The Diagnostic and Statistical Manual of Mental Disorders, Fourth Edition, Text Revision as an Appropriate Diagnostic for Premature Ejaculation

1468

The Diagnostic and Statistical Manual of Mental Disorders, Fourth Edition, Text Revision as an Appropriate Diagnostic for Premature Ejaculation Ridwan Shabsigh, MD,* and David Rowland, PhD† *Maimonides Medical Center, New York, NY, USA; †Valparaiso University, Valparaiso, IN, USA DOI: 10.1111/j.1743-6109.2007.00557.x

ABSTRACT

Introduction. The Diagnostic and Statistical Manual of Mental Disorders, fourth edition, text revision (DSM-IV-TR) criteria for premature ejaculation (PE) are multifactorial, and include concepts of latency, perceived control over ejaculation, and personal distress and interpersonal difficulty related to the condition. Recent publications have suggested that these criteria are not supported by empirical evidence, leading to the proposal that a PE diagnosis should be based solely on intravaginal ejaculatory latency time (IELT), while the presence of other factors (such as perceived lack of control over ejaculation) may be used to guide treatment decisions. Aim. To examine the evidence supporting the elements of the DSM-IV-TR criteria for PE. Main Outcome Measures. Literature searches on IELT, perceived control over ejaculation, and personal distress and interpersonal difficulty related to ejaculation. Results. From a historical perspective, there has been a lack of large observational studies that evaluated the contributions of the DSM-IV-TR components in men with PE. However, recently performed large observational studies have generated data supporting the inclusion of perceived control over ejaculation and personal distress related to ejaculation in the definition of PE. Furthermore, emerging evidence indicates that a perceived lack of control over ejaculation is directly associated with elevated personal distress related to ejaculation and decreased satisfaction with sexual intercourse, while the effects of IELT on these parameters are indirect, and mediated by perceived control over ejaculation. A key advantage of the DSM-IV-TR approach to the diagnosis of PE is that it firmly links PE to a negative outcome for the patient, which is an element common to diagnostic criteria for other conditions, including depression, hypertension, and osteoporosis. Conclusions. This new evidence strongly suggests that the DSM-IV-TR criteria for PE encompass aspects of the condition that patients describe as important. Shabsigh R, and Rowland D. The DSM-IV-TR as an appropriate diagnostic for premature ejaculation. J Sex Med 2007;4:1468–1478. Key Words. Premature Ejaculation; DSM-IV-TR; Diagnosis; Control over Ejaculation

Introduction

A

lthough premature ejaculation (PE) is among the most common sexual dysfunctions in men (21–25%) [1–3], few men with PE (9% or less) seek treatment from a health care professional (HCP) for the condition [1,4], and HCPs do not routinely inquire about symptoms of PE in their patients [4–6]. No approved medication is indicated for the treatment of PE, and the long-term efficacy of behavioral, cognitive, and sex therapy J Sex Med 2007;4:1468–1478

approaches is as yet undetermined [7]. It is perhaps, then, not surprising that the accurate diagnosis of PE has received little attention until recently, as new medications designed specifically for the treatment of PE are being tested in clinical trials [8]. For many years, practitioners have diagnosed PE using the criteria of the Diagnostic and Statistical Manual of Mental Disorders, fourth edition, text revision (DSM-IV-TR) and its predecessors [9], and other professional societies have proposed © 2007 International Society for Sexual Medicine

1469

Defining PE with the DSM-IV-TR Table 1

Definitions of premature ejaculation

Source

Definition

DSM-IV-TR [9]

1. “Persistent or recurrent ejaculation with minimal sexual stimulation before, on, or shortly after penetration and before the person wishes it” 2. Causes marked distress or interpersonal difficulty 3. Not due exclusively to the direct effects of a substance (e.g., withdrawal from opioids) IELT <1.0 or <1.5 minutes “Ejaculation that occurs sooner than desired, either before or shortly after penetration, causing distress to either one or both partners” “Persistent or recurrent ejaculation with minimal stimulation before, on, or shortly after penetration, and before the person wishes it, over which the sufferer has little or no voluntary control, which causes the sufferer and/or his partner bother or distress” Clinical description and diagnostic guidelines: “The inability to control ejaculation sufficiently for both partners to enjoy sexual interaction . . . Minimum duration of symptoms of 6 months” Diagnostic criteria for research: 1. General criteria for sexual dysfunction, including being unable to participate in a sexual relationship as he would wish, frequent occurrence, present for ⱖ6 months, and not entirely attributable to another disorder or drug treatment 2. Inability to delay ejaculation sufficiently to enjoy lovemaking, either due to ejaculation before or very soon after the beginning of intercourse (if a time limit is required: ⱕ15 seconds), or in the absence of sufficient erection to make intercourse possible 3. Not the result of prolonged abstinence

Waldinger and Schweitzer [10,11] American Urological Association [12] 2nd International Consultation [13]

International Classification of Diseases, 10th revision [33]

DSM-IV-TR = Diagnostic and Statistical Manual of Mental Disorders, fourth edition, text revision; IELT = intravaginal ejaculatory latency time.

similar criteria (Table 1) [10–14]. The DSMIV-TR definition of PE represents a multifactorial approach to the diagnosis of PE, with special attention to clinical relevance, and may be interpreted to include three principal components: short ejaculatory latency, perceived lack of control over ejaculation, and a negative impact of the condition on the person or relationship. One criticism of the DSM-IV-TR is that it provides a set of consensus diagnostic criteria, generated by experts, and that its criteria for PE in particular are not evidence based. Historically, there has been a lack of large, appropriately designed clinical trials and observational studies in men with PE to provide a solid base of empirical evidence needed to support any set of diagnostic criteria for PE. This is gradually changing— in recent years, several important studies have explored the relevance of specific aspects of PE [15,16], and sexual medicine specialists are gaining a better understanding of the relationships among the domains that characterize the condition. The potentially confusing status of the literature and the diversity of clinical practice and research have generated a significant need to have an article that brings together a comprehensive description and a critical analysis of the various criteria for PE. Our article describes the empirical evidence supporting a multifactorial approach to the diagnosis of PE and evaluates the appropriateness of the DSM-IV-TR criteria for PE as a clinical diagnostic tool.

