Women's Sexual Function and Dysfunction. Irwin Goldstein MD

Assessing the prevalence of female sexual dysfunction with surveys: what is feasible?

Cynthia A Graham, John Bancroft

There is current widespread interest in the prevalence of “female sexual dysfunction”, driven, in part, by industrial hopes that drugs such as sildenafil will have beneficial effects, generating a large market for women as well as men. At the same time, there has also been growing dissatisfaction with the current classification of sexual dysfunction, particularly for women.1,2

The purpose of this chapter is to review and evaluate the use of community-based surveys to assess the frequency of sexual dysfunction in women. As a first step, we consider the meaning of the term “female sexual dysfunction”. We conclude that, whereas sexual dysfunction is a useful clinical concept, only some of the criteria required to make such diagnoses can be established in large, population-based surveys. However, there are a number of important aspects relevant to prevalence and etiology that can be addressed, and suggestions will be made for improving survey design for such purposes.

In 1999, a paper in the Journal of the American Medical Association3 reported that 43% of American women have a sexual dysfunction. This figure, which will be examined more closely in this chapter, has been extensively cited in the scientific literature, in the media, and in self-help books for women*4,5 We argue that the very limited questions asked in this survey fell far short of what is needed to identify a “sexual dysfunction” (see fuller discussion in refs 6 and 7).

*As of 1 March 2004, over 300 citations in the medical and psychological literature were identified by the Institute for Scientific Information’s “Web of Science”.

What is “sexual dysfunction”?

So what does “sexual dysfunction” mean, and how does it differ from a more general term such as “sexual problem” or “sexual difficulty”? It was probably Masters and Johnson, in 1970, who initiated the current use of the term “sexual dysfunction”. It did appear in the medical literature prior to 1970, but always in reference to sexual side effects caused by medication or surgery.^10 Although in their first book Masters and Johnson11 put forward their idea of a sexual response cycle that is basically the same in men and women, in their second book12 they categorized sexual dysfunction for men and for women quite differently. With the male they focused on “impotence” (primary and secondary), with premature ejaculation and “ejaculatory incompetence” as additional categories. With the female, they focused on “orgasmic dysfunction”, with “vaginismus” as an additional, specifically female dysfunction. “Dyspareunia” was considered relevant to both men and women. Kaplan13 used the concept of “general female sexual dysfunction”, noting that this was “usually called frigidity”. In addition, she described “orgastic dysfunction”, drawing the distinction between “inhibition” of the “vasocon-gestive” and “orgasm” components of sexual response. Both Kaplan and Masters and Johnson therefore continued to use the conventional but pejorative terms for the principal types of male and female dysfunction, i.e., “impotence” and “frigidity”.

The first two editions of the Diagnostic and Statistical Manual of Mental Disorders (DSM-I and -II)14,15 did not use the concept “sexual dysfunction”. DSM-I,14 under the general heading “psychophysiological autonomic and visceral disorders”, included the terms “frigidity”, “impotence”, “premature ejaculation of semen”, and “vaginismus” in a list of supplementary terms for the urogenital system. In the second edition, DSM-II,15 only dyspareunia and impotence were listed as examples of “psychophysiological genito-urinary disorders”.

In 1980, the concept of “psychosexual dysfunction” appeared in the third edition of the DSM (DSM-III).16The DSM approach to sexual dysfunction was clearly based on Masters and Johnson’s ideas, although, interestingly, not their ideas about sexual dysfunction but rather their concept of the sexual response cycle,11 modified by Kaplan’s13 addition of “sexual desire” as a separate component. This resulted in male and female diagnostic categories that were conceptually similar: “inhibited sexual desire” (neither “inhibited” nor “sexual desire” were defined), “inhibited sexual excitement”, “inhibited orgasm”, and “functional dyspareunia”, each with a male and female version. “Functional vaginismus” was included as a specifically female dysfunction and premature ejaculation as a male dysfunction.

The current (fourth) edition of the DSM (DSM-IV)17 defines sexual dysfunction as characterized by disturbance in sexual desire and in the psychophysiologic changes that characterize the sexual response cycle, and as causing marked distress and interpersonal difficulty. The concept of “inhibition” no longer features. Thus, there is hypoactive sexual desire disorder and sexual aversion disorder, defined in the same way for men and women. Female sexual arousal disorder is defined as “a persistent or recurrent inability to attain, or to maintain until completion of the sexual activity, an adequate lubrication-swelling response of sexual excitement”, and for the male version, erection is the relevant response. Orgasmic disorder (i.e., delayed or absent orgasm) and dyspareunia are defined in basically the same way for men and women. Vaginismus is a specifically female disorder.

The International Statistical Classification of Diseases and Related Health Problems (ICD-10)18 covers sexual dysfunctions in one and a half pages, compared with nearly 30 pages in the current DSM. The basic categories of dysfunction are similar to those of the DSM, but there are few, if any, actual diagnostic criteria given for any of the dysfunctions. The ICD-10 also does not require that personal distress or interpersonal problems be present for a diagnosis to be made. Instead, it states, “sexual dysfunction covers the various ways in which an individual is unable to participate in a sexual relationship as he or she would wish”.

