Cardiovascular Risk Factors Among Internal Medicine Residents


  • Nicole L. Mihalopoulos MD, MPH,

    1. From the Department of Pediatrics, University of Utah, Salt Lake City, UT;1 and the Department of Epidemiology, Tulane University Health Sciences Center, New Orleans, LA2
    Search for more papers by this author
  • and 1 Gerald S. Berenson MD 2

    1. From the Department of Pediatrics, University of Utah, Salt Lake City, UT;1 and the Department of Epidemiology, Tulane University Health Sciences Center, New Orleans, LA2
    Search for more papers by this author

Nicole L. Mihalopoulos, MD, MPH, University of Utah Hospital, 50 North Medical Drive, 2A200 SOM, Salt Lake City, UT 84132


Modification of risk factors can reduce the number of deaths due to cardiovascular disease (CVD). Internal medicine (IM) residents devote significant clinical time to help patients modify CVD risk factors but may fail to recognize the presence of such factors in their own lives. The prevalence of major modifiable risk factors was assessed in IM residents. Of 101 eligible residents, 56 completed at least 1 component of the study. None had symptoms or history of CVD, and 11 (20%) had ≥2 CVD risk factors. Of 39 subjects who had physical assessment, 3 had systolic hypertension and 13 had a body mass index ≥25. Of 38 patients with fasting lipid measurements, 13 had total cholesterol >200 mg/dL; 28 had low-density lipoprotein ≥100 mg/dL, with 7 >160 mg/dL; and 7 had high-density lipoprotein <40 mg/dL. Three residents smoked 10 cigarettes per day and 15 were sedentary. Training in IM should include strategies to increase awareness of modifiable personal risk factors for CVD, as well as strategies to reduce or eliminate them.

Cardiovascular disease (CVD) is the most common cause of death among adult men and women in the United States. There are several modifiable risk factors for CVD identified by the American Heart Association: cigarette smoking, obesity, physical inactivity, poor nutrition, diabetes, high blood pressure (BP), and high blood cholesterol. Identification of these risk factors is a major part of caring for patients in internal medicine (IM) practice. Internists play an important role in influencing adult patients to modify cardiovascular (CV) risk factors.1,2 Further, internists who engage in behaviors that decrease their own CVD risk are more likely to counsel their patients to reduce their risks.3

IM residents have the potential to spend a significant portion of their clinical training learning how to counsel patients regarding decreasing their risk for CVD. Like experts in practice, the personal health habits and CV risk factors of IM residents such as smoking, physical activity, and nutrition are likely to affect how they counsel patients and the areas of prevention they choose to emphasize. There is little information in the medical literature regarding the prevalence of CV risk factors and their level of interest in modifying them among residents. A recent study by Yi and colleagues4 describes the self-rated health of residents and found that IM residents have a lower self-rating score compared with other specialties. The objective of our study was to assess CV risk factors among IM residents at a large university-based medical center in the southern United States.


All IM residents (including chief residents) were eligible for participation. For the purposes of this study, a resident is defined as a person currently enrolled in an American College of Graduate Medical Education accredited postgraduate training program working toward board certification in IM.

Recruitment and Retention

An introductory letter describing general details of the study was sent to all IM residents, followed by e-mail reminders. Consent and survey forms were completed at an IM noon conference. An additional effort to improve participation was conducted by a final recruitment letter and e-mails after the conference. The institutional review board of Tulane University approved the study. A follow-up report of risk factor findings was sent to each participating resident.

Residents were able to report to the General Clinical Research Center (GCRC) in the main hospital of their training and as their schedules allowed. Confidentiality was maintained by assigning random 4-digit study codes to each resident, which were used for all surveys and clinical assessment forms.


Participants completed a survey and clinical assessment. This included a physical examination (height, weight, BP, pulse, waist circumference), laboratory studies, dual energy x-ray absorptiometry (DEXA), and electrocardiography (ECG). Three GCRC nurses performed the physical examinations, ECG, and phlebotomy. A certified DEXA technician performed DEXA scans.

We defined modifiable risk factors using criteria set by the National Heart, Lung, and Blood Institute (NHLBI) and the Centers for Disease Control and Prevention (CDC). These included abnormal lipid concentrations (low-density lipoprotein cholesterol [LDL-C] ≥100 mg/dL, high-density lipoprotein cholesterol [HDL-C] <40 mg/dL for men and <50 mg/dL for women, triglycerides >150 mg/dL, and total cholesterol >200 mg/dL), systolic BP ≥130 mm Hg, diastolic BP ≥85 mm Hg, body mass index (BMI) ≥25.0, smoking (≥1 cigarette per month), and sedentary lifestyle (<30 min/d, <5 d/wk).5–8


Surveys from the Bogalusa Heart Study9 were adapted to gather information regarding medical history, current medications (specifically for diabetes mellitus, hypertension, dyslipidemia, and thyroid disorders), tobacco use (frequency of use and ever use), and physical activity. The physical activity assessment included time spent in specific activities, number of stairs climbed per day, and amount of time spent in sedentary behavior (watching television, sitting at a computer).

