APPENDIX 4. INFORMATIONAL MATERIALS ABOUT GENETIC TESTS

M. K. Cho^, Ph.D.; M. Arruda,^b N. A. Holtzman, M.D., M.P.H.

INTRODUCTION

As genetic testing moves into the mainstream of health care in the United States, patients and providers will have more need for basic test information. We surveyed informational materials available in 1995 and summarized our results here. In general, the materials tended to contain information that would aid in determining a patient’s eligibility for a genetic test, but lacked sufficient information about the tests themselves, such as their sensitivity, specificity or predictive value, the purpose of testing, and information concerning patient rights.

The increasing need for genetic testing information is stimulated by two main factors. First, the number of genetic tests ordered is increasing, in part because tests to predict genetic risk of common conditions such as cardiovascular disease, Alzheimer disease, or breast cancer are now available.

Second, the resultant increasing demand for and volume of tests may not occur with a corresponding increase in availability of genetic counseling for patients or consultations between genetic specialists and nonspecialist health care providers.

Consequently, genetic tests will increasingly be administered by providers without genetic training. The lack of training and of educational materials might already have contributed to providers’ and patients’ misunderstanding of test results and implications. A recent study of colon cancer testing found that physicians who had ordered the test misinterpreted the results nearly a third of the time.

Written educational materials can be used either with, or instead of, other educational forms such as video. Evidence shows that many patients prefer written materials, that they are effective for communicating medical information and that materials such as booklets effectively decrease anxiety about medical procedures. Written materials are especially useful as an adjunct to counseling sessions.

Written informational materials have been specifically recommended for genetic test . Use and interpretation of the tests is complex and testing can lead to social discrimination^,,,, and can have unanticipated psychological impact.^,,,,, Indeed, several policymaking bodies have been concerned about the risks and potential for inappropriate use of genetic tests, so they specified the key types of content that should be provided to patients if they are to make informed choices about testing.^,,,

We surveyed the extent to which existing informational materials on genetic tests contain key elements that are necessary if patients and providers are to evaluate the purpose, accuracy, risks, and benefits of that test. Further, we analyzed whether for-profit and nonprofit organizations differed in the inclusion of particular key elements.

METHODS

From the survey conducted for the Task Force on Genetic Testing (see Appendix 3) in early 1995, we identified 178 organizations, including biotechnology companies, molecular genetics laboratories, and cytogenetic laboratories, who said they had printed or audiovisual information about the genetic testing services or products they offered.

In a telephone survey, we contacted each of the 178 organizations to determine which of them provided printed or audiovisual information about genetic tests to providers, patients, or both. We then determined: (a) for whom the materials were intended, (b) the context in which information was disseminated (e.g., by request only or to providers or patients on mailing lists), (c) whether the information was accompanied by counseling, and (d) who developed the materials. The study was approved by the University of Pennsylvania Committee on Studies Involving Human Beings (Assurance #M1025). Written informed consent forms were waived.

The first set of calls to all 178 organizations was made between October 1 and November 28, 1995. We made up to three follow-up calls and made one written contact with a letter and questionnaire to each organization, continuing through March 1996. We interviewed the person in the organization who would be most knowledgeable about the development and distribution of informational materials for our telephone survey.

From the previous mailed questionnaire, (see Appendix 3) we also had data on whether the organization was for-profit (e.g., biotechnology companies) or nonprofit (e.g., cytogenetics or molecular genetics laboratories at universities).

We requested copies of printed or audiovisual informational materials from all organizations that indicated that they provided such material. We chose material for content analysis if it concerned either diagnostic or predictive DNA-based genetic tests. The selected tests included those to detect hereditary disorders such as Duchenne muscular dystrophy, as well as those to detect chromosome abnormalities in tumors, such as specific translocations. We excluded material if it concerned tests that were not DNA-based (e.g., a test that detected blood lipid levels) or tests not for use on human samples. We also excluded materials that involved only tests used in newborn screening or involved only amniocentesis, chorionic villus sampling, and/or triple tests (for Down syndrome). These materials generally do not describe specific, individual tests, and thus were not comparable to the rest of the content analysis.

Virtually all the material we collected was in the form of printed pamphlets. We analyzed only the English versions, and merely recorded the availability of non-English versions. Some organizations provided general pamphlets (e.g., about the organization, about genetic testing or genetic inheritance generally, or about counseling) along with specific pamphlets about a test. In these cases, we considered the two pamphlets together as the unit of analysis, and will refer to them as "the pamphlet."

Using recommendations by several policy-making bodies about minimal elements of information that patients would need in order to assess a genetic test, we chose 10 elements and determined if they were present in the pamphlets (see Table 1).^,,,, A statement fitting any part of the description of each content element was counted as a presence of the element. For example, any mention of either sensitivity, specificity, predictive value, false positive or false negative rate of a test was counted as a statement on test performance. One member of the research team (M.A.) coded all of the materials, and another member (M.C.) independently analyzed a 20 percent subsample to determine the inter-rater reliability of the coding scheme. The percent agreement between raters for each element ranged from 88 percent to 96 percent, with an average agreement of 93 percent. The corresponding Kappa scores of inter-rater agreement ranged from 0.74 to 0.92. Because of the high level of agreement, we used only the data coded by M.A. for our analyses.

