This FQHC slashed its patient no-show rate with AI in 3 months – TechToday

New York City’s Urban Health Plan used artificial intelligence to improve operational efficiency and patient care and controlled its high annual patient no-show rates with cost-effective patient interventions.

Above average missed appointments

No-shows are due to multiple factors, including social determinants of health such as transportation.

But they are also a long-term challenge for healthcare organizations that can affect patient care quality, increase healthcare organization costs, and present unnecessary reductions in overall patient access to healthcare appointments.

In March, Urban Health Plan had 42,000 health visits, the highest in its history, according to a presentation at an eClinicalWorks Health Summit in Boston last week.

“We took a multifaceted approach to just addressing access and patient engagement in general,” said Alison Connelly-Flores, director of medical information for the Urban Health Plan. Health informatics news.

This achievement is important because annually, UHP experiences a large number of missed appointments and appointments for waiting patients are sometimes “booked too far”. Providers are often overbooked to accommodate no-shows, Connelly-Flores said.

The federally qualified community health center organization, one of the largest in New York State, provides primary care, 18 specialties, diagnostic and other services for approximately 86,000 patients. The no-show rate was very high at UHP’s 12 sites in the South Bronx, Corona Queens, and Central Harlem neighborhoods, its 12 school health centers, and behavioral health services.

While UHP scheduled 794,322 visits in 2022, only 57.6% were completed compared to the national average for eClinicalWorks EHR data of 71%.

With 336,600 no-shows and overbooking to compensate, the results can range from long wait times, patient dissatisfaction and stress on providers, all of which can exacerbate burnout.

The organization had to change things to keep the health centers open.

Leadership wanted to know if the no-show rate was above or below the national average and which patients were continuing to miss their appointments.

Through the pilot, UHP learned that its no-show rate was 16.52% higher than its EHR peers.

It also learned that despite attrition among the low, moderate, and high probability groups, UHP show rates were consistent for each group between 2019 and 2022, even during the pandemic.

“It was interesting that these groups behaved in the same way. I didn’t anticipate that … it was definitely a pattern.” Connelly-Flores said after the session.

Machine learning analysis of network-level data

The no-show algorithm is a screening analysis that enables a systematic and consistent ordering of data that allows data analysts and users to view data at the network level.

“I can basically have a conversation with this data,” said Sameer Bhat, co-founder and vice president of sales at eClinicalWorks, before the discussion of the pilot test of UHP’s algorithm, which is part of a larger study of eClinicalWorks with numerous clients.

Bhat demonstrated how to look at demographics such as ethnicity and poverty level in one low dashboard and noted that the platform is independent of the EHR.

The platform can also aggregate discrete EHR data to identify gaps in care, according to the company’s website.

When he introduced Connelly-Flores and the pilot team, he said, “We’re blown away by some of the findings.”

They say the algorithm can find needles in haystacks and can help identify patients with a high probability of no-shows with 85% to 90% accuracy.

Missed Appointment Interventions That Worked

In addition to working with the no-show prediction model, the UHP team fine-tuned their outreach process using eClinicalMessenger.

UHP already handles more than one million annual voice messages, secure text messages and email reminders, according to eClinicalWorks’ announcement of the pilot.

After the model identified patients at the highest and moderate risks of missing their appointments, UHP tested new targeted interventions to ensure patients kept their scheduled appointments.

Would a phone call help? With 3,000 appointments a day, UHP can’t call all of its scheduled patients, Connelly-Flores said.

eClinicalWorks retained 38,431 with a high probability of not presenting in a control group. The company shared 18,061 with a high probability of no-show as well as 908 with a medium probability of no-show with UHP in order to test targeted interventions.

UHP distributed a high-risk no-show report to designated associates and provided a script to make calls and messages as consistent as possible. Designated associates documented the results of the call.

Patients who missed their appointments that day were offered the opportunity to switch to a same-day virtual visit by initiating telemedicine visits or rescheduling appointments. UHP doctors will call patients who have not visited themselves.

