November 14th, 2007
Recently I was reading Marty Klein’s controversial blog Sexual Intelligence. In one recent entry he debates the issue of sex addiction and linked to Patrick Carnes’ sex addiction screening test (SAST).
Now of all the things you know push my buttons are sex and (rather more sadly) questionnaires. So the chance to have a go at something that combined the two was too tempting to resist. Which is how I discovered I’m a sex addict.
Having taken 45 question test (which I answered completely honestly) I scored 9 on a scale that ranges from 0 through to 20. Apparently my score shows I have a problem that needed immediate therapeutic help. My answers also indicated:
“In addition there are certain subscales to further confirm that a problem exists. The following patterns emerged in your answers:
- A profile consistent with women who struggle with sexually compulsive behavior
The SAST measures key characteristics of addiction. The following dimensions of an addictive disorder appeared in your answers:
· Preoccupation: obsessive thinking about sexual behavior, opportunities, and fantasies
· Relationship disturbance: sexual behavior has created significant relationship problems
· Affect disturbance: significant depression, despair, or anxiety over sexual behavior”
Which is pretty scary. I’m a sex addict, and I’ve got significant relationship problems. I’m also depressed, desparing and anxious. Why didn’t anybody tell me?
Deciding to check this more thoroughly I did a bit of digging and discovered the Sexual Addiction Screening Test (SAST) was apparently “developed in cooperation with hospitals, treatment programs, private therapists, and community groups, the SAST provides a profile of responses which help to discriminate between addictive and non-addictive behavior” (according to the SAST page). A search of the scientific literature indicates it has been used in research and cited by a number of people to describe sexually addictive behaviours. In short, it seems to be something that’s used consistently to assess sex addiction.
Scarier still. I’d been diagnosed by a measure that lots of people seem to have been using in clinical and therapeutic practice. But I really don’t think I’m a sex addict. What’s wrong with me? Could I be in denial?
Well here are a few clues that something isn’t quite right with the SAST. Firstly I’m a new mum. So that might give you some idea about how much time, energy and enthusiasm I’ve got for sex right now. But presumably if I were a sex addict that wouldn’t be enough to stop me, so here are some further reasons why the SAST is faulty.
There are no parameters
In any well designed diagnostic tool you need to set parameters. That will help put any issues you are assessing into a timeframe and also give your respondents a clearer idea about what they’re supposed to be telling you in their answer.
The SAST completely lacks any parameters so you don’t know if you’re supposed to be answering in relation to recent events, or something you did in the past. You’re also unable to indicate the duration of anything you are answering about.
Let’s take questions 15 (Do you feel controlled by your sexual desire?), 16 (Have important parts of your life (such as job, family, friends, leisure activities) been neglected because you were spending too much time on sex?) and 17 (Do you ever think your sexual desire is stronger than you are?). To each of these you have to answer yes or no. You can’t indicate whether you are constantly driven by your desires or once you were besotted with someone and felt utterly overwhelmed by your feelings. So if on occasion you’ve totally been swept along by passion (and I sincerely hope most of us have at least once) you have to answer yes to all these questions.
For this to be a more reliable diagnostic you’d need to ask people to indicate not only how they are feeling, but how long they have been feeling this way for. And that you are looking for repetative and recent behaviour.
You can see how I started on my way to a sex addict score by questions like 8 (Has anyone been hurt emotionally because of your sexual behavior?). As a teenager I definitely upset a couple of boyfriends either by refusing sex or by making remarks about their willys. Now because this question doesn’t define sexual behaviour properly, refusing sex or discussing sex could still count as behaviour and can still cause emotional hurt. It forces people into suggesting they cause hurt through sexual activity when they could just as easily be answering they caused hurt through inactivity. And it can bring you one step up the addict scale without even necessarily having to be sexually active.
There are reliability and validity problems
Outcome measures such as this need to be unambiguous. That means everyone reading the questions will understand them similarly and they’ll actually measure what they claim to be measuring every time the questionnaire is used. What you don’t want is questions written in a way that people all respond to differently. In that case you’ll get an answer you think is genuine but you may have actually measured something else. Which happens throughout the SAST.
Let’s take questions 3 (Do you often find yourself preoccupied with sexual thoughts?) and 21 (Are you in crisis over sexual matters?). Giving a yes answer to these implies that sex (in terms of addiction) is taking over your life. But a yes answer could just as easily be given by a jealous partner, a person in the first flush of a new relationship, or someone falling out of love.
When you get a questionnaire that reveals problem answers like this it’s usually a sign of poor design and lack of piloting. Because enough testing of this tool would indicate you can have multiple responses to things that may seem straightforward.
This usually tends to happen when the person designing the outcome measure has a fixed idea about the thing they want to test and cannot see there is any other interpretation of their question. Which may explain the ludicrously unreliable question 11 (Do you hide some of your sexual behaviors from others?). No doubt this was designed to capture those people who hide their porn use, masturbation, affairs or casual encounters from others. But actually nobody can answer anything other than ‘yes’ to this question. After all, we all refrain from masturbating on the bus, shagging in front of our parents, or stripping during jury service. Don’t we?