“Political correctness”

The term “frigidity” has more or less disappeared from clinical and research use, although the term “impotence” continues to be used by some sections of the medical profession. Both terms can be regarded as male-centered; impotence, from erectile failure, rendering the man powerless; frigid indicating absence of warmth, with a consequent chilling effect on the male partner. The use by the DSM of the “sexual response cycle” as a predictable sequence of basically physiologic events, essentially the same in women and men, can be seen as an attempt to redress the societal view, which has prevailed in the past, that women’s sexuality is something fundamentally different from that of men. Since such a distinction was central to the long-standing societal repression of female sexuality, challenging this distinction had obvious sociopolitical significance, and may have been regarded by some as “politically correct” for that reason.

However, although Masters and Johnson proposed a “sexual response cycle” that in physiologic terms was basically the same for women and men (and we accept Tiefer’s19 criticisms of that concept), they also emphasized the impact of sociocultural factors on women’s sexuality. “Negation of female sexuality, which discourages the development of an effectively useful sexual value system, has been an exercise of the so-called double standard and its socio-cultural precursors”.12 They add, “Socio-cultural influence more often than not places woman in a position in which she must adapt, sublimate, inhibit or even distort her natural capacity to function sexually in order to fulfill her genetically assigned role. Herein lies a major source of woman’s sexual dysfunction”. Masters and Johnson did not appear to equate the sexual dysfunctions of men and women, either in their diagnostic categories, or in the above, telling statement. But if their other ideas on the sexual response cycle were used by clinicians to achieve “political correctness” through the DSM process, “political correctness” can be regarded as no longer the same on this issue. There is now growing recognition of, and emphasis on, gender differences in sexuality. Alternative models of sexual response20 and new, women-centered definitions of sexual problems in women21 have been proposed. In a recent national survey of heterosexual American women,6 the main predictors of distress about sex were not the more physiologic factors that are the basis of the DSM-IV (e.g., lubrication, frequency of orgasm) but were factors related to a woman’s mood and the subjective quality of the sexual interaction with the partner. And, as early studies evaluating the effects of sildenafil in women have proven disappointing,22 even the pharmaceutical industry is moving toward a “politically correct” view that there are fundamental differences between men and women’s sexuality. Pfizer recently publicly announced that, after 8 years of work, they had terminated their research program evaluating the effects of sildenafil on women, citing greater complexity of women’s sexual response and a “disconnect between genital changes and mental changes” in women as reasons.23

The issue of distress

Although an essential criterion for the DSM-IV17 diagnosis of any sexual dysfunction is that the problem causes “marked distress or interpersonal difficulty”, this is a relatively recent development. In the revised third edition of the DSM (DSM- III-R),24 distress or interpersonal difficulty was not required. Why this requirement that a sexual problem cause “marked distress” before being considered a “dysfunction” was added to the DSM-IV is not clear. The recent International Consensus Development Conference Report, written by a multidisciplinary group of experts who reappraised the DSM categories and proposed a slightly modified classification system for female sexual dysfunction,25 recommended that the presence of “personal distress” be a requirement for any diagnosis, as distinct from the “marked distress or interpersonal difficulty” required in the DSM. The rationale for this emphasis on specifically “personal” distress, while excluding interpersonal distress, was not given. The issue of “distress” is particularly relevant to low sexual desire, which is often not a cause of concern for the individual but may cause problems in the relationship; in such cases, using the new consensus classification system, a diagnosis of sexual dysfunction would not be given. The whole issue of “distress” as an essential diagnostic criterion is complex. Distress, it could be argued, is a reaction to the dysfunction and thus has no explanatory value. It is of clinical relevance, but mainly because it determines whether the individual is motivated to change or wants help; the person who is not distressed by the dysfunction, or whose sexual relationship is not affected by it, is unlikely to seek treatment. It does not help in deciding which treatment is most likely to be effective. On the other hand, if our goal is to conceptualize “sexual dysfunction” in women’s terms, capturing the problems and difficulties which are the principal concern for women, then we need to understand what worries or distresses them. But it may well turn out not to be the DSM diagnostic criteria.

Previous community-based surveys

In a review article on the prevalence of sexual dysfunctions, Simons and Carey26 pointed to a lack of methodological rigor in many of the studies. Of the 52 studies reviewed, more than one- third did not provide any operational definition of the dysfunction being assessed. Moreover, of studies that purportedly used the DSM criteria to establish sexual dysfunction, few assessed the presence of “marked distress or interpersonal difficulty” or the absence of an Axis I disorder (both essential criteria for a DSM diagnosis).

