Cognitive Emulation: Insights from Iodine Data Scientists

By Lance Eason, Chief Data Scientist & Jon Matthews, Data Science Manager 

Q: Iodine has pioneered a new machine learning approach called Cognitive Emulation. How would you summarize this approach?

Lance Eason (LE): There are two sides of the picture to consider when thinking about clinical documentation integrity (CDI): 1) what is really happening to the patient, and 2) what has been documented and written about the patient. CDI is all about making those two pictures align, specifically making sure the documentation actually reflects what is clinically going on with the patient. Iodine’s Cognitive Emulation approach helps identify where there are discrepancies between these two sides by looking at the entire clinical picture, not just what has been documented or what aligns with certain rules or thresholds.

Q: Iodine has iterated on its intellectual property over time as new technology becomes available. How have different types of artificial intelligence been utilized, and what has been the experience with each?

Jon Matthews (JM) and LE: Iodine started off with a rules-based approach just using NLP, which calculated the probability conditions were present based on a simple set of rules. We started identifying challenges and limitations – issues with false positives and false negatives, for example – so, we decided we needed a more advanced technique. We explored coupling  Machine Learning and advanced Natural Language Processing (NLP) to holistically review the entire patient record, and found the approach worked well. This allowed us to move past surface-level documentation improvements constrained to specificity and capture the full spectrum of possible discrepancies between documentation and clinical reality. 

Q: What underlying technology does Cognitive Emulation rely on, and how is it different from legacy solutions on the market?

LE: Iodine’s combination of machine learning and NLP allows it to make separate judgments about the clinical state of the patient versus what is documented. This is a more statistical approach and does not rely on specific hard factors to determine if the patient’s symptoms are above or below certain markers. Instead, it calculates a statistical likelihood that a documentation opportunity exists based on clinical evidence. We let the technology find the patterns and connections between the data rather than arbitrarily defining rules ourselves.

Q: What is a marker-based or rules-based approach, and how is Iodine different?

JM: Rules- or marker-based solutions require either the client or the software authors to define what each condition means, so there is not a lot of flexibility. These approaches follow “if, then” logic and make simple yes or no decisions based on whether inputs are above or below certain values. The problem with this strategy is that one could spend forever iterating on those thresholds manually and still get poor results. You lose a lot of nuance in the data because what is apparently useful might actually be missing a large subset of important features in the models. Machine learning algorithms are able to find these features on their own and do so automatically, which takes the guesswork and years of iteration out of the picture.

 Q: How is Iodine’s Cognitive Emulation approach impacting healthcare?

JM: The biggest advantage of Iodine’s Cognitive Emulation approach is that it is able to leverage the experiences and knowledge of physicians, coders, and CDI Specialists (CDIS) across the country into one product. Iodine’s very broad data set allows it to define new and interesting features that are helpful for predictions, which you would not be able to do if you were just coming up with rules on your own. In this way, Iodine helps hospitals work with the resources they have to capture more opportunities for documentation improvement. With Iodine’s vast dataset, experience, and clinical expertise, we are able to build products that drive revenue integrity and labor optimization. So many functions in a hospital require guessing or people to spend time doing unproductive work. The sky’s the limit for what Iodine can do for health systems with our machine learning models. 

LE: There are a lot of patients in the hospital at any given time, and their medical records are each constantly being updated with new data. The job of a CDIS is to survey all of the patients all of the time. With an infinite amount of CDIS you would review every case every day, but this is unrealistic without the help of technology. Iodine does the job of filtering and prioritizing, emulating what the CDIS would have been doing, but constantly and for every case, so that CDI teams are spending their time on the cases that most require human review. This allows them to do more productive work, not just searching through all of the cases to find the ones to isolate for review. If you think of CDI work as searching for needles in haystacks, queries would be the needles and the hay would be all of the cases that do not require a review. Iodine removes most of the hay so the needles can easily be identified.

Machine Learning versus Natural Language Processing: What is the Difference?

