Logo of nihpaAbout Author manuscriptsSubmit a manuscriptHHS Public Access; Author Manuscript; Accepted for publication in peer reviewed journal;
Obesity (Silver Spring). Author manuscript; available in PMC 2009 Sep 28.
Published in final edited form as:
PMCID: PMC2753419
NIHMSID: NIHMS140728

Ethnic Differences in Self-reported and Measured Obesity

Abstract

As use of self-reported data to classify obesity continues, ethnic differences in reporting errors remain unclear. The objective of this study is to elucidate misreporting disparities between African Americans (AAs) and European Americans (EAs). The Pennington Center Longitudinal Study (PCLS) is an ongoing investigation of environmental, behavioral, and biological factors associated with obesity, diabetes, and other common diseases. Self-reported and measured height and weight were collected during initial screening for eligibility in various studies by telephone and clinic visits. All ethnicity–sex groups (15,656 adults aged 18–65 years, 53% obese, 34% AA, 37% men) misreported heights and weights increasingly as measured values increased (P < 0.0001). More AA vs. EA women (P < 0.001) misreported height and weight, but more EA vs. AA men misreported their weight (P < 0.02). Obesity was underestimated more in AA vs. EA women (self-reported – measured prevalence = −4.0% (AA) vs. −2.6% (EA), P < 0.0001), but less in AA vs. EA men (−3.2% (AA) vs. −4.2% (EA), P < 0.0001)). With measured obesity prevalence equalized at 53% in all groups, the self-reported obesity prevalence in women was 50.4% (AA) vs. 49.6% (EA), and in men 49.8% (AA) vs. 47.3 (EA). Underestimation in women was −2.6% (AA) vs. −3.4% (EA); in men it was −3.2% (AA) vs. −5.7% (EA), P < 0.003. Self-reported height and weight portend underestimation of obesity prevalence and the effect varies by ethnicity and gender. However, comparisons depend on the true prevalence within ethnicity–gender groups. After controlling for obesity prevalence, disparity in underestimation was greater in EA than in AA men (P < 0.003) but not women.

INTRODUCTION

Population surveillance and clinical screening for overweight and obesity is a major public health priority (1,2), and current recommendations support the use of the BMI for these efforts. In many cases, due to logistical or fiscal constraints, there is a reliance on self-reported rather than directly measured values for height and weight to estimate BMI and the prevalence of overweight and obesity (3-5). There is good evidence that relying on self-reported height and weight produces a lower mean BMI and, hence, a lower prevalence of overweight and obesity (6). These findings have been reported in many countries throughout the world and some report that attenuating effects on obesity prevalence vary with age, ethnicity, gender, and level of adiposity (7-21).

An important consideration is whether there are ethnic differences in the relationship between self-reported and measured data for both height and weight. The existence of such differences and subsequent effects on calculated BMI have important consequences for both the population surveillance of obesity and the clinical assessment of obesity-related health risks if self-reported information continues to be used. Conflicting results persist; for example, Rowland did not find significant ethnic differences in self-reporting errors whereas Villanueva reported that among men more non-Hispanic AAs and Mexican Americans (MAs) overestimated their weight compared to non-Hispanic EAs, but ethnic differences were not significant in women (15,20). Gillum and Sempos found that obesity prevalence was consistently and substantially underestimated for women using self-reported vs. measured height and weight with the gap being greatest for MA. Underestimation was also noteworthy for MA men, but estimates of obesity prevalence based on self-reported vs. measured data were equivocal for men in other ethnic groups (10). The purpose of this study was to further investigate ethnic differences between AA and EA men and women in self-reported vs. measured height, weight, BMI, and the subsequent estimates of prevalence of overweight and obesity. The focus was on determining whether ethnic disparities were attributable to true reporting differences or in fact associated with other ethnic differences.

METHODS AND PROCEDURES

Source of data

The Pennington Center Longitudinal Study (PCLS) is an ongoing investigation of obesity, lifestyle, and the development of cardiovascular disease, type 2 diabetes, cancer, and other common diseases. The PCLS sample comprised volunteers who have participated in clinical studies conducted at the Pennington Biomedical Research Center since 1992. The present investigation is limited to cross-sectional analyses of the sample at baseline (1992–2008). The Pennington Biomedical Research Center, located in Baton Rouge, Louisiana, is part of the Louisiana State University System. Thus, most of the participants reside in the Baton Rouge area.

