Appendix C The evidence

This appendix lists the evidence statements from 4 reviews (3 settings-based reviews and a views review) provided by the public health collaborating centre (see appendix A) and links them to the relevant recommendations. (See appendix B for the key to quality assessments.) The evidence statements are presented here without references – these can be found in the full review (see appendix E for details). It also lists 5 expert reports and their links to the recommendations and sets out a brief summary of findings from the economic analysis and the fieldwork.

The 4 reviews of effectiveness are:

  • A review of the effectiveness and cost effectiveness of contraceptive services and interventions to encourage use of those services for socially disadvantaged young people: services and interventions in education settings.

  • A review of the effectiveness and cost effectiveness of contraceptive services and interventions to encourage use of those services for socially disadvantaged young people: views review.

  • A review of the effectiveness and cost effectiveness of contraceptive services and interventions to encourage use of those services for socially disadvantaged young people: services and interventions in healthcare settings.

  • A review of the effectiveness and cost effectiveness of contraceptive services and interventions to encourage use of those services for socially disadvantaged young people: services and interventions in community settings.

Evidence statement E1a indicates that the linked statement is numbered 1a in the review 'A review of the effectiveness and cost effectiveness of contraceptive services and interventions to encourage use of those services for socially disadvantaged young people: services and interventions in education settings'.

Evidence statement V1a indicates that the linked statement is numbered 1a in the review 'A review of the effectiveness and cost effectiveness of contraceptive services and interventions to encourage use of those services for socially disadvantaged young people: views review'.

Evidence statement H1a indicates that the linked statement is numbered 1a in the review 'A review of the effectiveness and cost effectiveness of contraceptive services and interventions to encourage use of those services for socially disadvantaged young people: services and interventions in healthcare settings'.

Evidence statement C1a indicates that the linked statement is numbered 1a in the review 'A review of the effectiveness and cost effectiveness of contraceptive services and interventions to encourage use of those services for socially disadvantaged young people: services and interventions in community settings'.

ER-IHYP indicates evidence in the expert report 'Improving healthcare for young people'.

ER-TPS indicates evidence in the expert report 'Teenage pregnancy strategy'.

ER-AHC indicates evidence in the expert report 'Access to health care: how do we reach vulnerable groups?'

ER-CSSDP indicates evidence in the expert report 'Contribution to NICE guidance on contraceptive services for socially disadvantaged young people'.

ER-DH indicates evidence in the expert report 'DH evidence: NICE guidance on contraception for socially disadvantaged young people'.

See also the full reviews, expert reports, economic analysis and fieldwork report. Where a recommendation is not directly taken from the evidence statements, but is inferred from the evidence, this is indicated by IDE (inference derived from the evidence).

Recommendation 1: evidence statements V15, V25, V26, ER-IHYP, IDE

Recommendation 2: evidence statements V2, V12, V15, V19, V22, V25, V26, ER-IHYP, IDE

Recommendation 3: evidence statements C1a, C1c, H4, V1a, V1b, V2, V11, V12, V15, V16, V19, ER-IHYP, IDE

Recommendation 4: evidence statements C2d, H1b, H4, E6a, V1a, V2, V12, V15, V19, ER-IHYP, IDE

Recommendation 5: evidence statements V2, V12, V14, V17, V18, ER IHYP, IDE

Recommendation 6: evidence statements C2a, H1b, E6a, V1a, IDE

Recommendation 7: evidence statements C2a, H1b, E6a, V1a, IDE

Recommendation 8: evidence statements E3a, V1a, V2, V11, V12, V15, V19, IDE

Recommendation 9: evidence statements H2, V1a, V1b, V1c, V11, V20, IDE

Recommendation 10: evidence statements H3, V1a, V1b, V7, IDE

Recommendation 11: evidence statements C1a, C1c, E6b, V1a, V2, V11, V17, IDE

Recommendation 12: evidence statements V1a, V1b, V2, V12, V14, V18, V19, V27, IDE

Evidence statements

Please note that the wording of some evidence statements has been altered slightly from those in the review team's report to make them more consistent with each other and NICE's standard house style.

Community review evidence statements

C1: Media based interventions

There is mixed evidence from 5 studies to suggest that media based interventions may reduce teenage pregnancy, increase contraceptive use and improve the knowledge and attitudes of young people in relation to these outcomes:

C1a. Computer based interventions

Moderate evidence from 1 randomised controlled trial (RCT) (++) showed that a computer based intervention could significantly reduce pregnancy and improve emergency hormonal contraception (EHC) use, as well as improving knowledge and attitudes based outcomes).

