Every week, we email a case-based board-style question to tens of thousands of clinicians as part of our NEJM Knowledge+ Question of the Week program. Our initial offering focused on internal medicine, but we’ve expanded to include family medicine board review questions as well.
So far, the response to our family medicine content has been overwhelmingly positive. In May, we started inviting respondents to comment publicly on the Question of the Week, resulting in engaging and informative dialogue. Clinicians are using the space to highlight areas of controversy, ask their colleagues for advice, and share their clinical knowledge.
Through that forum, we’ve received requests for more content in pediatric and adolescent medicine, more geriatrics, more cardiovascular disease, and more mental health. We promise to deliver in all those areas and more, but in the meantime, let’s take a look at how folks did on the 23 family medicine board review questions that we’ve featured so far in NEJM Knowledge+ Question of the Week.
Are You Challenged by Our Family Medicine Board Review Questions?
Each week, about 14,500 physicians answer our featured question — and the average score for those questions is just over 50%. As you might expect, we see a lot of variation from question to question — with some answered correctly by 80% of respondents and some by as few as 20%.
When the vast majority of respondents get a particular question right, we reason that the case must have been straightforward and the correct answer fairly clear-cut.
For example, in mid-August, we asked for the next step in evaluation for a 72-year-old woman with intermittent upper abdominal discomfort accompanied by fatigue and diaphoresis who had normal liver-function tests, normal pancreatic enzyme levels, and a normal CT scan of the abdomen. The correct answer? An electrocardiogram to rule out myocardial infarction. Nearly 80% of respondents answered correctly, and several told us that while the case was simple, it was a good reminder to be vigilant for atypical signs of MI, particularly in women and elders.
These were our easiest cases, with clear patient histories and presentations along with uncontroversial answer choices. Not all of our questions are so straightforward — others have proven troublesome, especially the ones hinging on recent (and contested) cholesterol guidelines.
Why Are We Testing on the Lipid Guidelines?
Writing a case-based question that draws on a controversial guideline is difficult, but we think it’s worthwhile because it helps physicians to (a) be aware of major guidelines on important topics and (b) understand where their clinical decisions are (or are not) in alignment with those guidelines. With that rationale in mind, we featured two questions this year related to the 2013 AHA/ACC guidelines on lipid management.
Measuring Lipid Levels
In one case, we asked readers when to remeasure lipid levels in a healthy 42-year-old woman whose LDL-cholesterol level was 157 mg/dL when last measured 2 years ago. About 20% of respondents answered in accordance with guidelines that they would remeasure this patient’s lipid levels in 2 years — but over half answered that they would measure her levels now, at the current visit.
A healthy 42-year-old woman had an LDL-cholesterol level of 157 mg/dL when measured 2 years ago. When should her lipid levels be remeasured?
a) When she develops a cardiovascular risk factor (9%)
b) When she becomes postmenopausal (9%)
c) Now (53%)
d) In 2 years (21%)
e) When she reaches age 50 (8%)
The ACC/AHA guidelines recommend that patients like this one — age 20 to 79 without ASCVD — have their lipid levels measured every 4 to 6 years, so that their 10-year risk of ASCVD can be recalculated. However, many patients still expect annual lipid assessments, and many clinicians still provide them, thus creating a disconnect between guideline-recommended practice and real-world clinical practice.
In a second case related to the 2013 ACC/AHA lipid management guidelines, we asked our audience this whopper of a question:
Which one of the following options is most appropriate for a 46-year-old man with a BMI of 33, a blood pressure of 138/86 mm Hg while taking daily hydrochlorothiazide, a family history of early cardiac disease, an LDL-cholesterol level of 150 mg/dL (reference range, <130), a high-sensitivity C-reactive protein level of 3.10 mg/liter (0.02–8.00), a fasting glucose level of 97 mg/dL (70–100), and an estimated 9.3% 10-year risk for atherosclerotic cardiovascular disease, who stopped rosuvastatin after reading that it increases the risk for type 2 diabetes?
a) Initiate ezetimibe and valsartan (9%)
b) Continue to withhold rosuvastatin and initiate atenolol (3%)
c) Continue to withhold rosuvastatin and initiate aspirin (8%)
d) Continue to withhold rosuvastatin and initiate metformin (5%)
e) Reassure the patient and restart rosuvastatin (75%)
Three quarters of respondents answered in a manner consistent with guidelines, choosing to reassure the patient and restart rosuvastatin — but the online comments about this question were filled with concerns, mostly centered on two themes:
- Lifestyle modification should be recommended for this patient.
- Aspirin would be a better choice than a statin for this patient — or should at least be used in addition to the statin.
Given these insightful comments, we carefully reevaluated the question and ultimately decided to rewrite it. The new question (shown below) still tests knowledge of the ACC/AHA lipid management guideline — but in a much more straightforward way.
