Saturday, March 28, 2009

Case for a Collaborative Health-Support Software System - Part 3 of 3

Introducing the ReAsure HealthNode™ Software Technology

In my previous two post (click here for the 1st in the series), I (a) presented a case for using collaborative health-support software systems in loosely-coupled networks of healthcare professionals and consumers and (b) explained how to implement such systems and networks in a way that increases value to the consumer. In this post I introduce a next-generation software system we've been developing for quite some time, called the ReAsure HealthNode™ software technology, which is designed to help these networks transform our current low-value, fragmented, healthcare industry into a high-value one by enabling people in widely scattered localities to:

  • Collaborate around personalized health-support software
  • Evolve those programs
  • Use the programs to help prevent, treat, and manage health problems efficiently and effectively …
Resulting in:
  • Better care coordination
  • Greater consumer-centered cognitive support
  • superior shared decision-making
  • More competent self-maintenance (self-care).
The system utilizes innovative methods and the patented CP Split™ technology[1] to deliver a unique set of benefits and advantages, which follows.

Benefits & Advantages #1:
A Simple, Flexible, Low-Cost Way to Exchange Health-Support Software and Data Files

The ReAsure HealthNode™ software apps provide a simple, flexible, low-cost way to exchange health-support software and data files rapidly and securely via peer-to-peer mesh networks. These networks operate in a node-based[2] communications architecture in which each computerized device loaded with a ReAsure HealthNode™ software module and a health-support software program becomes a node connected to other nodes in the network. Each node, furthermore, has publisher-subscriber functionality,[3] which means certain nodes serve as publishers and other as subscribers. The publisher nodes send the files and the subscriber nodes receive them. The system uses a "desktop-to-desktop" (application-to-application) architecture and exchanges the files via FTP (file transfer protocol), e-mail attachments, or other methods of file transfer through Internet connections.

At one end of the connection, the node serving as the publisher must authorize the transfer of a health-support software file and/or data file by authenticating that each subscriber node is allowed to receive them. At the other end of the connection, the subscribing nodes must allow the publisher node to deposit any necessary file(s) into the subscriber's computer.

Exchanging Health-Support Software Files

The health-support software files designed for the ReAsure HealthNode™ apps consist of particular types of software templates. These templates are electronic files that run (execute) the health-support software by applying particular rules (algorithms) consisting of mathematical formulas, logic, text functions, and formatting instructions (which reflect the software's underlying model). The software templates have two basic functions:
  • Create Data Files (also called "content files") for distribution, which is a publisher function
  • Consume (extract data from) the content file and render it in reports or enable new data to be input, which is a subscriber function.
The health-support software template files are sent from the publisher node to its subscribers when the initial connections are made, and again if the templates have been modified. The files are also exchanged when network collaborators want to compare and examine the different software programs, discuss their findings, debate the pros and cons of competing programs, modify the programs, build new programs, etc. This process is crucial for the software's evolution (continuous improvement) and can be facilitated by giving the nodes access to knowledge management software containing a virtual forum.[4] The oil & gas industry, for example, has used a node-based virtual forum, which enable participants to share lessons learned, best practices, after action reviews, drilling reports, and health & safety alerts on deep water drilling rigs around the globe.[5] Such a program can be applied to healthcare to help collaborators evolve the health-support software.

Exchanging the Data that the Software Uses

In addition to exchanging health-support software files, the ReAsure HealthNode™ Network system uses node networks to exchange and transform data files containing consumers' health information. Figure 1 (below) describes the basic components and processes of its node-based data file exchange method, which are explained following the figure.

Click image to enlarge

Starting on the left of Figure 1, the Software Template File box depicts the health-support software template file and the arrow through it depicts the publisher node sending the file to the subscriber nodes as discussed above. Note that ReAsure HealthNode™ apps can also interface with other software programs via application programming interfaces APIs[6] (which is not depicted in the Figure 1).

