Submitted by Gabe Deem. Published: 3/12/2015. Twitter @GabeDeem

Nothing Adds Up in Dubious Study: Youthful Subjects' ED Left Unexplained

A critique of "Viewing Sexual Stimuli Associated with Greater Sexual Responsiveness, Not Erectile Dysfunction" by Nicole Prause and Jim Pfaus

A study claims to have investigated porn-induced erectile dysfunction! This would be exciting news, if, in fact, the researchers actually gathered data that is essential to investigate porn-induced ED (PIED). Let me start off by making something clear, that I will explain in further detail below; this study doesn't, and due to its poor design can’t, tell us anything about whether or not today's high-speed internet porn leads to erectile dysfunction with a partner.

Why can’t this study tell us anything important about the possibility of PIED? Because of what it doesn't do, and the many, many flaws in what it claims to have done.

What the study doesn't do:

1) The study doesn't investigate men complaining of erectile dysfunction. The study doesn't examine young men with years of porn use and unexplained ED (that is, men for whom organic, below-the-belt problems have been ruled out). Nor does the study investigate porn-induced ED in such men by having them remove porn use and monitor possible changes. In fact, the researchers didn't even furnish details for their subjects who revealed they had erectile function problems on the IIEF [erectile-function] questionnaire (later). Yet the authors draw far reaching conclusions about the non-existence of porn-induced ED.

2) The study doesn't study men with porn addiction, or even "heavy" porn users. Just non-compulsive users. From the study's conclusion:

"These data did not include hypersexual patients. Results are probably best interpreted as limited to men with normal, regular VSS use."

Translation: The study did not include "hypersexuals", which is the authors' term for "porn addicts". Excluding hypersexuals is a huge weakness, considering that most men with chronic porn-induced ED self-identify as porn addicts. A small minority of men with porn-induced ED do not appear to be addicted, but they usually have a history of years of porn use.

Not only does this study not examine men with chronic ED, it excludes heavy porn users and porn addicts. Nothing like not looking at something if you don't want to find evidence of it!

3) The college age subjects were not asked about years of porn use! The subjects, as far as I know, could have started using porn merely weeks before the study, or they could have given up their porn watching ways just before the study was conducted after watching for years. Some could have started at age 10, or started in their sophomore year of college, or they could have just broken up with their girlfriend last month, and are now heavy users.

4) The study doesn't assess actual erections in relation to hours of use, contrary to what its title implies.

The study claims (more below) that men were asked a single question about how aroused they were after they were shown some porn. As the study said,

"No physiological genital response data were included to support men’s self-reported experience."

To summarize, this study:

  1. Did not assess individuals complaining of erectile dysfunction
  2. Did not include heavy porn users or porn addicts
  3. Did not assess "sexual response" (contrary to the misleading title)
  4. Did not ask men to attempt masturbation without porn (the way to test for porn-induced ED)
  5. Did not have men remove porn to see if erectile functioning eventually improved (the only way to know it's porn-induced)
  6. Did not ask about years of porn use, age guys started using porn, type of porn, or escalation of use.
  7. Did not ask about delayed ejaculation or anorgasmia (precursors to PIED)

What the study claims to do (link to study):

The claims are almost irrelevant as this jumbled data-salad is not even a true study with subjects chosen for this investigation. Instead, lead author Prause claims to have cannibalized bits and pieces of four of her older studies to construct this ED "study." However, those four studies were not about erectile dysfunction, nor did any of them report correlations between porn use and erectile function. Far more egregious is that the collective data from those four studies in no way line up with the data claimed for this ED study. The forthcoming details will have you asking, "How in the world did this mess pass peer-review?"

Before I explore the discrepancies, omissions, and sleights of hand the authors employed, you will need a few basics on the study. Using primarily university psychology students (average age 23), the study claimed to examine the relationship between:

  1. Some subjects' weekly hours of porn use and self-reported arousal after viewing porn in the lab (based on a single question that did not ask about erections), and
  2. Some subjects' weekly hours of porn use and some subjects' scores on the International Index of Erectile Function (IIEF).

