Family history represents the unique genomic, ecologic, and gene-environment interactions that affect the metabolic profile and life course of a family and its members. It is well known that a strong family history of coronary heart disease (CHD) is a significant predictor of an individual's risk of CHD even after adjusting for an individual's own established risk factors such as hypertension, smoking, and abnormal lipoprotein levels. The explanation for this familial disease aggregation is not well understood except for the general knowledge that genetic and environmental factors predisposing to CHD also aggregate in families. Given the multifactorial nature of an individual's risk, it can be argued that an individual's familial risk of disease may, in fact, be a better indicator of the many complex interactions among genetic and environmental factors that predispose to disease than can be captured by an individual's own risk factors. Issues of how to assess, quantitate, and apply family history information in clinical settings still need to be resolved. Some clinical risk indicators such as the NCEP III guidelines take into account family history while others, such as the Framingham Risk Score, do not. Moreover, several family-centered intervention studies have demonstrated the particular advantages of focusing on families rather than just individuals. Although there has been tremendous progress in primary prevention of CHD over the last 20 years, substantial advancements may still be achieved by focusing on the family as its own unit of inference and as a specific target for disease prevention.

Epidemiologic studies consistently observe an approximate two-fold increased risk of colorectal cancer in persons having one or more first-degree relatives with the disease (Potter, 1993). Studies of extended pedigrees in Utah (Cannon-Albright 1994) found that the excess familiality was 1.3 for colon cancer, approximately the same as for breast and the other common epithelial cancers. Twin studies (Lichenstein, 2000) suggest a heritability of 35% (95% CI 10%-48%) for colon cancer, within the range observed for breast and prostate cancer. Colon cancer often occurs in aggregation with other cancers, including ovarian cancer and breast cancer.

The aggregation of colon cancer in families may be due one or more causal explanations (segregation of major genes, aggregation of minor genes, aggregation of environmental factors, and interactions between minor genes and environmental factors), as well as non-causal explanations:(chance and bias).

Causal explanations.

Major genes. The identification of familial syndromes that include colon cancer was initially reported by Lynch and co-workers in the 1970s. Several genes have been identified that account for these syndromes: the list includes APC as locus of susceptibility for Familial Adenomatous Polyposis Coli, and a group of mismatch DNA repair genes (including hMSH2, hMLH1) as loci of susceptibility for Hereditary Non-Polyposis Coli. Taken together, highly penetrant mutations in these genes probably account for at most 15% of colon cancer cases in the general population. The contribution of more common, less-highly penetrant alleles in these and other major genes have not been fully investigated.

Minor genes. Common, low-penetrance polymorphisms in a variety of metabolism genes have been associated with increased risk of colon cancer, including N-actyltransferases 1 and 2 (NAT1, NAT2) and methylenetetrahydrofolate reductase (MTHFR). Polymorphisms in DNA repair genes and other loci are the subject of continued investigation. Many of these low penetrant alleles appear to exhibit greater than additive joint effects with diet and other environmental risk factors for colon cancer.

Non-causal explanations

Some of the aggregation of colon cancer in families could be due to chance or bias. Strict criteria for defining family history (e.g. Amsterdam criteria for HNPCC) have fairly high sensitivity and specificity for identifying high-risk kindreds. However, the informativeness of family history within population-based studies has not been well studied. Aitken (1995) reported that the sensitivity and specificity of reported family history of colon cancer are less than 100 percent, and misclassification is usually differential, but the degree of misclassification is "unlikely to inflate estimates sufficiently materially alter conclusions from case-control studies."

Research priorities:

Pharoah et al. (2002) recently showed that risk of breast cancer in women who do not carry mutations in BRCA1 or BRCA2 is mostly likely attributable to multiplicative interactions among minor genes. Such a polygenic model may also apply to colon cancer. Potentially, there may be groups of common polymorphisms that once identified, can differentiate high-risk from low-risk persons in the general population. Combinations of genotypes and traditional risk factors (such as diet) may have practical value for risk evaluation and designing interventions to prevent colon cancer in the general population. In the absence of comprehensive knowledge of these genes and environmental factors, family history may serve as a useful proxy for identifying high-risk individuals. However, studies that determine the extent to which family history of colon cancer can be attributed to currently identified genes and environmental factors are needed, as well as additional studies to examine the contribution of misclassification and other sources of bias.

Can Family History of Type 2 Diabetes be used as a tool for Public Health?
Karen Edwards, PhD, University of Washington
RCM Wines, T. Harrison, LA Hindorff, H. Kim

Given the substantial morbidity and mortality associated with Type 2 diabetes it is important that public health seek ways to delay or prevent the onset of this condition. Risk factors for Type 2 diabetes are well established, including underlying genetic susceptibility. Despite this knowledge as well as significant advances in understanding the human genome, the prevalence of Type 2 diabetes continues to rise at an alarming rate. Because Type 2 diabetes is a complex condition and likely involves a combination of genetic and environmental factors, DNA testing for this condition is not yet warranted. However, because family history reflects underlying genetic susceptibility, in addition to other factors, this may be a useful public health tool for disease prevention. There are several important issues that need to be considered when evaluating family history as a public health tool. These issues, including the accuracy of recall by family members, analytic and clinical validity, clinical utility, ethical, legal and social issues, as well as potential unintended consequences will be outlined and discussed. Areas for future research based on current limitations in knowledge will be identified.