Artificial intelligence is utilized for many use cases across the healthcare industry. However, just as all humans have different cognitive abilities, each type of artificial intelligence is distinct, and some applications are more advanced than others.

Artificial intelligence (AI) is a broad term referring to the field of technology that teaches machines to think and learn in order to perform tasks and solve problems like people. 

There are many types of artificial intelligence, but Iodine focuses on two: 

  1. Natural Language Processing, commonly referred to as NLP, interprets raw, arbitrary written text and transforms it into something a computer can understand. 
  2. Machine Learning, a form of applied statistics, solves problems based on large amounts of data by connecting the dots between many inputs without any human intervention. It answers questions similarly to how humans do, but automatically and on a much larger scale. 

What is the difference between the two? NLP interprets written language, whereas Machine Learning makes predictions based on patterns learned from experience. 

Iodine leverages both Machine Learning and NLP to power its CognitiveML™ Engine. These AI technologies work together to analyze, interpret, and understand the information within a patient’s medical record. Specifically, Iodine uses NLP to identify mentions of symptoms, diseases, procedures, medications, anatomical parts, and other key information. These mentions augment discrete clinical evidence including orders, results, medications, and demographic information and are evaluated by Iodine’s machine learning models. Machine learning models are mathematical representations of real world processes that are trained by analyzing vast amounts of data–billions of data points in Iodine’s case. Through pattern matching, these machine learning models emulate physician thought processes to determine the likelihood specific conditions exist and whether there are discrepancies between the patient’s clinical state and what has been documented.

Both NLP and Machine Learning are essential components of Iodine’s technology. Like CDI offerings available from other companies, Iodine uses NLP to determine what the documentation says which can help identify inconsistencies within the documentation and issues with specificity. By itself, however, NLP cannot find many financial and quality accuracy improvement opportunities because:

  • NLP cannot determine when patient information that is supported by medical data is not written in a patient’s chart. 
  • NLP cannot perform clinical validation, in which clinical evidence does not support something that has been documented, which increases the risk of audit. 

The objective of CDI is to determine whether the written documentation aligns with a patient’s clinical reality, and this requires more than NLP. This is where Machine Learning comes in. Machine Learning considers the entire clinical picture and makes connections or predictions based on learnings from other data, including: lab results, vital signs, medications, radiology results.

The result of combining NLP and Machine Learning is intelligent software that evaluates complex medical data similar to a physician’s approach to diagnosing and treating patients. Iodine’s Cognitive Emulation approach (via CognitiveML™ engine) augments the work of health system professionals with software that can actually “think” and learn from each new data point, just as physicians do. Iodine applies physician-like judgment to the clinical evidence in a patient’s chart and leverages previous learnings to more accurately determine the likelihood a condition exists. Conditions often present in a variety of ways, and by relying on clinical evidence rather than ambiguous thresholds, Iodine is able to identify and learn from these unique instances. 

Reducing Revenue Cycle Leakage with Cognitive Emulation

Today, most healthcare technology solutions that support revenue cycle billing, coding and documentation teams use systems and workflows that “think” like computers – not clinicians.

They leverage rules and check-lists, which only consider narrative documentation and can lead to unforeseen errors given the many nuances of the healthcare revenue cycle. 

Iodine Software takes a different approach. The company has pioneered a new application of artificial intelligence and machine learning called Cognitive Emulation. This approach uses proprietary AI technology and machine learning algorithms which allow a machine to interpret clinical data in the same way a clinician thinks, and emulate clinical judgement in a manner very similar to how providers of care assess and treat their patients. 

Augmenting the work of health system professionals with software that can actually think enables Iodine to solve numerous healthcare problems in new ways with proven results. Iodine analyzes the full clinical record for each patient much the way a clinician would, but at a massive scale: across over 20 million admissions worth of data and 1.5 billion medical concepts from over 480 hospitals across the country.

Iodine’s first application of Cognitive Emulation? Examining new ways to fix mid-revenue cycle leakage.