Part of my sex addict diagnostic arose from answering yes to the following questions:
30. Have you spent considerable time and money on strip clubs, adult bookstores and movie houses?
28. Have you subscribed to or regularly purchased or rented sexually explicit materials (magazines, videos, books or online pornography)?
34. Have you regularly purchased romantic novels or sexually explicit magazines?
I research sex. So as part of my work I’ve visited sex clubs, stores and cinemas. I’ve bought materials to use in research. Okay I accept this isn’t an excuse everyone can use, but despite this there are a number of traps due to question wording that could make people seem like they have a problem when they don’t.
For example question 30 asks if you’ve spent time and money in a club. You could answer yes but only refer to one thing. So a cleaner who works in a strip club spends a lot of time there, so does a customer paying cash. Question 34 lumps together everything from explicit magazines, romantic novels and lads mags. Which when I last looked weren’t the same thing at all.
The SAST uses vague terms
When you’re trying to diagnose behaviour you need to give people concrete ideas to respond to. If you ask them things that aren’t clearly explained they won’t be able to answer properly. And there are key terms like ‘often’, ‘regularly’ or ‘frequently’ you should avoid because what one person’s definition of regularly is will differ from someone else’s.
Taking question 39 (Have you regularly engaged in sadomasochistic behavior?) ‘regularly’ for one respondent might be being tied up and spanked on a nightly basis, for others it could represent their annual birthday treat of being handcuffed to the bedpost. Going back to the lack of parameters there’s also the issue of no time limit so if you regularly engaged in SM in a relationship two decades ago you still have to answer yes even if you’ve never done it since. Moreover ‘behaviour’ is a vague term as in this case it could be anything from reading the Marquis de Sade to keeping willing slaves in your home-made dungeon.
It asks people impossible questions
We always try, where possible, to ensure terms in diagnostics and sexual health questionnaires are not judgemental or value laden. It’s for a simple reason, if you use very loaded terms people will either lie or refuse to answer. In many cases it’s not possible for people to accurately say whether they think their behaviour is okay or not. Someone who overdoes the alcohol or food may not think they have a drink problem or are heading towards obesity. While someone who has been led to think they’ve got a health problem may be hard to convince they’re fine.
The SAST asks people questions like ‘do you feel that your sexual behavior is not normal?’ (4). Now if you’re answering a survey entitled ‘sex addiction’ you’re probably half way to thinking you’re not normal. But also people have no real idea what is normal in terms of sexual behaviour. It’s normal for some people to not want sex at all. It’s normal for some people to like dressing up as a furry animal before sex. It’s normal for some people to enjoy a romantic dinner before sex every Thursday evening. We may think we’re normal because of what we do, or we may think we’re abnormal because we think other people don’t do as we do.
Asking people to say if they think their behaviour is normal is useless without asking them to list what behaviours they think are abnormal. That would give you an idea about whether they have a genuine problem or just lack basic sex information.
There’s no additional measure of distress
Throughout the SAST there are questions that indicates what the clinician who designed the measure thinks are problematic sexual behaviours. And I’m sure we’d all agree that abusing children or being abused yourself are both issues that are serious and need to be dealt with. However, these are put alongside other activities in the SAST like reading romantic novels and having sex in public places which may not be everyone’s cup of tea but aren’t a sexual dysfunction either.
Within any reputable measure of sexual behaviour we always include a question about distress. Because if a person is engaged in SM, public sex, activities that make them feel degraded, and lots of porn use but they are happy with it and not hurting anyone else this should be taken as a sign they are open about their sex life, rather than slapping a clinical diagnosis on their pervy ass.
So having been through the measure I think I can safely conclude I’m not a sex addict. I also used an additional measure that the SAST neglects. I obtained an independent assessment. I checked with my partner whether I seemed addicted to sex and if our relationship was in tatters. Their reply was a sarcastic ‘yes of course’. Confirming I’m not a sex addict and our relationship is just fine.
The problem with the SAST is that it has been used extensively by practitioners and featured in numerous journal publications. But at no time does anyone seem to have actually asked whether this measure has been piloted, subjected to any tests for reliability and validity, or standardised in any way. Which one would usually expect for a diagnostic tool of this kind.
Clearly I knew what I was doing when I answered the survey and I wasn’t answering it as someone who thought they had a problem with sex addiction. Yet if I can get a score that suggests I’m a sex addict, how many countless others will have been misdiagnosed?
This isn’t just a magazine quiz. This is a diagnostic measure that is being used by clinicians on a daily basis. It’s badly designed but clearly has been accepted on face value by people who really should know better. More shockingly this diagnostic that’s shot full of holes is also live on the web for people who worry they might have a problem to complete – and most likely have their worst fears confirmed.
So if you are a therapist or practitioner currently using the SAST I would recommend you stop until it has been fully tested, piloted, and the actual measure published in a peer reviewed journal.
For me this is either an example of a practitioner with poor research skills who hasn’t been able to design, test or validate a measure in a way you would expect within clinical practice. Or it’s someone who has deliberately created a diagnostic that is suitably vague to artificially increase the prevalence of sex addiction. Either case is unacceptable and I sincerely hope the therapeutic community considers investigating this measure, and its use in clinical practice.Tweet