Introduction

Oncologists and their patients are facing disruptive changes in health care practice, research, governmental oversight, business practices, communication, and reimbursement—changes brought on by the individual growth and merging of the fields of information technology with medical technology, medical practice, biology, and physics. We now practice in large hospital systems or accountable care organizations rather than in small group practice silos; deal with changing reimbursement methods as we transition from fee-for-service to bundling payments and defined encounters of care; document our patient care using electronic media rather than paper; and see the doctor–patient relationship become more patient-centric. We are increasingly communicating electronically with all health care stakeholders, including our patients.

This dramatic increase in the quantity, quality, and ease of finding information—brought about by the use of computers and the effortlessness of connecting everyone and everything—all of which has been brought about by the Internet, has changed our lives forever. However, we remain frustrated with our electronic devices needed to communicate with our health care system because they still require typing or dictation for data entry, and the data entered are not easily understood by the computer or able to be shared, reused, or reported easily because of the absence of seamless data interoperability where data entered into one device could be understood, used, and reused effectively in another device.

Clayton Christensen presciently warned us about the disruptive effects of this transformation in 1997,¹ and he warned us specifically about its effects on health care in 2008.² Indeed, for the first time, this new, disruptive digital world of medicine is increasingly defined by information becoming electronically mobile, cheap, available to all, and consumer-oriented to such an extent that almost all recent information technology advances in hardware and software begin with the consumer rather than business. Oncologists and their patients must begin to understand this transformative event that will directly address our current frustrations.

An oncology digital health information technology system

Common data elements and value sets

Common data elements define the data to be collected in the medical record by specifically identifying the label of each data field and the appropriate value set to choose for data entry. When possible, the value set choice should reference a standard code system such as a clinical vocabulary used to describe the clinical encounter (e.g., SNOMED-CT), the laboratory data (e.g., LOINC), the terms describing basic and clinical research activities [e.g., cancer Biomedical Informatics Grid (caBIG)], and the elements and codes that define specific disease classifications used for billing and registry reporting purposes (e.g., ICD-9 and ICD-10 codes).

Data types

Computers capture electronic structured and unstructured data. Data can be structured using data elements and value sets that are machine readable, reusable, shareable, actionable, multiuser-able, and multipurposeful. Unstructured or narrative data are entered by typing or dictation and adds context making it is understood by humans but not by computers unless sophisticated natural language processing tools are used or the data can be tagged by metadata that renders them searchable. Searchable or machine-readable data would permit the EHR (electronic health record) to provide clinical decision support (CDS), user education, and secondary reporting purposes. In reality, both types of data are important in capturing the complete patient’s health record.

Software functionalities and documentation

Software functionalities refer to the capabilities that a software program should provide. For all EHRs, ePrescription software is necessary, whereas for medical oncology, EHRs chemotherapy administration is required. Table 1 lists EHR functional elements defined by ASCO’s EHR workgroup and Table 2 lists oncology-specific documentation required in an oncology EHR.

Internet

The Internet serves as the communication and messaging component of the digital health care system. The development of the Internet can be separated into four stages. At its inception, the Internet was used for “searching” for information and “communicating” with text, files and pictures, and any media. Users were mostly academicians and business people. E-mail and bulletin boards were popular. Next, we were able to “do” things using web sites. Consumers discovered the Internet and commerce exploded. Then, we were able to “socialize” and “collaborate” using social media sites like Facebook. Apps exploded as mobile computing dominated the Internet. Now, we are entering the Internet of “Things,” where any medical device can inexpensively communicate over the Internet with other devices and computers.

Computer interoperability and data exchange

Computer interoperability and data exchange or the sharing of data requires that the data created in one EHR be sent to another EHR with full understanding and context so that it will be placed in the appropriate data field of the receiving program. ASCO and the National Cancer Institute’s caBIG and Community Cancer Center Program created the Clinical Oncology Requirements for the EHR (CORE) that delineated the core common data elements, functionalities, and interoperability standards for oncology that helped to define the original oncology EHR certification criteria used by the Certification Commission for Healthcare Information Technology (CCHIT).³ Seamless interoperability standards have foundational importance by defining and transmitting medical information in a way that will make our data entry, computerized workflows, and result reporting dramatically simplified and efficient.