Evidence Supporting the Components of the DSM-IV-TR

The diagnosis of PE using the DSM-IV-TR criteria is based on the presentation of the patient with specific symptoms or elements of the condition, similar to the diagnosis of migraine and depression. The current debate regarding the appropriate diagnosis of PE exists because there is no universally agreed-upon symptom or set of symptoms that objectively defines the condition. Rather, experts have evaluated the evidence in the literature, and have made recommendations as to which factors are important or relevant. Some have come to the conclusion that PE is best described as a multifactorial condition that is linked to a negative outcome for the patient [9–15,17,18], while others have recently proposed that PE be defined solely by a short intravaginal ejaculatory latency time (IELT) relative to the overall population, with other dimensions used to guide treatment decisions [10,11,19].

Evidence for the Role of Latency Time Latency time is commonly assessed in clinical trials using IELT, defined as the time from vaginal penetration to the start of intravaginal ejaculation [20]. A majority of men with PE have reported IELTs that are shorter than those observed in the general population, regardless of the diagnostic approach used. For example, an observational study of men with self-reported lifelong PE recruited through newspaper articles about the J Sex Med 2007;4:1468–1478

1470 psychopharmacological treatment of PE (n = 110) found that 80% of these men had a stopwatchmeasured IELT of <30 seconds, and 90% had an IELT of <60 seconds [20]. Results from the Premature Ejaculation Prevalence and Attitudes (PEPA) survey, a large, web-based, survey of men in the United States, Germany, and Italy demonstrated that among men with symptoms of PE (i.e., lack of control over ejaculation and a latency time that was a problem for them or their partner, n = 2754), 58.2% also reported an estimated latency time of <5 minutes [1]. Another study reported that 87% of men with PE diagnosed using the DSM-IV-TR criteria estimated their IELT to be <2 minutes (vs. 22% of men without PE, P < 0.001) [21]. Men with PE recognize that their IELT is shorter than that of the “average” man. In the PEPA survey, a majority of men with PE reported an estimated IELT <5 minutes, and estimated that the “average” man lasts for 8.9 minutes (mean of responses) [1]. Interestingly, this perceived “average” value is similar to the stopwatchmeasured IELT of men without PE in a U.S. observational study (mean IELT 9.2 minutes) [15]; both of these values are somewhat higher than that reported for generally healthy men in a study conducted in five European countries (median IELT, 5.4 minutes) [22]. Men with self-reported symptoms of PE in the PEPA survey also reported that a man should “ideally” last for 15.8 minutes (mean of responses). Together, these results suggest that men with PE believe that they should last longer, based on a comparison to real or perceived societal or personal norms, or partner expectations. Lastly, IELT has been demonstrated to be an integral part of statistical models describing PE—among men (n = 189) diagnosed with PE by clinicians based on the DSM-IV-TR criteria in a large, U.S. observational study, models describing the relationships between PE-specific dimensions did not represent the condition accurately if IELT was not included [17].

Limitations of Latency Time In both concept and application, IELT has significant shortcomings; therefore, it should not be construed as a “gold standard” for diagnosing PE and assessing its treatment. Conceptual Limitations: Latency Time and “Minimal Stimulation” The DSM-IV-TR and the 2nd International Consultation definitions of PE have put forth the J Sex Med 2007;4:1468–1478

Shabsigh and Rowland construct of minimal stimulation (Table 1). Both latency time and IELT are imprecise and unvalidated proxy measures for sexual stimulation—time does not cause ejaculation; it is the amount and degree of sexual stimulation over time, coupled with subjective feelings of arousal, that result in the biobehavioral response of ejaculation [23,24]. IELT is relevant only insofar as it assumes a certain amount of penile stimulation (or thrusting). These conceptual arguments are supported by recent studies demonstrating that while stopwatchmeasured IELT and assessments of penile sensitivity are highly repeatable in each subject, the two assessments are not correlated (r 2 < 0.08) [25]. The inadequacy of IELT in the assessment of PE is best illustrated through three examples. The first involves men with anteportal ejaculation, for whom there is no IELT, because ejaculation occurs without penetration; up to 49% of men with PE have reported anteportal ejaculation to occur at least “sometimes” [20]. In such men, not only does IELT have no heuristic value, but the primary role of stimulation (whether intentional or not on the part of the man and his partner) over latency time in assessing PE becomes obvious: unlike IELT, stimulation may occur within or outside the vagina. Second, the IELT distribution of men with PE diagnosed using the DSM-IV-TR shows significant overlap with that of men not meeting the DSM-IV-TR criteria for PE [15]. Some have suggested that this overlap is evidence for the inadequacy of the DSM-IV-TR criteria [10,11]; in fact, it identifies a serious shortcoming of IELT. Clinical experience indicates that many men with PE postpone their ejaculation as long as possible by minimizing their thrusting during intercourse. Measures such as IELT, which assume a constant and continuous rate of thrusting, fail to capture such “quiescent” periods. Another conceptual argument for the utility of “stimulation” measures over “time latency” measures is found in the animal literature [26,27], where the standard measure for ejaculatory latency is not time, but rather the number of intromissions to ejaculation, where each intromission is considered one thrust. Indeed, in rat studies on sexual and ejaculatory response, intromission latency, ejaculatory latency, mount frequency, and intromission frequency may all affect ejaculatory behavior [27].

