(2:00 pm, Friday)

Martha R. Tarrant, BA, MS
University of Virginia

Patricia M. Shannon
University of Virginia

A Blueprint for Home Health Care Support via the Internet and the World Wide Web

ABSTRACT

The presentation will demonstrate how Internet tools, particularly Mosaic, can be used to develop an open access community education, information, and support application. Mosaic offers a short learning curve, ease of use and visual appeal to overcome initial resistance. For information service providers, its management structure allows seamless invisible gopher capabilities to a wide range of disease subject resources, where they can be found, and for developers, relatively simple programming to produce resources, where they cannot be found.

PAPER

Introduction

The cost of healthcare in the United States has grown at a faster rate than the cost of living (Gleicher, 1991) and expenditures from that sector of the economy alone now represent nearly 15% of the gross domestic product (GDP) (Ashby & Greene, 1993). To control future costs and spending, the medical establishment has been forced to adopt treatment policy reforms, such as quicker hospital discharge and its corollary, greater reliance on home care for extended recovery (Glasheen, 1994). Yet, if home care is to be truly effective as a costcontrolling alternative to inpatient care, the same medical establishment that made the reforms must guard against hospital reentry for care receivers or stress illnesses in caregivers by providing a "seamless support system" in the home (Glasheen, 1994). This paper describes how we, as members of the University of Virginia (UVa) Health Sciences community, propose using the Internet and the World Wide Web (Web) via a local public access network for informational and emotional home care outreach. We have identified four primary objectives for the system we envision: (1) to insure network access for all members of the community; (2) to provide training for information providers and users; (3) to identify local home health care needs; and (4) to locate or produce resources that satisfy those needs. All four objectives are now in progress.

I. Network Access for the Community

A. Original Concept

Our plan for a community network was first delineated in June, 1994 in the form of a grant proposal to the Telecommunications and Information Infrastructure Assistance Program (TIIAP) of the U. S. Department of Commerce. The specifics of that plan as written would be to develop an electronic information service officially named the Monticello Area Virtual Village project but popularly known as "Monticello Avenue".

Driven by the proposed TIIAP grant funds and matching commitments from project partners including the University of Virginia, the local city and county governments, Adelphia Cable Communications and Sprint/Centel, Monticello Avenue would provide free Internet/Web access to all citizens of Charlottesville and its seven surrounding counties. A Web graphical viewer (Netscape) would be used to support access from Macintosh 6100 PowerPCs located in library branches, government buildings and schools. Dialup capability and atextual viewer (Lynx) would be builtin for indirect textbased VT100 access elsewhere. Adelphia and Sprint/Centel would conduct trials of a highbandwidth metropolitan area network prototype that would ultimately evolve into a direct home access system provided at low cost to users. Adelphia would initially test twoway 0.5 Mbps access using a portion of their existing fiberoptic and coaxial cable plant; Sprint/Centel would develop a 10Mbps Ethernet service with further construction of their fiber ring around Charlottesville. The plan's implementation would take place in three phases over eighteen months. In each phase, project partners expected to accomplish the following:

Phase 1: From summer 1994 through spring, 1995, (1) an RS6000 server would be donated by UVa and moved to the main branch of the JeffersonMadison Regional Library system, which would serve as the hub of the network. Project partners would acquire additional equipment (including direct fiber lines) to establish the designated public access facilities, while detailed planning for dialup capability would begin. Adelphia and Sprint would set up their first highspeed connections for the direct home access trial. (2) Community workgroups would be established to explore issues such as a permanent management structure, the appropriate use and social implications of networking, the extent of and boundaries to commercial activity, security and the potential liability of misinformation, denial of access, etc. (3) Public or private agencies and community groups (e.g., churches, scout troops, etc.) would apply for space as information providers on the service and volunteers would be recruited to work with them. Project partners and volunteers together would design and develop a Web "frontend" as well as train the information providers in how to update its contents. A readonly pilot version of the service would be tested online. There would be no individual accounts on the pilot service except those of the design and development team and the volunteer workers; however, a plan for the management of future individual accounts would be written and staff trainers would be hired to help future account recipients get acclimated to the service.

Phase 2: Phase 2 would take place in the summer and fall of 1995. (1) A limited number of individual accounts would be given to citizens to test the accounts management procedure. More accounts would be added as the ability to manage them wasverified. (2) The commercial partners would expand their test of highspeed access into the homes. (3) A commercial activity plan would be written, including several models of operation and fee structures to be tested. Tests of those prototype models would begin. (4) Project partners would work to develop news and discussion groups that would be of particular interest to citizens in the target area.