Most of the data elements that we use in medicine have been defined, standardized, and harmonized into code sets and SNOMED-CT, a clinical vocabulary. Meaningful Use Stage 1 brought some interoperability to public health reporting. Meaningful use stage 2 brought additional interoperability to well-defined content, vocabulary, and transport standards for transitions of care. Our leading Standards Developing Organization (SDO), Health Level Seven International (HL7®) has defined and standardized the use of clinically useful summary documents known as consolidated clinical document architecture (C-CDA) documents.^{4, 5}

ASCO’s Data Standards and Interoperability Task Force, which I chair, has developed two oncology-specific interoperable technology standards. We have taken two ASCO Treatment Plan and Summary paper templates developed for adjuvant breast cancer and adjuvant colorectal cancer^6–8 and transformed them into implementation guides (IGs). The IGs can be used to create an interoperable document that can be electronically exchanged between any computer system that adheres to the HL7 C-CDA standard.⁹ CDAs define content and transport of a document that can be large. This document standard [entitled Clinical Oncology Treatment Plan and Summary (COTPS)] summarizes cancer data from almost all medical reports that are created by physicians, hospitals, laboratories, imaging centers, pharmacies, and patients during the cancer journey and gives review and guidance to the patient and their providers involved with their care. We plan on creating similar C-CDA documents for other common cancers, patient-reported outcomes, and cancer survivorship plans.

At the HIMSS Interoperability Showcase for 2014, ASCO’s COTPS was used to demonstrate how the use of common data elements, standardized reports, adherence to interoperability standards, and use of innovative data entry tools can make provider workflows more efficient and our notes and reports sharable, actionable, and multipurposeful. There, 10 vendors used these standards for exchanging data in real time. Novel data entry tools including voice-recognition transcription to text to data to structured data, a care plan manager, medical devices in the patient’s home demonstrating data capture and alerting capabilities, patient education, patient-reported outcomes and questionnaires, and actionable reports for quality, research, population health, and big data analytic registries were shown. A health information exchange captured and displayed the compilation of clinical reports in real time and made them available to all vendor systems while recording the cancer journey of a woman given adjuvant treatment for her breast cancer. The care plan manager was able to electronically create the ASCO COTPS for the patient, caregivers, and all providers.

HL7 is closer to approving a new, nimble, and simpler interoperability and data exchange standard called Fast Healthcare Interoperability Resources (FHIR).¹⁰ FHIR can represent granular or discrete data elements and documents. Part of the enthusiasm surrounding FHIR is due to the elegant simplicity of the technology. It combines the best features of HL7 V2, HL7 V3, and CDA, while leveraging the latest web service technologies. The major technology change embodied in FHIR is a fundamental move away from a document-centric approach to a data-level access approach using application programming interfaces (APIs). The ability of this interface to exchange data with any software including the EHR using a public API can foster an ecosystem of interoperability among health IT systems. Specifically, FHIR features a modular concept called Resources that can be extended and adapted as a step forward from today’s document-based exchange architecture using a very basic set of structured data like a medication list or lab result. FHIR provides for a plug-and-play platform working within and between systems that is conceptually similar to the Apple app system. FHIR would also allow developers to create new and innovative apps by using the public APIs and following a well-defined set of rules. Most vendors would likely embrace a public API offering a unified approach to share data with any application. Systems can easily read the extensions using the same framework as other resources, and when FHIR and the SMART platform are combined, we have a nimble interoperable standard that can capture the functionalities of Apple’s App model. Recently, many EHR vendors and health care organizations have formed the Argonaut Project to accelerate query/response interoperability under the auspices of ANSI-certified HL7 standards development organization processes for sharing electronic health care information, a most promising change from groups that have in the past encouraged being proprietary.¹¹

For the oncologist, we will have true interoperability and data exchange whereby small nimble apps will be able to accept data from another app or an enterprise-wide EHR, use that data effectively, send it to another app and eventually return it to the originating application source with updated data that can be understood and incorporated into the originating program’s database. Efficient data entry tools, educational links for the oncologist or for their patients (think CDS), wellness data incorporation and useful personal health records (PHRs) creation, important current and meaningful alerts, and automated secondary uses of data can be reported such as quality measures, state registry data, and rapid learning systems (RLSs). Enterprise-wide EHRs that takes years for updates and do very little in educating the user can rely on these nimble apps from trusted sources to enhance their products. Centers for Medicare and Medicaid Services (CMS) has made interoperability a major goal for the country in its recent report “Connecting Health and Care for the Nation.”¹²

