ORIGINAL RESEARCH—MEN'S SEXUAL HEALTH
A Psychometric Investigation of the Hypersexual Disorder Screening Inventory among Highly Sexually Active Gay and Bisexual Men: An Item Response Theory Analysis
Article first published online: 27 MAR 2013
© 2013 International Society for Sexual Medicine
The Journal of Sexual Medicine
Volume 10, Issue 12, pages 3088–3101, December 2013
How to Cite
Parsons, J. T., Rendina, H. J., Ventuneac, A., Cook, K. F., Grov, C. and Mustanski, B. (2013), A Psychometric Investigation of the Hypersexual Disorder Screening Inventory among Highly Sexually Active Gay and Bisexual Men: An Item Response Theory Analysis. Journal of Sexual Medicine, 10: 3088–3101. doi: 10.1111/jsm.12117
- Issue published online: 2 DEC 2013
- Article first published online: 27 MAR 2013
- National Institute of Mental Health. Grant Number: R01-MH087714
- Hypersexual Disorder;
- Gay and Bisexual Men;
- Sexual Compulsivity;
- Item Response Theory;
- Confirmatory Factor Analysis
The Hypersexual Disorder Screening Inventory (HDSI) was designed as an instrument for the screening of hypersexuality by the American Psychiatric Association's taskforce for the fifth edition of the Diagnostic and Statistical Manual of Mental Disorders.
Our study sought to conduct a psychometric analysis of the HDSI, including an investigation of its underlying structure and reliability utilizing item response theory (IRT) modeling, and an examination of its polythetic scoring criteria in comparison to a standard dimensionally based cutoff score.
We examined a diverse group of 202 highly sexually active gay and bisexual men in New York City. We conducted psychometric analyses of the HDSI, including both confirmatory factor analysis of its structure and IRT analysis of the item and scale reliabilities.
Main Outcome Measures
We utilized the HDSI.
The HDSI adequately fit a single-factor solution, although there was evidence that two of the items may measure a second factor that taps into sex as a form of coping. The scale showed evidence of strong reliability across much of the continuum of hypersexuality, and results suggested that, in addition to the proposed polythetic scoring criteria, a cutoff score of 20 on the severity index might be used for preliminary classification of HD.
The HDSI was found to be highly reliable, and results suggested that a unidimensional, quantitative conception of hypersexuality with a clinically relevant cutoff score may be more appropriate than a qualitative syndrome comprised of multiple distinct clusters of problems. However, we also found preliminary evidence that three clusters of symptoms may constitute an HD syndrome as opposed to the two clusters initially proposed. Future research is needed to determine which of these issues are characteristic of the hypersexuality and HD constructs themselves and which are more likely to be methodological artifacts of the HDSI. Parsons JT, Rendina HJ, Ventuneac A, Cook KF, Grov C, and Mustanski B. A psychometric investigation of the Hypersexual Disorder Screening Inventory among highly sexually active gay and bisexual men: An item response theory analysis. J Sex Med 2013;10:3088–3101.