Epidemiological studies document parental or sibling history of asthma as a risk factor for asthma. However, the strength of the association and its correlation with specific clinical outcomes has varied. Differences in populations studied, methods for obtaining or defining a positive family history, and documentation of asthma or related clinical endpoints may account for some variation in the estimated effect of family history. In addition, little is known about interactions between family history and known environmental risk factors. Further understanding of the relationship between family history of asthma and asthma outcomes could be of value to public health efforts in asthma prevention. Family history represents a potential tool to identify children at risk, in order to provide targeted prevention efforts. In addition, selection of patients at increased risk via family history could improve the power of clinical and epidemiologic studies. However, the use of family history information for asthma prevention requires a careful evaluation of (1) family history in the context of other known risk factors, including genetic polymorphisms associated with asthma risk and environmental contributors to the disease; (2) the efficacy of intervention programs based on family history; and (3) the meaning applied to family history risk information by patients and health care providers. Family history of asthma is likely to be most useful as a prevention tool if it can be shown either to identify a subset of patients at high risk who would benefit from intensive prevention efforts or to motivate behavioral change to decrease environmental risk.

Documenting the Family Health History: An Overview of Available Tools
Kristin Peterson Oehlke, MS, CGC, ATPM-CDA
Office of Genomics and Disease Prevention, CDC

Family health history information is collected and used by medical professionals, families and others for a variety of reasons such as medical risk assessment, preventive screening, adoption and other social situations, research and genealogy. Family history tools are available in multiple formats, each tailored to collect the type and complexity of information required. The available tools may provide a foundation for developing family history tools suitable for use in public health practice. This presentation will provide an overview of family history tools that are currently available, their relevant features and their intended applications.

Family History as a Tool for Risk Assessment in Public Health Research and Practice and in Preventive Medicine

Hoda Anton-Culver, PhD, University of California, Irvine

Background. While family history information on cancer is collected in clinical and research settings and is used to infer risk of the disease in population-based, case-control, cohort, or family-based studies, little information is available on the accuracy of a proband's report. In this study we sought to validate the reporting of family history of cancer by cancer-affected probands in population-based and clinic-based family registries of breast, ovarian, and colorectal cancer.

Methods. To assess the accuracy of reported family history of cancer in the relatives of probands, we compared the family history from the personal interview of the proband to a gold standard that included pathology reports or self-reports from the relative or death certificates on deceased relatives. Our study included 670 breast cancer families, 123 ovarian cancer families, and 318 colorectal cancer families that accounted for 3222 relatives of probands. To account for the within-family correlations on the responses, we used a generalized estimating equation (GEE) approach.

Results: Estimates of sensitivity varied across cancer sites and by degree of relation to the proband. In particular, sensitivity of the proband's personal interview among first-degree relatives was 95.4% (95%CI 92-6-98.3) for female breast cancer, 83.3% (95%CI 72.8-93.8) for ovarian cancer, 89.7% (95%CI 85.4-94.0) for colorectal cancer and 79.3% (95%CI 70.0-88.6) for prostate cancer. Sensitivity of the proband's personal interview among second-degree relatives fell to 82.4% (95%CI 76-6-88.2) for female breast cancer, 44.1% (95%CI 30.9-57.3) for ovarian cancer, 57.8% (95%CI 49.5-66.0) for colorectal cancer and 66.7 (95%CI 55.1-78.2) for prostate cancer. In addition to the relative's degree of relationship to the proband and age at diagnosis of the proband, the best predictor of reporting accuracy was the mode of ascertainment of the proband, with clinic-based ascertained probands more accurate compared to population-based ascertained probands.

Conclusion: We found high reliability for most cancer sites among first-degree relatives and moderate for second- and third-degree relatives. Over-reporting of cancer was rare (2.4%). Race/ethnicity and sex of the proband did not influence the accuracy of reporting. However, degree of relationship to the proband, age at diagnosis of the proband, and source of ascertainment of probands were statistically significant predictors on accuracy of reporting.

Family History Collection and Pedigree Analysis
Maren T. Scheuner, MD, MPH, Cedars-Sinai Medical Center

The systematic collection of family history information currently appears to be the most appropriate approach for identification of individuals with a genetic susceptibility to most common diseases. A positive family history is common among many chronic conditions and it is quantitatively significant (King et al., 1992). For many common diseases it is one of the greatest risk factors with relative risks ranging from 2 to 10 times those of the general population as seen in Table 1. Furthermore, for most common diseases, an even greater increase in relative risk is associated with an increasing number of affected relatives, as well as earlier ages of disease onset. Qualitative characterization of risk can also be accomplished by reviewing the family history. Characteristics of a high-risk family history include multifocal or bilateral disease, higher rates of disease recurrence, and the occurrence of related diagnoses. Recognition of these quantitative and qualitative familial disease risks has important implications for identifying family members at risk, arriving at the appropriate diagnosis, and for recommending appropriate genetic tests as well as individualized management and prevention for those individuals.

