It is not a secret that the healthcare industry is transitioning to more data-driven healthcare for efficiency, cost-savings and improved outcomes. However, the industry is still in its infancy in dealing with big data, according to Dan Riskin, MD, CEO of Health Fidelity, a healthcare data company, as well as a consulting assistant professor at Stanford (Calif.) University and a previous member of the Health Advisory Committee for the first Obama Presidential Campaign.
Big data is defined as very large sets of complex data that become difficult to process using on-hand database management tools. According to Dr. Riskin, the promise of this big data in healthcare is revolutionary.
"Use of big data will ease the transition to authentic data-driven healthcare, allowing healthcare professionals to improve the standard of care based on millions of cases, define needs for subpopulations and identify and intervene for population groups at risk for poor outcomes," says Dr. Riskin.
Big data issues for hospital executives
However, big data can be complicated to work with, especially in healthcare — an industry with intricate data that requires many algorithms. According to Dr. Riskin, big data is very important to three key areas of healthcare: revenue cycle, compliance and quality. However, these areas may face issues with big data down the road, so hospital executives need to prioritize decision-making on how to prepare and report metrics in the future. "Executives have data problems in revenue cycle, compliance and analytics. In each of these spaces they have huge spend as well as initiatives that can feel like impending doom," says Dr. Riskin.
1. Revenue cycle. According to Dr. Riskin, big data for the revenue cycle may become a problem area for hospitals because they have relied primarily on manual labor — physicians, nurses and the coding department — to aggregate the necessary information. While manual processes have worked in the past, they are very expensive. "With ICD-10 set for 2014, the idea of expanding and growing manual processes to deal with big data for the revenue cycle may become overwhelming to healthcare executives”, says Dr. Riskin.
2. Compliance. Many hospitals are planning to expand their manual processes in order to have the necessary data for complying with meaningful use, but this will lead to a massive increase in spending, according to Dr. Riskin. "Right now, most hospitals meet meaningful use stage 1. The manual process for pulling compliance data may be manageable at this stage. When stage 2 begins in 2014, the reporting requirements will increase. How will hospitals manage the increase? This is something that healthcare executives may not recognize yet, but they need to address," says Dr. Riskin.
3. Quality. While hospitals need to track quality data so they can improve patient care, smaller to mid-size hospitals may not be able to hire the same amount of employees as larger hospitals to pull quality data to create dashboards and drive improvement processes, says Dr. Riskin.
"They can't afford what the large health systems are doing now, which is to create and expand manual processes for aggregating this information. However, it is a key driver for increasing transparency of quality levels to patients and driving higher patient volumes," says Dr. Riskin. "If patients know what the hospital's quality level is, the hospital has a strong motivator to increase its quality beyond what it may have been when it was not recorded or reported," says Dr. Riskin. The problem is that the mechanisms for aggregating the data are currently reliant on a human work force, which is expensive and may be not make sense when reporting requirements and data amounts increase.
Solution: Automate the process
According to Dr. Riskin, hospital executives cannot continue to rely on manual processes for reporting revenue cycle, compliance and quality data. "[The reporting] needs to be automated from the point of care onward. Currently, hospitals plan to create infrastructure for manually pulling the measures and using software primarily to report, but the workflow does not need to be so manual. You can use clinical natural language processing software to automatically extract the information," says Dr. Riskin.
The narrative note typically entered into an electronic medical record has a lot of information. Automated software engines with clinical natural language processing can correlate clinical findings and body locations in progress notes, radiology reports, and other documents; extract keywords for revenue cycle, compliance and quality; evaluate electronic discharge notes to assess quality of care; and use semantic features to classify patient smoking status, among other tasks.
The choices that executives make now could change their data experience two years from now, especially as they begin to address ICD-10 and meaningful use stage 2. "They can meet meaningful use stage 1 with low-level analytics just by improving manual processes and using basic software. Does that really sustain them when the requirements will double for each of these areas in a few years? Are they going to look back and wish that they had created an automated infrastructure instead of doubling their staff in each of these areas?" asks Dr. Riskin. "I'm a huge proponent of cost-effective healthcare. I'm a believer that no one should increase cost in the system if it can be avoided. [However,], building up manual processes, even though perhaps a little cheaper than automated processes now, will be much more expensive in a few years," continues Dr. Riskin.
Software services that easily automate the full workflow of reporting for revenue cycle, compliance and quality may not be realistic for all hospitals. Regardless, hospital executives need to carefully assess their health information management and health IT infrastructure and make informed decisions that meet present and future needs.