The Node as Publisher box in Figure 1 depicts a node that:
  • Uses a health-support software template to:
    • Acquire data and information (content) from specified sources, as depicted by the arrow from the Data Input to Data/Information Sources box, which is a template input function. The content may include numbers and text, as well as links to electronic documents, images, and web sites.
    • Process (manipulate, transform, and organize) the content as defined by the template's algorithms and then package that content in encrypted, delimited (e.g., comma separated value)[7] content file (Data File). Since the content file do not contain any formatting instructions or "markup tags,"[8] they are very small to transmit and efficient to use. These template output functions are depicted by the arrow to the Content File box.
  • Uses its publisher module to transmit the content file to its subscribers via e-mail or any other methods, as depicted by the arrow to the Node as Subscriber box, which is a publisher node transmission function.
The Node as Subscriber box depicts a node that:
  • Uses its subscriber module to retrieve the data & information from the content file, which is a subscriber node input function, as depicted by the arrow coming out of the Content Files box.
  • Uses a health-support software template to:
    • Format the content and displays it in interactive reports, which is a template output function, as depicted by the arrow to the Report Display box.
    • Enable a person to enter new and modified content through its user interface, as depicted by the arrow to the Data Input box, which is template input function. Note that his added or changed content may be used to update the existing content file stored locally in the person's computer, as depicted by the arrow to the Existing Local content file box, or the content may be exported to any databases or other data sources, as depicted by the arrow back to the Data Input to Data/Information Sources box; both of which are template input functions.
Since the content files are typically much smaller than the software template files, transmitting the content files as encrypted e-mail attachments has several advantages:
  • It provides a convenient and transparent automated method that accommodates everyone's needs—from people with continuous broadband to those connected to the Internet only occasionally, as well as those using low speed dial-up services
  • It saves money by eliminating the need for expensive hardware, high-speed networks, costly IT support, and complex software systems.
Figure 2 (below) depicts how a network of nodes operates to exchange information.

Click image to enlarge

Following is a description of the six steps shown in Figure 2.

Step 1: Line #1 depicts the node at the top retrieving content from databases (including EMR/EHRs, laboratory systems, etc.), as well as from electronic files and other sources, and then processing that content to create a content file using node functions defined in its health-support software file.

Step 2: The solid blue arrows of line #2 show the node at the top using the publisher functions to send content files via encrypted e-mail attachments to the node at the upper right, the nodes on the left, and the node at the bottom.

Step 3: This dashed arrow (line #3) shows the top node, after sending content files to the node on it left, subsequently receives the content file from that same node and retrieves it via its subscriber functionality. This means both these nodes invoke their publisher and subscriber functionality.

Step 4: These two nodes receive and retrieve content files; only their subscriber functionality is invoked.

Step 5: These dotted arrows show content files being passed sequentially from one node to the next, with each node adding to or modifying the files it receives, before sending extended content files to the next node; their provider and subscriber functions are invoked.

Step 6: The bottom node receives content files from two other nodes via its subscriber functionality. After forming a composite content file from the accumulated content, as defined by its health-support software templates, it sends the composite content file back to the node at the top using its publisher functionality.
Thus, based on Figures 1 and 2, networks of nodes using publisher subscriber modules and health-support software can:
  • Exchange and evaluate software template files
  • Create and share content files
  • Generate personalized, interactive reports
  • Modify the contents file and then reuse it to generate updated reports or export it to other data stores.
Following is a practical example of a collaborative health-support software system in action.

A Practical Example

Imagine a consumer having the ReAsure HealthNode™ modules in his computer, along with a health-support software program. When he installed the node modules, he went through a registration process during which he sent requests, as a publisher, to the nodes of people to whom he wants to send certain of his health information. He also sent requests, a subscriber, to people from whom he wanted to receive information. In addition, he received requests from others who want to receive his health data (i.e., they subscribed to him). The nodes used e-mail to transmit the content files, so the e-mail addresses of the corresponding nodes were stored during registration. As each request was approved by the appropriate parties, the transfer of software template files and content file began.

Now, whenever he wants to sends particular data to the healthcare providers he authorized as his subscribers, he click his mouse to initiates certain publisher node functions, which instruct his node to:
  • Retrieve and decrypt his content file
  • Extract from his content file the particular data authorized for each provider; these authorizations are based on the provider's "role" (e.g., the provider's area of specialty)[9] Pack the extracted data into other content files and encrypts them
  • Attach the content file to e-mails and sends them to the appropriate providers.
Upon receipt of the e-mails by his providers, each of their subscriber nodes:
  • Retrieves the content file from the e-mail
  • Decrypts the content file
  • Extract the data from the content file
  • Exports the data to the provider's electronic medical record system.
All these steps taken by the publisher and subscriber nodes are done automatically.