Assessment Modalities It is not clear which assessment of latency time most accurately reflects this domain. In clinical trials,

Defining PE with the DSM-IV-TR IELT has been self-estimated, partner-estimated, and measured via stopwatch. In one study of men presenting with a complaint of PE (n = 110), the mean stopwatch-measured IELT was 28 seconds; when these men were asked to estimate their IELT using several different response methods (including written and oral, with and without a timer or scale), the mean estimated IELT among the various approaches ranged from 16 to 41 seconds [20]. All of the methods of estimating IELT correlated moderately with the stopwatch measurement (range r 2 = 0.50–0.59); estimated IELT reported either orally or using a list of values was higher than the stopwatch-measured value. Yet, results from another study of men with and without PE revealed that self-estimated and stopwatch-measured IELT were interchangeable: IELT assessed using either approach predicted a DSM-IV-TR diagnosis of PE with 80% sensitivity and 80% specificity [28]. In that study, no systematic bias (i.e., either over- or underestimation) was observed [28].

Use of the Stopwatch Use of the stopwatch may constitute a disruption of coitus that could affect IELT, because the measurement of IELT imposes an artificial coital regimen upon the couple by requiring a single penetration to ejaculation. The extent to which this “IELT-constrained coitus” actually simulates a typical encounter for most couples has not been explored. More importantly, this unnatural, regimented, and uniform approach to assessing the couple’s sexual response, in which their behavior may have to be altered to fit the demands of the measurement variable, is likely to affect arousal, ejaculatory control, and the man’s typical ejaculatory latency. In one study, 21% of men enrolled “could not or were not willing to” use the stopwatch, but no detailed explanations were given [20]. Interestingly, these men reported significantly lower control over ejaculation during foreplay than men who agreed to use the stopwatch (P < 0.001). It would seem logical to conclude that only a particular kind of patient in a particular kind of relationship would agree to measure and record his IELT every time he has intercourse. How this potential bias in patient selection may influence study results is not known. The previous discussion illustrates problems with the conceptualization of IELT as a defining measure of PE, and that the proposed use of IELT as the sole pathognomic dimension for defining PE is not well established as a valid index of minimal stimulation or of PE in general.

1471

Sexual Stimulation and its Relationship With IELT Assessment of sexual stimulation is a significant challenge that has yet to be addressed satisfactorily in clinical trials. The number of thrusts is virtually the only assessment that attempts to capture sexual stimulation, but optimal strategies for operationalizing and measuring this construct have received only minimal attention. The fact that IELT is only weakly correlated (r 2 = 0.28) with the number of thrusts underscores the problem of using IELT as a proxy for sexual stimulation [20]. Furthermore, assessments of penile sensitivity are not correlated with IELT [25]. Equally important, no studies evaluating men with PE or treatments for PE have assessed the total duration of all forms of stimulation and concomitant sexual arousal and excitement, including those beginning before penetration. Although it is possible to standardize the level of sexual stimulation in a controlled, laboratory setting, using stimulation devices [29–32], it is very difficult to assess the level of sexual stimulation or the subject’s subjective experience of that stimulation in a sexual encounter. Thus, of all of the various criteria assessed in men with PE, “minimal stimulation” has received the least amount of investigation; however, it is arguably one of the most important components, as it firmly roots the definition in the context of the physiology of ejaculation. Although measures of sexual stimulation have greater alignment with the DSM-IV-TR than do measures of time, until more expedient methods are developed to measure penile stroking and/or thrusting, IELT will undoubtedly continue to be the measure of choice within clinical trials, as it is easy both to measure and interpret. However, this variable must be applied in ways that contribute to, rather than sharply define, the classification of men into PE and non-PE categories. Researchers and clinicians would do well to apply this criterion gingerly, so as to minimize errors of either omission and/or of inclusion when diagnosing and assessing PE. Evidence for the Role of Perceived Control over Ejaculation Perceived control over ejaculation may generally be viewed as a measure of self-efficacy—the degree to which a person feels that he has the ability to affect an outcome, in this case, the timing of ejaculation. Perceived control over ejaculation is usually assessed in clinical trials using a Likert-type question or series of questions [15,20,33]. In one study of men presenting a complaint of PE (n = 110), J Sex Med 2007;4:1468–1478

1472 98% reported that their feeling of control over ejaculation during intercourse was “none” or “little”; none of these men reported “almost full” or “full” control over ejaculation [20]. Among men diagnosed with PE by clinicians (n = 207) using the DSM-IV-TR criteria, 72% reported “very poor” or “poor” control over ejaculation (vs. 5% of men without PE, n = 1380, P < 0.001) [15]. Across the entire study population, perceived control over ejaculation was correlated moderately (r 2 = 0.51) with stopwatch-measured IELT. Similar results were reported in a recent observational study conducted in five European countries [34]. New analyses, such as those arising from the aforementioned U.S. observational study, are crucial to expanding our understanding of PE; however, the validity of that study has been questioned recently [10]. Therefore, we would like to clarify the following methodological points. In that study, men were diagnosed with PE by clinicians who were either trained or experienced in using the DSM-IV-TR for PE diagnosis. Importantly, physicians were not provided with questionnaires, response scales, or stopwatch-measured IELT to guide their diagnosis. At baseline and after 2 and 4 weeks, patients provided event logs containing stopwatch-measured IELT for each intercourse event and responded to four singleitem measures, which assessed perceived control over ejaculation, satisfaction with sexual intercourse (5-point scales, from “very poor” to “very good”), and personal distress and interpersonal difficulty related to ejaculation (5-point scales, from “not at all” to “extreme”). IELT and responses to the single-item measures were then compared between men with and without PE, classified based on their physician’s diagnosis. Therefore, results from this study can provide important insights into the relationship between specific PE parameters and a clinician diagnosis of PE. The importance of perceived control over ejaculation in the characterization of PE was highlighted in a recent analysis of data from that U.S. observational study, which demonstrated that perceived control over ejaculation and personal distress related to ejaculation were the two most important explanatory variables in every model predicting PE status in which they were included; in fact, distress related to ejaculation reported by the female partner was more predictive of a DSMIV-TR diagnosis of PE than either estimated or measured IELT in every model in which it was considered [28]. Additional analyses have demonstrated that all of these domains were correlated J Sex Med 2007;4:1468–1478