Phase 3: Finally, in the spring and summer of 1996, project partners would conduct pilot tests of electronic mail between individual citizens and governmental or community agencies. The number of individual accounts would increase as Adelphia and Sprint expand the availability of connections into the home.

B. Modifications and Current Strategies

As can be expected of all plans, changes have been made to the original. The most outstanding difference between the blueprint and reality has been in funding. Results of the Department of Commerce grant process were released in October and Monticello Avenue was not one of its grant recipients, presumably because the considerable assets of the project partners conveyed an impression of insufficient need. Despite the loss of half its proposed budget, however, the community network project is moving ahead  albeit with a tightened focus and longer timeline. The decision to continue was made in light of several recent developments. (1) The City Board of Supervisors and UVa have extended their financial commitments so that, between money spent and promised, the effect of losing the TIIAP grant has been somewhat minimized. (2) Adelphia and Sprint/Centel have stepped up their plans for fiber installation in and around the city. (3) The Albemarle County School Board has approved an incremental budget item of $180,000 over two years to install a highspeed fiber line of its own that would link the 23 county schools to each other and the Internet. Although not intended as part of the Monticello Avenue initiative, this move was seen as indirectly benefiting the community network if the school system's administration would agree to open up any or all computer labs to the public after hours. A second difference of note has been in the level of volunteer involvement which was significantly curtailed as the result of destructive behavior exhibited in personal access accounts. In June of 1994, 150 volunteers were randomly recruited from the community and given user identification codes on "Hopper", an RS6000 server loaned to the project by UVa. Dedicated newsgroups were set up for communications between volunteers and training was provided for those who felt they needed help in getting on or around the system. A clash of cultures and personalities quickly became apparent in the public newsgroups, mirroring similar problems in private email exchanges. Abusive and vulgar language was used; unsubstantiated accusations and threats were made; First Amendments rights were called into question. A chilling effect on posting resulted to the point that almost no activity took place in any of the newsgroups and all no cost, full service accounts had to be terminated after 6 months. As in the case of funding, we have been able to convert the negative experience with volunteer accounts to a positive by revising the strategy for development of "front end" menus and adaptation of information. (1) Whereas the original plan envisioned a design and development team that included selfselected volunteers, the individuals that now make up that team have been specifically invited to participate or were assigned to it by UVa and the information providers represented in the prototype. A steering committee that draws from all volunteer cultures, yet is committed to working together productively despite their differences, was established within the team to set up the topic hierarchy, metaphors and control structure of the prototype. (2) The decision was made to do noncontroversial topics first (e.g., government, library, schools, etc.) and to delay commercial activity until after the prototype is introduced to the public. Established politicallyneutral organizations and agencies have been asked to accept ownership of major topic subdirectories for the initial design (e.g., United Way for "Community", the Piedmont Council for the Arts for "Arts", etc.). Each organization or agency that has agreed to participate has appointed its own Monticello Avenue representative(s) to the development team. Representatives have been appointed on the basis of their authority to speak for the organization and their level of computer knowledge, as they will ultimately be responsible for content monitoring and updates. (3) Weekly meetings with representatives have been set up to assess progress and discuss any problems they may have. Highlights of those meetings are converted to HTML and added to a "Developers' Sketchbook" that is maintained on the Web. Reviews of other community networks, style suggestions and standards, graphic design possibilities, etc. are also included in the Sketchbook as points of discussion among team members.

II. Training for Information Providers and Users

While the network is in development, three streams of training are taking place. One is aimed at current Monticello Avenue community representatives; the second, at nonspecific future groups and users; the third, at potential information providers for home health care support.

A. Training of Current Community Representatives

One Instructional Technology (IT) graduate student from UVa has been appointed to each Monticello Avenue representative to teach the multimedia Web production process oneonone. The students themselves are enrolled in a twosemester sequence entitled "Courseware Tools and Design", which consists of approximately one hour's instruction, one hour's supervised lab time for exercises and one hour's meeting time with their assigned representatives per week. Their instruction (which they pass on to the representatives) covers software program options for creating Web documents on both Macintosh & Windows platforms (such as HTML SuperText for Macs versus HTML Assistant for Windows or Fetch for Macs versus Rapid Filer for Windows); the hardware and software requirements for networking (such as how to set up a SLIP connection); and necessary Unix language commands. Student competencies are demonstrated through implementation of two class projects: the prototype Web interface for Monticello Avenue and Anthology, the statewide public school network which will be linked into Monticello Avenue via its "Education" option.