C1c. Social marketing campaigns

Weak evidence from 2 before-and-after studies (2 −) suggest that social marketing campaigns may have a significant effect on the use of contraception or EHC as well as knowledge and attitude based outcomes. In the first study (−), compared with the controls, participants in the intervention group were significantly more likely to have heard of EHC, know the mechanism of action of EHC, have discussed EHC with a care provider, received an advanced prescription for EHC, and intend to use EHC in the future if needed. The second study (−) showed that increased exposure to the social marketing campaign was associated with a significant increase in condom use at last sexual experience.

C2: Interventions to prevent repeat pregnancy

There is inconsistent evidence from 8 studies to suggest that community based interventions may be effective in preventing repeat pregnancy:

C2a. Home visitor interventions

Inconsistent evidence from 3 RCTs (2 ++, 1 +) suggests that home visitors may be effective in preventing repeat pregnancy; only 2 of the 3 studies measured repeat pregnancy rate as an outcome and only 1 of these provided evidence of clear benefit. The first RCT (++) showed a significant reduction in repeat birth for the intervention group. The second RCT (+) showed a significant improvement in parenting scores for the intervention group, but the effect on repeat pregnancy was not significant. The third RCT (++) showed a significant improvement in contraceptive use for the intervention group, but did not measure repeat pregnancy.

C2d. Generic programmes for teenage mothers

Moderate evidence from 1 RCT (+) suggests that generic programmes for teen mothers (to prevent repeat pregnancy, increase school retention, reduce substance abuse, and improve wellbeing) could be effective in significantly reducing repeat pregnancy and consequent births.

Healthcare review evidence statements

H1: Interventions to provide new adolescent services and to encourage access to existing services
H1b. Outreach to existing mainstream services

Moderate evidence from 5 studies (2 +, 3 −) suggests that outreach programmes to encourage young people to attend mainstream sexual health services may be effective in increasing service use, but the effect on reducing teenage pregnancy rates is unclear. In the non-RCT study (+), compared with control, the outreach group was significantly more likely to likely to report consistent contraception use, and women were also less likely to report pregnancy. In the first cohort study (+) condom use increased and pregnancy decreased, but the impact of the intervention is unclear because of poor reporting. In the second cohort study (−), during the 5 years of the intervention, the number of attendees at family planning clinics aged under 20 and under 16 significantly increased. Pregnancy is reported to have 'remained low' but no data is given. In the third cohort study (−), those who attended an orientation session were significantly more likely to start using services, and attendance at the 3-month booster session was associated with significantly higher continued clinic contact at 1 year. In the interrupted time series study

(−), the number of new users of family planning services aged under 26 years increased significantly in the first 18 months of the outreach programme.

H2: Advance supply of emergency hormonal contraception

There is strong evidence from 4 RCTs (3 ++, 1 +) to support the advance provision of EHC to young people to increase EHC use. In most cases increased use was not at the expense of other contraceptive use, and did not promote risky sexual behaviour; the exception was 1 study (+) with adolescent mothers. In the first study (++), at 6-month follow up EHC use was significantly higher in the intervention (advanced provision) group than the control, and the mean time to use EHC was significantly shorter in the intervention group compared with the control group. There were no differences in hormonal contraception or condom use between the groups. In the second study (++) (with random allocation to receive EHC via pharmacy, clinical access or advance provision) EHC use at 6-month follow up was significantly greater in the advance provision group than the clinical access group. Pharmacy access did not affect EHC use when compared with clinic access. In the third study (++), the advance EHC group reported (non-significantly) higher emergency contraception use and significantly sooner use. In the fourth study (+), at 12-month follow up those in the advance provision group were significantly more likely than the controls to have used EHC, but also more likely to have had unprotected sex in the past 6 months.

H3: Interventions to promote adolescent condom use

There is strong evidence from 5 studies (4 ++, 1 −) to support interventions that combine discussion and demonstration of condom use to increase adolescent condom use and engagement with clinical services. In the first study (++), at 6 month follow up intervention subjects reported statistically significant increase in condom use by their sexual partner for protection against STIs. In the second study (++), at 1 year clients were twice as likely to report having received condoms from the clinic. In the third study (++), of 2 methods of cognitive behavioural therapy (CBT) to reduce unprotected sex, those in the skills-based CBT group were significantly less likely to have unprotected sex at 12 months than those in the information-based CBT group or control group. In the fourth study (++), more of the intervention group than the comparator group returned for their scheduled clinic revisits (statistical significance not clear). In the fifth study (−) it is suggested that, compared with the rest of the country, attendance at the GUM clinic by young people is much higher, particularly at sites offering daily access and located geographically close to a school (no statistical data are given to validate this).