Applying Cognitive Emulation to the healthcare revenue cycle

Today, each stage of the CDI or documentation process represents an opportunity for leakage via incorrect or incomplete coding and documentation. Legacy software solutions focus on natural language processing – which means they are only looking at documentation in the patient record and not taking into account lab results, vital signs, cardiology and radiology results, and patient history (for example). Even health systems with mature CDI programs supported by legacy software solutions can see significant documentation improvement opportunities not making it into the final code – leaving millions of dollars on the table. 

For clinical documentation to be accurate and support reimbursement objectives, a CDI team would ideally review every record, every day throughout the stay and ensure all conditions are documented appropriately. In the current staffing environment, with typical case volumes, that is essentially impossible without the right type of technology.

Similar to how physicians base decisions on their knowledge of medical concepts, experience in the field, and clinical judgment, Iodine leverages its Cognitive Emulation approach (via its CognitiveML Engine) to consider and analyze the entire patient record. For example, rather than drawing conclusions simply from documentation, Iodine considers: 

  • Lab results
  • Vital signs
  • Orders
  • Medications
  • Demographic information
  • Patient history
  • Radiology results
  • Cardiology results
  • Working, target, and final codes
  • and Documentation

By matching this evidence against patterns observed across the billions of data points in Iodine’s database, the CognitiveML Engine determines the statistical likelihood a meaningful difference exists between a patient’s clinical state and what has been documented. Not only does Iodine identify instances of highly-likely but undocumented conditions, it also can identify where the clinical data contradicts what is documented to help avoid denials. As new data is reviewed, Iodine expands its knowledge base, becoming increasingly more intelligent and effective over time.

Proven results using Cognitive Emulation

Iodine applied its Cognitive Emulation approach to clinical judgement with the AwareCDI Suite to reliably identify areas of potential opportunity to accelerate productivity, data accuracy, and financial return. 

Cognitive Emulation drives smarter prioritization through the Iodine Concurrent module, which considers the entirety of the clinical record and patient experience to predict likely conditions, identify gaps between evidence and documentation, and help CDI teams query more effectively and efficiently. 

Cognitive Emulation is also the foundation of the Iodine Forecast module, which predicts final DRGs early in the patient’s stay–well ahead of the availability of final coded diagnoses which are typically only available days to weeks after discharge. Forecast™ increases the accuracy and efficiency of assigning working DRGs, supporting CDI workflow and enabling teams across the health system to make real-time, evidence-based predictions – without needing to wait for final coded data.  

Iodine clients consistently achieve substantial results utilizing Cognitive Emulation and the AwareCDI Suite to improve documentation integrity. Irrespective of any preexisting legacy tools, Iodine clients have seen:

  • A median 75% increase in query volume
  • A mean 4.3%, or 0.0633 point average, increase in CMI per facility
  • An increase in MCC capture at 94% of facilities after using Iodine 
  • $1.5 billion in additional appropriate reimbursement recognized annually

To learn more about Iodine’s Cognitive Emulation approach to solving mid-revenue cycle leakage, contact info@iodinesoftware.com.

Figures are based on a $6000 modeled base rate and actual measured MCC capture performance from the 2019 Iodine Performance Cohort Analysis of 339 facilities that compared measured MCC capture and CMI impact for the Iodine usage period 9/1/2018-8/31/2019 against pre-Iodine baseline performance.

Examining the Unforeseen Financial and Clinical Impact of COVID-19

Leveraging Iodine Software’s Cognitive Emulation approach to better understand current and future state 

Iodine Software is a healthcare AI company that has pioneered a new machine learning approach— Cognitive Emulation—to help healthcare finance leaders build resilient organizations. To date, Iodine has partnered with over 600 hospitals in the United States to create a large and diverse clinical data set that can provide insights into the COVID-19 pandemic.