Shabsigh and Rowland significantly with each other (P ⱕ 0.05), with the exception of IELT and interpersonal difficulty related to ejaculation [17]. The highest correlation coefficients were observed between perceived control over ejaculation and personal distress related to ejaculation (r 2 = 0.60) and perceived control over ejaculation and satisfaction with sexual intercourse (r 2 = 0.58) [17].

Relationship Between Perceived Control over Ejaculation and IELT One new investigation has explored the relationships between PE-specific variables through a path analysis, which is a specific type of regression analysis that allows the determination of direct and indirect effects among variables in a model [17]. Perceived control over ejaculation had a significant (P < 0.05) direct effect on both satisfaction with sexual intercourse and personal distress related to ejaculation. The only significant direct effect of IELT was on perceived control over ejaculation. Satisfaction with sexual intercourse and personal distress related to ejaculation each in turn had a significant direct effect on interpersonal difficulty related to ejaculation (Figure 1). These results suggest that it is a patient’s report of low or absent control over ejaculation that is directly associated with his distress and low sexual satisfaction, not his IELT alone. Furthermore, perceived control over ejaculation mediates the effect of IELT on the other variables that describe more clinically relevant parameters characterizing PE. A logical extension of these conclusions is that for the negative impact of PE to improve following treatment, a man with PE must perceive that his control over ejaculation has improved. Evidence for the Role of Distress and/or Interpersonal Difficulty All disorders defined in the DSM-IV-TR assume some type of functional impairment, whether interpersonal, social, psychological, behavioral, or mental, which is expressed as distress and/or interpersonal difficulty in its definition of PE. It has been suggested that inclusion of this domain in the definition of PE is arbitrary, and not supported by empirical evidence. The requirement for distress and/or interpersonal difficulty is not unique to the DSM-IV-TR characterization of PE—all consensus definitions of PE include some component of negative influence on the man’s life, whether described as distress, bother, upset, frustration, interpersonal difficulty, relationship difficulty, decreased enjoyment of lovemaking,

Defining PE with the DSM-IV-TR

1473

Figure 1 The relationships among components of the DSM-IV-TR definition of PE. Results from a path analysis of data from an observational study of men with PE demonstrated that IELT only exerts a significant direct effect on perceived control over ejaculation. Perceived control over ejaculation has direct effects on satisfaction with sexual intercourse and personal distress related to ejaculation, which in turn, have a direct effect on interpersonal difficulty related to ejaculation. IELT = intravaginal ejaculatory latency time.

etc. (see Table 1) [9,12,13,35]. These dimensions firmly link the presence of specific symptoms to a negative outcome for the patient. Results from the U.S. observational study demonstrated that 88% of men with PE reported at least “moderate” personal distress related to ejaculation, and 56% reported at least “moderate” interpersonal difficulty related to ejaculation [15]. New analyses of results from this study, in which the classification criteria for men with PE were further restricted to include a clinician diagnosis using the DSM-IV-TR and an IELT ⱕ2 minutes (n = 89) [36], demonstrated that men with PE reported a significantly lower frequency of and satisfaction with intercourse, greater personal distress and interpersonal difficulty related to ejaculation, and lower self-esteem and confidence in their sexual and overall relationship with their partner. Furthermore, men with PE rated their overall health-related quality of life lower than men without PE. Results from other, smaller studies support these findings [5,21,37]. Furthermore, when assessed with standardized psychological instruments, men with PE typically view their problem as distressful as men with erectile dysfunction (ED) [36]. PE also affects men’s partners and relationships. Nearly two-thirds of men with PE have reported that they avoided discussing sexual problems with their partner [37], and one in two have reported distress about their relationship due to PE [5]. Single men not in relationships reported a reluctance to establish new relationships due to PE, while men with PE in relationships reported that they worry about an inability to satisfy their partner [5]. Other studies of men who selfidentified as having a “PE problem” described a negative impact on their personal functioning and

sexual relationship [36,38]. In comparison with men with no sexual dysfunction (n = 43), men with PE had significantly (P < 0.01) lower scores for intimacy, including total intimacy and each intimacy subscale assessed (emotional, social, sexual, recreational, and intellectual) [39]. While very few men with PE seek treatment for their condition, those who do have reported that the negative impact of PE on their lives and relationships is a key motivating factor to seek treatment. In the PEPA survey, only 9% of men with symptoms of PE reported that they had discussed their condition with a physician [1]. The majority of these men (75%) indicated that they had sought treatment to better satisfy their partner sexually; a moderate proportion of these men reported that they had become very depressed or upset (31%) or that their partner was concerned, upset, or angry, about their PE (21%). Furthermore, even in those studies in which the negative impact of PE was not specifically assessed (e.g., [19,20,40,41]), the complaint of PE by the man that drove him to participate in a clinical trial could be interpreted as an indication that he was bothered by his condition. Strengths and Weaknesses of the Multifactorial Approach and the Single-Factor Approach

