0
Original Research |

Effect of Adding Systematic Family History Enquiry to Cardiovascular Disease Risk Assessment in Primary Care : A Matched-Pair, Cluster Randomized Trial FREE

Nadeem Qureshi, DM; Sarah Armstrong, PhD; Paula Dhiman, MSc; Paula Saukko, PhD; Joan Middlemass, MPhil; Philip H. Evans, MPhil; Joe Kai, MD, ADDFAM (Added Value of Family History in CVD Risk Assessment) Study Group
[+] Article and Author Information

Disclaimer: The views expressed are those of the authors and not necessarily those of the United Kingdom Department of Health.

Acknowledgment: The authors thank the participants and practices for their contribution; a study external advisory committee (Michael Modell, Richard McManus, and Stirling Bryan); Uzair Suhail for his assistance in collating data extracted from general practitioner computer records; Paula Yoon for her advice on the protocol; Judith Allanson, Penny Standen, and Brenda Wilson for their literature review; and Tim Coleman for his helpful comments on a draft version of the paper.

Grant Support: By grant HSR 36A from the Genetics Health Services Research program of the United Kingdom Department of Health.

Potential Conflicts of Interest: Disclosures can be viewed at www.acponline.org/authors/icmje/ConflictOfInterestForms.do?msNum=M11-1217.

Reproducible Research Statement:Study protocol: Available (14). Statistical code and data set: Available from Dr. Qureshi (e-mail, mailto:nadeem.qureshi@nottingham.ac.uk), subject to approval by national ethics and study steering committees.

Requests for Single Reprints: Nadeem Qureshi, DM, Division of Primary Care, University of Nottingham, Tower Building, 13th Floor, University Park, Nottingham NG7 2RD, United Kingdom; e-mail, mailto:nadeem.qureshi@nottingham.ac.uk.

Current Author Addresses: Drs. Qureshi, Armstrong, and Kai and Ms. Dhiman: Division of Primary Care, University of Nottingham, Tower Building, 13th Floor, University Park, Nottingham NG7 2RD, United Kingdom.

Dr. Saukko: Department of Social Sciences, Brockington Building, Loughborough University, Loughborough, Leicestershire LE11 3TU, United Kingdom.

Ms. Middlemass: Primary Care, School of Health and Social Care, University of Lincoln, Brayford Pool, Lincoln, Lincolnshire LN6 7TS, United Kingdom.

Mr. Evans: Peninsula Medical School, St. Luke's Campus, Smeall Building, Magdalen Road, Exeter, Devon EX1 2LU, United Kingdom.

Author Contributions: Conception and design: N. Qureshi, S. Armstrong, P. Saukko, P.H. Evans, J. Kai.

Analysis and interpretation of the data: N. Qureshi, S. Armstrong, P. Dhiman, J. Kai.

Drafting of the article: N. Qureshi, S. Armstrong, J. Kai.

Critical revision of the article for important intellectual content: N. Qureshi, S. Armstrong, P. Saukko, P.H. Evans, J. Kai.

Final approval of the article: N. Qureshi, S. Armstrong, P. Dhiman, P. Saukko, J. Middlemass, P.H. Evans, J. Kai.

Provision of study materials or patients: N. Qureshi, P. Saukko, J. Middlemass.

Statistical expertise: S. Armstrong, P. Dhiman.

Obtaining of funding: N. Qureshi, S. Armstrong, P. Saukko, J. Middlemas, J. Kai.

Administrative, technical, or logistic support: J. Middlemass, P.H. Evans.

Collection and assembly of data: N. Qureshi, P. Saukko, J. Middlemass.

* For additional members of the ADDFAM study group, see Appendix 1.


From University of Nottingham, Nottingham; Loughborough University, Loughborough; University of Lincoln, Brayford Pool, Lincoln; and Peninsula Medical School, Exeter, United Kingdom.


Ann Intern Med. 2012;156(4):253-262. doi:10.7326/0003-4819-156-4-201202210-00002
Text Size: A A A

Background: Evidence of the value of systematically collecting family history in primary care is limited.

Objective: To evaluate the feasibility of systematically collecting family history of coronary heart disease in primary care and the effect of incorporating these data into cardiovascular risk assessment.

Design: Pragmatic, matched-pair, cluster randomized, controlled trial. (International Standardized Randomized Controlled Trial Number Register: ISRCTN 17943542).

Setting: 24 family practices in the United Kingdom.

Participants: 748 persons aged 30 to 65 years with no previously diagnosed cardiovascular risk, seen between July 2007 and March 2009.

Intervention: Participants in control practices had the usual Framingham-based cardiovascular risk assessment with and without use of existing family history information in their medical records. Participants in intervention practices also completed a questionnaire to systematically collect their family history. All participants were informed of their risk status. Participants with high cardiovascular risk were invited for a consultation.

Measurements: The primary outcome was the proportion of participants with high cardiovascular risk (10-year risk ≥20%). Other measures included questionnaire completion rate and anxiety score.

Results: 98% of participants completed the family history questionnaire. The mean increase in proportion of participants classified as having high cardiovascular risk was 4.8 percentage points in the intervention practices, compared with 0.3 percentage point in control practices when family history from patient records was incorporated. The 4.5–percentage point difference between groups (95% CI, 1.7 to 7.2 percentage points) remained significant after adjustment for participant and practice characteristics (P = 0.007). Anxiety scores were similar between groups.

Limitations: Relatively few participants were from ethnic minority or less-educated groups. The potential to explore behavioral change and clinical outcomes was limited. Many data were missing for anxiety scores.

Conclusion: Systematically collecting family history increases the proportion of persons identified as having high cardiovascular risk for further targeted prevention and seems to have little or no effect on anxiety.

Primary Funding Source: Genetics Health Services Research program of the United Kingdom Department of Health.

Editors' Notes
Context

  • Information about a patient's family history can improve risk assessment for cardiovascular disease.

Contribution

  • This cluster randomized trial, which involved 748 adult patients with no previously diagnosed cardiovascular risk across 24 family practices, found that collecting family history data with mailed questionnaires identified more high-risk patients who were eligible for targeted prevention than did usual practice procedures (4.8% vs. 0.3%).

Caution

  • Effect on long-term clinical outcomes of patients was not evaluated.

Implication

  • Systematic collection of family history data is a feasible practice-level intervention that could improve cardiovascular risk assessment and help target patients who are most in need of preventive interventions.

—The Editors


Family history is a recognized risk factor for many chronic diseases (1) and is traditionally part of history taking in clinical practice (2). However, a recent National Institutes of Health State-of-the-Science conference (23) identified the need for evidence, from controlled trials, on the value of systematically collecting and using family history in primary care. Family history is rarely used in isolation but instead is part of a multifactorial risk assessment, such as for coronary heart disease (CHD). The Joint British Societies 2 (JBS2) cardiovascular risk assessment guidelines use the Framingham algorithm and are widely adopted in the United Kingdom (46). The cardiovascular risk prediction charts in these guidelines (5) use the core cardiovascular factors of age, sex, smoking status, systolic blood pressure, and ratio of total to high-density lipoprotein (HDL) cholesterol levels. However, family history of premature CHD can also be subjectively incorporated at the clinician's discretion. Like the U.S. National Cholesterol Education Program Adult Treatment Panel III guideline (5, 7), the JBS guidelines define a significant family history as CHD in a male first-degree relative younger than 55 years or a female first-degree relative younger than 65 years.

Available observational epidemiologic studies (89) suggest that adding family history might identify more than 60% of persons who have the greatest risk for CHD and who might benefit from preventive care. However, experimental evidence from pragmatic randomized trials is required to evaluate the effect of family history interventions and the feasibility of implementing them in usual primary care practice (23). During consideration of appropriate interventions, it should be recognized that family history is poorly recorded in electronic health records in family practice (1011). A more systematic approach to collecting family history is needed to improve identification of significant familial risk. Self-administered family history questionnaires may provide a solution (1213).

Two key questions arise for clinicians and policymakers. First, in a healthy population that accepts an invitation to cardiovascular risk assessment, how feasible is it for primary care physicians to collect more detailed family history information? Second, if such information is collected and used systematically, how many more persons at high risk for cardiovascular disease will be identified? We compared an intervention in which family history of premature CHD was systematically collected and incorporated into cardiovascular risk assessment in primary care with risk assessment based on usual practice. We hypothesized that systematically collecting family history in family practice would improve identification of persons with undiagnosed high cardiovascular risk. We also explored any effect on participant anxiety and potential changes in self-reported behavior.