Although the studies were mostly well designed, the data were not always well analysed and reported, which may have affected reliability. Applicability in the UK may also be limited because most of the studies were conducted in the USA/Canada (2 in populations that were majority black American and 1 population who were African American/Latino).

H4: Adolescent contraceptive use

Strong evidence from 2 RCTs (2 ++) and 1 non-RCT (+) suggests that interventions aimed to improve adolescent contraceptive use by additional service provision can be effective, but this depends on the intervention. The first study (++) was of a nurse led one-to-one intervention, the intervention group reported significantly greater oral contraception adherence than the controls. The second study (+) was of a computer based contraception decision aid intervention. At 1-year follow up the first intervention group had significantly higher contraception knowledge and (non-significantly) fewer pregnancies than the non-intervention group. This finding was not replicated in a second study population. The third study (++) was of an intervention to administer 'quick start' of contraception (immediately administered contraceptive injection), at 6-month follow up there were no differences in continuation rates or pregnancy rates between the groups.

Education review evidence statements

E3: School based health centres
E3a. On site dispensing

Strong evidence from 3 papers (2 +, 1 −) supports the direct provision of contraceptives dispensed on site from school based health centres as a way to increase contraceptive provision. However, the use of those contraceptives or any subsequent outcomes is unclear. In the first study (+), significantly more of the intervention cohort selected hormonal contraception at the first or second visit than the control cohort, and were also significantly less likely to select no contraception. In the second study (+), adolescents in the intervention group were significantly more likely to receive condom/HIV instruction, and significantly less likely to report lifetime or recent sexual intercourse. Sexually active adolescents in the intervention group were twice as likely to use condoms but less likely to use other contraceptives. In the third study (−), direct provision led to a statistically significant increase in the number of contraceptives prescribed to young people. The data analysis in this paper is poor, giving only percentage increases, but it does appear to indicate that on site dispensing increases contraceptive provision.

E6: Curriculum interventions with additional components
E6a. Community outreach

Strong evidence from 3 studies (3 +) suggests curriculum interventions that include community outreach components can be effective in preventing teenage pregnancy and risky sexual behaviour. In the first study (+), rates of pregnancy, along with rate of school failure and academic suspension, were significantly lower in the Teen Outreach group than the control group. In the second study (+) Teen Outreach was again shown to be effective, especially for those who were already teen parents. In the third study (+) Reach for Health participants were significantly less likely than controls to report sexual initiation or recent sex.

E6b. Virtual world intervention

Moderate evidence from 1 study (+) suggests that a virtual world intervention was effective when associated with a curriculum based intervention about sexual risk behaviour. The intervention group had significantly better understanding than the control group of how reproduction works and the possible consequences of sex, and of the importance of behaving in ways that limit sexual experience.

Views review evidence statements

V1. Lack of knowledge
V1a. Gaps in knowledge about sexual activity

Three qualitative studies (1 ++, 1 +, 1 −) describe a lack of knowledge among young people about potential consequences of sexual activity. One paper covering interviews with 16–21 year olds as part of a mixed method study (−) describes a lack of knowledge before first sexual experience and lack of knowledge about the consequences of sexual activity. This was echoed in interviews with 16–23 year olds from black and ethnic minority groups, who reported a lack of knowledge about risky sexual activity (+). Also, interviews with young mothers aged 14–16 years reported gaps in their knowledge about becoming pregnant and abortion (++).

V1b. Gaps in knowledge about use of contraception

Three qualitative studies (2 ++, 1 −) describe a lack of knowledge about correct use of contraception among young people. Gaps in knowledge about aspects of contraception were reported in young mothers aged 14–16 (++), in a mixed group of 16–25-year-old women (++) and in a mixed group of 15–18 year olds (−). One qualitative study (++) suggests that a lack of knowledge about contraception methods may be greater in young people from deprived areas and found that lack of knowledge regarding contraception methods was greater in socially disadvantaged young women aged 16–20.

