This article has Open Peer Review reports available.
Measuring the accuracy of self-reported height and weight in a community-based sample of young people
© Bowring et al.; licensee BioMed Central Ltd. 2012
Received: 8 June 2012
Accepted: 22 October 2012
Published: 21 November 2012
Self-reported anthropometric data are commonly used to estimate prevalence of obesity in population and community-based studies. We aim to: 1) Determine whether survey participants are able and willing to self-report height and weight; 2) Assess the accuracy of self-reported compared to measured anthropometric data in a community-based sample of young people.
Participants (16–29 years) of a behaviour survey, recruited at a Melbourne music festival (January 2011), were asked to self-report height and weight; researchers independently weighed and measured a sub-sample. Body Mass Index was calculated and overweight/obesity classified as ≥25kg/m2. Differences between measured and self-reported values were assessed using paired t-test/Wilcoxon signed ranks test. Accurate report of height and weight were defined as <2cm and <2kg difference between self-report and measured values, respectively. Agreement between classification of overweight/obesity by self-report and measured values was assessed using McNemar’s test.
Of 1405 survey participants, 82% of males and 72% of females self-reported their height and weight. Among 67 participants who were also independently measured, self-reported height and weight were significantly less than measured height (p=0.01) and weight (p<0.01) among females, but no differences were detected among males. Overall, 52% accurately self-reported height, 30% under-reported, and 18% over-reported; 34% accurately self-reported weight, 52% under-reported and 13% over-reported. More females (70%) than males (35%) under-reported weight (p=0.01). Prevalence of overweight/obesity was 33% based on self-report data and 39% based on measured data (p=0.16).
Self-reported measurements may underestimate weight but accurately identified overweight/obesity in the majority of this sample of young people.
In Australia, obesity is a national health priority, and approximately one quarter of young adult Australians are overweight or obese . This is a concern for immediate social and health problems, as well as risk of future obesity in adulthood and associated chronic health problems, such as type II diabetes, cardiovascular diseases, musculoskeletal disease and some cancers . Data on prevalence and trends in obesity in young people are needed to inform, monitor, and evaluate appropriate policy and interventions.
Body mass index (BMI) is widely used as a measure of obesity due to its simple derivation from height and weight. Height and weight are commonly self-reported in population health surveys for ease of collection . In contrast to direct measurement, self-report enables a large number of individuals to be sampled at relatively minimal cost, time, and resources and the survey tool can be administered face to face, by telephone or online.
Self-reported height and weight may be affected by response or recall bias. It is commonly found that weight is underestimated while height is often overestimated [3–8], leading to underestimation of BMI [5, 7, 8]. Some studies have identified a greater discrepancy in self-reported and measured height and weight among certain groups, including overweight or obese individuals compared to healthy or underweight individuals [6, 7, 9–12], females [7–9, 12, 13], dieters compared to non-dieters, non-smokers compared to smokers [11, 14], and older individuals [7, 9, 11]. Such systematic error may lead to lower prevalence estimates of overweight and obesity and systematic bias in studies examining the relationship between obesity and health outcomes dependent on self-reported measurements. In the Australian context, from a telephone-recruited population survey in Adelaide, classification of overweight/obesity among 18–24 year olds was 6.3% through self-report data and 10.1% through measured data,; while in the 1995 Australian National Health Survey, classification of overweight/obesity among 15–19 year olds was 18% and 12% of males and females, respectively, through self-report data and 25% and 19% of males and females, respectively, through measured data .
In spite of these findings, there remains a lack of data reporting the willingness and accuracy of young people to self-report height, weight, and derived BMI, particularly in community-based settings. Traditional methods of recruiting young people into population surveys, such as random digit dialling and school-based recruitment, are subject to increasing barriers affecting participation, tracking, and retention, such as high mobility, changes in telephone use and technology, and non-school attendance [15, 16]. Community-based recruitment provides an alternative means to recruit large numbers of young people, and we have previously shown how periodic recruitment from a music festival can be used for surveillance in young people [17, 18]. In this setting, use of self-reported measurements would be preferable, given the substantial added time and resources required to independently measure participants.
To determine whether young people participating in a risk behaviour survey are able and willing to provide self-reported height and weight data; and
To assess the accuracy of self-reported height, weight and BMI compared to measured values in a community-based sample of young people.