Electronic health record database and repository

The EHR is a systematic database that collects longitudinal, attributable, secure, nonmodifiable digital health information and data about an individual patient or population. It should be capable of being shared across all health care settings, providers, and with the patient. It should represent data that accurately captures and presents the latest health status of the patient and their clinical encounters. It does a superb job in documenting EHR-defined clinical activities and administrative requirements, but today it is unacceptable in providing efficient data entry, educational and safety features, effective provider workflows, helpful CDS, seamless interoperability across all EHRs, plug-and-play modular designs and automated registry, and other secondary reporting needs. This is because the initial EHR creators modeled it after a paper chart that was passive and a nonparticipant in health care. It lacked an understanding of the clinical workflow or a provider’s or the health care system’s needs. Today, government financial incentives for purchasing EHRs and meeting meaningful use that require some interoperability, quality improvement, CDS, and registry reporting have persuaded the majority of oncologists to adopt them. Unfortunately, they are frustrated because these meaningful use functionalities are poorly designed into today’s EHRs. EHR data must not only document our clinical activities with relevance, succinctness, and context that paper systems may have more easily captured in the past, but it also must be used for secondary purposes such as clinical and population research, quality measurements, registry reporting, reimbursement documentation, and RLS analytics to name a few that current EHRs do not effectively address. We spend less face-to-face interaction time with our cancer patients, more time with the computer screen using menus, interfaces, and alerts,¹³ and our workdays are averaging 48 min longer due to EHR use.¹⁴

EHRs must become more adept in data entry providing voice to text to data functionality (dictate and have the computer use natural language and machine reading skills to place the dictated material into corresponding data fields). Also, by opening up their software, EHR vendors could unleash a creative army of developers that could help their users and improve the product.

The Medical Informatics Committee of the American College of Physicians has recently explored EHR documentation issues and written an excellent position paper with cogent recommendations for changes that are listed in Table 3. Yet, computers will remain our individual link to our health care system for the foreseeable future. We must find more efficient ways to capture and reuse data and have computers do more of our work.

Source: Data from Ref. 15.

Personal health records

Patients will continue to take a more responsive role in health care as they pay a larger share of its cost, make known their values and wishes, and help make key decisions. With the unlimited educational resources of the Internet, our patients have access to the same medical literature and textbooks that we have. With meaningful use requirements for providing electronic patient reports to the patient, with the CMS proposing that patients have access to their laboratory test results, with e-mailing between patients and oncologists becoming commonplace (and reimbursable with some payers), and with most medical reports becoming available in digitized form, patients will be in control of their medical record. A pilot project that makes almost the entire medical record electronically available to patients has been successfully implemented at the MD Anderson Cancer Center; the majority of patients are more than satisfied, and most doctors, many of whom were skeptical initially, have become converted proponents of “open access” care.¹⁶ The highly successful OpenNotes project is an example of how eager patients are receiving more of their medical information.¹⁷

Knowledge and education

Oncology knowledge bases

After aggregating a cancer patient’s actionable health data into our EHR using our data gathering and entry tools, we need to have our EHR assist us with CDS tools to search our available knowledge bases today and prepare for contributing to a RLS for oncology with our EHRs to hopefully provide us with a “technology to aggregate individual patient data across populations that share common characteristics defining a health state in order to generate knowledge and learning” into improving patient outcomes, quality, and health care value utilizing practice guidelines or algorithms as so eloquently discussed by Yu.¹⁸ Our knowledge bases in oncology consist of clinical trials data, systems biology data, patient-sourced data (health and wellness), and health systems data on operational processes and patient outcomes.

Clinical trials data come from being published in peer-reviewed journals or systematic reviews of the medical literature and being vetted and weighed for quality and knowledge gaps by experts gathered by a professional society and published as a clinical practice guideline.

