September 2003

Grant Results

SUMMARY

From 1998 through 2002, researchers at the University of Virginia Health Sciences Center developed and pilot tested a Web-based tool for collecting family health history. Called Health Heritage, it can enhance communication between primary care providers and individuals about their risks for common diseases with genetic components.

Key Results
The Health Heritage Web site guides individuals in entering family history information. Then it applies a set of 89 evidence-based rules, or "algorithms," to this data to assess the individual's risk of developing certain conditions in five disease areas: oncology, cardiology, vascular disease, neurology and endocrinology.

It also provides health care recommendations for each of the conditions, written so both the individual and the physician will be able to understand. Researchers then conducted a pilot test of the Web site and compared its effectiveness to the usual method of patient chart review and the "gold standard" of the genetic counselor interview.

Key Findings

  • Health Heritage outperformed the usual care collection method with regard to the number of relatives identified, the identification of relationships, ages and disease conditions and the completeness of relatives' health histories. Health Heritage also performed favorably in comparison to results obtained through the genetic counselor interview. However, while more accurate in identifying conditions than the usual care method, Health Heritage missed a significant number of conditions identified by the genetic counselor.
  • A majority of participants found completing the family health history on Health Heritage not difficult and found it easy to create a family tree — i.e., to outline the family connections. Obtaining the information needed to complete the health history was more difficult.
  • Most participants were satisfied with the confidentiality and security of Health Heritage.

Researchers reported the project results and the findings of the pilot study in the peer-reviewed Journal of the American Medical Informatics Association.

Funding
The Robert Wood Johnson Foundation (RWJF) provided two grants totaling $2,101,456 to the center for this project.

 See Grant Detail & Contact Information
 Back to the Table of Contents


THE PROBLEM

The sequencing of all human genes by the Human Genome Project has greatly expanded the knowledge of the heritability of common diseases. With an accurate record of family health history in hand, physicians now have the capability to assess an individual's risk of chronic diseases that have genetic components, and to initiate measures of disease prevention and management.

However, even though primary care providers recognize that the collection of family health history is an important part of routine care, it is not often systematically collected, documented or used. Even when patients have a complete family health history available, few primary care providers have the time or expertise to interpret the data.

In September 1997, the federal Task Force on Genetic Testing, convened by the National Institutes of Health, identified the need for increased knowledge among primary care providers and patients regarding the interpretation of genetic vulnerability to disease.

According to their report (see Task Force report), providers and patients must combine knowledge of genetic risk with appropriate behavioral, pharmacological and other approaches to reduce the likelihood and severity of diseases such as cancer, heart disease and hypertension.

However, this requires a level of genetics expertise that many primary care practitioners, who are already overloaded with information, do not possess. Genetic counselors provide a level of knowledge considered the "gold standard" in this regard, but their numbers are insufficient to meet demand. Given the efficiency of the Internet in accessing, storing and sharing information, Web-based tools may offer a way to meet the need for the collection and interpretation of accurate family health history information within the primary care practice.

 Back to the Table of Contents


RWJF STRATEGY

RWJF has made significant investments in enhancing the generalist physician workforce, including efforts to increase both their number and their capacity to provide basic health services. Genetics, on the other hand, is an exploratory area of grantmaking at RWJF. Efforts to date have included support for conferences to explore the implications for health care of advances in human genetics, the establishment of a coalition to educate health professionals in genetics, the development of interactive courseware for physicians and exploring options to improve public and private decision-making about genetics.

The principal investigator for this project, Dr. William Knaus, is a former RWJF National Program Director (Program on the Care of Critically Ill Hospitalized Adults — SUPPORT), with extensive experience in developing computer-based clinical tools.

 Back to the Table of Contents


THE PROJECT

RWJF provided two grants to the University of Virginia Health Sciences Center to support the development and pilot testing of a Web-based family health history collection tool that can enhance communications between primary care providers and patients about their risks for common diseases with genetic components.