V1c. Gaps in knowledge about emergency hormonal contraception

One qualitative interview study (++) highlights emergency hormonal contraception as an area of particular lack of knowledge among young women aged 16–25. Survey data suggest knowledge of emergency contraception in 78–90% of school aged girls. One survey linked less knowledge of emergency contraception with being a pupil at a school with lower academic achievement.

V2: The obstacle of embarrassment
V2a.Embarrassment about discussing sex

One qualitative study (++) reports that discussion of sex and contraception is embarrassing. A study of mixed young city dwellers aged 16–25 reported that the younger participants reported that discussing sex or any type of contraception was embarrassing.

V2b. Embarrassment about using contraceptive services

The potential for feelings of embarrassment to inhibit young people from using contraceptive services is outlined in 7 papers (1 ++, 5 +, 1 −) reporting views from a variety of groups of young people. Clients of family planning clinics describe embarrassment or stigma associated with accessing contraceptive supplies (++). Young people from ethnic minorities also describe embarrassment if they are seen accessing a service (−). At a male drop-in service, 66% of clients reported that embarrassment would stop them using a service. Young people of school age (2 +) echo this embarrassment about accessing services. One survey reports 20–24% of 11–39 year old women had been embarrassed, scared or concerned about using a sexual health service. Another paper (+) describes women of 16–25 years old feeling embarrassed when using contraceptive services. Mixed groups of young people described embarrassment as a barrier to obtaining and using condoms (+). The importance of clinics overcoming young people's feelings of embarrassment was also recognised by staff (GPs and nurses) (2 +).

V2c. Embarrassment about particular services

Two papers (1 +, 1 ungraded survey) report embarrassment related specifically to particular services. One (+) reports that young people aged 14–25 perceive that at times teachers are clearly embarrassed when discussing sexual issues, leading to the young people also feeling embarrassed. The other states that 63% of young women and 46% of young men aged 15–16 years reported embarrassment about attending a consultation with a GP in regard to sexual health.

V2d. Embarrassment at reception

One study (+) describes a particular aspect of accessing a service that is embarrassing. It reports that young people aged up to 24 feel embarrassed when giving their name and address at a reception desk.

V7: Views of condoms

Three studies (1 ++, 1 +, 1 −) suggest that condoms can be perceived negatively, as uncomfortable or a barrier to intimacy, among some teenagers. Two (1 ++, 1 +) report these negative views among teenagers aged 14–15 and teenagers including those who were young mothers or pregnant, and another study (−) reports a mix of positive and negative perceptions of condom use among 12–13 year olds and 16–17 year olds. Four studies (1 ++, 1 +, 2 −) suggest some young people think that there are negative connotations for young women carrying condoms.

V11: Knowledge of local services

Three studies (1 ++, 2 +) describe uncertainty among young people about where to go to access contraceptives, especially among young men and younger participants.

V12: Perception of trust in services

Five papers (1 ++, 2 +, 2 −) describe the importance of young people perceiving that contraceptive services are trustworthy and legitimate, enabling them to feel confident, and being in control when using them.

V14: Concerns regarding GP-based services

Five studies (1 ++, 4 +) report that some young people have concerns about attending a GP practice for contraceptive services because of a perceived potential loss of confidentiality. This seems to be a particular concern in rural communities.

V15: Accessibility of services

Eleven studies (5 ++, 5 +, 1 −) suggest the importance of accessibility of services for young people, with convenient location, extended opening hours, and choice in location as important elements.

V16: Appointment systems

Studies report varying views about whether an appointment system or a drop-in service provides greater accessibility for young people. Four (2 ++, 1 +, 1 −) suggest an appointment-free system offers convenience. However, 1 (+) reports that staff perceive that waiting times in a clinic are not an obstacle to accessibility. One survey of young people reported that 62% would prefer a walk-in service. Another survey suggested that young people may appreciate the option of making appointments by telephone.

V17: The importance of anonymity

Eight studies (1 ++, 6 +, 1 −) report that preserving anonymity when accessing services is a significant concern for young people. These concerns regarding anonymity are also perceived by staff (1 ++, 3 +, 1 −).

V18: The importance of confidentiality

Eleven papers (3 ++, 6 +, 2 −) report that confidentiality is a key concern for young people in accessing a sexual health service. Concerns regarding confidentiality feature particularly in regard to rural areas and GPs.

V19: The importance of respectful and non-judgemental staff

A range of qualitative studies and survey data highlights that young people value staff who have a respectful and non-judgemental attitude towards them.