The authors' claims for 1 & 2 above are as follows:

  1. Those who used 2+ hours of porn per week reported a somewhat higher sexual arousal score (6/9) than the two lower categories of porn use (5/9).
  2. No significant correlation was found between moderate porn use and erectile function scores on the IIEF.

I dissect the claims under number 1 and number 2 below. With each claim I hark back to the discrepancies and omissions that I will now detail.

A closer look at the study: Missing subjects, omissions, discrepancies & misrepresentations

1) The starting point: We are told that the subjects and data for this ED study were culled from four other studies, which have already been published (study 1, study 2, study 3, study 4). As noted, none of the studies assessed the relationship between porn use and erectile dysfunction. Only one study reported erectile functioning scores, for only 47 men.

2) Number of total subjects: Lead author Prause tweeted several times about the study, letting the world know that 280 subjects were involved, and that they had "no problems at home". However, the four underlying studies contained only 234 male subjects. While 280 appears once in this study's Table 1 as the number of subjects reporting "intercourse partners last year", so do the numbers 262, 257, 212 and 127. Yet, none of these numbers match anything reported in the 4 underlying studies, and only 47 men took the erection questionnaire. Contrary to her tweet, the average score (21.4) for erectile function placed these 47 young men, on average, squarely in the mild ED category. Oops.

Prause tweet Jan, 2015

  • Discrepancy 1: 46 subjects appear out of nowhere in the claim of 280 subjects, while the actual number of subjects (234) is found nowhere in the ED study.
  • Discrepancy 2: Subject numbers in Table 1: 280, 262, 257, 212 and 127 - match nothing from the 4 underlying studies.
  • Unsupported claim: Prause tweets that the study involved 280 subjects.
  • Missing: Any explanation of how Prause conjured the number "280" for her subjects.
  • Unsupported claim 2: Prause tweeted they had no problems, but their erection scores indicate ED on average.

3) Number of subjects who took the IIEF (erectile-function test): The ED study claims that 127 men took the IIEF (pg 11 also says 133). However, only one of the four studies reported IIEF scores, and the number of subjects who took it was 47. Where did Prause get the extra 80 men? She doesn't explain. This study did not assess the erectile functioning of 280 subjects, nor 234, and not even 127. Again, only 47 subjects took the IIEF.

  • Discrepancy: Study claims that 127 subjects took the IIEF, but it's really 47.
  • Unsupported claim: Prause tweets that 280 subjects were involved.
  • Missing: Any raw data on the mysterious 127

4) Average IIEF score for 47 subjects same as for the missing 80: As described above, only one study, with 47 men, reported an IIEF score. That study only reported a score for the full 15-question IIEF, not the 6-question "erection subscale" reported in the current study. Wherever it came from, the average score for the 6-question erection subscale was 21.4, and indicates "mild erectile dysfunction". In addition, the current ED study also claims an average IIEF score of 21.4 for the entire 127. Say what? We know the "accounted for" 47 men averaged 21.4, and the 127 averaged 21.4. This means that the 80 missing men had to also average 21.4. What is the probability that happening? The only sane conclusion is that the missing 80 never existed.

  • Unbelievable coincidence: The average IIEF scores for the 47 men must be the same as the unaccounted for 80 men.
  • Misleading: The average score (21.4) indicates "mild erectile dysfunction", while the study claims the men had "relatively good erectile function" (maybe relative to a 70-year old man?).
  • Missing: IIEF scores for erection sub-scale on original study.
  • Missing: The IIEF scores for any subject. No raw data, no scatter plot, no graph.

5) Number of subjects for hours/week porn viewing: The ED study claims to have porn viewing data on 136 men. Instead, only 90 subjects, from 2 studies, report hours viewing per week. Where did the authors conjure up 46 extra subjects? In addition, this study claims to correlate hours of porn viewing per week with the IIEF scores, but 90 men (hrs/week) doesn't match 47 men (IIEF scores).