B. Training of NonSpecific Future Groups and Users

As the network expands, groups that have not as yet been targeted will want to participate. Because we cannot anticipate which specific groups they will be and there is a finite number of Instructional Technology students to serve their needs, part of the original plan called for training Monticello Avenue volunteers who would themselves teach broader array of users. As in all predominantly volunteer organizations, there was a wide range of skill levels among the forty who signed up to become trainers. Orientation classes were consequently devised to demonstrate how and what to teach as well as to create a common minimum proficiency in Internet skills among them. After orientation, participants were asked to conduct workshops or miniclasses of their own, demonstrating for other volunteers the resources and skills they use most frequently. This was intended, from our perspective, as a means to evaluate their ability to teach, their presentation styles and their Internet proficiency; from their perspective, it was meant as an opportunity to aquaint themselves with stateoftheart instructional equipment and the dynamics of adult education. In view of the illwill that grew from the volunteers differences with each other over the newsgroups, it is not known how many of the original trainers could be rerecruited. Therefore, part of the current strategy provides that (1) the inplace design and development team will evolve into an advisory council. This assumes that they too will agree to continue with the project. If they choose not to continue, a contingency plan is being implemented. Each representative and his or her IT student have been asked to document their experience, so that the next representative(s) can benefit from the lessons they learned. A "Curator's Handbook" for every organization and one combining all organizations will be produced from that documentation. A Frequently Asked Questions (FAQ) subset of the Developers' Sketchbook will be culled from the generic version of the Handbook and put on the Web as well. (2) Another strategy is to provide training for the JeffersonMadison Regional Library's reference staff members who will be the immediate contact points for the public using the system. Librarians  particularly those working alone at nights and on the weekends  are now being instructed in gophers, the Web, and all available options within Netscape. They are being shown how to locate and navigate within a range of general interest sites and how to search for others. Probably most important of all objectives, however, is that they develop a familiarity with the PowerPcs and understand how to troubleshoot in the event of a system crash or malfunction. The systems administrator, who is located at the main library branch, is slated to provide oneonone troubleshooting instruction in the spring.

C. Training of Potential Home Health Care Information Providers

Many of those who have now or soon will have subject matter expertise for a home health care support system on the Internet are being formally trained in how to use the medium. Since January 1994, handson Internet classes (including an introduction to the Web) have been offered at least bimonthly to medical and nursing students, medical and nursing faculty, and staff of the University's Health Sciences Center. Beyond teaching basic skills, however, our mission as their trainers is to facilitate the use of those skills in whatever context possible. We have, therefore, developed lines of activity around the advantages of Internet participation, particularly for medical students and medical faculty. Medical Students. As part of a belief that medical careers are becoming increasingly more dependent upon computing and telecommunications, UVa's School of Medicine and the Computers in Medical Education Committee have instituted several changes over the last two years that effect medical students.

(1) Internet training has been incorporated into the fall firstyear orientation program. (2) Grades are now being distributed through electronic mail. (3) Newsgroups have been set up for specific first and secondyear courses. Faculty monitor their own newsgroups, posting open discussion questions, practice test items and/or assignments as they see a need. (4) An online tutoring service is provided through another dedicated newsgroup, "medical student Q&A", to which firstyear students are automatically subscribed. Fourthyear medical students as well as fourth and fifth year MD/PhDs are paid by the Dean's Office to monitor and respond to any posted academic question. (5) Professors have incorporated Web study slide sets into the curriculum with the help of the Office of Medical Education. Students can test their recognition of images on any of three platforms running a graphical browser from any of three groundswide locations: the Health Sciences Library (which supports Macintosh and Windows), the Academic Computing/Health Sciences lab (which supports Unix) or the Thornton Engineering School Stacks (which supports Windows). Given a high baud rate modem and SLIP connection or ARA software, students may also be able to dial in from home. Medical Faculty. Just as teaching faculty are developing Web projects for students, clinical faculty are developing sites for housestaff and continuing education credit. Departments are actively using Web pages as worktools, setting them up as miniservers to other medical sites in their specialty areas. Some have released residents to create specialized courses on the Web  courses that would not otherwise be provided but that relate to recent cases. With the help of IT students, others have undertaken to develop community medical education materials aimed at school children. (It is anticipated that these materials will ultimately be delivered to the schools through Monticello Avenue).

