Article Text

Download PDFPDF

Self-report is a valid measure of circumcision status in homosexual men
  1. D J Templeton1,2,
  2. L Mao3,
  3. G P Prestage1,
  4. F Jin1,
  5. J M Kaldor1,
  6. A E Grulich1
  1. 1
    National Centre in HIV Epidemiology and Clinical Research, University of New South Wales, Sydney, Australia
  2. 2
    RPA Sexual Health, Royal Prince Alfred Hospital, Camperdown, Australia
  3. 3
    National Centre in HIV Social Research, University of New South Wales, Sydney, Australia
  1. Dr David J Templeton, National Centre in HIV Epidemiology and Clinical Research, The University of New South Wales, Level 2, 376 Victoria Street, Sydney NSW 2010, Australia; dtempleton{at}nchecr.unsw.edu.au

Abstract

Objective: Misreporting of circumcision status may affect observed relationships between circumcision status and HIV or other sexually transmissible infections. As no data exist on the validity of self-reported circumcision status among homosexual men, we investigated the agreement between self-report and examination findings in a subgroup of participants in the Health in Men (HIM) study in Sydney, Australia.

Methods: A subgroup of 240 participants in the community based HIM cohort study attending annual interview agreed to a brief genital examination by a trained study nurse who was unaware of their previous self-reported circumcision status.

Results: Five participants reported being uncircumcised at baseline but were classified as circumcised on examination. All participants who self-reported being circumcised were found on examination to be circumcised. Three cases in which the examining study nurse was unsure of participants’ circumcision status were excluded. Of the remaining 237 participants, 155 (65.4%) were classified as circumcised on examination, including five men who self identified as uncircumcised. Compared with examination, self-reported circumcision status resulted in a sensitivity of 96.8%, specificity of 100%, positive predictive value of 100% and negative predictive value of 94.3%. The overall agreement between circumcision status on examination and self-report was 97.9% (κ score, 0.95; p<0.001)

Conclusion: Self-report was a valid measure of circumcision status in this group of predominantly Anglo gay-community-attached men. We believe our findings can be generalised to similarly aged gay-community-attached men in other developed countries.

Statistics from Altmetric.com

Request Permissions

If you wish to reuse any or all of this article please use the link below which will take you to the Copyright Clearance Center’s RightsLink service. You will be able to get a quick price and instant permission to reuse the content in many different ways.

Validation of self-reported circumcision status by clinical examination has long been considered the “gold standard” in assessing circumcision status.1 Minimising measurement bias is important in observational research, as incorrect self-report may affect observed relationships in comparative studies of circumcision status and HIV or other sexually transmissible infections.2 Of published studies in the last decade that have examined the validity of self-reported circumcision status compared with clinical examination, overall agreement has ranged from 81% to almost 100% in developing countries.24 Over a quarter of predominantly black US adolescents did not know their circumcision status, although more than 90% of those who reported their status did so correctly.5 The only published Australian study reported over 98% agreement between self-report and clinical examination among 1300 adult sexually transmitted disease clinic attendees.6 As no published studies have addressed validation of self-reported circumcision status among homosexual men, we investigated the agreement between self-report and examination findings among a sub-group of participants in the Health in Men (HIM) study.

METHODS

Participants were HIV-negative men who were recruited mainly from community based sources from June 2001 to December 2004. The methods of the HIM study have been described in detail elsewhere.7

All participants were asked at baseline face-to-face interview to report their circumcision status. Between February and June 2007, a validation study was performed on participants attending for their annual face-to-face interview. A registered nurse obtained informed consent to perform a brief genital examination to assess participants’ circumcision status. Independent of this validation study, an additional indepth circumcision questionnaire was administered to a subgroup of consenting participants (n = 922) in the final year of the study. Where baseline self-report was uncircumcised but examination finding was circumcised, self-reported circumcision status in the additional questionnaire was also examined to exclude the possibility that the men were circumcised during the study. The male study nurse was extensively educated to recognise differing circumcision status and specifically directed not to discuss circumcision status with participants. The nurse was unaware of participants’ self-reported circumcision status recorded at both the baseline and the additional circumcision questionnaire.

Statistical analyses were performed using STATA 10.0 (STATA Corporation, College Station, Texas, USA). The validity of self-reported circumcision status at baseline was compared with physical examination and assessed using the κ score.

