Beyond the hype of big data and artificial intelligence: building foundations for knowledge and wisdom.

Big data, coupled with the use of advanced analytical approaches, such as artificial intelligence (AI), have the potential to improve medical outcomes and population health. Data that are routinely generated from, for example, electronic medical records and smart devices have become progressively easier and cheaper to collect, process, and analyze. In recent decades, this has prompted a substantial increase in biomedical research efforts outside traditional clinical trial settings. Despite the apparent enthusiasm of researchers, funders, and the media, evidence is scarce for successful implementation of products, algorithms, and services arising that make a real difference to clinical care. This article collection provides concrete examples of how "big data" can be used to advance healthcare and discusses some of the limitations and challenges encountered with this type of research. It primarily focuses on real-world data, such as electronic medical records and genomic medicine, considers new developments in AI and digital health, and discusses ethical considerations and issues related to data sharing. Overall, we remain positive that big data studies and associated new technologies will continue to guide novel, exciting research that will ultimately improve healthcare and medicine-but we are also realistic that concerns remain about privacy, equity, security, and benefit to all.

Digital Trespass: Ethical and Terms-of-Use Violations by Researchers Accessing Data From an Online Patient Community.

With the expansion and popularity of research on websites such as Facebook and Twitter, there has been increasing concern about investigator conduct and social media ethics. The availability of large data sets has attracted researchers who are not traditionally associated with health data and its associated ethical considerations, such as computer and data scientists. Reliance on oversight by ethics review boards is inadequate and, due to the public availability of social media data, there is often confusion between public and private spaces. In addition, social media participants and researchers may pay little attention to traditional terms of use. In this paper, we review four cases involving ethical and terms-of-use violations by researchers seeking to conduct social media studies in an online patient research network. These violations involved unauthorized scraping of social media data, entry of false information, misrepresentation of researcher identities of participants on forums, lack of ethical approval and informed consent, use of member quotations, and presentation of findings at conferences and in journals without verifying accurate potential biases and limitations of the data. The correction of these ethical lapses often involves much effort in detecting and responding to violators, addressing these lapses with members of an online community, and correcting inaccuracies in the literature (including retraction of publications and conference presentations). Despite these corrective actions, we do not regard these episodes solely as violations. Instead, they represent broader ethical issues that may arise from potential sources of confusion, misinformation, inadequacies in applying traditional informed consent procedures to social media research, and differences in ethics training and scientific methodology across research disciplines. Social media research stakeholders need to assure participants that their studies will not compromise anonymity or lead to harmful outcomes while preserving the societal value of their health-related studies. Based on our experience and published recommendations by social media researchers, we offer potential directions for future prevention-oriented measures that can be applied by data producers, computer/data scientists, institutional review boards, research ethics committees, and publishers.

A Modular Health-Related Quality of Life Instrument for Electronic Assessment and Treatment Monitoring: Web-Based Development and Psychometric Validation of Core Thrive Items.