Participants and procedure
The structure of the market stall set up meant there was one measuring station for height and weight. After completing the questionnaire, participants were invited to have their height and weight measured if approximately three or less participants were waiting to use the measuring station.
A core set of questions about drug and alcohol use and sexual risk behaviours is asked annually in the BDO survey [17, 18]. In 2011, participants were additionally asked to self-report their height in centimetres (without shoes) and weight in kilograms (without clothes and shoes) with an option for “don’t know” in both questions.
When submitting their completed survey, a subset of individuals was invited to have their height and weight measured by a trained research assistant (Figure 1). As outlined above, invitation was based upon the availability of the measurement station. Participants were measured without shoes and wearing light clothes using digital scales and a portable stadiometer. Height and weight were measured to the nearest 1cm and 0.1kg, respectively.
To link measurements to the participants’ surveys, measurements were labelled using either the survey identification number or a “porn star name”, a novel anonymous identifier derived from name of first pet and name of first street , which was also reported in the questionnaire.
BMI for both self-reported and measured values was calculated as weight (kg) divided by height squared (m2). For participants aged 18 years and older, overweight and obesity were classified using the standard international adult BMI ranges: underweight (BMI <18.5), healthy weight (BMI = 18.5-24.9), overweight (BMI = 25.0-29.9), or obese (BMI >30). [21, 22] For participants aged under 18 years, age and sex specific BMI cut-offs for overweight and obesity were used, as defined by the International Obesity Task Force. For the purpose of this research, obesity classification was dichotomised as non-overweight (BMI<25.0) and overweight/obese (BMI≥25.0).
A range of self-reported demographic indicators were assessed: sex (male, female); age (16–19, 20–29 years); born in Australia (yes, no); area of residence (major, non-major city); living arrangements (if live with parents/partner; yes, no); post-high school education (if completed or in the process of completing; yes, no); recreational income (<$120, ≥$120 per week). Area of residence was classified from Australian postcodes of residence using the Australian Standard Geographical Classification (ASGC) Remoteness Areas system . Recreational income was defined as money available for spending or saving, not including essential living costs or tax.
Data were entered into a Microsoft Access database and statistical analysis was conducted in Stata version 11 .
To assess factors associated with survey participants’ (n=1405) ability and willingness to provide self-report height and weight measurements, participants who did and did not self-report height and weight were compared by demographic indicators using chi-square tests of proportion (Figure 1).
Among those with self-reported height and weight (n=1059), participants with and without measurements taken were compared using chi-square tests of proportion and Wilcoxon rank-sum test.
Test of equality between measured and self-reported mean or median was assessed using a paired t-test or Wilcoxon signed ranks test, respectively, both overall and stratified by sex, age group, and obesity classification categories.
Correlation of self-report to measured values was calculated using Pearson correlation coefficient or Spearman rank correlation coefficient according to distribution of both self-reported and measured height, weight and BMI.
The proportion of individuals who accurately, under- and over- reported their weight/height was compared using a Fisher’s exact test. Accurate report of weight/height was defined as less than 2kg/2cm absolute difference between self-report and measured weight/height given an acceptable margin of error, based on previous studies . Under-report of height/weight was defined as self-report ≥2cm/2kg less than measured values; over-report of weight/height was defined as self-report being ≥2cm/2kg more than measured values. The proportion of participants misreporting their height or weight by five or more kilograms was also calculated.
Prevalence of overweight/obesity based on self-report and measured data was compared using two-sided exact McNemar’s test. Misclassification of overweight/obesity was measured in terms of sensitivity and specificity with exact binomial 95% confidence intervals (CI), whereby 100% indicates maximum agreement between self-reported and measured classification.
Significance level was 0.05 in all analyses.
Ethical approval was granted by the Alfred Hospital Human Research Ethics Committee (number 326/08).
Comparison of participant characteristics and risk behaviours by self-report height and weight status
Self-reported height and weight
Self-reported height and/or weight missing/unknown
Median age (IQR)
Post high-school education 2
Lives with parent/s
Lives with partner
Recreational income 3
<$120 per week
≥$120 per week
Area of residence 4
Eighty survey participants had height and weight independently measured. Amongst survey participants providing self-reported height and weight, participants with (6%) and without (94%) independent anthropometric measurement did not differ by demographic characteristics (all p≥0.08).
