Five Reasons why the FDA got it Right with Digital Health

The FDA recently announced its Digital Health Innovation Action Plan.  This plan was both important and necessary.  Digital health differs significantly from drugs and traditional medical devices around which the FDA has developed its structure and policies over decades. Digital health software technologies have far less involvement by clinicians in product development than Pharma or medical device companies.  Innovation and improvements in digital health occur at lightning speed in comparison to other FDA regulated industries and require different approval and oversight strategies (see below).  Multifunctional digital health technology interacting with both drugs and devices creates a need for a new agency framework and workflow for such combined products. The Action Plan is a major step in the right direction because it recognizes digital health’s unique attributes which span both IT and clinical arenas. In addition it recognizes the need for infrastructural and process changes in the agency itself. It will address lingering issues such as multifunctional software (technology which has software qualifying as a medical device combined with that which doesn’t).

1. The Action Plan builds on and improves previous digital health guidances. The FDA has previously released guidance documents on Mobile Medical Apps and General Principles of Software Validation. These were a well thought out and developed group of guidelines with patient safety as a focus. The Action plan includes a plan to issue a new draft guidance by the end of 2017 on mobile medical apps, medical device data systems (used for the electronic transfer, storage, display, or conversion of medical device data), medical image storage devices, medical image communications devices, low-risk general wellness products, and laboratory workflow.  This guidance will reflect  the impact of The 21st Century Cures Act passed by Congress last year on previous FDA guidelines on software. The FDASIA Health IT Report from the Office of the National Coordinator left many unanswered questions about clinical decision support technologies. The Action Plan is in part designed to address that gap by planning a Guidance on Clinical Decision Support by Q1 of 2018.
2. The FDA will have new digital health experts.  The Innovation plan reflects the need for hiring digital health technology experts by the agency.  According to the Action Plan, the FDA’s goal is “is to build a cadre of experts with a deep understanding and experience with software development and its application to medical devices. This new staff will work with reviewers, compliance officers, and others within the FDA to improve the quality, predictability, consistency, timeliness, and efficiency of decision-making on individual products and firms.”  In addition, the agency is to launch an Entrepreneurs in Residence program for “…thought leaders and others with real experience in software development to build and structure the digital health function within CDRH [FDA’s Center for Devices and Radiologic Health] as it – and the market – grows.”  This input from industry leaders will provide the leadership in designing the FDA’s structure and approach in order for it to get digital health addressed in the manner it deserves.
3. A Pre-cert program will be about the developer not the software. A simplistic explanation of this program would describe it as analogous to the TSA’s Precheck Program. According to the Action Plan, this Software Certification Pilot Program  “… could “pre-certify” eligible digital health developers who demonstrate a culture of quality and organizational excellence based on objective criteria, for example, that they can and do excel in software design, development, and validation (testing). Pre-certified developers could then qualify to be able to market their lower-risk devices without additional FDA review or with a more streamlined premarket review.”   This program would both set quality development standards and streamline approval processes. This pre-cert pilot program is admirably designed  “…to leverage customer input to develop a program that can help reduce the time and cost of market entry for software developers that FDA determines reliably manufacture high-quality, safe and effective digital health devices while providing appropriate patient safeguards.”
4. The FDA will work with other global agencies. Last year the International Medical Device Regulators Forum (IMDRF) published a proposal for the clinical evaluation of software as a medical device.  The FDA issued this as a draft guidance and it is expected to adopt the final IMDRF version as a final guidance. This global initiative will allow for faster, less expensive and standardized adoption and oversight of this type of technology.
5. The upgrade approval process will be clarified. The Action Plan calls for a final version guidance by the end of 2017 of its draft proposal of Deciding When to Submit a 510(k) Software Change to an Existing Device. The Plan states that this final guidance “includes guiding principles as well as a simple to follow flow chart, with the aim of setting clear, practical, and transparent regulatory expectations.”