Accuracy of family history information has been investigated for coronary heart disease, diabetes, hypertension and several cancers. In assessing the reliability of reported family histories of myocardial infarction, Kee et al. (1993) performed a case-control study in which reported histories of first-degree relatives were validated using death certificates, physician records, and hospital records. In the 174 cases, the sensitivity, positive predictive value, and specificity of a reported history of infarction in first-degree relatives were 67.3%, 70.5%, and 96.5%, respectively. These values did not differ significantly from the corresponding figures for the 175 controls (68.5%, 73.8%, and 97.7%, respectively). In this study, only small differences were observed between odds ratios based on reported and verified data, indicating that neither misclassification nor recall bias had a substantial impact on the measurement of the effect of the family history. In a study assessing the accuracy of family history reports of diabetes, Hispanic and non-Hispanic white patients and controls were interviewed at clinic visits (Kahn et al, 1990). Verification of these reports was obtained by subsequently interviewing family members. There was complete agreement between the information given by the proband regarding diabetic status and answers given by the respective family members. The Family Heart Study also characterized the validity of reports of coronary artery disease, diabetes and hypertension by probands (Bensen et al., 1999). Using a relative's self report as a standard, the sensitivity of the proband's report for coronary artery disease was 85% and 81% for parents and siblings, respectively. The report for hypertension was 76% and 56% for these relatives, and for diabetes it was 87% and 72%, respectively. Most specificity values were above 90%.

Love and co-workers (1985) have studied the accuracy of patient reports of a family history of cancer. Verification of cancer histories was done by reviewing pathology and operative reports, hospital admission and discharge summaries, death certificates, and autopsy reports. Verification of negative histories was not performed. The accuracy of cancer site identification by the participant was 83.7% in first degree, 71.3% in second degree, and 71% in third degree relatives. Participants were correct in 91% and 89% of the cases for all relatives in which they reported breast and colon as the primary sites, respectively. For first-degree relatives, 94% of reported breast cancer cases and 93% of colon cancer cases were confirmed. For pancreas, lung, and liver as the primary cancer site, the accuracy rates of cancer reports were less at 75%, 60%, and 5%, respectively.

Overall, these studies suggest that a positive family history report can generally be used with a high degree of confidence for the identification of individuals who may be at increased risk for developing disease. The lower sensitivity values do indicate some under-reporting of disease in relatives; thus, a negative report should not be used as an indicator of a minimum or decreased disease risk (below the general population risk).

Estimates for the frequency of family history reports among cases with several common diseases are available in the literature (Table 1). However, knowledge of the prevalence of family history reports of common diseases among individuals within the population at large, is necessary for shaping policy regarding genetic risk assessment for these conditions. The population frequency of average, moderate, or high risk family history reports for heart disease, stroke, hypertension, diabetes, and colon, breast, ovarian, endometrial and prostate cancers have been determined by reviewing pedigrees obtained by genetic counselors in a prenatal diagnosis clinic (Scheuner et al., 1997a). The population consisted of 400 "healthy,” employed, middle-class people age 18 to 66 years. None of the consultants were seeking counseling because of a family history of one the chronic disorders under study. Forty-three percent (170) reported a family history of at least one of the selected common disorders (130 individuals were at risk for one disorder, 33 were at risk for two, and 7 were at risk for 3). Depending on the specific disorder, most consultands had an average risk (general population risk) for any given disorder, approximately 5 to 15% were at moderate risk (2 to 3 times the population risk), and 1 to 10% were at high risk (risks approaching those associated with Mendelian disorders). Family history of cardiovascular diseases, including coronary artery disease, stroke and syndrome X, were most common and reported by 34% of consultands. Family history of cancer was reported by 8% of the consultands and several individuals had histories suggestive of dominant cancer conditions.

The genetic evaluation or risk assessment for common diseases typically begins with pedigree analysis. The pedigree structure is created and information for each family member is collected. Demographic information is documented including each relative's name, current age or age at death, locale, and ethnicity. Medical history is documented for each family member including age at diagnosis and known interventions or procedures. Information is also collected regarding important risk factors for a disease, such as smoking for heart disease or emphysema. Validation of the medical history of each family member is performed by reviewing records when possible.

After collecting the family history a qualitative assessment of disease risk can be performed which might include a differential diagnosis for identified familial traits. For example, when considering an inherited form of breast cancer there are at least six different genetic syndromes that can feature breast cancer, including site-specific breast cancer, breast-ovary syndrome, Li-Fraumeni syndrome, Cowden syndrome, Peutz-Jeghers syndrome and HNPCC (Hoskins et al., 1995). The types of cancers and other conditions reported in the family distinguish each of these syndromes. Mutations in different genes may underlie the genetic susceptibility in these syndromes and genetic testing can help to confirm a suspected diagnosis. Thus, estimation of the probability that genetic testing will provide information that will change the risk assessment and management and prevention of disease should also be discussed as part of the risk assessment and diagnosis process.

Quantitative risk assessment can also be performed for specific conditions relating to the family history using mathematical models or published estimates (Gail et al., 1989; Amos et al., 1992; Claus et al., 1991; Claus et al., 1993; Benichou, 1993; Schildkraut et al., 1989). For example, in the case of breast cancer a woman's empiric risk based on the family history of breast and/or ovarian cancer can be provided and she can contrast this to the population risk and the risk that would be associated with an inherited cancer susceptibility mutation. However, most estimates using these models and algorithms have limitations and they should not be the only means for risk assessment.

Thus, accurate genetic diagnosis and risk assessment requires the expertise of professionals who are familiar with the characteristics of genetic susceptibilities to common disease, the differential diagnosis of a suspected disease susceptibility, the resources available for determining disease risk based on a family history, as well as the likelihood that genetic testing may be useful in risk assessment, diagnosis and planning management and prevention.