In a similar manner, the consumer may want to obtain data from a provider or lab, that is, he may want to be one of their subscribers. If the provider (or lab) has a ReAsure HealthNode™ and the registration process is completed, the consumer would send a request to their nodes for a content file containing certain information. In this scenario, the consumer's node uses its subscriber functionality, and the provider (or lab) node serves as the publisher.

Once the consumer's node receives the e-mail with the content file attached, it automatically retrieves and decrypts the content file, extracts its contents, and merges the new data into his own (locally stored) content file. Depending on the software rules, the data may replace (i.e., overwrite) existing data in his content file or it may be added to the existing data. His content file can then be accessed, formatted, and presented by any appropriate health-support software program, such as the Personal Health Profiler™.[10]

In yet another scenario, a provider may request particular data from the consumer, or certain rules the guide in the consumer's node might be executed on behalf of the provider (e.g., if the consumer has diabetes, a rule might tell his node to send his glucose readings to his primary care physicians, endocrinologists and wellness coach once a week). In this case, the consumer's node is once again the publisher, which automatically extracts those data from his content file and sends it to the subscriber's (i.e., providers') nodes.

Data Transformation and Translation

Transmitting content files between nodes and rendering them as reports is only part of the process. What happens when a publisher node sends a content file to subscriber nodes utilized by people who use different terminologies, speak different languages, or have different data storage and presentation requirements? How are the contents of the content file transformed and translated to meet the diverse needs of everyone in such loosely-coupled networks?

Transforming Data

Data often has to be transformed when being sent from one database to another. This happens when, for example, the databases have different table field[11] names (e.g., "birth_date" and "dob") or data formats/syntax (e.g., whether or not dashes should be included in a phone number). What is required, therefore, is either to force everyone to use the same data standard, such as transforming everything to XML[12] using a common "schema" (data structure). Another is to transform the original data names and formats so the data are received in the proper configurations.

XML data standards can be used in the node network to transform data, and the content files can be constructed of data transmitted in XML files. But simpler and most efficient ways include having rules included in the node functionality that instruct:
  • The publisher node to do the necessary data transformations prior to packing the data into the content file before sending it to its subscribers
  • The subscriber nodes to do the transformation upon receipt of the content file
  • Intermediate nodes make the transformations as the content file passes through them.
Note that these transformation methods require that the publisher and subscriber nodes be notified in advance as to the required transformations. This notification process can happen during the registration or upon a subscriber node's request for data from the publisher.

Translating Information

When people in loosely-coupled networks share information there is often the need for it to be translated. In addition to language translation (e.g., English to Spanish), the issue of terminology translation must be addressed. This refers to the problem that occurs when different people use different terms (their local standards) to refer to the same concept.

One common strategy used to avoid such problems is to force everyone to adopt the same global terminology standards[13] by agreeing on one set of terms (semantics). While setting arbitrary global standards for health-related terms is a way to foster widespread communications between people from different regions, organizations and healthcare cultures/communities, there's also a downside to eliminating the local standards people rely upon, i.e., they lose information due to reduced semantic precision and nuance.

Take, for example, the term "high blood pressure;" there are 126 different terms referring to this concept of elevated blood pressure levels. These terms include "malignant hypertension," which refers to very high blood pressure with swelling of the optic nerve behind the eye; it's a condition usually accompanied by other organ damage such as heart failure, kidney failure, and hypertensive encephalopathy. "Pregnancy-induced hypertension," on the other hand, is when blood pressure rises during pregnancy (also called toxemia or preeclampsia). These are very different types of hypertension. So, while referring to a person's condition using a global standard term such as "hypertension" clearly conveys that the person has high blood pressure, the standardized term loses important details found in the more detailed local standard terms. These lost details, in turn, could very well affect treatment decisions and outcomes. So, there is a good reason to have multiple terms for a health-related concept.

Furthermore, relying on global standards are problematic because, as standards evolve, it can be very difficult and costly to change the global standards Consider, for example, the clamor over by switching to the new ICD-10 global standard of diagnostic terms (codes), which have evolved from the ICD-9 standard.[14]

It would be much better, therefore, to keep local standards, support their evolution, and use the data translation described above to ensure everyone gets the information needed using the terms they need and understand. In a node network, this can be accomplished in a similar way data transformation occurs, but instead of transforming the data, the terms that the subscriber node requires either replace, or are added to, the terms in the original content file.

Composite Reporting

In addition to modifying a content file data through transformations and translations through node networks, the nodes support composite reporting. Composite reports consist of information sent from multiple publisher nodes to a single subscriber node. The subscriber node takes all that information and combines it into a single integrated content file, which is then used to generate composite reports containing information from multiple sources.

