BeyondApples.Org

About 24 million Americans, or 8 percent of the U.S. population, have diabetes. Although people with diabetes are at risk for serious medical complications, with close monitoring and treatment they need not suffer from them. The problem is 1 in 4 people with diabetes don’t even know it.

Compared to many other diseases, diabetes can be diagnosed easily. In people without any symptoms of the illness, two routine blood tests can provide the diagnosis. The first is to screen for the disease; if abnormal, the test is repeated. The most commonly used screening test for diabetes is a fasting plasma glucose test, or FPG. Because diabetes is characterized by abnormally high sugar levels in the blood (also called plasma), an FPG of 126 mg/dl or higher is suggestive of the condition.

While this sounds simple enough, diagnosing diabetes can be more challenging in practice. The problem with FPG is that it requires not eating anything for at least 8 hours prior, which is typically done by having the patient return on another day after skipping breakfast. This creates at least two barriers. First, for many patients coming to the doctor is an ordeal. My patients often have to take time off of work to see me. This is particularly true for those with blue-collar jobs that require them to report to work early in the morning, the exact time that the blood sample must be drawn. The second barrier is that returning for labs makes follow up more difficult. It is easier for doctors to keep track of tests that are done in the same office visit than those done later. As I discussed in a previous entry (http://beyondapples.org/2009/07/30/forgetting-to-break-bad-news/), studies show that over seven percent of clinically significant abnormal test results are not reported back to patients.

New recommendations from the American Diabetes Association (ADA) support the use of a new screening test, hemoglobin A1c (HbA1c), that promises to make the process easier.1 HbA1c avoids the pitfalls above because it does not require an overnight fast. Blood glucose levels change moment-to-moment with food, stress, and time of day; in contrast, HbA1c is a measure of a person’s average blood sugar levels over the previous two to three months. As a “running average,” it is less subject to daily variation and is already being used to monitor people with diabetes. An HbA1c of 6.5% or higher is suggestive of diabetes.

Another advantage of HbA1c is that it allows doctors to more reliably diagnose diabetes in hospitalized patients. Any illness creates a sympathetic “fight-or-flight” response that drives up heart rate, blood pressure, and, it turns out, glucose levels. As a result, a patient who has an elevated FPG in the hospital may have diabetes or just be hyperstimulated. With HbA1c, doctors will be able to detect diabetes in those who are often at greatest risk of having it — those who slip through the cracks of good primary care and wind up in the hospital.    

So should you be screened for diabetes? The United States Preventive Services Task Force (USPSTF) recommends routine screening for diabetes in individuals with a blood pressure above 135/80 mmHg.2 The ADA casts a wider net and recommends screening in all adults ages 45 and older, and younger adults who are overweight (body mass index or BMI > 25kg/m2) and have any additional risk factors:  

• physical inactivity
• first-degree relative with diabetes (parents, siblings, children)
• members of a high-risk ethnic population (e.g., African American, Latino, Native American, Asian American, Pacific Islander)
• women who delivered a baby weighing > 9 lbs or were diagnosed with gestational diabetes
• hypertension (> 140/90 mmHg or on therapy for hypertension)
• HDL cholesterol level < 35 mg/dl or a triglyceride level > 250 mg/dl
• women with polycystic ovary syndrome
• other clinical conditions associated with insulin resistance (e.g., severe obesity, acanthosis nigricans)
• history of cardiovascular disease

To calculate your BMI, go to http://www.nhlbisupport.com/bmi/. 

Like FPG, HbA1c can also help identify individuals with pre-diabetes, those with sugar levels that are abnormal but not enough to diagnose diabetes. For FGP, individuals with a blood sugar level of 100-125 mg/dl have pre-diabetes; with HbA1c the range is 5.7%-6.4%. One-third of people with pre-diabetes will develop diabetes within 3 years; recognizing the condition is critical because studies show that through aggressive lifestyle modifications people with pre-diabetes can prevent the onset of diabetes (see http://beyondapples.org/2009/11/14/world-diabetes-day-2009/). 

Why go through all this trouble to talk about a new test for diabetes? After all, isn’t your doctor the one who decides whether to order blood work and what tests to order? The trouble is that we’ve learned the hard way that it takes anywhere from 7 to 10 years for the medical community to adopt a new guideline. For those of us at risk for diabetes that’s not soon enough. The next time your doctor orders blood work ask him or her about getting screened for diabetes. If you don’t need fasting blood work for another reason and are worried about having to come back, ask about HbA1c.

- Shantanu Nundy, M.D.

1 Standards of Medical Care in Diabetes — 2010. American Diabetes Association. Diabetes Care, Volume 33, Supplement 1, January 2010. 

2 http://www.ahrq.gov/clinic/uspstf08/type2/type2summ.htm

Last week the internal medicine residents in my program participated in a simple but profound experiment. Using a standardized patient communication tool we created, called the “Heart Health Checklist,” we wanted to see if we could improve our patients’ satisfaction with their preventive cardiovascular care.

In the weeks leading up to the experiment, a group of us decided what the patient communication tool would look like and then actually created it. We designed the tool around three critical communication gaps. First, our patients often didn’t realize what steps we were taking to prevent cardiovascular disease. Second, even though most of us counsel our patients on preventing heart disease, they often didn’t have a complete understanding of their risk factors, including where their numbers stood with respect to their goals. And thirdly, our patients were often uncertain about the steps they needed to take to reduce their risk of cardiovascular disease, or if they were taking them, such as starting a new medication, they did not fully understand why.