Our trial methods have been described in detail elsewhere (14). The hypotheses were tested in a pragmatic, matched-pair, cluster randomized, controlled trial between July 2007 and March 2009. Ethical approval was obtained from a United Kingdom Medical Research Ethics Committee (reference 06/MRE10/9).

Setting and Randomization

All family practices in the research networks of the central and southwestern regions of England were contacted. To prevent an imbalance between control and intervention groups, eligible practices that were willing to participate were matched into pairs according to United Kingdom Index of Multiple Deprivation score and ethnicity (<10% or ≥10% ethnic minorities originating from the Indian subcontinent), as recognized risk factors for CHD (5, 1517). One practice in each pair was randomly assigned to the family history intervention group and the other to the control (usual care) group by using the Web-based randomization service of the Nottingham Clinical Trials Unit, Nottingham, United Kingdom. Randomization was stratified by the 2 regions. Physicians were aware of the intervention used in their own practices but not in other practices. For practical reasons, we could not blind research fellows; however, the primary outcome was an objective measure obtained from a predefined algorithm (JBS2 risk calculator), and secondary outcomes were self-reported by participants. The data entry clerks and statisticians were blinded to group assignment until all analyses had been completed.

Before participant recruitment, research fellows facilitated a standardized training session at each practice. For both groups, this involved the use and interpretation of cardiovascular risk scores and current public health recommendations for lifestyle advice (5). In the intervention group, clinicians were given information on interpreting and communicating the risk associated with a family history of premature CHD, on the basis of a pilot study (18).

The standard cardiovascular risk score is calculated by inputting core risk factors (age, sex, smoking status, systolic blood pressure, and total–HDL cholesterol ratio) into the cardiovascular risk calculator. This score was multiplied by 1.5 if a family history of premature CHD was identified (5). Appendix 2 provides further details.

Participants and Recruitment

To be included, participants had to be aged 30 to 65 years and had to request or be offered a cardiovascular risk assessment by their family physician, as per usual practice. Participants were excluded if they had previously diagnosed diabetes or atherosclerotic disease (CHD, stroke, or peripheral vascular disease), were already receiving lipid-lowering medications, or were excluded by their family physicians for psychological or social reasons.

The initial cardiovascular risk assessment consultation was similar across sites, with the family physician or office nurse checking for exclusion criteria in the JBS2 cardiovascular risk assessment, measuring blood pressure, and documenting smoking status (Appendix 2). Serum cholesterol samples were collected by local phlebotomy services. The physician or nurse also checked study exclusion criteria and gave eligible participants an invitation letter, consent form, information leaflet, study questionnaire for secondary outcomes, and a family health questionnaire (intervention group only). Consent included access to participants' electronic health records to extract anonymized data on family history information, other cardiovascular risk factors, investigations, and medication. Two weeks after the consent forms were returned to the research office, practices were contacted to confirm the eligibility of potential participants and to gather data on core cardiovascular risk factors.

Interventions

Participants in both groups received standard cardiovascular risk assessment for core risk factors (5). In the intervention group, family history of CHD was also systematically collected by using a self-administered questionnaire at recruitment and this was incorporated into the risk assessment score (5).

The self-administered family history questionnaire was designed to collect information in primary care on cardiovascular disease, cancer, and reproductive carrier status (12, 19). This tool has been successfully validated and demonstrated 90% agreement in identification of family history of premature CHD with a criterion standard (clinical genetic 3-generation pedigree-drawing interview) and with confirmation of face and content validity from participants and family physicians (10, 12). The questionnaire was further tested with family practices that performed cardiovascular risk assessment (18). The questionnaire covers personal medical history, details of CHD in parents and grandparents, and family size (number of siblings, offspring, uncles, and aunts). Further sections are completed if participants recall any relatives listed in the family size section who had heart disease or died.

The cardiovascular risk scores were calculated by research fellows. The results were then sent back to the participants, along with a lifestyle advice leaflet, within 4 weeks of the original assessment. A duplicate copy was also sent to the family physician. Consistent with usual clinical practice, all participants who had a 20% or greater risk for cardiovascular disease over the next 10 years were offered an appointment to see their family physician or office nurse about 2 weeks after the result letter was posted. In this consultation, the increased risk was explained and lifestyle advice was offered. In the intervention group, the effect of a family history of premature CHD on cardiovascular risk was also discussed.

Outcomes and Follow-up

To determine the feasibility of using the family history questionnaire, we calculated the response and completion rates. The former was defined as the proportion of participants assigned to the family history intervention group who returned the questionnaire; the latter as the proportion of participants who entered information on their family size and their parents or grandparents.

The primary outcome measure was proportion of participants classified as having high risk for cardiovascular disease (10-year risk ≥20%). The electronic health records at each practice were reviewed to identify available data on family history of premature CHD and to extract anonymized data on other cardiovascular risk factors, investigations, medication, and newly diagnosed CHD.

For all participants, secondary outcome measures were assessed at baseline and 6 months by using a self-administered questionnaire to collect information on anxiety (6-item Spielberger State-Trait Anxiety Inventory), smoking, exercise (stage of change), and fat intake (1923). Full details are described elsewhere (14).

Sample Size

The sample size for the primary outcome measure was based on a comparison of the change in the percentage of participants classified as having high cardiovascular risk between study groups. This assumed that the proportion of participants identified as having high cardiovascular risk would increase by 3 percentage points in the family history intervention group and would not increase in the control group, based on usual practice at the start of the study (10). Assuming a power of 80% and a 2-tailed α of 5%, 265 participants per group were required to detect a difference of 3 percentage points. To allow for the cluster design, we assumed an intraclass correlation coefficient of 0.01 and a cluster size of 40. This gave a sample size of 369 participants per group completing cardiovascular risk assessment (10 practices per group).

Statistical Analysis

Analysis was undertaken on an intention-to-treat basis because practices (and thus participants) were analyzed in the groups to which they were randomly assigned. The time points of interest were immediately after the cardiovascular risk score calculation for the primary outcome measure and 6 months after the intervention for the secondary outcome measures. The primary comparison was the difference in the increase in the proportion of participants classified as having high cardiovascular risk in the intervention practices (with addition of systematic family history in cardiovascular disease assessment) compared with control practices. This comparison was calculated twice, first without taking account of available family history in electronic health records in the control group and, second, with the family history information included. Family history of CHD was assumed to be negative if no information on family history was available in the electronic health record. For the secondary outcomes of anxiety score, exercise, and dietary intake, the 6-month follow-up measures were compared between the groups. For smoking, change between baseline and 6-month follow-up was compared. Two statisticians performed independent analyses by using STATA, version 11.0 (StataCorp, College Station, Texas).

Similar to other studies with matched-pair designs (2426), primary and secondary outcomes were compared between groups by using a 2-stage procedure. In the first stage, we adjusted for practice- and participant-level variables. In the second stage, we tested for the intervention effect—that is, the difference in event rates (for example, the mean increase in proportion of participants at high cardiovascular risk)—between the intervention and control group pairs. In the first stage, we fitted a logistic model containing practice- and patient-level covariates but no intervention indicators. We used the residuals obtained from this model (the difference between the predicted outcome rate and the observed outcome rate for each practice) in the second stage of the analysis, in which we applied the weighted paired t test to the difference between the residuals corresponding to the practices in a matched pair to test for the intervention effect. The number of participants in each practice pair was used as the weights in this analysis. If the assumptions of the paired t test were not met, an unweighted Wilcoxon matched-pairs signed rank test was used instead. A permutation test was used to assess the robustness of the t test (27). All analyses were adjusted for resident training status of practice, a proxy measure of practice workload, and the ethnicity and educational status of the participants. Both practice variables would affect the family physicians who offered opportunistic cardiovascular risk assessment, and the participant variables would increase cardiovascular risk (5, 1517, 2830). Secondary outcomes were also adjusted for baseline measures, and the participant variables used to calculate the cardiovascular risk score (such as sex) were adjusted for by the inclusion of baseline cardiovascular risk in the models. In the analysis of secondary outcomes with high proportions of missing values, the values were replaced by using the ICE command in STATA to carry out multiple imputation of all predictor variables in the model (10 imputations).

Role of the Funding Source

Our study was funded by the Genetics Health Services Research program of the United Kingdom Department of Health. The funders and sponsors did not participate in the design or conduct of this study, analysis or interpretation of data, or writing of or decision to submit the manuscript for publication.