Study sample

The sample includes 15,656 adults aged 18–65 years (mean 39.9 years (95% confidence interval: 39.7–40.1)); 5,784 were EA women, 4,013 AA women, 4,494 EA men, and 1,365 AA men. Participants in the clinical trials were recruited and screened in a multistep process. As individuals showed interest in participating in a specific study, they were interviewed by telephone to verify that study-specific inclusion criteria were met before the volunteer was asked to come into the clinic. During this interview, the volunteer's age, gender, ethnicity, health history, and height and weight were recorded. Once the person was deemed eligible to participate in a particular study, a screening appointment was made in which the individual was asked to come into the clinic for further evaluation. At the initial screening visit, the participants provided written informed consent, and a physical examination was conducted that included anthropometric measurements, blood pressure assessment, and phlebotomy.

Anthropometry

During the telephone screening interview, self-reported height and weight were recorded in inches and pounds, respectively. These measures were converted to meters and kilograms, respectively, and self-reported BMI was calculated as weight in kilogram/height in meter2. During the initial visit to the Pennington Biomedical Research Center screening site, a nurse measured the volunteer's height to the nearest 0.1 cm using a stadiometer. The participant was required to remove his or her shoes and have his or her heels, buttocks, and upper part of the back remain in contact with the stadiometer. The subject was asked to inhale and hold his or her breath, while a second nurse lightly applied traction to the patient's head to maintain alignment with the Frankfort Plane. This process was repeated, and the average of the two heights was recorded. The subject's weight was measured to the nearest 0.1 kg using a digital scale after he or she removed outer clothing, heavy pocket items, and shoes. The subject stepped onto the middle of the scale and remained still with arms at his or her sides. This process was repeated, and the average of two weights was recorded. Directly measured BMI was calculated as weight in kg/height in m2. All procedures employed by the PCLS were approved by the Pennington Biomedical Research Center Institutional Review Board.

Statistical analysis

Data management and analysis were performed using SAS 9.1 software (SAS Institute, Cary, NC). Adults were categorized by gender (women, men), ethnicity (EA = European Americans, AA = African Americans), age in years (18–29, 30–39, 40–49, 50–65), and BMI in kg/m2 (<25 = not overweight or obese, 25–29 = overweight, ≥30 = obese).

Individual differences between self-reported and measured height, weight, and calculated BMI were plotted against their respective measured values. Positive differences are indicative of overestimation in the self-reported values, whereas negative differences are indicative of underestimation. Superimposed on each of these scatter plots is a regression line of the form y = bx + a, where y = self-reported value – measured value, and x = measured value. Ideally, the points in these plots should be randomly scattered rather closely about a horizontal line that passes through zero on the vertical axis, and the regression line should coincide with that horizontal line if differences do not tend to increase (or decrease) as the measured values increase.

Regression models were used to calculate age-ethnicity-gender–adjusted predicted values for measured height, weight, and BMI, respectively, from their observed counterparts. Means and 95% confidence intervals were reported for age, height, weight, and BMI. The paired t test was employed to assess the statistical significance of gender–ethnic-specific differences in the respective means for self-reported vs. measured height, weight, and BMI. Age-, gender- and ethnicity-specific prevalences (%) of obesity were calculated for both self-reported and measured BMI. McNemar's test was employed at the 0.05 level to assess the statistical significance of discrepancies between self-reported and measured overweight and obesity prevalence. Sensitivity and specificity of measured obesity classification based on self-reported data were calculated. The two-sample χ2-test was employed to assess gender–age-specific discrepancies in sensitivity and specificity between EA and AA.

There were differences among the four ethnicity–gender groups with respect to their BMI distributions and prevalence of obesity. To overcome this potential source of bias, each ethnicity–gender group was stratified by BMI categories and a random sample of 800 participants was selected whereby each BMI category was proportionate to the BMI distribution for the total cohort. Thus, in the overall cohort and in each ethnicity–gender group of 800, the BMI distribution was as follows: <25 (21%), 25–29.9 (26%), 30–34.9 (27%), 35–39.9 (16%), and 40+ (10%).