III. Identification of Community Home Health Care Needs

A. Equipment Needs

Home health care presents its own set of specialized equipment needs for the disabled and the homebound  needs that were not part of the original Monticello Avenue plan beyond experimentation with direct home connections through the cable and telephone companies. Once we have secured a service provider with a current patient base from which to assess specific requirements, we intend to pursue equipment grants to cover those needs, computer science students to install whatever equipment we receive and outreach trainers to teach the patrons in their own homes. We are currently producing a videotape that describes the installation/configuration of a SLIP connection and we plan to draft a hardcopy reference manual that describes how to use Netscape to find resources. As we are still in the process of developing the backbone for the network, however, this part of the objective has not been fully addressed. Our primary focus has been identification of information needs for the frontend menus.

B. Information Needs

A study of resource utilization for home health care conducted by researchers from UVa's School of Medicine in 1990 formed the basis of our preliminary needs assessment. Data collected on a random sample of enrollees with home health care agencies of the Virginia Department of Health between October 19821983 and July 19841985 revealed that (1) twothirds of enrollees were in home care directly after hospital discharge, (2) twothirds were over the age of 65, and (3) the primary diagnoses of most fell into the categories of stroke, bone fractures, cancer, diabetes, or urinary incontinence (Williams, Phillips, Torner & Irvine, 1990).

Stroke, some fractures (particularly, hip) and incontinence are ageassociated and, while cancer or diabetes may be seen in all age groups, their frequency of diagnosis runs high among the elderly (Ettinger, 1991; Brown & Jackson, 1994). The general conclusion for our system, therefore, has to be that information resources should strongly but not exclusively relate to care of the elderly. Three limitations of the preliminary needs assessment study are important for our resource planning: (1) results did not include private or hospitalbased home care enrollees; (2) those with chronic functional impairments were omitted; and (3) data are now dated by one decade. Although we do not suspect a significant difference in our general conclusion as a result of data collected only from public agency enrollees, private and hospitalbased home care flourishes in this area and to the extent that their data might differ, gaps in our target resources would need to be identified. We do suspect that the incidence and prevalence of specific ageassociated diseases might change with the inclusion of chronic functional impairments because frequency data typically vary between localities and there are highly visible organizations in Charlottesville and Albemarle County built around chronic functional impairments like Alzheimer's disease. Also, given the growth rate of Acquired Immune Deficiency Syndrome (AIDS) which was first identified in the early 1980s when study data were collected, it is likely that local demographics of home care will have changed, such that not only care of the elderly but also care of infectious diseases should be part of our system.

To test for these possible differences, we plan to undertake a second, more current needs assessment using the same methodology but data of local home health care groups (public, private and hospitalbased). Yet, to accomplish this task requires more manpower than is currently available, given the drive for development of the network and frontend menus. While still in Phase 1, therefore, we are dependent for change data on the impressions and expertise of local service providers. The three service providers we have targeted initially are the Albemarle County Health Department (for public information), Martha Jefferson HomeCare (for private hospital based data), and UVa's Continuum (for public hospitalbased data).

IV. Location or Production of Needed Information Resources

Just as local home health care needs must be identified, resources to meet those needs must be found or created. To be functional and helpful to the greatest number of citizens, resources  where they exist  should appeal to the lowest common denominator reader and novice computer user. They should be viewed equally well with or without images for both graphical browser and textbased applications. They should have a specific component for care of patients and/or home care guidelines. Ideally, they should also have an outlet for emotional support in as much as caregiving increases both perceived subjective and objective strain (Lalonde & Kasprzyk, 1993); for legal support to insure that caregivers understand their rights and responsibilities; and for financial support so that they can afford to care for their loved ones no matter what the duration of the care receiver's illness. To find such sites, we are currently soliciting URLs of applicable Web resources  with associated gophers where possible  from readers of newsgroups in diseasespecific specialty medicine, family law, tax law and business. Where resources do not exist, cannot be found or do not satisfy other criteria, our design and development team, the HealthLink community representatives and/or the IT student assigned to HealthLink will be responsible for producing them.