What we came up with was the Heart Health Checklist, a paper form that we would fill out and discuss in clinic and then give to our patients to take home. It would serve as a “report card” of sorts. Instead of History and Math, patients were being “graded” on their cholesterol and blood pressure. The checklist showed them exactly where they stood compared to where they should be and included instructions for how to improve. We decided to focus on five elements of cardiovascular health: blood pressure, smoking, cholesterol (total, HDL, and LDL cholesterol), weight (BMI), and, for those patients with diabetes, sugar control (HbA1c). As discussed in an earlier blog post, “Are You Five for Five?”, 95 percent of heart attacks occur in people with at least one of the following five risk factors: high blood pressure, cholesterol disorder, smoking, diabetes, and obesity. By focusing on these five measures, the checklist targeted the major controllable risk factors for heart disease. Graphically the Heart Health Checklist was designed as a table. Each row was dedicated to one of the five measures of cardiovascular health. In the first column we put the patient’s goal (e.g., blood pressure < 140/90); in the second column, his or her current status (e.g.,149/86) ; and in the third column our plan for improvement (e.g., reduce salt, increase dose of HCTZ).

As part of our experiment, we will be collecting and analyzing data about patient satisfaction before and after we implementing the Heart Health Checklist. While the data won’t be available for some time, in my mind the checklist has already proved its merits. Two cases illustrate the point:

Case 1: Ms. PG is a 68 year-old woman who I have been seeing in clinic for over a year. She has been struggling with weight loss and anxiety over this time, but also with a smoking habit that just won’t quit. Every time I see her in clinic it’s pretty much the same story – ”Yes Dr. Nundy, I want to quit. No Dr. Nundy, I’m just not ready yet.” Her visit last week was no different. But just as she was heading out the door, I realized I had forgotten about the Heart Health Checklist. I reached for the checklist and frantically started filling it out, already late for my next appointment. As I went through her blood pressure, cholesterol, weight and sugar control; I wrote “excellent” in the column where I was supposed to write out her plan for improvement. As I did, she remarked at how amazed she was about how good her health was. Only one measure stood out — smoking — next to which I wrote “make a plan to quit”. Her sentiment was palpable: ”You mean to say that if I quit smoking then my heart health will be overall ’excellent’?”

Case 2: Mr. YM is a young professional in his late 20s. Since graduating from college, he has put on weight and developed prehypertension.* With relatively few medical issues, we always have plenty of time to discuss his blood pressure and weight and review proper lifestyle modifications. Like Ms. PG I went through the entire clinic session forgetting to take out the Heart Health Checklist. At the last moment I remembered, and used the checklist to go over (yet again) his blood pressure and weight and our plan for controlling them. But this time the conversation took a different tenor. With the numbers staring right at him, I could almost see the light bulb go off in his head as I explained what I was concerned about and what we could do to improve. He left the office firmly grasping the checklist in his hands, telling me that this new tool was “extremely helpful” and a “great idea.”

Much has been written about the art of doctor-patient communication — how we are supposed to listen attentively, speak with minimal medical jargon, and ask patients what they understand about their conditions. But I seldom hear much about the science of doctor-patient communication. While simple, our experiment is already having a profound influence on how I think about relaying health information to my patients. While talking clearly and openly with our patients about their medical conditions is no doubt essential, so too I am finding is simply writing it down.

Preventive health is sometimes viewed as a futile discipline. Old habits die hard; patients will never change. We look for the next breakthrough pill or wonder diet, thinking that they will curb the tide of chronic disease that is washing over our nation. But the fact of the matter is that obesity, diabetes, and even heart disease are lifestyle diseases. The problem is inherently in how we live — so too then must be the solution. It’s not that people don’t want to be healthy — everybody does — but rather that they don’t understand what is their risk of serious illness, where they stand compared to where they should be, and how to make small steps in the right direction. The next time we reach for a prescription pad to control a lifestyle disease, we should consider reaching for a blank sheet of paper, or better yet, a Heart Health Checklist.

- Shantanu Nundy, M.D.

* elevated blood pressure not high enough to be classified as hypertension; by definition, systolic BP between 130-140 mmHg and/or diastolic BP between 80-90 mmHg

Recently a close relative of mine was diagnosed with advanced colon cancer. The experience has been at once saddening, terrifying and sobering. Why did this happen? How did it happen? Why wasn’t it caught earlier? From an objective perspective, the case is unusual on a number of fronts, and illuminates the major limitations of the current state-of-art of colon cancer screening.

1. Ethnicity. Like myself, my uncle-in-law is of Indian descent. Colon cancer is much less common in Asian ethnicities than Western ones. In Japan, where the best data is available, the disease is uncommon enough that doctors do not routinely perform colon cancer screening (in contrast they routinely screen for stomach cancer, which we don’t do in the U.S.). The risks and benefits of screening for a disease are related to its prevalence. Though my uncle is 53, three years beyond the age at which screening is routinely recommended in the United States, I would not have necessarily recommended it for him given its low prevalence in India.

2. Age: Even for people living in the U.S., developing advanced colon cancer at 53 is rare. Though screening is recommended for average-risk individuals starting at age 50, the average age at diagnosis is 64. The goal of screening is to detect and remove pre-cancers before they progress into cancer or if cancer has already developed to catch it early at a curable stage. Thus had we screened my uncle-in-law for colon cancer we would be expecting, if anything, to find precancerous polyps, not full-blown cancer.

3. Family history: Every American over ages 50 to 75 needs to be screened for colon cancer, regardless of lack of family history.1 In those with a family history, screening may need to begin at an earlier age. For example, experts recommend screening first-degree relatives of people who developed cancer before age 60 starting at age 40 or 10 years before the age of diagnosis, whichever comes first.2 As far as we know, there is no history of colon cancer in our extended family. Such a history may have tipped us off to begin our prevention efforts earlier.

Our understanding of cancer and other diseases is still limited, and as a result our abilities to detect diseases early and to prevent them altogether are still crude. Most diseases result from a combination of genes and environmental exposure. The better our understanding of these elements the greater our ability for targeted prevention. Family history of colon cancer is a proxy for our genetic makeup. Thus people with a history of colon cancer in first-degree relatives with whom we share 50% of our genes or in multiple second degree relatives are at higher risk than the average person. Ethnicity is another proxy for genetic risk and one that is even more crude. In the future, we will hopefully be able to move beyond these blunt measures and better assess genetic risk. Such understanding would allow us to better target screening to those at highest risk. This would prevent unnecessary testing in low-risk adults and promote screening in people at truly increased risk. Already in diabetes researchers have identified specific genetic variants that predispose people to diabetes; in oncology, BRCA1 and BRCA2 are already being used in the clinics to identify women at very high risk of breast and ovarian cancers.

