Circumcision in Australia: further evidence on its effects on sexual health and wellbeing
Abstract
Objective: To report on the prevalence and demographic variation in circumcision in Australia and examine sexual health outcomes in comparison with earlier research.
Methods: A representative household sample of 4,290 Australian men aged 16–64 years completed a computer‐assisted telephone interview including questions on circumcision status, demographic variables, reported lifetime experience of selected sexually transmissible infections (STIs), experience of sexual difficulties in the previous 12 months, masturbation, and sexual practices at last heterosexual encounter.
Results: More than half the men (58%) were circumcised. Circumcision was less common (33%) among men under 30 and more common (66%) among those born in Australia. After adjustment for age and number of partners, circumcision was unrelated to STI history except for non‐specific urethritis (higher among circumcised men, OR=2.11, p<0.001) and penile candidiasis (lower among circumcised men, OR=0.49, p<0.001).
Circumcision was unrelated to any of the sexual difficulties we asked about (after adjusting for age) except that circumcised men were somewhat less likely to have worried during sex about whether their bodies looked unattractive (OR=0.77, p=0.04). No association between lack of circumcision and erection difficulties was detected. After correction for age, circumcised men were somewhat more likely to have masturbated alone in the previous 12 months (OR=1.20, p=0.02).
Conclusions: Circumcision appears to have minimal protective effects on sexual health in Australia.
Controversy about the benefits and ill‐effects of circumcision continues to rage in the scientific and popular literature.1-5 Earlier Australian findings from a large representative‐sample survey, Sex in Australia,6 have been used to support arguments for the desirability of the procedure,7 although no STI‐prevention benefit was found.
Both the Sex in Australia survey and the US National Health and Social Life Survey8 found that uncircumcised men had higher rates of sexual difficulties, specifically erection difficulties among the over‐50s. Both the Australian and US researchers hypothesised that the difference was due to underlying social differences between the circumcised and uncircumcised population. The majority of Australian‐born men aged over 30 in 2001 were born in hospital, where circumcision was more or less routine until the 1970s.
Here we report data from a study that commenced four years after the earlier survey with a different sample recruited in the same way. This analysis aims to confirm or challenge the earlier results. Although this study has a smaller total sample, all respondents were asked all questions, whereas in the Sex in Australia survey certain subsamples were asked an abbreviated set of questions and the results statistically adjusted to compensate for this. As a result, the standard errors from this study are smaller, resulting in tighter confidence intervals.
Methods
The Australian Longitudinal Study of Health and Relationships involved annual computer‐assisted telephone interviews with a large broadly representative national sample of Australians aged 16–64 years. This analysis uses cross‐sectional data from the 2005 intake wave.
Eligible households were identified through random digit dialling. We used separate sampling frames for men and women; where there was more than one eligible person in a household the participant was selected randomly. Interviews lasted about 25 minutes. People with insufficient proficiency in English (about 5% of households) could not participate. Full methodological details are published elsewhere.9
This research was approved by the Human Ethics Committee of La Trobe University.
The intake interview included a wide range of topics concerning sexual behaviour and attitudes and relationship issues. In this paper we analyse demographic data, lifetime experience of specific sexually transmissible infections (STIs), sexual difficulties experienced for at least a month in the previous year, and sexual practice at last heterosexual encounter within the previous year.
Both descriptive and logistic analyses were conducted using Stata (Release 10.1, 2005). The odds ratios for sexual difficulties, STIs and sexual practice were age‐adjusted and, in addition, the STI analysis was adjusted for lifetime numbers of opposite‐sex partners and of same‐sex partners.
Results
In all, 4,290 men were interviewed; the response rate was 54%. All analyses are conducted using weighted data to adjust for the study design. More than half, 58%, were circumcised. No respondents said they did not know whether they were circumcised, and only eight men (0.2%) refused to answer the question.
