Self-Rated Cardiovascular Risk and 15-Year Cardiovascular Mortality

In the article “Self-Rated Cardiovascular Risk and 15-Year Cardiovascular Mortality” Gramling et al. (1) found statistically significant lower hazard ratios of CVD mortality among men rating their CVD risk to be lower-than-average risk for men at their age. Data did, however, not support such conclusions among female participants. The study contributes with very interesting new knowledge to Family Medicine concerning the usefulness of measuring self-rated CVD risk. The result that men who had an optimistic perception of their risk had an advantage would be interesting to elaborate further, e.g. among those who actually would have been diagnosed as healthy and at risk of CVD by a physician. Idler et al. (2) found, in a study of self-rated health (SRH) and mortality among participants in NHANES, that healthy participants who rated their health poor or fair as compared to excellent had increased mortality risk. However, among participants diagnosed by a physician as suffering from circulatory system disease the risk was only increased among those reporting a history of circulatory disease, but not among those who did not report any history of circulatory disease. The latter may show that the effect of self-reported health on mortality – and maybe also the effect of self-reported CVD – could reflect the attitude towards having a chronic condition. Taking the above knowledge into account, the findings of Gramling et al. call for further research into whether it is the respondents’ knowledge of own health condition at baseline that explains the effect of self-rated CVD risk on mortality or whether it is the respondents’ subsequent health related behaviour / attitude during the observation period. Furthermore, we need to clarify in which group of patients a self-rated CVD risk measurement provide us with useful information not gained by lab tests or by clinical examination. Research into these areas is needed before we implement the question of SRR in the clinical consultation.

The findings of Gramling et al. underline the fact that having high scores in SRH and/or having a perception of one’s cardiovascular risk as being low somehow decreases mortality in certain groups of patients. Further research may give us a better understanding of this exciting area and possible source of good health for our patients. Until we unravel this mystery, which presumably will take some time or perhaps never happen, we, as clinicians, would like to point to the still underestimated value of patient empowerment through a dialogue based on a salutogenic perspective on disease and health (3). By activating and using patients’ self-assessed health resources, unlocked by the use of key- questions in the clinical encounter (4, 5), we might just activate those health preserving mechanisms that account for the effect shown in Gramling et al.’s magnificent paper.

(1) Gramling R, Klein W, Roberts M, Waring ME, Gramling D, Eaton CB. Self-rated cardiovascular risk and 15-year cardiovascular mortality. Ann Fam Med 2008; 6(4):302-306.

(2) Idler E, Leventhal H, McLaughlin J, Leventhal E. In sickness but not in health: self-ratings, identity, and mortality. J Health Soc Behav 2004; 45(3):336-356.

(3) Hollnagel H, Malterud K. From risk factors to health resources in medical practice. Med Health Care and Phil 2000;3:257-64.

(4) Malterud K, Hollnagel H. Talking with Women About Personal Health Resources in General Practice. Key Questions about Salutogenesis. Scand J Prim Health Care 1998; 16: 66-71

(5) Sångren H, Nielsen ABS. Ressourcetænkning og selvvurderede helbredsressourcer. Månedskr Prakt Lægegern 2008; 8: 809-18.

Competing interests:   None declared

Gramling et al. report that lower self-ratings of cardiovascular disease (CVD) risk are independently associated with lower rates of CVD death among men (1). This finding highlights the potential impact of yet undefined factors on the prediction of CVD events. Although the authors acknowledge important limitations, we suggest that physician perception on patient CVD risk and its potential influence on patient outcome should be discussed and incorporated in future work.

Indeed, while individuals may sometime rate their CVD risk independently from their physician, it is likely – and expected – that most of the time, "self-rated" CVD risk is influenced by physician perception and expectations. Physicians and patients are building and sharing an "our" self-rated risk, in the same manner as they share medical decision-making. Although fundamental, the contribution of physicians in risk elaboration has been overlooked. This is unfortunate since physician optimism may by itself improve patient outcome (2). This phenomenon has been referred to as the "curabo effect" and has been described in chronic diseases (2,3,4).

