利用区块链来加强患者报告的观察指标发展.pdf

The Use of a Blockchain to Foster the Development of Patient-Reported Outcome Measures JASON C.GOLDWATER,MA,MPA August 8,2016 2 Introduction to Patient Related Outcome Measures Patient-related outcome measures(PROMs),which focus on outcomes that are directly related to the patient,have taken on added importance and significance over the past several years.1 This is due,in part,to the increased attention focused on the patient experience of care and to provide a patient-focused assessment on the burden and impact of disease.PROMs can include symptoms and other aspects of health related quality of life indicators such as physical or social function,treatment adherence,and satisfaction with treatment.They can also facilitate more accurate patient-physician communication in terms of the burden of treatment-related morbidities by providing a more detailed and complete evaluation of treatments for specific conditions,such as cancer or multiple sclerosis.2 PROMs are distinct from traditional clinical efficacy measures(e.g.,survival in cancer,smoking cessation)because they directly reflect the impact of disease and its treatment from the patients perspective.1 It can examine the balance between the efficiency of the treatment and its burden on the patient.It is also effective in looking at areas such as physical functioning and overall well-being,and highlighting the efficacy and safety of treatments in relation to its overall clinical benefit.Because the measures themselves are developed from the patients perspective,it can also facilitate greater patient involvement in treatment decision-making as well as providing guidance for health care decisions.1 However,there is difficulty in the development,implementation and sustainability of PROMs over the long-term.3 The success of a well-developed PROM that adequately and accurately reflect that patients perspective relies on the use of an instrument that is psychometrically tested and validated and can capture the burden of disease or treatment.This means the instrument must be reliable,in that the PROM yields the same metric for evaluation each time it is administered,providing that the construct being measured has not changed.Additionally,the PROM must always measure its intended objective,and not demonstrate significant variation which would obscure the results and provide little value to either the patient or provider.One of the most effective ways of demonstrating reliability and validity of the psychometric properties of an instrument designed to collected patient-reported data is to provide enough evidence to demonstrate it can reliably measure its intended target.3 The development and testing of the instrument,in addition to the development and implementation of the PROM,is both time-consuming and expensive.4 In order to appropriately identify patient experiences,trends and issues that have the most significance,an instrument must be tested through a variety of methods.These may include cognitive interviewing among patient groups,focus groups and behavior coding.To ensure that the results are statistically valid,the appropriate patient population must be included,with the right combination of sociodemographic characteristics,such as race,age,gender,marital status and others.While numerous instruments have been developed successfully to accurately assess patient-based outcomes in areas such as neurology and depression,the process itself can be burdensome,causing the actual development and implementation of a PROM to take two to three years to complete.5 3 The Internet of Things for PROM Development A potential alternative to the use of standardized instruments to collect data and provide a foundation for the development of a PROM is the use of technologies associated with the Internet of Things(IoT).The US National Intelligence Council defines this term as,“the general idea of things,especially everyday objects,that are readable,recognizable,locatable,addressable and controllable via the Internet whether through Remote Frequency Identification(RFID),wireless local-area networks(LANs),wide-area networks(WANs)or other means.6 In a larger context,while the most familiar Internet-connected devices are computer such as laptops,servers,smartphones,and tables,the IoT concept refers to everyday objects that are starting to come online with the use of embedded sensors and microprocessors,communicating with each other and the Internet.In the area of health,this refers to the use of self-tracking devices and personal environment monitoring applications,such as those that provide automate digital health monitoring,ambient behavior management suggestions and other real-time personalized recommendations.7 This use of these technologies has rapidly increased as over 80%of all adults use the Internet to search for health information,connect with patients similar to themselves,and share information regarding their condition or the facilities in which they seek treatment.7 There are currently over 220 million smartphone users in the United States that have downloaded over seven billion applications,of which approximately 22,000 are related to health.8 In the area of wearable devices and sensors,over 15%of individuals from the ages of 18-55 have devices such as Fitbit and Jawbone UP and 10%of the population from the ages of 65 and over have either wearable sensor technology or sensors integrated into their living environment to manage conditions such as falls,chronic disease and medication management.9 The utilization of these technologies within the IoT has led to the development of personalized data streams that can identify and manage physical and mental health outcomes.These outcomes are greatly expanded from the traditional and nearly exclusive focus on cure to branch upwards to a wider set of endpoints,such as establishing baseline and variability levels of phenotypic measure in individuals.Additionally,a greater focus can be placed on improving,normalizing and preventing conditions,while proactively targeting longevity,enhancement and health.This shifting in health outcomes is already starting to be seen in the area of physical health with a new focus on prevention and wellness maintenance.Furthermore,the therapeutic delivery platform of the mobile phone with applications such as Code Blue and Breathe2Relax,demonstrate that optimization of the technology can affect mental health and mental performance among populations of all ages.The significant amount of data being generated through these devices,such as electronic medical records(EMRs),quantified self-tracking devices,smartphone applications and personal health records(PHRs)provide an opportunity to gather insight into a patients health status that was previously only available through the administration of a psychometrically validated instrument.This expanding ecosystem provides a more proactive approach to health as the data streams from these devices can be intermingled with social networks,crowdsourced studies and the Quantified Self community,which collects and shares biophysical assessments.With the increasing interest and need in PROMs,the integrated data 4 streams coming from these tightly coupled software solutions that encompass a large array of patient-based data provide an ability to develop these measures independent of a validated tool.For example,the emerging quantified data streams could be helpful in elucidating the mental health of both