Our environment also contributes to our risk of colon cancer. Population studies have demonstrated that as people immigrate from Japan to the United States their risk of colon cancer increases. Such “natural experiments” provide strong evidence that something about our Western way of life predisposes us to colon cancer, but the exact reasons are still unclear. Though some have suggested that high fat, low fiber diets may play a role, these data have not established causality. Contrast this degree of understanding with cervical or lung cancer. In cervical cancer we have established that the cause of the disease is infection with the human papillomavirus or HPV; this understanding has led to the development of the first vaccine against cancer. Likewise, while it is easy to take for granted today, the knowledge that smoking is the major risk factor for lung cancer has had enormous implications and saved countless lives. Hopefully one day we will learn more about the modifiable risk factors for colon cancer and be able to prevent it altogether.

Given these limitations, it’s remarkable to consider that our current state-of-art — routinely screening  all Americans over age 50 with fecal occult blood testing, sigmoidoscopy or colonoscopy — is proven to reduce death from colon cancer by 33%. If we simply increased the proportion of Americans being screened today to 90%, we would prevent 19,000 deaths annually.3 Still, tragedies like the one in my family are a harsh reminder that while we have made much progress we have much farther to go.

- Shantanu Nundy, M.D.

1 http://www.ahrq.gov/clinic/uspstf08/colocancer/colosum.htm

2 http://caonline.amcancersoc.org/cgi/content/full/CA.2007.0018v1

3 http://www.prevent.org/content/view/50/120/

You never know it’s coming until it’s right in front of you – a teachable moment, the chance to make a lasting impact in the life of another person.

Yesterday, I was working in urgent care. Rarely glamorous enough to be found on House or Grey’s Anatomy, urgent care is where busy primary care offices often schedule patients who need to be seen for acute complaints on a 24- to 48-hour basis. The vast majority of cases are upper respiratory illnesses (e.g., “my nose is running), acute musculoskeletal pain (“my back hurts”), or social issues ( “ran out of meds”). It can get busy, especially in the winter, and the pace is quick — get the story, figure out what they got, and send them on their way. Preventive issues are left for the full doctor visits; the focus here is on the problem at hand.

I was having a fairly typical afternoon as I headed in to see my fifth patient. The triage sheet listed “cough ” as the chief complaint, and the temp was recorded as 97.8. Glancing at the birthdate on the chart, I calculated that the patient was 53 as I marched into the examination room. From the doorway, it was clear she was having an upper respiratory illness. Her face was slightly swollen and red, and she was breathing through her mouth. She quickly offered that her symptoms actually started improving the night before, but that she didn’t want to cancel her appointment just in case. A bit more history confirmed the diagnosis — she had a routine cold. A minute later, nearing the end of my physical exam, examining her mouth, it hit me — the smell of cigarettes.

“You smoke?” I asked half-rhetorically.

“Yea doc, I smoke,” she said, shaking her head.

“That’s no good. You know smoking’s probably a major reason why you’re here, don’t ya? Smokers get the worst cases of the flu.”

I decided to leave it at that. Get the point across, plant the seed, and hope that the over time personalized messages like that push her from pre-contemplative to being serious about quitting.

When I presented the patient to the supervising attending doctor, I kept it simple: “URI, starting to get better.” Never sure how much to say, I quickly added, “though it’s a kind of a teachable moment — she’s a smoker.”

According to Wikipedia, a teachable moment is “the time at which learning a particular topic or idea becomes easiest or possible.”* In medicine, the term is used similarly to describe a situation in which a patient is most apt to adopt a healthy behavior; for example, telling someone the importance of wearing seat belts during an ER visit for a car accident. Patient education is an essential component to being a doctor — the word “doctor” in fact comes from the Latin word “docere” meaning “to teach” – but in reality, without a meaningful context, it’s easy for patients to simply ignore a doctor’s advice. Teachable moments then become critical avenues to effectively communicate health information and affect behavior change.

Though I said it only half-seriously (“teachable moment” being a jargon term), my attending immediately answered the call. After emphasizing the benefits of quitting — not the least of which is to the pocketbook (“with the money you save, treat yourself to a vacation”) — my attending learned that over the past two weeks, while the patient was sick, she had cut down smoking from a pack a day to two cigarettes a day. Not myself realizing the significance of this, my attending pointed out to the patient that she had already gotten over the biggest hurdle to quitting – overcoming the awful symptoms of nicotine withdrawal. “Take these two miserable weeks you’ve had as an opportunity to change your life for the better.”

It was clear from the patient’s body language that the message was sticking. With my attending back seeing other patients, I pushed on and asked about her previous quit attempts. She had tried quitting in the past but would relapse after a week or so. Quitting made her stressed out and tired, which I explained were typical symptoms of nicotine withdrawal. She was definitely motivated. Both her brother and her father had passed away from lung cancer; they too were heavy smokers. But because of her failed past attempts she had given up hope that she could actually quit. I again encouraged her, saying that today doctors had more ways than ever to help people quit and stay quit. She had heard of Chantix (varenicline) and asked if it could help. I described how varenicline worked — how it reduced the cravings for cigarettes and blunted the chemical satisfaction of smoking. After advising her about its side effects (vivid dreams) and setting a quit date (next Wednesday), I wrote her a prescription, scheduled her a follow up appointment, and emailed her primary care doctor about the plan.

Though not 15 minutes earlier her mind was preoccupied with her cough, something that would have improved and gone without our help, here she was leaving with a plan to quit smoking, something that if successful would pay dividends the rest of her life. It was a teachable moment, for her and for me.