Of the 33 practices that expressed interest, 2 were outside the geographic area. The remaining 31 practices used the JBS2 risk assessment tool to assess cardiovascular risk. After receiving further study information, 7 practices declined to participate (1 practice already incorporated family history in risk assessment). The remaining 24 practices did not use family history in cardiovascular risk assessment and were divided into 12 pairs (6 in each region).

Overall, 748 eligible participants were recruited from the 12 pairs of family practices. Practices and participants in both groups had similar characteristics, apart from a higher proportion of male and Asian participants in the intervention group (Tables 1 and 2).

Table Jump Placeholder Table 1. Characteristics of Study Practices at Baseline
Table Jump Placeholder Table 2. Characteristics of Participants at Baseline
Study Progress

The Figure summarizes participant flow through the study. All practices received the intervention to which they were allocated, and none of the practices withdrew after recruitment. A total of 1828 recruitment packs were distributed to the 24 practices (915 to the intervention group and 913 to the control group). The recruitment rates were 45.0% (412 consenting participants) in the intervention group and 41.6% (380 consenting participants) in the control group. In the multivariate analysis of the primary outcome measure, 13 participants were excluded because of missing covariate data (ethnicity or education) and an additional 110 were excluded for the secondary outcome measures because they did not return the final study questionnaire.

The response rate to the 6-month questionnaire was 86.6% (310 of 358 participants) in the control group and 84.1% (328 of 390 participants) in the intervention group. For the 105 participants identified as having high cardiovascular risk, the median time between posting the results letter and the date of consultation was 12 days (interquartile range, 7 to 16.5 days). The initial offer of a consultation was accepted by 27 participants (75%) in the control group and 48 (69.6%) in the intervention group.

Of the 412 participants in the intervention group, 7 did not return their family history questionnaires, for a response rate of 98.3%. Among the 405 participants who returned the questionnaire, 18 (4.4%) did not complete the family size section and 19 (4.7%) provided no information on the age or medical history of their parents or grandparents. Overall, 24 questionnaires (5.9%) had at least 1 incomplete section and were considered poorly completed.

Primary Outcome Measure

The primary outcome measure was analyzed for 735 (98.3%) of allocated participants (Figure). Table 3 shows the proportion of participants at high cardiovascular risk after standard cardiovascular risk assessment (control group) and enhanced cardiovascular risk assessment incorporating systematically collected family history of premature CHD (family history intervention group). For participants in the intervention group, the percentage of participants at high risk increased by 5.1 percentage points, compared with a 0.5–percentage point increase in the control group when family history from electronic health records was incorporated. The number of participants at high risk in the intervention group increased from 49 to 69 (40.8%) after family history from the questionnaire was incorporated, compared with a 5.6% increase in the control group (from 36 to 38 participants).

Table Jump Placeholder Table 3. Proportion of Patients With High CVD Risk

Across the intervention practices, the percentage of participants at high cardiovascular risk increased by a mean of 4.8 percentage points (Table 3), compared with a mean increase of 0.3 percentage point in the control practices when family history from electronic health records was incorporated. Thus, the mean difference between study groups was 4.5 percentage points (95% CI, 1.7 to 7.2 percentage points) when family history from the control group was incorporated and 4.8 percentage points (CI, 2.0 to 7.7 percentage points) when it was excluded. After adjustment for participant- and practice-level variables, the difference between groups was significant (P = 0.007 when incorporating family history from the control group; P = 0.005 when excluding it). The P value of the permutation test was identical to that obtained from the original analysis, which confirms the results of our original paired t test (27).

Table 4 presents the proportion of all participants with a family history of CHD in each group, regardless of the cardiovascular risk status of the participant. About 5% of participants in each group (5.9% in the control group vs. 5.4% in the intervention group) had a family history of premature CHD recorded in their health records, which increased to 29.2% in the intervention group when information from the family history questionnaire was added. In both groups, more than 50% of electronic health records had no information on any positive or negative family history of CHD.

Table Jump Placeholder Table 4. Prevalence of Family History of CHD From Electronic Health Records and Family History Questionnaires
Secondary Outcome Measures

Appendix Table 1 summarizes secondary outcome measures at baseline and at 6-month follow-up. In the multivariate analysis, multiple imputation was used to replace missing values for these outcomes because of the high proportion of participants with missing data (anxiety, 180 participants [24%]; stage of exercise, 142 participants [19.0%]; total fat intake, 607 [81.1%]). Study groups did not significantly differ in anxiety levels or any other outcome measure (Appendix Table 2). Because of the small number of smokers at baseline, multiple imputation was not performed. Among participants who smoked at baseline and who completed the 6-month questionnaire, none in the control group stopped smoking; however, 6 of 30 (20.0%) smoked less. In the intervention group, 10 quit smoking and 8 smoked less (18 of 29 [62.1%]). The study groups significantly differed in smoking cessation or reduction (chi-square = 9.14; P = 0.001).

Table Jump Placeholder Appendix Table 1. Secondary Outcome Variables at Baseline and 6-Month Follow-up for Participants Who Completed Baseline and 6-Month Questionnaires
Table Jump Placeholder Appendix Table 2. Difference in Secondary Outcome Measures Between Practices in Intervention and Control Groups

For changes in medication use from baseline to 6 months among all participants, self-reported aspirin use was greater in the intervention group than in the control group (29 to 43 participants [48.3% increase] vs. 19 to 25 participants [31.6% increase]). At 6-month follow-up, 58 participants (16.2%) in the control group were receiving statins, compared with 56 (14.4%) in the intervention group. Over the 6-month study, 2 participants (0.5%) in the intervention group and 7 (2.0%) in the control group were diagnosed with CHD.

We found that adding systematic collection of family history improves the identification of persons with high cardiovascular risk in primary care practice. Compared with practices that offered the usual Framingham-based assessment, the intervention practices identified an additional 5% of persons with high cardiovascular risk. The substantial participant response and completion rates suggest that collecting the necessary family history data by using a self-completed questionnaire is feasible.

To our knowledge, ours is the first controlled trial to evaluate the clinical utility of systematically collecting family history as part of a multifactorial cardiovascular risk assessment in primary care (3, 15). This use within a multifactorial assessment of clinical risk to inform preventive clinical intervention adds to existing evidence from standalone assessments of family history for identifying familial risk for cancer, diabetes, and cardiovascular disease (3, 15, 3133).

The number of persons classified as having high cardiovascular risk increased by 41% in the intervention group (from 49 to 69), compared with a 6% increase in the control group (from 36 to 38), after family history available in electronic health records was taken into account. The persons identified would originally have been classified as having moderate cardiovascular risk (10-year risk of 10% to 19%) and would not be considered for statins or aspirin (5). Being identified as having high cardiovascular risk would justify targeting of intensive lifestyle change and preventive medication. In particular, statins could cost-effectively reduce future CHD events by more than 30% in these persons (3437). Consistent with previous work (3, 19, 31), our intervention did not lead to undue anxiety.

Our trial benefited from minimal missing cases for primary outcome measures and a high response rate to study questionnaires. To improve generalizability and anticipate implementation in practice, we used a pragmatic intervention and process for cardiovascular risk assessment that reflected procedures currently used in clinical practice. Our prior pilot work (12, 18) indicated that recording family history of CHD is very limited in practice. As confirmed at practice recruitment, it was also uncommon for family physicians to incorporate family history into cardiovascular risk assessment, even when this information was available in electronic health records (18). Thus, our original primary outcome measure compared cardiovascular risk assessment using systematic collection of family history with assessment involving no family history information. However, the potential value of family history in cardiovascular risk assessment has recently become more prominent in primary care practice and policy in the United Kingdom (3839). Thus, our revised primary outcome measure compared cardiovascular risk assessment that included systematically collected family history with an assessment that took family history available in patient records into account. Our study confirms that using the latter method only minimally increases the proportion of participants identified as having high cardiovascular risk. Moreover, a large proportion of records contains no information on family history at all. As definitions of relevant family history of CHD for risk assessment tools become more complex, systematic recording of family history will probably increase in importance; our findings further support this (16, 3940).

One of the greatest limitations to implementing systematic family history enquiry is its collection within the health consultation itself. Our findings suggest that collating and summarizing the information before the consultation might facilitate its use in cardiovascular risk assessment or in other clinical contexts. For example, these data could be collected when a person registers with a family practice (12, 19) and inserted into electronic health records by using predefined codes for relevant family histories (41). In the future, online questionnaires, similar to the Web-based Surgeon General's Family Health Portrait tool (42), could be developed.

