- Research article
- Open Access
- Open Peer Review
Sexual behaviour, sexually transmitted infections and attitudes to chlamydia testing among a unique national sample of young Australians: baseline data from a randomised controlled trial
© Kang et al.; licensee BioMed Central Ltd. 2014
- Received: 20 September 2013
- Accepted: 4 January 2014
- Published: 8 January 2014
Chlamydia infection is the most common notifiable sexually transmitted infection (STI) in Australia and mostly affects young people (15 – 25 years). This paper presents baseline data from a randomised controlled trial that aimed to increase chlamydia testing among sexually active young people. The objectives were to identify associations between sexual behaviour, substance use and STI history and explore attitudes to chlamydia testing.
This study was conducted in cyberspace. Study recruitment, allocation, delivery of interventions and baseline and follow up data collection all took place online. Participants were 16 – 25 years old and resided in Australia. Substance use correlates of sexual activity; predictors of history of STIs; barriers to and facilitators of chlamydia testing were analysed.
Of 856 participants (79.1% female), 704 had experienced penetrative intercourse. Sexually active participants were more likely to smoke regularly or daily, to drink alcohol, or to have binge drunk or used marijuana or other illicit substances recently. Risk factors for having a history of any STI were 3 or more sexual partners ever, 6 or more partners in the past 12 months, condom non-use and being 20 years or older. Almost all sexually active participants said that they would have a chlamydia test if their doctor recommended it.
Sexually active young people are at risk of STIs and may engage in substance use risk behaviours. Where one health risk behaviour is identified, it is important to seek information about others. Chlamydia testing can be facilitated by doctors and nurses recommending it. Primary care providers have a useful role in chlamydia control.
Australian and New Zealand Trials Registry ACTRN12607000582459
- Young people
- Sexually transmitted infections
- Health risk behaviour
Most people experience their first partnered sexual interactions in the second decade of life . The average age of first penetrative vaginal intercourse in Australia has been 16 years for females and males for over three decades [2, 3]. Despite the initiation of sexual intercourse in the teenage years being normative, the 1980s’ HIV pandemic and changing patterns of sexual relationships among young people have contributed to worldwide concern about adolescent sexual behaviour as a public health issue. Genital chlamydia infection is more prevalent than HIV globally and affects people under 25 years more than any other age group . To address the rapid rise of chlamydia notifications in Australia since 1999, the first (2005 – 2008) and second (2010 – 2013) national sexually transmissible infections (STI) strategies included chlamydia control and young people as specific priorities [5, 6]. One of the key components of both strategies was to increase chlamydia testing among target groups.
Research into young people’s sex lives in the public health context has focused on sexual risk behaviour. This has been variously defined as penetrative intercourse without consistent condom use, intercourse with casual partners, having three or more sexual partners in twelve months and/or initiation of sexual intercourse before 16 years of age [7–9]. Many studies have found that adolescent ‘risk behaviours’ tend to cluster, for example smoking, binge drinking and sexual risk behaviours [9, 10]. The technical ease with which chlamydia infection can be tested in urine samples has added a new dimension to recent studies, with attention to young people’s interaction with the health system and the barriers and facilitators to testing of prime importance.
Provide demographic, substance use and sexual history characteristics of a unique sample of young Australians (16 – 25 years).
Identify associations between sexual behavior, substance use and STI history.
Report on chlamydia knowledge and attitudes to chlamydia testing.
This paper reports on baseline data from a randomised controlled trial (RCT)  of 16 – 25 year-old Australians. The study was conducted in cyberspace. A website about chlamydia was launched in March 2007 and invited eligible visitors to participate in a study to promote chlamydia testing. This website was developed with input from 20 youth consultants (16–25 years) who were recruited through professional and collegiate networks. The website was promoted through paid advertising, existing youth-related websites, social networking sites and opportunistic radio interviews. Website traffic was monitored using Google Analytics.
The website invited eligible visitors to participate in the study via clickable links on the homepage and all other webpages. To be eligible, visitors had to be aged 16 – 25 years, reside in Australia, answer ‘yes’ to whether they had ever had penetrative (vaginal or anal) intercourse and provide an email address. Potential participants were taken to a Participant Information Statement, entered a current email address and ticked a consent box. This step took them to the baseline questionnaire housed within the website.