Iodine’s Cognitive Emulation approach analyzes the full clinical record for each patient much the way a clinician would, but on a massive scale. Coupled with proprietary technology that forecasts diagnosis-related groups (DRGs) for every patient still in the hospital, Iodine is able to look in real time at trends emerging in this data set without having to rely on final coded data that is typically only available post-discharge, after a considerable delay.

For this report, Iodine reviewed more than 60,000 COVID-19 cases from over 600 hospitals— spanning cases from the entire country, including both infection hot spots and emerging areas of concern. Using the Iodine CognitiveMLTM engine and Iodine ForecastTM product, the Iodine Data Science team identified likely COVID-19 patients and predicted the DRG for recently discharged, but not final-coded, cases. Both structured and unstructured data were leveraged to generate this analysis. The timeframe for this analysis is from March 4 through May 3, 2020. Within this data set, 50.1% of patients were male and 49.9% were female.

This information is meant to help healthcare providers nationwide more accurately forecast their resource needs (including staff, ICU beds, ventilators and other critical care equipment) and understand the demographics most vulnerable to COVID-19, as well as support healthcare finance leaders in determining the right strategies to ensure financial resilience both in the near- and long-term.

Download the full report here

 

The Impact of COVID-19 on Inpatient Admissions

By The Iodine Data Science Team on June 4, 2020

Overview

Iodine Software is a healthcare AI company that has pioneered a new machine learning approach—Cognitive Emulation—to help healthcare finance leaders build resilient organizations. Using Iodine’s CognitiveML™ engine, we reviewed inpatient admission data from 350+ facilities and 6M+ admissions. This analysis is a comparison of data from 2019, with data from the January 1 through May 19, 2020 timeframe.

Prior to the COVID-19 pandemic, hospitals were experiencing similar year-over-year levels of inpatient admissions for both elective and non-elective procedures. However, medical and surgical admissions began to decline in early March with admission counts dropping to their lowest point in early April. Following this sharp decline, both medical and surgical admissions began to recover but have still not reached their respective 2019 volumes.

Figure 1: Year-over-year Weekly Admissions

This figure shows that both elective and non-elective inpatient admissions began a significant decline around the week of March 4, 2020. After hitting their lowest volumes during the week of April 1, admissions began to steadily increase but are still below 2019 volumes. Elective admissions are trending to be close to pre-COVID levels.

Figure 2: 2020 Elective Surgery Inpatient Admissions

This figure shows that elective surgery inpatient admissions experienced a precipitous decline around the week of March 11 and dropped to a significantly low volume of 23% of 2019 admissions during the week of April 1. However, that low was short lived and beginning the week of April 22, admissions began to increase almost as quickly as they declined. Currently, elective surgery admission volume is at 76% of 2019 admissions.

Figure 3: Year-over-year Weekly Admissions: Selected Elective Admission Types

This figure analyzes selected elective admission types, omitting those with insufficient data, including palliative, pediatrics, and psychiatry. Maternity admissions, which are typically stable, remained relatively consistent with 2019 admissions. However, medical and surgical admissions began to decline around the week of March 4 and arrived at the lowest admission volume the week of April 8. Following this sharp decline, both medical and surgical admissions began to recover but have not reached their respective 2019 volumes.

Closing

As Iodine showed in previous analyses, hospitals are experiencing a significant drop in expected overall reimbursement this year because of COVID-19. As hospitals exit the pandemic, medical dollars are expected to go down as COVID admissions decrease. Unless surgical volumes appreciably recover, we expect overall reimbursement to sink once more. In reviewing this information, it will be important for CFOs and Revenue Cycle departments to look at new strategies for operating in this “new normal”, both short- and long-term.

Analyzing the Financial Impact of COVID-19

By Iodine Data Science Team – May 13, 2020 

This is Part III of Iodine’s series on the clinical and financial impact of COVID-19 and what healthcare finance leaders need to be aware of as they plan for short- and long-term resiliency. Part I focused on the mortality impacts of COVID-19 and Part II on COVID-19 length of stay and ventilator demand.