Using a cluster randomized design reduced the risk for contamination between groups, whereas using matched pairs helped preserve power by reducing the imbalance between groups. We acknowledge that a relatively small proportion of less-educated persons and members of ethnic minority groups were recruited, and that higher self-reported levels of risk-reducing behavior at baseline may affect the generalizability of our study (43). Further, the intervention was not designed to change participant behavior. However, the potential for any effects on lifestyle was explored to inform future interventions (44). This had several limitations, including follow-up of only 6 months and use of self-reported information and data extracted from electronic records, both of which had missing values. Our study therefore lacks clinical outcome data.

No major or statistically significant changes in participants' risk-reducing behaviors were identified. Other studies (43, 45) have produced mixed findings: A combined familial risk assessment for cardiovascular diseases and cancer demonstrated moderate improvement in self-reported dietary intake and physical activity, whereas another study showed that participants reported a healthier diet after identification of familial diabetes risk. Although use of family history may assist in directing preventive efforts to persons at greatest cardiovascular risk, further research could assess whether more intensive interventions, including counseling that emphasized familial predisposition, would improve objectively measured risk-reducing behavior and clinical outcomes.

Compared with universal screening of untreated persons, recent modeling suggests that using a targeted strategy to identify about 60% of the population at highest risk could prevent almost all cardiovascular disease (8, 46). Our study shows that using systematic family history information increases the proportion of persons who can be identified as having the highest cardiovascular risk in the general primary care population. Although we did not compare targeted screening with universal screening, our findings highlight the promising role that greater use of systematic family history collection could play in a targeted strategy in primary care. This potentially low-cost approach also seems feasible in practice and is acceptable to patients.

Valdez R, Yoon PW, Qureshi N, Green RF, Khoury MJ.  Family history in public health practice: a genomic tool for disease prevention and health promotion. Annu Rev Public Health. 2010; 31.69-87 PubMed
 
Berg AO, Baird MA, Botkin JR, Driscoll DA, Fishman PA, Guarino PD. et al.  National Institutes of Health state-of-the-science conference statement: family history and improving health. Ann Intern Med. 2009; 151.872-7 PubMed
 
Wilson BJ, Qureshi N, Santaguida P, Little J, Carroll JC, Allanson J. et al.  Systematic review: family history in risk assessment for common diseases. Ann Intern Med. 2009; 151.878-85 PubMed
 
Anderson KM, Wilson PW, Odell PM, Kannel WB.  An updated coronary risk profile. A statement for health professionals. Circulation. 1991; 83.356-62 PubMed
 
British Cardiac Society.  JBS 2: Joint British Societies' guidelines on prevention of cardiovascular disease in clinical practice. Heart. 2005; 91.Suppl 5v1-52 PubMed
 
Haq IU, Ramsay LE, Yeo WW, Jackson PR, Wallis EJ.  Is the Framingham risk function valid for northern European populations? A comparison of methods for estimating absolute coronary risk in high risk men. Heart. 1999; 81.40-6 PubMed
 
National Cholesterol Education Program (NCEP) Expert Panel on Detection, Evaluation, and Treatment of High Blood Cholesterol in Adults (Adult Treatment Panel III).  Third report of the National Cholesterol Education Program (NCEP) expert panel on detection, evaluation, and treatment of high blood cholesterol in adults (adult treatment panel III) final report. Circulation. 2002; 106.3143-421 PubMed
 
Lawson KD, Fenwick EA, Pell AC, Pell JP.  Comparison of mass and targeted screening strategies for cardiovascular risk: simulation of the effectiveness, cost-effectiveness and coverage using a cross-sectional survey of 3921 people. Heart. 2010; 96.208-12 PubMed
 
Williams RR, Hunt SC, Heiss G, Province MA, Bensen JT, Higgins M. et al.  Usefulness of cardiovascular family history data for population-based preventive medicine and medical research (the Health Family Tree Study and the NHLBI Family Heart Study). Am J Cardiol. 2001; 87.129-35 PubMed
 
Qureshi N.  Exploring the Potential of the Family History to Identify Genetic Risk in Primary Care [doctoral thesis]. Nottingham, UK: Univ Nottingham; 2006.
 
De Sutter J, De Bacquer D, Kotseva K, Sans S, Pyörälä K, Wood D, et al. EUROpean Action on Secondary Prevention through Intervention to Reduce Events II study group.  Screening of family members of patients with premature coronary heart disease; results from the EUROASPIRE II family survey. Eur Heart J. 2003; 24.249-57 PubMed
 
Qureshi N, Bethea J, Modell B, Brennan P, Papageorgiou A, Raeburn S. et al.  Collecting genetic information in primary care: evaluating a new family history tool. Fam Pract. 2005; 22.663-9 PubMed
 
Frezzo TM, Rubinstein WS, Dunham D, Ormond KE.  The genetic family history as a risk assessment tool in internal medicine. Genet Med. 2003; 5.84-91 PubMed
 
Qureshi N, Armstrong S, Saukko P, Sach T, Middlemass J, Evans PH. et al.  Realising the potential of the family history in risk assessment and primary prevention of coronary heart disease in primary care: ADDFAM study protocol. BMC Health Serv Res. 2009; 9.184 PubMed
 
Greenland P, Alpert JS, Beller GA, Benjamin EJ, Budoff MJ, Fayad ZA, et al. American College of Cardiology Foundation/American Heart Association Task Force on Practice Guidelines.  2010 ACCF/AHA guideline for assessment of cardiovascular risk in asymptomatic adults: a report of the American College of Cardiology Foundation/American Heart Association Task Force on Practice Guidelines. Circulation. 2010; 122.584-636 PubMed
 
Woodward M, Brindle P, Tunstall-Pedoe H, SIGN group on risk estimation.  Adding social deprivation and family history to cardiovascular risk assessment: the ASSIGN score from the Scottish Heart Health Extended Cohort (SHHEC). Heart. 2007; 93.172-6 PubMed
 
Bhopal R, Fischbacher C, Vartiainen E, Unwin N, White M, Alberti G.  Predicted and observed cardiovascular disease in South Asians: application of FINRISK, Framingham and SCORE models to Newcastle Heart Project data. J Public Health (Oxf). 2005; 27.93-100 PubMed
 
Hall R, Saukko PM, Evans PH, Qureshi N, Humphries SE.  Assessing family history of heart disease in primary care consultations: a qualitative study. Fam Pract. 2007; 24.435-42 PubMed
 
Qureshi N, Standen PJ, Hapgood R, Hayes J.  A randomized controlled trial to assess the psychological impact of a family history screening questionnaire in general practice. Fam Pract. 2001; 18.78-83 PubMed
 
Marteau TM, Bekker H.  The development of a six-item short-form of the state scale of the Spielberger State-Trait Anxiety Inventory (STAI). Br J Clin Psychol. 1992; 31. (Pt 3) 301-6 PubMed
 
Roe L, Strong C, Whiteside C, Neil A, Mant D.  Dietary intervention in primary care: validity of the DINE method for diet assessment. Fam Pract. 1994; 11.375-81 PubMed
 
Prochaska JO, Velicer WF.  The transtheoretical model of health behavior change. Am J Health Promot. 1997; 12.38-48 PubMed
 
Doherty SC, Steptoe A, Rink E, Kendrick T, Hilton S.  Readiness to change health behaviours among patients at high risk of cardiovascular disease. J Cardiovasc Risk. 1998; 5.147-53 PubMed
 
Stiell IG, Grimshaw J, Wells GA, Coyle D, Lesiuk HJ, Rowe BH. et al.  A matched-pair cluster design study protocol to evaluate implementation of the Canadian C-spine rule in hospital emergency departments: phase III. Implement Sci. 2007; 2.4 PubMed
 
Stiell IG, Clement CM, Grimshaw J, Brison RJ, Rowe BH, Schull MJ. et al.  Implementation of the Canadian C-Spine Rule: prospective 12 centre cluster randomised trial. BMJ. 2009; 339.b4146 PubMed
 
Gail MH, Byar DP, Pechacek TF, Corle DK.  Aspects of statistical design for the Community Intervention Trial for Smoking Cessation (COMMIT). Control Clin Trials. 1992; 13.6-21 PubMed
 
Donner A, Klar N.  Design and Analysis of Cluster Randomized Trials in Health Research. New York: Oxford Univ Pr; 2000.100-8
 
Kaplan GA, Keil JE.  Socioeconomic factors and cardiovascular disease: a review of the literature. Circulation. 1993; 88.1973-98 PubMed
 
