These stories, and countless others like them in the global media today suggested abstinence only sex education is more effective than other forms of sex education. The press coverage was based on a study by Jemmott et al entitled ‘Efficacy of a Theory-Based Abstinence-Only Intervention Over 24 Months’ in the journal Archives of Pediatrics and Adolescent Medicine (unfortunately not open access). But did the research actually find abstinence sex education is superior to other approaches?
Before we can answer that question, let’s look at the study itself.
What did the research involve?
The research was based on 662 male and female low income African American students aged between 11-15 (grades 6-7) mean age 12 at the time the study started in 2001. Students were recruited through announcements in school assemblies, adverts in schools and youth clubs, and letters to parents as part of a pre-existing health promotion project. Those who volunteered were randomly assigned to one of four interventions that lasted an hour and ran over an eight week period in schools on Saturdays. 16 male and 51 female African American facilitators (mean age 43) were trained to deliver the interventions all of which included group work, games, watching videos, and skill building activities. The interventions were:
Abstinence information only Focused on abstinence (not having sex) to “eliminate the risk of pregnancy and STIs including HIV. It was designed to (1) increase HIV/STI knowledge, (2) strengthen behavioural beliefs supporting abstinence including the belief that abstinence can prevent pregnancy, STIs and HIV, and that abstinence can foster attainment of future goals and (3) increase skills to negotiate abstinence and reduce pressure to have sex. It was not designed to meet federal criteria for abstinence-only programs. For instance, the target behaviour was abstaining from vaginal, anal or oral intercourse until a time later in life when the adolescent is more prepared to handle the consequences of sex. The intervention did not contain inaccurate information, portray sex in a negative light, or use a moralistic tone. The training and curriculum manual explicitly instructed the facilitators not to disparage the efficacy of condoms or allow the view that condoms are ineffective to go uncorrected” (p.153)
Safer sex information only
Promoted condom use as a means of preventing HIV and other STIs and pregnancy, it did not mention abstaining from/delaying sex.
Comprehensive intervention
Combined both abstinence/delay and safer sex messaging (some of these sessions were 8 hour sessions in total, some were 12 hours in total).
Health promotion intervention
This activity served as a control and covered health-related education on topics such as preventing heart disease, stroke, or diabetes, and encouraging exercise plus avoiding cigarettes.
Participants in all of the interventions were asked to complete questionnaires before the study began, straight after the last intervention session and during follow up at 3,6,12, 18 and 24 months afterwards. The self-reported questionnaires assessed whether students had engaged in sexual activity, number of sexual partners, unprotected sex, and condom use.
What was the aim of the study?
The researchers sought to identify which approach to delivering sex education resulted in teenagers delaying sexual activity, and the researchers hypothesised the abstinence based approach would be most successful at delivering this.
Why focus on these participants?
In the US the rates of HIV and other STIs and unplanned pregnancy are higher among low income African American youth than other groups of teenagers. The study aimed to identify effective means of enabling this group of young people to reduce their risks of pregnancy and infection.
What did the study find?
The results indicated those who received the abstinence only intervention were significantly more likely than students in the health promotion (control) group to have delayed having sex at 24 months post-intervention. From the paper “the abstinence-only intervention reduced sexual initiation (P=.03). The model-estimated probability of ever having sexual intercourse by the 24-month follow-up was 33.5% in the abstinence-only intervention and 48.5% in the health-promotion control group. The safer sex and comprehensive interventions did not differ from the control group in sexual initiation” (p.156). This suggests the abstinence based approach was not significantly more effective than condom promotion only or combining delay plus condom use messaging. [Update 03/02/10: I was emailed about my review of the research by Joe Paxton from Harvard who pointed out I'd previously not been clear about the differences between the conditions and to question the outcomes presented in the paper - specifically that the claims that abstinence only education was 'better'. It appears better than control in delaying sexual activity, but not signficantly different to the other conditions (safer sex or combined interventions). This has led some critics to question the study further in terms of its outcomes and claim of the effectiveness of abstinence based education].
What were the limitations of the research?