The sampling frame was all eligible visitors to the website, but participation rates were not measurable since the denominator was unknown. The target sample size for the RCT was 1000. A detailed description of the RCT methods has been published elsewhere .
One hundred and fifty-two young people who completed baseline data had not had penetrative sex and were not enrolled in the RCT. Their substantial number provided a comparison group when analysing sociodemographic correlates of health risk behaviours, as many differences achieved statistical significance.
The baseline questionnaire collected sociodemographic data, sexual and substance use history, knowledge about chlamydia, attitudes towards chlamydia testing and testing preferences. Some attitude questions explored known barriers to accessing health care for Australian young people , such as concerns about confidentiality and owning one’s own ‘Medicare card’. Australia has a universal health insurance scheme (‘Medicare’) and doctors and pathology providers can choose to bill a service directly to Medicare, meaning the patient has no upfront fee. Possessing one’s own card (separate from a family card) is not essential, but can facilitate this direct billing process.
Data were analysed using SPSS v19.0 (IBM, USA). Categorical variables were compared between groups using chi-square tests with Yate’s correction for 2 × 2 tables. Continuous variables were compared using independent samples t-tests. Logistic regression analyses were conducted on the whole sample to assess substance use behaviours that were associated with being sexually active while controlling for age and sex, and on the sexually active group to predict risk factors for STIs.
Ethical approval was obtained from The University of Sydney Human Research Ethics Committee and the trial was registered with the Australian and New Zealand Trials Registry (ACTRN12607000582459).
Recruitment took place between March 2007 and January 2008 and 856 people aged 16 – 25 years provided complete datasets. Seven hundred and four young people reported they had ever had sexual intercourse and 152 reported they had never had sexual intercourse.
Sociodemographic characteristics of the sample (n = 856) compared to the population aged 15–24 years from the 2006 Census 
Sample at baseline 2007
16 – 25 years
15 – 24 years in 2006
Sex = female
Born in Australia
Born overseas in English speaking country
Born overseas in non-English speaking country
Aboriginal and/or Torres Strait Islander
Speak English at home
State of residence
New South Wales
Fully engaged in education and/or employment
Education Level (highest attained)
Completed Year 12 or above
At home with parents/guardian
Away from home (private rental, campus based, with friends)
Refuge or supported accommodation
Sexually active and non-sexually active respondents compared
Substance use factors associated with being sexually active (n = 856)
Sexually active (n = 704)
Not sexually active (n = 152)
Odds ratio (95% CI)*
15.1 (5.5, 41.6)
Drinking alcohol ever
6.6 (4.1, 10.7)
Drinking > once/week
4.9 (2.9, 8.3)
6.2 (3.9, 9.9)
Marijuana in last 30 days
5.0 (2.3, 10.6)
Illicit drugs last 30 days
11.1 (2.6, 46.0)
Differences in chlamydia knowledge between sexually active and non-sexually active young people
Sexually active% who answered correctly (n = 704)
Not sexually active% who answered correctly (n = 152)
Chlamydia is a sexually transmissible infection that affects only women
Chlamydia can lead to sterility among women
A woman can have chlamydia without any obvious symptoms
A man can have chlamydia without any obvious symptoms
Chlamydia is curable
Chlamydia can be prevented by using condoms when you have sex
Chlamydia can be tested for with a urine sample
Mean composite score/7
Characteristics of the sexually active group
The SA group (n = 704) was analysed to explore sexual history (number of sexual partners, condom use, history of chlamydia or other STIs) as well as attitudes to and preferences for chlamydia testing.
Differences in sexual history between females and males
N = 547
N = 154
Mean age (SD)
20.0 years (3.0)
21.4 years (2.6)
Mean age of first intercourse (SD)
16.1 years (2.3)
16.7 years (2.3)
Mean no. of years since first intercourse (SD)
3.9 years (3.0)
4.7 years (3.0)
Number of sexual partners ever
Three to five
Six to ten
Eleven or more
Number of sexual partners in last 12 months
Three to five
Six to ten
Eleven or more
Use condoms always or mostly
History of chlamydia
History of HPV
History of gonorrhoea
History of genital herpes
History of HIV
One hundred and eighteen young people (16.9%) reported a history of ever being diagnosed with chlamydia at baseline. The next most frequently reported STI was human papillomavirus (defined as ‘human papillomavirus also called HPV or the wart virus’) (8.3%) followed by genital herpes (3.9%), gonorrhoea (2.1%) and HIV (1.0%).