Calnan M, Cant S, Williams S, Killoran A.  Involvement of the primary health care team in coronary heart disease prevention. Br J Gen Pract. 1994; 44.224-8 PubMed
 
Ford AS, Ford WS.  Health education and the primary care physician: the practitioner's perspective. Soc Sci Med. 1983; 17.1505-12 PubMed
 
Qureshi N, Wilson B, Santaguida P, Little J, Carroll J, Allanson J. et al.  Family history and improving health. Evid Rep Technol Assess (Full Rep). 2009.1-135 PubMed
 
Qureshi N, Wilson B, Santaguida P, Carroll J, Allanson J, Culebro CR. et al.  Collection and use of cancer family history in primary care. Evid Rep Technol Assess (Full Rep). 2007.1-84 PubMed
 
O'Neill SM, Rubinstein WS, Wang C, Yoon PW, Acheson LS, Rothrock N, et al. Family Healthware Impact Trial group.  Familial risk for common diseases in primary care: the Family Healthware Impact Trial. Am J Prev Med. 2009; 36.506-14 PubMed
 
Maron DJ, Fazio S, Linton MF.  Current perspectives on statins. Circulation. 2000; 101.207-13 PubMed
 
Colhoun HM, Betteridge DJ, Durrington PN, Hitman GA, Neil HA, Livingstone SJ, et al. CARDS investigators.  Primary prevention of cardiovascular disease with atorvastatin in type 2 diabetes in the Collaborative Atorvastatin Diabetes Study (CARDS): multicentre randomised placebo-controlled trial. Lancet. 2004; 364.685-96 PubMed
 
Sever PS, Dahlöf B, Poulter NR, Wedel H, Beevers G, Caulfield M, et al. ASCOT investigators.  Prevention of coronary and stroke events with atorvastatin in hypertensive patients who have average or lower-than-average cholesterol concentrations, in the Anglo-Scandinavian Cardiac Outcomes Trial—Lipid Lowering Arm (ASCOT-LLA): a multicentre randomised controlled trial. Lancet. 2003; 361.1149-58 PubMed
 
Pickin DM, McCabe CJ, Ramsay LE, Payne N, Haq IU, Yeo WW. et al.  Cost effectiveness of HMG-CoA reductase inhibitor (statin) treatment related to the risk of coronary heart disease and cost of drug treatment. Heart. 1999; 82.325-32 PubMed
 
National Health Service.  Putting Prevention First—Vascular Checks: Risk Assessment and Management. London: UK Department of Health; 2008.
 
Cooper A, Nherera L, Calvert N, O'Flynn N, Turnbull N, Robson J. et al.  Clinical Guidelines and Evidence Review for Lipid Modification: Cardiovascular Risk Assessment and the Primary and Secondary Prevention of Cardiovascular Disease. London: National Collaborating Centre for Primary Care and Royal College of General Practitioners; 2008.
 
Qureshi N, Humphries SE, Seed M, Rowlands P, Minhas R, NICE Guideline Development Group.  Identification and management of familial hypercholesterolaemia: what does it mean to primary care? Br J Gen Pract. 2009; 59.773-6 PubMed
 
Feero WG, Bigley MB, Brinner KM, Family Health History Multi-Stakeholder Workgroup of the American Health Information Community.  New standards and enhanced utility for family health history information in the electronic health record: an update from the American Health Information Community's Family Health History Multi-Stakeholder Workgroup. J Am Med Inform Assoc. 2008; 15.723-8 PubMed
 
Guttmacher AE, Collins FS, Carmona RH.  The family history—more important than ever. N Engl J Med. 2004; 351.2333-6 PubMed
 
Ruffin MT 4th, Nease DE Jr, Sen A, Pace WD, Wang C, Acheson LS, et al. Family History Impact Trial (FHITr) Group.  Effect of preventive messages tailored to family history on health behaviors: the Family Healthware Impact Trial. Ann Fam Med. 2011; 9.3-11 PubMed
 
Campbell M, Fitzpatrick R, Haines A, Kinmonth AL, Sandercock P, Spiegelhalter D. et al.  Framework for design and evaluation of complex interventions to improve health. BMJ. 2000; 321.694-6 PubMed
 
Pijl M, Timmermans DR, Claassen L, Janssens AC, Nijpels G, Dekker JM. et al.  Impact of communicating familial risk of diabetes on illness perceptions and self-reported behavioral outcomes: a randomized controlled trial. Diabetes Care. 2009; 32.597-9 PubMed
 
Chamnan P, Simmons RK, Khaw KT, Wareham NJ, Griffin SJ.  Estimating the population impact of screening strategies for identifying and treating people at high risk of cardiovascular disease: modelling study. BMJ. 2010; 340.c1693 PubMed
 
Sheridan S, Pignone M, Mulrow C.  Framingham-based tools to calculate the global risk of coronary heart disease: a systematic review of tools for clinicians. J Gen Intern Med. 2003; 18.1039-52 PubMed
 
Jones AF, Walker J, Jewkes C, Game FL, Bartlett WA, Marshall T. et al.  Comparative accuracy of cardiovascular risk prediction methods in primary care patients. Heart. 2001; 85.37-43 PubMed
 
Appendix 1: The ADDFAM Study Group

The ADDFAM study group also includes Laura Cross-Bardell, Hannah Farrimond, Steve Humphries, Matthew Jones, and Tracey Sach.

Appendix 2: Identification of Asymptomatic Persons at Increased Risk for Cardiovascular Disease
Background on JBS2 Cardiovascular Risk Assessment

The JBS cardiovascular risk assessment guidelines have been widely disseminated in British primary care as cardiovascular disease risk prediction charts since their first iteration in 1998. The charts combine ease of use and good sensitivity and specificity (4748). However, the algorithm overestimates risk in younger persons (age <49 years) and has not been validated for additional risk factors for cardiovascular disease, such as ethnic minority ancestry and family history (5, 17). The charts are populated with objective core risk factors (age, sex, smoking status, systolic blood pressure, and total–HDL cholesterol ratio). As for other Framingham-based risk calculations, the JBS2 charts (5) categorize persons into 3 risk strata for cardiovascular disease: high risk (10-year risk ≥20%), moderate risk (10% to 19%), and average risk (<10%). The cardiovascular disease scores can be subjectively refined by physicians by taking other cardiovascular risk factors into account, such as family history, ethnicity, or triglyceride levels (5). On the basis of review of the epidemiologic data and consensus opinion, the JBS2 committee has recommended that the cardiovascular disease risk score can be multiplied by a factor of 1.5 for persons with a family history of premature CHD, defined as a male first-degree relative younger than 55 years or a female first-degree relative younger than 65 years (5). Although the practices used paper-based cardiovascular disease risk prediction charts, the study team used the standalone JBS2 cardiovascular risk calculator, based on the Framingham equation, to accurately incorporate family history information into the cardiovascular disease risk score.

How to Assess for Total Cardiovascular Disease Risk

(Adopted from reference (5).) A short history, focused clinical examination, and a blood sample provide a simple, quick, practical assessment of an asymptomatic person's total cardiovascular risk. Using the JBS cardiovascular risk prediction charts (5), you can estimate the probability (percentage) of developing cardiovascular disease over 10 years.

Which risk factors to measure: age (years), sex (male or female), current or former cigarette smoker (yes or no), systolic blood pressure (measure twice and use the mean), total nonfasting cholesterol level, nonfasting HDL cholesterol level, and total–HDL cholesterol ratio (assume 1.0 if no measurement available).

Calculate total cardiovascular disease risk from the cardiovascular risk prediction charts. Risk of 20% or greater over 10 years is defined as high risk and justifies professional lifestyle intervention and appropriate use of antithrombotic, antihypertensive, and lipid-lowering therapy.

Who Should Not Have Cardiovascular Disease Risk Calculated

(Adopted from reference (5).) Persons with atherosclerotic cardiovascular disease; persistently elevated blood pressure (≥160/100 mm Hg) or target organ damage due to hypertension; a total–HDL cholesterol ratio of 6 or greater; type 1 or 2 diabetes mellitus; renal dysfunction, including diabetic nephropathy; or familial hypercholesterolemia, familial combined hyperlipidemia, or other inherited dyslipidemia are at sufficiently high risk to justify professional lifestyle intervention and appropriate treatment and do not need formal risk calculation.