Systems biology data in oncology help us characterize the molecular, cell growth regulatory and immunologic changes found in different cancers as they progress, permitting us to understand their causation, classification, and therapies. As computational biology and their analytical tools begin to deal with the terabytes of panomic data from an individual patient whose genome, transcriptome, proteome, and metabolome could be routinely assessed in the near future, nimble resources such as The Cancer Genome Atlas¹⁹ and the Global Alliance for Genomics and Health²⁰ will support providers and researchers in their quest for providing personalized medicine and improved patient outcomes. These knowledge bases working together will permit a more rapid drug evaluation and approval process for targeted drugs by identifying exceptional cancer responses associated with specific panomic aberrations.

Finally, the Digital Health Information Technology System will use our EHR clinical data, patient-reported data, laboratory information systems and data repositories, administrative claim data, cancer registries, wellness, activity and nutritional data, and postmarketing surveillance data to capture health care systems data that reflect the real-world patient population experience with our therapies.

Clinical decision support systems (CDSS)

CDSS (clinical decision support systems) are user-chosen or situational awareness invoking application tools that have access to knowledge bases or practice guidelines that are used by EHRs to educate and/or alert their users about patient-specific clinical decision making and health management at the point of service. Musen et al. have described the following three types of CDSS²¹:

• Information management (direct link to a textbook section or Internet site or a tool such as an Infobutton that is a context-sensitive link embedded within an EHR, which allows easy retrieval and subsequent electronic reuse of the relevant information residing in the EHR or at an Internet link)²²
• Situational awareness (alerts or patient dashboards)
• Patient specific logic-based guidance on diagnostic and therapeutic choices such as could be derived from Knowledge bases or a RLS like CancerLinQ.¹⁸

Rapid learning systems

As cancer patients increasingly have their medical records digitized, it becomes obvious that there is a treasure trove of clinical data in these records that have the potential to benefit society by opening up what happens to the 97% of cancer patients who do not go on clinical trials. By learning about the comparative benefits or harm of our new treatments and procedures in nonclinical trial patients after these therapies have been granted regulatory approval, we can apply findings and improve treatments on a continuous, ongoing basis. The National Cancer Policy Forum of the Institute of Medicine workshop entitled “A Foundation for Evidence-Driven Practice: A Rapid Learning System for Cancer Care” examined the elements of a RLS for cancer.²³ It recommended that the elements of such a system include registries and databases, emerging information technology, tools for patient-centered and patient-driven CDS and patient engagement, ways of accommodating culture change, clinical practice guidelines, point-of-care needs in clinical oncology, and federal policy issues and implications. ASCO and others have begun to define a RLS for cancer called CancerLinQ.^{24, 25} It will use these tools to develop a more thorough understanding of cancer biology, defining a cancer based on a molecularly driven diagnosis; it will also incorporate a therapeutic development system that uses oncology EHR registries to produce smarter and faster clinical trials. Because of its more seamless integration of clinical and translational research, it has the potential to ensure that every cancer patient’s experience can inform research and improve care.

RLS will require computational tools for data compilation and analysis from pooled massive databases combining machine learning, data visualization, statistics, and programming. For our cancer patients, this Big Data operation could incorporate all clinical data, genomic data, wellness and health data derived from medical devices and medical payment data obtained from Medicare and the private payers. ASCO’s CancerLinQ hopes to obtain all EHR and clinical knowledge base data for all cancer patients residing in the United States from our offices, cancer centers, and hospitals. An oncologist could enter all the clinical information on his patient and ask CancerLinQ to electronically match the patient to the closest group of patients with similar clinical characteristics and reveal the treatments that produced the best outcomes for these similar patients. In addition, it could generate knowledge for improving quality and cost effectiveness of care by creating an informatics infrastructure for practice-based research networks to collect practice-based evidence using CDSS functionality built into our EHRs to link machine-readable oncology knowledge bases and guidelines repositories as originally described by Sim et al.²⁶

The mobile computing revolution

Today, general-purpose Windows and Mac OS X computers are the devices most commonly used for EHRs, but they carry the burden of 30 years of rapid, unplanned change. Many physicians and patients are frustrated with a lot of the existing EHR software choices that rely on the existing PC and Mac paradigm that uses windows, icons, menus, and pointers for human–computer interaction.

