Considerations in the Measurement and Reporting of Withdrawal: Findings from the 2018 National Survey of Sexual Health and Behavior

Tsung chieh Fu, Devon J. Hensel, Jonathon J. Beckmeyer, B. Dodge, D. Herbenick

Research output: Contribution to journalArticle

2 Scopus citations


Introduction: Although withdrawal use is routinely measured as part of contraceptive surveillance in the United States, its assessment may be prone to underreporting and measurement errors. Additionally, at the population level, little is known about subjective experiences of withdrawal. Aim: To measure respondents’ frequency of and reasons for engaging in extra-vaginal ejaculation in the past year, and to compare contraceptive withdrawal vs extra-vaginal/anal ejaculation during the most recent sexual event. Methods: The 2018 National Survey of Sexual Health and Behavior is a U.S. nationally representative probability survey of adolescents and adults. This study was administered in February–March 2018 via GfK Research's KnowledgePanel. Respondents who reported consensual penile-vaginal intercourse in the past year were included in the analysis. Logistic regression was used to assess factors associated with finding extra-vaginal ejaculation sexy/arousing. Main Outcome Measures: Main outcomes were the frequency of and reasons for extra-vaginal ejaculation in the past year, as well as reported contraceptive withdrawal use and the location of ejaculation at the most recent penile-vaginal intercourse. Results: In the past year, approximately 60% of respondents reported engaging in any extra-vaginal ejaculation; pregnancy prevention and sexual pleasure were given as common reasons. Those who were men, white, or aged 25–29 were more likely to find extra-vaginal ejaculation sexy or arousing; no gender differences were observed in their partner's perception of this act. Prevalence of selecting withdrawal on the contraceptive inventory was 12.2% (95% CI 10.6–14.0) at last sexual event, whereas extra-vaginal/anal ejaculation was reported by 20.8% (95% CI 18.7–23.0). This lack of concordance in differing measures was observed consistently across all age groups, and approximately 12% to 14% of individuals reported an ejaculation location that did not correspond to their withdrawal use. Clinical Implications: The results of this study have implications for how healthcare providers discuss withdrawal with their patients and counsel them on using extra-vaginal ejaculation as part of their contraceptive use or part of their sexual repertoire. Strengths & Limitations: This is the first nationally representative study to directly compare contraceptive withdrawal to extra-vaginal ejaculation. Limited data were collected at event level regarding motivations for extra-vaginal ejaculation, pregnancy attitudes, perceptions of condoms and sexually transmitted infection prevention, or exposure to erotic media, thus limiting further assessment of relationships between various factors and extra-vaginal/anal ejaculation. Conclusion: We found that reporting of withdrawal as a contraceptive method vs extra-vaginal/anal ejaculation was inconsistent and that extra-vaginal/anal ejaculation was commonly used for reasons other than pregnancy prevention; future research should use more precise measures of “withdrawal.” Fu TC, Hensel DJ, Beckmeyer JJ, et al. Considerations in the Measurement and Reporting of Withdrawal: Findings from the 2018 National Survey of Sexual Health and Behavior. J Sex Med 2019;16:1170–1177.

Original languageEnglish (US)
Pages (from-to)1170-1177
Number of pages8
JournalJournal of Sexual Medicine
Issue number8
StatePublished - Aug 2019


  • Coitus Interruptus
  • Contraception
  • Ejaculation
  • Orgasm
  • Sexual Health
  • United States

ASJC Scopus subject areas

  • Reproductive Medicine
  • Obstetrics and Gynecology
  • Urology

Fingerprint Dive into the research topics of 'Considerations in the Measurement and Reporting of Withdrawal: Findings from the 2018 National Survey of Sexual Health and Behavior'. Together they form a unique fingerprint.

  • Cite this