Participation in effective early detection and prevention strategies should have significant benefit for individuals with a genetic susceptibility to common diseases. Under-utilization of these services by genetically susceptible individuals has been documented (Scheuner et al., 1997b). A genetic risk assessment survey of 176 managed care members found that the only significant predictor of cholesterol screening was advanced age (p=0.05). Gender, ethnicity, time since last visit to the doctor, and family history of cardiovascular disease (heart disease, stroke, and carotid, aortic or peripheral vascular disease), diabetes, or a cholesterol abnormality were not predictive of who was likely to have had cholesterol checked. Similarly, participation in mammography, sigmoidoscopy, fecal occult blood testing and serum prostate-specific antigen testing was associated with older age rather than a cancer family history. This included individuals from hereditary colon cancer and high-risk breast and ovarian cancer pedigrees. Among the respondents to the mailed genetic risk assessment survey, 53 charts from 15 primary care physicians selected randomly were reviewed for family history. This included data collected from 223 patient visits (4.2 visits per patient) that occurred over the 5-year period prior to the mailing of the survey. Among the charts reviewed, 39 belonged to subjects who gave self-reports of a family history of at least one common disease; the physician documented this family history in only 36%. The number of first-degree relatives that were reported by patients as having one of the conditions under study was 115, compared to only 23 documented by the physician. The corresponding number of self-reported affected second-degree relatives was 213, and the physicians documented only 4. The type of disease did not influence the family history documentation by the physician; they were dismal across all disease categories.

Failure of the primary care physician to collect the appropriate family history data appears to be an important factor that precluded adequate genetic risk assessment and access to appropriate preventive services for subjects with a genetic susceptibility to many common, chronic conditions. Acheson et al. (2000) have found that family practice physicians discuss family history about half of the time during new patient visits, and only 22% of the time during established patient visits. This included discussion of “family issues” that might represent a broad range of topics in addition to medical issues. The quality of information collected was likely limited (although this was not determined) since the average duration of family history discussions was less than 2.5 minutes. Only 11% of patients' records included a family tree. This was probably due to the limited time available for a patient visit, 10 minutes. It is estimated that construction of a pedigree in the family practice setting obtained via semi-structured interview takes 15 to 20 minutes.

In a survey of 339 primary care providers in the United Kingdom the majority felt a need to provide genetic services (Suchard et al., 1999). However, only 29% felt sufficiently prepared to take family histories and draw pedigrees and only 15% felt prepared to counsel patients about genetic test results. Almost two-thirds would participate in training and education regarding genetics and would participate in providing genetic services, yet not all were willing to provide these services. For practitioners who do feel comfortable with transmitting genetic information, the increased time demands required to research the family history and provide genetic counseling (Surh et al., 1995) may act as disincentives to routinely offering adequate genetic risk assessment.

Thus, rather than providing comprehensive genetic services for patients with complex genetics service needs, the role of the primary care provider should include: 1) identification of patients who may benefit from genetic services, 2) provision of basic genetic information to facilitate the referral process, 3) recognition of the special psychosocial issues for a family with a genetic condition, and 4) coordination of care and monitoring of health (Hayflick et al., 1998). Primary care providers' lack of general knowledge in genetics (Hoffman et al., 1993) may preclude appropriate referrals to genetic professionals for genetic risk assessment, counseling, and when appropriate, DNA testing (Hayflick et al., 1998). This is compounded by incentive arrangements that reward "gate-keeper" physicians to limit patient access to specialty referrals and treatments (Swartz and Brennan, 1996). Unfortunately, there are a limited number of professionals who are trained in genetics (Rowley et al., 1995), and this may be the most significant factor in related to provision of clinical genetic services for common diseases. Clinicians may also be reluctant to pursue genetic risk assessment for their patients since evidence regarding the efficacy and utility of using genetic information in disease management and prevention is limited. This lack of evidence should not, however, deter clinicians from utilizing genetic information regarding common disease, as it clearly has value in providing risk information and in many instances can guide decision making for disease management and prevention.