After I wished her best of luck, she said aloud but really for herself, “I think I can do it… No. I have to. I have to do it.”

- Shantanu Nundy, M.D.

* http://en.wikipedia.org/wiki/Teachable_moment

In the past two decades lowering LDL cholesterol has become synonymous with the fight against heart disease. Millions of people are on statins — the most popular class of cholesterol-lowering medications — including Lipitor and Crestor. Despite this, doctors still aren’t sure who to prescribe statins to.

While most doctors agree that everyone who has had a heart attack should be on a statin, we are less clear about when to use statins to prevent coronary heart disease (CHD) in people who have never had an acute event before. The current guidelines for cholesterol-lowering therapy come from the National Cholesterol Education Program Adult Treatment Panel III guidelines (or ATP III) that were originally published in 2002. The guidelines suggest that the decision to start a statin be based on an individual’s risk of a heart attack and their LDL (or bad) cholesterol level.1 Below is a table that summarizes the recommendations:

Thus, deciding whether to start a cholesterol-lowering medication is a three-step process. First, doctors need to perform a risk assessment, typically by using an established risk calculator such as the Framingham risk score (to calculate your risk visit http://hp2010.nhlbihin.net/atpiii/calculator.asp?usertype=prof). Second, doctors measure LDL cholesterol. And third, using the table above, doctors decide whether to initiate treatment based on whether the patient is at or above their goal LDL.

There are two major problems with our current treatment approach. The first is that we are woefully bad are implementing it. A recent study published in JAMA found that many Americans have LDL cholesterols above their goal levels.2 Using national data, the study investigators found that among those with high LDLs, 35 percent had not been screened, 25 percent were screened but did not know their results, and 40 percent were inadequately treated. The data was even worse for high-risk individuals — those who would benefit most from statins; in this group, two-thirds of those eligible did not receive any treatment.

The second problem is that following the ATP III guidelines sometimes leads to perverse clinical scenarios. I have thirty year-olds with 0 risk factors for coronary heart disease on statins because their LDL is 190 mg/dl. At the same time, I have older men who smoke and have diabetes not taking statins because their LDL is 90 mg/dl. Clearly, of these two types of patients, the latter are at much higher risk of CHD and therefore would benefit more from statins. And yet, following the guidelines leads to the exact opposite of what good clinical judgment would suggest.

Two accompanying editorials in JAMA propose provocative new approaches for preventing coronary heart disease.3,4 While the two editorials make two very different suggestions, both sets of authors move away from using LDL cholesterol to decide whether to initiate statin therapy:

1. Age-based:The Hingorani and Psaty editorial proposes starting statins on everyone above a certain age (note: they don’t specify what age). As rationale, they cite the limitations of current risk calculators in distinguishing who is at risk for coronary heart disease and who isn’t. Because statins have become relatively cheap and are widely considered safe, simply starting everyone above a certain age on statins would reduce heart attacks and strokes in people at high risk of CHD without significantly harming those at low risk.

2. Risk-based. The Gaziano and Gaziano editorial proposes simplifying the current algorithm so that individual risk of coronary heart disease alone be used to guide statin therapy. Thus, for example, any person with above a 20 percent 10-year risk of CHD would be placed on a statin regardless of their LDL. As evidence, they cite data that the reduction in CHD risk in people treated with statins is independent of baseline LDL; instead percent reduction in LDL seems to be more relevant.

As a first pass, prescribing a statin to everyone above a certain age seems radical. No treatment is without harm and certainly not without cost. However, there is precedent within preventive health for such a blanket approach. We give the flu vaccine to everyone over age 50 and the pneumonia vaccine to everyone at age 65, simply because above this age the risk of serious infections is sufficient enough to warrant the costs. Likewise, we give antibiotic eye drops to all infants at birth to prevent gonorrhea eye infection, regardless of their mothers’ sexual history. There is also precedent for the risk-based approach proposed by Gaziano and Gaziano. In fact, in the UK, it is standard-of-care to place patients on statins based solely on their risk of CHD.

While there is a good evidence to support these two proposals, moving to a new CHD prevention strategy is ultimately a practical decision. Do I want to underprevent heart disease or overprevent it? The data shows that our current system is simply not working. While there are some clinics that no doubt have good systems in place to effectively screen, risk-stratify, and treat patients with high LDL; most clinics, including mine, do not. While we can and should argue that these clinics need urgent redesign, until such systems are in place, it makes sense to take a simpler, more common sense approach to preventing our nation’s number one killer.*

- Shantanu Nundy, M.D.

*Author’s Note: My views on statins for primary prevention have shifted since I last wrote about them (see http://beyondapples.org/2009/08/16/statins-panacea-or-to-good-to-be-true/). This change in thinking was prompted by the increasing realization that the sophistication of evidence-based medical guidelines often exceeds our ability to effectively deliver them in practice. Is a perfect treatment algorithm performed 50% of the time really better than an imperfect treatment algorithm performed 100% of the time?  In this particular case, there is decent evidence that our algorithm itself is imperfect. That is, used correctly 100% of the time we would still misclassify and over- and undertreat a significant portion of patients. Add to this our poor track record in implementing this algorithm and you create a huge opportunity to improve CHD prevention by simplifying our decision-making algorithm.

