Recommended Care, Says Who?

Guidelines for care increasingly help guide medical decision-making. If you’ve followed top health news over the past several years you’ve read conflicting statements about the utility of various medical procedures and tests. As a general internist I’ve devoured these reports with particular interest trying to wade through bias to formulate views that I believe will be of most benefit to my patients.

Recently I put together a talk about recommendations for prevention in women. As I prepared the talk I sorted through the data and opinions of various medical special interest groups.  Guidelines for care are usually developed by professional organizations or “expert panels,”  such as the Center for Disease Control (CDC), or the American Cancer Society (ACS).  One might like to think of these panels as unbiased in their advice, but clearly each expert panel is tainted, at least to a degree, by its own special interests.  Conflicting guidelines make it hard for health care consumers to figure out who to believe.

For example, several days ago the United States Preventive Services Task Force (USPSTF) came out with its final guideline on using PSA for the purpose of screening for prostate cancer. The task force recommended against using the test for screening healthy men, giving the screening test a D level rating (not recommended).

What is the USPSTF?

The U.S. Preventive Services Task Force (USPSTF), first convened by the U.S. Public Health Service in 1984, and since 1998 sponsored by the Agency for Healthcare Research and Quality (AHRQ), is the leading independent panel of private-sector experts in prevention and primary care. The USPSTF conducts rigorous, impartial assessments of the scientific evidence for the effectiveness of a broad range of clinical preventive services, including screening, counseling, and preventive medications. Its recommendations are considered the "gold standard" for clinical preventive services.

As a general internist I was indoctrinated to trust the USPSTF with a degree of allegiance that rivals my commitment to the University of Michigan Wolverines (my alma mater).  Nonetheless, there are other expert opinions that I also pay attention to.  In the case of PSA screening the American Urological Association (AUA) continues to make a strong argument supporting its use in screening. I discuss the debate more in my previous post:  Should Doctors Stop Using PSA to Screen for Prostate Cancer? 

Another example of conflicting opinions in medical guidelines:  Who should be screened for diabetes?  The American Diabetes Association (ADA) recommends that all adults with a body mass index >25 kg/m2 (overweight or obese adults) and one additional risk factor for diabetes be screened at least every three years, and that all adults with a BMI>25kg/m2 who are over 45 years of age be screened. In contrast, the USPSTF recommends for screening only adults with blood pressure greater than 135/80, a grade B (recommended) rating. For other healthy adults, overweight or not, it advises that the evidence is insufficient to recommend for or against screening.

These are but two examples of the numerous medical guidelines and recommendations that conflict:  who should be screened for HIV, whether women ages 40 to 50 should be screened routinely for breast cancer with mammograms, how often women over 50 should be screened with mammograms, how often women under 30 should have a pap smear, whether women need an annual pelvic exam, whether women need an annual breast exam, and at what age screening for colon cancer with colonoscopy should be discontinued.  The ADA,  the USPSTF, the CDC, the American College of Obstetrics and Gynecology (ACOG), the ACS, the AUA, the American Heart Association and American College of Cardiology—along with numerous other professional societies and expert groups have not reached consensus on various matters. 

Primary care doctors, as a group, have largely been supportive of proposed measures to reduce medical “waste”—including “unnecessary” tests and procedures that promise to bankrupt Medicare if they continue to go unchecked. In turn we stand to benefit from proposals to re-organize health care delivery and to strengthen primary care--the Medical Home Model, which may channel additional funds toward primary care. In comparison to specialists, primary care physicians have suffered in recent decades, with reimbursement policies that favor physicians who do procedures, as opposed to those who primarily talk to patients for a living.  We have been historically under-represented in the American Medical Association, the largest physician’s professional group in our country. I am part of the American College of Physicians (ACP), a national organization of internists — physicians who specialize in the prevention, detection and treatment of illnesses in adults--but recognize that we also have bias—currently, in my view, a growing skepticism of tests and procedures,  which in many ways is overdue. Yet I would wager that it’s a sentiment that primary care physicians find easier to rally behind than do many specialists, whose interests are more likely to be threatened by changes to reimbursement policy.

