Biomedical Engineering - E-health Methods and Applications

Completed notes of the course - Prof. C.

Complete course

m-Health, mobile applications and regulation Class EHApp Type Lecture Materials Date Riguardare Smartphones are incredible devices because they have capatibilites to be transformed into a medical device for a number of reasons, or to exploit GPS location from the phone for emergency services. The potential for smartphones is on decision support systems, telemedicine, lifestyle wearables, patient education, remote monitoring etc.; the applications could be a lot.m-Health, mobile applications and regulation 1 @23/09/2021 Barriers for physicians to adopt digital health Software as Medical Device What is a medical device What is an Active device Defining a medical device Medical Device Software Guidance on qualification and classification of software Classification process of medical devices Conformity assessment procedure When is clinical evaluation needed? Clinical data What is equivalence? Clinical evidence of MDSW Scientific validity Technical validation Clinical Validation Expert Panels Post-Market Surveillance ������ m-Health: a medical and public health practice supported by mobile devices, such as mobile phones, patient monitoring devices, personal digital assistants, and other wireless devices. Wearables are part of mobile health and they increase in popularity since they contain additional sensors for more advanced applications, always using the smartphone as an interface of a smartwatch with the display embedded. All these devices are not meant for medical conditions but for fitness purposes, but anyway they are capable of ECG recording capabilities. This leads to misunderstandings because the consumers might be misled to think that it is a medical technology even though it is not. ������ The mobile health is the future of healthcare because it allows for data collection in real time, empowering individuals to assume a more active role in monitoring. Usually, the doctor will instead check on you at his office one or two hourse every few months. So the continuous monitoring might be a disruptive innovation. Not only for chronic patients, but to monitor also lifestyle and wellness of an healthy person. Also the medical guidelines that are rewritten every 5 years, (constitute the basis for best practice), are starting to talk about the smartphone use on telemonitoring and other smartphone applications, since they see the advantages in the use of these new technologies. ������ The problem with wearables and apps is that not all of them are medical devices (actually very few of them is)m-Health, mobile applications and regulation 2 Inversion in the patient paradigm: the individual presents own health data to the physician for clinical intepretation. The problems in this case could be the low specificity in the results and diagnosis that a wearable gives. There is a broad range of apps that need regulation, control, validation also in informing the patients that sometimes he can’t rely on the app so much if it is not validated. Usually with drugs the filter between the prescriptions and the patients is the doctor, while with apps there is no intermediary. Apomediation: process of disintermediation, where previous intermediaries (healthcare professionals) are bypassed by new apomediaries (web, google, on-line groups) in guiding the access to health information to citizens . There are of course benefits in the access to information but many risks in the trueness of the results because you can’t be sure of a reliable source. With apps in the stores, it is the developer that decides the category (primary and secondary) in which the app will be listed! With the ranking algorithms what happens is that apps that are “first” are the most popular but not necessarily the most valid... Usually the most reliable apps result from collaboration among developers, physicians and experts in health law. The majority of app developers have no formal medical training and are unaware of patient safety issues ; if your app is intended to have a medical claim, you must undergo regulatory processes before selling it. Barriers for physicians to adopt digital health Technology prescription I have to be sure that the technology is accurate in measurements, otherwise it will be never adopted. Security Data integrity Data interpretation If the patient will acquire the data by itself, I need to be sure that the data captured is good enough and representative of reality, otherwise I can’t interpret those data and it’s useless Automated intepretation, what is the reliability of automated intepretation? m-Health, mobile applications and regulation 3 Committment to review and interpret data, if I suggest something to the patient, I assume that if the patient is using the device and sending me the data, I will have to look at it, but when? The physician needs more commitment Patient-physician communication: it implies a different way to communicate with the physician, what about the privacy in digital messaging apps like Whatsapp that is being used nowadays? The communication must be traceable somehow if something happens to the patient, this generates some barriers to implementation Lack of reimbursement Software as Medical Device Software has always been integral part of medical devices, and not considered separately from its dedicated hardware. With the development of new digital health solutions, software is acquiring more and more its “own life”, being platform- independent and thus potentially acting as stand-alone medical device. The previous directive started in 2007 for softwares listed by itself and not connected to a medical device. But it was not yet a regulation. Nowadays for softwares there is a regulation of 2017/745 (MDR fully in force on May 26 2020), but what is the difference between regulation and directive? ������ Regulation → binding legislative act that must be applied in its entirety across EU. ������ Directive → legislative act that sets out a goal that all EU countries must achieve, but it is up to the individual countries to decide how, it is not automatically adopted and each country could also change somehow the implementation in its own legislation The regulation that came into play in May 2020 is actually already old because it takes time to be put in place and the businesses are already pressuring the policy makers to issue a new regulation, especially to target the AI-driven softwares, that they believe are endangered by the 2017/745 Regulation. There is another regulation of in-vitro m-Health, mobile applications and regulation 4 diagnostic medical device that will enter in May 2022, and these two regulations change everything in the market. There are also guidance documents that try to explain in depth how to apply a certain rule with practical examples, but they’re not binding. What is a medical device ������ 'medical device' means any instrument, apparatus, appliance, material or other article, whether used alone or in combination, including the software necessary for its proper application intended by the manufacturer to be used for human beings for the purpose of: —— and which does not achieve its principle intended action in or on the human body by pharmacological, immunological or metabolic means, but which may be assisted in its function by such means. diagnosis, prevention, monitoring, treatment or alleviation of disease + prediction and prognosis (in red it’s being introduced by the new regulation). With prediction and prognosis, medical device will also include everything that tracks a vital process or anything that is aimed and has the purpose of predicting something . These two terms are usually associated to machine learning, so whatever algorithm