Recently, Bancroft and colleagues6 compared their survey of women with four earlier studies involving substantial community based samples of women.3,27-30 Three relevant conclusions were drawn from this comparison: (1) arbitrary definitions of what constituted a “sexual problem” were used, resulting in considerable variability across studies in the prevalence of specific sexual problems; (2) there was limited overlap between women who defined themselves as having a sexual problem and those assigned to one of the researcher-defined problem categories; (3) a strong association was found between the presence of “sexual problems” and other problems in the women’s lives, most notably mood problems and relationship difficulties.3,6,28,30

In 2003, two new British studies using community-based samples were reported. Mercer and colleagues31 used secondary data from a national probability sample survey of sexual attitudes and lifestyles carried out in 2000.32,33 In this survey, which obtained a response rate of 65.4%, and included 5530 women aged 16-44 years who reported at least one heterosexual partner in the previous year, 53.8% of the women indicated at least one sexual problem lasting at least 1 month during the previous year, the most common problem being “lacked interest in sex” (40.6% of women vs 17.1% of men). “Persistent problems” were defined as those lasting at least 6 months; 15.6% of women reported at least one persistent problem, with 10.2% citing lack of interest in sex as a persistent problem (compared with 1.8% of the men). Overall, 21.0% of women and 10.5% of men had sought help for a sexual problem in the previous year. The striking contrast between the prevalence of “short duration” (53.8%) and “persistent” problems (15.6%), particularly relating to sexual interest, confronts us with the possibility that transient impairment may result from situational factors, and may not indicate “dysfunction” per se. On the other hand, recurrence of short-term adaptive impairments of this kind may indicate vulnerability to sexual problems.34 We will consider this further later in the chapter.

Nazareth and colleagues35 assessed prevalence of sexual dysfunction in a sample of 1065 women and 447 men attending general practitioners in England. This paper, published in the British Medical Journal, is brief. We also find it perplexing. Much is made in the paper of the use of ICD-10 diagnostic criteria, “stringently applied”, yet, as stated earlier, ICD-10 gives little more than headings and brief descriptions, and certainly no diagnostic criteria. In this study, the “diagnostic criteria” are based on the past 4 weeks. While this time period may be preferable to obtain reliable estimates of frequencies (see below), no clinician is going to make a diagnosis of sexual dysfunction based only on the past 4 weeks.

In an early but still highly relevant review of the survey literature, Nathan36 suggested five ways in which epidemiologic data on rates of sexual dysfunction could prove useful: service planning, prevention efforts (e.g. by highlighting high-risk groups), providing normative data, generating hypotheses about causal factors, and, lastly, by providing data that might challenge existing definitions of “dysfunction”. Recent survey efforts have focused on the first of these applications, but much less attention has been paid to the other four possibilities. For example, in reporting their conclusion that 43% of women have a sexual dysfunction, Laumann and colleagues3 stated that “with the affected population rarely receiving medical therapy for sexual dysfunction, service delivery efforts should be augmented to target high-risk populations”, what has been described as a clear example of “medicalization”.37

How do we conceptualize women’s sexual problems? The need for a theoretical approach

Simons and Carey25 concluded that use of “a common classification system”, such as the DSM, will lead to better surveys. We would argue that any such classificatory system should be based on good, clinically relevant research that identifies what women see as important sexual problems. This does not appear to be the case with the DSM. Research has shown that sexual satisfaction for many women is closely linked to relationship factors, such as feeling “connected” with a sexual partner.6,38 As Tiefer21 has argued, the DSM “bypasses” these relational aspects of women’s sexuality and assumes an exclusively individual approach to sexuality. Other researchers have highlighted the importance of partner-related factors to women’s sexual satisfaction.38,39 In a survey, Ellison38 asked over 2000 women to distinguish between experiences that they considered “problems” and those that they thought of as “just the way life is”. Experiences such as feeling “too tired” or “too busy” for sex were less likely to be reported as problems and were most likely to be considered as “just the way life is”. At a minimum, studies should assess whether women perceive situations such as inability to reach orgasm as a “problem” or not. It is also important to establish whether a woman is receiving what is for her appropriate stimulation during sexual activity with her partner. In some cases, a woman’s failure to experience an orgasm may have more to do with her partner’s behavior than whether she is “dysfunctional”. But we also need a much better understanding of how women experience sexual problems, in their own terms. To do this requires research, with both nonclinical and clinical samples, into basic questions such as how women experience sexual desire and arousal.

Notably absent from this literature is a theoretical approach which might enable us to conceptualize sexual dysfunctions in ways that are relevant and appropriate to each gender, and which can be empirically tested. Recently, the dual control model has been presented as one way to fill this gap.40,41 This postulates that, for the majority of individuals, inhibition of sexual response is an adaptive mechanism that reduces the likelihood of sexual responses in circumstances where they would be inappropriate or disadvantageous, a mechanism of relevance across species. This reintroduces the concept of inhibition, but with the important difference that it is not necessarily dysfunctional. If we accept this basic premise, we are then faced with the fundamental challenge of distinguishing between a reduction in sexual responsiveness that is appropriate or adaptive given the current circumstances, and one that is a manifestation of some malfunctioning of the sexual response system that might appropriately be called a dysfunction. This may result from increased inhibition or reduced excitation. The focus of any intervention, for the first pattern, would then be the circumstances that elicited the adaptive response (e.g., improving a relationship problem, or dealing with stressful life circumstances). The focus for the second pattern would involve, in some way, treating the sexual dysfunction.

