Learning Health Systems (LHS)
In 2007, the Washington-based Institute of Medicine (IOM), a nonprofit, nongovernmental organization that is part of the National Academies of Science, released a book-length report titled The Learning Healthcare System (LHS). This report was the first of a series of dozen reports from IOM’s Roundtable on Evidence-Based Medicine, now the Roundtable on Value & Science-Driven Health Care. This report envisioned the creation of LHS by integrating two disparate fields, clinical research and clinical medicine. The report defined LHS as:
A learning healthcare system is [one that] is designed to generate and apply the best evidence for the collaborative healthcare choices of each patient and provider; to drive the process of discovery as a natural outgrowth of patient care; and to ensure innovation, quality, safety, and value in health care .
Part of the reason for envisioning such a system was that clinical decision making based on clinical practice guidelines were not adequately supported by high quality evidence . Evidence generated by randomized clinical trials are considered as gold standards, but most often they are not generalizable due to rigid inclusion criteria. Moreover, such trials are costly and time consuming. An interconnected system of Electronic Health Records (EHRs) and databases with the ability to share data and generate insight using state of the art information technology and analytics would provide better evidence to guide the decisions made by health systems, care providers, and patients and their families.
Elements of LHS
Four elements essential for LHS are :
1. An organizational architecture that facilitates formation of communities of patients, families, front-line clinicians, researchers and health system leaders who collaborate to produce and use big data;
2. Large electronic health and health care data sets (big data);
3. Quality improvement for each patient at the point of care brought about by the integration of relevant new knowledge generated through research; and
4. Observational research and clinical trials done in routine clinical care settings.
Challenges to Building the Learning Healthcare System
1. EHRs, Data Standards, Interoperability & Computable Phenotypes
A major challenge to LHS is lack of interoperable EHRs. Mutually agreed upon data standards and definitions, such as those developed by HL7, will enable more interoperable systems and will facilitate data exchange. There is also the need for computable electronic phenotypes in order to identify patients with similar conditions.
2.Professional/Health System Interactions
Another major obstacle to LHS system is the friction between the two disparate set of professionals - healthcare providers and researchers. Many healthcare systems are designed to primarily support the delivery of care with research being a secondary objective. Research activities require a different set of infrastructure leading to two parallel systems incapable of informing each other. Establishing organizational policies to bring research mainstream and creating a cyclical system in which research informs care and vice versa, is very important for the creation of LHS.
Research under the LHS model essentially uses clinical data collected during the routine delivery of care. Such type of research has blurred the line between research and clinical practice. The new paradigm of doing research has created new challenges for regulatory bodies, such as Institutional Review Boards (IRBs). IRBs are often wary of research activities conducted using patient data and have sought to maintain the distinction between research and standard of care activities. Quality Improvement (QI) activities aim to improve processes and patient management and are exempt from the requirements of IRBs (such a patient consent) and Health Insurance Portability and Accountability Act (HIPAA). However, there is increasing debate over which QI and LHS studies should be classified as research, and regulations surrounding this issue are complex and may be inconsistently applied [4,5].
The need for customized statistical methods to address the complexities of big data in healthcare is also challenging.
Incorporating patient generated data in EHR can enable better clinical decision making and improve clinical care. Patient can send data from wearable devices to his/her EHR but such a system creates security challenges requiring data security, availability of an audit trail, data backup, and appropriate access to investigators and clinical staff .
A 2015 paper by Friedman et al. lays down the research agenda for high functioning LHS .
Examples of LHS
AcademyHealth - [https://www.academyhealth.org/
Veterans Health Administration - https://www.va.gov/health/
1.Institute of Medicine. The Learning Healthcare System: Workshop Summary. Olsen L, Aisner D, McGinnis JM, eds. Washington, DC: National Academies Press; 2007. Available at: http://www.iom.edu/Reports/2007/The-Learning-Healthcare-System-Workshop-Summary.aspx. Accessed April 4, 2014.
2.Tricoci P, Allen JM, Kramer JM, et al. Scientific evidence underlying the ACC/AHA clinical practice guidelines. JAMA 2009;301:831–841. PMID: 19244190. doi: 10.1001/jama.2009.205.
3.Baily MA, Bottrell M, Lynn J, et al. The ethics of using QI methods to improve health care quality and safety. Hastings Cent Rep 2006;36:S1–S40. PMID: 16898359.
4.Baily MA. Harming through protection? N Engl J Med 2008;358:768–769. PMID: 18287599. doi: 10.1056/NEJMp0800372
5.U.S. Food and Drug Administration. Guidance for Industry: Patient-Reported Outcome Measures: Use in Medical Product Development to Support Labeling Claims. 2009. Available at: http://www.fda.gov/downloads/drugs/guidancecomplianceregulatoryinformation/guidances/ucm193282.pdf. Accessed April 9, 2014.
6.Friedman C, Rubin J, Brown J, Buntin M, Corn M, Etheredge L, et al. Toward a science of learning systems: a research agenda for the high-functioning Learning Health System. Journal of the American Medical Informatics Association : JAMIA. 2015;22(1):43-50.
Submitted by Meenakshi Mishra