This Action Plan seeks to improve on previous FDA guidances and initiatives.  It recognizes the need for infrastructural changes of the FDA itself to support digital health, and necessary input from industry leaders.  Bakul Patel and the FDA should be commended for developing this Action Plan in response to industry leaders looking for detailed support for improving, crystallizing, standardizing, and streamlining oversight policies. I look forward to downstream developments arising from it and encourage those developers clamoring for a voice at the table to apply for the Pre-Cert Pilot Program.

from The Digital Health Corner http://ift.tt/2veopqW

Five Digital Health Imperatives for Patient Safety

The most discussed issues in healthcare today are cost savings and access to care.  There is no denying their rightful preeminence in the public discourse.  Something which has received more attention in the past but has been surpassed by the aforementioned topics in both lay and healthcare policy press has been patient safety. Perhaps the best definition of patient safety is that of the World Health Organization. Simply put, it is “…the prevention of errors and adverse effects to patients associated with health care.” Patient safety initiatives have taken many forms. Most are aware of public awareness (ex. healthcare worker hand washing) and regulatory clinical requirements (ex. wrong site/procedure/ person surgeries) publicized in the past. Digital tools now present opportunities for unprecedented improvements in patient safety. I would like to highlight a few.

1. Improving health literacy. Health literacy is defined as the degree to which individuals have the capacity to obtain, process, and understand basic health information and services needed to make appropriate health  What does health literacy have to do with patient safety? People who do not know their diagnoses when they leave an office visit or hospital (some estimates are 60%) or do not know the reasons for each medication they are prescribed are much more likely to be non-adherent to medical advice or recommended treatment regimens. In fact, low health literacy is related to higher mortality rates. In a post I wrote here in 2012, I discussed some of the implications of low health literacy. In one study last year examining the extent of use of digital health tools in relation to health literacy, it was found that those subjects with low health literacy used these tools less. Though there is much to be desired regarding the analysis in this study, the authors are to be commended for the number of participants (>5000) and for highlighting how health literacy and digital technology can affect each other.  According to the Department of Health and Human Services, only 12 percent of US adults are health literate. Improving health literacy via digital tools is not only an opportunity it is necessary. The 15 or 20 minute office visit does not afford even the best physician to explain a diagnosis or proposed treatment plan in a digestible way a patient and caregiver deserve. Language barriers compound the health literacy dilemma. There is a well-established legal framework for language access. The cost to providers and inaccessibility outside of the clinic or hospital make mainstream language translation technology insufficient to close gaps in care.  Mobile non-human digital language translation tools will hopefully soon become the preferred alternative.
2. The use of AI in workflow. There’s been a lot of buzz about artificial intelligence (AI). The present popular conversations surrounding AI involve speculation whether AI will replace humans in every possible aspect of healthcare. There are even statistics bantered about as to percentage of physicians, nurses, and others who will lose their jobs to AI technology.  That aside, AI can have a significant role in improving patient safety.  Medtronic, Masimo, and others are working with the Patient Safety Movement Foundation by sharing de-identified data from medical devices in designing predictive analytics programs with the ultimate goal of improving patient safety. Many software platforms exist which claim they improve patient safety but evidence is scarce.  In advertizing their software, many companies equate hospital readmissions with a patient safety metric though strictly speaking it isn’t by the definition above unless it is related to a complication of treatment rendered during the prior hospitalization. Predictive analytics and other AI may however be relevant to patient safety by gathering data from the patient’s EHR (and heaven help us one day from data sets of other systems) to predict both minor and major preventable adverse events. AI platforms have the opportunity to make all the data about a patient come to life, integrate and provide a real-time picture of what is happening to the patient and provide alerts which might change treatment. AI is technology which needs to be incorporated into workflows which themselves are designed with maximum patient safety in mind. In this manner, the human aspect in patient safety is still important.
3. The use of digital tools for inventory and other tracking. Inventory tracking in both the enterprise and ambulatory clinic is an important patient safety issue.  Equipment servicing schedules, drug, medical device, patient and personnel tracking and other logistical considerations are important in patient safety. RFID and other advanced digital technologies have been developed to help in this regard.
4. Improving the electronic medical record (EHR). A 2016 study published in the Journal of Patient Safety demonstrated a 17-30% reduction of in-hospital adverse events in patients with cardiovascular diagnoses, pneumonia or a diagnosis leading to surgery when a fully electronic EHR is used. Specifically the adverse events included hospital-acquired infections, adverse drug events (based on selected medications), and post-procedural events. As this studied playing field has leveled today with the vast majority of institutions utilizing completely electronic records, there exist yet more opportunities for improving patient safety with EHR associated technologies. One hot button issue is that of matching of patients with their own EHR record. The challenge lies in the strong legislative opposition to a unique national patient identifier.  To this end the Office of the National Coordinator has launched a Patient Matching Algorithm Challenge.  Other ways in which the EHR can improve patient safety involve the use of AI (see above). Input from patients in reviewing their records via the patient portal is arguably the first place to start.
5. IT security. IT security is a direct and indirect threat to patient safety. This has been demonstrated with cardiac implantable electronic devices and recent global IT hacking which involved hospital systems. A technology called blockchain was developed in 2008 and has found its biggest interest in the financial industry. The use of blockchain in healthcare has been discussed for a number of years.  However there exist significant challenges in the utilization of blockchain for patient records.