RESULTS

Assessment of self-reporting errors

Table 1 presents the descriptive characteristics of the sample. EA women were slightly older and AA men slightly younger on average. Based on measured height and weight, 52.9% were obese (BMI 30+) and 26.4% were overweight (BMI 25–29). Obesity prevalence ranged from 47.4 to 64.7% over the four gender–ethnic groups and was lower in men and strikingly higher (64.7%) in AA women.

Table 1
Characteristics of the sample of 15,656 AA and EA adults from the Pennington Center Longitudinal Study

Figure 1 presents the differences between self-reported and measured (i) height, (ii) weight, and (iii) BMI plotted against the measured values. For height, it can be seen that many of the differences are rather large indicating many sizable errors in the self-reported heights. In addition, the regression line indicates the predicted difference increased about 0.03cm for every 1.0cm increase in the measured height, suggesting that there was a general increase in the tendency of people to overestimate their height as their true height increased. Thus, people who were 160cm tall overestimated their height by 0.74cm on average, whereas those 190cm tall overestimated by 1.65cm on average. Sizable errors are also seen in the self-reported weights. Here, the predicted difference decreased by 0.034kg for every 1.0 kilogram increase in measured weight. Hence, people who weighed 50kg actually overestimated their weight by about 0.30kg, whereas those who measured 150kg underestimated their weight by about 3.13kg on average. Sizable errors are also seen in the BMI's calculated from self-reported height and weight. Here, the predicted difference decreased by 0.077kg/m2for every 1.0kg/m2 increase in BMI that was calculated from measured height and weight. Hence, people with measured BMI equal to 20kg/m2 actually overestimated their BMI by about 0.11kg/m2, whereas those with measured BMI equal to 40kg/m2 underestimated their BMI by about 1.43kg/ m2 on average.

Figure 1
Plot of the differences between self-report and measured (a) height, (b) weight, and (c) BMI against the measured value in a sample of 15,656 African-American and European-American adults from the Pennington Center Longitudinal Study.

Results of regression analyses in which the measured values for height, weight, and BMI were taken in turn as the dependent variable in a model that included the corresponding self-reported value together with the independent variables (predictors) age, gender, and ethnicity are displayed in Figure 2. These plots suggest reasonably good prediction on measured values that may be accomplished from the self-reported data. However, it is also clear from Figure 2 that sizable discrepancies were observed on an individual basis.

Figure 2
Plot of predicted against measured (a) height, (b) weight, and (c) BMI in a sample of 15,656 African-American and European-American adults from the Pennington Center Longitudinal Study. Predicted values were predicted from age, ethnicity, and self-reported ...

Gender–ethnicity-specific means for self-reported and measured height, weight, and BMI are shown in Table 2. The mean self-reported height significantly overestimated the mean measured height in all four gender–ethnicity groups, ranging from 0.58 cm higher for EA women to 1.54 cm higher for AA men. The amount of overestimation was significantly greater in men compared to women. It was also significantly greater in AA women relative to EA women (0.96 cm vs. 0.58 cm, P < 0.0001), but not significantly different by ethnic group in men. On the other hand, the mean self-reported weight significantly underestimated the mean measured weight in all subject groups, ranging from −0.24 kg for AA men to −1.55 kg for AA women. The amount of underestimation was significantly greater in women compared to men. In women, the underestimation was also significantly greater in AA (−1.55 kg vs. −1.18 kg, P < 0.0001) and in EA men relative to AA men (−0.56 vs. −0.24, P < 0.02). The mean calculated self-reported BMI significantly underestimated the calculated measured BMI in all four subject groups, ranging from −0.59 kg/m2 for AA men to −0.97 kg/m2 for AA women. The amount of underestimation was significantly greater in AA women than in any of the other three subject groups, but the differences among EA women, EA men, and AA men were not statistically significant.

Table 2
Means for self-reported vs. measured height, weight, and BMI among 15,656 AA and EA adults from the Pennington Center Longitudinal Study

Estimating obesity prevalence

The prevalence of obesity based on self-reported vs. measured BMI is shown in Table 3 by gender–ethnicity–age groups. Overall, the measured prevalence was virtually the same in EA (47.8%) and AA (47.4%) men, but it was much higher in AA (64.7%) compared to EA (49.9%) women. It increased with age in all four gender–ethnicity groups. The discrepancy in prevalence using self-reported vs. measured BMI was statistically significant in all gender–ethnicity–age groups except in 30- to 39-year-old AA men (although the magnitude of the discrepancy was 3.4%—larger than some other groups that achieved significant discrepancy). Overall, the discrepancy was lowest in EA women and highest in EA men. It tended to decrease with age in women and increase with age in men.