1 National Cholesterol Education Program Adult Treatment Panel III report,  http://www.nhlbi.nih.gov/guidelines/cholesterol/index.htm

2 Kuklina EV, Yoon PW, Keenan NL. Trends in High Levels of Low-density Lipoprotein Cholesterol in the United States, 1999-2006. JAMA 2009;302(19):2104-10.  http://jama.ama-assn.org/cgi/content/abstract/302/19/2104

3 Gaziano JM, Gaziano TA. Simplifying the Approach to the Management of Dyslipidemia. JAMA 2009; 302(19):2148-2149. http://jama.ama-assn.org/cgi/content/short/302/19/2148

4 Hingorani AD, Psaty BM. Primary Prevention of Cardiovascular Disease: Time to Get More or Less Personal? JAMA 2009;302(19):2144-2145. http://jama.ama-assn.org/cgi/content/extract/302/19/2144

This past weekend was spent in the “mecca” of wellness and prevention — Southern California. As I strolled through a Santa Monica farmer’s market surveying farm fresh foods, learning about organic farming, and yes, trying raw (unpasteurized) milk, I could not help wondering if yet again I was getting nutrition all wrong (see http://tinyurl.com/y9qr9yy). The visit was all the more illuminating because it followed a recent reading of “In Defense of Food” by Michael Pollan*, which challenged the way I think about food and eating. In this blog post I summarize the key points of the book and reflect on the lessons learned.

Synopsis

So what kind of diet does Michael Pollan espouse in his book? Low-fat? Low-carb? Low-calorie? None of these, actually. In fact, “In Defense of Food” rejects the notion that food can be understood by its constituent parts. “Food is more than the sum of its parts.” Thus thinking about food in reductionist terms – grams of fat, protein, and carbohydrates — is overly simplistic and plain wrong.

Instead Mr. Pollan makes the case that we need to get back to thinking about food not nutrients, to eating real food not artificial products. His recommendations on a healthful diet are summarized in three simple sentences: “Eat food. Not too much. Mostly plants.”

While the second two assertions are self-evident, the first deserves some explanation. Eat food? What else do we eat, if not food? “In Defense of Food” argues that much of we eat today is not really food. Rather the Western diet is replete with “food-like” substances. As an example, Michael Pollan cites Wonder Bread. While at first glance Wonder Bread appears to be food (I mean, isn’t it just bread?), it’s really a commercial product. In fact, typing “wonder bread ingredients” into Google brings up “Wonder Bread History — Invention of Wonder Bread.” Here is a list of the ingredients in Classic Wonder Bread**:

Enriched Wheat Flour [Flour, Barley Malt, Ferrous Sulfate (Iron) B Vitamins (Niacin, Thiamine Mononitrate {B1}Riboflavin {B2}Folic Acid)Water, High Fructose Corn Syrup or Sugar, Yeast. contains 2% or less of Calcium Sulfate (Ingredients in Excess of Amount Present in Regular Enriched White Bread) Wheat Gluten, Soybean Oil, Salt, Dough Conditioners (May contain Mono and Diglycerides, Sodium Stearoyl Lactylate, Dicalcium Phosphate, Datem, Sorbic Acid, and/or Calcium Dioxide) Vinegar, Soy Flour, Tricalcium Phosphate (Ingredients in Excess of Amount Present in Regular Enriched White Bread)Yeast Nutrients (May Contain: Ammonium Phosphate, Monocalcium Phosphate, Calcium Carbonate, Ammonium Sulfate, Ammonium Chloride, and/or Diammonium Phosphate)Whey, Cornstarch, Wheat Starch, Enzymes, Calcium Propionate, to Retain Freshness) Soy Lecithin. Kosher Information – Contains Less Than 1.6 % of Whey (A Milk Derivative).

Some of these ingredients are no doubt essential to making bread. But one has to “wonder” about the need for high fructose corn syrup, dough conditioners, and cornstarch. Thus when we stop to think about it, the recommendation to “eat food” is not as straightforward as it would seem. The food industry has created an array of products that distort the natural composition of foods. There is a variety of reasons for this (cost, storability, fortification) but the end result is that our “foods” often contain a host of ingredients that are unwanted and have undesirable or unknown effects on our health.

The broader concept is that we need to pay attention not so much to what’s in our food but where our food comes from. “You are what you eat eats.” Not all pieces of steak are created equal. Some cattle are fed almost exclusively soy, raised in cell-like quarters, and given periodic shots of hormones and antibiotics. Others are grass fed, free roaming, and organic. The concept applies to vegetables too. Vegetables from your backyard are different from those in your local Whole Foods’ organic aisle, which are different from those in the local megachain grocery store. And it extends to food products such as yogurts and orange juice.

Comment

My major criticism of “In Defense of Food” is that it is light on data. Most of its assertions are presented without supporting evidence. Those data that are provided are often interesting but seldom definitive. For example, Mr. Pollan notes that over the past two decades as the percentage of gross domestic product (GDP) spent on food has declined the proportion spent on health care has skyrocketed. (The implication is that as we have gone from high quality, expensive farm fresh foods to mass produced, cheap food products we have gotten sicker.)

But where the book lacks in data it makes up in concepts. So much of what we read today in nutrition are variations on the same theme. While a low-fat diet and low-carb diet are on one level radically different, they both follow the prevailing ideology of “nutritionism” — the notion that the effect of food on health can be understood by studying its nutritional properties. What “In Defense of Food” presents is a different point of view.

It’s easy to forget that just a generation or two ago people did not think in terms of protein, carbs, and fat. When we eat a steak or a can of beans, we often think about “getting our proteins in;” when we deny ourselves ice cream we are shying away from carbs and fats. But many of our parents and certainly our grandparents grew up in a world in which foods were just foods. It’s equally easy to forget that the value of thinking of food through the lens of nutrition is just a hypothesis. (Clearly from a macro view — in terms of prevalence of diabetes and obesity – our health has only worsened since this paradigm shift.) Simply being reminded that nutrition and food aren’t necessarily the same thing was one of my major take aways from the book.

As many observers have noted, the vast majority of clinical trials of one diet versus another have been null or negative trials. That is, most studies that try to increase carbohydrate consumption or decrease fat consumption do not show any differences in health outcomes. One intriguing explanation for this is that all these trials are just variations on the same theme. Thus the lack of positive nutrition trials indirectly lends credence to idea that nutritionism is misguided and ultimately defeatist. The problem is with the entire “Western diet” not simply a component or two of it.