Demographics
As in the earlier study, circumcision rates differed strongly by age group (Table 1). The majority of men over 30 were circumcised, but less than a third of those under 30. Circumcision was much less common among those speaking a language other than English at home, those born outside Australia and those with Buddhist religion. These differences were reflected in lower circumcision rates among the less educated, those living in major cities, those with low incomes and unskilled workers.
| Sample size | Circumcised % | ORa | 95% CI | p value | |
|---|---|---|---|---|---|
| Age | |||||
| 16–19 | 366 | 27.0 | 0.68 | (0.47–0.97) | 0.03 |
| 20–29 | 737 | 35.3 | 1.00 | – | – |
| 30–39 | 807 | 62.0 | 2.99 | (2.38–3.76) | <0.001 |
| 40–49 | 1,053 | 67.2 | 3.74 | (2.99–4.68) | <0.001 |
| 50–59 | 987 | 70.3 | 4.33 | (3.43–5.46) | <0.001 |
| 60–64 | 333 | 66.2 | 3.61 | (2.71–4.79) | <0.001 |
| Language spoken at home | |||||
| Other | 238 | 21.0 | 1.00 | – | – |
| English | 4,043 | 60.1 | 5.66 | 3.96–8.07 | <0.001 |
| Country of birth | |||||
| Australia | 3,308 | 66.2 | 1.00 | – | – |
| Other English‐speaking country | 476 | 34.5 | 0.27 | (0.22–0.33) | <0.001 |
| Rest of Europe | 154 | 15.0 | 0.09 | (0.06–0.14) | <0.001 |
| Asia | 231 | 24.2 | 0.16 | (0.11–0.23) | <0.001 |
| Other | 111 | 41.5 | 0.36 | (0.24–0.55) | <0.001 |
| Religion | |||||
| None | 2,158 | 57.8 | 1.00 | – | – |
| Christian | 1,934 | 58.8 | 1.04 | (0.90–1.19) | 0.6 |
| Buddhist | 48 | 27.1 | 0.27 | (0.13–0.55) | <0.001 |
| Muslim | 42 | 74.5 | 2.13 | (0.85–5.34) | 0.1 |
| Other non‐Christian | 93 | 52.2 | 0.80 | (0.51–1.24) | 0.3 |
| Religious attendance | |||||
| Less than monthly | 1,496 | 59.5 | 1.00 | – | – |
| Monthly or more often | 626 | 54.5 | 0.81 | (0.66–1.01) | 0.06 |
| Education | |||||
| Lower secondary | 1,054 | 54.4 | 0.85 | (0.72–1.00) | 0.05 |
| Secondary | 2,199 | 58.5 | 1.00 | – | – |
| Post‐secondary | 1,028 | 60.4 | 1.08 | (0.92–1.28) | 0.3 |
| Region of residence | |||||
| Major city | 2,191 | 54.7 | 1.00 | – | – |
| Regional | 1,857 | 61.3 | 1.31 | (1.14–1.51) | <0.001 |
| Remote | 173 | 68.1 | 1.77 | (1.24–2.52) | 0.002 |
| Household income | |||||
| Low (<$20,000) | 316 | 49.4 | 0.70 | (0.53–0.91) | 0.007 |
| Middle ($20,000–$60,000) | 1,310 | 58.3 | 1.00 | – | – |
| High (>$60,000) | 2,260 | 62.5 | 1.19 | (1.03–1.38) | 0.02 |
| Occupational category | |||||
| Professional | 1,592 | 65.9 | 1.00 | ‐ | ‐ |
| Associate professional | 823 | 56.9 | 0.69 | (0.57–0.83) | <0.001 |
| Tradesperson | 1,148 | 58.9 | 0.74 | (0.63–0.88) | <0.001 |
| Unskilled | 633 | 42.6 | 0.38 | (0.31–0.48) | <0.001 |
- Notes: a) Odds of being circumcised in comparison with reference group.
- CI = confidence interval.