One hundred and eighty-eight young people (26.7%) reported having had a chlamydia test in the past six months, of these 70 (37.2%) reported that the test was positive. All but one who reported a positive chlamydia test stated that they had received antibiotic treatment and 65/70 had informed their partner of the result. Forty-seven (67.1%) of those who had tested positive reported that they had returned for a second test. Most who had not returned for a second test stated that insufficient time (to test for re-infection) had passed.
Predictors of STI history in participants who had sexual intercourse ever
Percent (%) in exposed group
Percent (%) in unexposed group
Unadjusted odds ratio
Adjusted odds ratio (95% CI)
Aboriginal and/or Torres Strait Islander*
English at home*
Age 20 years and older
1.95 (1.22, 3.12)
Age first intercourse < = 15 years*
Condom use never or sometimes
1.48 (1.02, 2.16)
Three or more sexual partners ever
4.78 (2.67, 8.54)
Six or more sexual partners last 12 months
2.25 (1.34, 3.79)
Early initiation of sexual intercourse (< 16 years) was associated with being Aboriginal and/ or Torres Strait Islander (71.4% cf 31.6%, p <0.0001); no/low condom use (54.7% cf 42.7%; p = 0.004); 6 or more partners in the past 12 months (17.5% cf 7.7%; p < 0.0001); smoking regularly (47.9% cf 27.2%, p < 0.0001) and using cannabis (46.0% cf 29.9%; p = 0.001) or illicit drugs (44.6% cf 31.3%, p = 0.02). There was no association between early sex and gender, alcohol use or history of chlamydia testing.
Attitudinal questions explored barriers and facilitators to testing. The strongest facilitators of testing in the SA group were ‘if my doctor recommended it’ (97.4% agreed), ‘because I don’t want to give it to my partner if I have it’ (95.4% agreed), ‘to prevent long term health problems’ (94.5% agreed) and ‘if my partner wanted me to’ (88.5% agreed). Seventy-six percent agreed that they ‘would feel comfortable visiting a doctor or nurse for a chlamydia test’ , although 43.2% also agreed that they ‘would feel too embarrassed to talk to a doctor or a nurse about chlamydia’. Less than half the sample (46.2%) agreed that they did not want to talk about their sexual history with a doctor or nurse. Barriers to testing included being scared of what the test might show (40.0% agreed), their partner/s finding out (25.1% agreed) concerns about cost (36.5% agreed) confidentiality (19.8% agreed), transport (17.3% agreed) and not having their own Medicare card (11.2% agreed). Thirty-three percent agreed that they did not know how to get a chlamydia test. Seventy-nine percent agreed that they would like to see a doctor or nurse the same sex as themselves to get a chlamydia test. Young women were significantly more likely than young men to prefer seeing a doctor of the same sex (83.2% cf 66.0%, p <0.0001) but there were no gender differences in relation to attitudes to testing.
Two hundred and ninety-six young people in the SA group (42.0%) indicated that they would prefer to have a chlamydia test done by their usual general practitioner (GP), and 144 (20.5%) preferred a different GP. One hundred and fifty-five young people (22.0%) preferred to be tested at a sexual health clinic. The remainder preferred a Youth Health Service (4.5%), Family Planning Clinic (2.8%), or did not state a preference (8.2%).
This unique sample, recruited entirely in cyberspace, adds to knowledge about sexual behaviours of young Australians and their socio-demographic and substance use correlates. For those who had had sexual intercourse, the mean age of sexual debut was 16.2 years. Sexually active young people were more likely than their non-sexually active counterparts to speak English at home, to smoke regularly, drink alcohol, binge drink and use marijuana or other illicit substances. They also had greater knowledge about chlamydia infection and testing. Sexually active young people reported that they would be willing to have a chlamydia test if their doctor recommended it, or to protect their own or their sexual partner/s’ health. Barriers to testing included embarrassment, concerns about confidentiality, cost, transport and not having a Medicare card. Most preferred to see a general practitioner (GP) for chlamydia testing.