Table 3
Prevalence of obesity (BMI ≥30) based on self-reported vs. measured height and weight 15,656 AA and EA adults from the Pennington Center Longitudinal Study

Taking prevalence of obesity as defined by measured BMI (≥30 kg/m2) as the reference, sensitivity, and specificity of prevalence as defined by self-reported BMI is given in Table 4 by gender–ethnicity–age group. Overall, sensitivity ranged from 87.6% for AA men to 91.2% for EA women, whereas specificity ranged from 94.7% for AA women to 97.6% for EA men. The ethnic differences in sensitivity were not significant in any age group for women. Sensitivity was significantly higher for AA men (92.1%) compared to EA men (81.6%) in the age group of 18–29 years but significantly lower for AA men (83.4%) than EA men (91.7%) in the age group of 30–39 years. The ethnic differences were not significant in men over 40 years of age. Specificity was significantly lower in AA women (96.5%) compared to EA women (99.4%) ages 18–29, but the discrepancies were not significant in any other age groups. Specificity was significantly lower in AA men compared to EA men in the three oldest age groups, but the ethnic difference was virtually zero in men 18–29 years of age.

Table 4
Sensitivity and specificity of obesity classification (BMI ≥30) based on self-reported data among 15,656 AA and EA adults from the Pennington Center Longitudinal Study

When the BMI distribution was forced to be equal across the four ethnicity–gender groups with a measured obesity prevalence of 53% in all groups, the self-reported prevalence of obesity was 50.4% in AA women, 49.6% in EA women, 49.8% in AA men, and 47.3% in EA men. The differences between the self-reported and measured prevalence of obesity in the total sample vs. those in the random subsample are shown in Figure 3. In the total sample, the prevalence of self-reported obesity is underestimated more in AA women than in EA women; however, once the sample is adjusted for differences in the prevalence of obesity, EA women have a greater difference in self-reported vs. measured obesity. Among men, EAs tend to have greater differences between self-reported and measured obesity; however, the difference is greater when the random sample is applied.

Figure 3
Plot of the differences between self-reported and measured prevalences of obesity in the total sample of 15,656 and a random sample of 3,200 African-American (AA) and European-American (EA) adults matched on the distribution of BMI from the Pennington ...

DISCUSSION

The results of this study affirm the propensity of adults to overestimate their height and underestimate their weight. The combined effect of an underestimated numerator and an overestimated denominator led to a pattern of underestimation when self-reported height and weight were used to calculate BMI. This in turn led to an underestimation of the true obesity prevalence. Adult AA women on average overestimated their heights and underestimated their weights more than EA women leading to a greater underestimation of BMI and obesity. Adult AA and EA men both tended to overestimate their height but to a similar degree. However, although EA men underestimated their weight more on average, the discrepancy had little impact on BMI, perhaps because they also tended to overestimate their height. As a result, the prevalence of obesity was underestimated for both AA and EA men, but the underestimation was somewhat greater for EA men. The sensitivity and specificity of obesity classification based on self-reported height and weight exceeded 90% overall and was only slightly greater for EA compared to AA women.

Age

Kuczmarski et al. observed that men of all ages on average overestimated their heights increasingly with age, but women did so only after they reached the age of 50 years (11). Men overestimated their weight with no clear age relationship except overestimation was greater in the elderly; women, on the other hand, underestimated their weight prior to age 60 by amounts that declined rapidly at older ages. Despite these discrepancies, BMI remained a good marker for overweight and obesity. Our study found similar success with BMI.

Gender

Connor-Gorber et al. reviewed 64 relevant investigations that were published during the period 1979–2005 (6). The authors identified 12 studies that published (self-report-direct measure) mean differences for height in the general population; 10 reported overestimating heights on average, and most found comparable results in men and women. In contrast, we observed overestimation of mean height was significantly greater in men compared to women. Connor-Gorber et al. further identified 11 of 15 studies that reported underestimating mean weight. Women in the PCLS underestimated their weights more on average compared to men, and this gender discrepancy was also seen for BMI, findings that conflict with the preponderance of evidence in other studies (6).