BACKGROUND: Patient-reported outcome (PRO) measures describe natural history, manage disease, and measure the effects of interventions in trials. Patients themselves increasingly use Web-based PRO tools to track their progress, share their data, and even self-experiment. However, existing PROs have limitations such as being: designed for paper (not screens), long and burdensome, negatively framed, under onerous licensing restrictions, either too generic or too specific. OBJECTIVE: This study aimed to develop and validate the core items of a modular, patient-centric, PRO system (Thrive) that could measure health status across a range of chronic conditions with minimal burden. METHODS: Thrive was developed in 4 phases, largely consistent with Food and Drug Administration guidance regarding PRO development. First, preliminary core items (common across multiple conditions: core Thrive items) were developed through literature review, analysis of approximately 20 existing PROs on PatientsLikeMe, and feedback from psychometric and content experts. Second, 2 rounds of cognitive interviews were iteratively conducted with patients (N=14) to obtain feedback on the preliminary items. Third, core Thrive items were administered electronically along with comparator measures, including 20-item Short-Form General Health Survey (SF)-20 and Patient Health Questionnaire (PHQ)-9, to a large sample (N=2002) of adults with chronic diseases through the PatientsLikeMe platform. On the basis of theoretical and empirical rationale, items were revised or removed. Fourth, the revised core Thrive items were administered to another sample of patients (N=704) with generic and condition-specific comparator measures. A psychometric evaluation, which included both modern and classical test theory approaches, was conducted on these items, and several more items were removed. RESULTS: Cognitive interviews helped to remove confusing or redundant items. Empirical testing of subscales revealed good internal consistency (Cronbach alpha=.712-.879), test-retest reliability (absolute intraclass correlations=.749-.912), and convergent validity with legacy PRO scales (eg, Pearson r=.5-.75 between Thrive subscales and PHQ-9 total). The finalized instrument consists of a 19-item core including 5 multi-item subscales: Core symptoms, Abilities, Mobility, Sleep, and Thriving. Results provide evidence of construct (content, convergent) validity, high levels of test-retest and internal consistency reliability, and the ability to detect change over time. The items did not exhibit bias based on gender or age, and the items generally functioned similarly across conditions. These results support the use of Thrive Core items across diverse chronic patient populations. CONCLUSIONS: Thrive appears to be a useful approach for capturing important domains for patients with chronic conditions. This core set serves as a foundation to begin developing modular condition-specific versions in the near future. Cross-walking against traditional PROs from the PatientsLikeMe platform is underway, in addition to clinical validation and comparison with biomarkers. Thrive is licensed under Creative Commons Attribution ShareAlike 4.0.

Patient, study thyself.

The past 15 years have seen the emergence of a new paradigm in medical research, namely of people living with medical conditions (whether patients, parents, or caregivers) using digital tools to conduct N-of-1 trials and scientifically grounded research on themselves, whilst using the Internet to form communities of like-minded individuals willing to self-experiment. Prominent examples can be found in amyotrophic lateral sclerosis/motor neurone disease (the 'lithium study' on PatientsLikeMe), Parkinson's disease ('digital patient' Sara Riggare), and diabetes (the 'open artificial pancreas' of the #WeAreNotWaiting movement). Through transparency, data sharing, open source code, and publication in the peer-reviewed scientific literature, such activities conform to expected scientific conventions. However, other conventions, such as ethical oversight, regulation, professionalization, and the ability to translate this new form of relatively biased data into generalizable decisions, remain challenged. While critics worry such participant-led research merely muddies the waters of high-quality medical research and exposes patients to new harms, the potential is there to enroll millions of active minds in unravelling the wicked problems of complex medical disorders that degrade the human health span.

DigitalMe: a journey towards personalized health and thriving.

The use of information and communication technologies for health (eHealth) delivered via mobile-based or digitally enhanced solutions (mHealth) have rapidly evolved. When used together across various mobile applications and devices eHealth and mHealth technologies have the ability to passively monitor behavior as an indicator of socialization and mood; accumulate a range of biomedical data such as weight and heart rate; and track metrics associated with activities including steps taken and hours slept. Yet, these technologies are insufficient for measuring the full array of data about an individual and the impact of that data on a person's current and future health. Digital health converges eHealth and mHealth with patient data about their health, healthcare, living, and environment with genomics. An innovative opportunity to unravel the complexities of disease and aging is increasingly possible with an integrative multi-omics approach informed by multidisciplinary sciences including medicine, design, biomedical informatics and engineering. The digitization of individual level data from all available sources makes possible the development of DigitalMe™, a personalized virtual avatar of a real person. The combination of longitudinally collected person generated data and molecular data derived from biospecimens offers researchers unique opportunities to better understand the mechanisms of disease while advancing person-centric hypotheses generation related to treatments, diagnostics, and prognostics.

Scaling PatientsLikeMe via a "Generalized Platform" for Members with Chronic Illness: Web-Based Survey Study of Benefits Arising.