Keeping this fundamental distinction in mind, we can then look at each individual’s case through three conceptual “win- dows”.6

(1) Through the first window, we look at the woman’s current situation. To what extent are problems in her relationship with her sexual partner, or in her life more generally, resulting in adaptive inhibition of sexual responsiveness?

(2) Through the second window, we consider the individual’s sexual history. To what extent has the current pattern of nonresponsiveness been evident earlier in her life? Here we can use a second basic premise from the dual control model: that individuals vary in their propensity for sexual inhibition. Our attempts to measure such propensities in men42 and, more recently, in women (Graham, Sanders, and Milhausen, unpublished data) have supported this premise; the majority of individuals score somewhere in the middle, what we can call “adaptive”, range for inhibition proneness. But some individuals score high in this respect, and we have found a strong association between such scores and clinically established erectile dysfunction in men (Bancroft, Herbenick, Barnes et al., unpublished data). We have also found that men scoring low on inhibition are more likely to take sexual risks.43,44 Comparable research with women is still at an early stage, but we anticipate similar associations, with the possibility that inhibition proneness may play a greater role in the sexuality of women, with a wider range of inhibition-inducing circumstances, and hence with a greater likelihood of inhibition being “adaptive” rather than dysfunctional. The origins or determinants of such individual differences, whether they be genetic factors or early learning, or the effects of early traumatic sexual experiences, remain to be established.

(3) Through the third window, we can look for evidence of aging, medical problems, hormonal changes or the side effects of drugs, which might directly impair sexual functioning.

We can meaningfully consider problems explained through the second and third windows as “sexual dysfunctions”. To make these distinctions requires careful clinical assessment, and a not inconsiderable amount of clinical judgment, taking into account a woman’s ethnic, cultural, religious, and social background as well as age and experience, current health status, frequency and duration of the sexual problems, and distress.

The importance of establishing norms from community surveys is particularly pertinent to studying the prevalence of “sexual desire problems”. Prevalence rates for low sexual desire vary greatly across different studies, more so than for other dys- functions6 but are consistently higher for women than for men.3,31 Surveys reporting prevalence rates for sexual desire disorders have typically asked women about whether they have experienced “problems” such as “lacking desire for sex” 3 or “decreased interest in sex”.29 Yet, rarely have researchers looked at how women interpret “lacking desire for sex”. This is particularly important given the comorbidity between arousal and desire disorders25 and the observation that women often do not differentiate “arousal” from “desire”.38,45 Although there are no established norms, there is an extensive literature on gender differences in the frequency of thoughts and fantasies about sex, the desired frequency of sexual activity and of initiating sex.46 Research has consistently found that men report more frequent sexual thoughts and fantasies. In one study of 40-year-old Danish women, 33% reported that they never experienced “spontaneous libido”, yet most of those women enjoyed sexual activity once their partners had initiated it.47 With our theoretical model, there is also the question of whether inhibition of sexual interest, as an “adaptive” reaction to current circumstances, and hence transitory, is more common in women than in men. This is highly relevant to deciding whether in a particular case “low sexual desire” should be regarded as a dysfunction. And we are still left with the question of how appropriate it is to assess the frequency of sexual thoughts as an indicator of “sexual interest” in women. As Heiman48 has argued, we need to consider what we specifically want to capture in our assessment of sexual desire in women; “desire” is an elusive concept.

Designing better community-based studies - methodological issues

Survey design

Most community-based surveys have been and will continue to be cross-sectional. Longitudinal studies have obvious and considerable advantages, although because of their long-term nature, they are exceptionally difficult to implement. Denner- stein and colleagues39 have reported on a prospective, population-based study of Australian-born women, assessed annually for 8 years, as they progressed through the menopausal transition. The purpose of this study was not to establish prevalence of sexual dysfunction, but to assess factors that affect women’s sexual functioning and how these change over time. Another longitudinal study that, while principally focused on issues on mental health, assessed sexual problems is the Zurich cohort study.34 A sample of men and women were interviewed on four occasions over a 10-year period, between the ages of 20 and 30. Women were more likely than men to report chronic or recurrent sexual difficulties, problems that were also more strongly associated with minor psychiatric or personality problems.

Longitudinal studies involving repeated assessment over a relatively short period (e.g., 2 years) are becoming increasingly feasible with the development of established survey panels (see below).

Obtaining a sample

Surveys that aim to establish prevalence of sexual problems and associated factors clearly need to be representative, if any gener alization to the general population is to be made. This presents a challenge. Currently, the only feasible way to generate a representative national sample is to employ one of the survey organizations that have the mechanisms for sampling in place.3,31 This is extremely costly. Whereas a few years ago, random digit dialing was the usual method for generating a sample, this is no longer considered acceptable as a result of problems with telephone surveys in general (discussed further below). Obtaining a sample of a known but limited population, such as general practitioner’s lists,35 is more feasible, but problems of generalizability and participation biases remain.

When selecting a representative sample, the specific aims of the project should be taken into consideration. Thus, if age is an important factor, the use of probability-based oversampling of otherwise underrepresented age groups will allow sufficient power to test the age-related hypotheses. Other examples of groups that, if a focus of the study, may require oversampling are gay men and lesbian women, celibate individuals, and various ethnic groups.