For example, let's say a primary care physician (PCP) wants to keep track of the treatment a patient is receiving from several provider specialists. The PCP's node, which serves as the subscriber, would send a request for certain data from all the patient's specialists. Upon receipt of the data request, the specialists' nodes, which serve as the publishers, retrieve the requested data from their different electronic health record databases and send the data automatically to the PCP's node. The PCP's node then incorporates the data into a composite report tailored to the PCP's needs and preferences, and then presents the report on screen for the PCP to view. The PCP's subscriber node could also be instructed to request data from the publisher node connected to the patient's personal health record and, upon receipt (and as authorized by the patient), add the data into the same report as authorized by the patient. Likewise, a consumer using the Personal Health Profiler™ software can create composite reports in a similar manner from data sent by multiple provider nodes.

Protecting Personal Health Information

With all this personal health data being sent around, a powerful method is needed to protect people's privacy. While encryption and authentication handles security issues (e.g., making it safe to send content file by e-mail), it doesn't deal with privacy issues (i.e., who is allowed to receive a person's health information). Instead, protecting the privacy of information sent by consumers' nodes requires strategies such as transmitting "limited data sets" and enabling "granular level" of control. That is, consumers should be able to implement one-time authorization to share certain parts of their content files with specific types of providers. They should also have granular control over whom, if anyone, gets to see their information by authorizing particular types of providers to receive particular pieces of information. Warnings and alerts inform the consumer if certain information not being authorized ought to be shared with certain providers who need those data to help make diagnostic and treatment decisions.[15] The Personal Health Profiler™, for example, deploys these privacy safeguards.

Benefits & Advantages #2:
An Elegant Way to Maintain a Complete and Evolving Data Set

A second set of benefits and advantages of the ReAsure HealthNode™ Network system is its ability to maintain a complete and evolving data set that excludes nothing. That is, every possible piece of health information—about one's health status, risks, conditions, treatments received, the clinical outcomes (results) and costs of the treatments, etc.—can be collected, stored and used over a person's entire lifetime. It does it using an indexing and categorizing method similar to the Dewey Decimal classification system libraries use to organize books and magazines.

This all-encompassing method enables rich, detailed health-support software to be built. And it also permits development of holistic (mind-body-environment) health-support software that improve care value by promoting greater understanding of a consumer's health problems, threats, and needs. The holistic approach provides feedback, guidance, and instruction about people's:
  • Signs & symptoms[16] and their possible relationship to medications side-effects treatment procedures, including complementary and alternative medicine[17]
  • Psychological (emotional, cognitive, and behavioral) health and its relationship to biomedical conditions (e.g., the effect of mental stress of blood glucose control in diabetic persons)[18]
  • Genetic markers and the associated health risk factors[19]
  • Health trends based on health status changes and lab test results over time
  • Methods for coping with and solving personal problems.
Benefits & Advantages #3:
A Way to Increase Care Value to Consumers while Protecting Populations

And a third set of benefits and advantages refers to the ReAsure HealthNode™ Network system's ability to support evidence-based research across all healthcare disciplines focused on improving healthcare value to consumers, and to protect populations through surveillance and support of emergency personnel.

Increasing Care Value to Consumers through Evidence-Based Research

Increasing and sustaining care value to consumers requires a simple, flexible, low-cost way to exchange health-support software; a convenient, secure, economical way to exchange the data used by the software; and an effective, efficient way to maintain a complete and evolving data set. By adding the system's ability to de-identify consumers' health data[20] to these previously discussed advantages and benefits, it becomes apparent that the ReAsure HealthNode™ Network system is ideal for researchers involved in creating, studying, and evolving high-value evidence-based guidelines across all healthcare disciplines (both sick-care and well-care, as well as conventional and complementary & alternative care[21]).

This kind of research is essential for development of evidence-based healthcare decision support systems utilizing quality metrics, practice guidelines, knowledge services and tools, and continuous quality improvement feedback loops.[22]

Protecting Populations
  • The ReAsure HealthNode™ Network system also addresses the need to protect populations through:
    Comprehensive biosurveillance and post-market drug & medical device surveillance[23]
  • Support for first-responder and trauma center staff during disasters[24][25]
  • Has no "single point of failure,", which leaves centralized systems vulnerable in disaster situations.[26][27]
Summary and Conclusion

Collaboration among loosely-coupled networks of healthcare providers, researchers, and consumers using health-support software is an essential strategy for increasing healthcare value. ReAsure HealthNode™ Network system demonstrates how to enable efforts to bring consumers high value care.