BACKGROUND: Launched in 2006 for patients with amyotrophic lateral sclerosis, PatientsLikeMe is an online community offering patient-reported outcomes, symptom tracking, and social features. Every member of the site can see all the data reported by every other member, view aggregated reports, identify "patients like them," and learn about treatment options in order to live better with their condition. In previous studies, members reported benefits such as improved condition knowledge, increased medication adherence, and better management of side effects. However, the site evolved in 2011 from condition-specific "vertical" communities consisting only of people with the same disease to a "generalized platform," in which every patient could connect with every other patient regardless of condition and with generic, rather than condition-specific, data tools. Some, but not all, communities received further custom tracking tools. OBJECTIVE: We aimed to understand (1) whether members of PatientsLikeMe using the generalized platform still reported similar benefits and (2) assess factors associated with benefits, such as community customization, site use, and patient activation. METHODS: A cross-sectional retrospective custom survey was fielded to 377,625 members between 2016 and 2017 including the Patient Activation Measure (PAM). A benefit index was developed for comparability across conditions. RESULTS: The invitation was viewed by 26,048 members of whom 11,915 did not respond, 5091 opted out, 1591 provided partial data, and 17 were screened out. Complete responses were received from 7434 participants. Users perceived greatest benefit in understanding how their condition may affect them (4530/6770, 66.91% participants, excluding "does not apply" answers), understanding what might help them live better with their condition (4247/6750, 62.92%), which treatments were available (4143/6898, 60.06%), understanding treatment side effects (4182/6902, 60.59%), and important factors in making treatment decisions (3919/6813, 57.52%). The benefit index was 29% higher for the "most activated" patients (PAM level 4 vs PAM level 1; relative risk [RR]=1.29, P<.001), 21% higher for conditions with some community customization versus none (RR=1.21, P<.001), and 11% higher in those using the site most often versus least (RR=1.11, P<.001). CONCLUSIONS: Members of the generalized platform reported a range of benefits related to improved knowledge and understanding of their condition and treatment management. Condition-specific customization may improve their experience still further. Future studies will explore longitudinal changes to patient activation.

Clinical trials from the patient perspective: survey in an online patient community.

BACKGROUND: Developing new medicines relies on the successful conduct of clinical trials. As trial protocols become more arduous, it becomes harder to recruit and retain patient volunteers, although recent efforts such as OMERACT and I-SPY2 show that partnering with patients can be beneficial. We sought to describe drivers and barriers to trial participation, as well as condition-specific trial preferences. METHODS: An online survey was fielded via the patient-powered research network PatientsLikeMe to 1,621 members living with nine selected chronic health conditions. Questions included demographics, trial experience, reasons for non-participation, questions relating to aspects of trial design, and an adaptation of the Net Promoter Score (NPS) for trial satisfaction. RESULTS: Mean age of respondents was 55 years; most patients were white (93%), female (67%), and living in the United States (72%). Primary conditions were MS (21%), Parkinson's (20%), fibromyalgia (15%), ALS (10%), type 2 diabetes (10%), rheumatoid arthritis (RA, 8%), epilepsy (8%), major depressive disorder (MDD, 5%) and systemic lupus erythematosus (SLE, 3%). Most patients had not discussed a trial with their physician and only 21% had ever enrolled, with rates highest in ALS (36%), Parkinson's disease (36%) and MS (20%) and lowest among SLE (9%), MDD (11%) and Fibromyalgia (11%). Common reasons for non-participation were eligibility criteria, inconvenience of travel and concerns about side effects. NPS suggested that many patients were unsatisfied; patients with lupus, epilepsy, RA, and fibromyalgia reported negative scores, i.e. they would dissuade other patients like them from taking part in trials. The most important considerations in trial participation were the opportunity to improve one's own health and that of others, the reputation of the institution, and having medical bills covered in case of injury. Least important were remuneration and possibility of receiving a placebo. ALS patients were more willing to tolerate undesirable aspects of trials. CONCLUSIONS: Most patients are willing to enroll yet very few are invited. When they do, trial participation is often burdensome, but patients are willing to help improve their design. Researchers should let patients help design better trials to overcome recruitment and retention issues and hasten the development of new medicines.