References

1. Acheson LS, Wiesner GL, Zyzanski SJ, Goodwin MA, Stange KC (2000) Family history-taking in community family practice: Implications for genetic screening. Genetics in Medicine 2:180-185
2. Acton RT, Go R, Roseman J (1989) Strategies for the utilization of genetic information to target preventive interventions. Proceedings of the 25th Annual Meeting of the Society of Prospective Medicine Society of Prospective Medicine, Indianapolis, Indiana , pp. 88-100
3. Alberts MJ (1990) Genetic aspects of cerebrovascular disease. Current Concepts of Cerbrovascular Disease and Stroke 25:25-30
4. Amos CI, Shaw GL, Tucker MA, Hartge P (1992) Age at onset for familial epithelial ovarian cancer. Journal of the American Medical Association 268:1896-1899
5. Amos CI, Struewing JP (1993) Genetic epidemiology of epithelial ovarian cancer. Cancer 71:566-572
6. Anderson DE, Badzioch MD (1993) Familial breast cancer risks. Effects of prostate and other cancers. Cancer 72:114-119
7. Benichou J (1993) A computer program for estimating individualized probabilities of breast cancer. Comp and Biomed Res 26:373-382
8. Bensen JT, Liese AD, Rushing JT, Province M, Folsom AR, Rich SS, Higgins M (1999) Accuracy of proband reported family history: The NHLBI Family Heart Study (FHS). Genetic Epidemiology 17:141-150
9. Boltenberg A, Furgyik S, Kullander S (1990) Familial cancer aggregation in cases of adenocarcinoma corporis uteri. Acta Obstet Gynecol Scand 69: 249-258
10. Carter BS, Beaty TH, Steinberg GD, Childs B, Walsh PC (1992) Mendelian inheritance of familial prostate cancer. Proc Natl Acad Sci 89:3367-337
11. Claus EB, Risch N, Thompson WD (1991) Genetic analysis of breast cancer in the cancer and steroid hormone study. Am J Hum Genet 48:232-242
12. Claus EB, Risch N, Thompson WD (1993) The calculation of breast cancer risk for women with a first degree family history of ovarian cancer. Breast Cancer Res and Treat 28:115-120
13. Colditz GA, Stampfer MJ, Willett WC, Rosner B, Speizer FE, Henneckens CH (1986) A prospective study of parental history of myocardial infarction and coronary heart disease in women. Am J Epidemiol 123:48-58
14. Colditz GA, Rimm EB, Giovannucci E, Stampfer MJ, Rosner B, Willett WC (1991) A prospective study of parental history of myocardial infarction and coronary artery disease in men. Am J Cardiol 67:933-938
15. Eddy DM, Nugent FW, Eddy JF, Coller J, Gilbertsen V, Gottlieb LS, Rice R, Sherlock P, Winawer S (1987) Screening for colorectal cancer in a high-risk population, results of a mathematical model. Gastroenterology 92: 682-92
16. Egan KM, Newcomb PA, Longnecker MP, Trentham-Dietz A, Baron JA, Trichopoulos D, Stamfer MJ, Willett WC (1996) Jewish religion and risk of breast cancer. Lancet 347:1645-1646
17. Evans RA, Marel GM, Lancaster ED, Kos S, Evans M, Wong SY (1998) Bone mass is low in relatives of osteoporortic patients. Ann Intern Med 109:870-873
18. Gail MH, Brinton LA, Byar DP, Corle DK, Green SB, Schairer C, Mulvihill JJ (1989) Projecting individualized probabilities of developing breast cancer for white females who are being examined annually. Journal of the National Cancer Institute 81:1879-86
19. Graffagnino C, Gasecki AP, Doig GS, Hachinski VC (1994) The importance of family history in cerebrovascular disease. Stroke 25:1599-1604
20. Greggi S, Genuardi M, Benedetti-Panici P, Cento R, Scambia G, Neri G, Mancuso S (1990) Analysis of 138 consecutive ovarian cancer patients: Incidence and characteristics of familial cases. Gynecologic Oncology 39:300-304
21. Grover S, Quinn MA, Weideman P (1973) Patterns of inheritance of ovarian cancer. An analysis from an ovarian cancer screening program. Cancer 72:526-530
22. Gruber SB, Thompson WD (1996) A population-based study of endometrial cancer and familial risk in younger women. Cancer and Steroid Hormone Study Group. Cancer Epidemiol Biomark Prev 5:411-417
23. Hayflick SJ, Eiff MP, Carpenter L, Steinberger J (1998) Primary care physicians' utilization and perceptions of genetics services. Genetics in Medicine1:13-21
24. Henderson LB, Adams JS, Goldstein DR, Braunstein GD, Rotter JI, Scheuner MT (2000) A familial risk profile for osteoporosis. Genetics in Medicine 2:222-225
25. Hoffman KJ, Tambor ES, Chase GA, Geller G, Faden RR, Holtzman NA (1993) Physician's knowledge of genetics and genetic tests. Acad Med 68:625-632
26. Hoskins KF, Stopfer JE, Calzone KA, Merajver SD, Rebbeck TR, Garber JE, Weber BL (1995) Assessment and counseling for women with a family history of breast cancer. Journal of the American Medical Association 273:577-585
27. Houlston RS, Bourne TH, Davies A, Whitehead MI, Campbell S, Collins WP, Slack J (1992) Use of family history in a screening clinic for familial ovarian cancer. Gynecol Oncol 47:247-252
28. Kee F, Tiret L, Robo JY, Nicaud V, McCrum E, Evans A, Cambien F (1993) Reliability of reported family history of myocardial infarction. British Medical Journal 307:1528-1530
29. Kahn LB, Marshall JA, Baxter J, Shetterly SM, Hamman RF (1990) Accuracy of reported family history of diabetes mellitus. Results from San Luis Valley Diabetes Study. Diabetes Care 13:796
30. Keen H, Track NS (1968) Age of onset and inheritance of diabetes: The importance of examining relatives. Diabetologia 4:317-321
31. Kiely DK, Wolf PA, Cupples A, Beiser AS, Myers RH (1993) Familial aggregation of stroke. The Framingham Study. Stroke 24:1366-1371
32. King RA, Rotter JI, Motulsky AG (1992) The approach to genetic bases of common diseases. In: King RA, Rotter JI, Motulsky AG (eds). The Genetic Basis of Common Diseases. Oxford University Press, Inc., New York, pp 3-18
33. Köbberling J, Tattersall R (1982) The genetics of diabetes mellitus. Academic Press, London.
34. Krall EA, Dawson-Hughes B (1993) Heritable and life determinants of bone mineral density. J Bone Miner Res 8:1-9
35. Love RR, Evans AM, Josten DM (1985) The accuracy of patient reports of a family history of cancer. Journal Chronic Disorders 38:289-93
36. Lovett E (1976) Family studies in cancer of the colon and rectum. Br J Surg 63:13-18
37. Macklin MT (1960) Inheritance of cancer of the stomach and large intestine in man. J Natl Cancer Inst 24:551-571
38. Mecklin J-P(1987) Frequency of hereditary colorectal carcinoma. Gastroenterology 93: 1021-25
39. Mitchell BD, Kammerer CM, Reinhart LJ, Stern MP (1994) NIDDM in Mexican-American families. Heterogeneity by age of onset. Diabetes Care 17:567-573
40. Pocock NA, Eisman JA, Hopper JL, Yeates MG, Sambrook PN, Eberl S (1987) Genetic determinants of bone mass in adults: A twin study. J Clin Invest 80:706-710
41. Ponz de Leon M, Scapoli C, Zanghieri G, Sassatelli R, Sacchetti C, Barrai I (1992) Genetic transmission of colorectal cancer: exploratory data analysis from a population based registry. J Med Genet 29: 531-538
42. Reaven GM (1988) Role of insulin resistance in human disease. Diabetes 37:595-607
43. Rissanen AM (1979) Familial occurrence of coronary heart disease: Effect of age at diagnosis. Am J Cardiol 44:60-66
44. Rose G (1964) Familial patterns in ischaemic heart disease. Br J Prev Soc Med 18:75-80
45. Sandles LG, Shulman LP, Elias S, Photopulos GJ, Smiley LM, Posten WM, Simpson JL (1992) Endometrial carcinoma: genetic analysis suggesting heritable site-specific uterine cancer. Gynecologic Oncology 47:167-171
46. Rowley PT, Loader S, Sutera CJ, Kozyra A (1995) Prenatal genetic counseling for hemoglobinopathy carriers: A comparison of primary providers of prenatal care and professional genetic counselors. American Journal of Human Genetics 56: 769-776>
    