Physical Examination

A standardized protocol for the physical examinations (anthropometry, BP, pulse, skinfold, waist circumference) was used.10,11 Patients wore a hospital gown without shoes. Weight (kg) was measured once using a standard physician's scale (Model 5002, Serial #5002–2873; Scale-Tronix, Inc, Wheaton, IL). Height (cm) was measured once using a standard wall-mounted stadiometer. Right arm BP (mm Hg) was measured 3 times with residents seated and rested by trained observers with a mercury sphygmomanometer. Pulse rate was measured in increments of 30 seconds. The average of 3 BP and 3 pulse rate values were calculated. Waist circumference (cm) was measured 3 times with a plastic tape and averaged.

Laboratory Studies

Fasting serum lipid and lipoprotein (triglycerides, LDL-C, HDL-C, and total cholesterol) concentrations were determined by the laboratory at the Medical Center of Louisiana at New Orleans.

Dual Energy X-Ray Absorptiometry

DEXA scan results determined body mass composition and calculation of percentage of body fat.


Participants had a standard 12-lead resting ECG performed. ECG findings were interpreted by a cardiologist (GB) who was blinded to the other measures.

Statistical Analysis

Descriptive statistics were calculated for BMI, percentage of body fat, waist circumference, BP, pulse rate, lipids, and lipoproteins. Physical activity was categorized as sedentary, low, moderate, or high based on CDC criteria.8 Smoking status was categorized as nonsmoker, previous smoker, or current smoker (at least 1 cigarette per month).5



Of 101 eligible residents, 56 (54% male; mean age, 30.4 years [SD 3.02], 59% white, 12.5% East Indian, 11% Asian, 9% black) completed either the survey or parts of the physical examination. Of these, 22 completed the entire clinical assessment and the survey and 16 did not complete any component of the clinical assessment.

Risk Factors

None of the residents had symptoms or a history of CVD. Prevalence of CVD risk factors is given in percents where n represents the number of participants who underwent measurements for each risk factor. As shown in Figure 1, 20% (n=11) had ≥2 modifiable CVD risk factors. The most common risk factor was elevated LDL-C (>100 mg/mL), present in 75% of the participants (n=38). Borderline high total cholesterol (>200 mg/dL) was present in 63% of participants and elevated total cholesterol (>240 mg/dL) in 26% (n=38). Excessive body fat (≥32% for women, ≥25% for men) was present in 50% of participants (n=26), 7 women and 6 men.12,13 Being overweight or obese (BMI ≥25) was the next most common risk factor (33%, n=39), followed by low HDL-C (28.9%, n=38), sedentary lifestyle (28%, n=53), and elevated total cholesterol levels. The least common risk factors were systolic hypertension (>140 mm Hg) (7%, n=40), cigarette smoking (7%, n=53), and elevated triglyceride level (>150 mg/dL), present in 1 participant (n=38). None of the participants had abnormal ECG findings (n=36) or a prior diagnosis of diabetes mellitus, hypertension, or dyslipidemia. Means and SDs of each risk factor are presented in the Table. Figure 2 presents the percentage of participants with each modifiable risk factor.6,7

Figure 1.

Participants with 0 to 3+ modifiable cardiovascular risk factors.

Table Table.  Modifiable Risk Factor Descriptive Statistics
Risk FactorMean (±SD)RangeResidents, % (Abnormal Value)
Body fat, % (DEXA scan) (n=26)
Total fat, %28.9 (8.84)14–43.550.0
 Male25.1 (9.0)14–38.623.1 (≥24.9%)
 Female32.2 (7.6)17.3–43.526.9 (≥31.9%)
Trunk fat, %30.0 (9.51)14.7–45.353.8
 Male28.8 (10.9)16.4–43.526.9
 Female31.1 (8.4)14.7–45.326.9
Waist circumference (n=39)
 Male89.3 (8.4)79.7–104.25.1 (>102 cm)
 Female81.1 (12.6)66.3–1207.7 (>88 cm)
Hemodynamics (n=40)
 Systolic blood pressure, mm Hg120 (14)99–1617 (>140 mm Hg)
 Diastolic blood pressure, mm Hg72 (7)57–905 (>85 mm Hg)
Heart rate, beats per min70 (11.3)49–1085 (>100 beats per min)
Lipids, mg/dL (n=38)
 Total cholesterol190 (37.1)129–27434 (>200 mg/dL)
 Low-density lipoprotein cholesterol127 (39.4)61–22475 (>100 mg/dL)
 High-density lipoprotein cholesterol50 (13.6)29–8328.9 (male, <40 mg/dL; female, <50 mg/dL)
Triglycerides73 (34.5)35–1883 (>150 mg/dL)
Sedentary lifestyle (n=53)28 (<30 minute/d, <3 d/wk)
Smoking (n=53)7 (>1 cigarette per month)
Abbreviations: DEXA, dual energy x-ray absorptiometry.
Figure 2.

Percentage of participants with each modifiable cardiovascular risk factor. Tchol indicates total cholesterol; LDL, low-density lipoprotein; HDL, high-density lipoprotein; Trigs, triglycerides; BMI, body mass index; and SBP, systolic blood pressure.