For studies that focus on relationships between factors of possible causal relevance, rather than prevalence rates, less representative samples can be useful, provided that the limitations of the sample and the need for replication are kept in mind.

Methods of data collection

Surveys conventionally have used self-report questionnaires, interview methods, or a combination of the two. There are advantages and disadvantages with each of these methods. In face-to-face interviews, the interviewer can clarify the meaning of questions and probe for more detailed responses if neces- sary,3,32 but this also can result in greater variance in how questions are asked and meanings conveyed. The training of the interviewer is of crucial importance here, as are possible biases related to the gender, age, and ethnic background of the inter- viewer.49 A disadvantage of interviews, particularly for large- scale surveys, is that they are expensive and labor-intensive. Problems of underreporting of sexual problems due to concerns about stigmatization may also be more likely if interviews are used, particularly when conditions of privacy cannot be assured. In Laumann et al.’s survey,3 21% of the face-to-face interviews were conducted with a third person present (most often children or stepchildren, but in some cases spouses or sexual partners). Although this may not be a problem for many survey topics, asking questions about sensitive issues such as sex requires confidentiality if not anonymity.

Telephone surveys have been used in several of the large- scale acquired immune deficiency syndrome (AIDS) behavioral surveys, such as the survey carried out in France.50 In a direct comparison of telephone survey and face-to-face interviewing, although questions were more easily answered on the telephone, they were more likely to be influenced by social desirability than with the interview method.51 Although telephone surveys are less expensive than interviews, it has become increasingly difficult to recruit participants by telephone.

The use of various kinds of computer-assisted interviews in surveys has increased. There are a number of advantage of these methods: use of “branching” and “skip” questions, checks for inconsistencies, and automatic data entry and “time stamping”, as well as the increased confidentiality and anonymity that they provide respondents. Studies have found that computerized methods elicit more accurate reports of “sensitive” behaviors such as same-sex sexual activity.52 If a method such as audiocomputer-assisted self-interview (A-CASI) is used in a respondent’s home, an interviewer might be present to answer questions about the survey, although this adds to the expense. Such methods have been incorporated into recent telephone surveys, either with the interviewer entering the participant’s responses directly into a computer (computer-assisted telephone interview [CATI]), or with respondents entering responses directly into the computer with touch-tone telephones (telephone-audio-computer-assisted self-interview (T-ACASI).Such methods do not, however, solve the participation problems with telephone surveys.

A newly emerging method, which has not as yet been used to collect sexuality-related information, but is being increasingly used for health-related surveys, and which, in our view, has considerable potential, is the use of Internet panels. An example of this is the Knowledge Networks Panel,53 which involves approximately 40 000 individuals aged over 18. In return for receiving cable connections and computer facilities, an individual agrees to complete a short survey every few weeks. This is a fairly representative sample demographically, although the representation is limited by the original use of random digit dialing for establishing the panel. There are a number of distinct advantages to this approach. First, all of the advantages of computerized interviewing are present. A substantial amount of information about the panelists is already available, which not only can be used to augment the data collected in a specific survey, but also allows better description of those who decline to participate. A large number of individuals can be screened before selecting probability-based subsamples, and a substantial number of questions can be asked by using a series of short surveys. Second, repeated surveying allows assessment of stability of responses and change over time. In a recent comparative study of this Internet approach and telephone interviewing, Chang and Krosnik (unpublished data) found the Internet method to have some advantages in terms of data quality. It remains to be seen whether this approach is effective in collecting sexuality-related data.

Survey questions

Surveys assessing the prevalence of factors relevant to sexual dysfunction need careful consideration of issues such as question wording and comprehension. Items should be as clear and specific as possible.54 Pretesting of items, either with individuals or focus groups, should be utilized,55 and checked in pilot surveys where subjects can be asked, after completing the survey, how they interpreted the questions.

Assessing the frequency of sexual activity or sexual problems raises important methodological issues, mainly because of recall error; accuracy of retrospective recall of sexual behavior declines significantly with longer recall periods.56,57 At the same time, assessing the duration of a problem is clearly important, and often requires reporting over relatively long time periods. This is where the “windows” concept, described earlier, can be useful. Assessment of a short and recent time period6,27,35 allows reasonably accurate assessment of the current situation (i.e., through the first “window”). Care should be taken, however, with how such frequency is recorded. With a short time interval, such as a month, the number of occasions of sexual activity can be realistically recorded, followed by the proportion of those occasions in which a particular response (e.g., orgasm, vaginal lubrication) did not occur.6

In addition to the recent time period, it is clearly important to establish how long a particular pattern has been evident, partly to enable a more comprehensive assessment through the “first window” (e.g., did this altered pattern precede or follow some other factor of potential causal relevance?), but also to address the questions through the “second window” (e.g., Is this a lifelong pattern? Have their been periods of “normality”?). A combination of a relatively precise frequency for a recent time period, and a more realistic estimate of duration or intermittence is therefore appropriate.