Medicine and the future of health: reflecting on the past to forge ahead.

The development of new therapies has a rich history, evolves quickly with societal trends, and will have an exciting future. The last century has seen an exponential increase in complex interactions between medical practitioners, pharmaceutical companies, governments and patients. We believe technology and societal expectations will open up the opportunity for more individuals to participate as information becomes more freely available and inequality less acceptable. Corporations must recognize that usual market forces do not function ideally in a setting where health is regarded as a human right, and as modern consumers, patients will increasingly take control of their own data, wellbeing, and even the means of production for developing their own treatments. Ethics and legislation will increasingly impact the processes that facilitate drug development, distribution and administration. This article collection is a cross-journal collaboration, between the Journal of Pharmaceutical Policy and Practice (JoPPP) and BMC Medicine that seeks to cover recent advances in drug development, medicines use, policy and access with high clinical and public health relevance in the future.The Medicine and the Future of Health article collection is a joint collection between BMC Medicine and Journal of Pharmaceutical Policy and Practice. Therefore, this Editorial by the guest editors has been published in both journals.

It's a long shot, but it just might work! Perspectives on the future of medicine.

What does the future of medicine hold? We asked six researchers to share their most ambitious and optimistic views of the future, grounded in the present but looking out a decade or more from now to consider what's possible. They paint a picture of a connected and data-driven world in which patient value, patient feedback, and patient empowerment shape a continually learning system that ensures each patient's experience contributes to the improved outcome of every patient like them, whether it be through clinical trials, data from consumer devices, hacking their medical devices, or defining value in thoughtful new ways.

Concept Elicitation Within Patient-Powered Research Networks: A Feasibility Study in Chronic Lymphocytic Leukemia.

OBJECTIVES: To explore the feasibility of using social media-based patient networks to gather qualitative data on patient-reported outcome (PRO) concepts relevant to chronic lymphocytic leukemia (CLL). METHODS: Between August and November 2013, US-residing members of the PatientsLikeMe online CLL patient community completed open-ended web-based surveys designed to elicit descriptions of CLL symptoms, impacts, and treatment-related perceptions. Qualitative telephone follow-up interviews were conducted with a subsample of respondents. Survey responses and interview transcripts were coded for qualitative analysis using Atlas.ti. RESULTS: Fifty survey responses were included in the analyses. Participants were age 60.5 ± 6.9 years, 54% female, and 96% white. When surveyed, 20% were receiving current treatment, 16% were in remission, and 64% were treatment-naïve. Among respondents, 369 descriptions of CLL symptoms were coded. Fatigue-related symptoms were expressed most frequently, with 54% reporting "fatigue," "tiredness," or both in their responses. These concepts were followed by night sweats (38%), swollen lymph nodes (32%), and frequent infections (28%). Among impacts of CLL, worry and fear (66% of respondents), depressed feelings (52%), and work limitations (50%) were noted most frequently. CONCLUSIONS: Survey results identified constitutional symptoms of CLL included in existing PRO instruments and the literature. Although the findings suggest that qualitative data obtained through social media applications can be potentially useful in supporting concept identification for newly developed PRO instruments, they also indicate that online approaches alone may not be sufficient to achieve efficient and exhaustive concept elicitation. Further research is needed to identify whether the results can support content validity in the same way as established qualitative research methods.

Increasing Patient Involvement in Drug Development.