Tables

Table Jump Placeholder Table 1. Characteristics of Study Practices at Baseline
Table Jump Placeholder Table 2. Characteristics of Participants at Baseline
Table Jump Placeholder Table 3. Proportion of Patients With High CVD Risk
Table Jump Placeholder Table 4. Prevalence of Family History of CHD From Electronic Health Records and Family History Questionnaires
Table Jump Placeholder Appendix Table 1. Secondary Outcome Variables at Baseline and 6-Month Follow-up for Participants Who Completed Baseline and 6-Month Questionnaires
Table Jump Placeholder Appendix Table 2. Difference in Secondary Outcome Measures Between Practices in Intervention and Control Groups

References

Valdez R, Yoon PW, Qureshi N, Green RF, Khoury MJ.  Family history in public health practice: a genomic tool for disease prevention and health promotion. Annu Rev Public Health. 2010; 31.69-87 PubMed
 
Berg AO, Baird MA, Botkin JR, Driscoll DA, Fishman PA, Guarino PD. et al.  National Institutes of Health state-of-the-science conference statement: family history and improving health. Ann Intern Med. 2009; 151.872-7 PubMed
 
Wilson BJ, Qureshi N, Santaguida P, Little J, Carroll JC, Allanson J. et al.  Systematic review: family history in risk assessment for common diseases. Ann Intern Med. 2009; 151.878-85 PubMed
 
Anderson KM, Wilson PW, Odell PM, Kannel WB.  An updated coronary risk profile. A statement for health professionals. Circulation. 1991; 83.356-62 PubMed
 
British Cardiac Society.  JBS 2: Joint British Societies' guidelines on prevention of cardiovascular disease in clinical practice. Heart. 2005; 91.Suppl 5v1-52 PubMed
 
Haq IU, Ramsay LE, Yeo WW, Jackson PR, Wallis EJ.  Is the Framingham risk function valid for northern European populations? A comparison of methods for estimating absolute coronary risk in high risk men. Heart. 1999; 81.40-6 PubMed
 
National Cholesterol Education Program (NCEP) Expert Panel on Detection, Evaluation, and Treatment of High Blood Cholesterol in Adults (Adult Treatment Panel III).  Third report of the National Cholesterol Education Program (NCEP) expert panel on detection, evaluation, and treatment of high blood cholesterol in adults (adult treatment panel III) final report. Circulation. 2002; 106.3143-421 PubMed
 
Lawson KD, Fenwick EA, Pell AC, Pell JP.  Comparison of mass and targeted screening strategies for cardiovascular risk: simulation of the effectiveness, cost-effectiveness and coverage using a cross-sectional survey of 3921 people. Heart. 2010; 96.208-12 PubMed
 
Williams RR, Hunt SC, Heiss G, Province MA, Bensen JT, Higgins M. et al.  Usefulness of cardiovascular family history data for population-based preventive medicine and medical research (the Health Family Tree Study and the NHLBI Family Heart Study). Am J Cardiol. 2001; 87.129-35 PubMed
 
Qureshi N.  Exploring the Potential of the Family History to Identify Genetic Risk in Primary Care [doctoral thesis]. Nottingham, UK: Univ Nottingham; 2006.
 
De Sutter J, De Bacquer D, Kotseva K, Sans S, Pyörälä K, Wood D, et al. EUROpean Action on Secondary Prevention through Intervention to Reduce Events II study group.  Screening of family members of patients with premature coronary heart disease; results from the EUROASPIRE II family survey. Eur Heart J. 2003; 24.249-57 PubMed
 
Qureshi N, Bethea J, Modell B, Brennan P, Papageorgiou A, Raeburn S. et al.  Collecting genetic information in primary care: evaluating a new family history tool. Fam Pract. 2005; 22.663-9 PubMed
 
Frezzo TM, Rubinstein WS, Dunham D, Ormond KE.  The genetic family history as a risk assessment tool in internal medicine. Genet Med. 2003; 5.84-91 PubMed
 
Qureshi N, Armstrong S, Saukko P, Sach T, Middlemass J, Evans PH. et al.  Realising the potential of the family history in risk assessment and primary prevention of coronary heart disease in primary care: ADDFAM study protocol. BMC Health Serv Res. 2009; 9.184 PubMed
 
Greenland P, Alpert JS, Beller GA, Benjamin EJ, Budoff MJ, Fayad ZA, et al. American College of Cardiology Foundation/American Heart Association Task Force on Practice Guidelines.  2010 ACCF/AHA guideline for assessment of cardiovascular risk in asymptomatic adults: a report of the American College of Cardiology Foundation/American Heart Association Task Force on Practice Guidelines. Circulation. 2010; 122.584-636 PubMed
 
Woodward M, Brindle P, Tunstall-Pedoe H, SIGN group on risk estimation.  Adding social deprivation and family history to cardiovascular risk assessment: the ASSIGN score from the Scottish Heart Health Extended Cohort (SHHEC). Heart. 2007; 93.172-6 PubMed
 
Bhopal R, Fischbacher C, Vartiainen E, Unwin N, White M, Alberti G.  Predicted and observed cardiovascular disease in South Asians: application of FINRISK, Framingham and SCORE models to Newcastle Heart Project data. J Public Health (Oxf). 2005; 27.93-100 PubMed
 
Hall R, Saukko PM, Evans PH, Qureshi N, Humphries SE.  Assessing family history of heart disease in primary care consultations: a qualitative study. Fam Pract. 2007; 24.435-42 PubMed
 
Qureshi N, Standen PJ, Hapgood R, Hayes J.  A randomized controlled trial to assess the psychological impact of a family history screening questionnaire in general practice. Fam Pract. 2001; 18.78-83 PubMed
 
Marteau TM, Bekker H.  The development of a six-item short-form of the state scale of the Spielberger State-Trait Anxiety Inventory (STAI). Br J Clin Psychol. 1992; 31. (Pt 3) 301-6 PubMed
 
Roe L, Strong C, Whiteside C, Neil A, Mant D.  Dietary intervention in primary care: validity of the DINE method for diet assessment. Fam Pract. 1994; 11.375-81 PubMed
 
Prochaska JO, Velicer WF.  The transtheoretical model of health behavior change. Am J Health Promot. 1997; 12.38-48 PubMed
 
Doherty SC, Steptoe A, Rink E, Kendrick T, Hilton S.  Readiness to change health behaviours among patients at high risk of cardiovascular disease. J Cardiovasc Risk. 1998; 5.147-53 PubMed
 
Stiell IG, Grimshaw J, Wells GA, Coyle D, Lesiuk HJ, Rowe BH. et al.  A matched-pair cluster design study protocol to evaluate implementation of the Canadian C-spine rule in hospital emergency departments: phase III. Implement Sci. 2007; 2.4 PubMed
 
Stiell IG, Clement CM, Grimshaw J, Brison RJ, Rowe BH, Schull MJ. et al.  Implementation of the Canadian C-Spine Rule: prospective 12 centre cluster randomised trial. BMJ. 2009; 339.b4146 PubMed
 
Gail MH, Byar DP, Pechacek TF, Corle DK.  Aspects of statistical design for the Community Intervention Trial for Smoking Cessation (COMMIT). Control Clin Trials. 1992; 13.6-21 PubMed
 
Donner A, Klar N.  Design and Analysis of Cluster Randomized Trials in Health Research. New York: Oxford Univ Pr; 2000.100-8
 
Kaplan GA, Keil JE.  Socioeconomic factors and cardiovascular disease: a review of the literature. Circulation. 1993; 88.1973-98 PubMed
 
Calnan M, Cant S, Williams S, Killoran A.  Involvement of the primary health care team in coronary heart disease prevention. Br J Gen Pract. 1994; 44.224-8 PubMed
 
Ford AS, Ford WS.  Health education and the primary care physician: the practitioner's perspective. Soc Sci Med. 1983; 17.1505-12 PubMed
 
Qureshi N, Wilson B, Santaguida P, Little J, Carroll J, Allanson J. et al.  Family history and improving health. Evid Rep Technol Assess (Full Rep). 2009.1-135 PubMed
 
Qureshi N, Wilson B, Santaguida P, Carroll J, Allanson J, Culebro CR. et al.  Collection and use of cancer family history in primary care. Evid Rep Technol Assess (Full Rep). 2007.1-84 PubMed
 
O'Neill SM, Rubinstein WS, Wang C, Yoon PW, Acheson LS, Rothrock N, et al. Family Healthware Impact Trial group.  Familial risk for common diseases in primary care: the Family Healthware Impact Trial. Am J Prev Med. 2009; 36.506-14 PubMed
 
