COMMENTS FROM DR. SEXTON:
Overall, this is an excellent article on an important topic. I only have a few comments to add to Dr. Siwek and the
reviewers. Please carefully review their comments, especially Reviewer #3. My only contradiction would be to
2
avoid adding more details of the studies as suggested by Reviewer #3. We have space limitations that should be
reserved for explaining your recommendations.
RESPONSE: We agree that adding more details of the studies would add to the length without adding
substantially to the overall message of this paper. We did add one sentence about the effects of global CHD
risk +/- counseling reducing predicted CHD risk from 0.2-2%.
In the section about the evidence behind global risk calculation, please briefly define in parenthesis "receiver
operator characteristic".
RESPONSE: We have added a brief parenthetical description “(graphical techniques to assess the accuracy
of tests)” of this term.
In the section on evidence of effectiveness, the last paragraph should be rewritten in the third person per AFP style.
RESPONSE: We changed to third-person style.
I agree with Reviewer #3's concern about potentially endorsing CRP testing. Please further clarify the role of CRP in
global risk assessment, if any.
RESPONSE: We have removed the sentences on the potential use of CRP for reclassifying people at
intermediate risk and added a sentence on the lack of evidence that using CRP to guide statin management is
better than global CHD assessment alone.
I agree with Dr. Siwek's comments about adding a table for physicians to better understand and explain risk
reduction/interventions in the section on presenting the information to patients. Again in that section please do not
use first person "our systematic review".
RESPONSE: We have added the new table (see also under Dr. Siwek’s comments). We changed “our…” to
“a recent systematic review….”
In the uncertainties section you mention the role of diabetes as a CHD equivalent, but what about chronic kidney
disease? How does this play a role in risk assessment? Should we be doing a CKD calculation too?
RESPONSE: CVD is highly prevalent in patients with CKD and may account for 50% of all deaths. The
Dialysis Outcomes Quality Initiative (DOQI) publication on the evaluation, classification, and stratification of
CKD states that a reduced GFR identifies individuals at greater risk for CVD and death. This risk is the
result of traditional and nontraditional CVD risk factors. AHA guidelines recommend treating patients with
CKD as highest risk. We have added this information to the paper (and the reference).
Also, please consider writing a patient education handout to accompany your article. Dr. Siwek has provided further
details below. Before you write it, please take a look at the section on "health calculators" on familydoctor.org. You
can click on http://familydoctor.org/online/famdocen/home/tools/calculators.html which then links you to revolution
health http://www.revolutionhealth.com/calculators/heart-attack-risk. If there is too much overlap, please let me
know. Otherwise, creating a unique handout on this topic will be a great addition to the family doctor site.
RESPONSE: Thank you for inviting us to write a patient handout. We did not think there was too much
overlap. We do, however, think that the www.med-decisions.com website is better than the revolutionhealth
site because it allows people to “see” the average risk reductions potentially achievable by instituting therapy
(and also does not have advertising). We have included this information in the patient handout. (We did not
include the address to the NCEP calculator because it also does not provide information on risk reduction.)
Let us know what you think.
3
COMMENTS FROM DR. SIWEK:
Overall, this is a well-written, professionally prepared manuscript. Many thanks for sending it to us. It makes an
important point about assessing global risk, and using risk calculators. And, you provide some handy
references/links to get them into the hands of any reader who doesn't already have one.
RESPONSE: Thank you Dr. Siwek.
Page 3: in the introduction, you do a good job of telling readers to treat absolute risk, rather than just risk factors.
Anything you can do to emphasize this point would be helpful. There is evidence that clinicians spend too much
time treating individuals who have risk factors but relatively low risk, while not being aggressive enough with
individuals who have high risk, some of whom have modest levels of "risk factors." One concrete example that
drives this point home is that people who have already had a cardiovascular event (MI or CVA) are those at highest
risk for recurrence; yet too often those people aren't treated aggressively enough.
RESPONSE: We added a sentence emphasizing the important point that treatment should be based on a
person’s absolute CHD risk rather than just his or her risk factors. We did not add the specific example of a
person with a previous event only because that gets into secondary prevention, for which Framingham
calculations would not be applicable. As pointed out, they are “automatically” in the highest risk group. We
added a sentence in the first paragraph of the introduction reminding the reader of this fact.
Page 4: "Because the risk-benefit ratio of taking aspirin for primary prevention transitions from harmful to helpful at
a 10-year CHD risk of between 6% to 10%, 13 clinicians need to be able to estimate risk with relatively fine
gradations in order to effectively counsel patients and make recommendations." Good, concrete example.
RESPONSE: Thank you.