Relatively little systematic attention has been paid to assessing duration of a problem or pattern of behavior. In the study by Mercer and colleagues,31 participants were asked: “In the last year ... have you experienced any of the following for one month or longer?” This was followed by the same list of seven “problems” used in Laumann et al.’s study.3 For any item that was answered “yes”, respondents were asked: “For how long did that period last when you.?” (this sequence of questioning was facilitated by computer-assisted self-interview). There were four response options: at least 1 month, but less than 3 months; at least 3 months, but less than 6 months; at least 6 months, but less than 1 year; and 1 year or longer. In contrast, Laumann et al.3 simply asked whether the participant had experienced each of the seven difficulties “for several months or longer” during the previous year, with the response options “yes” or “no”.

Ernst and colleagues34 showed the benefits of a longitudinal study, with their finding that women, more often than men, reported relatively transient “disturbances” of sexual function that were nevertheless recurring. This pointed not only to circumstantial factors inducing the disturbance (i.e., “first window’), but also a vulnerability to such factors (i.e., “second window”).

The use of standardized or psychometrically validated measures in community-based surveys has been infrequent. A small number of studies have used structured interviews, such as the Diagnostic Interview Schedule (DIS), to estimate the prevalence of psychosexual dysfunction.26 In some cases, it has been stated that such measures permitted clinically accurate identification of the DSM or ICD diagnostic categories,35,58 although such claims do not bear close examination.

Recently, there has been much emphasis on the need for “validated” measures of sexual problems, particularly for use in clinical trials of female sexual dysfunction.25 A number of brief, self-report measures have been developed, such as the Female Sexual Function Index (FSFI)59 and the Sexual Function Questionnaire (SFQ).60 However, most of these measures were developed for use with clinical populations and to assess outcome after therapy, that is, change in sexual functioning, rather than the presence or absence of sexual dysfunction.61 Few measures have been validated by comparison with other questionnaires; most have established “construct validity” only by differentiating between clinical and nonclinical populations. Development of questionnaires explicitly designed to assess sexual dysfunction in community-based surveys is required.

Exploring causal relationships

Which associated factors should be assessed will vary with the hypotheses being tested. Increasingly, surveys are showing an important relation between mood and sexual functioning, particularly in women, and some appropriate measure of mood and well-being, such as the Short-Form Health Survey (SF12),62 should be considered. The more subjective features of the woman’s sexual relationship with her partner are also emerging as important, and warrant direct and well-considered inquiry. Issues relevant to the “third window”, such as current physical health or use of medication, together with duration, also require attention.

In evaluating the relevance of the woman’s previous history to her current sexual functioning, earlier traumatic sexual experiences, including child sexual abuse, should be considered. The age of onset of sexual activity, including masturbation, may be informative. Evidence suggests considerable individual variability in women for when they first become interested and responsive sexually.63,64 It remains to be determined whether women with later (or earlier) onset of sexual responsiveness are more vulnerable to later sexual problems.


Previous reviews in this area have stressed the need for standard definitions of sexual dysfunction65 and the use of a common classification system.26 At this stage, we would recommend something different. Firstly, before adopting standard definitions, it seems that considerable progress needs to be made in understanding the nature of women’s sexual problems. Qualitative approaches could be particularly valuable in providing insights into factors that influence distress about a sexual problem.66 While a common classification system would be extremely useful, and enable comparison across studies, if the typology of sexual problems does not reflect the reality of women’s experiences, “standardization” across studies does not, in our view, equal improvement. To date, the rationale for revising the DSM diagnostic categories of sexual dysfunction has not been made clear, resulting in changes that appear to be arbitrary. It has even been suggested that we should preserve the current DSM diagnostic categories of sexual dysfunction to maintain “continuity in research and clinical practice”,25 but this could effectively turn our attention away from important research questions.

In this chapter we have argued that much of the information required to arrive at a clinical diagnosis of sexual dysfunction is difficult if not impossible to obtain in large, community-based surveys. However, establishing the prevalence of problematic patterns of sexual response in women and their association with factors of potential causal relevance is clearly needed. Of crucial importance is the distinction between inhibition of sexual interest or response that is “adaptive” given the current circumstances, and inhibition or lack of responsiveness, which is “dysfunctional”, a distinction which is probably more crucial for women than for men.


1. Segraves RT. Editor’s comments. JSzMOiMaLThei 2001; 27: 81.

2. Tiefer L. Critique of the DMS-III-R nosology of sexual dysfunctions. Psychiatr Med 1992; 10: 227-45.

3. Laumann EO, Paik A, Rosen RC. Sexual dysfunctions in the United States: prevalence and predictors. JAMA 1999; 281: 537-44.

4. Foley S, Kope SA, Sugrue DP. Sex Matters for Women: A Complete Guide to Taking Care of Your Sexual Self. New York: Guilford Press, 2002.

5. Leiblum S, Sachs J. Getting the Sex You Want: A Woman’s Guide to Becoming Proud, Passionate, and Pleased in Bed. New York: Crown Publishers, 2002.