  • Discrepancy 1: Study claims hrs/week viewing porn data for 136 subjects, but it's really 90.
  • Discrepancy 2: Study claims to correlate hrs/week porn viewing with IIEF scores, but 90 doesn't equal 47
  • Unsupported claim: Prause tweets N=280, but the true N=47.
  • Missing: Hours viewed for the subjects. No raw data, no scatter plot, no graph, no mean or standard deviation.
  • Missing: No legitimate data on the correlation between porn use and hours viewed per week.

6) Sexual arousal ratings: On page 8 the authors state that men rated their sexual arousal after viewing porn on a scale from 1 to 9.

""men were asked to indicate their level of “sexual arousal” ranging from 1 “not at all” to 9 “extremely.”

In reality, only 1 of the 4 underlying studies used a 1 to 9 scale. One used a 0 to 7 scale, one used a 1 to 7 scale, and one study did not report sexual arousal ratings. By the way, the study misleads the press, and readers, by implying in its title that erections were measured in the lab and found to be more "responsive" in association with more porn viewing. This did not happen. At best, the scores imply craving or horniness.

  • Discrepancy: Arousal scales in ED paper don't match arousal scales in 3 underlying studies.
  • Misleading: This study did not assess "sexual responsiveness" or erectile response.
  • Missing: No raw data or scatter plot for the subjects.

7) Stimulus used for sexual arousal ratings: The authors make a big deal about the sexual arousal ratings being slightly higher for the 2 plus hour/ per week group. Wouldn't a good study use the same stimulus for all subjects? Of course. But not this study. Three different types of sexual stimuli were used in the 4 underlying studies: Two studies used a 3-minute film, one study used a 20-second film, and one study used still images. It's well established that films are far more arousing than photos. What's shocking is that in this study Prause claims all 4 studies used sexual films:

"The VSS presented in the studies were all films."

Absolutely false! Only 2 studies with 90 men reported scores, and 47 of those men viewed only pictures of naked women, not films.

  • Discrepancy 1: Four different studies, and 3 different types of sexual stimuli...but one graph.
  • Discrepancy 2: In the graph below are 136 subjects, yet only 90 subjects actually reported hours of porn/week in any of the underlying studies.
  • Discrepancy 3: The sexual arousal scale is 1 - 7 in the graph below, yet the study said the scale was 1 - 9 (which was claimed to have been used in 1 of the 4 studies)
  • Unsupported claim: Prause claims all 4 studies used films.

Keep in mind that these porn viewing subjects are the same group as in number 5 above, and in the graph under number 1. Both claim 136 men, but the data says otherwise.

8) No data correlating porn use with IIEF scores: What's the headline news from this study? The authors claim there was no relationship between erectile functioning scores and hours of porn viewed per week. Big news, but no data. All they offer are a few sentences (pg 11-12) reassuring us that no correlation was found. No data, no graph, no scores, nothing. Only an allusion to the mysterious 127 men, 80 of whom are unaccounted for, discussed in 3 and 4 above. From the study:

"Men (N = 127) reported relatively good erectile functioning (see Table 1). Neither the total scale score, nor the erectile subscale score, on the International Index of Erectile Functioning was related to the hours of VSS viewed in the average week."

  • Missing 1: Any graph or table showing us a correlation between hours of porn viewed/week and IIEF scores.
  • Missing 2: Raw data. Any data.
  • Discrepancy: They appear to claim 127 subjects, yet only 47 men took the IIEF.
  • Misleading: Claiming the men "reported relatively good erectile functioning", while the average score (21.4) indicates mild ED.

With absolutely nothing in the 4 underlying studies matching the ED study, and with 80 subjects nowhere to be found, excuse me if I don't take the authors' word on the lack of correlation with hours of use. To illustrate this point, the study's conclusion opens with a string of inaccuracies:

"Data from a large sample of men (N = 280) across similar studies were aggregated to test the hypothesis that consuming more VSS was related to erectile problems."

In just this one sentence, I can identify a host of misrepresentations:

  • "N = 280": Nope, only 47 men took the IIEF
  • "across similar studies": Nope, the studies were not similar.
  • "were aggregated": Nothing matches the underlying 4 studies
  • "to test the hypothesis": No data were presented for the authors' hypothesis.