To illustrate structure and contents, we are in the process of producing a prototype homecare homepage (with disclaimer) which we will use to present Monticello Avenue and its potential to the three targeted service providers. The prototype will consist of the following subtopics, hyperlinks and global Usenet newsgroups that we have thus far identified:

1. General Information for Caregivers

  • Local Service Providers
  • Medical Services (nursing aides, intravenous and physical therapy)
  • Continuum
  • Martha Jefferson HomeCare
  • Hospice of the Piedmont
  • Care Management
  • InHome Attendants
  • Day and Respite Care
  • Local Hotlines
  • Equipment Suppliers
  • Emotional Support and Counseling
  • Coping with Terminal Illness and Death

    (http://asa.ugl.lib.umich.edu/chdocs/support/emotion.html)

    alt.support.depression

  • Legal Issues of Home Health Caregiving
  • Durable Medical Power of Attorney
  • Living Will

    Financial Issues of Home Health Caregiving

  • Budgeting for LongTerm Illness
  • Taxes (What You Can and Cannot Deduct)
  • Responsibilities and Liabilities of InHome

    Attendants

    2. Topics Sorted by Disease and Disability

  • Alzheimer's  UC Irvine's information about dementia

    Counseling chronic illness section on Alzheimer's

    - Cancer -- Oncolink (http://cancer.med.upenn.edu/)

  • Cancernet (gopher://odie.niaid.nih.gov/11/clin/cancernet) combined with Physician Data Query

    (gopher://odie.niaid.nih.gov/11/clin/cancernet/pdqinfo/pif)

    (gopher://una.hh.lib.umich.edu/00ubetdursstacjs/cancer%3agourhin)

  • Usenet newsgroup: alt.support.cancer

  • AIDS - US Dept. of Health & Human Svcs
    (http://www.os.dhhs.gov)

    (gopher://odie.niaid.nih.gov/11/aids/sri)

  • Usenet newsgroups:
    clari.tw.health.aids, sci.med.aids

    Phase 2 should allow us to create limitedaccount (i.e., reasonably secure) local newsgroups after the model of the Cleveland Freenet Alzheimer's Forum where caregivers post their frustrations, share ideas about the care they are giving and get free advice from professionals [freenet.hosp.alzheimer.forum]. Other proposed local newsgroups might be devoted to homebound job opportunities in the area or interactive telecourses for home study. In the third and final phase, it is hoped that an electronic mail system will emerge that would enable direct correspondence with volunteer physicians or home health nurses and participation in broaderbased discussion groups that function through electronic mailings [e.g., MAJORDOMO@wudios.wustl.edu for Alzheimer's].

    The home health care system we have described is a response to two dynamic forces in our society. One is the growth in costs for inpatient hospital care and its consequent "up and out" policies; the other, the promise of information technologies to make those policies succeed. In developing this blueprint, we have followed the vision of Dr. Don Detmer, Vice President for Health Sciences at UVa, when he said that:

    The challenge [of information technologies in healthcare] is not simply to transfer to computers all the tasks we currently accomplish on paper. Rather, taking advantage of the capabilities of these technologies, we should rethink all of our goals and objectives with an eye out for costeffectiveness; out of this will come a fundamental reinvention of the work we do. (Detmer,1994, p.2)

    References

    Ashby, J. L. & Greene, T. F. (1993). Implications of a global budget for facilitybased health spending. Inquiry, 30, 362271.

    Brown, D. F. & Jackson, T. W. (1994). Diabetes: 'tight control' in a comprehensive treatment plan. Geriatrics, 49(6), 2429, 3336.

    Detmer, D. E. (1994). 'Let's write our own script': An interview with Dr. Don E. Detmer, Vice President and Provost for Health Sciences, on informatics at the University of Virginia. Inside Information, 1(1), 12.

    Ettinger, W. H. (1991). Forces of change in the health care system: implications for cancer care in the 1990s. Cancer, 67, 17281731.

    Glasheen, L. K. (1994). Homeward bound. AARP Bulletin, 35(8), 7& 11.

    Gleicher, N. (1991). Expansion of health care to the uninsured and underinsured has to be cost neutral. JAMA, 265(18), 23882390.

    Lalonde, B. & Kasprzyk, D. (1993). Correlates of caregiver strain in home health care. Home Health Services Quarterly, 14(2/3), 97110.

    Williams, B. C., Phillips, E. K., Torner, J. C., & Irvine, A. A. (1990). Predicting utilization of home health resources: Important data from routinely collected information. Medical Care, 28(5), 37991.

    [Previous