To me the most convincing evidence is that we are getting nutrition all wrong is that today diseases that were once rare or nonexistent in indigenous populations have become commonplace. These “diseases of civilization” not only include diabetes and obesity but also appendicitis, cancers (including of the  breast and colon), and arguably coronary heart disease. We know this from reports of Western-trained physicians working in indigenous populations from Africa to Asia at the turn of the last century and more recently from population studies of Asian immigrants to the United States.

The simple fact is that something about our current way of life is dangerously unhealthy. While much has changed about us in the past 200 years (e.g., exposure to pollution, stress, urbanization), it seems likely that our changing diet is the most likely culprit. While many of us have focused on the changing nutrition in our diet to explain this trend, it stands to reason that we should at least consider the possibility that the changing nature of our diet is at least part of the problem.

- Shantanu Nundy, M.D.

To read Michael Pollan in his own words read ”Unhappy Meals”: http://www.nytimes.com/2007/01/28/magazine/28nutritionism.t.html?pagewanted=all

* “In Defense of Food — An Eater’s Manifesto” by Michael Pollan, April 2009, Penguin Press

** from www.zeer.com; Wonder Bread does not publish the ingredients of its products on its website www.wonderbread.com

In case you didn’t hear, the world ended last week. No, I’m not talking about the film “2012.” I’m talking about the latest guidelines from the United States Preventive Task Force (USPSTF) on breast cancer screening.

The guidelines, which came out early last week to much fanfare, make two major changes to the previous USPSTF recommendations from 2002. One, the task force recommends that routine screening for breast cancer begin at age 50, rather than age 40. Two, screening should occur every two years, rather than every one to two years.

My 2 cents are that 1) the change is not as major as we are being led to believe and b) the change is in the right direction because it recognizes the limitations of mammography in younger women and increases individualized decision-making.

First off, how significant is this change in guidelines? While many women are accustomed to the idea getting mammograms every year starting at age 40, we have known for some time that the data supporting screening in women ages 40-49 and screening every year are limited. In fact, in the previous guidelines, while the USPSTF recommended screening beginning at age 40, it noted that the “precise age at which the benefits from screening mammography justify the potential harms is a subjective judgment and should take into account patient preferences.” While studies have shown that mammography in women ages 50-69 reduces deaths from breast cancer by about 33 percent, most studies in women ages 40-49 have shown no mortality benefit. Although pooling all these smaller studies together shows a 15 percent decrease in deaths, this must be weighed against the increased risk of false positives, unnecessary biopsies, and treatment of cancers that if left alone would not cause any harm. In terms of screening interval, the previous guidelines wrote, “In the trials that demonstrated the effectiveness of mammography in lowering breast cancer mortality, screening was performed every 12-33 months… there is little evidence to suggest that annual mammography is more effective than mammography done every other year.” Thus the previous guidelines recommended screening every 1-2 years, while the guidelines now call for screening every 2 years. Thus the guidelines are a small change from what doctor offices should have been doing — that is, individualized decision-making in younger women and not-necessarily-annual mammograms in older women.

Second, why do I think the new guidelines are better? One of the more sobering facts those of us passionate about preventive health have come to realize is that the benefits of cancer screening have been oversold. Cancer screening, while effective, is limited and certainly no panacea. Even if we increased mammography rates to 100 percent, there would still be lots of breast cancer and lots of death from breast cancer. As described in a recent NY Times article, there are really three types of cancer. The first are cancers that grow so quickly that no matter when we catch them the outcome is the same. The second are cancers that grow so slowly (or even regress) such that we are almost better off not finding them (we’re better off because once found we treat them with aggressive measures — surgery, chemotherapy, and radiation — without any improvement in outcome). And the third are cancers that grow moderately fast and that if found early can be effectively treated and even cured. Unfortunately, with our current state of technology, we cannot distinguish between these three types of cancer at diagnosis. Thus once found, all cancers are treated the same, even though screening and early detection only benefit the third type of cancer.

So part of this change in the guidelines reflects the fact that mammography is imperfect. At the same time, it also reflects the fact that mammograms are not benign procedures. To prevent one death from breast cancer in women ages 40-49, 1900 women would need to be screened for 10 years, of which hundreds would undergo unnecessary biopsies and a few would undergo unnecessary treatment with breast surgery and radiation. It not just the financial cost that makes mammography in lower risk women not make sense. Dollars and cents aside, from a purely patient care perspective, it is not at all obvious that the benefits of screening in the average woman ages 40-49 is worth the harms.

In the end, the guidelines are simply that — guidelines. The USPSTF is not arguing that women should not be screened for breast cancer in their 40s or should not be screened every year. What the task force is saying is that mammograms should not be routinely recommended to women ages 40-49 and should not routinely be done more frequently than every 2 years. Instead, women ages 40-49 should have a conversation with their doctors about screening. In that conversation doctors should explain the risks and benefits of screening, and women should discuss their family history of breast cancer, their risk factors, and their preferences, and come up with a shared decision. (That sounds familiar doesn’t it? Yes, because that’s what doctors are supposed to do with every patient every time for every test.)

What this comes down to then is the word routine. Routine means everybody gets it, that doctors order it without thinking about it, that patients get it without necessarily be explained the full benefits and harms. In busy office practices, this means that patients have mammograms scheduled by office coordinators or nurse practitioners, or by the radiology suite that did their mammogram last year, often without even a word in passing between the patient and the doctor. It also means, in our increasingly quality improvement driven environment, that insurers and the accreditation agencies evaluate doctors based on how many of their female patients over a certain age get their mammogram within the recommended screening interval. Routine is a buzz term for the standardization of a medical service in order to apply it uniformly to a defined population. Put this way, given the data above, mammograms should not be routine in women ages 40-49. Mammograms should not be done uniformly, but rather be an individualized decision, just as the previous 2002 USPSTF guidelines had called for.