The final sample is based on 67 participants with complete self-reported and measured height and weight (Figure 1); 51% were male, the median age was 20.1 years, 37% had post-high school education, 67% lived in a major-city, 57% lived with their parent(s), and 58% had $120 or less recreational income per week.
Accuracy of self-reported height, weight, and BMI compared to measured values
Comparison of measured and self-reported values for height, weight, and BMI overall and by sex, age group, and obesity classification
Pearson correlation between measured and self-reported
Test of equality between measured and self-reported, p-value
Age group (years)
16-19 (n=33 )
20-29 (n=34 )
Correlations between self-reported and measured values for height and weight were high (Table 2); correlation was equal to or greater than 0.92 for all values and sub-categories with the exception of height among females, with correlation of 0.84. Correlation of self-reported and measured BMI was moderately high for all sub-groups with the exception among overweight or obese individuals, in which correlation was only 0.40.
Proportion of participants who accurately, under- and over-reported their height and weight stratified by sex, age group, and obesity classification
Age group (years)
Age group (years)
Overall, 34% of participants accurately self-reported their weight (within 2 kg), 52% under-reported and 13% over-reported their weight (Table 3). Significantly more females (70%) than males (35%) under-reported their weight by at least two kilograms (p=0.01). Compared to non-overweight participants, overweight/obese participants were less likely to accurately report weight and more likely to under-report weight, but this was non-significant (p=0.24).
In total, 3 (9%) males and 9 (27%) females inaccurately self-reported their height by five centimetres or more (p=0.05), and 12 (35%) males and 5 (15%) females inaccurately self-reported their weight by five kilograms or more (p=0.06). Overweight/obese participants were more likely to misreport their weight by at least 5kg than non-overweight participants (42% vs 15%, p=0.01).
Classification of overweight/obesity
Based on self-report data, 22 (33%) participants were classified as overweight or obese, compared to 26 (39%) participants based on measured data (p=0.29). In total, 59 (88%) were correctly classified. The sensitivity of self-reported data was 77% (95%CI 56%-91%) and specificity was 95% (95%CI 83%-99%).
In this study of young people attending a music festival, we determined the feasibility and accuracy of collecting self-reported height and weight. Our results confirm that at a group level, self-report measures in a community-based setting is a useful tool for estimating the prevalence of overweight and obesity, particularly when impractical to take independent measurements.
Three-quarters of survey participants provided both self-report height and weight. In the remainder, the majority reported not knowing their height and/or weight, with only a few missing values, although the “don’t know” category might have also included some refusals. Of concern, participants who did not self-report their height or weight were systematically different from those who did; they were more likely to be female, younger, and less educated. Because height and weight measurements were not taken for all participants, we could not determine whether self-reporting height and weight was influenced by bodyweight status. Further research is needed to explore if there are biases in bodyweight status influencing willingness and ability to self-report height and weight in a community-based setting.
Self-reported and measured height did not significantly differ, and approximately half of males and females reported their height within two centimetres of measured height. However, females were more likely than males to underreport their height, and approximately one quarter of females misreported their height by more than five centimetres, compared to only nine percent of males. Previous studies have observed over-report of height [6, 7] or decreased accuracy of height with increasing age, perhaps due to decreasing opportunities to regularly measure height or changes in height over time [4, 27]. Of note, we did not detect a difference in accurate report by age group in our study.
The difference between self-reported and measured weight was more pronounced than for height, particularly among females and overweight or obese individuals. Although one third of all individuals reported their weight within two kilograms of measured weight, a notable 35% of males and 15% of females misreported their weight by five or more kilograms. Females were more likely than males to under-report their weight. Our results are consistent with previous studies reporting systematic under-reporting of weight by females and overweight/obese individuals [4, 12]. It has been postulated that social desirability bias may explain the underreporting of weight, particularly among females and overweight/obese individuals [8, 28]. However, other research that has included a measure of social desirability has challenged this notion [26, 28].