you’re introducing that has a goal of prediction, risk stratification (high risk patient for example) so that the clinical path is then different because of the medical guidelines, so this software needs to be certified. In 2017 AI was still new in medicine, so they tried to cover also new implementations that would have come in the future. Now the European commission is working on a new regulation for AI since this one of 2017 doesn’t cover all the aspects of AI. Since a model is continuously trained and it’s dynamic, then if I certify a software that changes everytime, how can I regulate it? Do I need to repeat again the measurements, because the software has changed? If you want to use a computer for aided assistance to the physician, it needs to be certified and proven valid. Even things that are very standard, for example formulas that doctors use to produce risk scores. If this calculation is made on a paper then it’s fine, if you put it on a medical app then it needs certification... because it’s a medical app for clinical assistance to the physician.m-Health, mobile applications and regulation 5 All these changes will shift a lot of products currently regulated as class I into a higher class of risk automatically. These rules apply to any SW installed and used by users in the EU. What is an Active device ������ Device which operation depends on a source of energy other than that generated by the human body for that purpose, or by gravity, and which acts by changing the density of or converting that energy. Software is also deemed to be an active device . ������ ‘active implantable medical device’ means any active medical device which is intended to be totally or partially introduced, surgically or medically, into the human body or by medical intervention into a natural orifice, and which is intended to remain after the procedure. High risk medical device, even if it is not active because it has to remain in contact with live skin and it is very dangerous. ������ ‘in vitro diagnostic medical device’ means any medical device which is a reagent, reagent product, calibrator, control material, kit, instrument, apparatus, equipment, or system, whether used alone or in combination, intended by the manufacturer to be used in vitro for the examination of specimens, including blood and tissue donations, derived from the human body. in vitro diagnostic medical device is not covered in the 2017/745 regulation because a new regulation only on this type of devices is coming in May 2022. Old regulations didn’t consider all the advancements of proteomics, gene testing, DNA testing that were unregulated or not regulated properly Defining a medical devicem-Health, mobile applications and regulation 6 The first question is if the software is a medical device or not. If yes of course you need to classify your device. If it is not intended as a medical device you can’t enter the market of medical devices, at best the wellness and wellbeing. For classification I need then to give documents on my device. It’s important to assess if your software is actually a medical device or not: For industry and developers: to avoid incurring in penalties due to false claims to guarantee safety, privacy, accuracy to properly enter in certificated medical device market sector For physicians: to avoid liability problems by using not approved software inside the hospital to be aware of the effective intended use of a software, in particular if it is recommended to a patient For patients: to fully understand the reliability and accuracy of what they are using Medical Device Software ������ Medical device software is software that is intended to be used, alone or in combination , for a medical purpose as specified in the definition of a “medical device” in the Medical Devices Regulation. If the software creates new information, it is considered a medical device, as long as the information created is intended to have a medical purpose . For example image processing software that retrieve the exams and visualize the images is not a medical device, but the part of the software that could do some segmentation is a medical device if that information is claimed to be used for diagnosis. Software that does not meet the definition of a medical device but is intended by the manufacturer to be an accessory to a medical device, falls respectively under the scope of the MDR or IVDR. m-Health, mobile applications and regulation 7 Software that is driving or influencing the use of a medical device is covered by the MDR or IVDR either as a part/component of a device or as an accessory to a medical device. Example: software used to operate a device (app controlling a medical device) Guidance on qualification and classification of software The Guidance is the document that if followed will precisely qualify your product as a medical device or not; it doesn’t have to be followed strictly by the manufacturers because it is not part of the regulation itself, but it is very useful to understand if it fits in the category. Is software an MDR Annex XVI device, an accessory to a MD, or a SW driving or influencing the use of a MD? Annex XVI products are those for which a manufacturer claims only an aesthetic or another non-medical purpose but which are similar to medical devices in terms of functioning and risk profile. These products were previously unregulated, like for example body implants, botox or contact lenses, electromagnetic brain stimulation. Any software related to these devices, also the SW needs regulation. If yes → Is software performing an action on data different from storage, archival, communication, simple search or basic arithmetic? If yes → Is the action for the benefit of the individual patient? If yes → Is the SW a MDSW according to the definition of this Guidance? If yes → Once we know if our software is a medical device, we have to go through the process of risk assessment and all the other steps that lead to CE mark. The regulation process for MD is completely different from vaccines and drugs, since the process of conformity assessment is made by private companies (for vaccines it’s a government agency). So you pay this company that will assess your device and give you the certificate if compliant to all the regulations. If your device is rejected, you could go to another notified body without any trace of conformity not passed previously until you pass the assessment. These notified bodies should be vigilated by the national regulatory agency of health; of course this is done by samples and not all the certifications that notifided bodies issue every year.m-Health, mobile applications and regulation 8 Classification process of medical devices The strictest rule applies , if the software is connected to a class IIa and the software is class I, both the MD and SW will fall into the stringest category i.e. class IIa. Software which drives of influences the use of a device, falls within the same class of the device while, if software is independent of any other device, it is classified in its own right Rule 11 in the regulation 2017/745 is the one dealing with softwares. Software intended to provide information which is used to take decisions with diagnostic or therapeutic purpose is classified as Class IIa , except if such decision have an impact that can cause death (class III) or serious deterioration or surgical intervention (class IIb). Software intended to monitor physiological processes is classified as class IIa , except if it is indended for vital parameters monitoring, then it’s class IIb. Class I: all other medical device software, but it is difficult to find a software that doesn’t follow the previous intended purposes in health. Devices for contraception are Class IIb (app for fertility in