If we step back and are honest with ourselves, we would ideally have such a conversation about mammograms in all women we order it for. Mammograms are not a benign procedure and require some explanation and discussion. But the reality of medical practice is that we don’t have the time to counsel every patient about every test we do. Thus we consign ourselves to only consenting patients for those procedures that are out-of-the-ordinary, that are not routine, in hopes that we strike a balance between doing what makes sense for most patients and doing what makes sense for the patient in front of us.

Put in this light, this change in the recommendation for breast cancer screening — from a routine screening in women ages 40-49 to a non-routine procedure — compels us that for the marginal case – a woman in her 40s for whom screening is not the obvious choice — we sit down and have a honest conservation about what makes sense for her. And that I think is never a bad idea.

- Shantanu Nundy, M.D.

USPSTF breast cancer screening guidelines: http://www.ahrq.gov/clinic/USpstf/uspsbrca.htm

NY Times article about the new guidelines: http://www.nytimes.com/2009/11/17/health/17cancer.html?bl

NY Times article about data behind new mammography guidelines: http://www.nytimes.com/2009/11/23/health/23cancer.html?ref=health

NY Times articles about history of mammogram guidelines: http://www.nytimes.com/2009/11/22/weekinreview/22kolata.html?ref=health

NY Times Op-Ed piece, “Addicted to Mammograms”: http://www.nytimes.com/2009/11/20/opinion/20aronowitz.html?sudsredirect=true

My blog post on pros and cons of mammograms: http://beyondapples.org/2009/07/14/the-case-of-the-woman-who-refused-her-mammogram/

One of the recurring themes of BeyondApples.Org is that preventive health guidelines are and should be based on solid evidence. Most people in the general public appreciate that new drugs are rigorously tested in clinical trials — highly controlled experiments in which some people are given the new drug and others an old drug or placebo in order to measure its true effects. But many do not realize that similarly designed experiments are done to test for non-pharmacologic interventions, such as advice to quit smoking or lose weight. Thus when doctors tell us to stop smoking or exercise, it isn’t just lip service. When done in an evidence-based way, their recommendations are based on scientific data that shows that for people like us this advice has a real health benefit. In this blog entry, I discuss a groundbreaking study that shows that through intensive lifestyle interventions diabetes can be prevented. This topic is timely because today, November 14th 2009, is World Diabetes Day (www.worlddiabetesday.org).

Diabetes is a terrible disease. It affects 8 percent of the U.S. population and is a leading cause of end-stage kidney disease, blindness, and heart attacks. Overall diabetes is the fifth leading cause of death worldwide. Although effective treatments exist, like most diseases it is better not to get it in the first place. Controlling diabetes usually require lifelong medications and often daily insulin injections, and even those with well-controlled disease have frequent appointments with an array of specialists. We have known for years what the major risk factors for diabetes are*: family history, overweight (body mass index or BMI > 25 kg/m2), abdominal obesity (waist circumference > 35 inches in women, > 40 inches in men), sedentary lifestyle (low levels of physical activity), and impaired glucose tolerance (sugar levels that are abnormally high but not enough to diagnose diabetes). Given these risk factors, the big question on many people’s minds is can diabetes be prevented?

To evaluate this question, researchers in the late 1990s conducted a now landmark clinical trial called the Diabetes Prevention Program or DPP. In this study, over 3200 people at high risk for diabetes were randomly assigned to one of three groups — intensive lifestyle modification, daily oral medication, or placebo — and then followed for an average of three years. People in the lifestyle modification group were enrolled in a wellness program with the goal of achieving and maintaining a 7 percent weight loss through a low calorie, low fat diet and moderate intensity exercise (at least 150 minutes a week of brisk walking). Those in the medication group were given metformin, a medicine that is often used to treat diabetes. And those in the placebo group were given a placebo or dummy pill.

The results were impressive. Over the course of the study, about one-third of the people in the placebo group developed diabetes. In comparison, those taking metformin had a 31 percent lower risk of developing diabetes. Even more remarkable was that those in the lifestyle modification group were 58 percent less likely to develop diabetes than placebo. Weight loss was the most important predictor of risk of diabetes — for every kilogram (2.2 lbs) of weight loss risk of diabetes dropped by 16 percent. The study, which was published in the New England Journal of Medicine in 2002, showed that diabetes is preventable. For every 7 people that participate in similar lifestyle-intervention program over three years, one case of diabetes would be prevented (number needed to treat or NNT=6.9). Through an intense dieting and exercise program, it is possible to keep at-risk people from developing diabetes. 

When the study came out many experts wondered how long the benefits would last. Are we simply delaying the onset of diabetes or are we actually preventing it? Just this past month a follow up study was published in Lancet by the DPP group that helps answer that question. After 3 years in the lifestyle, metformin or placebo group all study participants were enrolled in a less-intense lifestyle counseling program for an average of 7 years. (Those in the metformin group received both medication and counseling.) They found that over the 7 years of follow up the incidence of diabetes in the three groups was the same — not too surprising given that all the participants received the same treatment. However, what was remarkable was that the differences in rates of diabetes from the first 3 years of the study persisted over the entire 10 years. Overall the rate of diabetes was 18 percent lower in the metformin group and 34 percent lower in the intensive lifestyle group. Thus the study concluded that the “prevention or delay of diabetes with lifestyle intervention or metformin can persist for at least 10 years.”

This study provided many new insights into diabetes prevention. However, science is a iterative, continuous process; like all studies with these results came new questions:

1. The lifestyle intervention in this study was intensive. Each patient in the lifestyle group participated in a 16-lesson curriculum covering diet, exercise, and behavior modfication taught by case managers on a one-to-one basis over 24 weeks. After this patients received individual sessions (usually monthly) and group sessions designed to reinforce the behavioral changes. This begs the question: is the intensity of this intervention realistic for most primary care practices? If not, can less intense interventions have similar impact?

2. The participants were at high risk for diabetes. They not only were overweight but already had signs of impaired glucose tolerance (called pre-diabetes). In fact, the study found that one-third of those with the placebo group developed diabetes over three years (10 percent risk per year). Do the results of this study apply to people at lower risk (e.g., overweight with normal blood sugars, family history of diabetes)?