BACKGROUND: To ensure the creation of treatments that maximize value at the lowest cost, all aspects of the health care system need to align with patient needs and preferences. Despite growing efforts to engage patients in research and regulatory activities, the pharmaceutical industry has yet to maximize patient involvement in the drug development process. OBJECTIVE: To gain a better understanding of the present state of patient involvement in drug development. METHODS: Through a semistructured interview methodology, we sought to identify opportunities, barriers, and examples of patient involvement in the drug development process. Telephone interviews were conducted with six senior leaders of evidence generation within the pharmaceutical industry and four patients. These interviews were supplemented with interviews with a research funder, a regulator, a patient advocacy group, and a caregiver. RESULTS: Although our interviewees spoke of the potential benefits of aligning research around the needs of patients, there were few examples they could share to suggest this was occurring at scale. A number of barriers were identified including the added burden associated with adverse event reporting, concerns about patient representativeness or their ability to participate in drug development conversations, and the costs in time and resources involved relative to returns on investment. CONCLUSIONS: As health care systems continue to evolve and establish patients as the primary stakeholder in their health care decision making, the pharmaceutical industry will need to be innovative to demonstrate the value of their products relative to the outcomes experienced by patients. Pharmaceutical companies should recognize the value of involving patients across the entire product life cycle and work to transform present perceptions and practices throughout their organizations.

PatientsLikeMe® Online Epilepsy Community: Patient characteristics and predictors of poor health-related quality of life.

OBJECTIVE: The online PatientsLikeMe® Epilepsy Community allows patients with epilepsy to record, monitor, and share their demographic, disease, and treatment characteristics, providing valuable insights into patient perceptions and understanding of epilepsy. The objective of this retrospective analysis was to characterize the profile of users and their disease and identify factors predictive of poor health-related quality of life (HRQoL), while assessing the platform's potential in providing patient-reported data for research purposes. METHODS: Data recorded (January 2010-November 2011) by Epilepsy Community members, with an epilepsy diagnosis and who reported >1 seizure, included the following: sociodemographic and disease characteristics, treatments, symptoms, side effects perceived as medication-related, seizure occurrence, and standardized questionnaires (Quality of Life in Epilepsy Inventory [QOLIE-31/P], EuroQoL 5-Dimensions Scale, 3 Levels [EQ-5D-3L], and Hospital Anxiety and Depression Scale [HADS]). Univariate and multivariate logistic regressions were conducted to identify predictors of poor HRQoL. RESULTS: During the study period, the Epilepsy Community comprised 3073 patients, of whom 71.5% were female, had a mean age of 37.8years, and had a mean epilepsy duration of 17.7years. The most frequently reported moderate/severe symptoms (n=2135) included memory problems (60.2%), problems concentrating (53.8%), and fatigue (50.0%). Medication-related side effects (n=639) included somnolence (23.2%), fatigue (17.2%), and memory impairment (13.8%). The QOLIE-31/P scores (n=1121) were significantly worse in patients who experienced a recent seizure. For QOLIE-31/P, highly predictive factors for poor HRQoL included the following: mild/moderate problems concentrating, depression, memory problems, treatment side effects, occurrence of tonic-clonic seizures, and epilepsy duration ≤1year. For EQ-5D-3L, highly predictive factors for poor HRQoL included the following: pain, depression, and comorbidities. Patients on newer AEDs were less likely to report poor HRQoL (QOLIE-31/P). SIGNIFICANCE: These findings move further towards supporting the feasibility and usefulness of collecting real-world, anonymized data recorded by patients online. The data provide insights into factors impacting HRQoL, suggesting that a holistic treatment approach beyond seizure control should be considered in epilepsy.

Research led by participants: a new social contract for a new kind of research.

In recent years, there have been prominent calls for a new social contract that accords a more central role to citizens in health research. Typically, this has been understood as citizens and patients having a greater voice and role within the standard research enterprise. Beyond this, however, it is important that the renegotiated contract specifically addresses the oversight of a new, path-breaking approach to health research: participant-led research. In light of the momentum behind participant-led research and its potential to advance health knowledge by challenging and complementing traditional research, it is vital for all stakeholders to work together in securing the conditions that will enable it to flourish.

Exploring Concordance of Patient-Reported Information on PatientsLikeMe and Medical Claims Data at the Patient Level.