6. Bancroft J, Loftus J, Long JS. Distress about sex: a national survey of women in heterosexual relationships. Arch Sex Behav 2003; 32: 193-208.

7. Bancroft J, Loftus J, Long JS. Reply to Rosen and Laumann (2003). Arch Sex Behav 2003; 32: 213-16.

8. Shader RI. Sexual dysfunction associated with thioridazine hydrochloride. JAMA 1964; 188: 1007-9.

9. May RE. Sexual dysfunction following rectal excision for ulcerative colitis. BrlSug 1966; 53: 29-30.

10. Bernstein WC, Bernstein EF. Sexual dysfunction following radical surgery for cancer of the rectum. Dis Colon Rectum 1966; 9: 328-32.

11. Masters WH, Johnson VE. Human Sexual Response. Boston: Little, Brown, 1966.

12. Masters WH, Johnson VE. Human Sexual Inadequacy. Boston: Little, Brown, 1970.

13. Kaplan HS. The New Sex Therapy. New York: Brunner/Mazel, 1974.

14. American Psychiatric Association. Diagnostic and Statistical Manual of Mental Disorders (DSM-I). Washington, DC: 1952.

15. American Psychiatric Association. Diagnostic and Statistical Manual of Mental Disorders, 2nd edn (DSM-II). Washington, DC: 1968.

16. American Psychiatric Association. Diagnostic and Statistical Manual of Mental Disorders, 3rd edn (DSM-III). Washington, DC: 1980.

17. American Psychiatric Association. Diagnostic and Statistical Manual of Mental Disorders, 4th edn (DSM-IV). Washington, DC: 2000.

18. World Health Organization. ICD-10: International Statistical Classification of Diseases and Related Health Problems, 10th edn. Geneva: 1992.

19. Tiefer L. Historical, scientific, clinical and feminist criticisms of “The human sexual response cycle”. Annu Rev Sex Res 1999; 2: 1-23.

20. Basson R. The female sexual response: a different model. J Sex Marital Ther 2000; 26: 51-64.

21. Tiefer L. Arriving at a “new view” of women’s sexual problems: background, theory, and activism. In E Kaschak, E Tiefer, eds. A New View of Women’s Sexual Problems. Binghamton: Haworth Press, 2001: 63-98.

22. Basson R, McInnes R, Smith MD et al. Efficacy and safety of sildenafil citrate in women with sexual dysfunction associated with female sexual arousal disorder. J Womens Health 2002; 11: 367-77.

23. Harris G. Pfizer gives up testing Viagra on women. New York Times 2004; Feb 28.

24. American Psychiatric Association. Diagnostic and Statistical Manual of Mental Disorders, 3rd edn rev. (DSM-III-R). Washington, DC: 1987.

25. Basson R, Berman J, Burnett A et al. Report of the International Consensus Development Conference on Female Sexual Dysfunction: definitions and classification. J Urol 2000; 163: 888-93.

26. Simons JS, Carey MP. Prevalence of sexual dysfunctions: results from a decade of research. ArchSexB^tav 2001; 30: 177-219.

27. Dunn KM, Croft PR, Hackett GI. Sexual problems: a study of the prevalence and need for health care in the general population. Fam Pract 1998; 15: 519-24.

28. Dunn KM, Croft PR, Hackett GI. Association of sexual problems with social, psychological, and physical problems in men and women: a cross sectional population survey. J Epidemiol Community Health 1999; 53: 144-8.

29. Fugl-Meyer AR, Fugl-Meyer K. Sexual disabilities, problems and satisfaction in 18-74 year old Swedes. Scand J Sexol 1999; 2: 79-105.

30. Osborn M, Hawton K, Gath D. Sexual dysfunction among middle- aged women in the community. BMJ 1988; 296: 959-62.

31. Mercer CH, Fenton KA, Johnson AM et al. Sexual function problems and help seeking behaviour in Britain: national probability sample survey. BMJ 2003; 327: 426-7.

32. Johnson AM, Mercer CII, Erens B et al. Sexual behaviour in Britain: partnerships, practices, and HIV risk behaviours. Lancet 2001; 358: 1835-42.

33. Erens B, McManus S, Field J et al. National survey of sexual attitudes and lifestyles II: technical report. London: National Centre for Social Research, 2001.

34. Ernst C, Foldényi M, Angst J. The Zurich study: XXI. Sexual dysfunctions and disturbances in young adults. Eur Arch Psychiatry Clin Neurosci 1993; 243: 179-88.

35. Nazareth I, Boynton P, King M. Problems with sexual function in people attending London general practitioners: cross sectional study. BMJ 2003; 327: 423-9.

36. Nathan SG. The epidemiology of the DSM-III psychosexual dysfunctions. JSxMaialTh&L 1986; 12: 267-81.

37. Moynihan R. The making of a disease: female sexual dysfunction. BMJ 2003; 326: 45-7.

38. Ellison C. Women’s Sexualities. Oakland: New Harbinger, 2000.

39. Dennerstein L, Lehert P, Burger H et al. Factors affecting sexual functioning of women in the midlife years. Climacteric 1999; 2: 254-62.