Physician expectations are certainly based on CVD risk factors and on the Framingham Risk Score at baseline, for which the authors adjusted. However, physician expectations and therefore "self-rated" risk are also influenced by the strength of the working alliance and the quality of the treatment alliance, which are neither captured nor adjusted for using actual baseline risk. Would physician and patient rate CVD risk independently from the level of compliance with cardioprotective drugs, the magnitude of change in lipid levels, and the effort to quit smoking? Whether the strength of the working alliance and the quality of the treatment alliance can explain in part the optimistically-biased risk perception found and/or the inconsistencies of findings between men and women is unknown, but data should be collected and the association adjusted accordingly.

We suggest that investigators should attempt to gather information on physician expectations and the strength of the working alliance while trying to assess the long-term impact of "self-rated" risk.

1. Gramling R, Klein W, Roberts M, Waring ME, Gramling D, Eaton CB. Self-rated cardiovascular risk and 15-year cardiovascular mortality. Ann Fam Med. 2008;6(4):302-6.
2. Graz B, Wietlisbach V, Porchet F, Vader JP. Prognosis or "curabo effect?": physician prediction and patient outcome of surgery for low back pain and sciatica. Spine. 2005;30(12):1448-52.
3. Horvath AO. The therapeutic relationship: from transference to alliance. J Clin Psychol. 2000;56(2):163-73.
4. Priebe S, Gruyters T. The importance of the first three days: predictors of treatment outcome in depressed in-patients. Br J Clin Psychol. 1995;34(pt 2):229-36.

Competing interests:   None declared

The relationship between perceived risk and subsequent illness and mortality is compelling and also mysterious. In cancer settings, for example, patients with advanced cancer who overestimate their prognoses (by comparison to published reports) tend to live longer, although not as long as they think (or, obviously not as long as they would like). They tend to accept more aggressive treatment, but aggressive treatment does not affect longevity – it is not a mediating factor.1 It seems that the only people who accurately estimate their prognoses are those who are depressed.2 We know that depression shortens life, not only via suicide, but probably also because of activation of cytokines, increased cardiovascular reactivity and other processes. Or, put another way, there is a cognitive neuroimmunology of hope, which is only beginning to be explored in laboratory and real life settings.

Self-rating of CVD risk is similar is many ways to self-rating of health. Self-rated health is based on an implied comparison with peers, it is affected by dispositional attitude, and is powerfully and related to mortality, independently of many measures of morbidity.3,4 A key question for future research is how are these constructs related. For example, would self rating of CVD risk predict mortality independently on self- rated health?

One conclusion that can be drawn from studies in cancer, and Gramling’s compelling study of risk perception in cardiovascular disease, is that some degree of unrealistic optimism is healthy.5 But, is the optimism itself healthy, or simply a marker for other factors, such as depression, social support, a sense of meaning, or anxiety? Given that we, as clinicians, want to encourage behavior change, how can we communicate enough accurate information so that patients can make informed decisions and not feel overwhelmed? How can we present the information in ways that enhance rather than squelch motivation for self-care? The literature on risk communication may provide some clues. Patients vary in their information needs and information processing styles. So, one might initiate discussions about risk with “What would you like to know about ….?”. Then clinicians can follow up with questions about format of the presentation of risk; many patients prefer qualitative estimates to numbers, but a significant proportion do like numerical estimates, preferably in terms of absolute rather than relative risk.6 Other patients prefer graphics, and among graphical displays, some formats are preferred over others. Motivation is enhanced by a sense of personal autonomy and control, feeling that one can accomplish desired ends (self-efficacy) and healing relationships.7 Another lesson that we learn from Gramling’s data is that comparative perceptions (“I have greater risk than others”) may not be as motivating as perceptions that one is not different from others with similar risk factors. Thus, among men with high cholesterol, rather than emphasizing the absolute or relative level of risk compared to the general population, perhaps the message should be, “There are many others in your situation, and many of them have been able to lower their cholesterol and live longer.”

These questions can be studied in an organized fashion. Focus groups, conjoint analyses and video vignettes can evince from patients their emotional and cognitive responses to different ways of presenting messages. Madison Avenue uses these approaches all the time to “inform” and motivate behavior changes with considerable “success”. In health communication, we have been behind the times in terms of the sophistication of our health messages, both within the patient-physician relationship as well as utilizing the mass media. Gramling’s study provides important information that can guide these future efforts.