In this study we found that self-report height and weight is a reasonable predictor of overweight and obesity among young people, particularly when pertaining to population-level applications such as monitoring trends in overweight, program evaluation, and advocacy for funding . Notwithstanding inaccuracies in the self-reporting of weight, the effect on BMI was small, with the median difference less than one unit. Sensitivity of classification of overweight/obesity was around 77% − similar to the 70% found in 15–19 year olds participants of the Australian National Health Survey . Nonetheless, in this sample, approximately two-fifths of overweight/obese individuals would have been incorrectly classified as non-overweight based on self-report (false negatives), which is consistent with previous findings . Methods to limit this inaccuracy and bias might include a correction algorithm to account for generalised misreporting based on certain characteristics [2, 11, 29], periodically measuring a sub-sample, or where feasible, advising participants ahead of time to weigh and measure themselves before participating .
Obesity prevention and control is a national priority in Australia, with increasing millions of dollars being invested to its cause . Community-based settings, including web-based studies, are ideal alternatives to traditional means of population-based recruitment in order to both inform and evaluate obesity interventions – particularly among young adults who are difficult to access through household telephone surveys and school settings. They provide novel means to reach a large number of people, including hard-to-reach populations . In these settings self-reported anthropometric measures are the most practical means to define and estimate the prevalence of overweight and obesity. To our knowledge, this is the first study to investigate accuracy of self-report in a community-based sample; our findings confirm that it is possible to use self-report data to estimate and monitor trends in prevalence of overweight and obesity, particularly when bias is expected to remain constant.
This study has a number of limitations. First, the results are based on a convenience-sample and may not be representative of young people in Australia. Second, the sample size for the independent measurement of height and weight was relatively small and may have limited our ability to detect differences and associations with self-reported height and weight. This sample size was limited by the use of only one measuring station, and in the future we recommend more stations are utilised to obtain a larger sample. Third, small differences in self-reported and measured values may be attributable to non-differential measurement bias; multiple researchers were responsible for taking measurements throughout the day and the recruitment day was characterised by high temperatures, reaching up to 40 degrees Celsius. Some participants may have been dehydrated while others were drinking large quantities of water (the study team was distributing water); both factors may have impacted on the accurate measurement of weight. Natural daily weight fluctuation may also partly explain some weight discrepancies; at the festival measurements were taken between 10am-3pm, which may differ to participants’ customary time for weighing themselves. Fourth, selection bias may have been introduced because individuals were not systematically randomised to have their height and weight measured by a researcher; there was potential unrecognised selection bias by the investigators inviting participants to be measured, as well in participants who declined to be measured. However, no demographic differences were identified between those with and without measurements taken. Further research is needed to confirm findings in a larger sample and to see whether a convenience music festival audience differs from the general young adult population in terms of self-reported weight and height accuracy.
In conclusion, our results suggest that the majority of young people participating in a risk behavioural survey are able and willing to self-report their height and weight. The classification of overweight/obesity based on derived self-report BMI has high sensitivity against independent measurement in this setting. Given the wide reach and efficiency of surveys including self-reported measures, these results suggest that that self-reported height and weight are a suitable proxy to estimate the prevalence of overweight and obesity by BMI in a community-based sample of young people, but some underestimation of overweight/obesity is likely and should be taken into consideration.
MH receives funding from the NH&MRC; AP is supported by a VicHealth fellowship; RFP is supported by an Australian Postgraduate Award and a Monash/BakerIDI Departmental Scholarship; MSCL receives funding from an NH&MRC early career research fellowship. The 2011 survey was funded by the Victorian Department of Health. The authors gratefully acknowledge the contribution to this work of the Victorian Operational Infrastructure Support Program. The authors thank Bianca Fiebeger for allowing us to recruit at the Big Day Out music festival, the trained study recruiters, and the study participants. Showbag contents were donated by the Victorian Department of Justice, Marie Stopes International, Youth Projects, Hepatitis Victoria, Department of Health and Ageing, the Free Condom Project and Cadbury Schweppes.