The entire study is like this, with subjects, numbers, methodologies, and claims appearing from nowhere, and unsupported by the underlying studies.


Let's look more closely at what the researchers claim to have investigated

NUMBER 1: Weekly hours of porn use and self-reported arousal after viewing porn in the lab

The researchers claim to have placed 136 participants in three groups based on weekly porn use (graph below). Discrepancy: Weekly porn use is only reported for 90 subjects in 2 studies.

Bar graph

Men were shown porn in the lab, and the study claimed they rated their arousal using a scale of 1 to 9. Discrepancy 1: One study used 0-9, three studies used 0-7. Discrepancy 2: Apples and oranges - One study used still images, one a 20 second film, two used a 3 minute video.

The bar graph allowed the authors to avoid plotting arousal scores clearly. Thus, readers cannot contemplate variations in self-reported arousal relative to hours of porn use for themselves. The researchers imply that answering a question about "sexual arousal" is solid evidence of erectile function. In fact, there's a footnote in one study saying that the researchers ignored questionnaire results on "penile erection" because they presumed that "sexual arousal" would gather the same information. However, that is most definitely not a reasonable assumption for guys with porn-induced erectile dysfunction (who are very aroused by porn but can't get erections with partners), and it may not be true of participants here either.

Another, more legitimate, way to interpret this arousal difference between the two porn-use groups is probably that men in '2+ hours per week' category experienced slightly greater cravings to use porn. Interestingly, they had less desire for sex with a partner and more desire to masturbate than those who logged .01-2 hours watching porn. (Figure 2 in study). This is quite possibly evidence of sensitization, which is greater reward circuit (brain) activation and craving when exposed to (porn) cues. Sensitization can be a precursor to addiction.

Recently, two Cambridge University studies demonstrated sensitization in compulsive porn users. Participants' brains were hyper-aroused in response to porn video clips, even though they didn't "like" some of the sexual stimuli more than control participants. In a dramatic example of how sensitization can affect sexual performance, 60% of the Cambridge subjects reported arousal/erectile problems with partners, but not with porn. From the Cambridge study:

"CSB subjects reported that as a result of excessive use of sexually explicit materials.....they experienced diminished libido or erectile function specifically in physical relationships with women (although not in relationship to the sexually explicit material)"

Put simply, a heavy porn user can experience higher subjective arousal (cravings) yet also experience erection problems with a partner. In short, his arousal in response to porn is not evidence of his "sexual responsiveness"/erectile function.

  • Watching more porn will improve erections??

Astoundingly, the authors of the current study suggest that "VSS viewing might even improve erectile functioning." Their advice is based on arousal and desire scores (not erectile-function scores). This is the worst advice possible if these "aroused" young men are, in fact, becoming sensitized (addicted) to porn. Their porn viewing arousal would not translate to their erectile function during real sex, which tends to decline in those who develop porn-induced ED as their sensitization to porn grows. Such a decline is precisely what the Cambridge subjects reported.

Certainly, viewing porn might improve erections while viewers watch, but the problem for those reporting porn-induced ED is overwhelmingly erectile function with partners. Moreover, there is no evidence in this study that viewing porn, or, as the authors obligingly suggest, a variety of porn, improves erectile function with partners. If this were correct, I would think the 47 young men tested for erectile function would have reported better boners the more porn they watched. Instead, they reported "mild erectile dysfunction" as a group.

It's worth noting that the Cambridge researchers addressed both compulsive porn users (CSB) and young men with ED while investigating porn addicts' brains. The current study missed both aspects, while purporting to investigate ED in young porn users.

NUMBER 2: Weekly hours of porn use and scores on a questionnaire entitled the International Index of Erectile Function (IIEF)

Here's where things get really ugly. The authors claimed that 127 young men completed a questionnaire called the IIEF, a 15-item survey (not a "19-item survey" as the authors state), in which men score their erectile health, desire and sexual satisfaction during masturbation and, primarily, sexual intercourse. Again, no actual penis responses were measured to confirm these self-reported scores. Discrepancy: only 47 men took the IIEF. Note: they also say on page 11 that 133 men took the IIEF. Does it ever end?