Compared to Census data from 2006 (the closest year to our study) , young people living in major cities, born in Australia and speaking English at home were over-represented but differences were minor. Ninety-one per cent of our sample were engaged in education or paid employment, which is higher than comparable national data from 2007 which found 83.6% of 15 – 24 year old Australians fully engaged in education or training and/or work . We found that being Aboriginal and/or Torres Strait Islander was a predictor of sexual debut < 16 years. This is entirely consistent with sexual experience reported by the first systematic survey of child health among Aboriginal children and young people in Western Australia in 2005 . Reasons for earlier sexual activity are thought to reflect earlier social maturity and the cultural acceptability of earlier child bearing . Rates of substance use in this sample are difficult to compare with either the 2007 national drug strategy household survey  or the 2008 national survey of secondary students and sexual health  because of different age groupings, definitions, and wording of questions, however there were no stark differences. In the national drug strategy household survey, 32.4% of 14 – 29 year olds smoked daily or weekly, compared to 24.2% of our sample of 16 – 25 year olds. Patterns of alcohol use were very similar between our sample for those aged 16 and 17 and the 2008 National Secondary Schools and Sexual Health survey . Rates of marijuana and other illicit substance use were reasonably congruent between our sample and the national household survey . The correlation between being sexually active and substance use and the clustering of health risk behaviours are consistent with international studies;  this is one of very few Australian studies  to report explicitly on this.
The high proportion who had had a recent positive chlamydia test is not surprising. We would expect those who had reason to be concerned to be searching for relevant information and come across our website. Therefore it was pleasing that the majority had not had a test and that, although unanticipated, young people who had never had sex also chose to complete the baseline questionnaire. Knowledge about chlamydia was considerably higher for comparable questions than in the national survey of secondary students in 2008 but this may reflect the older age of study participants as well as the bias towards young people interested in chlamydia.
Most young people indicated that they would have a chlamydia test to protect their, or their partner/s’ health. This, combined with the fact that one-third of the sample did not know how to get a chlamydia test, while a smaller proportion expressed concerns about confidentiality suggests that education about access to testing is also warranted.
These findings highlight the important role of GPs in chlamydia control. Over 60% of sexually active young people in this study preferred to see a GP for chlamydia testing and almost all (males as well as females) would be willing to have a test if their doctor recommended it. National general practice surveillance data has shown that the “opportunity to test” for chlamydia is the strongest predictor for being tested . Yet over the same study period as ours (2007 – 2008), Medicare data showed that only 12.5% of sexually active young women and 3.7% of sexually active young men were tested for chlamydia . This lower testing rate among young men is consistent with sex differences found in the United Kingdom, one of few countries with a national screening program  and one of few that recommends, as in Australia, that asymptomatic, sexually active young men as well as young women are routinely screened. As with our study, this research also found that young men would find screening in general practice acceptable.
Australian GPs identify lack of time and knowledge as barriers to testing, but they also report concern about ‘patient embarrassment’ as a factor [24, 25]. Recent evidence suggests that this concern is misplaced and that most patients are willing to discuss STIs with their GPs .
While forty percent of our sample also nominated embarrassment as a barrier, the majority stated that they would be comfortable visiting their doctor or nurse for a chlamydia test. Our finding that less than 50% of young people did not want a sexual history taken contrasts with earlier research among young Australian women which found that having a sexual history taken was a dominant concern . These apparent contradictions might reflect differences in our sample’s experience with chlamydia testing compared with those of the general population. But they highlight nuances associated with, and need for sensitivity when discussing, sexuality. The onus should be on doctors or nurses to raise the issue, explore sexual histories sensitively, explain confidentiality, normalise testing as part of routine health care and recommend testing when appropriate. Promoting messages about taking care of one’s health and one’s partner/s’ health might also be effective. We also acknowledge that our data exploring attitudes to testing are now a few years old. Awareness of chlamydia might have increased significantly due to national media campaigns. Young people’s use of technology and social media might have changed considerably since we conducted this study. Whilst the impact of these changes might inform new health promotion strategies, we believe that our findings clearly support that having health professionals directly offer testing opportunistically is likely to be effective at increasing testing rates.
The main limitation of this study is the unique sampling method which brings into question the study’s external validity. Although we believe this was an innovative way to deliver a behavioural intervention, it is a self-selected sample likely to be more interested in chlamydia. While the sample is not generalisable in the same way as a random sample would be, we can gain a sense of how well basic demographic groups are represented (gender, indigenous status, country of birth, region of residence). Website traffic can only infer possible reach of a website rather than provide a denominator for measuring participation rates, which is another challenge in online research. Another limitation of this study is the reliance on self-report: this may have under-estimated exposure to STIs and/or over-estimated screening. However the anonymity of the internet might have facilitated disclosure of personal information.