RESULTS

Of 284 HIM participants who attended for their annual interview between February and June 2007, 240 gave consent for the circumcision validation substudy: a participation rate of 85%. self-reported uncircumcised men were less likely than self-reported circumcised men to consent (p = 0.051), as were younger participants (p = 0.017) and those of non-Anglo ethnicity (p = 0.018). Country of birth, educational level attained and religious affiliation were not related to non-consent. When controlling for ethnicity and self-reported circumcision status, only younger age remained independently associated with non-consent (p = 0.045). Median age of participants was 41 years (range 22–75 years), 55% had a university education and 75% were of Anglo ethnicity. Five participants reported being uncircumcised at baseline but were classified as circumcised on examination. These five participants also reported being uncircumcised on the additional circumcision questionnaire thus excluding the possibility that they were circumcised during the study period. Four of these five participants had postgraduate university qualifications and four were of Anglo ethnicity. All participants who self-reported being circumcised were found on examination to be circumcised. In three cases the examining study nurse was unsure of participants’ circumcision status. Of these three participants, one reported being uncircumcised while the other two reported being circumcised in both questionnaires. The two participants who reported being circumcised were noted by the study nurse to have had past surgery involving partial removal of preputial tissue. Excluding these three cases (n = 237), 155 (65.4%) participants were classified as circumcised on examination. Compared with examination, self-reported circumcision status resulted in a sensitivity of 96.8% (150 of 155 circumcised participants correctly reported their status), specificity of 100% (all 82 uncircumcised participants correctly reported their status), positive predictive value of 100% (all 150 self-reports of being circumcised were confirmed on examination) and negative predictive value of 94.3% (82 of 87 self-reports of being uncircumcised were confirmed on examination). The overall agreement between circumcision status on examination and self-report was 97.9% (κ score 0.95; p<0.001)

DISCUSSION

Almost all participants in our study correctly identified their circumcision status. This is unsurprising, given the familiarity of homosexual men with the penis and the fact that circumcision has long been the subject of sexual fetish among many homosexual men.8

Misclassification of circumcision status among heterosexual men may have been more common decades ago1 because issues of sexuality were not as openly discussed as they are today. In addition, poor correlation observed in some studies from developing countries may be attributed to different surgical techniques employed in various parts of the world. Circumcision is performed more uniformly in developed countries compared with some parts of Africa where several different surgical techniques are used resulting in a wide range of post-circumcision penile appearances.9

It is possible that our study nurse may have incorrectly classified the circumcision status of participants. Although uncommon, misclassification by clinicians has been previously reported.10 We feel this possibility is unlikely given the training our male study nurse received. In addition, “partial circumcision” was not an optional field in data collection forms for interviews or examinations, which may have resulted in occasional discrepancies between self-report and circumcision status on examination. The positive predictive value of self-report depends on circumcision prevalence and may therefore be slightly lower among homosexual men of different ethnicity, country of birth and younger age in whom the procedure is less common.11

To our knowledge, this is the first study validating self-reported circumcision status with examination findings in homosexual men. The use of self-reported circumcision status appears to be a valid measure among this group of predominantly Anglo gay-community-attached men. We believe our findings can be generalised to similarly aged gay-community-attached men in other developed countries.

Key messages

  • In this first study assessing the validity of self reported circumcision status in men who have sex with men (MSM), Sydney community based MSM correctly reported their circumcision status in almost 98% of cases when compared to clinical examination.

  • All five cases of incorrect self reported circumcision status were in circumcised men.

  • The excellent validity of self reported circumcision status can be generalised to MSM of similar age in other developed countries.

Ethics approval was obtained from the University of New South Wales, Australia.

Acknowledgments

The authors wish to thank Andrew Frankland for assistance with data management, the anonymous reviewers of the original manuscript, all the HIM participants and the dedicated HIM study team.

REFERENCES

Footnotes

  • Contributors: DJT had the original idea and developed the protocol for the validation study, performed the analyses and drafted the manuscript; AEG took overall responsibility for the project and assisted in the analyses and drafting of the manuscript; LM, GPP, FJ and JMK assisted in formulating the analyses and drafting the manuscript.

  • Funding: DJT is supported by National Health and Medical Research Council Public Health Scholarship no. 351044. The National Centre in HIV Epidemiology and Clinical Research and the National Centre in HIV Social Research are funded by the Australian Government Department of Health and Ageing. The Health in Men Cohort study was funded by the National Institutes of Health, a component of the US Department of Health and Human Services (NIH/NIAID/DAIDS: HVDDT Award N01-AI-05395), the Australian Government Department of Health and Ageing (Canberra) and the New South Wales Health Department (Sydney), and the National Health and Medical Research Council (project grant # 400944).

  • Competing interests: None.

Linked Articles

  • Brief encounters
    Helen Ward Rob Miller