Sexually transmissible infections
There was no statistically significant difference in circumcision rates between those who had and had not experienced most STIs (Table 2). However, men who reported having had non‐specific urethritis were much more likely to be circumcised, and those who had had penile candidiasis (thrush infection) were much less likely to be circumcised. The interview also asked about experience of syphilis and test results for HIV. However, these diseases were too rare (<0.5%) in the sample, and too heavily correlated with homosexual activity, for analysis by circumcision status to be practicable.
| Sample size | % with STI | ORa | 95% CI | p value | |
|---|---|---|---|---|---|
| Genital warts | |||||
| Uncircumcised | 1,623 | 2.8 | 1.00 | – | – |
| Circumcised | 2,370 | 4.5 | 1.37 | (0.97–1.95) | 0.08 |
| Chlamydiab | |||||
| Uncircumcised | 1,601 | 1.9 | 1.00 | – | – |
| Circumcised | 2,353 | 2.5 | 1.39 | (0.90–2.15) | 0.1 |
| Genital herpes | |||||
| Uncircumcised | 1,624 | 1.7 | 1.00 | – | – |
| Circumcised | 2,372 | 2.3 | 1.10 | (0.68–1.76) | 0.7 |
| Gonorrhoeab | |||||
| Uncircumcised | 1,602 | 1.8 | 1.00 | – | – |
| Circumcised | 2,354 | 2.2 | 0.90 | (0.55–1.50) | 0.7 |
| Non‐specific urethritisb | |||||
| Uncircumcised | 1,601 | 2.1 | 1.00 | – | – |
| Circumcised | 2,345 | 5.8 | 2.16 | (1.46–3.21) | <0.001 |
| Penile candidiasis | |||||
| Uncircumcised | 1,620 | 7.7 | 1.00 | – | – |
| Circumcised | 2,364 | 4.9 | 0.49 | (0.38–0.64) | <0.001 |
| Pubic lice | |||||
| Uncircumcised | 1,624 | 5.7 | 1.00 | – | – |
| Circumcised | 2,374 | 8.7 | 1.25 | (0.93–1.67) | 0.1 |
- Notes: a) Odds of reporting the condition in comparison with reference group (uncircumcised), adjusted for age category, lifetime number of opposite‐sex partners and lifetime number of same‐sex partners.
- b) Asked only of those men who had had vaginal or anal intercourse.
- CI = confidence interval
Sexual difficulties
Most sexual difficulties did not appear to affect circumcised men differently from uncircumcised men (Table 3). There was no reported significant association between erectile difficulties and circumcision, even when accounting for any age interaction (analysis available on request). Fewer circumcised men than uncircumcised men said they worried during sex whether their body looked unattractive, and though this worry is more common in younger men, the association was statistically significant after adjustment for age category (p=0.04).
| Sample size | % with difficulty | ORa | 95% CI | p value | |
|---|---|---|---|---|---|
| Lacked interest in having sex | |||||
| Uncircumcised | 1,484 | 21.3 | 1.00 | – | – |
| Circumcised | 2,185 | 18.8 | 0.86 | (0.72–1.03) | 0.1 |
| Came to orgasm too quickly | |||||
| Uncircumcised | 1,477 | 9.5 | 1.00 | – | – |
| Circumcised | 2,173 | 12.9 | 1.24 | (0.97–1.57) | 0.09 |
| Felt anxious about ability to perform sexually | |||||
| Uncircumcised | 1,478 | 11.2 | 1.00 | – | – |
| Circumcised | 2,178 | 10.6 | 0.87 | (0.68–1.11) | 0.3 |
| Trouble keeping erection | |||||
| Uncircumcised | 1,476 | 6.6 | 1.00 | – | – |
| Circumcised | 2,180 | 8.9 | 1.09 | (0.82–1.46) | 0.5 |
| Did not find sex pleasurable | |||||
| Uncircumcised | 1,478 | 3.8 | 1.00 | – | – |
| Circumcised | 2,178 | 3.3 | 0.88 | (0.57–1.36) | 0.6 |
| Unable to come to orgasm | |||||
| Uncircumcised | 1,482 | 3.9 | 1.00 | – | – |
| Circumcised | 2,178 | 3.8 | 0.76 | (0.53–1.09) | 0.1 |
| Physical pain during intercourse | |||||
| Uncircumcised | 1,468 | 1.7 | 1.00 | – | – |
| Circumcised | 2,176 | 1.5 | 0.85 | (0.46–1.56) | 0.6 |
| Worried during sex about whether body looked attractive | |||||
| Uncircumcised | 1,480 | 13.1 | 1.00 | – | – |
| Circumcised | 2,182 | 9.1 | 0.77 | (0.61–0.99) | 0.04 |
| Any of the above difficulties | |||||
| Uncircumcised | 1,484 | 40.4 | 1.00 | – | – |
| Circumcised | 2,185 | 39.3 | 0.95 | (0.82–1.11) | 0.5 |
- Notes: a) Odds of reporting difficulty in comparison with reference group (uncircumcised), adjusted for age category.