The DSM-IV-TR (and other consensus definitions) does not provide a checklist of symptoms and precisely defined thresholds to confirm the diagnosis of PE; as a result, it is subject to interpretation error by the practitioner. In the search for a more quantitative diagnostic tool for PE, recent reports have proposed that IELT should be the only factor that defines PE status [10,11]. As previously noted, advocates of the IELT-only approach have proposed that the lower limit of the J Sex Med 2007;4:1468–1478

1474 95% and 99% confidence intervals for the IELT distribution of the overall population (approximately 1.5 and 1.0 minutes, respectively) should serve as threshold points for the classification of men with PE [10,11]. A similar approach has been used for other conditions, such as hypertension and bone mineral density. Advocates of the IELT-only approach propose that the severity of PE and treatment decisions could be guided by considering the subject’s perceived control over ejaculation and associated distress. The DSM-IV-TR definition does not conflict with this theory; rather, it indicates that although IELT should be considered, a short ejaculatory latency, on its own, is insufficient to establish a diagnosis of PE. This proposal for a new diagnostic approach to PE was based on two postulates: (i) that PE is a solely “biologic” condition, and (ii) that all biologic conditions can be assessed using measurable physical or tangible markers. First, describing PE (or any condition) as biologic has no useful meaning in the discussion of biobehavioral responses. All psychological processes (such as arousal, anxiety, and libido) are underlain by, and cannot occur independent of, biological events—that is, biochemical release and receptor activation at specific neuroanatomical sites that comprise larger physiological response systems [42]. In this context, “biologic” may have been intended to mean that lifelong PE is innate or genetically predetermined. However, a proportion of men with PE have reported that they acquired PE later in life [1], and the eventual development of PE in these men may also be genetically predetermined, in the same manner as male pattern baldness or myopia. Second, not all innate or genetically predetermined conditions can be assessed with physical or tangible markers; indeed, many conditions with known genetic or hereditary components (e.g., depression) cannot be assessed using a laboratory test, and rely on careful screening of patients by appropriately trained physicians exercising their clinical judgment. The combined use of physical and laboratory tests with psychological and/or cognitive assessments is common for a range of conditions, and this has long been the standard for assessing PE. These postulates arose from the theory that PE is not a dysfunction—there does not appear to be a pathophysiologic cause for PE, and all elements of the sexual response cycle are thought to function normally in these men, albeit at a faster rate [22,43,44]. Therefore, it has been proposed that J Sex Med 2007;4:1468–1478

Shabsigh and Rowland ejaculatory latency, like other biologic variables, exists as a spectrum, in which some men have very short latencies, most have moderate latencies, and some have very long latencies [22,43,44]. Of note, this theory is entirely compatible with the DSMIV-TR approach; the question remains as to whether a very short ejaculatory latency alone is sufficient to diagnose a man with PE. Diagnostic thresholds for other conditions were developed based on an increased probability of negative outcomes for the patient [45]; for example, osteoporosis is diagnosed using bone mineral density measurements, and the validity of statistically derived diagnostic thresholds is based on their association with an elevated risk for bone fracture [46]. Similarly, in the case of hypertension, the division between normal and high blood pressure is arbitrary, and was established to define patients at elevated risk for cardiovascular events and those who might benefit from medical therapy [47]. The Joint National Committee on Prevention, Detection, Evaluation, and Treatment of High Blood Pressure, which created the current blood pressure diagnostic categories for hypertension, noted that although the classification of blood pressure is somewhat arbitrary, clinicians may find it to be a useful tool [48]. Thus, a statistical variation from the norm has clinical importance when it is associated with a negative outcome. In contrast, the statistical relationship used to derive the IELT thresholds of 1.0 and 1.5 minutes was that between men with very short IELTs and the overall population, not between those with and without negative outcomes related to their ejaculatory latency.

Challenges with Implementing the IELT-Only Approach Any single-factor diagnostic approach relies heavily on the accuracy and precision of the measurement and its predictive value for an adverse outcome. IELT is a reliable measure of the duration of penetration [15], and estimated and stopwatch-measured IELT are correlated moderately in men with PE [28]; IELT does not assess the amount or level of sexual stimulation during intercourse, and does not account for stimulation that may occur prior to intercourse. Many feel that the DSM-IV-TR could be more precise, and that diagnostic thresholds of some kind would be highly desirable; however, IELT should not be considered a laboratory test for the diagnosis of PE. Many (if not most) laboratory diagnostics still require careful interpretation and the application of the clinician’s judgment in treatment decisions.

Defining PE with the DSM-IV-TR For example, not all patients with low bone mineral density will experience bone fracture; however, a bone mineral density that is 2.5 standard deviations less than the mean of a “young normal” population is associated with a level of risk for bone fracture that warrants treatment. Implementing this IELT-only approach in clinical practice could lead to men being diagnosed with PE who are completely satisfied with their latency time, which fails to recognize that the preference for a given duration of intercourse differs from person to person and from couple to couple—there are men who have a short IELT that is not a problem for them or their partner. In contrast, the DSM-IV-TR approach requires that if the latency time does not lead to any negative consequences for the man or his partner, then there is no basis for a diagnosis of PE. However, practitioners must exercise clinical judgment when applying the DSM-IV-TR criteria; for example, distress related to ejaculation may result from a partner with a long latency to orgasm, rather than the man’s rapid ejaculatory response.

