Hum. Reprod. Advance Access originally published online on June 3, 2006
Human Reproduction 2006 21(9):2296-2303; doi:10.1093/humrep/del168
| ||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||
Specialist contraceptive counselling and provision after termination of pregnancy improves uptake of long-acting methods but does not prevent repeat abortion: a randomized trial
1 Lothian NHS Family Planning and Well Woman Services 2 Department of Clinical Science and Community Health University of Edinburgh, Edinburgh, Scotland and 3 Department of Public Health and Policy, University of London School of Hygiene and Tropical Medicine, London, UK
4 To whom correspondence should be addressed at: NHS Lothian Family Planning and Well Woman Services, 18 Dean Terrace, Edinburgh EH4 1NL, Scotland. E-mail: anna.glasier{at}lpct.scot.nhs.uk
 |
Abstract |
|---|
 Top Abstract IntroductionMaterials and methodsResultsDiscussionAcknowledgementsReferences |
|---|
BACKGROUND: One in four abortions in the UK is undertaken for women who have had one before. Women undergoing abortion in Edinburgh were targeted for improved contraceptive advice and provision in this randomized trial. METHODS: Between November 2001 and May 2002, women recruited at assessment for abortion were randomized at admission to receive specialist contraceptive advice and enhanced provision (316 women) or standard care (297 women). Randomization was based on the week of admission. Contraceptive use 16 weeks after abortion was assessed by questionnaire and subsequent abortions by review of the hospital records 2 years later. RESULTS: Women receiving specialist advice and enhanced provision were more likely to leave the hospital with contraception (271 versus 115, P < 0.001), which was more likely to be a long-acting method (141 versus 78, P < 0.001) than women receiving standard care. Four months later, there was no significant difference in contraceptive prevalence or continuation, but women in the intervention group were more likely to be using contraceptive implants (32 versus 6, P < 0.001). Two years later, 14.6% of women in the intervention group (44/302) and 10% of controls (27/268) had undergone another abortion in the same hospital (P = 0.267). CONCLUSIONS: Specialist contraceptive advice and enhanced provision had a short-lived effect on contraceptive uptake and increased the use of long-acting methods but did not appear to reduce repeat abortions.
Key words: abortion/contraception/intervention/long-acting/repeat abortion
|
Introduction |
|---|
|
TopAbstractIntroductionMaterials and methodsResultsDiscussionAcknowledgementsReferences |
|---|
Every year in the UK, just under 200 000 pregnancies end in therapeutic abortion (Department of Health, 2004
; Information and Statistics Division, Scottish Health Service, 2004). Most women having an abortion will have had unprotected intercourse or will have been using a contraceptive method associated with high failure rates when used typically, mainly condoms or pills (Garg et al., 2001; Jones et al., 2002). Methods of contraception which are independent of daily compliance for their effectiveness (implants, intrauterine methods and injectable contraceptives) are associated with very low failure rates even with typical use (Trussell, 2004) but are used by <8% of women in the UK (Dawe and Rainford, 2004). Moreover, for all methods of contraception, discontinuation rates are disappointingly high (Rosenberg et al., 1995; Grigorieva et al., 2005).
In England and Wales, more than one in four women will have an abortion at some time in their lives, and over 31 women in every 1000 aged 18–24 years will have an abortion in 1 year (Department of Health, 2004). Identifying women at risk of unintended pregnancy to target them for intensive contraceptive advice might offer a strategy for reducing abortion rates. In the UK, 20–25% of abortions each year are done for women who have had a previous pregnancy terminated (Department of Health, 2004; Information and Statistics Division, Scottish Health Service, 2004). However, advice about contraceptive use after abortion can be cursory, and many women leave abortion services with just condoms or with no contraceptive method (Penney et al., 1994; Kumar et al., 2004). Identifying women undergoing termination of pregnancy as a target for better contraceptive counselling, we undertook a randomized trial to determine whether specialist advice and improved provision of contraception result in increased uptake of more effective methods of contraception and improved contraceptive continuation. We undertook a case-note review 2 years later to determine the number of women undergoing repeat abortion.