With the introduction of the iPhone by Apple in 2007, Apple introduced a natural human-like interface that could sense touch, touch force, voice, vision, hearing, location, body position, height, and direction. It could communicate over Wi-Fi, Bluetooth, cellular, and NFC (near field communication). It has become the most easily learned and most efficient mobile operating system available. Most EHR vendors today are providing access to their programs via iPhones, iPads, and Android devices, and others are calling for an Apple-like app platform for EHRs.²⁷

Government and medical leadership consensus on a mobile HIT evolution

For the first time, our thought leaders in health information technology (HIT), governmental groups overseeing health information and EHR technology, Congress, our President and his expert panels, our medical leaders and specialty organizations, medical standards organizations, informatic experts, the health care industry and even many, vendors/developers agree that the technologic infrastructure, and recent advances in hardware, software, and services can provide a new relationship between physicians and their computers where computers serve us as an efficient, up to date, personalized, customizable, secure, shareable, and knowledgeable assistant. Halamka popularized the term “electronic medicine”²⁸ to urge EHRs to adopt these mobile computing advances in 2011, and I have taken the liberty to include additional items that have been developed over the past 4 years that will help oncologists participate in a true Oncology Digital Health Information Technology System.

Cloud computing

Cloud computing refers to storing patient data in centralized servers connected to the Internet rather than in our offices and provides the ability to scale up complex software that can be updated frequently and simultaneously with all software users at different sites thus removing the need of our offices to deal with database administration, server hosting, and data security at great financial savings. By being Internet connected, EHRs can more easily share data with patients and other physicians, send reports to external registries and have sophisticated tools to assist with learning the software initially and later on with updates.

Apps

Apps are versatile, small programs that use web-based APIs, that have extensions which can work with data passed on from another app or an EHR program. Apps could provide new functionalities for the EHR or could collect data from the EHR and send it to registries or RLSs that could analyze this new data and provide CDS for the EHR program. APIs are simply a collection of functions that a programmer can call to use another programs services and data. APIs contain tiny, atomistic chunks of data that must use consistent units and terminology that are handled much more efficiently. Natural language processing is being used to extract these tiny chunks of data from our narrative dictations and create structured machine-readable content that can be used more effectively by the computer for any purpose. It is an ideal method for permitting a medical specialty society to provide specialty-specific updated tools for EHRs that find it hard to keep their software up to date with the needs of the medical specialist.

Modular software and application programming interfaces (APIs)

The SMART (substitutable medical apps and reusable technology) platform designed by Mandl and Kohane²⁷ is an ideal platform to bring Apple’s App model system into the EHR ecosystem. Mandl recently pointed out how this platform could work in an emergency situation to permit the Center for Disease Control and Prevention (CDC) to create an App that could work with any cooperating EHR and reshape ER triage workflows to emphasize travel history and recommend rapidly updated assessment and isolation guidelines for handling patients who present to an ER with fever and recent travel to an Ebola-affected region.²⁹ The App could be immediately used by all EHRs, updated easily, and specific data about all patients quarantined by a hospital could be sent back to the CDC for monitoring and reporting purposes on a national level. He points out that this hypothetical app could be rapidly written once and run nearly everywhere.

Engaged, digitally connected digitized patients, medical devices, and the Internet of Things

Over this time period, our health system has been transformed so that patients instead of being passive participants in the health care system are now very engaged as they have access to the same medical literature that we have on the Internet and their complete medical record permitting joint decision making with their providers leading to better outcomes and fewer malpractice suits. Using digital educational tools now available on web sites or in Apps, patient have a better understanding of their diseases, treatments, upcoming procedures, suitable clinical trials, informed consent, and can provide their values, wishes, and misconceptions directly to shared EHRs in some institutions.

As we enter the Internet of Things stage with inexpensive, accurate, intelligent medical devices, disease prevention and early medical interventions along with the increasing interest in the wellness of the patient will lead to the rapid uptake of these medical devices and wearable computing garments that connect wirelessly with our smartphones, facilitating reimbursable telemedicine consults with our patients with chronic illnesses at home and early detection of adverse events from our therapies or their diseases. Using digital devices to measure vital signs and digital EKGs, stethoscopes, cameras, ophthalmoscopes, otoscopes, and ultrasound devices attached to our smartphone, the physical exam can be completed remotely.³⁰ Monitoring and analyzing this digital data will require Big Data tools that are rapidly being developed to transform medicine as it is currently practiced. Also, certain regulatory and compliance issues arise when using medical devices such as security, auditability, and the need for provenance that are currently being addressed.