We welcome any feedback and opportunities for collaboration.

References and Notes
[2] A node is an electronic device (e.g., a PC, laptop, cell phone, and hand-help device) attached to a network, which contains a software module enabling it to send, receive, or forward information across that network.
[4] A virtual forum enables people share information over the Internet, online or offline, through “threaded” discussions in messages on the same topic are grouped together for easy retrieval and reading.
[9] Each provider’s role is established during the registration process and used by the node thereafter.
[20] The ReAsure HealthNode™ system de-identifies consumers’ health data by “decompositing” their content files and extracting the data without identifying to whom it belongs. See

Saturday, March 21, 2009

Case for a Collaborative Health-Support Software System - Part 2 of 3

In my previous post, I presented a case for using collaborative health-support software systems in loosely-coupled networks of healthcare professionals and consumers. This post focuses on how to implement such systems and networks in a way that increases value to the consumer.

How Using Health-Support Software in Loosely-Coupled Networks Promotes High-Value Care

When people use health-support software in loosely-coupled networks, they increase care value by:
  • Fostering coordinated care
  • Delivering consumer-centered cognitive support
  • Sharing decision-making.

Fostering Coordinated Care

Coordinated care is a strategy for ensuring that consumers being treated by multiple providers in loosely-coupled networks—including primary care physicians, biomedical and mental health specialists, and wellness coaches—who work together in a synchronized manner to deliver high-value services. Methods for implementing this strategy include the use of personalized health-support software to deliver whole-person integrated care and support medical home deployment.

Using Personalized Health-Support Software

Healthcare providers delivering coordinated care in loosely-coupled networks require different types of health-support software. That's because providers in the network typically work at different locations, use disparate information systems, and are from different healthcare disciplines that require different health information. For example, a primary care physician, being a generalist, needs a broad spectrum of patient information covering patients' biomedical, psychological, and environmental factors (including data about current medical conditions, medications being taken, allergies, vital signs, basic lab results, medical history, stress and other emotional factors, etc.). While all providers would benefit from similar information, a specialist would benefit from a more in-depth sub-set of data related to their area of specialization.

A cardiologist, for example, would benefit from data related to heart functioning.[1] A dentist would benefit from data about previous dental work done, dental x-rays, existing medical conditions affecting teeth and gums, etc. A mental health practitioner would benefit from detailed information about the relationship between a person's thoughts, emotions, and behavior, as well as psychosocial data, etc. A holistic (integrative/integrated medicine) practitioner would benefit from addition information about the mind-body connection, metabolic functioning, etc. Furthermore, the consumer would benefit from a personal health profile report that includes risk-appraisal, current health status, and self-management information in lay language. That means successfully coordinating care requires use of many different health-support software in loosely-coupled networks.

In other words, health-support software should be personalized, that is, tailored to each person's particular requirements.

Delivering Whole-Person Integrated Care

Delivering whole-person integrated care has two components:[2]
  1. The whole-person part focuses on improving a person's health and wellbeing by addressing one's physical health (body), mental/psychological health (mind), and the mind-body connection ("holistic" health). In other words, it views an individual as a whole entity, whose body and mind are interconnected.
  2. The integration part refers to integrating well-care with sick-care. Instead of viewing sick-care (treatment of illness and dysfunction) and well-care (preventions and self-management of chronic conditions) as two separate avenues in the road to health, sick-care/well-care integration refers to an integrated care delivery system.
    Instead of viewing sick-care and well-care as two separate avenues in the road to health, this integrated approach involves a new kind of coordination and collaboration between (a) medical and related sick-care practitioners focused on the diagnosis and treatment of health problems and (b) well-care practitioners focused on prevention, recovery and well-being, as well as peak performance.[3]
An example of the whole-person integrated care approach is when people with chronic illnesses receive care from their medical providers, along with coaching, instruction and social support from a health education and support program. While the people's doctors deal with medical issues (including medical procedures and medication subscriptions), the health program focuses on enabling and motivating the individual to self-manage chronic conditions (such as diabetes) by developing healthy attitudes and beliefs, emotions, coping strategies, psychosocial relationships, and life satisfaction, as well as gaining essential awareness, knowledge and skills.[4]

Since this type of whole-person integrated care involves multidisciplinary teams of providers, as well as the consumer's social connections, the individuals involved form a loosely-coupled network. The health-support software they use, therefore, should focus on:
  • Giving the providers timely information about consumers' progress and blocks
  • Enabling the professionals to collaborate easily and efficiently
  • Sparking helpful conversations among the social networks
  • Educating and instructing the consumers, as well as providing ongoing health status feedback.