|
Materials and methods |
|---|
|
TopAbstractIntroductionMaterials and methodsResultsDiscussionAcknowledgementsReferences |
|---|
From November 2001 to May 2002, all women presenting to the abortion clinic at the Royal Infirmary of Edinburgh were invited to participate in a survey of contraception after the termination of pregnancy. Routine management of abortion in the hospital offers women who present before 9 weeks’ gestation a choice of surgical or medical abortion. Second trimester termination of pregnancy (>13 weeks gestation) is performed medically. Women undergoing termination for fetal abnormality, those whose command of English was poor and a few women thought by the nursing staff who were distributing the questionnaire to be too distressed to be approached, were excluded from the study. During the 7 months of recruitment to the study, calendar weeks were randomized by the statistician using random number tables. The unit staff was informed of the allocation verbally by the investigator (C.S.) on the first day of each working week. Women admitted for abortion during an ‘intervention week’ were offered expert advice and enhanced provision of contraception (intervention group), whereas those admitted during a ‘control week’ received the standard care available at the time of the study (control group). Routine management of women being referred for abortion comprised of an initial consultation with the gynaecology clinic followed by admission to the hospital the following week for the abortion procedure. Patient’s consent to participate in the study was obtained by the nurses at the initial presentation to the gynaecology clinic. These nurses, who also arranged for the women to be admitted to the unit the following week to have the pregnancy terminated, were blinded to the randomization.
Standard care comprised of a brief discussion about contraception at the initial outpatient consultation before the abortion procedure was arranged and further discussion by the nursing staff before the women left hospital after the abortion. Women requesting oral contraception were offered only one packet of combined or progestogen-only pill, and only one brand of each was available. Copper intrauterine devices (IUD) or levonorgestrel-releasing systems [LNG-intrauterine system (IUS)] could be inserted at the time of surgical, but not medical, termination. Similarly, Depo Provera was administered before discharge from the hospital following surgical termination of pregnancy (STOP) but only if products of conception had been identified at the time of medical termination of pregnancy (MTOP). Contraceptive implants were not provided. Condoms and written information about all the methods of contraception were freely available. Routine follow-up at the hospital 2 weeks after the procedure was arranged for women undergoing MTOP, but women choosing surgical abortion were advised to see their general practitioner at this time.
Women consenting to take part in the study and admitted for abortion during an intervention week underwent the same procedures, including brief discussion about future contraception during the initial consultation and assessment in the outpatient clinic. They were subsequently interviewed either pre-operatively before STOP or whilst in the ward undergoing MTOP by a doctor with specialist training in contraception. A few women undergoing surgical abortion were interviewed post-operatively if there was no time to see them before the procedure. Interviews lasted 
15–20 min. Demographic details, educational qualifications and a full reproductive history, including ever-use of contraception and reasons for discontinuation, were recorded. Detailed information about contraceptive use at the time of conception was sought. The preferred method of contraception for use after abortion was discussed, and a supply, sufficient for 3 months, of the chosen method (and brand if relevant) was dispensed. Depo Provera was injected, or implants inserted, before discharge from hospital. An IUD/IUS was inserted in the operating theatre following STOP but not immediately after MTOP; rather, firm arrangements were made for insertion at a local family planning clinic (FPC) after routine hospital follow-up at 2 weeks. Condoms and written information were provided as required. The differences between the two groups in contraceptive provision after the abortion are summarized in Table I.
|
During intervention weeks, women undergoing STOP were encouraged to see their referring doctor 2 weeks after the procedure; women undergoing MTOP were given an appointment for follow-up at the hospital at this time. A second follow-up appointment at 12 weeks was recommended for all women treated during intervention weeks regardless of the method of pregnancy termination. Women wishing to attend their general practitioner or a dedicated youth clinic were advised to return to that provider for further supplies of contraception. Women attending the FPC had an appointment formally arranged for them by telephone before they left hospital.