(1) Weeks JC, Cook EF, O'Day SJ et al. Relationship between cancer patients' predictions of prognosis and their treatment preferences. JAMA. 1998;279:1709-1714.

(2) Parker SM, Clayton JM, Hancock K et al. A systematic review of prognostic/end-of-life communication with adults in the advanced stages of a life-limiting illness: patient/caregiver preferences for the content, style, and timing of information. [Review] [63 refs]. Journal of Pain & Symptom Management. 2007;34:81-93.

(3) Idler EL, Benyamini Y. Self-rated health and mortality: a review of twenty-seven community studies. [Review] [95 refs]. Journal of Health & Social Behavior. 1997;38:21-37.

(4) Winter L, Lawton MP, Langston CA, Ruckdeschel K, Sando R. Symptoms, affects, and self-rated health: evidence for a subjective trajectory of health. Journal of Aging & Health. 2007;19:453-469.

(5) Taylor SE, Brown JD. Illusion and well-being: a social psychological perspective on mental health. [Review] [235 refs]. Psychological Bulletin. 1988;103:193-210.

(6) Epstein RM, Alper BS, Quill TE. Communicating evidence for participatory decision making. JAMA. 2004;291:2359-2366.

(7) Williams GC, Rodin GC, Ryan RM, Grolnick WS, Deci EL. Autonomous regulation and long-term medication adherence in adult outpatients. Health Psychol. 1998;17:269-276.

Competing interests:   None declared

Fascinating study! Gramling et al interpret these novel findings in the context of minimization of fear-induced response, presumably less activation of the sympathetic nervous system.1 They cite previous literature in support of their hypothesis. It is consistent with the finding that patients with the least favorable cholesterol tests and worse cardiovascular risk scores are most likely to recall their tests and risk as more favorable than assessed.2

A related explanation is that self-rated risk represents the flip side of self-rated health. There is a large and robust literature documenting that self-rated health predicts mortality independently of many objective measures of health.3,4 Self-rated health is among the most powerful predictors of mortality; persons who rate their health as poor have double the mortality of those rating their health as excellent after controlling for a range of health predictors.4 Poor self rated health is related to circulating cytokine levels,5 personality and negative affect,6 SES,6 and disease labeling.7

While explanations for the relationship between self-rated health and mortality are poorly understood, these findings suggest that people have an intuitive sense of their health beyond that captured by standard medical measures. It is conceivable that self-rated risk (cardiovascular risk being a major determinant of overall risk) represents the converse of this phenomenon. Future work might assess whether similar factors affect self-rated cardiovascular risk and the extent to which self-rated cardiovascular risk and self-rated health represent a continuum of the same construct.

1. Gramling R, Klein W, Roberts M, Waring ME, Gramling D, Eaton CB. Self-Rated Cardiovascular Risk and 15-Year Cardiovascular Mortality. Ann Fam Med 2008;6:302-306.

2. Croyle RT, Loftus EF, Barger SD, Sun YC, Hart M, Gettig J. How well do people recall risk factor test results? Accuracy and bias among cholesterol screening participants. Health Psychol 2006;25:425-432.

3. Idler EL, Benyamini Y. Self-rated health and mortality: a review of twenty-seven community studies. J Health Soc Behav 1997;38:21-37.

4. DeSalvo KB, Bloser N, Reynolds K, He J, Muntner P. Mortality prediction with a single general self-rated health question. A meta- analysis. J Gen Intern Med 2006;21:267-275.

5. Lekander M, Elofsson S, Neve IM, Hansson LO, Unden AL. Self-rated health is related to levels of circulating cytokines. Psychosomat Med 2004;66:559-563.

6. Barger SD. Do psychological characteristics explain socioeconomic stratification of self-rated health? J Health Psychol 2006;11:21-35.

7. Barger SD, Muldoon MF. Hypertension labelling was associated with poorer self-rated health in the Third US National Health and Nutrition Examination Survey. J Hum Hypertens 2006;20:117-123.

Competing interests:   None declared