Challenges with Implementing the DSM-IV-TR The DSM-IV-TR is not the perfect diagnostic for PE—there are many aspects that could be modified to improve its interpretability and applicability. The language of the DSM-IV-TR could be modified to clarify how the concepts currently included are meant to be used in practice. While it is unlikely that the DSM-IV-TR will endorse diagnostic thresholds, as none of its definitions employ this type of tool, some additional guidance as to the severity of short ejaculatory latency and lack of control over ejaculation may be appropriate. The definition should also eliminate aspects that are now known not to be directly associated with PE, such as age. The DSM-IV-TR also currently gives equal qualitative weight to its dimensions—it would appear that short latency time, lack of control over ejaculation, and distress related to ejaculation each carry the same importance. Furthermore, each of these dimensions may also be multifactorial. Similar to the unresolved relationship between IELT and sexual stimulation, perceived control over ejaculation may be influenced by several factors, including frequency of sexual activity, partner relationships, alcohol use, and drug use. Distress may have vast influences from physical and psychological factors other than PE. These challenges in the diagnosis of PE using the DSM-IV-TR are not unique; indeed, they are typical of the DSM-IV-TR approach to the diag-

1475 nosis of biobehavioral conditions. For example, stuttering is defined in the DSM-IV-TR as frequent occurrences of two or more specific speech patterns (a list of eight is provided) that interfere with “academic or occupational achievement or with social communication” and are not due to another deficit/dysfunction [9]. Similar to the definition of PE, this definition of stuttering does not specify a frequency threshold, and requires a negative consequence to the patient. The DSM-IV-TR definitions of other sexual dysfunctions are formulated similarly, including those for ED, male orgasmic disorder, female orgasmic disorder, and female sexual arousal disorder.

Using the DSM-IV-TR for the Diagnosis of PE It is undoubtedly useful, and perhaps necessary in specific circumstances, to employ some form of IELT threshold for the diagnosis of PE. In clinical trials, in which study populations must be defined consistently, the use of diagnostic thresholds of some kind, in conjunction with the DSM-IV-TR criteria, may be appropriate. Nearly all clinical trials of treatments for PE have applied an IELT threshold in the definition of the eligible study population [8,20,41,49–51], and some have also included other criteria, based on the DSM-IV-TR [8,33,52]. Ideally, the diagnosis of men in clinical trials should mimic a diagnosis in clinical practice; otherwise, the results of clinical trials cannot be translated effectively to the routine diagnosis and treatment of men with PE. Patients presenting for the treatment of PE in clinical practice are not subject to the strict criteria of clinical trials—single men not in long-term relationships and those who do not wish to measure IELT with a stopwatch should not be excluded from the treatment of PE. Because the negative impact of the condition (i.e., distress, interpersonal difficulty, etc.) drives the patient to seek treatment [1], and perceived control over ejaculation has a direct effect on these outcomes [17,28], the DSM-IV-TR definition is highly appropriate. While it may not be practical to ask patients complaining of PE to provide stopwatch assessments of IELT, estimated IELT could certainly be considered in the diagnosis, based on the clinician’s judgment. Large epidemiological studies are limited in their ability to obtain certain types of data; certainly, stopwatch-measured IELT is not a feasible option with this methodology. The PEPA survey applied a modified DSM-IV-TR set of criteria, based on self-reported low or absent control over J Sex Med 2007;4:1468–1478

1476 ejaculation, and an ejaculatory latency that was a problem for a man or his partner [1]. The National Health and Social Life Survey used latent class analysis of results from 90-minute, in-person interviews to assign PE status, as well as other factors [3]. The Global Study of Sexual Attitudes and Behaviors used a simple “yes” or “no” response to the question, “During the last 12 months have you ever experienced any of the following for a period of 2 months or more when you . . . reached climax (experienced orgasm) too quickly” [53]. It is likely that the inability to replicate a clinical diagnosis of PE in the context of a large survey limits the interpretation of results from these studies. Conclusions

The ability of the DSM-IV-TR criteria to diagnose PE accurately has been challenged, and, as a result, a recommendation was made to a move away from the multifactorial approach and a move toward an IELT-only approach. Several recent studies have clearly supported the importance of each of the domains contained in the DSM-IVTR, including latency time, perceived control over ejaculation, and personal distress and interpersonal difficulty related to ejaculation in the diagnosis of PE. Furthermore, new analyses have demonstrated that it is the patient’s sense of poor control over ejaculation that contributes directly to his elevated personal distress and decreased sexual satisfaction, while IELT only has an indirect effect on these outcomes. These results strongly suggest that the multifactorial DSMIV-TR approach is a highly appropriate diagnostic for PE. Corresponding Author: David Rowland, PhD, Valparaiso University—Kretzmann Hall 116, 1700 Chapel Drive, Valparaiso, IN, USA 46383-6493. Tel: 219 464 5446; Fax: 219 464 5381; E-mail: david. [email protected] Conflict of Interest: Dr. Shabsigh is a consultant/advisor and investigator for Johnson & Johnson, Raritan, NJ. Dr. Rowland is a consultant/advisor for Johnson & Johnson, Raritan, NJ. Statement of Authorship

Category 1 (a) Conception and Design David Rowland; Ridwan Shabsigh (b) Acquisition of Data Not applicable J Sex Med 2007;4:1468–1478

Shabsigh and Rowland (c) Analysis and Interpretation of Data David Rowland; Ridwan Shabsigh

Category 2 (a) Drafting the Article David Rowland; Ridwan Shabsigh (b) Revising It for Intellectual Content David Rowland; Ridwan Shabsigh

Category 3 (a) Final Approval of the Completed Article David Rowland; Ridwan Shabsigh