Iodine Software is a healthcare AI company that has pioneered a new machine learning approach—Cognitive Emulation—to help healthcare finance leaders build resilient organizations. To date, Iodine has partnered with over 600 hospitals in the United States to create a large and diverse clinical data set that can provide insights into the COVID-19 pandemic. 

Iodine’s Cognitive Emulation approach analyzes the full clinical record for each patient much the way a clinician would, but on a massive scale. Coupled with proprietary technology that forecasts DRGs for every patient still in the hospital, Iodine is able to look in real time at trends emerging in its data set without having to rely on final coded data that is typically only available post-discharge, after a considerable delay. 

Using Iodine’s CognitiveML engine and the Iodine Forecast product, we identified likely COVID-19 patients and predicted the DRG for recently discharged, but not final-coded, cases. More than 60,000 COVID-19 cases from the 600+ hospitals in Iodine’s current data set—spanning cases from the entire country including both infection hot spots and emerging areas of concern–were reviewed. The timeframe for this analysis is from March 4 through May 3, 2020. Both structured and unstructured data were leveraged to generate this analysis.

COVID-19 Patient Impact on Revenue
This analysis shows the impact on year-over-year weekly reimbursement an average hospital would expect as it cares for an increasing volume of COVID-19 patients. 

Even prior to handling an appreciable volume of COVID-19 patients, hospitals were already experiencing an approximate 30% dip year-over-year in expected reimbursement. In anticipation of increased admissions due to COVID-19 and the potential risk of transmission to other patients, facilities across the country cancelled elective surgeries and postponed non-urgent procedures. These actions drove a significant drop in expected overall reimbursement. 

Although we see overall reimbursement rising as hospitals treat higher volumes of COVID patients, surgical reimbursement continues to be suppressed. This could be worrisome as hospitals exit the pandemic, as medical dollars are expected to go down as COVID admissions decrease. Unless surgical volumes appreciably recover, we expect overall reimbursement to sink once more.

The figure below shows the following: 

  • Overall reimbursement starts off steeply lower than 2019 levels with a 30% YoY decline.
  • This depressed level of reimbursement remains fairly consistent until COVID-19 patients account for 7% of weekly admissions.
  • Reimbursement begins to recover as the volume of COVID-19 patients surpass 7% new weekly admissions. We have found that two scenarios contribute to this recovery:
    • For a certain number of facilities, the volume of medical admissions during COVID-19 is similar to their historical volume, but case mix index (CMI) increased dramatically due to a higher percentage of COVID-19 cases. The medical CMI increase of 30%-70% on a per-facility basis has lessened the depression in year-over-year reimbursement.
    • A smaller number of facilities also experienced an increase in medical case volume due to more COVID-19 admissions. In this situation, the total medical admissions increase coupled by a rise in medical CMI drives significant medical reimbursement improvement.
  • Reimbursement reaches about 100% of the previous year’s reimbursement level when the volume of COVID-19 patients reaches 10% of weekly admissions. However, this is only possible due to the influx of medical admissions. During this period, surgery reimbursement significantly drops to only 45% of the previous year’s surgery reimbursement level.

Note: This analysis only shows the expected reimbursement impact as COVID-19 admissions rise, and does not directly imply what a post-pandemic recovery may look like. Also this analysis does not account for additional expected reimbursement provided by the CARES Act, which increases Medicare reimbursement for COVID-19 cases.

Breakdown by MDC
The figure below shows the observed year-over-year change in admission volume percentage, broken down by Major Diagnostic Category (MDC), for COVID facilities, for March and April, 2019 and 2020. 

Note: This study defines a COVID facility as a facility where COVID-19 admissions represent at least 5% of the total admission for that month. Reference to a non-COVID facility is defined as a facility where COVID-19 admissions represent less than 5% of the total admissions for the month.

At COVID facilities, while most other MDCs experienced a decline in volume as expected, Respiratory and Infectious Disease admissions volumes were up, about 90% and 70% respectively. Trauma Surgery volume was up about 20%.