BACKGROUND: With the emergence of data generated by patient-powered research networks, it is informative to characterize their correspondence with health care system-generated data. OBJECTIVES: This study explored the linking of 2 disparate sources of real-world data: patient-reported data from a patient-powered research network (PatientsLikeMe) and insurance claims. METHODS: Active patients within the PatientsLikeMe community, residing in the United States, aged 18 years or older, with a self-reported diagnosis of multiple sclerosis or Parkinson's disease (PD) were invited to participate during a 2-week period in December 2014. Patient-reported data were anonymously matched and compared to IMS Health medical and pharmacy claims data with dates of service between December 2009 and December 2014. Patient-level match (identity), diagnosis, and usage of disease-modifying therapies (DMTs) were compared between data sources. RESULTS: Among 603 consenting patients, 94% had at least 1 record in the IMS Health dataset; of these, there was 93% agreement rate for multiple sclerosis diagnosis. Concordance on the use of any treatment was 59%, and agreement on reports of specific treatment usage (within an imputed 5-year period) ranged from 73.5% to 100%. CONCLUSIONS: It is possible to match patient identities between the 2 data sources, and the high concordance at multiple levels suggests that the matching process was accurate. Likewise, the high degree of concordance suggests that these patients were able to accurately self-report their diagnosis and, to a lesser degree, their treatment usage. Further studies of linked data types are warranted to evaluate the use of enriched datasets to generate novel insights.

Correction of: Exploring Concordance of Patient-Reported Information on PatientsLikeMe and Medical Claims Data at the Patient Level.

[This corrects the article DOI: 10.2196/jmir.5130.].

National Randomized Controlled Trial of Virtual House Calls for People with Parkinson's Disease: Interest and Barriers.

BACKGROUND: Delivering specialty care remotely directly into people's homes can enhance access for and improve the healthcare of individuals with chronic conditions. However, evidence supporting this approach is limited. MATERIALS AND METHODS: Connect.Parkinson is a randomized comparative effectiveness study that compares usual care of individuals with Parkinson's disease in the community with usual care augmented by virtual house calls with a Parkinson's disease specialist from 1 of 18 centers nationally. Individuals in the intervention arm receive four virtual visits from a Parkinson's disease specialist over 1 year via secure, Web-based videoconferencing directly into their homes. All study activities, including recruitment, enrollment, and assessments, are conducted remotely. Here we report on interest, feasibility, and barriers to enrollment in this ongoing study. RESULTS: During recruitment, 11,734 individuals visited the study's Web site, and 927 unique individuals submitted electronic interest forms. Two hundred ten individuals from 18 states enrolled in the study from March 2014 to June 2015, and 195 were randomized. Most participants were white (96%) and college educated (73%). Of the randomized participants, 73% had seen a Parkinson's disease specialist within the previous year. CONCLUSIONS: Among individuals with Parkinson's disease, national interest in receiving remote specialty care directly into the home is high. Remote enrollment in this care model is feasible but is likely affected by differential access to the Internet.

'Trust but verify'--five approaches to ensure safe medical apps.

Mobile health apps are health and wellness programs available on mobile devices such as smartphones or tablets. In three systematic assessments published in BMC Medicine, Huckvale and colleagues demonstrate that widely available health apps meant to help patients calculate their appropriate insulin dosage, educate themselves about asthma, or perform other important functions are methodologically weak. Insulin dose calculators lacked user input validation and made inappropriate dose recommendations, with a lack of documentation throughout. Since 2011, asthma apps have become more interactive, but have not improved in quality; peak flow calculators have the same issues as the insulin calculators. A review of the accredited National Health Service Health Apps Library found poor and inconsistent implementation of privacy and security, with 28% of apps lacking a privacy policy and one even transmitting personally identifying data the policy claimed would be anonymous. Ensuring patient safety might require a new approach, whether that be a consumer education program at one extreme or government regulation at the other. App store owners could ensure transparency of algorithms (whiteboxing), data sharing, and data quality. While a proper balance must be struck between innovation and caution, patient safety must be paramount.Please see related articles: http://dx.doi.org/10.1186/s12916-015-0444-y , http://www.biomedcentral.com/1741-7015/13/106 and http://www.biomedcentral.com/1741-7015/13/58.