women) while closed loop systems are class III (glucose levels monitoring pump). Nowadays the majority of softwares is classified in class I, with the new regulation they will fall into class II. For example AR glasses that provide information to the surgeon, with MDD is class I; with MDR is class IIb, because incorrect information display could lead to reoperation of the patient. The expected user population also changed the classification, if the device is intended to be used on high risk patients, or low risk patients. Conformity assessment procedure We have to produce a documentation explaining how the software in organized in key functional elements. Then produce pre-clinical and clinical data : software verification and validation i.e. prove that it has a benefit on the purpose that you’re intending with the device. Important to keep in mind specifically the intended purpose. It shall also address all of the different hardware configurations and operating systems. When is clinical evaluation needed?m-Health, mobile applications and regulation 9 If our claim implies a clinical benefit , something that I can measure, then I need clinical evidence within its own conformity assessment. How much clinical evidence is defined? ������ Sufficient amount and quality to allow a qualified assessment of whether the device is safe and achieves the intended clinical benefit, when used as intended by the manufacturer. Sufficient amount is of course subjective to the notified bodies and might be a problem of standardization of conformity assessments. Clinical data Information concerning safety or performance that is generated from the use of a device and is sourced from the following: clinical investigation(s) of the device concerned, clinical investigation(s) or other studies reported in scientific literature, of a device for which equivalence to the device in question can be demonstrated , reports published in peer reviewed scientific literature on other clinical experience of either the device in question or a device for which equivalence to the device in question can be demonstrated, clinically relevant information coming from post-market surveillance, in particular the post-market clinical follow-up; What is equivalence? Equivalence was used to certify new devices based on a very old study because it was seen that the two devices were equivalent. ������ Now a manufacturer of a device demonstrated to be equivalent to an already marketed device not manufactured by him, may not perform a clinical investigation, provided the following conditions are fulfilled:m-Health, mobile applications and regulation 10 ������ the two manufacturers have a contract in place that allows the manufacturer of the second device full access to the technical documentation on an ongoing basis. The concept of equivalence has been narrowed a lot and it forces the manufacturer to usually run a clinical evaluation assessment for its own device. Equivalence now has to be applied on three different fields: Technical: device is similar in design and specs Biological: same material in contact with same human tissue Clinical: used for the same clinical condition or purpose Clinical evidence of MDSW Scientific validity The extent to which the MDSW’s output (e.g. concept, conclusion, calculations), based on the inputs and algorithms selected, is associated with the targeted physiological state or clinical condition . This association should be clinically accepted or well founded, through literature research, professional guidelines, proof of concept studies.m-Health, mobile applications and regulation 11 Technical validation Demonstration of the ability to accurately, reliably and precisely generate the intended ouput through the production of clinical evidence. Clinical Validation Demonstration of the ability to yield clinically meaningful output (positive impact on different aspects) in accordance with the intended purpose . The benefit of such device has to be on the health of an individual expressed in terms of measurable, patient-relevant clinical outcomes; the impact can also be in improving USABILITY and reliable use. This is also considered a positive impact you shouldn’t produce only a hard impact on patient alone Expert Panels They have the role in checking the evaluation performed by the notified body specifically in answering to the clinical evidence being sufficient in quality and amount, provided by the manufacturers. They will scrutinize only devices in class III mainly but also class IIb active. They have 21 days to decide if they want to go into more detail of the evaluation, so they then have 39 more days remaining; in this case it means that the evidence provided by the manufacturers is not enough. How to decide? Novelty of the device Major clinical impact If for a specific category or group of devices there has been either a significant adverse change in the benefit-risk profile or a significantly increased rate of serious incidents Their role is consultive, they send back the clinical evaluation assessment and if for example they say it’s not enough evidence, the notified bodies can still go on with the CE mark because in the end they decide. Of course this will be visible and the notified body has to justify why it didn’t ask for new evidence. Another problem of digital health solution is how to measure and validate its benefit. It’s different from a drug because RCTs are no more the golden standard in evaluation. What is a placebo for digital health in a trial? It’s really tricky to think of these details. Post-Market Surveillancem-Health, mobile applications and regulation 12 Risk management system Unique device identifier (UDI) also applies to medical software and it will be included in the European database on medical devices (Eudamed); this is the result of big pressure of the industry to make the process all a lot blurrier, also with the new AI regulation that will smooth what has been done with the 2017/745. App and software recalls is much less severe than the recall of medical device; usually a patch can fix bugs, but it’s not an automatical update and you are not sure that all the users will have downloaded and fixed their app. It’s easier to be fixed but more difficult to be tracked in the reissuing.m-Health, mobile applications and regulation 13 Digital Health for monitoring adherence Class EHApp Type Lecture Materials Date RiguardareDigital Health for monitoring adherence 1 @06/10/2021 Terminology Five dimensions of adherence Measurements Pharmacoepidemiologic Methods for self-report subjective measurement Questionnaires and Scales Solutions to track adherence Packaging solutions Electronic medication packaging (EMP) Smart boxes Smart blisters Electronic medication trays Pill tracking systems Alarm-based solutions Wearable devices Digital technologies for controlling adherence graph Hypertension patients ESH Care InTake Care Unintentional non-adherence Intentional non-adherence The OODA loop How to improve orientation? How to gain engagement? Mantaining engagement Terminology ������ Adherence: the extent to which the persons’ behavior (including medication- taking) corresponds with agreed recommendations from a healthcare provider. It’s a more general definition and doesn’t strictly include only medications Definitions of nonadherence Primary : the frequency with which patients fail to fill prescriptions (fill the bottle with a number of pills because the insurance gives you the bottle) when new medications are started, so I don’t get the drug Secondary : I get the drug at home but I’m not taking it as I’m supposed to take it Compliance: the act of taking the medication on schedule and as prescribed Persistence: the duration of time from initiation to discontinuation of therapy Adherence is considered as a cronic pathology because of its impact on healthcare. Poor medication adherence is a cause of morbidity, increased hospitalization, mortality and increased healthcare and waste of money. 