Who does one believe? Professional guidelines are likely to shape what is covered by Medicare and private health insurance. If you’re a gynecologist you might trust ACOG, if you’re a primary care doctor, the USPSTF or the ACP, a urologist, the American Urological Association--the politics of medicine are as real as the patients that we treat. Moreover, advocacy groups, such as the Susan G. Komen Foundation and Planned Parenthood, play a role in shaping public opinion, which no doubt can also influence the opinion of even “expert panels,”  afraid to anger these factions.

If you are a medical professional it’s important to be aware of where guidelines conflict and to avoid reflexively supporting one own expert panels.  Even better (though more difficult), read through the original data. If you’re a patient, sometimes it’s good to get the opinion of both a specialist and a generalist in these matters—each can be equally valid and it can help one see both the trees and the forest.  

 

Why Doctors Interrupt

A few weeks ago I called a neurosurgeon to discuss a patient’s recent headaches.  My patient had been seen in the emergency room several days prior with the worst headache of his life. A complete work-up had not revealed a cause for the headache.  Although he was found to have a small aneurysm on CT angiogram, there was no evidence of bleeding by lumbar puncture.  The story, however, was slightly more complex than this. There had been several other findings that remained unexplained.  One of the findings led me to discuss the patient’s case with a cardiologist.  My patient had also undergone cervical spine decompression surgery several months prior to treat cervical myelopathy.  I wanted to engage the neurosurgeon and get his professional opinion about my patient’s headache, which had now recurred several days after his ER visit.

The surgeon was cordial, but about 5 seconds into my story he seemed inpatient and interrupted me.  “I heard about this guy,” he said, “What he needs is to be seen by one of our neurovascular specialists.”  I had more I wanted to say, but the doctor did not seem to want to listen.  I raised my voice slightly, interrupted him before he had a chance to end the conversation, and bulldozed through, telling the rest of the story in about two minutes.  “Now we’re talking,” he said, as I explained further about a family history of clotting and my concern about a dural thrombus as a potential etiology.  Together we formulated a plan that I was satisfied with--though the interaction left me with a feeling of unease. 

Interruption is a pervasive communication style with doctors.  In a well known study by Beckman and Frankel patients were allowed to complete their opening statement expressing their agenda in its entirety in only 23% of physician interviews.  The average time to interruption was 18 seconds.  This study’s findings have been replicated by several others.  In a more recent study of primary care residents, patients were allowed to speak for only 12 seconds on average before they were interrupted.  Female patients experience interruption more frequently than males. In contrast, studies have suggested higher rates of patient satisfaction with physician visits during which patients and doctors interrupt at similar frequency and also with visits in which there is more “reflective” silent time during the conversation.  Perhaps the tendency to interrupt extends to all physician derived professional communications, as in my case with the neurosurgeon on the phone.  

Why do physicians interrupt?  In practical terms, throughout the course of a given day a physician may be tasked with listening to twenty to thirty patient derived histories and with solving difficult problems for each of these patients in a matter of ten to fifteen minutes. This is a tough, if not impossible job.  Consequently, once a physician believes that the meat of the story is out there, he or she may respond and interrupt before hearing details that the patient (or colleague) feels are important.  In more abstract terms interruption is a communication strategy that reinforces physician dominance in the hierarchy of the patient-physician relationship.

The most frequent complaint that I hear from patients about other physicians is that a physician did not “listen,” or did not “seem to care” about their problem.  My advice to physicians and medical trainees: sit down, bite your tongue and wait. If you do interrupt, do so with brief questions allowing your patient to return to his or her agenda.  You might be surprised and learn something, and no doubt you’ll certainly have happier patients (and colleagues).

Who Should Take Aspirin for Prevention?

The answer is not entirely straightforward.  Aspirin has been shown to reduce the risk of cardiovascular disease, including heart attack and stroke. Aspirin inhibits the function of platelets, the blood cell line responsible for clot formation.  When a heart attack or stroke occur the cholesterol plaque that lines an artery ruptures and platelets aggregate, resulting in a cascade that results in acute occlusion of a blood vessel.  Patients who are treated with aspirin are less likely to clot.  However, the effects of aspirin are not entirely benign. With its platelet inhibition it also confers a higher risk of bleeding—in particular gastrointestinal bleeding and hemorrhagic stroke, which also may be life threatening.

Clinical trials have looked at aspirin intake, cardiovascular outcomes, and bleeding risk. Aspirin for acute cardiovascular events and for “secondary prevention” (prevention after the diagnosis of coronary artery disease or cerebrovascular disease has been established) is undisputed.  Trials suggest that the benefits of therapy outweigh the risks of bleeding. 