Deploying Medical Homes

Instead of a fragmented healthcare delivery system as exists today, continuity of care through medical home makes a person's primary care physician responsible for coordinating the services rendered by the multidisciplinary team of providers working with the person.[5][6] Health-support software for medical homes[7] focus on enabling the providers to know what each other is doing, their care the deliver is better connected, which helps improve quality and reduce costs by avoiding duplication of tests, medication conflicts (e.g., due to drug-drug interactions[8]), inappropriate or conflicting procedures, etc.

Delivering Consumer-Centered Cognitive Support

Consumer-centered cognitive support assists healthcare providers and consumers in finding useful answers, making valid health decisions, collaborating effectively, and taking competent, responsible action. This support increases care value through use of collaborative health-support software that:
  • Analyze and help interpret voluminous, complex diagnostic, treatment, and outcomes data
  • Identify existing and threatening physical and mental health problems
  • Search for scientifically validated options (diagnostic, treatment and preventive)
  • Help develop high-value (effective and efficient) plans of care
  • Enable loosely-coupled networks of providers, consumers and researchers to share and discuss lessons learned (e.g., observations, insights, hypotheses, explanations, and anecdotal information)
  • Continually evolve clinical knowledge by providing ongoing process and outcomes data to researchers for evaluating, building and modifying evidence-based guidelines.[9]

Promoting Provider Competence

Collaborative health-support software can promote provider competence by delivering consumer-centered cognitive support information. That is, provider's diagnostic and treatment decisions, and the implementation of those decisions, can be supported and improved through by the software's data analyses, care-plan assistance, and evidence-based guidelines.

Promoting Consumer Competence

In addition to assisting healthcare providers, consumer-centered cognitive support helps consumers acquire the knowledge, skills, and characteristics needed to make wise decisions and act in ways that improve one's health and wellbeing.

For health-support software to promote consumer competence, it should collect and analyze comprehensive, valid, reliable (complete, accurate, and dependable) data. These data are facts and figures, collected on multiple occasions (over time), which include:
  • Internal biological measures (biometrics) such as height, weight, blood pressure and other vital signs, cholesterol level, blood glucose and other blood component levels, imaging studies and lab tests, illnesses, allergies, genetics, etc.
  • Psychological measures of mood and emotions, cognitions (thoughts and beliefs), behaviors (e.g., exercise, eating, sleeping, smoking, substance and alcohol use), social relationships, learning styles and (dis)abilities, significant past experiences (e.g., memories, traumas), etc.
  • Healthcare treatments, including medications taken, procedures received, and their outcomes
  • Preventive actions, such as inoculations, wellness coaching, and health education
  • Demographics, which include age, gender, finances, ethnicity, etc.
  • Environmental conditions, past and present, including neighborhood crime levels, pollution, etc.
All these data must be analyzed using valid computational methods that generate actionable information that helps consumers make valid predictions, build useful knowledge and skills, make sound decisions, increase positive motivation, improve physical health, and have greater peace of mind (e.g., reducing mental stress and emotional distress). This information should include:
  • Trends and tendencies, which give insights into what is likely to happen in the future by studying what has been happening in the past and in the present
  • Associations (relationships, interactions) showing the connection between body, mind and behavior
  • Warnings and alerts indicating, for example, when it's time for a medical check-up or inoculation, when there are possible drug-drug interactions a person is having serious medication side-effects, if one's lab tests indicate a health problem, etc.
  • Guidelines and instructions informing a consumer about how to deal with a risk factor or health.[10]

Sharing Decision-Making

Shared decision-making occurs when healthcare consumers are able to make knowledgeable decisions about their own care in collaboration with their providers. It reflects the principle of self-determination and involves respecting people's right to define their own view of what is good for them and to pursue that view, as well as to give others the authority to make particular health care decisions for them. To make shared decision-making successful, healthcare providers must not only to understand consumers' needs and develop reasonable ways to meet those needs, but they must also present the alternatives in a way that enables consumers to make a knowledgeable choice. They must also take into account consumers' attitudes, preferences, and values.[11]

The Center for Shared Decision Making[12] employs decision-making coaches who use various decision-making aids, including pamphlets and videos, to help consumers examine and weigh their options about tests and treatments. Collaborative health-support software can be useful in shared decision-making by offering similar, computer-generated information to loosely coupled networks of coaches and consumers.