Under the first grant (ID# 035277), Dr. Knaus and his project team (consisting of experts in genetically related diseases, Web design and medical informatics, and physician consultants) sought to design a system that would fulfill two basic purposes: (1) assisting individuals in recording and maintaining their family health histories in a secure, confidential manner, and (2) supporting individuals and primary care providers through analysis of health histories, identification of potential risks and provision of health care recommendations.

The development of the family health history collection tool entailed a review of existing health history instruments, interviews and focus groups with patients and with primary care practitioners, surveys of patients about computer use, and observation of interactions between patients and genetic counselors.

Researchers selected 89 diseases and conditions from five basic areas — oncology, cardiology, neurology, endocrinology and vascular disease — and developed algorithms which, when applied to the family history data, would produce a risk assessment and prevention and management recommendations for each. Researchers selected the diseases and conditions based on their prevalence in the population, clinical importance or treatabilty and availability of an evidence-based approach to prevention and management.

Researchers evaluated the Health Heritage Web site through a pilot test that compared results from an intervention group and a "usual care" comparison group, and compared both of these to the "gold standard" of an interview with a genetic counselor.

The Health Heritage intervention group consisted of 159 individuals who used the Web site to record family history. Researchers used the usual method of patient chart review to collect the family histories of the 55 individuals who comprised the usual care group. Participants completed questionnaires prior to recording family histories or chart review.

After researchers had collected that information, participants had a phone interview with a genetic counselor. A genetic specialty team analyzed these data. While all participants in the usual care group completed the pilot study, 54 members of the intervention group completed the study. The total sample of 109 was 70 percent female and 92 percent non-Hispanic white; participants ranged in age from 18 to more over 75. Close to two-thirds (63 percent) had completed college or more.

The addition of grant ID# 040685 allowed the refinement of Health Heritage to make the site applicable to real-world practice settings. This effort included:

  • Making the site easier to use for practitioners and patients.
  • Developing a system to update the risk assessment and recommendation portion of the site regularly and investigating methods for interfacing with medical databases ("mapping capability"). Preliminary work revealed that the complete installation of mapping capability was a complex effort beyond the scope of this project, but a high priority for future work on Health Heritage.
  • Improving the completeness, structure, flexibility and comprehensibility of the health history questions.
  • Improving the administration of Health Heritage by evaluating and revising system security and developing Web usage statistical tools.

The center contracted with Charlottesville, Va.-based Dominion Digital for programming and supervisory maintenance of the Web site.

 Back to the Table of Contents


RESULTS

  • Researchers developed the Health Heritage Web site as a tool to assist in the collection and interpretation of family health history. The site guides individuals in the collection of family health history, and applies a set of 89 algorithms to the collected data to produce a risk assessment and health care prevention and management recommendations for a corresponding set of conditions within five disease areas: oncology, cardiology, neurology, endocrinology and vascular disease. The standard format for the recommendations section of the tool, written so that both the physician and individual patient can understand it, includes:
    1. Health history of family members considered for the recommendations.
    2. A "bottom line" summary statement.
    3. Explanation of likelihood related to this condition.
    4. Additional information about the recommendation.
    5. Frequently asked questions.
    6. Prevention.
    7. Management.
    8. Contributing factors.
    9. Impact on health.
    10. Related Web sites and literature.

Findings

Researchers reported pilot study findings in the July/August 2002 issue of the peer-reviewed Journal of the American Medical Informatics Association and in two additional papers prepared for publication.

