Patient similarity is the concept of researching the most and least effective treatments based on the health records of like individuals with comparable health conditions.
Patient similarity analytics can support a predictive model of health issues and treatment through accessing data about similar patients in similar situations and the treatments that worked most often for them. Patient similarity is a fundamental component of evidence-based medicine, which seeks to improve treatment by incorporating the most current research available with statistics and information from the records of huge numbers of patients.
Physicians generally aim to match best treatments based on their education, research and personal experience from previous patients. This narrow, individual perspective provides doctors with a relatively small window, in comparison with the number of patients with similar conditions treated statewide or nationally. Data analytics, including patient similarity algorithms, can be applied to records from the health indicator warehouse and other health record stores. This analysis enables a broader scope and larger sample size. In turn, the larger sample improves the confidence of analysis for more accurate diagnoses and prognoses, and more appropriate treatment of health issues with fewer occurrences of medicine interactions and other complications.
Patient similarity can also be used for predictive analytics on patients for proactive rather than reactive care. Predictive analysis can help healthcare professionals foresee potential health concerns and long-term health effects from different care options, rather than just the best treatments and likely short-term results of treatment.
Patient portals, which are web access points for patients of healthcare services to use self-serve health IT., provide access to healthcare information to patients themselves. These portals, such as PatientsLikeMe, enable greater patient education and involvement and, along with evidence-based medicine and patient similarity, enhance patient-provider communication and patient outcomes.