- CI = confidence interval
Sexual practice
Table 4 shows the sexual practices engaged in at last sexual encounter with an opposite‐sex partner. Other practices such as nipple stimulation and anal intercourse were not asked about. There was no evidence that uncircumcised men were less likely to be fellated by their partners than circumcised men. Fewer circumcised men (65.3%) than uncircumcised men (68.2%) had masturbated alone in the previous year, but after adjustment for age circumcised men were found to be more likely to have masturbated.
| Sample size | % engaged in activity | ORa | 95% CI | p value | |
|---|---|---|---|---|---|
| Masturbated alone in the last 12 months | |||||
| Uncircumcised | 1,651 | 68.2 | 1.00 | – | – |
| Circumcised | 2,410 | 65.3 | 1.20 | (1.03–1.40) | 0.02 |
| Vaginal intercourse | |||||
| Uncircumcised | 1,445 | 97.0 | 1.00 | – | – |
| Circumcised | 2,162 | 97.5 | 1.32 | (0.87–2.00) | 0.2 |
| Fellatio | |||||
| Uncircumcised | 1,404 | 38.3 | 1.00 | – | – |
| Circumcised | 2,125 | 31.7 | 0.93 | (0.79–1.09) | 0.4 |
| Cunnilingus | |||||
| Uncircumcised | 1,404 | 42.2 | 1.00 | – | – |
| Circumcised | 2,128 | 37.8 | 0.97 | (0.83–1.13) | 0.7 |
| Manual stimulation of male partner | |||||
| Uncircumcised | 906 | 69.9 | 1.00 | – | – |
| Circumcised | 1,478 | 71.2 | 1.06 | (0.87–1.30) | 0.6 |
| Manual stimulation of female partner | |||||
| Uncircumcised | 846 | 80.4 | 1.00 | – | – |
| Circumcised | 1,348 | 81.3 | 0.99 | (0.77–1.27) | 0.9 |
- Notes: a) Odds of engaging in the activity in comparison with reference group (uncircumcised), adjusted for age category.
- CI = confidence interval
Discussion
Self‐report of circumcision status is likely to be highly accurate,10, 11 and few (probably less than 10%)12 of the men would have been circumcised after infancy.
Overall, the results were very similar to the Sex in Australia survey,6 which is consistent with the respondents being sampled from the same population four years later.
The protective effect of circumcision against candidiasis is confirmed. Also, non‐specific urethritis was significantly more common among circumcised men. This effect was not statistically significant in the earlier study (OR 1.27, 95% CI 0.81–2.00), although the pattern of responses was in the same direction in the two studies. As the incidence and prevalence of syphilis and HIV among heterosexuals in Australia are very low it was not possible to evaluate whether circumcision had any protective effect. Recent modelling studies suggest that the protective effect of circumcision in the context of male‐to‐male sex would not be sufficient to justify circumcision promotion among gay men for HIV prevention.13, 14
In the area of sexual difficulties, two estimates were notably different from Sex in Australia: those for trouble maintaining an erection and pain during intercourse. In the earlier study, trouble with erections was less common among those who were circumcised. The difference was largely apparent among those aged 50–59. In this study, the effect appears to have vanished, both in general and when the men over 50 are examined separately. Physical pain during intercourse is rare among men (less than 2% of men reported it), yet it was significantly higher among uncircumcised men in the earlier study. Although physiologically plausible as a result of minor damage to the foreskin, there was no evidence for this phenomenon in this sample.