To summarize: High-value care—both provider-supplied and self-care—relies, in part, on the use of continually evolving, personalized, health-support software in loosely-coupled collaboratives to:
  • Foster coordinated care
  • Deliver consumer-centered cognitive support that promotes competent both professionally-rendered care and self-care
  • Enable shared decision-making.
In my next post, I'll introduce and describe the first collaborative health-support software system.

References and Notes:
[1] Data useful to a cardiologist includes the location of chest pain, ST elevation or depression, Q waves or left bundle branch block, T wave inversion or hyperacusis, CKMB and Troponin T or I levels, and heart imaging studies, etc.

Friday, March 06, 2009

Case for a Collaborative Health-Support Software System - Part 1 of 3

In our current economy, improving healthcare value is absolutely essential since the current system is unsustainable (unaffordable) long-term.[1] High-value comes from consistently delivering top quality care at a good price.[2] In other words, we have to focus on making healthcare much more cost-effective[3] by assuring consumers (healthcare patients, clients, customers)[4] receive the sick-care and well-care they need—delivered efficiently, safely and at a good price—without unnecessary tests and procedures, and without wasting money on products and services that provides little benefit to consumers.

Part of the solution is reforming healthcare economic models,[5] changing how care is paid,[6] and redirecting the way healthcare providers compete[7] because current methods fail to offer high value products and services to consumers.

Arguably more important, however, is the need for healthcare professionals and consumers to:

  • Know the most efficient and effective ways to prevent, treat and manage health problems
  • Use this knowledge competently and consistently to improve outcomes and control costs.
Sadly, while drowning in oceans of information, the healthcare industry lacks the knowledge and motivation needed to deliver high-value care.[8] This is evident by:
  • The disconnection between quality and cost; research has proven that more expensive care rarely results in better care
  • The fact that healthcare providers often don't know what treatments work best for a particular patient
  • The unaided human mind, no matter how competent, simply cannot focus on all the necessary details nor possess all the knowledge needed for continually making the best clinical decisions
  • Our country failure to focus adequately on making existing healthcare knowledge useful and applicable to clinical practice.
  • Obtaining the knowledge to improve decision-making requires a commitment to ongoing clinical outcomes research and a focus on continuous quality improvement, things that the healthcare industry has largely avoided
  • Knowledge about prescription medication safety and effectiveness can be greatly enhanced.
How can this troubling situation be improved? By using and evolving collaborative health-support software.

Health-Support Software Described

Let's begin by describing what health-support softwareis, what they do, and how using it in collaborative networks is an essential ingredient for increasing healthcare value.

What Health-Support Software Is and Does

Health-support software consists of computer programs (applications) designed to help people making health-related decisions or taking specific actions to deal with (potential) health problems. Good health-support software accomplishes two basic objectives; they are:
  1. Increasing awareness and understanding about people’s health status, threats, problems, and care options
  2. Guiding subsequent action to avoid the threats and deal with the problems.
Good health-support software accomplishes these objectives by providing personalized recommendations, instructions, feedback, and alerts.

To be truly useful, health-support software should manage (obtain, analyze, exchange, and present) comprehensive personal health information about a person’s:
  • Physical and psychological (body & mind, whole-person)[9] health and susceptibility, past and present
  • History of treatments, the clinical outcomes (the results of care), and the costs
  • Influences of one's environment and culture
  • Personal preferences.
Furthermore, the recommendations, instructions, feedback, and alerts generated by the software should be based on evidence-based guidelines that help people:
  • Assess (diagnose) and define health risks and conditions
  • Prevent risks from becoming serious health problems
  • Treat and cure existing problems
  • Manage chronic conditions (such as diabetes, obesity, and heart disease), so debilitating complications do not happen.
It's important to note that models differ in many ways; for example, different models:
  • Take into account different health data
  • Employ different math and logic to analyze the data
  • Make different assumptions
  • Base their recommendations, instructions, feedback, and alerts on different information resources (such as evidence-based guidelines[10])
  • Present their recommendations and instructions in different ways.
This means that certain health-support software is more useful than others depending on the people using them and the particular circumstances in which they are used. Yet despite these differences, they still share the same purpose: to increase care cost-effectiveness, a goal requiring collaboration.

