst reducing treatment costs by up to 50%.

There are multiple studies that support this estimation. For example, a study published in the UK which showed that using an AI system to interpret mammograms had an absolute reduction in false positives and false negatives by 5.7% and 9.4%, respectively.

2. Optimising Treatment Plans

Personalised medicine is becoming a reality with AI's ability to process vast amounts of data, including genetic information, lifestyle factors, and treatment responses. AI can recommend customised treatment plans tailored to each patient's unique needs, improving effectiveness and reducing adverse effects.

In a study performed by Sheu et al., the authors tried to predict the response to different antidepressants using electronic health records (EHR) of 17,556 patients and AI. They demonstrated that antidepressant response could be accurately predicted using real-world EHR data with AI modelling.

3. Improving Patient Monitoring

Wearable devices and remote monitoring tools powered by AI can continuously track patient health metrics, alerting healthcare providers to potential issues before they become critical. This real-time data collection enables more proactive care, particularly for chronic conditions.

4. Streamlining Administrative Tasks

AI can automate routine administrative tasks, such as scheduling, billing, and documentation. By reducing the administrative burden on healthcare professionals, AI allows them to focus more on patient care, enhancing the overall efficiency of healthcare systems.

5. Facilitating Drug Discovery

The process of drug discovery is traditionally time-consuming and expensive. AI can accelerate this process by analysing complex biological data to identify potential drug candidates, predict their effectiveness, and streamline clinical trials. This will lead to faster development of new treatments and therapies.

6. Supporting Healthcare Decision-Making

AI tools can assist healthcare providers in making more informed decisions by offering evidence-based recommendations, analysing patient data, and providing insights into potential outcomes. This support can lead to better patient care and more consistent treatment across different healthcare settings.

The impact of AI on healthcare should not be underestimated. Some estimates put the savings in the US alone at $360 billion annually. And that is before you also factor in benefits to healthcare quality, increased access, better patient experience, and greater clinician satisfaction.

The advances and improvements AI can bring to healthcare are immense. So, you could argue it will reduce the need for us humans to carry out any improvement work ourselves – AI will improve everything for us! But it is not quite that simple. For one, AI still needs us humans to deploy it on a problem and then actually implement and sustain the results. Plus, there is still the need for us to address the more human/behavioural side of healthcare processes and target improvements in those areas. Because for all the innovative technologies, cutting-edge medicines and advances in knowledge we have, healthcare organisations are still populated by humans organising themselves into collaborative processes - with varying levels of success. And it is these processes and the decisions we make as part of them that still provide plenty of room for improvement.

So, we will still need to do improvement work ourselves. But how will AI assist us in carrying out improvement work in the future?

Here are some thoughts on how QI/PI work will change with the advent of AI:

How AI will change how we do QI/PI

1. Automation of Routine Tasks

AI can automate many routine tasks in QI, such as data collection, monitoring, and analysis. This could free up human experts to focus on more complex and strategic aspects of QI, such as interpreting results, developing improvement strategies, and engaging stakeholders.

However, in 2024 these AI tools comes with a significant concern about accuracy and are not mature enough to successfully replace a typical QI stuff’s daily activities.

2. Enhanced Data Analysis

AI, particularly through machine learning, can analyse vast amounts of data more quickly and accurately than humans. This can lead to the discovery of patterns and insights that might be missed by traditional methods, thereby improving the quality of decision-making in QI initiatives.

3. Predictive Capabilities

AI can be used to predict outcomes and identify potential quality issues before they arise. For example, in healthcare, AI can predict patient outcomes based on historical data, helping to prevent adverse events and improve patient care quality. It could also be trained to analyse past improvement initiatives and predict the success of current ones based on how well current interventions have worked in the past.

An example comes from NYU’s Langone Medical Center where teams use AI tools to predict hospital readmission rates, a vital patient safety and care quality measure. The AI-powered predictive analytics technology is able to predict 80% of readmissions and performs 5% better than standard computer tools at calculating readmission risk.

4. Personalisation of Quality Interventions

AI can tailor quality improvement interventions to specific contexts or individuals. This is particularly valuable in areas like healthcare, where there are so many variables to consider when judging the likely success of an intervention. Just because it worked at another organisation, doesn’t mean it will work for you – AI can help adapt the intervention to work for your context.

5. Continuous Learning and Improvement

AI systems can continuously learn from new data and outcomes, enabling a dynamic approach to quality improvement that adapts over time, and more quickly than relying on human learning alone. This can lead to more effective and sustainable improvements.

However, AI is not without its challenges. It requires high-quality data, and biases in AI systems can lead to unintended consequences. Moreover, human oversight remains essential to ensure that AI-driven recommendations are practical, ethical, and aligned with organisational goals.

Conclusion

Rather than killing QI work, AI is likely to shift its focus from routine tasks to more strategic roles, where human judgment, creativity, and ethical considerations are crucial. The integration of AI in QI can lead to more efficient and effective improvement efforts, but it will require collaboration between AI technologies and human experts.

This is not a collaboration I foresee happening quickly given my experience of how technology is embraced in service-led/grassroots healthcare improvement work. But it is a collaboration that we at Life QI will continue to encourage and support in our mission to accelerate healthcare improvement in communities around the world.