Solving the Rubik’s Cube of payer data
March 21, 2019
By Mayur Yermaneni
For many health payers, making sense of their data is like trying to solve a Rubik’s Cube.
They have all of these individual data points. But the more they twist and turn them with their analytics, the further away from the goal they seem to get – and the more frustrated they get with the process.
Anyone who has learned to solve an actual Rubik’s Cube with regularity, however, knows the key is to understand and recognize the patterns that lead to success.
The Rubik’s Cube health payers are currently facing is the mountain of incredibly rich data they’re sitting on right now. America’s Health Insurance Plans (AHIP) says the typical regional payer processes $8 billion in claims each year. Each of those claims houses a wealth of interesting data. Yet the challenge they face is how to aggregate and parse it in ways that enable them to take actions that will improve health outcomes and reduce costs.
But it isn’t just the volume of data that makes it so valuable. It’s the unique view it offers into member/patient health.
Even today, in the electronic age, providers, for the most part, only see the clinical, laboratory, and pharmaceutical data captured by their own office, facility, or health system. Any care that occurs outside their boundaries is often a mystery (even though by now it shouldn’t be), leaving holes in their understanding of the member’s health.
The same is even more true for laboratories and pharmaceutical companies. They see tests given or prescriptions issued, but have no real data on why those tests were ordered or prescriptions written. It’s like knowing there are multiple colors on the Rubik’s Cube but only being able to distinguish one of them. The puzzle will be awfully difficult to solve.
Payers, however, can see all the “colors” because each of those entities submits claims to them. By incorporating all of those data points, along with additional information such as demographics, psychographics, and social determinants of health (SDoH) into their analytics, payers have a far greater ability to manipulate the data to drive more reliable, and more actionable, conclusions.
Deeper predictions
With the right analytics at their disposal, payers can take this breadth and depth of data and use it not just to identify members/patients who are already in the high-risk category but also predict those who are trending toward it but not there yet. The “trendings” are a very important group to identify because payers still have time to change their destiny, improving their health and lowering their own benefit costs.
Take diabetes for example. It is one of the most prevalent chronic conditions in America. In fact, according to the CDC more than 100 million adult Americans are now living with diabetes or pre-diabetes. Not only is it the seventh-leading cause of death in the U.S., it is an incredibly costly disease both in financial and quality of life terms. It can also lead to – or complicate – other chronic conditions as well.
The key issue is that once a member/patient is diagnosed as a diabetic, he/she is always considered a diabetic, and will require greater clinical and financial resources. If payers can unwind their Rubik’s Cube of data to identify which members/patients are trending toward acquiring diabetes, however, they can recommend interventions and work with members/patients (and their providers) to halt or even reverse the trend.
Payers can also use that data to identify which “Diabetes 101” interventions are likely (or unlikely) to work for a specific member/patient based on others who fit a similar profile or persona, so they can create alternatives. Suggesting an expensive medication for a diabetic member/patient who lives in an underserved community and is faced with the dilemma of paying rent or paying for medication is unlikely to drive member/patient compliance. Identifying that challenge through demographic or SDoH data will alert payers to look into alternatives.
Incorporating artificial intelligence (AI) into the analytics makes working with individuals and populations even more effective. It’s like having someone who already knows how to solve the Rubik’s Cube guide you through the process.
The advantage of AI is that it can work through that mountain of diverse data and uncover subtle relationships a human might miss, such as noticing that members/patients who respond that they have a pet on their self-assessments tend to be more compliant to their plans of care than those who do not.
By sharing that information, payers can help providers produce better health results for members/patients, again while reducing their own costs. It’s a win for everyone.
Taking action
Unlike providers, payers also have one other advantage: the financial resources to do something with the information once they have it.
In the value-based care era, providers know they must be proactive around member/patient health, especially for those who fall into the high-risk category. Yet it is difficult for most of them to dedicate the resources required to improve member/patient engagement to a significant degree.
Even mid-size regional payers, however, can usually afford to hire nurses or other clinicians – either internally or by outsourcing the function – to take what the predictive analytics have revealed and close the loop. These payer-sponsored clinicians can monitor adherence to plans of care, contact members who are trending toward increasing risk, and work with social services agencies to overcome SDoH challenges. The top payers will even have clinicians visit members in their homes, when required, to drive real change.
Solving the puzzle
While it may seem that having too much data about members is a first-world problem, the reality is that it can be overwhelming. And, like a Rubik’s Cube, sometimes the more organizations work to make sense of it the further away they get from their desired outcome.
Predictive and prescriptive analytics, especially when supported by AI and machine learning, can help take those maddening twists and turns of data and create a complete, clear picture that helps drive healthcare quality and member/patient satisfaction up while driving benefit costs down. Be sure you’re not the last health payer to discover it.