References

1 Porst H, Montorsi F, Rosen RC, Gaynor L, Grupe S, Alexander J. The Premature Ejaculation Prevalence and Attitudes (PEPA) Survey: Prevalence, comorbidities, and professional help-seeking. Eur Urol 2007;51:816–24. 2 Laumann EO, Nicolosi A, Glasser DB, Paik A, Gingell C, Moreira E, Wang T. Sexual problems among women and men aged 40–80 y: Prevalence and correlates identified in the Global Study of Sexual Attitudes and Behaviors. Int J Impot Res 2005;17:39–57. 3 Laumann EO, Paik A, Rosen RC. Sexual dysfunction in the United States: Prevalence and predictors. JAMA 1999;281:537–44. 4 Moreira ED Jr. Help-seeking behaviour for sexual problems: The Global Study of Sexual Attitudes and Behaviors. Int J Clin Pract 2005;59:6–16. 5 Symonds T, Roblin D, Hart K, Althof S. How does premature ejaculation impact a man’s life? J Sex Marital Ther 2003;29:361–70. 6 Aschka C, Himmel W, Ittner E, Kochen MM. Sexual problems of male patients in family practice. J Fam Pract 2001;50:773–8. 7 McCabe MP. Evaluation of a cognitive behavior therapy program for people with sexual dysfunction. J Sex Marital Ther 2001;27:259–71. 8 Pryor JL, Althof SE, Steidle C, Rosen RC, Hellstrom WJG, Shabsigh R, Miloslavsky M, Kell S, for the Dapoxetine Study Group. Efficacy and tolerability of dapoxetine in the treatment of premature ejaculation: Integrated analysis of two randomized, double-blind, placebo-controlled trials. Lancet 2006;368:929–37. 9 American Psychiatric Association. Diagnostic and statistical manual of mental disorders, DSM-IV. 4th edition, 4th text rev. Washington, DC: American Psychiatric Association; 2000. 10 Waldinger MD, Schweitzer DH. Changing paradigms from a historical DSM-III and DSM-IV view toward an evidence-based definition of premature ejaculation. Part I-Validity of DSM-IV-TR. J Sex Med 2006;3:682–92.

1477

Defining PE with the DSM-IV-TR 11 Waldinger MD, Schweitzer DH. Changing paradigms from a historical DSM-III and DSM-IV view toward an evidence-based definition of premature ejaculation. Part II—Proposals for DSM-V and ICD-11. J Sex Med 2006;3:693–705. 12 Montague DK, Jarow J, Broderick G, Dmochowsky RR, Heaton JP, Lue TF, Nehra A, Sharlip ID, AUA Erectile Dysfunction Guideline Update Panel. AUA guideline on the pharmacologic management of premature ejaculation. J Urol 2004;172:290–4. 13 McMahon CG, Abdo C, Incrocci L, Perelman M, Rowland D, Waldinger M, Xin Z. Disorders of orgasm and ejaculation in men. J Sex Med 2004; 1:58–65. 14 National Center for Health Statistics. International classification of diseases, 9th revision, clinical modifications (ICD-9-CM). Available at: http:// www.cdc.gov/nchs/about/otheract/icd9/ abticd9.htm (accessed December 7, 2005). 15 Patrick DL, Althof SE, Pryor JL, Rosen R, Rowland DL, Ho KF, McNulty P, Rothman M, Jamieson C. Premature ejaculation: An observational study of men and their partners. J Sex Med 2005;2:358–67. 16 Althof S, Rosen R, Symonds T, Mundayat R, May K, Abraham L. Development and validation of a new questionnaire to assess sexual satisfaction, control, and distress associated with premature ejaculation. J Sex Med 2006;3:465–75. 17 Patrick D, Rowland D, Rothman M. Interrelationships among measures of premature ejaculation: The central role of perceived control. J Sex Med 2007;4:780–8. 18 Hellstrom WJ. The DSM-IV-TR is an appropriate diagnostic tool for premature ejaculation. J Sex Med 2007;4:252. 19 Waldinger MD, Zwinderman AH, Olivier B, Schweitzer DH. Majority of men with lifelong premature ejaculation prefer daily drug treatment: An observational study in a conservative group of Dutch men. J Sex Med 2007 (in press). 20 Waldinger MD. An empirical operationalization study of DSM IV diagnostic criteria for premature ejaculation. Int J Psychiatry Clin Pract 1998;2:287– 93. 21 Hartmann U, Schedlowski M, Kruger TH. Cognitive and partner-related factors in rapid ejaculation: Differences between dysfunctional and functional men. World J Urol 2005;23:93–101. 22 Waldinger MD, Quinn P, Dilleen M, Mundayat R, Schweitzer DH, Boolell M. A multinational population survey of intravaginal ejaculation latency time. J Sex Med 2005;2:492–7. 23 Masters W, Johnson V. Human sexual inadequacy. Boston, MA: Little, Brown and Co.; 1970. 24 Bell AP. Human sexuality—a response. Int J Psychiatry 1972;10:99–102. 25 Vanden Broucke H, Everaert K, Peersman W, Claes H, Vanderschueren D, Van Kampen M. Ejaculation