Five papers (3 +, 1 −, 1 ungraded survey) report that staff also recognise the importance of being non-judgemental. However, they highlight that some staff may have ambivalent or varying attitudes towards young people and sexuality.

V20: Concerns regarding cost

Three studies (1 ++, 2 +) report that the cost of contraception is a concern for some young people.

V22: Clinic atmosphere

Four studies (1 ++, 2 +, 1 −) provide evidence from young people regarding the importance of a comfortable and welcoming atmosphere in sexual health service premises. This is echoed in a study of staff views.

V25: Availability of resources

There is evidence from 5 studies (1 ++, 1 +, 3 −) that staff have concerns regarding limited availability of resources for sexual health services.

V26: Agencies working together

There is evidence from 6 studies (3 ++, 1 +, 1 −, 1 ungraded survey) that staff perceive that well-organised services, and different agencies working together effectively, are important.

V27: Staff training

There is evidence from 6 studies (2 +, 4 −) that staff perceive a need for greater training in providing contraceptive services for young people.

Expert report/s

  • Improving healthcare for young people

  • Teenage pregnancy strategy: NICE meeting: 17 September

  • Access to health care: how do we reach vulnerable groups? Learning from the teenage health demonstration sites

  • Contribution to NICE guidance on contraceptive services focusing on socially disadvantaged young people

  • Department of Health evidence: NICE guidance on contraception focusing on socially disadvantaged young people

Cost-effectiveness evidence

The economic analysis indicates that, from a public sector perspective, providing contraceptives in schools and colleges is cost effective and results in net cost savings compared with no provision of contraceptives in these places. This result is robust to changes in the key model assumptions if the costs of government-funded benefits are included within the analysis. However, if government-funded benefits are excluded from the analysis, providing contraceptives within schools and colleges, while still being cost effective, has around a 50% probability of resulting in net cost savings.

The analysis also suggests that providing hormonal contraception within schools and colleges is likely to be more effective than providing condoms in terms of preventing pregnancies. This may also lead to greater cost savings than dispensing condoms. However, this comparison is subject to considerable uncertainty.

The economic analysis also suggests that, from a public sector perspective, intensive follow‑up and support after a teenage pregnancy results in a cost of £4000 for every repeat teenage pregnancy averted. This is in comparison with no follow-up after a teenage birth. Excluding government-funded benefits from the analysis leads to an estimated cost per repeat teenage pregnancy averted of £15,000.

From a public sector perspective, advance provision of emergency hormonal contraception is estimated to be more effective and less costly than not providing it in advance. However, when government-funded benefits are excluded from the analysis (that is, an NHS and personal social services perspective is adopted), the intervention is estimated to cost £310 per pregnancy averted among those aged 15–19, compared with no advance provision.

Finally, the analysis suggests that providing emergency hormonal contraception in advance is likely to be cost saving from a public sector perspective, when provided within schools and colleges alongside other contraceptives These results are informed by the following:

  • The Teenage Pregnancy Strategy's target to halve the under‑18 conception rate by 2010.

  • It has been assumed that before conception, the value of a future baby to society is neither positive nor negative. From this, it is clear that preventing conception cannot be measured in QALY terms, because future QALYs do not exist before conception. Thus the cost effectiveness of preventing a conception has been measured in terms of cost per pregnancy averted. However, once conceived and born, the baby is invested in a life expectancy, so that the loss of such a baby after birth can be measured as a loss of QALYs. A government-funded benefit given to young mothers can either be regarded as a transfer payment (from taxpayers to young mothers) or as a real resource cost. If it is seen as a transfer payment, the benefit to the mother is an equivalent cost to the taxpayer and these items cancel out. However, if the contraceptive intervention prevents a baby from being born, the money can be used by the government for other purposes without any opportunity cost. Following this logic, having fewer teenage births results in a much greater cost saving than if such benefits are considered as a transfer payment.

Fieldwork findings

Fieldwork aimed to test the relevance, usefulness and feasibility of putting the recommendations into practice. The PDG considered the findings when developing the final recommendations. For details, see the fieldwork section in appendix B and the full fieldwork report.

Fieldwork participants who work with socially deprived young people were very positive about the recommendations and their potential to help improve contraception service provision. Many participants stated that the recommendations represented best practice in the area, and although they did not offer an entirely new approach, they agreed that the measures had not been implemented universally. They believed wider and more systematic implementation would be achieved as a result of this guidance.

  • National Institute for Health and Care Excellence (NICE)