The next figure shows the same analysis for facilities where COVID-19 admission volume represents less than 5% of the hospital’s total admissions in a month. In anticipation of a COVID-19 case surge, and per CMS recommendations, facilities not immediately impacted by COVID cases had an overall reduction in admission volume across almost all MDCs. Trauma (Surgical) was the only MDC with any volume increase, and it only increased by 3.9%.

A comparison of the two figures also shows that at COVID facilities, the average MDC surgical volume decreased by 41%, while at non-COVID facilities, the average MDC surgical volume only decreased by 23%.

Conclusion
What Iodine has found to date ranges from verification of trends seen on far smaller data sets to seemingly new insights; and we’ve only just begun. Subsequent analyses will provide week-by-week updates, as well as extensions of this analysis into financial recovery questions.

Prior to COVID-19, health systems were already operating on generally thin margins, with many finance leaders acknowledging that a significant root cause was leakage from their mid-revenue cycle and that “average performance” was still well below optimal results. 

With health systems now facing an unexpected reimbursement decline of up to 30% YoY, CFOs and Revenue Cycle departments need to come up with both short- and long-term strategies to ensure resiliency. Iodine welcomes collaboration to accelerate the mutual understanding of this disease and to help hospitals cope with the challenges it presents them. Please contact us at info@iodinesoftware.com

 

Analyzing the Clinical Impact of COVID-19 Part II

By Iodine Data Science Team on May 8, 2020 

This is Part II of Iodine’s series on the clinical and financial impact of COVID-19 and what healthcare finance leaders need to be aware of as they plan for short- and long-term resiliency. To read Part I on the mortality impacts of COVID-19, click here

Iodine Software is a healthcare AI company that has pioneered a new machine learning approach—Cognitive Emulation—to help healthcare finance leaders build resilient organizations. To date, Iodine has partnered with over 600 hospitals in the United States to create a large and diverse clinical data set that can provide insights into the COVID-19 pandemic. 

Iodine’s Cognitive Emulation approach analyzes the full clinical record for each patient much the way a clinician would, but on a massive scale. Coupled with proprietary technology that forecasts DRGs for every patient still in the hospital, Iodine is able to look in real time at trends emerging in this data set without having to rely on final coded data that is typically only available post-discharge, after a considerable delay. 

This information is meant to help healthcare providers nationwide more accurately forecast their resource needs (including staff, ICU beds, ventilators and other critical care equipment) and understand the demographics most vulnerable to COVID-19, as well as support healthcare finance leaders in determining the right strategies to ensure financial resilience both in the near- and long-term.

Iodine reviewed more than 60,000 COVID-19 cases from the 600+ hospitals in its current data set—spanning cases from the entire country including both infection hot spots and emerging areas of concern. Within this data set, 50.1% of patients were male and 49.9% were female.

Length of Stay
Inpatients accounted for 56% of the 60,000 total COVID-19 admissions in Iodine’s data set. These 34,000 cases were analyzed by length of stay (LOS). The first figure (Percentage of Inpatients by Length of Stay) shows that 19.2% of people have an average length of stay between 3-5 days. The second figure (Inpatient Average Length of Stay by Age Group) shows the average length of stay by age group. On average, a COVID-19 inpatient was admitted for a length of stay of 7.6 days.

Ventilator Demand
The mortality rate of inpatients who were on a ventilator at some point during the inpatient stay is more than 5x higher than an inpatient who was never on a ventilator during their hospital stay (38.6% mortality rate for those on a ventilator compared to 7.3% mortality rate for those not on a ventilator).

Note: All mortality stats are in reference to admitted inpatients with COVID-19.

To learn more about Iodine’s Cognitive Emulation approach, please contact: info@iodinesoftware.com.