However, whether or not aspirin should be prescribed for “primary prevention” (prevention in a person who is disease-free) is more ambiguous.  In 2009 a meta-analysis of existing study data looking at this question was published in Lancet.  The analysis found that aspirin reduced the risk of non-fatal myocardial infarction by one fifth, but that aspirin therapy also significantly increased risk of major gastrointestinal and extracranial bleeding and did not improve overall mortality.  In 2009 the US Preventive Services Task Force (USPSTF) reviewed the existing data and concluded that while aspirin reduces the risk of myocardial infarction in men and ischemic stroke in women, it increases the risk of major extracranial bleeding.  The USPSTF recommended that the decision to use aspirin therapy for the purpose of primary prevention should take into account an individualized assessment of cardiovascular risk and also bleeding risk. Patients with higher cardiovascular risk may benefit most from therapy.  

Patients considering aspirin therapy for primary prevention should assess their cardiovascular risk profile with their personal physician. The Framingham Risk Calculator is a recommended tool for estimating one’s ten year risk of having a major cardiovascular event.   However, Framingham may not be as useful for women as it is for men, and some recommend use of the Reynolds Risk Calculator. The Reynolds Calculator incorporates the inflammatory marker hs-crp into its calculation of ten year risk.

In a similar vein, in 2010 the American Diabetes Association, the American Heart Association, and the American College of Cardiology issued a joint statement revising their recommendations for use of aspirin for thepurpose of primary prevention amongst diabetic patients.  In contrast to old guidelines, the new recommendations do not advise that all diabetics over age 40 receive aspirin therapy.  Rather, they advise aspirin therapy for primary prevention in male diabetics under 50 and female diabetics under 60 only if one additional cardiac risk factor is present (hypertension, high cholesterol, smoking, family history, microalbuminuria) The new recommendation is based in part on a subgroup analyses of diabetic patients in the meta-analysis of the Antithrombotic Trialists’ Collaboration showing  that diabetic patients benefited less from aspirin therapy than non-diabetics. In addition, several smaller studies conducted specifically on diabetics and looking at primary prevention failed to demonstrate a significant benefit of aspirin therapy in those without diagnosed cardiovascular disease.  Further study is ongoing to research the issue of primary prevention of cardiovascular disease with aspirin in diabetic patients.  For now the therapy is recommended for diabetics who are determined to be intermediate to high risk (Framingham risk of 10% or higher).

Who is most likely to suffer a complication related to daily use of aspirin? Risk factors for gastrointestinal bleeding with aspirin therapy have been identified:

• Non-steroid anti-inflammatory (NSAID) use (in particular, high dose NSAID use)
• Chronic steroid use
• Prior history of peptic ulcer disease (PUD)
• Advanced age (>60-65 years old)
• GERD or dyspepsia (less risk than PUD)
• Concomitant use of another anti-coagulant

Treatment with a proton pump inhibitor, or the prostaglandin E analog misoprostol, can reduce one’s risk of gastrointestinal bleeding from NSAIDS.  By contrast, H2 blockers are not effective in this regard.  In addition, using enteric coated aspirin does not reduce its gastrointestinal toxicity.  It is unclear what dose of aspirin is best for primary prevention, but most recommend low dose aspirin (81-162mg), which appears to be equal in efficacy to higher doses (though it has not been demonstrated to be safer).

It’s interesting to me that simultaneous with a growing emphasis on incorporating population-based strategies into healthcare delivery we are also becoming increasingly aware of the importance of identifying personalized risk factors in order to best counsel individual patients on medical care and prevention.  Aspirin therapy for primary prevention is an example of how a one-size-fits-all population-based strategy is hard to apply.  We have seen similar recent trends with mammography screening recommendations, using PSA for prostate cancer screening, and will likely soon be hearing more about using a personalized approach to recommending statins for the purpose of primary prevention of cardiovascular disease (given recent associations between statin use and reversible cognitive complaints and diabetes). How population medicine, its associated quality reporting, and pay-for-performance on the one hand, and personalized medicine on the other, are reconciled in medical practice will be a challenge to be dealt on the level of policy, practice, and reimbursement in years to come.