Precision medicine

Precision medicine is the prevention and treatment strategies that take individual variability into account using large-scale biologic databases (such as the human genome sequence), powerful methods for characterizing patients (such as proteomics, metabolomics, genomics, diverse cellular assays, and even mobile health technology), and computational tools for analyzing large sets of data. Whole genome sequencing is reaching the $1000 threshold and we are fast approaching the ability to sequence the “digital” patient’s germ line DNA, RNA, microbiome, epigenome along with their environment, anatomical, and clinical features.³¹ The breadth of this digital human data will permit President Obama to launch his Precision Medicine Initiative that supports a common medical data set, true interoperability of medical data that remains private and secure and a value- and quality-based payment system.^32,33

Apple-ization of medicine

Apple, through its unique ability to create, market, and integrate the best hardware, software, and computer services and its ability to work with other leading companies in many diverse fields, has caused a revolutionary transformation of many industries. Now with its latest mobile operating system, platform kits, mobile computing architecture and app models of software and their recent alliances with many of the largest health care industrial stakeholders, it has the ability to do the same for health care. Apple’s HealthKit that synchronizes health and wellness data, PHRs and medical device data, is a platform that resides on the iPhone that interfaces with cooperating medical devices. The iPhone gathers information on vital signs, body position, activity, behavior, nutrition, pulmonary function, and sleep using its accelerometer, microphone, gyroscope, and GPS (global positioning system) sensors and links to other medical devices to gain insight into a patient’s gait, motor impairment, fitness, speech, nutrition, memory, asthma inhaler use, and biomedical analytics such as glucose and oxygen. Data are safe, secure, and HIPPA compliant and is presented to the patient in a pleasant, graphic manner. It is a two-way medical communication platform that patients control. Data can be electronically sent to any cooperating EHR, analytics warehouse repository, or provider group and cannot be sold, misused, or seen by Apple.

Apple’s medical ResearchKit platform is an open-source software tool that will empower consumers and patients to decide if they want to participate in a medical study and how their data are to be shared with researchers. It is designed to give medical researchers a new way to gather information on patients residing anywhere in the world by using their iPhones and Apple Watch. The ResearchKit platform is designed to work hand in hand with Apple’s HealthKit software. Patients can digitally sign up for a trial, sign consent forms, communicate directly with the research center, have up to date information on the study, and have their results and pooled results. Innovative research apps are able to monitor the physical, mental, and behavioral effects of disease along with monitoring adverse events to treatments. A Stanford cardiac study signed up over 10,000 patients in 3 days compared with their expectation of getting 50 patients per institution over 1 year!

For oncologists, these innovative developments will permit our patients to wear devices that can be used to measure and send data over the Internet to their smartphones that will have health platform software for HIPAA compliant storage, analytic and graphical representation of the captured data that can then be sent to an individual’s providers, hospitals or medical alert capture stations all chosen by the patient.

Mobile computing, actionable, and shareable data provided by seamless interoperability standards and an oncologic RLS will continue to provide us with the tools to efficiently and effectively produce a health care system that “just works” for all stakeholders.

