For many Americans, health care will come to an abrupt end in 2020 as doctors’ offices and hospitals close their doors to screening and elective surgery. For people at higher risk of diseases such as breast cancer, the inability to obtain active medical care will not only delay diagnosis and treatment, but also delay time, which is essential for disease control. The reality of these delays is now emerging: Doctors report worrying rise in advanced cancer cases.
As the health system catches up with the backlog of (potentially life-saving) screenings, it is important that they prioritize high-risk groups to ensure they can get the preventive care they need.
One way is to work with the privacy team and design and execute the correct data strategy.
Use data and artificial intelligence to get better patient information
By using AI modeling, in addition to the patient’s age and certain demographic data, the payer and third-party data sets can also be considered. For example, Overweight and obese women are at higher risk of breast cancer Compared with women who maintain a healthy weight, especially after menopause. Users who currently or recently use HRT are diagnosed with a higher risk of breast cancer.
Utilizing this rich and personalized data is crucial, because although age is an important factor in many diseases, there are other factors/aspects that affect a person’s risk. For example, where they live (and what they are exposed to), the other diseases they have been diagnosed and treated, and their pregnancy history may all put someone at a higher risk of breast cancer.
Once the higher-risk individuals are identified, you can contact these patients through personalized and relevant communication to build trust, eliminate the fear of participating in healthcare during the pandemic, and gently remind and inform them of the importance of screening sex. Our experience is that this can facilitate screening and can make a diagnosis for patients who may not be scheduled for screening.
Effective and fair use of artificial intelligence
Despite the efforts of many health systems in terms of diversity, fairness, and inclusiveness, they rarely involve algorithms for communicating and participating with millions of patients. Therefore, it is essential for the team using this data to understand the existence of bias and take measures to reduce it.
First: Make sure you have a diverse team. Second: Assume that all models are biased.Third: Collect right type data. Fourth: Continuous testing and retesting.
See and think locally
Many organizations use national data sets or models when reaching patients, although many of the factors that put someone at risk for a certain disease are regional. Therefore, when designing a data strategy, it is important that health systems use data relevant to their patient communities. For example, if environmental factors put someone at a higher risk of breast cancer, data from the state on the other side of the country is not only irrelevant, but harmful.
Involve the entire organization
Any data strategy should include stakeholders from around the organization; it should not exist in departmental silos. For example, in marketing campaigns for women to undergo mammograms, clinical experts should be consulted to ensure accurate information and appropriate language.
For many health systems, members of the marketing, operations, and clinical teams collaborate regularly to drive positive health outcomes and improve the patient experience. These organizations have better overall results because they are working to eliminate silos and working together to focus on patients together.
Firm goals and flexible methods
With data and artificial intelligence, flexibility is part of the package. When you approach a group of patients, you may find that after you start, you need to adjust the algorithm slightly—especially when you find that certain patients are excluded due to unintentional bias. Remain firm on the goals of the strategy or activity, such as promoting screenings, but be flexible in the path of achieving the goals. And, when you find that you have made a mistake, recognize it and move on in a slightly different way.
Always put the patient first
In healthcare, it should always be relevant to the patient. Every strategy designed and adopted should aim to improve health outcomes and patient experience. As North Star, every data strategy should be designed with individual patients in mind-how they will react, what next steps you want them to take, and what information might resonate with them.
Photo: belchonock, Getty Images