IIEF scores from this study

  • The Unknown 59 (sic)

For a moment let's imagine we are in a parallel universe, and 127 men actually took the IIEF. The authors stated that only 59 had partners with whom they could observe their current erectile health. This makes the number of partnered subjects whose erectile health was actually investigated quite small. Yet these are the only participants who can help researchers understand current erectile function in relation to porn use. Why? Because, as the authors acknowledge, assessing current erectile function is dependent on the availability of a partner.

  • First, many young guys report a rapid decline in erectile health when they try to have sex with a partner after being on their own (with porn use) for a time. So "erectile function" tests based on remembered erectile function with partners would be of little value.
  • Second, men on recovery forums report that porn-induced ED is overwhelmingly likely to occur during partnered sex (or during masturbation without porn, a statistic the researchers didn't collect) - not with porn. In fact, some guys have dubbed this phenomenon "copulatory impotence."

So, why aren't the partnered men who took the IIEF the only subjects included in this study? And why aren't their data broken out clearly for readers? The researchers tell us that there was no association between viewing hours and erectile function when the partnered participants "were included in analyses." However, we learn nothing about those claimed analyses, or how they compare with the others. They're always lumped into larger, un-sourceable numbers, like 280 or 127. Exit the parallel universe and back to more shenanigans.

  • "Mild erectile dysfunction"

Let's take another look at the IIEF's "erectile function" subscale. The chart below shows the questions and scoring. (View entire test and subscale.) Possible scores for this subscale range from 1 to 30. For the men who are claimed to have completed this subscale, the mean (average) score was only 21.4 out of a possible 30. On average, they fell well within the "mild erectile dysfunction" category.

Keep in mind that these sorry erectile function scores were self-reported by 23-year old men, none of whom watched porn compulsively. This suggests internet porn, even consumed in a non-compulsive manner, may be having detrimental effects on youthful erections irrespective of (no) correlation with hours used.

In fact, these young men were well below previously established control group scores for much older men. In 1997, the studies conducted to validate the IIEF reported that erectile-function scores averaged 26.9 (average age 58), and 25.8 (average age 55). In short, older men in 1997 - before internet porn - had healthier erections even in middle age than these 23-year olds.

Unlikely coincidence? How could the 47 subjects who took the IIEF have exactly the same average (21.4) as the 80 ghostly subjects no one can find (21.4)?

Moreover, as 21.4 is the average score (for some, indeterminate N), it means that the scores for some participants were lower than 21.4. In fact, the SD (standard deviation) was large (9.8), so there was a wide range of erectile function scores. It's likely some fell into the "moderate" and "severe" erectile dysfunction categories. However, we don't know, because data are not provided - which brings me to...

  • Study graphics

Why didn't the authors in the current study do what conscientious researchers did in a recent study on porn users' brains, "Brain Structure and Functional Connectivity Associated With Pornography Consumption: The Brain on Porn," and plot all their data on a graph like that reproduced below? This allows the reader to see clearly that as pornography consumption rises, grey matter in the brain decreases. Why did the authors of this ED study hide individual data in mean scores and simplistic bar graphs?

Kuhn study scatter plot

  • Weekly Use?

The authors offer no support for their assumption that a correlation with weekly porn use is vital to establishing the existence of porn-induced erectile dysfunction, even though all their claims rest on the lack of correlation with weekly use scores. In 2011, German researchers found that porn-related problems correlate not with time spent, but rather with number of sex applications opened during porn sessions. Thus, the absence of a correlation between weekly hours of porn use and ED issues (let alone correlations with their other questionnaire results) isn't surprising, as novelty (number of clips, tabs open, etc.) appears to be more important than hours.

Moreover, precisely how were "weekly porn use" scores determined? The researchers do not say. Was it simply, "How much porn did you use last week?" If so, there could be new porn users who haven't had time to develop erectile problems in the "2+ hours" bin. And long-time users with porn-related problems, who had recently decided to cut out porn, perhaps due to sexual dysfunction symptoms, in the "0 hours" bin, making correlations even more unlikely.