References

1. 1 Christensen CM. The Innovator’s Dilemma: When New Technologies Cause Great Firms to Fail. Boston, MA: Harvard Business School Press; 1997.
2. 2 Christensen CM, Grossman JH, Hwang J. The Innovator’s Prescription: A Disruptive Solution for Health Care. New York, NY: McGraw-Hill; 2008.
3. 3 American Society of Clinical Oncology. Clinical oncology requirements for the EHR (CORE). October 6, 2009. Available from: http://www.asco.org/sites/default/files/oct_2009_-_asco_nci_core_white_paper.pdf. Accessed October 15, 2015.
4. 4 http://www.hl7.org
5. 5 https://en.wikipedia.org/wiki/Clinical_Document_Architecture
6. 6 Hewitt M, Greenfield S, Stovall E. From Cancer Patient to Cancer Survivor: Lost in Transition. Washington, DC: The National Academies Press; 2006.
7. 7 American Society of Clinical Oncology. Chemotherapy treatment plan and summary resources. Available from: http://www.instituteforquality.org/chemotherapy-treatment-plan-and-summaries. Accessed October 18, 2015.
8. 8 Salz T, Oeffinger KC, McCabe MS, et al. Survivorship care plans in research and practice. CA Cancer J Clin. 2012;62:101–117.
9. 9 HL7 CDA® R2 Implementation Guide: Clinical Oncology Treatment Plan and Summary, Release 1 Clinical Oncology Treatment Plan and Summary, DSTU Release 1 2013 [cited September 5, 2014]. http://www.hl7.org/implement/standards/product_brief.cfm?product_id=327. Accessed October 15, 2015.
10. 10 http://www.hl7.org/implement/standards/fhir/
11. 11 http://mycourses.med.harvard.edu/ec_res/nt/6209858F-CDDD-4518-ADF8-F94DF98B5ECF/Argonaut_Project-12_Dec_2014-v2.pdf
12. 12 http://www.healthit.gov/sites/default/files/nationwide-interoperability-roadmap-draft-version-1.0.pdf
13. 13 Electronic Health Records (EHRs) in the Oncology Clinic: How Clinician Interaction with EHRs Can Improve Communication with the Patient http://jop.ascopubs.org/content/early/2014/07/15/JOP.2014.001385.extract.
14. 14 McDonald CJ, Callaghan FM, Weissman A. Use of Internist’s free time by ambulatory care electronic medical record systems. JAMA Intern Med. 2014;174(11):1860–1863.
15. 15 Kuhn T, Basch P, Barr M, Yackel T. For the medical informatics committee of the American college of physicians. Clinical documentation in the 21st century: executive summary of a policy position paper from the American college of physicians. Ann Intern Med. 2015;162:301–303.
16. 16 Merrill M. Patients, referring docs at MD Anderson making good use of Web portal. Healthcare IT News July 6, 2010. Available from: http://www.healthcareitnews.com/news/patients-referring-docs-md-anderson-making-good-use-web-portal. Accessed March 15, 2012.
17. 17 ^{http://www.myopennotes.org}
18. 18 Yu PP. Knowledge bases, clinical decision support systems, and rapid learning in oncology. J Oncol Pract. 2015;11:1–4.
19. 19 http://cancergenome.nih.gov
20. 20 http://www.genomicsandhealth.org
21. 21 Musen MA, Greenes RA, Middleton B: Clinical decision-support systems, in Shortliffe EH, Cimino JJ (eds): Biomedical Informatics: Computer Applications in Health Care and Biomedicine. London, UK, Springer London, 2014
22. 22 http://www.hl7.org/implement/standards/product_brief.cfm?product_id=208
23. 23 Institute of Medicine. A foundation for evidence-driven practice: a rapid learning system for cancer. Available from: http://www.nap.edu/openbook.php?record_id=12868. Accessed March 12, 2012.
24. 24 Abernethy AP, Etheredge LM, Ganz PA, et al. Rapid-learning system for cancer care. J Clin Oncol. 2010;28:4268–4274.
25. 25 Accelerating Progress Against Cancer: ASCO’s blueprint for transforming clinical translational cancer research. November 2011. Available at http://www.cancerprogress.net/blueprint.html. Accessed March 12, 2012.
26. 26 Sim I, Gorman P, Greenes RA, et al. Clinical decision support systems for the practice of evidence-based medicine. J Am Med Inform Assoc. 2001;8:527–534.
27. 27 Mandl KD, Kohane IS. No small change for the health Information economy. N Engl J Med. 2009;360:1278–1281.
28. 28 Halamka JD The rise of electronic medicine. http://www.technologyreview.com/news/425298/the-rise-of-electronic-medicine/.
29. 29 Mandl KD. Ebola in the United States: EHRs as a public health tool at the point of care. JAMA Published online October. 2014;20. doi: 10.1001/jama.2014.15064.
30. 30 Sinofsky S. Patience, IoT is the new electronic. http://blog.learningbyshipping.com/2015/02/02/patience-iot-is-the-new-electronic/.
31. 31 Topol EJ. Individualized medicine from prewomb to tomb. Cell. 2014;157(1):241–253.
32. 32 Proposed Shared Nationwide Interoperability Roadmap http://www.healthit.gov/sites/default/files/nationwide-interoperability-roadmap-draft-version-1.0.pdf.
33. 33 Francis S, Collins FS, Varmus H. A new initiative on precision medicine. N Engl J Med. 2015;372:793–795.