40. Bancroft J. Central inhibition of sexual response in the male: a theoretical perspective. NeurociBobhavRv 1999; 23: 763-84.

41. Bancroft J, Janssen E. The dual control model of male sexual response: a theoretical approach to centrally mediated erectile dysfunction. Neurosci Biobehav Rev 2000; 24: 571-9.

42. Janssen E, Vorst H, Finn P et al. The Sexual Inhibition (SIS) and Sexual Excitation (SES) Scales. I. Measuring sexual inhibition and excitation proneness in men. JSxMes 2002; 39: 114-26.

43. Bancroft J, Janssen E, Strong D et al. Sexual risk taking in young heterosexual men. The relevance of personality factors. J Sex Res 2004; 41: 181-92.

44. Bancroft J, Janssen E, Strong D et al. Sexual risk taking in gay men: the relevance of sexual arousability, mood and sensation seeking. ArchSexBehav 2003; 32: 555-73.

45. Graham CA, Sanders SA, Milhausen RR et al. Turning on and turning off: a focus group study of the factors that affect women’s sexual arousal. ArchSexBhav 2004; 33: 527-38.

46. Baumeister RF, Catanese KR, Vohs KD. Is there a gender difference in strength of sex drive? Theoretical views, conceptual distinctions, and a review of relevant evidence. Pers Soc Psychol Rev 2001; 5: 242-73.

47. Garde K, Lunde I. Female sexual behaviour. A study in a random sample of 40-year-old women. Maturitas 1989; 2: 225-40.

48. Heiman JR. Sexual desire in human relationships. In W Everaerd, E Laan, S Both, eds. Sexual Appetite, Desire and Motivation: Energetics of the Sexual System. Amsterdam: Royal Netherlands Academy, 2001: 117-34.

49. Catania JA, Binson D, Canchola J et al. Effects of interviewer gender, interviewer choice, and item wording on responses to questions concerning sexual behavior. Public Opin Q 1996; 30: 345-75.

50. Spira A, Bajos N, the ACSF Group. Sexual Behaviour and AIDS. Aldershot: Avebury, 1994.

51. ACSF principal investigators and their associates. Analysis of sexual behaviour in France (ACSF). A comparison between two modes of investigation: telephone survey and face-to-face survey. AIDS 1992; 6: 315-23.

52. Turner CF, Miller HG, Rogers SM. Survey measurement of sexual 'margin-left:18.0pt;text-indent:-18.0pt'>53. Knowledge Networks, Inc. 1350 Willow Rd., Menlo Park, CA 94025 [www.knowledgenetworks.com\ganp].

54. Fenton KA, Johnson AM, McManus S et al. Measuring sexual behaviour: methodological challenges in survey research. Sex Transm Infect 2001; 77: 84-92.

55. Catania JA, Turner H, Pierce RC et al. Response bias in surveys of AIDS-related sexual behavior. In DG Ostrow, RC Kessler, eds. Methodological Issues in AIDS Behavioral Research. New York: Plenum Press, 1993: 133-62.

56. Catania JA, Gibson DR, Chitwood DD et al. Methodological problems in AIDS behavioral research: influences on measurement error and participation bias in studies of sexual behavior. Psychol Bull 1990; 108: 339-62.

57. Graham, CA. Catania JA, Brand R et al. Recalling sexual 40: 325-32.

58. Lindal E, Stefansson JG. The lifetime prevalence of psychosexual dysfunction among 55- to 57-year-olds in Iceland. Soc Psychiatry Psychiatr Epidemiol 1993; 28: 91-5.

59. Rosen R, Brown C, Heiman J et al. The female sexual function index (FSFI): a multidimensional self-report instrument for the assessment of female sexual function. J'SexMarMaLLheZ 2000; 26: 191-208.

60. Quirk FH, Heiman JR, Rosen RC et al. Development of a sexual function questionnaire for clinical trials of female sexual dysfunction. J Womens Health 2002; 11: 277-89.

61. Daker-White G. Reliable and valid self-report outcome measures in sexual (dys)function: a systematic review. Arch Sex Behav 2002; 31: 197-209.

62. Ware JE, Kosinski M, Keller S. A 12-item Short-Form Health Survey (SF-12): construction of scales and preliminary tests of reliability and validity. Med Care 1996; 34: 220-33.

63. Reynolds MA, Herbenick DL, Bancroft J. The nature of childhood sexual experiences: two studies 50 years apart. In J Bancroft, ed. Sexual Development in Childhood. Bloomington: Indiana University Press, 2003.

64. Bancroft J, Herbenick D, Reynolds M. Masturbation as a marker of sexual development. In J Bancroft, ed. Sexual Development in Childhood. Bloomington: Indiana University Press, 2003.

65. Dunn KM, Jordan K, Croft PR et al. Systematic review of sexual problems: epidemiology and methodology. J Sex Marital Ther 2002; 28: 399-422.

66. Tiefer, L, Rosen R, Giami A et al. Qualitative health research and sexual dysfunction. In TF Lue, R Basson, R Rosen et al., eds. International Consultation in Sexual Medicine. Paris: 2004: 161-70.