50% of patients in the world with long-term illnesses don’t take their medications as prescribed. Even with really severe and life-threatening diseases, patients have high non- adherence rate. Apart from the problems on the patients, there is also missed income estimated as 1/3 of the total income of the pharma market. Even a small solution could get to you a very small share of this income. Five dimensions of adherence Social/Economic → Age & race, but also low literacy because you may not understand the real benefit of taking a medication. Patient-Related → Forgetfulness, treatment anxiety, misunderstood instructions, lack of motivation. Therapy-Related → side effects or changes in therapy or the complexity of the regimen in case of chronic patients. Also the treatment duration can affect on the Digital Health for monitoring adherence 2 motivation to adhere to the therapy. Condition-Related → if you have a particular medical condition like depression, probably you’re not stimulated to take pills. Your medical condition like disability can also impact on adherence. HealthCare system → sometimes the doctor cares about prescribing you the drug but less about checking or stimulating the patient in adhering to the treatment. Healthcare professionals often don’t have the time to take care of this, but only to check you on the health. Maybe if the drugs are free you’re not motivated to take it. What if you don’t take it and it starts becoming an expense, maybe it’s a form of incentive (?) One way to try to control the adherence is to MEASURE it. Measurements Subjective : self-report by the patient and healthcare professional assessments, via a questionnaire . The drawback is that the patient doesn’t remember how many times he forgot the pill in the previous months or he tends to lie about it. Objective : pill counts, electronic monitoring, secondary database analysis and biochemical measures The best way is to have both the measurements together to exploit both the advantages. Direct measures: measurement of the drug or its metabolite concentration in body fluids, randomly or at specific intervals, or by direct patient observation. This direct measure is expensive, instrusive and the patient could take the pill only the day before the test and elude the measure (white coat adherence) Pharmacoepidemiologic Through some databases you can try to infer the adherence information , this is done usually on a group-based or population-based and for example you analyse insurence databases or electronic health records . Administrative data to understand the adherence. These tools assume that because we’re based on prescription and refilling of the drug, we could link these two events to the fact that the patient is taking the medication or not, Digital Health for monitoring adherence 3 but the patient could just refill the bottles and never take one of them. Usually there are two ways to calculate adherence using retrospective data: MPR : medication possession ratio number of days of medication supplied/number of days in refill interval. Doesn’t account for discontinuation at the end of the refill interval, so it is a little biased; if you fill the medication before completing the previous round, MPR might be above 100% PDC : proportion of days covered number of days that the medication is available / number of days in a time interval (refill interval); it avoids the ratio > 1 of MPR when the refill is prior to completion of medication. If you count the days then it can also include periods of discontinuation These two values assume that the patient is taking the pills at home, doesn’t take into account the change in insurance or if you don’t have the insurance because you lost your job, or the patient died. These data are not covered in the databases but anyway it’s still a cheap method because you’re not collecting other data and you’re just analysing already existing DB. Hawtorne effect is the bias when the patient knows that he’s being a part of the study and changes his/her behaviour. Methods for self-report subjective measurementDigital Health for monitoring adherence 4 Patient-kept diaries Patient interviews Motivational interview: not only understanding adherence but motivating you into adhering. Problems of time and also competences of the physician. Questionnaires and Scales There are many questionnaires that target specific things you want to assess. Morisky 8-item medication adherence questionnaireDigital Health for monitoring adherence 5 It is one of the most valid questionnaires to measure a good or bad predisposition of a patient at risk of non-adherence. We can build personas and clusters of patients that are stimulated accordingly. Solutions to track adherence Packaging solutionsDigital Health for monitoring adherence 6 Reminder in the blister on the packaging to know when you’ve taken it or missed it (like anti-conceptional pills) Electronic medication packaging (EMP) Smart boxes Electronic devices integrated into the containers in which pills are dispensed, for example MEMS is the most commonly used EMP device in medication adherence studies, because it records how many times it has been opened and transmit even directly the information to the physician. Drawbacks: open by mistake doesn’t mean that the pill has been taken when you open the box reminding the patient that he is being surveilled expensive Smart blisters Smart packagers are now coming to Europe (here we use blisters and not boxes of pills). These blisters have circuits attached that when it is broken to take out the pill, a count is recorded. Again you have to add the cost of packaging, accidental ruptures, additional cost for the patient. Electronic medication trays It opens only when it is the right time to be opened and the pill taken, and also a reminder on the smartphone to remind you to take the pill Stress to the patient need to refill correctly manage the change in therapy accordingly when necessary Pill tracking systemsDigital Health for monitoring adherence 7 You swallow the pill and a patch on the stomach detects when the pills pass through and an app records that you’ve taken the pill. It’s more difficult to accept that the patient needs to swallow small sensors. skin rash for patch because you keep it on for a long time delay in detection need a smartphone patient’s acceptance Alarm-based solutions Text messaging Smartphone with a medication adherence application Interactive voice response calls plus personalized messages Wearable devices Smartwatches, through accelerometers or gyroscopes to detect gestures and train the system to recognise particular movements like taking the medication (raising the hand towards the mouth for example) Digital technologies for controlling adherence graphDigital Health for monitoring adherence 8 Adherence also happens in the data transmission, because in self-report subjective measures, the patient has to send the data and remembers to switch on the system. Sometimes the patient forgets to do so and you lose the information. Hypertension patients it has been found a correlation between the risk of stroke and the level of systolic blood pressure, you reduce the risk by 10% if you reduce as small as 2 mmHg in BP. It is the disease which has the highest non-adherence. Solutions for these patients have to be based also on the age of the patient. ESH Care One solution is the ESH CARE, an application where the patient is asked to insert manually the values of his home measurements. The data can be sent to the doctor and he will receive data of the patients involved