3. This study showed conclusively that diabetes can be prevented. But in the end, the goal of preventive health care is to improve health, not necessarily to prevent diagnoses. So did the participants in the lifestyle modification group have improved health outcomes (e.g., less heart attacks, lower mortality)?

Over the next several years researchers in the DPP group and elsewhere will no doubt use science to answer these and other critical questions regarding diabetes prevention. Still I think those of us outside the walls of research can take home some important messages: a) diabetes is preventable, b) diabetes can be prevented by eating healthy and exercising, and c) that these benefits are long-lasting.

As we observe World Diabetes Day (www.worlddiabetesday.org) let us bear in mind these critical observations. I can think of no better way to “celebrate” World Diabetes Day than to renew each of our committments to a healthful lifestyle — a lifestyle which amongst other things will help us steer clear of diabetes and other chronic diseases.

- Shantanu Nundy, M.D.

* There are two major types of diabetes — type 1 or insulin-dependent diabetes and type 2 or non-insulin-dependent diabetes. Type 2 diabetes is by far more common and is associated with weight gain. In the blog entry, we use the term diabetes loosely to refer to type 2 diabetes.

To learn more about diabetes visit NIH ”Diabetes Overview” at http://diabetes.niddk.nih.gov/dm/pubs/overview/.

The lifestyle intervention course offered to participants in the DPP trial are available online at: http://www.bsc.gwu.edu/dpp/manuals.htmlvdoc.

To access original NEJM DPP article: http://content.nejm.org/cgi/content/abstract/346/6/393.

To access DPP followup article in Lancet: http://www.thelancet.com/journals/lancet/article/PIIS0140-6736(09)61457-4/abstract.

To access accompanying Lancet editorial on preventing diabetes: http://www.thelancet.com/journals/lancet/article/PIIS0140-6736(09)61631-7/fulltext.

Recently my wife, who is a cancer doctor, took care of a middle-aged woman with cervical cancer. As someone whose job it is to treat cancer, my wife took at face value that the woman had cervical cancer and focused her efforts on how to best help the patient overcome her terrible disease. As someone interested in preventive health, I took a different focus — how is it that this woman got cervical cancer at all? With appropriate screening, cervical cancer is almost entirely preventable (see http://beyondapples.org/2009/05/03/pap-smears-re-examined/). Since the advent of Pap smears rates of cervical cancer have plummeted, taking it from the number 1 most common malignancy in women to not even being in the top 10. These days, the 11,000 or so cases of cervical cancer annually almost exclusively occur in women who have never gotten a Pap smear or who get them infrequently. And sure enough, this patient was no different — her last Pap smear was during her last pregnancy over 20 years ago.

In considering the patient further, I naturally assumed that she had no primary care doctor. As we hear about almost daily on the news, there are millions of people in the United States today who because of lack of insurance and lack of access have no primary and preventive care. It is these people who are most at risk for preventable diseases such as cervical cancer and who sadly bear a disproportionate amount of the disease burden. But this patient HAD a primary care doctor. In reviewing her past medical history, my wife noted that the patient had high blood pressure and depression for which she was taking chronic medications. I stopped her right there: “You mean to say that this patient has a regular doctor and yet hasn’t had a Pap smear in over 20 years!?”

How could this be? How could a woman with a regular doctor not get a routine Pap smear — a universally accepted, life-saving preventive health measure – in over two decades? This past week an illuminating study was published in the Annals of Internal Medicine that evaluated how good primary care doctors are at following the screening guidelines for cervical cancer (see http://www.annals.org/content/151/9/602.abstract). In the study, primary care doctors were mailed a questionnaire with four common clinical scenarios and asked what the appropriate screening step would be. As an example, in one scenario, participants were presented with an 18-year old girl with first sexual intercourse 1 month ago and asked if she should get no Pap, Pap every 3 years, Pap every 2 years, Pap annually, or other. Although the guidelines from every major expert group state that cervical cancer screening should at age 21 or 3 years after first sexual intercourse, whichever comes first, almost half the doctors surveyed recommended Pap smear. Overall only 22.3 percent of clinicians answered all 4 clinical scenarios correctly. Put bluntly, the study found that less than 1 in 4 doctors know and practice the guidelines.

In many ways, this information is not a surprise. We already know that people consistently do not receive proven preventive health services. Thousands of people are not taking aspirin to prevent first heart attacks, getting routine mammograms, or receiving vaccines against common adult diseases. I started BeyondApples.Org to help tackle this very problem, to help get the word out among doctors and patients about the preventive health services we all need. But at the same time, this study was sobering and discouraging. This study was performed under artificial “test-like” conditions that if anything give doctors the benefit of the doubt. One could look at the statistic that less than half of Americans are getting appropriately screened for colon cancer and rationalize it in any number of ways: the doctor concluded that the guideline did not apply to the individual patient, the patient refused screening, the patient couldn’t afford to pay for the test. But you can’t make these excuses for the results of this study. Clinical vignettes such as the one used in this study are in many ways optimal settings to test clinical knowledge because they circumvent all these real-world confounders. The simple conclusion we are left with is that doctors just don’t know enough about preventive health.

We can argue about why this is the case. Clearly not all the blame falls on doctors. The science of medicine changes daily, different expert organizations put out conflicting guidelines, the health care system is broken. We can also debate solutions — we need better point-of-care decision support, better continuous medical education, better reimbursement for health maintenance. But from the patient’s perspective arguing about causes and debating solutions seems inadequate. For the patient that my wife is caring for these discussions are too late.

For the person out there wondering whether he or she is getting the right preventive care, I offer a simple solution. Get informed. Use the checklist we developed (www.beyondapples.org/checklist). Find out what preventive health services are recommended for you. And then partner with your doctor to make sure you are getting the best preventive health care possible.

- Shantanu Nundy, M.D.