Analyzing the Clinical Impact of COVID-19 Part I

[content-border]← Return to Blog   |   By Iodine Data Science Team on May 4, 2020

Iodine Software is a healthcare AI company that has pioneered a new machine learning approach—Cognitive Emulation—to help healthcare finance leaders build resilient organizations. To date, Iodine has partnered with over 600 hospitals in the United States to create a large and diverse clinical data set that can provide insights into the COVID-19 pandemic.[/content-border]

Iodine’s Cognitive Emulation approach analyzes the full clinical record for each patient much the way a clinician would, but on a massive scale. Coupled with proprietary technology that forecasts DRGs for every patient still in the hospital, Iodine is able to look in real time at trends emerging in this data set without having to rely on final coded data that is typically only available post-discharge, after a considerable delay. 

This information is meant to help healthcare providers nationwide more accurately forecast their resource needs (including staff, ICU beds, ventilators and other critical care equipment) and understand the demographics most vulnerable to COVID-19, as well as support healthcare finance leaders in determining the right strategies to ensure financial resilience both in the near- and long-term.

Iodine reviewed more than 60,000 COVID-19 cases from the 600+ hospitals in its current data set—spanning cases from the entire country including both infection hot spots and emerging areas of concern. Within this data set, 50.1% of patients were male and 49.9% were female.

This post begins a series on the clinical and financial impacts from COVID-19. The following analysis will focus on an initial set of observations looking at mortality and admission status. 

Admission Status

Inpatients accounted for 56% of the 60,000 total COVID-19 admissions in Iodine’s data set. These 34,000 cases were analyzed by age and gender, as shown in the figure below. 

Mortality Rate by Age Group

Iodine assessed the relative likelihood of mortality by age group of admitted inpatients with COVID-19, as shown in the graph below. The mean mortality rate was 16.3% across all inpatients, while the mean mortality rate for all inpatients on ventilators was 38.6%.  Additionally, individuals aged 90+ had a mortality rate that was nearly 2x the mean mortality rate for all COVID-19 patients in an inpatient hospital setting.

The graph below assesses the relative likelihood of mortality by age group and gender of admitted inpatients with COVID-19. Across all age groups, males had a higher rate of mortality than females.

Mortality Rate by Comorbidity 

Certain pre-existing conditions placed COVID-19 patients at higher risk for severe illness. Two conditions, diabetes and morbid obesity (BMI of 40 or greater), were analyzed in greater detail. Overall, the mortality rate for COVID-19 patients with either of these conditions was 18.5%, compared to the overall rate of 16.3% across all age groups.

The figure below shows that, as age increased, the relative likelihood of mortality due to obesity and diabetes decreased, suggesting that other factors were more likely to drive mortality for older COVID-19 patients.

Note: All mortality stats are in reference to admitted inpatients with COVID-19.  

To learn more about Iodine’s Cognitive Emulation approach, please contact info@iodinesoftware.com.

Return to Blog

Supporting the Front Lines during COVID-19 with Machine Learning Models

[content-border]← Return to Blog

In light of COVID-19, hospitals and health systems across the country are adapting to meet the needs of their communities. At Iodine Software, we also feel a responsibility to support our clients and their patients during this crisis.[/content-border]

After learning major pain points of our partner hospitals, our Iodine Data Science team applied their machine learning and artificial intelligence expertise to our 20 million inpatient admission database to create a predictive tool that allows for early identification of patients at risk for respiratory compromise. Early identification allows healthcare providers to plan care, intervene sooner and support positive patient care outcomes.

Iodine’s Patient Triage and Escalation Support Tool (PTEST) is an extension of our existing technology that has the ability to predict patient conditions and comorbidities. This tool is designed to assist with identifying patients at risk for pulmonary challenges, need for ventilators, and critical care support. This capability helps patient care teams to intervene early and potentially reduce the demand for scarce resources. In addition to supporting the quality of patient care, PTEST has the potential to aid healthcare leaders in more accurately forecasting their resource needs amidst equipment shortages and overwhelmed ICUs.

Help your organization to stay resilient during this healthcare crisis by learning more about PTEST and Iodine’s machine learning approach — cognitive emulation — to support accurate reporting of care provided. Please contact: info@iodinesoftware.com

Return to Blog