HJBR Sep/Oct 2023
HEALTHCARE JOURNAL OF BATON ROUGE I SEP / OCT 2023 51 Brooke Wyatt, MPH, DrPH Healthcare Informatics Program Manager Health Equity heart failure. This capacity also extends into the space of cancer diagnosis and treatment as well. Many physicians can be overwhelmed by the number of genet- ic variants, emerging drugs, and changing protocols to be taken into consideration, but with ML, we have the ability to greatly reduce this burden by establishing genetic profiles of those who would best benefit. Here in Louisiana, health networks are getting creative and even testing the use of AI products, specifically ambient sound technology, during patient visits to capture visit notes for review and approval by the provider. There are also instances where health institutions are exploring the use of ChatGPT features to help review and triage the influx of emails physicians consistently receive. In addition, through the use of AI, we are able to better gather and organize repositories of patients who have quali- fying indicators over varying time points/ encounters/claims as well as those who do not traditionally or consistently engage in care. Specifically, within Medicaid, there have been a number of studies showing the benefit of ML in better interpreting uti- lization of services as well as the needs and barriers enrollees face. This has ranged from understanding the predictive per- formance of social determinants of health measures to its intersection with clinical and behavioral health outcomes. For ex- ample, Medicaid recipients with access to support services, such as transportation, and who have housing or food security demonstrate more consistent engagement with their clinical care. This underscores the importance of prioritizing and ad- dressing the patients’ needs with the same assertiveness as traditional clinical care. From a more individualized perspective of a Medicaid patient’s experience, AI and ML can help create personalized care and support patient monitoring, as well as mit- igate billing and claims abuse. A personal- ized care plan puts the patient at the center of their own unique health needs instead of on the outside with a care plan based on broad and generalized guidelines. Through remote devices such as a smartwatch or phone application, patient monitoring for specific health issues or for check-ins is more tailored, accessible, and effective. Through identity detection, fraudulent claims can be more quickly identified be- cause the influx of data is more efficiently managed. In addition to, rather than instead of While there are many benefits to incor- porating AI and ML in healthcare, it is not a one-size-fits-all approach. If developed incorrectly or not assessed for its predic- tive validity and reliability as a viable re- source, these tools can become burden- some to both patients and providers. Their application must take into consideration the user experience, which is why phy- sician input and buy-in during develop- ment is critical. Additionally, ownership over tool updates is vital to ensuring that whatever decision-making being per- formed is reflective of any policy or proto- col changes. And while older providers can be overwhelmed by the task of learning AI and ML uses as well as trusting its output, as “technology natives,” younger physi- cians are more willing and able to inte- grate them into their practice and routine workday. Lastly, careful attention should be paid to the concerns around AI that exist beyond the healthcare setting: that it will lead to the automation of jobs and workforce displacement; that it will ex- ploit fear; and that, ultimately, it will mean the loss of control, privacy, and human value. But within healthcare, the touch of human delivery reigns supreme, and with the cost of implementing larger modeling into workflows and systems being so high, conservative use for the foreseeable future is expected. n REFERENCES 1 Gonzales, A.; Guruswamy, A.; Smith, S.R.; John- son, A. “Synthetic data in health care: A narra- tive reiew.” PLOS Digital Health 2, no. 1 (Janu- ary 2023). https://www.ncbi.nlm.nih.gov/pmc/ articles/PMC9931305/ 2 Gall, C.; Suzuki, E. ”5. Big data: A new dawn for public health.” Health in the 21st Century: Put- ting Data to Work for Stronger Health Systems (OECD iLibrary, Nov. 21, 2019). https://www. oecd-ilibrary.org/sites/f24cb567-en/index.html? itemId=/content/component/f24cb567-en 3 Johnson, K.B.; Wei, W.; Weeraratne, D.; et al. “Precision Medicine, AI, and the Future of Per- sonalized Health Care.” Clinical and Translational Science 14, no. 1 (January 2021): 86-93. doi: 10.1111/ cts.12884 4 Siwicki, B. “Epic-linked voice AI improves provider experience at Franciscan Mission- aries.” Healthcare IT News (July 13, 2023). https://www.healthcareitnews.com/news/epic- linked-voice-ai-improves-provider-experience- franciscan-missionaries Atrained researcher,study design, implementation, and evaluation expert, Brooke Wyatt, DrPH, is re- sponsible for driving the evaluation of health equity initiatives at Louisiana Healthcare Connections.The extent of her work focuses on investigating clinical risk factors — and social determinants of health — as they relate to less favorable outcomes in patient care,understanding the relationship between specific diseases or disease states and drawing relation to clinical or behavioral outcomes. More recently, this has taken a focus on synthesizing big data to iden- tify key clinical risk factors and indicators that can improve quality of care for patients.
Made with FlippingBook
RkJQdWJsaXNoZXIy MTcyMDMz