Connelly-Flores said if a doctor gets to a patient, almost 100 percent will accept the same-day virtual reschedule option. They use the video app to call these patients and only switch to a telehealth visit when a patient agrees.

“If you receive a ‘last minute’ cancellation or rescheduling, you may still be able to get your appointment back,” says an eClinicalWorks blog from March 2023.

“If the last-minute change is due to a transport or travel problem, maybe a television will be enough. This can save the appointment and even encourage more frequent visits for your convenience”.

For telehealth visits scheduled with patients who have a high or moderate likelihood of missing their appointment and who have not been seen in 15 months or more, UHP sent additional text messages.

UHP also increased access to virtual care by expanding hours to 89 per week.

To better support providers, the healthcare organization revised its workload analysis templates to account for each provider’s no-show rate and added same-day slots to accommodate a change to virtual attention.

Although these adjustments may require follow-up, overall, UHP found that the strategies could further reduce the no-show rate.

The algorithm was implemented in January 2023 and the intervention resulted in 4,432 more visits during the three-month pilot.

The low probability rate of no-show for 2023 so far is more than 5% higher than the previous four years.

The result between the two pilot patient groups also showed a 24.14% increase in the likelihood of making their appointments for patients at high risk of no-shows and an 8.08% improvement for those with moderate risk.

Although the no-show rate for patients most likely to miss their appointments increased by 154%, UHP interventions also increased the no-show rate by 19.17% for no-show patients of moderate risk.

UHP added a specific full-time role and adjusted an existing role to offer part-time support to call only those identified by the algorithm as being at higher risk of no-show.

During the pilot, targeted phone calls were about 400 per day, Connelly-Flores said.

“It wasn’t a great promotion.”

He said next steps for UHP include integrating the algorithm into its EHR, case management, addressing barriers to care, sending more personalized appointment reminders to patients with high risk of missed appointments and drill down into the data for more information.

“By harnessing the power of data and machine learning, we can help providers like Urban Health Plan deliver more effective care to their patients and reduce the burden of missed appointments,” said Girish Navani, CEO and co-founder of eClinicalWorks , in the company’s statement.

“This ultimately helps lower the cost of health care and helps improve patient outcomes.”

“When patients receive timely care, they see better health outcomes.” Connelly-Flores said.

Andrea Fox is a senior editor at Healthcare IT News.
Email: [email protected]

Healthcare IT News is a publication of HIMSS Media.

Source link
The latest in healthcare technology has made quite the splash, and no more so than at America’s largest federally qualified health center (FQHC): a sprawling 79-site network servicing millions of patients. This particular healthcare network recently made headlines when it announced that it had significantly slashed its patient no-show rate with the help of artificial intelligence (AI).

After the center’s launch of their AI technology in just three months, the FQHC reported seeing a 50% drop in patient non-attendance; a stunning success for any healthcare organization. The AI models identify individual risk factors in upcoming appointments and remind patients of their appointments via text, as well as autogenerate follow-up messages on patients’ engagements – leaving little room for error. Not only did the center reduce its no-show rate, but patient attendance confirmation improved exponentially.

Ikaroa, a leading full-stack tech company, is at the forefront of AI tools and applied technologies that have enabled the healthcare industry to unlock new milestones in precision, efficiency, and accuracy. With a deep understanding of the complexities of the healthcare industry, Ikaroa provided the AI solution to the FQHC and enabled them to deliver outstanding results in a short period of time.

The role of AI and ML solutions is rapidly increasing to improve healthcare organizations’ success, safety, compliance and communication goals. With the help of AI solutions, such as the one provided by Ikaroa, FQHC’s now have substantially greater control of their patient data. These systems enable FQHC’s to detect and respond with greater efficacy and accuracy, thus improving the quality of care for the patient.

The success of this FQHC’s AI implementation demonstrates the enormous potential of artificiall intelligent solutions in streamlining clinical workflows and enabling providers to deliver the best possible care to patients. With solutions developed by Ikaroa, healthcare organizations can reduce no-show rates, improve patient outcomes and experience, and maximize efficiency.


Leave a Reply

Your email address will not be published. Required fields are marked *