- Australian Insitute of Health and Welfare: Risk Factor Monitoring. A rising epidemic: obesity in Australian children and adolescents. 2004, Department of Health and Ageing, CanberraGoogle Scholar
- Hayes AJ, Kortt MA, Clarke PM, Brandrup JD: Estimating equations to correct self-reported height and weight: implications for prevalence of overweight and obesity in Australia. Aust N Z J Public Health. 2008, 32 (6): 542-545. 10.1111/j.1753-6405.2008.00306.x.View ArticlePubMedGoogle Scholar
- Gorber SC, 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 (4): 307-326. 10.1111/j.1467-789X.2007.00347.x.View ArticlePubMedGoogle Scholar
- Engstrom JL, Paterson SA, Doherty A, Trabulsi M, Speer KL: Accuracy of self-reported height and weight in women: an integrative review of the literature. J Midwifery Womens Health. 2003, 48 (5): 338-345. 10.1016/S1526-9523(03)00281-2.View ArticlePubMedGoogle Scholar
- Himes JH, Hannan P, Wall M, Neumark-Sztainer D: Factors associated with errors in self-reports of stature, weight, and body mass index in Minnesota adolescents. Ann Epidemiol. 2005, 15 (4): 272-278. 10.1016/j.annepidem.2004.08.010.View ArticlePubMedGoogle Scholar
- Wang Z, Patterson CM, Hills AP: A comparison of self-reported and measured height, weight and BMI in Australian adolescents. Aust N Z J Public Health. 2002, 26 (5): 473-478. 10.1111/j.1467-842X.2002.tb00350.x.View ArticlePubMedGoogle Scholar
- Elgar FJ, Stewart JM: Validity of self-report screening for overweight and obesity: Evidence from the Canadian Community Health Survey. Can J Public Health. 2009, 99 (5): 423-427.Google Scholar
- Brener ND, McManus T, Galuska DA, Lowry R, Wechsler H: Reliability and validity of self-reported height and weight among high school students. J Adolesc Health. 2003, 32 (4): 281-287. 10.1016/S1054-139X(02)00708-5.View ArticlePubMedGoogle Scholar
- Roberts RJ: Can self-reported data accurately describe the prevalence of overweight?. Public Health. 1995, 109 (4): 275-284. 10.1016/S0033-3506(95)80205-3.View ArticlePubMedGoogle Scholar
- 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 (04): 561-565. 10.1079/PHN2001322.View ArticlePubMedGoogle Scholar
- Nyholm M, Gullberg B, Merlo J, Lundqvist-Persson C, Rastam L, Lindblad U: The validity of obesity based on self-reported weight and height: Implications for population studies. Obesity. 2007, 15 (1): 197-208. 10.1038/oby.2007.536.View ArticlePubMedGoogle Scholar
- Sherry B, Jefferds ME, Grummer-Strawn LM: Accuracy of Adolescent Self-report of Height and Weight in Assessing Overweight Status: A Literature Review. Arch Pediatr Adolesc Med. 2007, 161 (12): 1154-1161. 10.1001/archpedi.161.12.1154.View ArticlePubMedGoogle Scholar
- Taylor AW, Grande ED, Gill TK, Chittleborough CR, Wilson DH, Adams RJ, Grant JF, Phillips P, Appleton S, Ruffin RE: How valid are self-reported height and weight? A comparison between CATI self-report and clinic measurements using a large cohort study. Aust N Z J Public Health. 2006, 30 (3): 238-246. 10.1111/j.1467-842X.2006.tb00864.x.View ArticlePubMedGoogle Scholar
- 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 (2): 143-148. 10.1136/jech.54.2.143.View ArticlePubMedPubMed CentralGoogle Scholar
- Faden VB, Day NL, Windle M, Windle R, Grube JW, Molina BSG, Pelham WE, Gnagy EM, Wilson TK, Jackson KM, et al: Collecting Longitudinal Data Through Childhood, Adolescence, and Young Adulthood: Methodological Challenges. Alcohol Clin Exp Res. 2004, 28 (2): 330-340. 10.1097/01.ALC.0000113411.33088.FE.View ArticlePubMedPubMed CentralGoogle Scholar
- Grande E, Taylor A: Sampling and coverage issues of telephone surveys used for collecting health information in Australia: results from a face-to-face survey from 1999 to 2008. BMC Med Res Methodol. 2010, 10 (1): 77-10.1186/1471-2288-10-77.View ArticlePubMed CentralGoogle Scholar