individuals and populations.The US National Institutes of Health estimates that 26.2%of Americans ages 18 and over suffer from a diagnosable mental disorder in any given year.10 Using PROMs to evaluate a patients mental status and functioning moves towards understanding their specific condition and helping move towards a more positive position of mental performance optimization,rather than seeking a cure or identifying a pharmacologic solution.Data streams from smartphone applications and patient-based web portals can provide quantified assessments of qualities such as empathy,loneliness,happiness and fulfillment.A measure of these types of qualities that come directly from a IoT device is possibly an early harbinger of what may become in both the development and utilization of a PROM that provides access to a patients condition.This insight can assist providers and health care administrators determine the most effective treatment protocols to ensure quality care is delivered to a patient that meets those specific parameters.This would also increase the probability that the measure matters directly to the patient as it is taking data from applications and devices that are continuously providing that data.The Drawbacks of Cloud Storage Solutions Yet,one of the most significant issues is using data from the IoT to crowdsource PROMs is the security of this individual health data being collected and stored from these devices.The most common mechanism to collect these data streams is through secure cloud storage systems,which give different levels of permissoning access to these data streams.The physical data storage spans multiple servers and is available and accessible through a web-based application programming interface(API),a cloud storage gateway or Web-content management systems.This method of storing and accessing data has become increasingly popular over the last several years,as organizations such as Amazon,Apple and Microsoft rely on this solution to store substantial amounts of consumer-based data.11 In 2013,the Office of Civil Rights(OCR)redefined a“Business Associate”(BA)and increased protections for the privacy and security of personal health information(PHI)under its“Omnibus Rules.”One of the new provisions stated that subcontractors,or organizations/individuals that act on behalf of a BA,become a BA themselves.11 This extended to cloud storage providers as covered entities and BAs use them to store health information,often citing that it is more cost effective and there are lower IT management costs.However,as consumers rely on crowd providers to store personal data,they relinquish direct control over that data and,as a result are unaware of who has access and where the data is geographically located.Even if an explicit business associate agreement is developed between the BA and the cloud storage provider,it would only provide the terms of who takes responsibility of the privacy and security of the data in the event a breach occurs.The consumer would potentially have control over access to these data streams,but would rely on the cloud storage provider to enforce those privileges.11 5 Furthermore,although the use of cloud storage is popular,there are still a number of risks that a consumer undertakes when using this mechanism for their personal data.In cloud-based architecture,data is replicated and moved frequently so the risks of unauthorized data use increases.Additionally,multiple individuals with access to the data,such as administrators,network engineers and technical experts that cover a wide area of servers in which the information is stored.11 This also increases the risk of unauthorized access and use.However,even if the data is secure through strict access controls and is encrypted at its point of origin and while in transit,it still poses a problem for the development of PROMs.The concept of a PROM is to develop a patient-focused measure that relates to an area or focus that is of concern to the patient,and one in which their engagement and feedback is essential for its successful implementation.Accessing large data streams from a variety of devices that are part of the IoT can provide a foundation on which to base a PROM,but it is difficult to ascertain whether that data will produce a measure that will have the intended meaning and relevancy for a patient.For example,there have been PROMs developed to evaluate common mental health disorders,such as depression,within patients.Using a structured and validated protocol such as the Personal Health Questionnaire(PHQ)can provide input into a measure that may tell a provider about the patients current status,functioning and quality of life.That knowledge may serve as a predicate to the type of therapy and potential pharmacologic treatment that is needed.An increase in overall quality of life and appropriate management of the underlying mental health condition would benefit the patient and would rely on their direct input to effectively produce a metric from the PROM that would evaluate whether the appropriate care was being delivered.Yet,if there was a reliance on devices through the IoT to produce data for these types of PROMs,it would not be as specific as the input taken from a questionnaire.The data streams within the cloud would contain numerous elements that would have to be mined and stratified to determine their relevancy to the measure.The consumer would have control over access to the data stream,but not to the elements within it.It would be up to the provider to sort through the data itself to populate the measure effectively.This would pose a significant and potentially insurmountable burden to a provider.Compounding this difficulty is the lack of an interoperable health network system in which disparate data streams from various systems and devices can be integrated together for a singular patient.In evaluating a patients quality of life,functional status and management of a mental health condition,a number of factors must be considered.The provider should know about potential co-morbid conditions that worsen the mental health status of a patients;the types of medications that the patient is currently on as well as their past medication history;the types of treatments the patient has undertaken to manage their illness;and their family history of mental illness,among other items.This information could be found across smartphone applications,patient web portals and other sensor-based technology,but the data cannot be integrated together.Furthermore,if the data from these personalized technologies needed to be integrated with data from an EMR to provide a more comprehensive health history for a patient,which may be significant in the development of a PROM,that combination of data would be challenging.The way the data in each of these systems is represented is neither similar nor consistent,as tracking data is often represented through web-based 6 languages,such as the Extensible Markup Language(XML),while data from an EMR uses robust and comprehensive clinical vocabularies.Additionally,the lack of a universal patient identifier would also pose burden in aligning the information directly with a patient to ensure their information is correctly represented in a quality measure.The Use of Blockchain Technology for PROMs The use of blockchain technology provides an ability to securely protect patient data from all devices,both in transit and at rest;securely provide access privileges to patients on