RESULTS

Some of the results of the content analysis were reported at the 1996 annual meeting of the American Society of Human Genetics and a detailed report will be published in the American Journal of Medical Genetics. Those results are summarized briefly:

Ninety-five percent (169/178) of organizations responded to our survey; and 131/169 (78 percent) confirmed using informational materials. We analyzed 115 pamphlets collected from 125/131 (95 percent) of the organizations.

Overall, information about genetic testing was highly variable in content and potentially deficient or misleading. For example, nearly half of the pamphlets included some statement about the accuracy of the test, such as "more than 99 percent accurate" or "will detect all carriers," but most of these did not specify whether their statements referred to sensitivity, specificity, or predictive value.

The elements least frequently included in the pamphlets were risks and benefits, patient rights, and intended use or purpose of the test. Only 6.9 percent (2/29) of pamphlets intended for physicians, 26 percent (10/39) of those intended for patients, and 17 percent (8/47) intended for both included any explanation of risks and benefits of testing. Also, only 3.4 percent (1/29) of those intended for physicians, 26 percent (10/39) of those intended for patients, and 47 percent (22/47) of those intended for both mentioned confidentiality, voluntariness, or the possibility of discrimination in conjunction with genetic testing.

The most frequently included elements were descriptions of the conditions detected by the test (99/115; 86 percent), and the appropriate patients for testing (85/115; 74 percent).

Less than half of the materials included a statement on test performance (48/115; 42 percent). Approximately one in five pamphlets (24/115; 21 percent) included information on test inter-pretation.

In addition, we performed analyses to determine when key elements included in the materials were associated with the type of organization that developed the material (i.e., for-profit or nonprofit). Of the 115 pamphlets, 82 percent (94/115) came from nonprofit organizations, and 18 percent (21/115) came from for-profit organizations. The materials used by nonprofit organizations were more likely to be developed by or in conjunction with an external group such as the March of Dimes or Cystic Fibrosis Foundation (38/94; 40 percent) than the materials used by for-profit organizations (2/21; 9.5 percent) (P=0.02). Whereas most of the pamphlets developed by for-profit organizations were designed for use by both providers and patients (17/21; 81 percent), a substantial proportion of materials from nonprofits was intended for providers only (30/94; 32 percent).

We found that pamphlets from for-profit and nonprofit organizations did not differ significantly in whether they included seven of the ten key content elements listed in Table 1. A substantial proportion of material from both for-profit (11/21, 52 percent) and nonprofit organizations (30/94; 32 percent) consisted of a list of available tests accompanied by pamphlets that could be used in counseling. In many of these cases, the lists included price and shipping information (directed at providers), and the pamphlets tended to have information about diseases, rather than about the genetic tests for the diseases.

Only about one-fifth of the materials from either for-profit or nonprofit organizations had any statements on interpretation of test results.

When interpretation statements were included, they typically referred to the fact that a negative test does not imply no risk of disease. For example:

For three key content elements, there were differences between materials from nonprofit and for-profit organizations: patient rights, the need for genetic counseling, and the purpose of the genetic test. Pamphlets from nonprofit organizations were likely to include information on patient rights, whereas none of the 21 pamphlets from for-profit organizations had such information (P<0.001) (see Table 2). Nonprofits were more likely to include the need for or availability of genetic counseling (P<0.001).

A typical example of a statement on patient rights in pamphlets from nonprofit organizations is, "All testing is voluntary and all information is confidential." A few pamphlets, however, had much lengthier statements including limitations on disclosure and confidentiality of family members:

• Overall, pamphlets tended to contain information that would aid in determining a patient’s eligibility for a genetic test, but did not contain sufficient information about the tests themselves.
• Several critical elements need to be added to most pamphlets to enhance informed choices by patients and providers. Price lists with shipping information and general pamphlets on genetic diseases do not provide adequate information about the tests.

• Information about test performance should specify whether it refers to sensitivity or specificity (relative to a specified gold standard) or predictive value (in a specified population). Deficiencies in information on test interpretation are particularly serious given the high proportion of physicians who misinterpret test results.
• Nonprofit organizations developing informational materials on genetic tests might consider enveloping pamphlets that are specifically aimed at patients, if they do not do so already. Their materials should add clear information about the intended purpose and uses of the relevant tests.
• Materials from for-profit organizations should add information on patients’ rights, such as the voluntary and confidential nature of testing, the need for informed consent, and the need for and availability of genetic counseling. This information needs to be included in materials designed for providers as well as patients.

ACKNOWLEDGMENTS

We thank Stephen Hilgartner for help in developing the telephone survey questions and the criteria for content analysis. This study was supported in part by a grant from the Robert Wood Johnson Foundation (to N. A. H.)

ABOUT THE AUTHOR