Ethnicity

Rowland reported nonsignificant EA vs. AA differences in reporting bias for both height and weight (15). Villanueva concluded that the odds for AA men overestimating their weight were 66% (95% confidence interval: 41–95%) greater than the corresponding odds for EA men, but did not find parallel significance for women (20). In contrast, we found that compared to the women in EA group, it is in the AA group the overestimation of height and underestimation of weight and BMI was more (P < 0.0001).

BMI

Two studies observed that reporting errors increase as BMI increases (22,23). To investigate whether differences in BMI levels were a potential source of bias in our estimates of ethnicity–gender-specific obesity prevalence, the overall sample distribution of BMI was replicated in subsamples from each subgroup. The results suggesting that obesity prevalence was underestimated more in AA than EA women was no longer evident in the subsample. These findings have important implications for studies examining differences in self-reported vs. directly measured BMI when differences in the prevalence of obesity exist across the groups (age, ethnicity, gender, etc.). For example, one study of self-reported vs. measured BMI reported that MA under-reported their BMI to a greater extent than EA and AA men and women; however, no attempt was made to control for the absolute difference in obesity prevalence across the three ethnic groups (10). Thus, researchers can accurately assess ethnic, gender, and other misreporting variations only after accounting for among-group differences in BMI distribution.

Strengths and limitations

The large bi-ethnic sample of AA and EA adults that have both self-reported and measured values for height, weight, and BMI is a marked strength of the study. Although the reported ethnic differences are highly statistically significant due to the large sample sizes, they are also large enough to be important in explaining some of the ethnic differences in obesity observed in population surveys. Use of self-reported data acquired on the telephone rather than by personal interview is also a strength because population surveillance systems tend to rely on telephone surveys (3,5). Self-reported prevalence of obesity has been reported to be significantly lower when data were collected on the telephone rather than by personal interview (4). Further, the ethnic distribution of the sample (34% AA) is only slightly different from the distribution in the adult population in Baton Rouge, LA (42% AA). Although there are no representative data for directly measured prevalence of obesity for Louisiana, the self-reported prevalence was recently reported as 27% (5). This is substantially lower than the self-reported 49.4% prevalence in the PCLS. Nevertheless, 21% of the PCLS participants had a measured BMI <25 kg/m2.

Concluding remarks

Our findings suggest that reported ethnic discrepancies in the magnitude of underestimation of obesity prevalence may in fact be an artifact of different levels of obesity in the groups. Thus, our results support the conclusion that researchers should preferentially investigate differences among groups that are homogenous with respect to obesity as measured in terms of their within-group BMI distributions. This finding also brings into focus the possibility that effects of other correlates of reporting errors such as age and gender may modulate when accounting for differential obesity prevalence. In summary, our study points to and emphasizes the need for exercising caution when making inferences based of self-reported height and weight, which cannot be overemphasized.

ACKNOWLEDGMENTS

This research was supported by the Pennington Biomedical Research Center. Special thanks to Emily Mire and Connie Murla for data management, and to the many clinical scientists and staff of the Pennington Biomedical Research Center clinic who have contributed screening data to the development of the Pennington Center Longitudinal Study. P.T.K. is supported, in part, by the Louisiana Public Facilities Authority Endowed Chair in Nutrition, and C.B. is funded, in part, by the George A. Bray Jr Chair in Nutrition.

Footnotes

DISCLOSURE

The authors declared no conflict of interest.