Maron DJ, Fazio S, Linton MF.  Current perspectives on statins. Circulation. 2000; 101.207-13 PubMed
 
Colhoun HM, Betteridge DJ, Durrington PN, Hitman GA, Neil HA, Livingstone SJ, et al. CARDS investigators.  Primary prevention of cardiovascular disease with atorvastatin in type 2 diabetes in the Collaborative Atorvastatin Diabetes Study (CARDS): multicentre randomised placebo-controlled trial. Lancet. 2004; 364.685-96 PubMed
 
Sever PS, Dahlöf B, Poulter NR, Wedel H, Beevers G, Caulfield M, et al. ASCOT investigators.  Prevention of coronary and stroke events with atorvastatin in hypertensive patients who have average or lower-than-average cholesterol concentrations, in the Anglo-Scandinavian Cardiac Outcomes Trial—Lipid Lowering Arm (ASCOT-LLA): a multicentre randomised controlled trial. Lancet. 2003; 361.1149-58 PubMed
 
Pickin DM, McCabe CJ, Ramsay LE, Payne N, Haq IU, Yeo WW. et al.  Cost effectiveness of HMG-CoA reductase inhibitor (statin) treatment related to the risk of coronary heart disease and cost of drug treatment. Heart. 1999; 82.325-32 PubMed
 
National Health Service.  Putting Prevention First—Vascular Checks: Risk Assessment and Management. London: UK Department of Health; 2008.
 
Cooper A, Nherera L, Calvert N, O'Flynn N, Turnbull N, Robson J. et al.  Clinical Guidelines and Evidence Review for Lipid Modification: Cardiovascular Risk Assessment and the Primary and Secondary Prevention of Cardiovascular Disease. London: National Collaborating Centre for Primary Care and Royal College of General Practitioners; 2008.
 
Qureshi N, Humphries SE, Seed M, Rowlands P, Minhas R, NICE Guideline Development Group.  Identification and management of familial hypercholesterolaemia: what does it mean to primary care? Br J Gen Pract. 2009; 59.773-6 PubMed
 
Feero WG, Bigley MB, Brinner KM, Family Health History Multi-Stakeholder Workgroup of the American Health Information Community.  New standards and enhanced utility for family health history information in the electronic health record: an update from the American Health Information Community's Family Health History Multi-Stakeholder Workgroup. J Am Med Inform Assoc. 2008; 15.723-8 PubMed
 
Guttmacher AE, Collins FS, Carmona RH.  The family history—more important than ever. N Engl J Med. 2004; 351.2333-6 PubMed
 
Ruffin MT 4th, Nease DE Jr, Sen A, Pace WD, Wang C, Acheson LS, et al. Family History Impact Trial (FHITr) Group.  Effect of preventive messages tailored to family history on health behaviors: the Family Healthware Impact Trial. Ann Fam Med. 2011; 9.3-11 PubMed
 
Campbell M, Fitzpatrick R, Haines A, Kinmonth AL, Sandercock P, Spiegelhalter D. et al.  Framework for design and evaluation of complex interventions to improve health. BMJ. 2000; 321.694-6 PubMed
 
Pijl M, Timmermans DR, Claassen L, Janssens AC, Nijpels G, Dekker JM. et al.  Impact of communicating familial risk of diabetes on illness perceptions and self-reported behavioral outcomes: a randomized controlled trial. Diabetes Care. 2009; 32.597-9 PubMed
 
Chamnan P, Simmons RK, Khaw KT, Wareham NJ, Griffin SJ.  Estimating the population impact of screening strategies for identifying and treating people at high risk of cardiovascular disease: modelling study. BMJ. 2010; 340.c1693 PubMed
 
Sheridan S, Pignone M, Mulrow C.  Framingham-based tools to calculate the global risk of coronary heart disease: a systematic review of tools for clinicians. J Gen Intern Med. 2003; 18.1039-52 PubMed
 
Jones AF, Walker J, Jewkes C, Game FL, Bartlett WA, Marshall T. et al.  Comparative accuracy of cardiovascular risk prediction methods in primary care patients. Heart. 2001; 85.37-43 PubMed
 

Letters

NOTE:
Citing articles are presented as examples only. In non-demo SCM6 implementation, integration with CrossRef’s "Cited By" API will populate this tab (http://www.crossref.org/citedby.html).

Comments

Submit a Comment
Easy math is not necessarily correct.
Posted on February 26, 2012
David C., Goff, Professor and Chair, Department of Epidemiology and Prevention
Wake Forest School of Medicine
Conflict of Interest: None Declared

I read with concern the article by Qureshi and colleagues. Whereas the clinical trial methodology is admirable, the scientific questions posed are not helpful to understanding the role of data on family history in improving risk assessment. On a superficial level, the idea that risk can only be upgraded is incorrect. If a positive family history is used to upgrade risk, a negative family history should be used to downgrade risk. Otherwise, the calibration of the underlying model is ignored. On a more substantive level, this study provides no evidence regarding whether data on family history provides value by improving model discrimination or calibration, or whether any reclassification is actually correct. As noted by the authors, this paper also does not address patient behaviors, treatments, or outcomes. These results do not support the incorporation of assessments of family history into routine clinical practice. We need results documenting improved discrimination, calibration, and correct reclassification. Easy math is not necessarily correct.

Conflict of Interest:

None declared

Family History already included in QRISK which is used widely in the NHS in the UK
Posted on February 28, 2012
John, Robson, Senior lecturer, Julia Hippisley-Cox, Peter Brindle
Centre for Health Sciences, Queen Mary's School of Medicine and Dentistry, London E1 2AT
Conflict of Interest: None Declared

We agree with the authors that a positive family history of ischaemic heart disease under age 60 years in a first degree relative has a marked impact on future risk of major cardiovascular [1]. In in the UK, we developed a new cardiovascular risk prediction algorithm QRISK[2] derived from 10.9 million person years and established that for both men and women, in all major ethnic groups, and at all levels of social deprivation, that a positive family history is associated with an increased risk of cardiovascular disease. On multivariate analysis, the magnitude of this increased risk varies with age from a nearly three-fold increase in patients aged 35-40 to a 30% increase among those aged 70-75 years which is similar to findings reported in other multi-ethnic studies[3]. QRISK (www.qrisk.org) has been externally validated by an independent team[4] and is included in national guidelines[5] and routinely used to assess cardiovascular risk as part of the NHS health Checks programme aiming to assess risk in some 15 million people nationally in England. Of these approximately 10%, 1 million people, will have positive family history. ASSIGN similarly incorporates family history in Scotland www.assign-score.com . Around 1 million people in England will have a 10yr risk of cardiovascular disease of 20% or more, for which positive family history will be a contributing factor. For these people current guidance recommends statins together with lifestyle change and antihypertensives where appropriate

REFERENCES

1. Qureshi N, Armstrong S, Dhiman P, Saukko P, Middlemass J, Evans PH, et al. Effect of Adding Systematic Family History Enquiry to Cardiovascular Disease Risk Assessment in Primary Care. Annals of Internal Medicine 2012;156(4):253-62.

2. Hippisley-Cox J, Coupland C, Vinogradova Y, Robson J, Minhas R, Sheikh A, et al. Predicting cardiovascular risk in England and Wales: prospective derivation and validation of QRISK2. BMJ 2008:bmj.39609.449676.25.

3. Nasir K, Budoff MJ, Wong ND, Scheuner M, Herrington D, Arnett DK, et al. Family history of premature coronary heart disease and coronary artery calcification: Multi-Ethnic Study of Atherosclerosis (MESA). Circulation 2007;116(6):619-26.

4. Collins GS, Altman DG. An independent and external validation of QRISK2 cardiovascular disease risk score: a prospective open cohort study. BMJ 2010;340:c2442-.

5. National Institute for Clinical Excellence. Lipid modification - Cardiovascular risk assessment and the modification of blood lipids for the primary and secondary prevention of cardiovascular disease. In: NICE, editor. London: NICE, 2008.

Conflict of Interest:

JR and PB were previously members of the NICE Guideline Development Group for Lipid Modification of which JR was chair. JHC is professor of clinical epidemiology at the University of Nottingham and co-director of QResearch? - a not-for-profit organisation which is a joint partnership between the University of Nottingham and EMIS (leading commercial supplier of IT for 60% of general practices in the UK). JHC is also director of ClinRisk Ltd which produces open and closed source software to ensure the reliable and updatable implementation of clinical risk algorithms. This work and any views expressed within it are solely those of the co-authors and not of any affiliated bodies or organisations. There are no other relationships or activities that could appear to have influenced the submitted work.