and checking the blood pressure with different graphs and indexes. The use of this platform resulted in an increase of the patient’s adherence and the reduction of blood pressure. InTake Care Dashboard of the physician that sets the alarm, this is transferred to the smartphone of the patient based on what the doctor has decided. There is a manual confirmation, taken or not taken, but also automatic gesture recognition and another tool related to the use of pictures of the blister in order to recognise if is the right medication and the number of pills taken. Moreover there is a document of assessment for adherence and trying to engage and motivate. A bracelet with accelerometer tracks the motion of the arms when taking the pills; Unintentional non-adherence It is usually related to some form of forgetfulness or confusion. The result is that patients may forget to take a dose, take the wrong amount, repeat a dose, or take the incorrect drug. Some solutions are refill reminder programs, decreasing the complexity of management of the disease. Auto-refill programs may help you in sending the pills at home as soon as you finish the previous box. Digital Health for monitoring adherence 9 Intentional non-adherence It’s the patient that decides that he’s done with taking the medication, these people don’t take medication because they don’t see an effect on their health. It’s a behavioural change that needs to be actuated (later in the course). The OODA loop It’s a commonly used model to describe decision making activities: Decision markers observe the information available, orientate the information to the patient’s life and the specific context, decide which are the available actions and then act . The barrier of this OODA loop is that many times the problem is that the patient doesn’t understand the complexity of the treatment that is being provided to him or still do not understand the impacts of missing the treatment. How to improve orientation? Simplify drug regimens can be a solution, for example multipill that replaces the multiple single pills that you should take so that it’s easier for the patient to follow the regimen.Digital Health for monitoring adherence 10 Name of the drugs are also important, the shape, the color or the smell. Shape perceptions are from a designer POV a solution How to gain engagement? The way the information is presented so that the patient is driven by a clear understanding and can orientate himself with it. Understanding the context Develop an emotional connection Information that users can relate to Mantaining engagement Maintaining engagement involves topics that minimize impact on patients’ lifestyle: Convenience Flexibility of use Transportability Time taken Complexity of useDigital Health for monitoring adherence 11 Patient empowerment Class EHApp Type Lecture Materials Date Riguardare Patient empowerment came out in the last decade for policy makers, it is considered the basic factor to allow the care delivery as it is given nowadays. In order to minimise costs and continue to give the therapy as it is today, the patient has to be the center. The chronic patients should learn how to manage their chronicity themselves and go into a process of shared decision-making with the physician This is the idea behing empowerment. There is still a lack of consensus on how to reach this goal it’s an ambiguous concept unclear how it is related to other concepts such as involvment, participation, enablment. Definition: “The people have the right and duty to participate individually and collectively in the planning and implementation of their healthcare.”Patient empowerment 1 @13/10/2021 Empowerment = Enablment + Engagement Behavioral part How to measure patient’s activation? Challenges to patient enablement Basis of user characterization and persona creation Personalization and personas Persona development The empowered patient Ideal Patient-Centered Care Approach (PCC) Self-management Shared decision making Coaching Four fundamental components to the process of empowerment: Undestanding his role acquisition of sufficient knowledge to engage with their healthcare provider patient skills for a self injection or dosage the presence of a facilitating environment to create stimuli of easier empowerment Based on these components you have the defition of empowerment : A process in which patients understand their role, are given the knowledge and skills by their health- care provider to perform a task in an environment that recognizes community and cultural differences and encourages patient participation. Another definition from the EU joint action on patient safety and quality of care: Empowerment is “a multi-dimensional process that helps people gain control over their own lives and increases their capacity to act on issues that they themselves define as important.” Empowerment = Enablment + Engagement Enablment : knowledge and skills for self-management but not necessarily sufficient motivationPatient empowerment 2 Engaged patients are thos who express a strong motivation but do not necessarily have sufficient ability and power for self-care. Patient empowerment is the combination of engagement and enablment Patient activation is a subset of patient empowerment referring to specific diseases. So heart failure patients, you do a study and you call it patient activation but it’s a synonym Behavioral part Participation: transfer of motivation, ability and power to patients through involvment. Involvment: Exploitation of motivation and ability and power to gain more involvment , through your involvment you establish more empowerment. This is important because when you do a study and start enrolling people, I should verify the level of enablement and engagement, because it could generate bias in the study. If the patient is not motivated, then the study will fail, therefore I need to find the barriers that block the empowerment. How to measure patient’s activation? The patient activation measure (PAM) is a commercial product that assesses an individual’s knowledge, skill, confidence for managing one’s health using a 0-100 metric. There are 4 levels with increasing activation. Each point increase in PAM score correlates to a 2% decrease in hospitalization and 2% increase in medication adherence. For example If i want to test a solution to increase adherence, I should test PAM measuring an index before and after a certain amount of time using my solution to see if there is an objective increase in the index. Challenges to patient enablement Health literacy The cognitive and social skills which determine the motivation and ability of individuals to gain access to, understand and use information in ways which Patient empowerment 3 promote and maintain good health. Without literacy you don’t have motivation, since you can’t grasp the benefits of therapy. We have different levels of literacy Functional: ability to process and apply the information to this own personal circumstances interactive: cognitive, literacy skills that enable active participation critical: critically analie and use information to participate in action that overcome structural barriers to health In Italy health literacy is really poor, there is a good correlation of health literacy being poor and more admissions, ER visits, costs Literacy also means the ability to follow complex treatments, and also leads to non- adherence etc. The majority of information is immediately forgotten if not written out. From the physician POV there are no methods followed because of no time, they could be: Teach-back: once you have given the instructions you ask the patient to repeat Facilitating: trying to use words that are less technical, it would help a lot in understanding the recommendations. Using NPL you