As with any study there will be limitations on research, and educational interventions are notoriously difficult to manage and directly demonstrate effects. The researchers are very clear about potential limitations on this current study which include reliance on self report data (which can be biased by recall problems and pressure to provide socially desirable answers). They also note that across all groups most of the young people had not had sex by 24 months follow up (probably due to the wide age range of participants and the fact most teens don’t have sex aged under 16). So although a difference in groups was noted, overall participants hadn’t had sex regardless of whatever intervention they received. And that would make it impossible to assess issues like safer sex, number of partners or condom usage reliably.
The students were perhaps also unrepresentative because they volunteered to be in the research and were willing to attend 8-12 hourly educational sessions in school at weekends. Not only were they motivated but they presumably had family support too. The researchers explain this makes it difficult to assess whether similar interventions on different kinds of pupils in the US or elsewhere would be as effective. This is particularly important given a growing awareness that while young people would like access to social clubs or sexual health clinics many find those offered at the weekend to be particularly inaccessible (particularly those with strict parents, or for girls who may be more likely to have their movements more closely monitored).
In addition to the concerns noted by the researchers, I’d raise some further questions that aren’t addressed in the paper. A large number of facilitators were involved in delivering the interventions. All of them were trained but with so many people included in a study there is potential for messages to differ. An account of how this was controlled for would have been very helpful within this paper.
There’s also the issue of the timing of this research. As a follow up study it naturally required time to run, however the study began in 2001 with data collected until 2004, yet it was only submitted for publication in 2009. While there is nothing suspicious about this, it does raise questions about the cohort being studied as over the past decade (during which time this research was completed, analysed and published) there have been many changes in Western culture in relation to a shifting consumerist and sexualised culture. So it may well be the outcomes obtained might be different were the research to be replicated. This is not a reason to dismiss the study but it would have been useful to see this issue addressed within the paper.
The biggest problem with the paper however is the use of the term ‘abstinence only’. To many this may well imply a faith based approach and yet if you read the description of the abstinence only intervention quoted above it is very clear this was not remotely faith based and in fact differed quite markedly from such approaches – particularly around the accurate mentions of condoms. The focus appeared to be about confidence, delay and anticipating sex as a positive event.
When I read about the paper in the press coverage I anticipated a study that compared a faith-based abstinence programme with other approaches. In fact this study is really about a delay-based programme that anticipates sex positively. And that is not what many people would understand as abstinence.
This is unfortunate as it may well be many working within sex education and healthcare will dismiss or perhaps not consult this paper believing it is promoting a faith based, sex negative approach. And faith groups who advocate abstinence will claim this paper supports their educational approaches which often spread misinformation about condoms and do not adequately cover issues about STIs or pregnancy. Indeed such approaches tend to use scare tactics and a lack of information to encourage young people not to have sex until they are married. This was not a feature of this current study. If I had reviewed this paper I would have recommended the term ‘abstinence’ be replaced by ‘delay messaging’ which would be more accurate and helpful to those searching for educational interventions that might inform their sex education practice.
Unfortunately the media for the most part didn’t make this clear. Admittedly a few journalists (and media blogs) did pick up on some of the issues I’ve touched on above (particularly relating to the idea of faith based abstinence approaches). However most did not explain the research – most likely because they did not read the paper (or failed to understand it). This is problematic because the coverage does not faithfully explain the study. It has suggested that faith based, sex negative abstinence approaches are better than other forms of sex education. And this study (and countless others) clearly show that’s not true.
So what’s the take home message? This is a useful paper and a fair piece of research. It has limitations which means it can’t reliably be used to inform sex educational policy, but it would certainly benefit from adaption and replication and a longer follow up. Unfortunately the problem is less about the study (which is clearly discussed by the researchers) and more about how the media has misreported it and how politicians and faith based groups are misrepresenting the findings to suit sex-negative abstinence programmes.
Rather than falling into that trap we should take this research as further evidence that sex education is effective when it is tailored to the individual needs of different children; builds confidence; resists peer pressure; addresses feelings and emotions as well as infections and contraception; promotes delay until a young person is ready for intimacy (see also here and here); and prepares them for positive relationships when they are older.
Teaching abstinence makes teens delay sex? Here’s the evidence behind the media hype
In the news today is coverage of research suggesting abstinence-based sex education leads to children delaying sex and is better than other approaches to sex ed. But is this what the study actually found? Here’s the low down on the paper and what the findings really mean for sex education.