Regardless of how the lead author calculated "weekly use," the most important data are still missing: total porn use and characteristics of use. Participants weren't asked about years of porn use or age (developmental stage) they began using. Moreover, the researchers didn't control for other factors that men on recovery forums often find are related to their performance issues: escalation to more extreme material, lengthy periods without partnered sex, need for novel porn, and masturbation only with internet porn.

Under the circumstances, and given the appalling numerical inconsistencies, a lack of correlations is of dubious significance, and the authors' dismissal of the phenomenon of porn-induced ED unwarranted.

Sexual conditioning: An idea worth exploring

The researchers correctly point out that:

Erections may become conditioned to aspects of VSS [porn] that do not transition easily to real-life partner situations. Sexual arousal may be conditioned to novel stimuli, including particular sexual images, specific sexual films or even non-sexual images. It is conceivable that experiencing the majority of sexual arousal within the context of VSS may result in a diminished erectile response during partnered sexual interactions. Similarly, young men who view VSS expect that partnered sex will occur with themes similar to what they view in VSS. Accordingly, when high stimulation expectations are not met, partnered sexual stimulation may not produce an erection.

Recognizing this possibility, one wonders why the researchers asked only about weekly hours and didn't ask their participants questions that would have helped reveal a possible link between their porn viewing and sexual conditioning, such as

  • at what age they began viewing porn videos
  • how many years they had viewed it
  • whether their tastes escalated over time to more extreme fetish porn
  • what percentages of their masturbations took place with and without porn.

If they wanted to find important data on porn-induced ED, they might also have asked the young men with low erectile function scores to masturbate both without porn and with it, and compare their experiences. Men with porn-induced ED generally have great difficulty masturbating without porn because they have conditioned their sexual arousal to screens, voyeurism, fetish content and/or constant novelty. Of course the researchers did not do that, because this wasn't a study specifically looking at the possibility of porn-induced ED.

Growing cause for concern

Highly regarded urologists have already spoken up about the issue of porn-induced ED, including academic urologists, such as Abraham Morgentaler, MD, Harvard urology professor and author, and Cornell urology professor and author Harry Fisch, MD. Said Morgentaler, "It’s hard to know exactly how many young men are suffering from porn-induced ED. But it’s clear that this is a new phenomenon, and it’s not rare." Fisch writes bluntly that porn is killing sex. In his book The New Naked, he zeroes in on the decisive element: the internet. It "provided ultra-easy access to something that is fine as an occasional treat but hell for your [sexual] health on a daily basis."

Interestingly, in the last few years a number of studies have reported unprecedented ED in young males, although none have inquired about internet porn use:

  1. Sexual functioning in military personnel: preliminary estimates and predictors. (2014) ED - 33%
  2. Sexual dysfunctions among young men: prevalence and associated factors. (2012) ED - 30%
  3. Erectile dysfunction among male active component service members, U.S. Armed Forces, 2004-2013. (2014) Annual incidence rates more than doubled between 2004 and 2013
  4. Prevalence and characteristics of sexual functioning among sexually experienced middle to late adolescents. (2014) 16-21 year olds:
  • Erectile Dysfunction - 27%
  • Low sexual desire - 24%
  • Problems with orgasm - 11%

Given that internet porn use is now nearly universal in young men, we should be slow to dismiss internet porn use as a potential cause of today's widespread youthful erectile dysfunction without a very thorough scientific investigation of subjects complaining of it. And slow as well to assume the authors are correct in their surmise that widespread youthful ED is due to "concerns about the partner’s STD status, relationship expectations, and concerns about one’s own attractiveness or penis size." Those factors have presumably been around for a much longer time than internet porn, and the spike in youthful ED problems is quite recent.

Most importantly, those concerns do not apply to guys who are unable to masturbate without porn, as they aren't worried about any of those concerns with their own hand.

While it is vital to publish all analyses about the subject of porn-induced sexual dysfunction that are based on solid research, this particular analysis raises a thicket of red flags. Today's youthful porn users deserve better.