  • Compared to the usual care collection method, Health Heritage captured more relatives, was more complete in identifying relationships and ages of relatives and in identifying disease conditions, and provided more complete histories of relatives with conditions. While Health Heritage captured fewer family members than the genetic counselor, it compared favorably with other results obtained through the genetic counselor interview, performing as well or better, depending upon family size or number of conditions.
  • The accuracy of Health Heritage in identifying conditions found by the genetic counselor was greater than that of the usual care method. However, it missed a significant number of conditions identified by the genetic counselor.
  • Over one-half (54.2 percent) of pilot study participants found completing the family health history on the Health Heritage Web site not at all difficult, with only four percent finding it very difficult. However, obtaining the health information needed to complete the family health history was more difficult, with 27.1 percent finding it not at all difficult and 10.4 percent finding it very difficult.
  • Ninety-four percent of participants found it easy to create a family tree — i.e., to outline the family connections — and 70 percent found it easy to update it. Most (70 percent) also agreed or strongly agreed that moving from one area on the Web site to another was easy.
  • Most participants were very (50 percent) or somewhat (40 percent) satisfied with the confidentiality of Health Heritage. Satisfaction with Web site security was even higher, with 48 percent very satisfied and 46 percent somewhat satisfied.
  • Participants spent between one and 120 hours gathering and recording information, with an average of 12.6 hours.

Limitations

The project sample was well educated and of a higher socioeconomic level than the general population. In completing the Health Heritage family health histories, they demonstrated a level of technical understanding that might not be present in a broader population. In addition, the pilot study focused on the experience of those who completed the Health Heritage family health history collection. The project did not address the reasons why two-thirds of those initially enrolled in the study did not complete it.

Communications

Researchers reported results in the peer-reviewed Journal of the American Medical Informatics Association. The team also presented results at multiple meetings and conferences. The Health Heritage Web site is not available for public viewing at this date. See the Bibliography for details.

 Back to the Table of Contents


LESSONS LEARNED

  1. Web site design should involve all those with a stake in the site in order to address critical needs. At the same time, the project team must balance the different agendas of these stakeholders. For this project, researchers gathered needs assessment information through focus groups, personal interviews and surveys. Stakeholders included patients/consumers, primary care physicians, disease specialists and medical geneticists. (Project Director)
  2. Web site design process should allow early user testing while the site is still paper-based, along with the ability to make key changes to the site after it is fully functional. The project team wanted to create an easy-to-use interface that required no separate instructions. This was accomplished by continuous testing at multiple stages in development, from paper prototypes to computer. (Project Director)

 Back to the Table of Contents


AFTER THE GRANT

The project team is seeking funding for projects that would use the Health Heritage Web site. Otherwise, they have no further plans at this time.

 Back to the Table of Contents


GRANT DETAILS & CONTACT INFORMATION

Project

Helping Primary Care Practitioners Assess Genetic Vulnerability to Common Chronic Diseases

Grantee

University of Virginia Health Sciences Center (Charlottesville,  VA)

  • Amount: $ 1,572,602
    Dates: November 1998 to September 2001
    ID#:  035277

  • Amount: $ 528,854
    Dates: February 2001 to May 2002
    ID#:  040685

Contact

William A. Knaus, M.D.
(434) 924-8430
wak4@virginia.edu

 Back to the Table of Contents


BIBLIOGRAPHY

(Current as of date of this report; as provided by grantee organization; not verified by RWJF; items not available from RWJF.)

Articles

Kinzie MB, Cohn WF, Julian MF and Knaus WA. "A User-Centered Model for Web Site Design: Needs Assessment, User Interface Design and Rapid Prototyping." Journal of the American Medical Informatics Association: 9(4): 320–330, 2002. Abstract available online.

Presentations and Testimony

W Knaus, W Cohn, J Barrett, M Kinzie, S Pelletier, M Julian and J Einbinder, "The Development of a Web-Based Family History Collection Tool," poster presentation at the Annual Meeting of the American Society of Human Genetics, October 19–23, 1999, San Francisco. American Journal of Human Genetics, 65(Suppl. 4): A387, 1999.

W Knaus, W Cohn, J Barrett, M Kinzie, S. Pelletier, M Julian and J Einbinder, "Health Heritage: A Web-Based Family Health History Tool," poster presentation at the Annual Meeting of the American Medical Informatics Association, November 6–10, 1999, Washington. Proceedings of the 1999 American Medical Informatics Association Annual Symposium available online.