47. Scheuner MT, Wang S-J, Raffel LJ, Larabell SK, Rotter JI (1997a) Family history: A comprehensive genetic risk assessment method for the chronic conditions of adulthood. American Journal of Medical Genetics 71:315-324
48. Scheuner MT, Tran LT, Henderson LB, Goldstein DR (1997b) Utilization of preventive strategies by genetically susceptible individuals. American Journal of Human Genetics 61:A57.
49. Scheuner MT (2001) Genetic predisposition to coronary artery disease. Current Opinion in Cardiology16:251-260
50. Schildkraut JM, Risch N, Thompson WD (1989) Evaluating genetic association among ovarian, breast, and endometrial cancer: Evidence for a breast/ovarian cancer relationship. American Journal of Human Genetics 45:521-529
51. Sholtz RI, Rosenman RH, Brand RJ (1975) The relationship of reported parental history to the incidence of coronary artery disease in the Western Collaborative Group Study. Am J Epidemiol 102:350-356
52. Slack J, Evans KA (1966) The increased risk of death from ischaemic heart disease in first-degree relatives of 121 men and 96 women with ischaemic heart disease. Journal of Medical Genetics 3:239-257
53. Slattery ML, Kerber RA (1993) A comprehensive evaluation of family history and breast cancer risk. JAMA 270:1563-1568
54. Spriggs DA, French JM, Murdy JM, Bates D, James OFW (1990) Historical risk factors for stroke: A case control study. Age and Ageing 19:280-287
55. Spitz MR, Currier RD, Fueger JJ, Babaian RJ, Newell GR (1991) Familial patterns of prostate cancer: A case-control analysis. J Urol 146:1305-1307
56. Steinberg GD, Carter BS, Beaty TH, Childs B, Walsh PC (1990) Family history and the risk of prostate cancer. The Prostate 17:337-347
57. Steinberg KK, Pernarelli JM, Marcus M, Khoury MJ, Schildkraut JM, Marchbanks PA (1998) Increased risk for familial ovarian cancer among Jewish women: A population-based case-control study Genet Epidemiol 15:51-59
58. Stephenson BM, Finan PJ, Gascoyne J, Garbett F, Murday VA, Bishop DT (1991) Frequency of familial colorectal cancer. Br J Surg 78:1162-66
59. Suchard MA, Yudkin P, Sinsheimer JS, Fowler GH (1999) General practitioners' views on genetic screening for common diseases. British Journal of General Practice 49:45-6
60. Surh LC, Wright PG, Cappelli M, Kasaboski A, Hastings VA, Hunter AG, and the Molecular Study Group (1995) Delivery of molecular genetic services within a health care system: Time analysis of the clinical workload. American Journal of Human Genetics 56: 760-768
61. Swartz K, Brennan TA (1996) Integrated health care, capitated payment, and quality: the role of regulation. Annals of Internal Medicine 124:442-48
62. Walker M (1999) Non-diabetic relatives: Characteristics and opportunities for intervention. In: Hitman GA (ed) Type 2 Diabetes Prediction and Prevention. John Wiley & Sons, New York, pp 303-322
63. Watson P, Lynch HT (1993) Extracolonic cancer in hereditary nonpolyposis colorectal cancer. Cancer 71:677-685
64. Woolf CM (1958) A genetic study of carcinoma of the large intestine. American Journal of Human Genetics 10:42-47