Page 7: "Clinicians should also provide information about what constitutes an actionable level of risk (e.g., 10% for
aspirin, 20% for cholesterol reduction, and any risk level for hypertension treatment and smoking cessation)
according to current practice guidelines." Please create a table that shows thresholds for intervention, based on
absolute risk. Also, please be specific/explicit: what does "10%" refer to?? There's no context, and for someone not
used to using these calculators, they're left in the dark guessing. Please be specific about "hard" endpoints--risk of
death and MI, and total CHD risk, including, for example, angina. Be explicit when setting thresholds for
intervention-what is the percentage, and what is the endpoint? Pretend you're telling someone who's never used one
of these calculators how to put it into practice. If this, then that etc.
RESPONSE: We created a new table with this information.
Page 7: "For example, a male, hypertensive smoker who has a 20% CHD risk over 10 years can reduce his risk to
10% (a 50% relative risk reduction) by smoking cessation." Please specify that this is global (if indeed that's what
you're referring to).
RESPONSE: Yes, that is what we are referring to. Changes made.
Also, I just noticed that global CHD risk does not seem to be defined in the text of the manuscript, but only in the
abstract. Please define it in the text, since this is such an important point (does CHD risk mean heart attack,
cardiovascular death, both, other?).
RESPONSE: Another excellent point. We added definition as first sentences of 2nd paragraph.
Page 8: "Finally, risk is a difficult concept, and there remains much to be learned regarding best methods of
presenting risk to patients, particularly those with low literacy or numeracy skills." Here's another opportunity to
distinguish between risk (where the money is), and risk factors. Emphasize to readers that they should be treating a
person's actual risk (for example, 20% risk of CHD in the next 10 years), rather than an isolated risk factor (elevated
LDL, which might translate into significant risk in some people, but not in others).
4
RESPONSE: We added a concluding sentence as suggested.
There seems to be some evidence that patients find it helpful when risk and benefit are presented using boxes
illustrating outcomes with treatment and without treatment, as exemplified here:
In AFP:
http://www.aafp.org/afp/20090415/668.html?aafpvlogin=5929344&aafpvpw=&URL_success=http%3A%2F%2Fw
ww.aafp.org%2Fafp%2F20090415%2F668.html
In the primary source:
http://nntonline.net/visualrx/examples/
<<in response to the following NICE guidance:
"People should be offered information about their absolute risk of CVD and about the absolute benefits and harms of
an intervention over a 10-year period. This information should be in a form that:
* presents individualized risk and benefit scenarios
* presents the absolute risk of events numerically
* uses appropriate diagrams and text."
The smiley face diagram made its first appearance in the BMJ in an article communicating risk; since then we
changed the colours of the faces to red, yellow and green as red and purple are not easy to distinguish for those who
are colour blind, and it seemed that traffic light colours might be more familiar to users.
There is also an interesting recent article on communicating risk by David Spiegelhalter in the Annals of Family
Medicine in which he refers to Visual Rx as one of the ways of trying to make risks accessible to patients. Links to
many further examples are available here. >>
RESPONSE: Thank you for this suggestion also. We added a Cates Plot as a new figure (along with the
website link to VisualRx) as well as information in the text and in the “best practices” for communicating
risk table.
References: if reference 20 is not accepted by the time of publication of this manuscript in AFP, then we won't be
able to list it (journals do not list "under review" papers in reference bibliographies). If this is the case, then please
renumber all references accordingly. If, as you request, we delay publication of this article until that one is accepted
(in which case, we would list it as "in press") or published, that's okay, provided that it is likely to get accepted. We
wouldn't want to hold this one indefinitely (no longer than a year from acceptance).
RESPONSE: The systematic review article has been accepted for publication by Archives of Internal
Medicine. The current citation is: Sheridan SL, Viera AJ, Krantz M, Ice CL, Steinman LE, Peters K, Kopin
L, Lungelow D. Does giving global coronary risk information to adults improve outcomes? A systematic
evidence review. Arch Intern Med [in press].
Please review, and cite as appropriate, this article: A Practical and Evidence-Based Approach to Cardiovascular
Disease Risk Reduction. Arch Intern Med 7/26/04 (especially if they have any tables or figures you might want to
adapt).
RESPONSE: Thank you for this reference article. We did not include it since it addressed secondary
prevention. There were no tables or figures to adapt.
Please see this reference on the issue of (over)treating risk factors, instead of actual risk.
http://www.bmj.com/cgi/content/extract/333/7576/988
As well as the issue of targeting risk, not just risk factors:
http://www.bmj.com/cgi/content/extract/324/7353/1570?maxtoshow=&HITS=10&hits=10&RESULTFORMAT=1
&title=risk+factors&andorexacttitle=and&andorexacttitleabs=and&andorexactfulltext=and&searchid=1&FIRSTIN
DEX=50&sortspec=date&fdate=1/1/2000&resourcetype=HWCIT
RESPONSES: Thank you for these references as well. We particularly like the Law & Wald paper and use it
in one of the lessons in our MD-MPH course on prevention.