We found that 20% of IM residents had at least 2 previously unrecognized modifiable CVD risk factors. These findings are similar to studies of nonphysicians of the same age and sex.14 Earlier attempts to assess CVD risk factors among physicians or medical students include the Physicians' Health Study (PHS)15; an analysis of fitness, body fat, and perceived stress in a small group of primary care residents16; and a screening program to detect medical students at risk for coronary artery disease.17

The PHS is a long-term study of physicians that began in 1991. Compared with the PHS, we found that fewer IM residents were smokers (5.8% vs 34%) or had high BP (7% vs 12%). We found, however, that residents had a higher prevalence of elevated cholesterol (10% vs 7.9%), overweight (23% vs 13.6%), and sedentary behavior (28% vs 13.6%). This higher prevalence of risk factors is concerning given that the IM residents were younger than participants in the PHS (average age, 30 years compared with 53 years, respectively). It should be noted, however, that the PHS thresholds for overweight (BMI ≥27.8), hypertension (systolic BP ≥160 mm Hg or diastolic BP ≥95 mm Hg, or being treated), and hyperlipidemia (total cholesterol ≥250 mg/dL or being treated) were considerably higher than those used in the current study. Unlike the PHS, which concluded that the physicians in their study represent a population at lower risk for CVD than the general male population, the residents in our study appear to have CVD risk similar to the general population.14,18 The most striking finding is the 75% of participants with elevated LDL-C levels.

Previous studies that have examined the health of medical students and residents have reported conflicting results. Although Frank and colleagues19 found that few medical students had CVD risk factors, a study by Troyer colleagues20 reported that >50% of medical students had ≥1 CVD risk factors, including sedentary behavior, overweight, and dyslipidemia. A longitudinal study conducted at the University of Virginia found that residents reported increased stress and decreased physical activity at the end of internship and, by the end of residency, men weighed more and had more body fat and lower HDL-C levels.21 Our results are similar to these findings and further emphasize that IM residents are at increased risk for having unrecognized CVD risk factors.

Although it is reassuring that none of our participants had abnormal ECG findings or a history of angina, the high prevalence of risk factors combined with autopsy data indicating that among individuals who are the same age as the IM residents, approximately 70% already have significant coronary atherosclerosis suggests that they are at high risk for future serious cardiac events.22,23 This further strengthens the need to educate IM residents about their CV risk factors and ways in which they can be reduced. The fact that none of the residents in our study were aware of their risk factors indicates that few were following the screening recommendations from the NHLBI for cholesterol and BP.8,9 The reluctance to participate is an indication of lack of self-motivation to engage in preventive care for themselves and relay prevention to their medical practice. There is evidence that few residents have a primary care provider, and many feel uncomfortable seeking care for their medical conditions or do not have the time to seek care.24 These barriers to primary care result in inadequate screening and treatment of risk factors for CVD and in general medical care. Based on these observations, it might be suggested that training programs for IM should incorporate risk factor profiling and training in risk factor counseling.


There are some limitations to our study. First, our results are based on only 56 residents, and not all participants completed all aspects of the study. This may have resulted in an underestimation of risk factors among those who did not have, for example, a blood draw or DEXA scan. We did not measure serum glucose or insulin concentrations and were, thus, less able to identify residents with the metabolic syndrome. Residents who already had manifestations of the metabolic syndrome would be at even higher risk. Because our results are based on a single residency program and not all residents in the program participated, the generalizability of our results to all IM residents is limited. We found that it was very difficult to achieve high levels of participation in a postgraduate training program that is busy with clinical responsibilities. Reasons for nonparticipation or incompletion of all segments of the study included time constraints, refusal/fear of blood draw, and “not wanting to know how fat I am” by one resident. Our assessments were cross-sectional rather than serial or longitudinal. Repeat screening conducted over time may have produced different findings.

Despite these limitations, the finding that at least 20% of IM residents had CVD risk factors suggests the need for further study of IM training programs and dedication to prevention. For example, studying the effect on the behavior of the residents who were provided information regarding their risk factors with respect to either their attempts to reduce them or to counsel their patients to do so would be an important future step and potential yield from this program. Ideally, a baseline assessment of CVD risk factors and patient counseling on CVD prevention should be performed as part of “Intern Orientation” and then followed up during each year of residency, with appropriate interventions for residents and determination of the effect on patient care.


One in 5 IM residents had risk factors for CVD that were potentially modifiable. None were aware of the presence of these factors and the need to make changes in their lifestyle to reduce them. Because the health of a physician has been shown to be correlated with the prevention counseling and screening of patients seen by the physician, training in IM should include strategies to increase awareness of personal modifiable risk factors for CVD, practical plans to reduce or eliminate them, and have an impact on implementing and counseling for risk factors in their patients.25

Acknowledgments and disclosures:

We thank Paul Young, MD, Professor of Pediatrics at the University of Utah, for critical review and editing of the manuscript. None of the authors have a conflict of interest. Funding was provided by The Tulane-Charity-LSU General Clinical Research Center (grant #5M01 RR05096).