Study: abstinence programs most effective at delaying sex among youths – CNN
Teaching abstinence makes children delay first sexual intercourse: research – Telegraph
Study: abstinence only program shows promise – Baltimore Sun
Study finds benefit in abstinence programme – New York Times
These stories, and countless others like them in the global media today suggested abstinence only sex education is more effective than other forms of sex education. The press coverage was based on a study by Jemmott et al entitled ‘Efficacy of a Theory-Based Abstinence-Only Intervention Over 24 Months’ in the journal Archives of Pediatrics and Adolescent Medicine (unfortunately not open access). But did the research actually find abstinence sex education is superior to other approaches?
Before we can answer that question, let’s look at the study itself.
What did the research involve?
The research was based on 662 male and female low income African American students aged between 11-15 (grades 6-7) mean age 12 at the time the study started in 2001. Students were recruited through announcements in school assemblies, adverts in schools and youth clubs, and letters to parents as part of a pre-existing health promotion project. Those who volunteered were randomly assigned to one of four interventions that lasted an hour and ran over an eight week period in schools on Saturdays. 16 male and 51 female African American facilitators (mean age 43) were trained to deliver the interventions all of which included group work, games, watching videos, and skill building activities. The interventions were:
Abstinence information only
Focused on abstinence (not having sex) to “eliminate the risk of pregnancy and STIs including HIV. It was designed to (1) increase HIV/STI knowledge, (2) strengthen behavioural beliefs supporting abstinence including the belief that abstinence can prevent pregnancy, STIs and HIV, and that abstinence can foster attainment of future goals and (3) increase skills to negotiate abstinence and reduce pressure to have sex. It was not designed to meet federal criteria for abstinence-only programs. For instance, the target behaviour was abstaining from vaginal, anal or oral intercourse until a time later in life when the adolescent is more prepared to handle the consequences of sex. The intervention did not contain inaccurate information, portray sex in a negative light, or use a moralistic tone. The training and curriculum manual explicitly instructed the facilitators not to disparage the efficacy of condoms or allow the view that condoms are ineffective to go uncorrected” (p.153)
Safer sex information only
Promoted condom use as a means of preventing HIV and other STIs and pregnancy, it did not mention abstaining from/delaying sex.
Comprehensive intervention
Combined both abstinence/delay and safer sex messaging (some of these sessions were 8 hour sessions in total, some were 12 hours in total).
Health promotion intervention
This activity served as a control and covered health-related education on topics such as preventing heart disease, stroke, or diabetes, and encouraging exercise plus avoiding cigarettes.
Participants in all of the interventions were asked to complete questionnaires before the study began, straight after the last intervention session and during follow up at 3,6,12, 18 and 24 months afterwards. The self-reported questionnaires assessed whether students had engaged in sexual activity, number of sexual partners, unprotected sex, and condom use.
What was the aim of the study?
The researchers sought to identify which approach to delivering sex education resulted in teenagers delaying sexual activity, and the researchers hypothesised the abstinence based approach would be most successful at delivering this.
Why focus on these participants?
In the US the rates of HIV and other STIs and unplanned pregnancy are higher among low income African American youth than other groups of teenagers. The study aimed to identify effective means of enabling this group of young people to reduce their risks of pregnancy and infection.
What did the study find?
The results indicated those who received the abstinence only intervention were significantly more likely than students in the health promotion (control) group to have delayed having sex at 24 months post-intervention. From the paper “the abstinence-only intervention reduced sexual initiation (P=.03). The model-estimated probability of ever having sexual intercourse by the 24-month follow-up was 33.5% in the abstinence-only intervention and 48.5% in the health-promotion control group. The safer sex and comprehensive interventions did not differ from the control group in sexual initiation” (p.156). This suggests the abstinence based approach was not significantly more effective than condom promotion only or combining delay plus condom use messaging. [Update 03/02/10: I was emailed about my review of the research by Joe Paxton from Harvard who pointed out I'd previously not been clear about the differences between the conditions and to question the outcomes presented in the paper - specifically that the claims that abstinence only education was 'better'. It appears better than control in delaying sexual activity, but not signficantly different to the other conditions (safer sex or combined interventions). This has led some critics to question the study further in terms of its outcomes and claim of the effectiveness of abstinence based education].
What were the limitations of the research?