In summary, Australian young people begin partnered sexual activity in their mid-teen years, and many are at risk of chlamydia and other STIs. The availability of affordable testing and treatment for chlamydia, as well as young people’s willingness to consult with doctors and nurses in general practice and other primary care settings and take up testing, should make it a routine part of all consultations with young people.
Where one health risk behaviour is identified, it is important to seek information about others. Chlamydia testing can be facilitated by doctors and nurses recommending it. Primary care providers have a useful role in chlamydia control.
We would like to acknowledge the Youth Consultants who helped to guide this project and all the young people who participated in the study. The study was funded by the Commonwealth Department of Health and Ageing, Australia.
- Bearinger LH, Sieving RE, Ferguson J, Sharma V: Global perspectives on the sexual and reproductive health of adolescents: patterns, prevention, and potential. Lancet. 2007, 369: 1220-1231. 10.1016/S0140-6736(07)60367-5.View ArticlePubMedGoogle Scholar
- Agius PA, Pitts MK, Smith AMA, Mitchell A: Sexual behaviour and related knowledge among a representative sample of secondary school students between 1997 and 2008. Aust NZ J Public Health. 2010, 34 (5): 476-481. 10.1111/j.1753-6405.2010.00593.x.View ArticleGoogle Scholar
- Rissel CE, Richters J, Grulich AE: deVisser RO, Smith AMA. First experiences of vaginal intercourse and oral sex among a representative sample of adults. Aust NZ J Public Health. 2003, 27: 131-137. 10.1111/j.1467-842X.2003.tb00800.x.View ArticleGoogle Scholar
- Glasier A, Gülmezoglu AM, Schmid GP, Moreno CG, Van Look PFA: Sexual and reproductive health: a matter of life and death. Lancet. 2006, 368: 1595-1607. 10.1016/S0140-6736(06)69478-6.View ArticlePubMedGoogle Scholar
- Department of Health and Ageing: National sexually transmissible infections strategy 2005 – 2008. 2005, Canberra, Australia: Commonwealth of AustraliaGoogle Scholar
- Department of Health and Ageing: Second national sexually transmissible infections strategy 2010 – 2013. 2010, Canberra, Australia: Commonwealth of AustraliaGoogle Scholar
- de Visser RO, Rissel CE, Smith AMA, Richters J: Sociodemographic correlates of selected health risk behaviours in a representative sample of Australian young people. Int J Behav Med. 2006, 13 (2): 153-162. 10.1207/s15327558ijbm1302_7.View ArticlePubMedGoogle Scholar
- Jackson CA, Henderson M, Frank JW, Haw SJ: An overview of prevention of multiple risk behaviour in adolescence and young adulthood. J Pub Health. 2012, 34 (S1): i31-i40.View ArticleGoogle Scholar
- Madkour AS, Farhart T, Halpern CT, et al: Early adolescent sexual initiation as a problem behavior: a comparative study of five nations. J Adol Health. 2010, 47: 389-398. 10.1016/j.jadohealth.2010.02.008.View ArticleGoogle Scholar
- Wiefferink CH, Peters L, Hoekstra F, et al: Clustering of health-related behaviours and their determinants: possible consequences for school health interventions. Prev Sci. 2006, 7: 127-149. 10.1007/s11121-005-0021-2.View ArticlePubMedGoogle Scholar
- Kang M, Rochford A, Skinner SR, Mindel A, Webb M, Peat J, Usherwood T: Facilitating chlamydia testing among young people: a randomised controlled trial in cyberspace. Sex Transm Infect. 2012, 88: 568-573. 10.1136/sextrans-2011-050124.View ArticlePubMedGoogle Scholar
- Booth ML, Bernard D, Quine S, Kang M, Usherwood T, Alperstein G: Bennett DL Access to health care among Australian adolescents: young people's perspectives and their socio-demographic distribution. J Adolesc Health. 2004, 34 (1): 97-103.View ArticlePubMedGoogle Scholar