While popular discussions focus on health insurance coverage, shortcomings of EHRs and the health system in general, we must never lose sight of what should be the focus of care itself, patients.  Healthcare must be accessible, affordable, efficient, of high quality with good outcomes and most of all as safe as possible. Patient safety should always be a metric of good care and new programs and technologies.

from The Digital Health Corner http://ift.tt/2u1JmEP

The The Five Biggest Areas of Opportunity for Digital Health

Digital health is unquestionably becoming part of healthcare lexicon and fabric. Electronic health records (EHRs) and personal fitness trackers have helped create awareness through use.  The entrepreneurial enthusiasm for the healthcare space is evident by the volume of digital health incubators, medical school innovation centers,  and angel investors.  Though there has been significant sector investment, the road to success of adoption in the healthcare enterprise has been challenging.  I’d like to discuss what I believe are five areas of significant opportunity for quality technologies.

1. EHRs. According to most recent statistics from the Office of the National Coordinator, use of EHRs has increased from 20% in 2004 to 87% in 2015. EHRs were designed as documentation centers for billing and regulatory purposes. Relevant clinical patient management data workflow was not a priority and remains a major pain point for clinicians today. According to a study in the American Journal of Emergency Medicine ER physicians spend only 28% of their time in direct face to face patient contact and can go through 4000 computer mouse clicks in one shift.  From a provider standpoint. the regulatory and billing data entry should be performed by someone else and relegated to an (almost) invisible part of the EHR.  We need EHRs which are clinically oriented with good user interfaces. Interoperability [defined by the federal Office of the National Coordinator for health information technology (HIT) as the ability of information systems to exchange patients’ electronic health information and use information from other EHR systems without any special effort from the user] is another major pain point that needs to be addressed. .Six years into Meaningful Use we have yet to achieve any significant interoperability of EHRs. There are hospitals within the same healthcare system in many places with disparate EHRs which do not talk to each other or exchange information.  Increasing healthcare consolidation of hospitals has exacerbated the problem of lack of interoperability. Health Information Exchanges (HIEs) have been woefully underfunded and have fallen short of their vision. There remain many opportunities for technologies to assist in achieving true interoperability.
2. Clinical trials. CIOs are constantly inundated with requests to purchase new technologies which will “save money, improve patient satisfaction and outcomes and decrease readmissions.” What is in fact lacking in most cases is evidence for these claims.  The hesitation of many entrepreneurs to embrace the intuitive adoption requirement of proof of claim (which needs to be said should not differ from the adoption of product in any field of endeavor making claims) is the misconception that time-consuming large costly randomized clinical trials are what I am referring to. This should not however translate to “take my word for it” is all you need. I agree that traditional trials are neither practical nor necessary for most tools. Even the FDA has now recognized with thoughtful and cautious restraint a role for ‘real world evidence’ (defined by the legislation as “data regarding the usage, or the potential benefits or risks, of a drug derived from sources other than randomized clinical trials,” including sources such as “ongoing safety surveillance, observational studies, registries, claims, and patient-centered outcomes research activities.” in the approval process of drugs. Thus, the opportunity for trials utilizing digital registries, mobile clinical trial platforms, quality communications and analytics tools is significant.