JS Einbinder, J Barrett, W Cohn and LH Einbinder, "Workshop: How to Develop Web-Based Health Risk Assessment Tools," at the Annual Meeting of the American Medical Informatics Association, November 6–10, 1999, Washington. Proceedings of the 1999 American Medical Informatics Association Annual Symposium available online.

JS Einbinder, W Cohn, J Barrett, S Pelletier, A Wenzel, M Harrison, S Miesfeldt, M Reitmeyer, A Tucker, B Worrall and W Knaus, "A Knowledge Framework for Representing Family Health History," poster presentation at the Annual Meeting of the American Medical Informatics Association, November 4–8, 2000, Los Angeles. Proceedings of the 2000 American Medical Informatics Association Annual Symposium available online.

M Kinzie, W Cohn, M Julian and W Knaus, "Best Design Practices for a Family Health History Web Site: Needs Assessment, Rapid Prototyping and A Focus on the User Interface," poster presentation at the Annual Meeting of the American Medical Informatics Association, November 4–8, 2000, Los Angeles. Proceedings of the 2000 American Medical Informatics Association Annual Symposiumavailable online.

JR Barrett, WD Woolfolk, DP Wagner, AJ Wenzel and WA Knaus, "Securing the HealthHeritage Web, a Family Health History Web Site," poster presentation at the Annual Meeting of the American Medical Informatics Association, November 4–8, 2000, Los Angeles. Proceedings of the 2000 American Medical Informatics Association Annual Symposium available online.

SL Pelletier, BB Worrall, MB Harrison, S. Miesfeldt, AL Tucker, TE Kelly, P Schnatterly, J Einbinder, M Reitmeyer, W Cohn and W Knaus, "The Development of Algorithms to Assess Family Health History in a Web Based Application," poster presentation at the Annual Meeting of the American Society of Human Genetics, October 3–7, 2000, Philadelphia. American Journal of Human Genetics, 67(Suppl. 4): A203, 2001, available online.

WM Novicoff, W Cohn, J Barrett, M Kinzie, M Ropka, MB Harrison, BB Worrall, SL Pelletier, T Kelly and W Knaus, "Assessing Genetic Risk Using Web-Based Family Health History Tool," poster presentation at the American Association for the Advancement of Science Conference, June 11–14, 2000, Ashland, Ore. Proceedings of the 2000 American Association for the Advancement of Science Conference, Pacific Division.

Wendy C, W N, S Pelletier, M Ropka, J Barrett, J Gibson and W Knaus, "Evaluation of a Web-Based Tool to Gather and Assess Family Health History," poster presentation at the Annual Meeting of the American Medical Informatics Association, November 3–7, 2001, Washington. Proceedings of the 2001 American Medical Informatics Association Annual Symposium available online.

W Cohn, S. Pelletier, M Kinzie, J Barrett, A Wenzel and W Knaus, "Health Heritage: A Web-Based Family Health History Tool," poster presentation at the Annual Meeting of the American Medical Informatics Association, November 3–7, 2001, Washington. Proceedings of the 2001 American Medical Informatics Association Annual Symposium available online.

M Kinzie, W Cohn and W Knaus, "The Importance of Heuristic Evaluation and Usability Testing in the User Interface Design for a Family Health History Web Site," poster presentation at the Annual Meeting of the American Medical Informatics Association, November 3–7, 2001, Washington. Proceedings of the 2001 American Medical Informatics Association Annual Symposium available online.

W Cohn, S. Pelletier and W Knaus, "Health Heritage: Web-Based Tool to Gather and Assess Family Health History," (theater demonstration) at the Annual Meeting of the American Medical Informatics Association, November 9–13, 2002, San Antonio. Proceedings of the 2002 American Medical Informatics Association Annual Symposium available online.

 Back to the Table of Contents


Report prepared by: Mary B. Geisz
Reviewed by: Robert Crum
Reviewed by: Marian Bass
Program Officer: Constance M. Pechura

Most Requested