How Health-Support Software Enables Effective Collaboration

Collaboration is essential for bringing ever-greater value to the healthcare consumer. Increasing healthcare efficiency and effectiveness depends, in part, on loosely-coupled networks of people collaborating to support shared decision-making and facilitating care coordination. This is why collaborative health-support software is so important.

Loosely-Coupled Networks

A "loosely-coupled network" is a dispersed group of people from multiple locations and with different roles, responsibilities, experiences, and awareness. These people collaborate to make more valid decisions and competent actions by taking advantage of their collective intelligence.[11] When such collaboration is among people with wide diversities of knowledge, ideas and points of view, the collective provides a greater assortment of intellectual resources, and offers access to a wider variety of non-redundant information and understandings [12] on which to base decisions and guide actions. This is unlike a tightly-coupled network that limits participation to people within the same discipline, department, region, etc. who have access to the same information sources and who share similar experiences. Loosely-coupled collectives provide the greatest opportunities for stimulating multifaceted discussions and creative solutions.

Evolving Health-Support Software through Collaboration in Loosely-Coupled Networks

Collaboration helps evolve (improve) health-support software by making them more effective in:
  • Increasing people’s awareness and understanding of health risks, problems, and remedies
  • Providing useful recommendations, guidance, feedback, and instruction.
Evolving health-support software in this manner is accomplished when collaborators share the software and scrutinize it by:
  • Comparing and testing its ability to reflect reality accurately
  • Testing it in different scenarios
  • Discussing the assumptions and results it produces.
When they find models that disagree or generate invalid results, they:
  • Examine the fundamental assumptions built into the software’s underlying models[10]
  • Look for logical flaws and inconsistencies in the software’s models
  • Question the program’s authors’ perception of reality
  • Debate about the assumptions and practical value of the software.
By challenging the model's assumptions in these ways:
  • Useful counterintuitive insights often emerge
  • Innovative thought is sparked
  • New questions arise
  • Relationships between collaborators are developed
  • The influence of an organization's culture and politics are revealed
  • Compelling and unexpected management issues are discovered.
This all means that sharing and playing with health-support software is an effective way to evolve the software to increase its validity and usefulness.

Failure to evolve models in this manner can be disastrous! Take, for example, the famous mathematical model[13] that helped bring Wall Street to its knees. The model was used for the past five years to justify risky mortgage-based investments, even though few understood it and even fewer questioned its assumptions. Invented by a mathematician, the model was based on the faulty assumption that home prices would continue to rise and few homeowners would be unable to pay back their mortgages. While some financial experts realized this was a seriously flawed model, it helped bring great wealth to many for years, so there was little incentive to criticize it. This very model is a root cause of the world's current financial meltdown.

The healthcare industry ought to learn a valuable lesson from that fiasco by continually evaluating end evolving the diagnostic and treatment models underlying clinical guidelines and decision-support tools!

Enabling Loosely-Coupled Networks with Decentralized Communication Architectures

Collaborators in loosely-coupled networks would benefit from a low-cost decentralized (peer-to-peer) communications architecture. One example is the “mesh” network architecture,[14] which grew out of the need for a distribution network in telephone, power, and water, and oil pipeline businesses. It enables anyone to communicate directly with anyone else at any time, from anywhere. It’s like using the telephone to communicate with a single person, or to set up a conference call where each person can communicate to many other people. The mesh network architecture, therefore, provides a simple, flexible, low-cost way for loosely-coupled networks to collaborate.

In my next post, I'll discuss how to use health -support software in loosely-coupled networks to provide high-value care by promoting coordinated care, delivering consumer-centered cognitive support, and fostering shared decision-making.

[3] Cost-effective mean economical in terms of tangible benefits produced by money spent (
[4] The term “consumer” is used throughout this paper to refer to people who “consume” (use) healthcare products and services. Depending on the healthcare products and serviced used, a consumer may be a patient, client, customer, member, etc. In contrast, the term “provider” is used to refer to any professional who provides (delivers, renders) healthcare services. They include physicians and non-physician doctors, nurses, therapists, counselors, wellness coaches, physical trainers, health educators, etc.
[12] Software programs have underlying models that guide its operations. The models may include mathematical formulas, logic rules, text functions, formatting instructions, and other procedures (routines).