Which one of the following options is most appropriate for a 46-year-old black man with a family history of early cardiac disease, a BMI of 33, a blood pressure of 138/86 mm Hg while taking daily hydrochlorothiazide, an LDL-cholesterol level of 126 mg/dL (reference range, <130), a triglyceride level of 260 mg/dL (<150), a fasting glucose level of 90 mg/dL (70–100), and an estimated 9.3% 10-year risk for atherosclerotic cardiovascular disease who has tried, with minimal success, to change his diet and exercise level in the past year?
a) Initiate simvastatin 40 mg once daily
b) Initiate simvastatin 10 mg once daily
c) Initiate gemfibrozil 600 mg twice daily
d) Initiate metformin 500 mg twice daily
e) Initiate lisinopril 10 mg
In this newly written case, the patient clearly meets criteria for moderate-intensity or high-intensity statin therapy, according to the ACC/AHA guidelines: he has an LDL-cholesterol level between 70 and 189 mg/dL, he does not have diabetes mellitus or clinical ASCVD, and he has an estimated 10-year ASCVD risk (based on the ACC/AHA risk calculator) of ≥7.5%. Therefore, a moderate-intensity statin such as simvastatin 40 mg is a reasonable choice; the 10-mg dose is considered low-intensity.
This new version of the question is now in NEJM Knowledge+ Family Medicine Board Review, and we’ll be tracking it closely to ensure that physicians find it both challenging and valuable.
Less Controversial, But Still Difficult
Of course, most of the NEJM Knowledge+ questions that prove difficult are not necessarily controversial. Many are backed by multiple guidelines and based on current standards of care. The following two examples fall into this category.
Glycemic Control in Critically Ill Patients
On September 8, we featured a question on glycemic control in a critically ill patient with persistent hyperglycemia and no known history of diabetes mellitus. Only 30% of respondents correctly answered that they would provide IV insulin with a target glucose level of 140 to 200 mg/dL. Most of the others preferred sliding-scale insulin or wanted to initiate treatment only if the glucose level exceeded 220 mg/mL.
What is the recommended approach for glycemic control in a 42-year-old previously healthy man who is admitted to the intensive care unit for severe community-acquired pneumonia and whose random blood glucose levels are 188, 193, and 212 mg/dL (reference range, <140) during the first few hours after admission?
a) Intravenous insulin infusion with a target glucose level of 110 to 140 mg/dL (18%)
b) Subcutaneous basal glargine insulin injection, with daily adjustment until the morning glucose level is c) Sliding-scale insulin injection (in which dose is dependent on the glucose value) whenever glucose levels exceed 180 mg/dL (25%)
d) Continued monitoring, with treatment initiated only if the glucose level exceeds 220 mg/dL (22%)
e) Intravenous insulin infusion with a target glucose level of 140 to 200 mg/dL (30%)
Several respondents asked online about the evidence to support our correct answer and expressed concern about hypoglycemia with IV insulin. James Hennessey, an endocrinologist at Beth Israel Deaconess Medical Center and a section editor for NEJM Knowledge+, explains:
“A recent review of the eight contemporary U.S.–based guidelines on this topic illustrates the remarkably consistent recommendations made in not only the medical ICU but also surgical intensive care and non-ICU settings.
“When specified, hyperglycemia has been designated in all current recommendations to be present when the blood sugar is persistently >140 mg/dL. Recommended targets for blood sugars in the hospital setting are remarkably similar, with the American Diabetes Association report listing a target of 140–180 mg/dL and other guidelines listing 140–200, <150, or 180 mg/dL in the three remaining guidelines that note a starting threshold.
“Concerns about hypoglycemia with IV insulin are readily recognized in all of these guidelines. Continuous insulin infusion is considered to be the safest approach because the infusion (of short-acting insulin) can be readily shut off, thereby avoiding the prolonged insulin action seen with glargine administration and the reactive dosing seen with sliding-scale insulin administration.”
Treating Alcohol Dependence
On June 23, we described a patient with cirrhosis, esophageal varices, and acute alcoholic hepatitis who had a strong desire to stay sober but was worried about the possibility of relapse. When asked which medication would be most appropriate for this patient, most respondents chose disulfiram, but both this agent and naltrexone are hepatotoxic and should be avoided in patients with advanced liver disease. The more appropriate choice in this case is acamprosate, which is approved by the FDA for treating alcohol dependence and is contraindicated only in patients with severe renal insufficiency.
Which one of the following medications is most appropriate to help prevent a relapse of alcohol dependence in a patient with cirrhosis, esophageal varices, and acute alcoholic hepatitis?
a) Acamprosate (27%)
b) Flumazenil (5%)
c) Lorazepam (10%)
d) Disulfiram (38%)
e) Naltrexone (20%)
Testing yourself with challenging questions like these — and the many others offered through NEJM Knowledge+ — can help you improve your recall for the next time you need to call up this information, whether on a board exam or in a clinical setting.
Our Family Medicine Board Review Questions Help You Learn
The NEJM Knowledge+ Question of the Week provides an excellent way to test your core knowledge of family medicine topics. We draw from our bank of thousands of case-based questions across a range of diagnoses, settings, and patient demographics and aim to ensure that our questions are not only accurate and clear but also relevant to your daily practice.
Challenge yourself by signing up for NEJM Knowledge+ Question of the Week — it’s free, and it’s fun! If you’ve already signed up, let us know what you think of our questions and what we can do better.
Want more? NEJM Knowledge+ Family Medicine Board Review includes more than 1,500 case-based questions just like these. And it serves them up in multiple alternate formats on an adaptive platform that responds to your strengths and weaknesses to help you consistently improve and retain your knowledge. See a demo here.