Joint British Societies' Recommendations for Cardiovascular Disease Prevention: clarification
Posted on April 10, 2012
Paul N., Durrington, Professor of Medicine
University of Manchester
Conflict of Interest: None Declared

John Robson and colleagues are right to point out that family history has a greater influence on cardiovascular disease (CVD) risk in younger people. Thus, incorporating it into CVD risk assessment by simply multiplying by a single factor, as in the interesting article by Qureshi and colleagues, is not ideal. It is ironic, however, that the QRISK2 algorithm may not lead to intervention which is as effective as the Joint British Societies' (JBS) recommendations [1]. The validation to which John Robson et al refer [2] was not reassuring. In particular, the intervention group identified by QRISK2 had a mean CVD risk of 25% over the next 10 years whereas people with a 20% risk were targeted by the Framingham-based JBS2 assessment, exactly the level recommended for treatment with statins. The difference between the two approaches is exemplified by a man aged 68 having a serum cholesterol of 200mg/dL, LDL cholesterol 120mg/dL, HDL cholesterol 55 mg/dl, serum triglycerides 125 mg/dL, systolic blood pressure 160mmHg, height 172cm and body weight 90kg, whose 10-year risk is 20% by QRISK2 and 25% by JBS2. Prescribing a statin to achieve an LDL cholesterol target of 80mg/dL, will decrease his 10-year risk by one fifth (one fifth for each 40mg/dL decrease in LDL cholesterol [3]). A younger man aged 60 living in the same location with a much higher serum cholesterol of 280mg/dL and LDL cholesterol of 200mg/dL, but similar HDL cholesterol, triglycerides, height and weight, will have a 10-year risk of 16% by QRISK2 and 20% by JBS2. He would not receive a statin by QRISK2, but would according to JBS2. However, reducing his LDL cholesterol to 80mg/dL will decrease his cardiovascular risk by three fifths. Because examples such as these are commonplace, JBS2 will have a substantially greater impact in decreasing cardiovascular disease incidence in the general population. From the QRISK2 authors' own report [4] we know it identifies fewer young people than Framingham, but more elderly ones as at >20% 10-year risk. Whilst the concept of targeting younger people at high lifetime risk rather than waiting for them to cross a single threshold of risk is currently controversial [5], there is certainly no suggestion that we should adopt higher thresholds of risk for intervention in younger people and it is to be hoped the future iterations of QRISK2 will correct this problem.

REFERENCES.

1. Manuel D.G, Kwong K, Tanuseputro P, Lim J, Mustard C.A, Anderson G.M, et al. Effectiveness and efficiency of different guidelines on statin treatment for preventing deaths from coronary heart disease: modelling study BMJ 2006; 332: 1419-23.

2. Collins GS, Altman DG. An independent and external validation of QRISK2 cardiovascular disease risk score: a prospective open cohort study. BMJ 2010; 340: c2442.

3. Cholesterol Treatment Trialists' (CTT) Collaboration, Baigent C, Blackwell L, Emberson J, Holland LE, Reith C, Bhala N, et al. Efficacy and safety of more intensive lowering of LDL cholesterol: a meta-analysis of data from 170,000 participants in 26 randomised trials. Lancet 2010; 376: 1670-81.

4. Hippisley-Cox J, Coupland C, Vinogradova Y, Robson J, Minhas R, Sheikh A, et al. Predicting cardiovascular risk in England and Wales: prospective derivation and validation of QRISK2. BMJ 2008; 336: 1475-82.

5. Liew SM, Jackson R, Glasziou P. Should identical CVD risks in young and old patients be managed identically? Results from two models. BMJ Open 2012; 2: e000728.doi:10.1136/bmjopen-2011-000728 Accessed 4/10/12

Conflict of Interest:

Paul Durrington was a member of the JBS Committee developing recommendations for CVD prevention. His employer, the University of Manchester, holds the copyrights for the JBS2 charts and the JBS2 Cardiovascular Risk Assessor computer program, but does so not for profit, but to prevent their commercial exploitation. They are freely available to health professionals.

Authors' response
Posted on April 24, 2012
Nadeem, Qureshi, Professor, Paula Dhiman and Joe Kai
University of Nottingham
Conflict of Interest: None Declared

We agree with Goff that further study of the discriminatory accuracy of positive and negative family history in cardiovascular disease risk assessment algorithms is needed. The communication from Robson et al and other work (1) provide relevant information here. However, our research (2) had a clinical rather than epidemiological aim and so addressed different questions. We were concerned with clinical utility. We assessed the feasibility and impact of systematically collecting family history in comparison to its usual adhoc collection, and its subsequent use in cardiovascular risk assessment in family practice (3). Our paper offers the highest level of evidence for this sort of clinical intervention by use of a cluster randomised trial.

Robson et al provide helpful contextual information about the QRisk2 tool in the UK. On commencing our trial in 2007, the most widely adopted cardiovascular risk assessment tool in English family practice was the JBS tool as described and used in our study (2). This tool remains in the current British National Formulary, issued for use by all UK physicians in primary or secondary care. As the JBS tool uses the Framingham algorithm, which is employed in other developed countries, we hope our study will have wider international resonance, including in the US where it originated.

As noted in our paper, midway through our study, the potential value of family history became more prominent in the UK and we cite guidelines to which Robson et al also refer (4). In addition to JBS, this guidance, reissued with further information in 2010, alerts clinicians in the UK that they have an increasing choice of other risk assessment tools such as QRisk2, and notes the considerable debate about their relative merits (4). Whatever assessment tool is applied, one major challenge in actual practice remains the same: capturing and using authentic family history data. As Berg suggests in his editorial (5), our trial highlights the promise of a potentially low cost and feasible intervention to realise this in primary care practice.

It is encouraging there continues to be interest in the role of family history in cardiovascular risk assessment, and we look forward to further research in this field.

Nadeem Qureshi, Paula Dhiman and Joe Kai

References

1. Woodward M, Brindle P, Tunstall-Pedoe H, SIGN group on risk estimation. Adding social deprivation and family history to cardiovascular risk assessment: the ASSIGN score from the Scottish Heart Health Extended Cohort (SHHEC). Heart 2007; 93:172-6

2. Qureshi N, Armstrong S, Dhiman P, Saukko P, Middlemass J, Evans PH, Kai J. Effect of Adding Systematic Family History Enquiry to Cardiovascular Disease Risk Assessment in Primary Care. Ann Intern Med 2012; 156: 253-262.

3.Valdez R, Yoon PW, Qureshi N, Green RF, Khoury MJ. Family History in Public Health Practice: A Genomic Tool for Disease Prevention and Health Promotion. Annu. Rev. Public Health 2010; 31:69-87

4. Cooper A, Nherera L, Calvert N, O'Flynn N, Turnbull N, Robson J, et al. Clinical Guidelines and Evidence Review for Lipid Modification: Cardiovascular Risk Assessment and the Primary and Secondary Prevention of Cardiovascular Disease. London: National Collaborating Centre for Primary Care and Royal College of General Practitioners, 2008 (reissued March 2010).

5. Berg AO. Family History Gets a Boost. Ann Intern Med 2012; 156: 315-316.

Conflict of Interest:

None declared

Submit a Comment

Summary for Patients

Clinical Slide Sets

Terms of Use

The In the Clinic® slide sets are owned and copyrighted by the American College of Physicians (ACP). All text, graphics, trademarks, and other intellectual property incorporated into the slide sets remain the sole and exclusive property of the ACP. The slide sets may be used only by the person who downloads or purchases them and only for the purpose of presenting them during not-for-profit educational activities. Users may incorporate the entire slide set or selected individual slides into their own teaching presentations but may not alter the content of the slides in any way or remove the ACP copyright notice. Users may make print copies for use as hand-outs for the audience the user is personally addressing but may not otherwise reproduce or distribute the slides by any means or media, including but not limited to sending them as e-mail attachments, posting them on Internet or Intranet sites, publishing them in meeting proceedings, or making them available for sale or distribution in any unauthorized form, without the express written permission of the ACP. Unauthorized use of the In the Clinic slide sets will constitute copyright infringement.

Toolkit

Want to Subscribe?

Learn more about subscription options

Advertisement
Related Articles
Topic Collections
PubMed Articles

Want to Subscribe?

Learn more about subscription options

Forgot your password?
Enter your username and email address. We'll send you a reminder to the email address on record.
(Required)
(Required)