- Australian Bureau of Statistics (ABS): Experimental estimates of Aboriginal and Torres Strait Islander Australians, June 2006. ABS cat. No. 3238.0.55.001. 2008, Canberra: ABSGoogle Scholar
- Smith A, Agius P, Mitchell A, Barrett C, Pitts M: Secondary students and sexual health 2008, Monograph Series No. 70. 2009, Melbourne: Australian Research Centre in Sex, Health and Society, La Trobe UniversityGoogle Scholar
- Australian Bureau of Statistics: Population by Age and Sex, Australian States and Territories, ABS Cat. No. 3201.0. 2010, Canberra: ABSGoogle Scholar
- Australian Bureau of Statistics: Persons aged 15 – 24 who are fully engaged in education or training and/ or work, 2007. 6227.0 - Education and Work Australia, May 2009. 2010, Canberra: ABSGoogle Scholar
- Blair EM, Zubrick SR, Cox AH, on behalf of the WAACHS Steering Committee: The Western Australian Aboriginal Child Health Survey: findings to date on adolescents. Med J Aust. 2005, 183 (8): 433-435.PubMedGoogle Scholar
- AIHW: 2007 National Drug Strategy Household Survey: detailed findings. Drug statistics series no. 22. Cat. no. PHE 107. 2008, Canberra: AIHWGoogle Scholar
- Zimmer-Gembeck MJ, Helfand M: Ten years of longitudinal research on U.S. adolescent sexual behavior: Developmental correlates of sexual intercourse, and the importance of age, gender and ethnic background. Dev Rev. 2008, 28 (2): 153-224. 10.1016/j.dr.2007.06.001.View ArticleGoogle Scholar
- Hibbert M, Caust J, Patton G, Rosier M, Bowes G: The health of young people in Victoria. 1996, Centre for Adolescent Health: MelbourneGoogle Scholar
- Sawleshwarkar S, Harrison C, Britt H, Mindel A: Chlamydia testing in general practice in Australia. Sex Health. 2010, 7: 484-490. 10.1071/SH09110.View ArticlePubMedGoogle Scholar
- Kong FYS, Guy RJ, Hocking JS, Merritt T, Pirotta M, Heal C, Bergeri I, Donovan B, Hellard ME: Australian general practitioner chlamydia testing rates among young people. Med J Aust. 2011, 194 (5): 249-252.PubMedGoogle Scholar
- Saunders JM, Mercer CH, Sutcliffe LJ, Hart GJ, Cassell J, Estcourt CS: Where do young men want to access STI screening? A stratified random probability sample survey of young men in Great Britain. Sex Transm Infect. 2012, 88: 427-432. 10.1136/sextrans-2011-050406.View ArticlePubMedPubMed CentralGoogle Scholar
- Hocking JS, Parker RM, Pavlin N, Fairley CK, Gunn JM: What needs to change to increase chlamydia screening in general practice in Australia? The views of general practitioners. BMC Public Health. 2008, 8: 425-10.1186/1471-2458-8-425.View ArticlePubMedPubMed CentralGoogle Scholar
- Temple-Smith MJ, Mak D, Watson J, Bastian L, Smith A, Pitts M: Conversant or clueless? Chlamydia-related knowledge and practice of general practitioners in Western Australia. BMC Fam Pract. 2008, 9: 17-10.1186/1471-2296-9-17.View ArticlePubMedPubMed CentralGoogle Scholar
- Baker JR, Arnold-Reed DE, Brett T, Hince DA, O’Ferrall I, Bulsara MK: Perceptions of barriers to discussing and testing for sexually transmitted infections in a convenience sample of general practice patients. Aust J Prim Health. 2013, 19: 98-101. 10.1071/PY11110.View ArticlePubMedGoogle Scholar
- Pavlin N, Parker R, Gunn J, Fairley C, Hocking J: Take the sex out of STI screening! Views of young women on implementing chlamydia screening in general practice. BMC Infect Dis. 2008, 8: 62-10.1186/1471-2334-8-62.View ArticlePubMedPubMed CentralGoogle Scholar
- The pre-publication history for this paper can be accessed here:http://0-www.biomedcentral.com.brum.beds.ac.uk/1471-2458/14/12/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. The Creative Commons Public Domain Dedication waiver (http://creativecommons.org/publicdomain/zero/1.0/) applies to the data made available in this article, unless otherwise stated.