- Lim MSC, Hellard ME, Aitken CK, Hocking JS: Surveillance of STI risk behaviour among young people attending a music festival in Australia, 2005–08. Aust N Z J Public Health. 2009, 33: 482-484. 10.1111/j.1753-6405.2009.00434.x.View ArticlePubMedGoogle Scholar
- Lim MSC, Hellard ME, Hocking JS, Spelman TD, Aitken CK: Surveillance of drug use among young people attending a music festival in Australia, 2005–2008. Drug Alcohol Rev. 2010, 29 (2): 150-156.View ArticlePubMedGoogle Scholar
- Lim MSC, Hellard ME, Aitken CK, Hocking JS, Lim MSC, Hellard ME, Aitken CK, Hocking JS: Sexual-risk behaviour, self-perceived risk and knowledge of sexually transmissible infections among young Australians attending a music festival. Sex Health. 2007, 4 (1): 51-56. 10.1071/SH06031.View ArticlePubMedGoogle Scholar
- Lim MSC, Bowring A, Gold J, Hellard ME: What's your 'porn star' name? A novel method of identifying research participants (letter). Sex Transm Dis. 2011, 38 (2): 150-151.View ArticlePubMedGoogle Scholar
- National Health and Medical Research Council: Clinical Practice Guidelines for the Management of Overweight and Obesity in Adults. 2003, NHMRC, CanberraGoogle Scholar
- World Health Organization: Obesity: preventing and managing the global epidemic. Report of a WHO consultation. World Health Organ Tech Rep Ser. 2000, 894 (i-xii): 1-253.Google Scholar
- Cole TJ, Bellizzi MC, Flegal KM, Dietz WH: Establishing a standard definition for child overweight and obesity worldwide: international survey. BMJ. 2000, 320 (7244): 1240-10.1136/bmj.320.7244.1240.View ArticlePubMedPubMed CentralGoogle Scholar
- Australian Bureau of Statistics: Australian Standard Geographical Classification (ASGC) Remoteness Areas (cat. no. 1216.0). 2006, Australian Bureau of Statistics, CanberraGoogle Scholar
- StataCorp: Stata Statistical Software: Release 11. A College Station. 2009, StataCorp LP, TXGoogle Scholar
- Brestoff J, Perry I, Van den Broeck J: Challenging the role of social norms regarding body weight as an explanation for weight, height, and BMI misreporting biases: Development and application of a new approach to examining misreporting and misclassification bias in surveys. BMC Publ Health. 2011, 11 (1): 331-10.1186/1471-2458-11-331.View ArticleGoogle Scholar
- Australian Bureau of Statistics: How Australians Measure Up (cat. no. 4359.0). 1999, ABS, CanberraGoogle Scholar
- Larson M: Social desirability and self-reported weight and height. Int J Obes. 2000, 24 (5): 663-665. 10.1038/sj.ijo.0801233.View ArticleGoogle Scholar
- Ezzati M, Martin H, Skjold S, Hoorn SV, Murray CJL: Trends in national and state-level obesity in the USA after correction for self-report bias: analysis of health surveys. JRSM. 2006, 99 (5): 250-257. 10.1258/jrsm.99.5.250.View ArticleGoogle Scholar
- De Vriendt T, Huybrechts I, Ottevaere C, Van Trimpont I, De Henauw S: Validity of Self-Reported Weight and Height of Adolescents, Its Impact on Classification into BMI-Categories and the Association with Weighing Behaviour. Int J Environ Res Public Health. 2009, 6 (10): 2696-2711. 10.3390/ijerph6102696.View ArticlePubMedPubMed CentralGoogle Scholar
- National Health and Medical Research Council: Overweight and obesity. http://www.nhmrc.gov.au/grants/research-funding-statistics-and-data/overweight-and-obesity,
- Rhodes SD, Bowie DA, Hergenrather KC: Collecting behavioural data using the world wide web: considerations for researchers. J Epidemiol Community Health. 2003, 57 (1): 68-73. 10.1136/jech.57.1.68.View ArticlePubMedPubMed CentralGoogle Scholar
- The pre-publication history for this paper can be accessed here:http://www.biomedcentral.com/1471-2288/12/175/prepub
This article is published under license to BioMed Central Ltd. This is an Open Access article distributed under the terms of the Creative Commons Attribution License (http://creativecommons.org/licenses/by/2.0), which permits unrestricted use, distribution, and reproduction in any medium, provided the original work is properly cited.