Distraction during sex by concern about one's bodily appearance is much more common among women than men and diminishes with age.15 The reason for the correlation found here with being uncircumcised is unknown. Given the strong associations between lack of circumcision and socio‐demographic variables that might relate to sexual anxiety or body image issues (such as religion and non‐English‐speaking background), there is little reason to conclude that the association is due to concern about the uncircumcised penis per se.
After adjustment for age, circumcised men were found to be more likely to have masturbated in the past year and this finding was statistically significant (OR 1.20, p=0.02), though the actual difference was small. This result was not statistically significant in Sex in Australia (OR 1.15, p=0.2) but the direction of effect was the same.
The data analysed here were drawn from a large interview study on a range of sexual health and relationship issues. Thus more detailed data of interest specifically in relation to circumcision were not available. No questions were asked about age at circumcision, or about indications for circumcision for those not circumcised soon after birth. We also lack data on surgical or psychological complications, perceived penile sensitivity and satisfaction with the procedure. A further limitation of the study is that self‐report is not ideal for ascertaining lifetime STI history. However, studies of diagnosed STIs are limited to clinical or screened samples, rather than the large representative sample studied here.
Conclusion
At the population level, circumcision appears to have minimal protective effects on sexual health in the Australian context. Due to the low prevalence of HIV in Australia and its concentration among homosexually active men, the study provides no evidence about protection against HIV.
Acknowledgements
We thank all the men who took part in interviews, and the staff of the Hunter Valley Research Foundation who did the telephone interviews.
Funding
The Australian Longitudinal Study of Health and Relationships is funded by the National Health and Medical Research Council.
Notes :
- Notes: a) Odds of being circumcised in comparison with reference group.
- CI = confidence interval.
- Notes: a) Odds of reporting the condition in comparison with reference group (uncircumcised), adjusted for age category, lifetime number of opposite‐sex partners and lifetime number of same‐sex partners.
- b) Asked only of those men who had had vaginal or anal intercourse.
- CI = confidence interval
- Notes: a) Odds of reporting difficulty in comparison with reference group (uncircumcised), adjusted for age category.
- CI = confidence interval
- Notes: a) Odds of engaging in the activity in comparison with reference group (uncircumcised), adjusted for age category.
- CI = confidence interval
Number of times cited: 19
- Brian J Morris, Sean E Kennedy, Alex D Wodak, Adrian Mindel, David Golovsky, Leslie Schrieber, Eugenie R Lumbers, David J Handelsman and John B Ziegler, Early infant male circumcision: Systematic review, risk-benefit analysis, and progress in policy, World Journal of Clinical Pediatrics, 6, 1, (89), (2017).
- Monica P.C. Nordstrom, Nelli Westercamp, Walter Jaoko, Timothy Okeyo and Robert C. Bailey, Medical Male Circumcision Is Associated With Improvements in Pain During Intercourse and Sexual Satisfaction in Kenya, The Journal of Sexual Medicine, 14, 4, (601), (2017).
- Michael T. Brady, Newborn Male Circumcision with Parental Consent, as Stated in the AAP Circumcision Policy Statement, Is Both Legal and Ethical, The Journal of Law, Medicine & Ethics, 44, 2, (256), (2016).
- Andreas Wahl Blomkvist and Fredrik Eika, Re: Omskjæring av gutter, Tidsskrift for Den norske legeforening, 136, 1, (8), (2016).