could translate into synonyms what the doctor says, doesn’t exist this tool Assessment tools REALM The patient has to read at loud the words from the list, when you get to a word you can’t read, it means that they don’t understand its meaning. It’s a subtle thing but it generally works TOHFLA Test where you have to put words to complete the sentence and understand the context. Newest Vital Sign Bilingual screening tool that identifies patients at risk for low health literacy digital literacyPatient empowerment 4 the ability to locate, organize, understand, evaluate and analyze information using digital technology. Digital literacy is a prerequisite for effective use of e-Health. Assessing digital e-health literacy Questionnaires that try to catch some of the different literacies that can be declined (6 core skills) E-Health literacy scale (eHEALS) is a self-report tool composed by 8 qestions and is based on an individual’s perception of his skills and knowledge in the domain. In digital literacy it’s important to assess the general family literature since it’s often a shared competence to globally increase “power”. Digital tools to improve education Mono-directional: passive like youtube videos or interactive like apps, online courses, virtual reality. Again here you have the language barrier and the source, if it is reliable or not; I could provoke damage if I gave the wrong source to the people. Bi-directional: communication tools with doctor. Consider here privacy and confidentiality (social media messaging apps) Multi-directional: forums and social media. Again reliability here is the issue Apomediation The intermiediary role of the doctor is being surpassed by other intermediaries, like forums and internet, which are less reliable. The credibility of information is critical in this case, also with the case of data acquired with the smartphone. You lose the control of the context when you use digital solutions because you might lose reliability since the expert is not present when taking data. For example measuring arrythmia by yourself and then how is the doctor going to believe about the data you took by yourself? digital divide economic and social inequality with regard to access to an information or communication technology. Each region in italy has its economical development Patient empowerment 5 and territory limitations, so you could have problems especially for seniors, to have the same access to technologies. Basis of user characterization and persona creation In the context of eHealth it’s crucial to be aware of who these patients really are and what characteristic they have. If you don’t take this into account, you could design or implement a completely useless solution. Personalization and personas It is important to personalize digital solutions considering the clinical condition, context of utilization, digital health literacy etc, since it will allow better engagement. “Personas are not real people, but they represent them throughout the design process. They are hypothetical archetypes of actual users . Although they are imaginary, they are defined with significant rigor and precision. […] Personas are defined by their goals .”Patient empowerment 6 To verify if your assumptions in the end could fit the real users at the usability phase. Try to inform and facilitate the development of the solution. It is based on the information and data gathering of the potential consumers of your technology. Persona development 1. It starts with goal definition, why are you defining the persona? Then you define a survey with proper questions in order to acquire information that could be useful for you goals. Then you disseminate the surveys, it’s the best tool to reach specific patient groups, even online. If you do them online though you’re introducing a bias because the patient has access to a computer. 2. Once you have collected the raw data, you have data cleaning and managing, analize the missing data. Characterize the answers with numbers that will help you in making a score somehow for the management of the data. Clustering techniques like k-means or k-medoids create groups of data points. With some statistical analysis you can evaluate what is the parameter that is able to significantly differentiate between the clusters. 3. Finally you create personas based on clusters, that are differentiated regarding the parameter that you found the most relevant and so on. Moreover you can have focus groups, where you try to get from the history of what they tell you, particular aspects that can enrich the persona creation. The empowered patient communicates with healthcare professional effectively accesses and manages health information has health literacy makes decisions about personal health supports self-care Ideal Patient-Centered Care Approach (PCC) In the context of chronic diseases, patient empowerment is a process designed to facilitate self-directed behavioral changes.Patient empowerment 7 Self-management Self-management goes beyond simple education and literacy, its purpose is to manage life with disease, increase skills and self-confidence. Self-tracking is essential because the patient spends time with the doctor for a very short time during a year so you need a great amount of self-management to be able to follow treatments and deal with a disesase. E-patient: an empowered individual who uses online resources to obtain knowledge, to connect with other people in the same situation, and to communicate with different health care providers during their care process. Tracking and timing is important also because of prevention because usually a diagnosis is made years after the manifestation has already begun even if slightly. You need to have health literacy in order to do this otherwise how will you be able to understand if there are some symptoms. Quantified self: global movement for users and makers of self-tracking tools, to self- measure anything possible. Shared decision making Empowerment is a two-way process: the patient needs a counterpart in the “empowered health professional”. All health professionals need to develop the necessary skills and attitudes to adapt to the new patient role. CoachingPatient empowerment 8 Managing: making sure people do what they know how to do Training: teaching people to do what they don’t know how to do Mentoring: showing how Coaching: helping identify skills and capabilities and empowering people to use them as best as they canPatient empowerment 9 Patients and Individuals Class EHApp Type Lecture Materials Date Riguardare Each of us is a patient as soon as we interact with the healthcare systems in any way. We are not the same of course since each patient has all the variability of the human beings. It all comes back to the notion of personas of last lecture, since most solutions cannot fit all the patients, at least the clusters of patient is a solution. Patient’s rolePatients and Individuals 1 @20/10/2021 Patient’s role Empowered patient Behavioral medicine Why behavior is important? 