Do Americans Know Their Family History of Asthma and Heart Disease?
Joelyn Tonkin, MSPH, Paula Yoon, ScD, MPH, Muin Khoury, MD, PhD,
Office of Genomics and Disease Prevention, CDC

It has been proposed that family history could be used as a public health tool for risk stratification and screening. It is unclear, however, what proportion of the population is likely to know their family history. This study uses data from a national health survey to determine the proportion of people who can provide a family history of first-degree relatives for asthma and heart disease and examines the characteristics of people who do not know their family history for first-degree relatives. Healthstyles 2001, a population-based survey, included four family history questions. Participants were asked if their biological mother, father, or siblings had ever had asthma or heart disease. Results showed that 13.7% and 17.0% of respondents to this survey could not provide a complete family history of first-degree relatives for asthma and heart disease respectively. Logistic regression found that significant predictors for reporting an incomplete family history of asthma were age 65 and over (OR=1.8), income <$25,000 (OR=1.9) and $25,000-$49,999 (OR=1.4), Black (OR=2.2) and Hispanic (OR=1.5) race, governmental or no insurance (OR=1.4), and having asthma (OR=2.1). Significant predictors for reporting an incomplete FH of heart disease were age 65 and older (OR=1.7), income <$25,000 (OR=1.8) and $25,000-$49,999 (OR=1.3), Black (OR=1.6) and Other (OR=2.2) race, governmental or no insurance (OR=1.3), having asthma (OR=1.9), and not understanding health issues (OR=1.3). Some people who reported an incomplete family history of first-degree relatives, still provided enough information that could be used to stratify them into risk categories. Of the 511 people who had incomplete family histories for asthma, 17.0% could be classified as having at least a moderate family history and 6.5% could be classified as having a strong family history. Of the 630 people who had incomplete family histories heart disease, 30.4% could be classified as having at least a moderate family history and 15.3% could be classified as having a strong family history. In conclusion, a majority of the population can report a complete family history at least for first-degree relatives. Of those that cannot, even partial family history information may be useful for identifying people who are at higher risk than the population at large.

We investigated the effect of family history of cancer on likelihood of receiving cancer-screening tests. We analyzed nationally representative data from the year 2000 National Health Interview Survey Cancer Control Topical Module (CCTM) to evaluate whether individuals with a self-reported family history of either breast, cervical colorectal, or prostate cancers were more likely than those without a family history to undergo mammography, clinical breast exam, Pap smear, fecal occult blood test, colonoscopy/flexible sigmoidoscopy, or prostate specific antigen testing.

Preliminary results show a statistically significant difference in receipt of all cancer screening tests among those with a positive family history. In addition, among those screened, individuals with a positive family history were more likely to adhere to recommended screening guidelines and to receive screening tests within recommended intervals.

Population Versus High-Risk Prevention Strategies: Do We Need to Choose?
Steven C. Hunt, PhD, University of Utah

Family history is highly and independently predictive of future disease for most common diseases. While any individual risk factor may be predictive of disease, family history represents the combined influence of all risk factors, genetic, environmental, and behavioral, on disease expression in multiple family members. Families with a positive family history, regardless of what the risk factors are, have one thing in common: they express the disease. Assessing family history and determining shared (or unshared) risk factors leading to disease in each family is family-based medicine at its best.

The majority of cardiovascular disease occurs in a small proportion of families with a positive family history of cardiovascular disease. This means that effective identification of this relatively small subset of the population and targeted application of family-based intervention methods to these families could have a tremendous impact on reducing morbidity and mortality.

Studies of disease risk and risk factor distributions have suggested that the majority of events occur, not at the extremes of the distribution, but in the upper normal/borderline high range of the risk factor. For this reason, it has been questioned whether a high-risk approach using individual risk factors is cost effective compared to a population approach to risk factor reduction. However, the portion of the risk factor distribution from which most of the events arise is the main subgroup of people targeted by assessing positive family histories. Therefore, picking extremes based upon family history targets the portion of the population that has the most events and is not the same as targeting extremes of a single risk factor distribution.

Family history assessment identifies family members who are already affected who could benefit from secondary prevention and unaffected family members at risk who could benefit from primary prevention. In addition, even those families with an average risk would benefit by being reminded of current population recommendations for risk factor screening and control that should accompany a family history report.

Of course, collecting family history is not an easy task and can be costly. We have developed methods over the years that have significantly reduced the cost of assessing family history in large populations and have recently begun work on making Internet versions of these methods available. A validated and user-friendly Internet family history collection tool would be nearly cost-free and could be used by schools, health departments, in clinical settings, and by the general public. To accomplish this, some type of consensus document should be created concerning standard data that should be collected on each family and a standard definition of a positive family history should be adopted that has the greatest validity. Similar to the National Cholesterol Education Program guidelines or the Joint National Committee on Prevention, Detection, Evaluation, and Treatment of High Blood Pressure guidelines, which have evolved over time but provide the best current advice, guidelines on family history could help with the training of the health community, standardization of recommendations, and effective intervention development. These developments would lead to effective utilization of family history in the population to provide important health information to all families and to target high-risk families for special attention. Those individuals at highest risk usually cannot reduce their risk to appropriate levels without qualified help.

