For caregivers to use data properly, they must understand that much of the actionable data lives outside the healthcare system.
Healthcare experts are the first to say that data hold the key to developing an understanding of patients and their behaviors. Unfortunately, most of the data informing this understanding comes from limited or incomplete sources, such as large scale claims databases and customer relationship management (CRM) systems. According to a 2015 study from McKinsey &
Company, 95% of patient data resides outside of the care management system. This gap in data presents a significant barrier, as well as an opportunity, to a gain a more nuanced, actionable understanding of patient behavior. While this may seem like a minor issue that only affects clinicians and pharmaceutical companies, it ultimately reaches back to the patient, and many don’t receive the maximum benefit of care as a result.
To improve our understanding of patients, we must merge traditional data with new sources as a way of identifying influential behavior, attitudes, and life circumstances that lead to patient decision making and ultimately to clinical outcomes. For example, a 2016 study by the SAS Institute merged third party consumer data with claims data to predict healthcare utilization risk and costs.1 What they found was that television usage patterns, mail-order buying habits (including the purchasing of prescriptions via mail) and investments in stocks and bonds were all variables with predictive power helping to understand a patient’s risk for particular outcomes and the related cost.
Barriers and life circumstances are also critical. It’s no surprise that financial barriers play a major role in a patient’s decision process. A 2007 study evaluating hospital re-admission risk after heart attack found that patients facing financial barriers when attaining healthcare services or prescriptions were more likely to be admitted compared to those without barriers.2 While financial barriers will always be an issue when seeking medical care, being able to predict a patient’s actions based on his or her financial situation can greatly change the preemptive measures that will be taken during the diagnosis and treatment process. Similarly, understanding a patient’s social circumstance can shine new light on their future healthcare needs. For example, living alone or feelings of loneliness are often identified as hidden drivers of disease. This was found by a 2014 study by Cacioppo and Cacioppo which showed that older adults who reported feeling lonely were more likely to experience declines in mental health and physical well-being.3 So, if a patient lives alone, a physician may prescribe and recommend specific interventions which can greatly enhance their medical outcomes later in life.
Symphony Health Solutions (SHS) is committed to integrating and using data from inside and outside the system, and this commitment is evident in two key areas. First is our technology: SHS has made significant investments in building a technology infrastructure to support integration of all types of data with HIPAA compliance. Second is in our research and analytics: For example, in collaboration with SHS’s Audience & Media practice, the Symphony Health Scientific Studies team uses a broad range of anonymous socioeconomic data such as healthcare education, ethnicity, income, work environment, recreational activities and more to paint a more accurate picture of the patient journey. This data is typically inaccessible through traditional methods, but allows healthcare providers to better analyze and predict patient behavior, including their willingness and capacity to commit to prescribed therapies.
While this data is undoubtedly helpful, how can it be put into place to invoke change? We know that for almost every disease, the earlier the detection and diagnosis, the better the outcome. This is no different for heart disease, a chronic illness that impacts millions around the world. Congestive Heart Failure is classically categorized by NYHA Stages I-IV or AGA/ACC stages A-D. However, most patients aren’t diagnosed until stages III or C, meaning most have been living with the disease for some time before diagnosis and treatment. More comprehensive data and information, critical to the patient’s history or “journey,” such as family history, lifestyle and comorbid conditions (like diabetes or anemia) can lead to earlier detection. For instance, a main symptom of congestive heart failure is fatigue, which is not a telling symptom to the average patient.
However, with the help of other data a physician can triangulate their historical information with current signs and symptoms to identify heart disease at an earlier time. This is significant to a patient suffering from unrecognized heart disease who can have a materially improved quality of life with the help of early detection and appropriate treatment.
It’s an exciting time for healthcare as patient data has never been more accurate and accessible through the help of non-traditional information, wearable technologies and the cloud. The ability for the healthcare industry to build a better understanding of the patient through data is helping improve prevention, diagnosis, and quality of care. To make this more of a reality, it’s critical that the healthcare industry from practitioners to pharmaceutical companies look to new ways to understand patients that go beyond traditional data sources. In doing so, the industry will be able to provide earlier detection, more effective treatments, and ultimately strive toward the promise of longer, healthier lives for patients.
1. Garla S, Hopping A, Moaco R, Rittman S. What do your consumer habits say about your health risk? Using third-party data to predict individual health risk and costs. 2013; SAS Institute, Paper 170-2013.
2. Rahimi AR, Spertus JA, Reid KJ, et al. Financial barriers to health care and outcomes after acute myocardial infarction. JAMA. 2007;297(10):1063-72.
3. Cacioppo and Cacioppo. Social relationships and health: the toxic effects of perceived social isolation. Soc Personal Psychol Compass. 2014; 8(2): 58—72.10.1111/spc3.12087.