1. Cognitive behavior psychology Theory of planned behavior Theory of Operant conditioning Reinforcement theory 2. Social learning theory: observing Types of observational learning effects Necessary conditions for effective modelling Self-efficacy Behavioral change techniques Advantages of technology over human persuasion Persuasive technology Persuasive System Design (PSD) model Primary task support Dialogue support Credibility support Social support Different rules depending on the context of health care where the patient is, and the individual person taking on the role. Main patient role : to act as an administrator or coordinator of one’s own care , by communicating with different health care organizations or professionals. The thing we don’t have to forget is that the patient is the only one having the full picture of what is happening in terms of symptoms, care that is planned etc. From here comes the empowerment of the patient, to be able in the best way to deal with the complexity of his/her own care. Empowerment both as a process, when the purpose of an educational intervention is to increase one's ability to think critically and act autonomously, and as outcome when an enhanced sense of self-efficacy occurs as a result of the process. Empowerment is a process that evolves in time and is a continuous variable. Empowered patient «The patient is the most underused resource in the healthcare delivery system» From educational to patient empowerment, Bandura psychologist put the bases for behavioural psychology and in particular social learning theory , which describes how people learn new ways of behaving not just doing something himself but also by looking at others that do it. Self-efficacy: the strength of an individual’s belief that he or she can successfully perform a given task or activity. How good you think you’re going to perform the task independent on the competences you actually have. Self-efficacy is a strong predictor of behaviour change. Behavioral medicine Interdisciplinary field involving behavioral, psychosocial, and biomedical science relevant to the understanding of health and illness, and their application to prevention, diagnosis, treatment and rehabilitation. Why behavior is important? Through mind and beliefs we change the behavior and have an impact on health. So it is what is in the mind that shapes the behavior. Patients and Individuals 2 Research has proven that with technology, we can help people change their thoughts improve their behaviour and gain better health and wellbeing —> behavioral change! 1. Cognitive behavior psychology It focuses on explaining, predicting and changing our behavior, using processes (beliefs and thoughts, defined as cognitions) that are going on in the mind. Theory of planned behavior It assumes that individuals make more or less conscious planned decisions. A decision is planned through three factors: attitude : state of mind regarding certain behaviors based on your cognitive information subjective norm : perceived social pressure towards that behavior, facilitating or restricting a behavior perceived behavioral control : related to the self-efficacy, estimation of how difficult it will be to perform a certain behavior or to limit it. If we want to change the behavior we act to act on these factors. Theory of Operant conditioning It is mainly what you do if you train a kid i.e. based on reward or punishment. You behave based on what you receive (you’re the subject of this approach). Reinforcement theory individual behavior is a function of its consequence (based on law of effect) 2. Social learning theory: observing You can learn a behavior by looking at others, emulating the actions. The three concepts are: People can learn through observations and this is the important of models, to have a person that is a model for that kind of behavior to induce the person to imitate.Patients and Individuals 3 Internal mental states are an essential part of the process, you cannot predict a sure result because it depends on who you are Just because something has been learnt it doesn’t mean that it will result in a change in behavior (for example not quitting smoking if you know it is not good for your health) Types of observational learning effects What will the person do and how will he behave when observing someone else. Inhibition If something happens to somebody that has a certain behavior and you don’t like what’s happening you inhibit Disinhibition Learn to exhibit a behavior that is disapproved by most people because a model does it without being punished Facilitation Be prompted to do something not ordinary because of insufficient motivation in the first place Observational learning Learning a new behavior pattern by imitating the performance of someone else Necessary conditions for effective modelling Without these conditions, we probably could not change a behavior. Attention the person should feel a connection to the model Retention ability to code information and rehearse the model’s action in an easily remembered way Reproduction reproducing the behavior observed. The observer must be physically and intellectually capable of producing the act Motivation force that drives one to actPatients and Individuals 4 Self-efficacy It is not self-esteem but comes before self-esteem because it is how competent you BELIEVE you are, not how confident you actually are. General self-efficacy scale 10-item test to obtain a self-efficacy value So now we have to move from the understanding of what behaviors are and where do they come from, to how we can actually change a behavior. Behavioral change techniques Informing health literacy, giving useful information is a way to try to change behavior; the important thing is HOW to present the information Reinforcement Discussing Social comparison reflects the observational learning, you are stimulated to compare your performance with others. Fear appeal generate fear in the target group Skills training individuals learn from practicing behavior by themselves. Practice improves their confidence. These techniques go into the direction of persuasion: you are convinced for your own good to act differently. I want the user to understand what is better for him and help him to reach that behavior. Technology is the best way to introduce a behavioral change model. Advantages of technology over human persuasion Technology does not get tired of trying to persuade someone and it goes indefinitely. It is anonymous , it can create an easy way to help sensitive subjects, because you don’t Patients and Individuals 5 have a person in conveying the technology, which could be another barrier. Manage data Technology has the ability to process huge volumes of data to give more persuasive power. Scalability You can reach out multiple users at the same time, therefore it is scalable. Ubiquity You can find technology everywhere, you can deliver the message wherever the patient is with perfect timing. Multimodality The presentation of information can be received in many different ways to attract the user and not just a text message. Persuasive technology It’s the technology designed to change our attitudes and behaviors, and it is an intersection between behavioral sciences and computer sciences. It is about pesuasion, not about deception or coerction. By using technologies for persuasion, I can be more effective, people are more adherent to eHealth interventions when those interventions use more persuasive elements. It is still difficult to assess the persuasiveness of these technologies becuase there are too many factors that make up the context and it’s subjective to how the single individual will react to a technology, it doesn’t have a fixed effect on each and every person.Patients and Individuals 6 Persuasive System Design (PSD) model It is applicable to systems in the design of a solution (embedded from the beginning) to form, alter or reinforce attitudes, behaviors or an act of compliance without using deception. Ways people can be persuaded by means of technology Remember technology is never neutral but it is always used with an intention . People like their views and behaviours to be organized and consistent, if I’m making commitment, I would like a technology that helps me do so. Even before having made the commitment, if I have a system that is making me commit that will help even more to persuade. Persuasion is often incremental , you can’t think of changing behaviour overnight. You can both use direct and indirect use . A message can