REFERENCES

1. Clinical Guidelines on the Identification, Evaluation, and Treatment of Overweight and Obesity in Adults—The Evidence Report National Institutes of Health. Obes Res. 1998;6(Suppl 2):51S–209S. [PubMed]
2. World Health Organization . Report of a World Health Organization consultation on obesity, 1998. World Health Organization; Geneva: Jun, 1997. Obesity: preventing and managing the global epidemic; pp. 3–5.
3. Design and estimation for the National Health Interview Survey, 1995–2004. Vital Health Stat 2. 2000:1–31. [PubMed]
4. Beland Y. Canadian community health survey—methodological overview. Health Rep. 2002;13:9–14. [PubMed]
5. Hughes E, McCracken M, Roberts H, et al. Surveillance for certain health behaviors among states and selected local areas--behavioral risk factor surveillance system, United States, 2004. MMWR Surveill Summ. 2006;55:1–124. [PubMed]
6. Connor-Gorber S, Tremblay M, Moher D, Gorber B. A comparison of direct vs. self-report measures for assessing height, weight and body mass index: a systematic review. Obes Rev. 2007;8:307–326. [PubMed]
7. Alvarez-Torices JC, Franch-Nadal J, Alvarez-Guisasola F, Hernandez-Mejia R, Cueto-Espinar A. Self-reported height and weight and prevalence of obesity. Study in a Spanish population. Int J Obes Relat Metab Disord. 1993;17:663–667. [PubMed]
8. Avila-Funes JA, Gutierrez-Robledo LM, Ponce De Leon Rosales S. Validity of height and weight self-report in Mexican adults: results from the national health and aging study. J Nutr Health Aging. 2004;8:355–361. [PubMed]
9. Bolton-Smith C, Woodward M, Tunstall-Pedoe H, Morrison C. Accuracy of the estimated prevalence of obesity from self reported height and weight in an adult Scottish population. J Epidemiol Community Health. 2000;54:143–148. [PMC free article] [PubMed]
10. Gillum RF, Sempos CT. Ethnic variation in validity of classification of overweight and obesity using self-reported weight and height in American women and men: the Third National Health and Nutrition Examination Survey. Nutr J. 2005;4:27. [PMC free article] [PubMed]
11. Kuczmarski MF, Kuczmarski RJ, Najjar M. Effects of age on validity of self-reported height, weight, and body mass index: findings from the Third National Health and Nutrition Examination Survey, 1988–1994. J Am Diet Assoc. 2001;101:28–34. quiz 5–6. [PubMed]
12. Lawlor DA, Bedford C, Taylor M, Ebrahim S. Agreement between measured and self-reported weight in older women. Results from the British Women's Heart and Health Study. Age Ageing. 2002;31:169–174. [PubMed]
13. Niedhammer I, Bugel I, Bonenfant S, Goldberg M, Leclerc A. Validity of self-reported weight and height in the French GAZEL cohort. Int J Obes Relat Metab Disord. 2000;24:1111–1118. [PubMed]
14. Rona RJ, Chinn S, Manning R. The validity of reported parental height in inner city areas in England. Ann Hum Biol. 1989;16:41–44. [PubMed]
15. Rowland ML. Self-reported weight and height. Am J Clin Nutr. 1990;52:1125–1133. [PubMed]
16. Santillan AA, Camargo CA. Body mass index and asthma among Mexican adults: the effect of using self-reported vs measured weight and height. Int J Obes Relat Metab Disord. 2003;27:1430–1433. [PubMed]
17. Schmidt MI, Duncan BB, Tavares M, Polanczyk CA, Pellanda L, Zimmer PM. Validity of self-reported weight—a study of urban Brazilian adults. Rev Saude Publica. 1993;27:271–276. [PubMed]
18. Spencer EA, Appleby PN, Davey GK, Key TJ. Validity of self-reported height and weight in 4808 EPIC-Oxford participants. Public Health Nutr. 2002;5:561–565. [PubMed]
19. Stunkard AJ, Albaum JM. The accuracy of self-reported weights. Am J Clin Nutr. 1981;34:1593–1599. [PubMed]
20. Villanueva EV. The validity of self-reported weight in US adults: a population based cross-sectional study. BMC Public Health. 2001;1:11. [PMC free article] [PubMed]
21. Yannakoulia M, Panagiotakos DB, Pitsavos C, Stefanadis C. Correlates of BMI misreporting among apparently healthy individuals: the ATTICA study. Obesity (Silver Spring) 2006;14:894–901. [PubMed]
22. Nawaz H, Chan W, Abdulrahman M, Larson D, Katz DL. Self-reported weight and height: implications for obesity research. Am J Prev Med. 2001;20:294–298. [PubMed]
23. Ziebland S, Thorogood M, Fuller A, Muir J. Desire for the body normal: body image and discrepancies between self reported and measured height and weight in a British population. J Epidemiol Community Health. 1996;50:105–106. [PMC free article] [PubMed]
PubReader format: click here to try

Formats:

Save items

Related citations in PubMed

See reviews...See all...

Cited by other articles in PMC

See all...

Links

  • MedGen
    MedGen
    Related information in MedGen
  • PubMed
    PubMed
    PubMed citations for these articles

Recent Activity

Your browsing activity is empty.

Activity recording is turned off.

Turn recording back on

See more...