3 Imaging Issues for Hospital Executives to Address
Study: 40% of Hospitals Plan to Invest in Health IT Solutions
Big data is defined as very large sets of complex data that become difficult to process using on-hand database management tools. According to Dr. Riskin, the promise of this big data in healthcare is revolutionary.
"Use of big data will ease the transition to authentic data-driven healthcare, allowing healthcare professionals to improve the standard of care based on millions of cases, define needs for subpopulations and identify and intervene for population groups at risk for poor outcomes," says Dr. Riskin.
Big data issues for hospital executives
However, big data can be complicated to work with, especially in healthcare — an industry with intricate data that requires many algorithms. According to Dr. Riskin, big data is very important to three key areas of healthcare: revenue cycle, compliance and quality. However, these areas may face issues with big data down the road, so hospital executives need to prioritize decision-making on how to prepare and report metrics in the future. "Executives have data problems in revenue cycle, compliance and analytics. In each of these spaces they have huge spend as well as initiatives that can feel like impending doom," says Dr. Riskin.
1. Revenue cycle. According to Dr. Riskin, big data for the revenue cycle may become a problem area for hospitals because they have relied primarily on manual labor — physicians, nurses and the coding department — to aggregate the necessary information. While manual processes have worked in the past, they are very expensive. "With ICD-10 set for 2014, the idea of expanding and growing manual processes to deal with big data for the revenue cycle may become overwhelming to healthcare executives”, says Dr. Riskin.
2. Compliance. Many hospitals are planning to expand their manual processes in order to have the necessary data for complying with meaningful use, but this will lead to a massive increase in spending, according to Dr. Riskin. "Right now, most hospitals meet meaningful use stage 1. The manual process for pulling compliance data may be manageable at this stage. When stage 2 begins in 2014, the reporting requirements will increase. How will hospitals manage the increase? This is something that healthcare executives may not recognize yet, but they need to address," says Dr. Riskin.
3. Quality. While hospitals need to track quality data so they can improve patient care, smaller to mid-size hospitals may not be able to hire the same amount of employees as larger hospitals to pull quality data to create dashboards and drive improvement processes, says Dr. Riskin.
"They can't afford what the large health systems are doing now, which is to create and expand manual processes for aggregating this information. However, it is a key driver for increasing transparency of quality levels to patients and driving higher patient volumes," says Dr. Riskin. "If patients know what the hospital's quality level is, the hospital has a strong motivator to increase its quality beyond what it may have been when it was not recorded or reported," says Dr. Riskin. The problem is that the mechanisms for aggregating the data are currently reliant on a human work force, which is expensive and may be not make sense when reporting requirements and data amounts increase.
Solution: Automate the process
According to Dr. Riskin, hospital executives cannot continue to rely on manual processes for reporting revenue cycle, compliance and quality data. "[The reporting] needs to be automated from the point of care onward. Currently, hospitals plan to create infrastructure for manually pulling the measures and using software primarily to report, but the workflow does not need to be so manual. You can use clinical natural language processing software to automatically extract the information," says Dr. Riskin.
The narrative note typically entered into an electronic medical record has a lot of information. Automated software engines with clinical natural language processing can correlate clinical findings and body locations in progress notes, radiology reports, and other documents; extract keywords for revenue cycle, compliance and quality; evaluate electronic discharge notes to assess quality of care; and use semantic features to classify patient smoking status, among other tasks.
The choices that executives make now could change their data experience two years from now, especially as they begin to address ICD-10 and meaningful use stage 2. "They can meet meaningful use stage 1 with low-level analytics just by improving manual processes and using basic software. Does that really sustain them when the requirements will double for each of these areas in a few years? Are they going to look back and wish that they had created an automated infrastructure instead of doubling their staff in each of these areas?" asks Dr. Riskin. "I'm a huge proponent of cost-effective healthcare. I'm a believer that no one should increase cost in the system if it can be avoided. [However,], building up manual processes, even though perhaps a little cheaper than automated processes now, will be much more expensive in a few years," continues Dr. Riskin.
Software services that easily automate the full workflow of reporting for revenue cycle, compliance and quality may not be realistic for all hospitals. Regardless, hospital executives need to carefully assess their health information management and health IT infrastructure and make informed decisions that meet present and future needs.
More Articles on Health Information Management:
On Path to Electronic Health Information Exchange — Through an Unexpected Method3 Imaging Issues for Hospital Executives to Address
Study: 40% of Hospitals Plan to Invest in Health IT Solutions