About the author: Mayur Yermaneni is chief strategy and growth officer of eQHealth Solutions, a population health management and healthcare IT solutions company that touches millions of lives each year. The organization has more than 30 years of experience working with payers, providers, and government entities on increasing quality outcomes and optimizing payer and provider networks. He can be reached at myermaneni@eqhs.org.
For many health payers, making sense of their data is like trying to solve a Rubik’s Cube.
They have all of these individual data points. But the more they twist and turn them with their analytics, the further away from the goal they seem to get – and the more frustrated they get with the process.
Anyone who has learned to solve an actual Rubik’s Cube with regularity, however, knows the key is to understand and recognize the patterns that lead to success.
The Rubik’s Cube health payers are currently facing is the mountain of incredibly rich data they’re sitting on right now. America’s Health Insurance Plans (AHIP) says the typical regional payer processes $8 billion in claims each year. Each of those claims houses a wealth of interesting data. Yet the challenge they face is how to aggregate and parse it in ways that enable them to take actions that will improve health outcomes and reduce costs.
But it isn’t just the volume of data that makes it so valuable. It’s the unique view it offers into member/patient health.
Even today, in the electronic age, providers, for the most part, only see the clinical, laboratory, and pharmaceutical data captured by their own office, facility, or health system. Any care that occurs outside their boundaries is often a mystery (even though by now it shouldn’t be), leaving holes in their understanding of the member’s health.
The same is even more true for laboratories and pharmaceutical companies. They see tests given or prescriptions issued, but have no real data on why those tests were ordered or prescriptions written. It’s like knowing there are multiple colors on the Rubik’s Cube but only being able to distinguish one of them. The puzzle will be awfully difficult to solve.
Payers, however, can see all the “colors” because each of those entities submits claims to them. By incorporating all of those data points, along with additional information such as demographics, psychographics, and social determinants of health (SDoH) into their analytics, payers have a far greater ability to manipulate the data to drive more reliable, and more actionable, conclusions.
Deeper predictions
With the right analytics at their disposal, payers can take this breadth and depth of data and use it not just to identify members/patients who are already in the high-risk category but also predict those who are trending toward it but not there yet. The “trendings” are a very important group to identify because payers still have time to change their destiny, improving their health and lowering their own benefit costs.
Take diabetes for example. It is one of the most prevalent chronic conditions in America. In fact, according to the CDC more than 100 million adult Americans are now living with diabetes or pre-diabetes. Not only is it the seventh-leading cause of death in the U.S., it is an incredibly costly disease both in financial and quality of life terms. It can also lead to – or complicate – other chronic conditions as well.
The key issue is that once a member/patient is diagnosed as a diabetic, he/she is always considered a diabetic, and will require greater clinical and financial resources. If payers can unwind their Rubik’s Cube of data to identify which members/patients are trending toward acquiring diabetes, however, they can recommend interventions and work with members/patients (and their providers) to halt or even reverse the trend.
Payers can also use that data to identify which “Diabetes 101” interventions are likely (or unlikely) to work for a specific member/patient based on others who fit a similar profile or persona, so they can create alternatives. Suggesting an expensive medication for a diabetic member/patient who lives in an underserved community and is faced with the dilemma of paying rent or paying for medication is unlikely to drive member/patient compliance. Identifying that challenge through demographic or SDoH data will alert payers to look into alternatives.
Incorporating artificial intelligence (AI) into the analytics makes working with individuals and populations even more effective. It’s like having someone who already knows how to solve the Rubik’s Cube guide you through the process.
The advantage of AI is that it can work through that mountain of diverse data and uncover subtle relationships a human might miss, such as noticing that members/patients who respond that they have a pet on their self-assessments tend to be more compliant to their plans of care than those who do not.
By sharing that information, payers can help providers produce better health results for members/patients, again while reducing their own costs. It’s a win for everyone.
Taking action
Unlike providers, payers also have one other advantage: the financial resources to do something with the information once they have it.
In the value-based care era, providers know they must be proactive around member/patient health, especially for those who fall into the high-risk category. Yet it is difficult for most of them to dedicate the resources required to improve member/patient engagement to a significant degree.
Even mid-size regional payers, however, can usually afford to hire nurses or other clinicians – either internally or by outsourcing the function – to take what the predictive analytics have revealed and close the loop. These payer-sponsored clinicians can monitor adherence to plans of care, contact members who are trending toward increasing risk, and work with social services agencies to overcome SDoH challenges. The top payers will even have clinicians visit members in their homes, when required, to drive real change.
Solving the puzzle
While it may seem that having too much data about members is a first-world problem, the reality is that it can be overwhelming. And, like a Rubik’s Cube, sometimes the more organizations work to make sense of it the further away they get from their desired outcome.
Predictive and prescriptive analytics, especially when supported by AI and machine learning, can help take those maddening twists and turns of data and create a complete, clear picture that helps drive healthcare quality and member/patient satisfaction up while driving benefit costs down. Be sure you’re not the last health payer to discover it.
Mayur Yermaneni