26

27 28

29

30

31

32

33

34

35

36 37

38

latency times and their relationship to penile sensitivity in men with normal sexual function. J Urol 2007;177:237–40. Mos J, Mollet I, Tolboom JT, Waldinger MD, Olivier B. A comparison of the effects of different serotonin reuptake blockers on sexual behaviour of the male rat. Eur Neuropsychopharmacol 1999; 9:123–35. Waldinger MD, Olivier B. Animal models of premature and retarded ejaculation. World J Urol 2005;23:115–8. Rosen RC, McMahon CG, Niederberger C, Broderick GA, Jamieson C, Gagnon DD. Correlates to the clinical diagnosis of premature ejaculation: Results from a large observational study of men and their partners. J Urol 2007;177:1059– 64. Strassberg DS, de Gouveia Brazao CA, Rowland DL, Tan P, Slob AK. Clomipramine in the treatment of rapid (premature) ejaculation. J Sex Marital Ther 1999;25:89–101. Haensel SM, Rowland DL, Kallan KT, Koos Slob A. Clomipramine and sexual function in men with premature ejaculation and controls. J Urol 1996; 156:1310–5. Rowland DL, Cooper SE, Slob AK. Genital and psychoaffective response to erotic stimulation in sexually functional and dysfunctional men. J Abnorm Psychol 1996;105:194–203. McMahon CG, Stuckey BG, Andersen M, Purvis K, Koppiker N, Haughie S, Boolell M. Efficacy of sildenafil citrate (Viagra) in men with premature ejaculation. J Sex Med 2005;2:368–75. Althof S, Rosen R, Mundayat R, May K, Symonds T. A new premature ejaculation questionnaire to evaluate men’s level of control over ejaculation, satisfaction with sex life, and level of associated distress (Abstract). J Sex Med 2006;3(1 suppl):14. Giuliano F, Patrick DL, Porst H, La Pera G, Rothman M, Hashmonay R, Merchant S. Premature ejaculation: Results from a European observational study. Poster presented at the 2006 Annual Congress of the European Association of Urology; April 5–8, 2006; Paris, France. World Health Organization. International statistical classification of diseases and related health problems. ICD-10. Volume 1. Geneva: World Health Organization; 1992:355–6. Rowland DL, Patrick DL, Rothman M, Gagnon DD. The psychological burden of premature ejaculation. J Urol 2007;177:1065–70. Rowland D, Perelman M, Althof S, Barada J, McCullough A, Bull S, Jamieson C, Ho KF. Selfreported premature ejaculation and aspects of sexual functioning and satisfaction. J Sex Med 2004;1:225– 32. Byers ES, Grenier G. Premature or rapid ejaculation: Heterosexual couples’ perceptions of men’s ejaculatory behavior. Arch Sex Behav 2003;32:261–70. J Sex Med 2007;4:1468–1478

1478 39 McCabe MP. Intimacy and quality of life among sexually dysfunctional men and women. J Sex Marital Ther 1997;23:276–90. 40 Waldinger MD, Zwinderman AH, Olivier B. On-demand treatment of premature ejaculation with clomipramine and paroxetine: A randomized, double-blind fixed-dose study with stopwatch assessment. Eur Urol 2004;46:510–6. 41 Waldinger MD, Hengeveld MW, Zwinderman AH, Olivier B. Effect of SSRI antidepressants on ejaculation: A double-blind, randomized, placebocontrolled study with fluoxetine, fluvoxamine, paroxetine, and sertraline. J Clin Psychopharmacol 1998;18:274–81. 42 Rowland DL, Motofei IG. The aetiology of premature ejaculation and the mind-body problem: Implications for practice. Int J Clin Pract 2007;61:77– 82. 43 Waldinger MD, Zwinderman AH, Olivier B, Schweitzer DH. Proposal for a definition of lifelong premature ejaculation based on epidemiological stopwatch data. J Sex Med 2005;2:498–507. 44 Waldinger MD. The neurobiological approach to premature ejaculation. J Urol 2002;168:2359–67. 45 Fisher NDL, Williams GH. Hypertensive vascular disease. In: Kasper DL, Braunwald E, Fauci AS, Hauser S, Longo DL, Jameson JL, eds. Harrison’s principles of internal medicine. New York, NY: McGraw-Hill; 2005:1463–81. 46 Lindsay R, Cosman F. Osteoporosis. In: Kasper DL, Braunwald E, Fauci AS, Hauser S, Longo DL, Jameson JL, eds. Harrison’s principles of internal medicine. New York, NY: McGraw-Hill; 2005: 2268–78.

J Sex Med 2007;4:1468–1478

Shabsigh and Rowland 47 Williams GH. Approach to the patient with hypertension. In: Braunwald E, Fauci AS, Kasper DL, Hauser SL, Longo DL, Jameson JL, eds. Harrison’s principles of internal medicine. New York, NY: McGraw-Hill Medical Publishing Division; 2001: 211–4. 48 National Institutes of Health, National Heart, Lung, and Blood Institute, National High Blood Pressure Education Program. The sixth report of the joint national committee on prevention, detection, evaluation, and treatment of high blood pressure. NIH publication no. 98–4080. Bethesda, MD: National Institutes of Health; 1997. 49 Waldinger MD, Zwinderman AH, Olivier B. Antidepressants and ejaculation: A double-blind, randomized, placebo-controlled, fixed-dose study with paroxetine, sertraline, and nefazodone. J Clin Psychopharmacol 2001;21:293–7. 50 Waldinger MD, Hengeveld MW, Zwinderman AH. Paroxetine treatment of premature ejaculation: A double-blind, randomized, placebo-controlled study. Am J Psychiatry 1994;151:1377–9. 51 Althof SE, Levine SB, Corty EW, Risen CB, Stern EB, Kurit DM. A double-blind crossover trial of clomipramine for rapid ejaculation in 15 couples. J Clin Psychiatry 1995;56:402–7. 52 Althof SE, Rowland DL, McNulty P, Rothman M. Evaluation of the impact of premature ejaculation on a man’s self-esteem, confidence, and overall relationship (Abstract). J Sex Med 2006;3(1 suppl):14. 53 Nicolosi A, Laumann EO, Glasser DB, Moreira ED Jr, Paik A, Gingell C. Sexual behavior and sexual dysfunctions after age 40: The global study of sexual attitudes and behaviors. Urology 2004;64:991–7.