- Robert Darby, Targeting Patients Who Cannot Object? Re-Examining the Case for Non-Therapeutic Infant Circumcision, SAGE Open, 6, 2, (215824401664921), (2016).
- Brian J Morris, Richard G Wamai, Esther B Henebeng, Aaron AR Tobian, Jeffrey D Klausner, Joya Banerjee and Catherine A Hankins, Estimation of country-specific and global prevalence of male circumcision, Population Health Metrics, 14, 1, (2016).
- Gregory J. Boyle, Circumcision of Infants and Children: Short-Term Trauma and Long-Term Psychosexual Harm, Advances in Sexual Medicine, 05, 02, (22), (2015).
- Gregory J. Boyle, Does Male Circumcision Adversely Affect Sexual Sensation, Function, or Satisfaction? Critical Comment on Morris and Krieger (2013), Advances in Sexual Medicine, 05, 02, (7), (2015).
- Robert Darby, Syphilis 1855 and HIV-AIDS 2007: Historical reflections on the tendency to blame human anatomy for the action of micro-organisms, Global Public Health, 10, 5-6, (573), (2015).
- Angelika F Na, Sharman PT Tanny and John M Hutson, Circumcision: Is it worth it for 21st‐century Australian boys?, Journal of Paediatrics and Child Health, 51, 6, (580-583), (2015).
- Jennifer A. Bossio, Caroline F. Pukall and Stephen Steele, A Review of the Current State of the Male Circumcision Literature, The Journal of Sexual Medicine, 11, 12, (2847-2864), (2014).
- Brian J. Morris and John N. Krieger, Does Male Circumcision Affect Sexual Function, Sensitivity, or Satisfaction?—A Systematic Review, The Journal of Sexual Medicine, 10, 11, (2644-2657), (2013).
- Robert S. Van Howe, Sexually Transmitted Infections and Male Circumcision: A Systematic Review and Meta-Analysis, ISRN Urology, 2013, (1), (2013).
- B. J. Morris, A. D. Wodak, A. Mindel, L. Schrieber, K. A. Duggan, A. Dilley, R. J. Willcourt, M. Lowy and D. A. Cooper, The 2010 Royal Australasian College of Physicians' policy statement ‘Circumcision of infant males’ is not evidence based, Internal Medicine Journal, 42, 7, (822-828), (2012).
- Brian J. Morris, Alex D. Wodak, Adrian Mindel, Leslie Schrieber, Karen A. Duggan, Anthony Dilley, Robin J. Willcourt, Michael Lowy, David A. Cooper, Eugenie R. Lumbers, C. Terry Russell and Stephen R. Leeder, Infant male circumcision: An evidence-based policy statement, Open Journal of Preventive Medicine, 02, 01, (79), (2012).
- Brian J Morris, Jake H Waskett, Joya Banerjee, Richard G Wamai, Aaron AR Tobian, Ronald H Gray, Stefan A Bailis, Robert C Bailey, Jeffrey D Klausner, Robin J Willcourt, Daniel T Halperin, Thomas E Wiswell and Adrian Mindel, A 'snip' in time: what is the best age to circumcise?, BMC Pediatrics, 12, 1, (2012).
- Brian J. Morris, Robert C. Bailey, Jeffrey D. Klausner, Arleen Leibowitz, Richard G. Wamai, Jake H. Waskett, Joya Banerjee, Daniel T. Halperin, Laurie Zoloth, Helen A. Weiss and Catherine A. Hankins, Review: A critical evaluation of arguments opposing male circumcision for HIV prevention in developed countries, AIDS Care, 24, 12, (1565), (2012).
- Robert Darby and Robert Van Howe, Not a surgical vaccine: there is no case for boosting infant male circumcision to combat heterosexual transmission of HIV in Australia, Australian and New Zealand Journal of Public Health, 35, 5, (459-465), (2011).
- Brian J Morris and Alex Wodak, Circumcision survey misleading, Australian and New Zealand Journal of Public Health, 34, 6, (636-637), (2010).