3. Artificial Intelligence (AI). One early definition of Artificial Intelligence (AI) in medicine (1984) was “…the construction of AI programs that perform diagnosis and make therapy recommendations. Unlike medical applications based on other programming methods, such as purely statistical and probabilistic methods, medical AI programs are based on symbolic models of disease entities and their relationship to patient factors and clinical manifestations.” Today a broader definition may be applied: “the simulation of human intelligence processes by machines, especially computer systems. These processes include learning (the acquisition of information and rules for using the information), reasoning (using the rules to reach approximate or definite conclusions), and self-correction.” The use of artificial intelligence in medicine has been the subject of intense and rapidly growing interest in medical, computer science, and business arenas.  The market growth of AI is based on its projected impact on both technology and non-technology sectors. There have been arguments for and against the inevitability of replacement of physicians by AI technologies for a while now. The debate continues. BASF declared “We don’t make the household product, we make the product better.” An analogy can surely be made with AI. It runs in the background of technologies already in use but will make them run faster and more importantly will add a dimension of relevance of incoming data.
4. Personalized medicine. The National Cancer Institute’s definition of personalized medicine is “…a form of medicine that uses information about a person’s genes, proteins, and environment to prevent, diagnose, and treat disease…” Personalized medicine is medical care directed in whole or part from information specific to an individual.  Discoveries in the area of the genetics of cancer have resulted in the development of drugs no longer targeted towards an anatomical location but a specific genetic marker. A landmark clinical trial in which drugs are given solely on the basis of genetic markers identified in the cancer tissue itself is the NCI-MATCH Trial (Molecular Analysis for Therapy Choice). “Patients with advanced solid tumors, lymphomas, or myeloma may be eligible for MATCH, once they have progressed on standard treatment for their cancer or if they have a rare cancer for which there is no standard treatment.” The role of personally derived connected data (from sensors external or internal to the body) will also facilitate personalized medical care. Opportunities thus exist for life sciences and technology companies to develop products for this new therapeutic approach.
5. Social Media. An early observational study of synergistic impacts of healthcare and social media demonstrated that personal experiences and not data drive social media healthcare discussions. One early survey of physicians on their use of social media found that “85% of oncologists and primary care physicians use social media at least once a week or once a day to scan or explore health information. Sixty percent said social media improves the care they deliver.” The potential for social media to disseminate information from published clinical trials, the exchange of professional education among peers, and discussions surrounding disease states is invaluable.  To be sure there exist professional and regulatory guidelines for the use of social media for providers, vendors and other healthcare stakeholders.  Social media open platforms in healthcare have proven successful for patients, caregivers and others.  Examples are Treatment Diaries, patientslikeme, and   Potential opportunities here involve recruitment of patients for clinical trials, gleaning real world evidence data from discussions.

By no means is this a complete discussion of opportunities for digital health. These are what I consider the ‘biggest bang for the buck’ ones doable today. I look forward to comments and the sharing of experiences from others. As a consultant I am amazed on a daily basis at the high quality clinical, financial and personal experience energies devoted to the development and advocacy for digital health tools. Bring it!

 

from The Digital Health Corner http://ift.tt/2s0GloR