As with any study there will be limitations on research, and educational interventions are notoriously difficult to manage and directly demonstrate effects. The researchers are very clear about potential limitations on this current study which include reliance on self report data (which can be biased by recall problems and pressure to provide socially desirable answers). They also note that across all groups most of the young people had not had sex by 24 months follow up (probably due to the wide age range of participants and the fact most teens don’t have sex aged under 16). So although a difference in groups was noted, overall participants hadn’t had sex regardless of whatever intervention they received. And that would make it impossible to assess issues like safer sex, number of partners or condom usage reliably.
The students were perhaps also unrepresentative because they volunteered to be in the research and were willing to attend 8-12 hourly educational sessions in school at weekends. Not only were they motivated but they presumably had family support too. The researchers explain this makes it difficult to assess whether similar interventions on different kinds of pupils in the US or elsewhere would be as effective. This is particularly important given a growing awareness that while young people would like access to social clubs or sexual health clinics many find those offered at the weekend to be particularly inaccessible (particularly those with strict parents, or for girls who may be more likely to have their movements more closely monitored).
In addition to the concerns noted by the researchers, I’d raise some further questions that aren’t addressed in the paper. A large number of facilitators were involved in delivering the interventions. All of them were trained but with so many people included in a study there is potential for messages to differ. An account of how this was controlled for would have been very helpful within this paper.
There’s also the issue of the timing of this research. As a follow up study it naturally required time to run, however the study began in 2001 with data collected until 2004, yet it was only submitted for publication in 2009. While there is nothing suspicious about this, it does raise questions about the cohort being studied as over the past decade (during which time this research was completed, analysed and published) there have been many changes in Western culture in relation to a shifting consumerist and sexualised culture. So it may well be the outcomes obtained might be different were the research to be replicated. This is not a reason to dismiss the study but it would have been useful to see this issue addressed within the paper.
The biggest problem with the paper however is the use of the term ‘abstinence only’. To many this may well imply a faith based approach and yet if you read the description of the abstinence only intervention quoted above it is very clear this was not remotely faith based and in fact differed quite markedly from such approaches – particularly around the accurate mentions of condoms. The focus appeared to be about confidence, delay and anticipating sex as a positive event.
When I read about the paper in the press coverage I anticipated a study that compared a faith-based abstinence programme with other approaches. In fact this study is really about a delay-based programme that anticipates sex positively. And that is not what many people would understand as abstinence.
This is unfortunate as it may well be many working within sex education and healthcare will dismiss or perhaps not consult this paper believing it is promoting a faith based, sex negative approach. And faith groups who advocate abstinence will claim this paper supports their educational approaches which often spread misinformation about condoms and do not adequately cover issues about STIs or pregnancy. Indeed such approaches tend to use scare tactics and a lack of information to encourage young people not to have sex until they are married. This was not a feature of this current study. If I had reviewed this paper I would have recommended the term ‘abstinence’ be replaced by ‘delay messaging’ which would be more accurate and helpful to those searching for educational interventions that might inform their sex education practice.
Unfortunately the media for the most part didn’t make this clear. Admittedly a few journalists (and media blogs) did pick up on some of the issues I’ve touched on above (particularly relating to the idea of faith based abstinence approaches). However most did not explain the research – most likely because they did not read the paper (or failed to understand it). This is problematic because the coverage does not faithfully explain the study. It has suggested that faith based, sex negative abstinence approaches are better than other forms of sex education. And this study (and countless others) clearly show that’s not true.
So what’s the take home message?
This is a useful paper and a fair piece of research. It has limitations which means it can’t reliably be used to inform sex educational policy, but it would certainly benefit from adaption and replication and a longer follow up. Unfortunately the problem is less about the study (which is clearly discussed by the researchers) and more about how the media has misreported it and how politicians and faith based groups are misrepresenting the findings to suit sex-negative abstinence programmes.
Rather than falling into that trap we should take this research as further evidence that sex education is effective when it is tailored to the individual needs of different children; builds confidence; resists peer pressure; addresses feelings and emotions as well as infections and contraception; promotes delay until a young person is ready for intimacy (see also here and here); and prepares them for positive relationships when they are older.
Teaching abstinence makes teens delay sex? Here’s the evidence behind the media hypeRelated posts