be perceived also unconsciously but they can be helpful tools in reaching the goal. Essential Characteristics in a technology to persuade people Technology should be unobtrusive and fit into our daily lives. Technology should be open , the purpose should be clear from the beginning User-friendliness of the systemPatients and Individuals 7 For each support task, you can embed into your eHealth technology, one of the solutions mentioned below each task Primary task support It is the main block since it has the primary task of the intervention and persuasion. The principles of this category are: Tunneling Guide the users through a specific process, try to create an environment so that the person is persuaded to change a behaviour . For example persuade a person not to have a fear when performing a 1-hour long MRI. Self-monitoring It is at the base of behavioural change because it monitors and measures the performance in general. Rehearsal Rehearse a certain behaviour in order to enable the user to change their attitudes, for example how take a medication in the right way, how to inject a medication by yourself etc. Patients and Individuals 8 Reduction Try to divide the big task into small tasks so that it is easier to follow them Dialogue support How to interact with users, this solution of course changes with the technology but the main characteristics are: Reminders Liking Making the system attractive results in being more persuasive Social role The system adopts a social role for example a trainer or a coach Credibility support It is used to make the user realize that what you’re proposing is credible and can be trusted. People trust the technology if the chances of changing the behaviour is high. Expertise The system is viewed as having experts behind the project and therefore it can be more credible Third-party endorsements Real-world feel Highlighting the people behind the project, something more personal to explain who is behing the project Verifiability The system makes it easy to verify the credibility of the content via outside resources Social support Social comparison Forms or scoreboards to compare with somebody else, have a challenge Cooperation You can receive help from others, discussingPatients and Individuals 9 Normative influence You can receive messages that try to increase the peer pressure in making you adopt the target behaviourPatients and Individuals 10 Telehealth Class EHApp Type Lecture Materials Date RiguardareTelehealth 1 @17/11/2021 Definitions Telehealth vs biomedical informatics Why telehealth? Telehealth resources Electronic messages Telephone Remote monitoring Challenges of remote monitoring Remote monitoring scenario: CIED Self-tracking vs Remote monitoring Device classification based on technology Device classification based on acquisition Remote interpretation Teleradiology Teleopthalmology Telecardiology Video-based telehealth Synchronous video telehealth scenarios Emergency Remote intensive care Telepresence Limitations of televisits Health Pod Current trends in telehealth services Barriers for telehealth implantation Italy and telemedicine Televisit Telehealth and telemedicine approaches have the ability to bring professionals and patients close together. Historically, health care has usually involved travel. Travel has costs, both directly, in terms of gasoline or transportation tickets, and indirectly, in terms of travel time , delayed treatment, and lost productivity. Sometimes the patient is not able himself to go to the hospital so a caregiver has to lose work hours to accompany the patient. Definitions ������ Telemedicine involves the use of modern information technology, especially two-way interactive audio/video communications, computers and telemetry to deliver health services to remote patients and to facilitate information exchange between primary care physicians and specialists at some distance from each other. ������ Telehealth : newer and broader term referring to remote healthcare , including services provided using telemedicine, as well as interaction with automated systems or information resources . Telehealth is a broader concept and incorporates telemedicine. Emergency/Urgency : telehealth involved when you call the 911, they already know where you are and the possibility to exchange data from the ambulance to the hospital Relevant pathologies : specialized remote assistance for specific categories of patients Diagnosis and care : how to perform visits trying to detect the disease remotelyTelehealth 2 Risk of telemedicine marketization and abuse Suggestions to implement a good telemedicine service How to evaluate a telemedicine technology? MAST Framework of assessment Telemedicine in developing countries Education and Awareness Diagnostic support Monitoring : track some parameters useful to be followed for patients with possible risks of diseases. Telehealth vs biomedical informatics Telehealth emphasizes the distance where the care is provided, the patient is in a different location from the decision maker Biomedical informatics is more directed at the handling of the information, irrespective of the distance. Why telehealth? Because we don’t always have experts physicians in every place, it is effective and allows to save money and patients like it (more now during the pandemic); moreover it helps reducing the risks of infecting diseases and continue to help treating the patient during the pandemic. Telehealth resources We can divide the resources firstly based on the modality of provision of services Synchronous You have to be there so the physician and the patient need to be present Asynchronous store-and-forward, so all the people involved use the system with some delay but you lose on the in-time feedback of the patient Electronic messages It is the easiest, but for medical applications you are not sure if the delivery has happened and the message has been read . The security is problematic, and the messages are unstructured, so you need some effort to put the information inside. The worst part of an e-mail is that it is transient so you can’t convey the importance of a message and it gets lost inside the other messages. Solution: personal clinical electronic communications Web portals in which you sign-in as physician and patients, when the patient writes the message, he is sure that privacy and confidentiality are assured , it can be seen when Telehealth 3 the physician has read the communication and the physician can reply to the service, and what he wrote has legal validity and is recorded. One of the aspects of the medical act is the need to have a legal validity in order to preserve the patient from potential errors by the physician but also to preserve the physician by what the patient could do wrong when acting by himself and not following the instructions. Telephone One of the main resources when talking about telehealth, we usually focus on video because the phone call is not recorded and there is no trace of a medical act and it cannot be reimbursed . Phone consultations are not delivered as they could be. In the last years, the NHS has renewed interest in audio-only communication, in follow-up patients in chronic conditions. They found out that it was very effective in managing patients . Of course you need an investment to have somebody calling the patients, it can’t be the doctor. Remote monitoring Monitor the patient outside the hospital, transmitting data to a central location for review. Usually it involves the fact that multiple patients send their data to a center where trained personnel look at the data and activate actions and feedbacks. The parameters for remote monitoring are limitless and depend on the sensors that could be used by the patients to check some parameters. Some examples are bl