Association Between Family History of Colorectal Cancer and Preventive Health Behaviors in Women - United States, 1997
Jill M. Morris, PhD, Amanda Brown, PhD, Marta Gwinn, MD, MPH, Muin Khoury, MD, PhD, Office of Genomics and Disease Prevention, CDC

Background: Colorectal cancer (CRC) is the third leading cause of cancer-related mortality in the U.S., claiming over 56,000 lives annually. Although family history (FHx) is the second strongest risk factor for CRC, it is not known whether having a FHx of CRC motivates people to change lifestyle and screening behaviors to reduce their risk.

Methods: We conducted a cross-sectional analysis of data from 64,473 cancer-free women interviewed in 1997 for the American Cancer Society's Cancer Prevention Study II. FHx of CRC was categorized as strong (2 or more first-degree relatives with CRC, or one aged <50) or moderate (only 1 first-degree relative with CRC, not aged <50). We used multinomial logistic regression to estimate adjusted odds ratios (OR) for associations between FHx and screening/lifestyle factors. Regression models included age, education, type of health care coverage, and FHx of other cancers.

Results: The median age of participants was 67; 98% were white. Overall, 13.5% of respondents reported a FHx of CRC: 2.1% had a strong and 11.3% had a moderate FHx. Compared with women who had no FHx of CRC, those with a strong FHx were more likely to have undergone recent endoscopy (OR=2.3, 95% Confidence Interval [CI]=2.1-2.6), as were those with a moderate FHx (OR=1.8, CI=1.7-1.9). Women in both groups were slightly more likely to be nonsmokers, regular aspirin users, and to have had a rectal exam.

Conclusions: Awareness of a FHx of CRC may motivate people to seek endoscopy or engage in other preventive behaviors. Promoting awareness of FHx and incorporating FHx into prevention messages may be promising tools for increasing CRC screening and other preventive behaviors.

Using Decision Analytic Method to Assess the Utility of Family History for Public Health and Preventive Medicine
Anupam Tyagi, PhD, National Center for Chronic Disease Prevention and Health Promotion

In addition to genetic testing for common diseases, family history may be a useful tool for identifying people at increased risk of disease and developing targeted interventions for individuals at higher-than-average risk . This presentation addresses the issue of how to examine the utility of a family history tool for public health and preventive medicine. We propose the use of a decision analytic framework for the assessment of a family history tool, and outline the major elements of a decision-analytic approach including analytic perspective, costs, outcome measurements, and data needed to assess the value of a family history tool. We also describe the need for effective communication, implications of measurement error and imperfect information, and the use of sensitivity analysis to address uncertainty in parameter values of a decision model. Two examples are used to illustrate the framework. Sensitivity analysis is used to identify information gaps and research challenges that need to be addressed to make family history information a useful public health tool.

Can Knowledge of Family History Affect Behavior Change: Using Family History of Breast Cancer as an Example
Janet Audrain, PhD, University of Pennsylvania

Public health approaches to disease prevention have had a limited impact on promoting health behavior change. More intense interventions aimed at the general population may not be feasible and would likely be cost prohibitive. Approaches that are targeted to those with an elevated risk for developing disease may be more effective. Family history of disease could be used as a public health tool for risk stratification leading to improved disease prevention. Interventions directed toward those with a family history rather than the general population may result in cost-effective approaches aimed at subpopulations most in need. That is, if family history can aid in risk stratification, improve early detection and disease prevention, and influence health-promoting behaviors (e.g., routine screening, healthy diet, physical activity, weight maintenance), then family history could be used to target individuals at elevated risk who could benefit the most from interventions.

This presentation explores using family history of breast cancer as a public health tool for risk stratification and improved disease prevention. Specifically, the presentation will explore whether knowledge of family history can affect behavior change using family history of breast cancer as an example. Breast cancer is the most common neoplasm among women in the United States and second only to lung cancer in deaths attributable to cancer. It is estimated that in year 2002, over 200,000 women will be diagnosed with breast cancer and almost 40,000 will die of this disease. Having a family history of breast cancer in a first-degree relative is one of the most important risk factors for this disease. Women who have an affected first-degree relative have a 2- to 10-fold increased risk of developing breast cancer. This group may comprise an important target for family history risk education. Focusing efforts on women with a family history of breast cancer may improve early detection and breast cancer prevention, and influence health-promoting behaviors (e.g., cancer screening, physical activity, weight maintenance).

Through a review of the available research on women at moderate risk for developing breast cancer due to family history, the presentation will address important questions in considering family history as a public health tool and whether knowledge of family history impacts behavior change. For example, are women with a family history of breast cancer aware of their elevated risk compared to women without a family history? What are the characteristics of the women who lack this knowledge? Would women at increased risk due to family history be more accepting of recommendations about lifestyle changes and participation in heightened early detection and prevention strategies? Who is more likely to participate in breast cancer risk education programs? Does breast cancer risk counseling improve knowledge and comprehension of risk and does this translate into or motivate behavior change? Who is most likely to benefit? Are there behavioral and psychological benefits to receiving this knowledge? Are their behavioral and psychological characteristics that either promote or deter women who are aware of their elevated risk to engage in breast health promoting behaviors? Is knowledge of increased risk due to family history sufficient to change behavior or do skills training need to be part of the intervention. It is concluded that targeting interventions to individuals who have a family history of disease may be an effective strategy for improving health-promoting behaviors, although interventions that only focus on increasing knowledge of risk associated with family history may not be sufficient for long-term behavior change. Interventions may need to include behavior change skills training (e.g., goal setting, self-monitoring) and an assessment and management of barriers to behavior change.