The primary outcomes of the study were contraceptive method uptake and continuation rates at 16 weeks after termination of pregnancy. All participants (whether admitted during control or intervention weeks) were contacted at 16 weeks by post or by telephone (according to their stated preference at the time of recruitment). A questionnaire (self-administered if mailed or interviewer administered if telephoned), comprising 11 questions, sought information about follow-up consultations and contraceptive use. Three reminders with stamped addressed envelopes were posted at 2- to 4-week intervals to non-responders. For telephone follow-up, a message was left at the first unanswered call (if there was an answer phone), and three calls at different times of the day on different days of the week were made.
Two years after the start of the study, the case notes of all the participants were reviewed to determine how many women in each group had undergone another termination of pregnancy in the same hospital (the secondary outcome of the study).
The study protocol, including the case-note review, was approved by the Lothian Research Ethics Committee. All participants gave informed written consent to participation, and no reimbursement or payment was received.
Data collection, management and statistical analysis
For women admitted during control weeks (who received standard care), and for women who declined to take part in the study, baseline data were collected from the hospital case records. Although the records were structured, not all sections were routinely completed, and some data were missing. For women undergoing abortion during intervention weeks, baseline data were collected from case records and interviews. All baseline and follow-up data were coded and entered onto an excel database.
Deprivation category was calculated from the postal codes of each individual’s residential address (Carstairs and Morris, 1991). The Carstairs Score is a summary measure of relative deprivation between populations contained within small geographical localities. It is based on the proportion of overcrowded households, residents of lower social class (IV and V), male unemployment and non-owner-occupied properties.
A sample size of 600 women was calculated to provide 90% power to detect a difference of 15% in the percentage of women using contraception in each group at 16-week follow-up, significant at a 5% level, if the follow-up rate was around 50% and assuming an uptake of contraception in each group of at least 50% (based on previous internal audits). As the clusters in this trial were dependent on weeks rather than different locations or health care providers, the assumption was made that there would be very little variation between clusters and accordingly a low intra-cluster correlation coefficient with minimal loss of power. The above calculation was therefore computed as if the trial was individually randomized, with no inflation factor.
Associations between factors at baseline were investigated using Chi-squared tests and the Mantel–Haenszel test for linear association. Comparisons between outcomes in the two randomized groups took account of the cluster randomization by calculating weekly averages that were analysed with two-sample t-tests. McNemar’s test was used to examine changes in contraceptive use at different times.
|
Results |
|---|
|
TopAbstractIntroductionMaterials and methodsResultsDiscussionAcknowledgementsReferences |
|---|
A total of 1151 women were assessed for participation in the study; 613 were randomized (Figure 1). Mean age was similar between the three groups, 24 years for women in the intervention and control groups and 26 years for women who did not participate. There were no differences in age distribution nor deprivation score between women randomized to control or intervention weeks and those who declined or were not asked to participate (Table II). The mean age at the completion of full-time education was only available for women in the intervention group (17.7 years); 33.5% (n = 106) had left school at or before the age of 16 years, and 22% (n = 69) were still studying. The mean age of first intercourse in the study group was 16.7 years.
|
|
Contraceptive uptake
Contraceptive uptake before and after the abortion
For all three groups of women, the contraceptive method being ‘used’ in the cycle when conception occurred was compared with the method provided in hospital for use after the abortion as documented in the case notes at discharge, that is, method uptake. The data are shown separately for women undergoing surgical abortion in Table III and for medical abortion in Table IV. When the method of contraception the women claimed to have been using in the cycle the pregnancy was conceived was compared, there were no differences between the three groups of women who chose surgical abortion. There were two differences between the groups among women having MTOP (Table IV). Women randomized to the intervention group were more likely to have been ‘using’ NFP than women in the control group (P < 0.045) and less likely to have been using no contraception than women who declined to participate in the study (P < 0.016).
|
|
There were a number of differences among and between the groups when the method being ‘used’ during the cycle of conception was compared with the method chosen for use after the abortion (contraceptive uptake) (Table V for STOP and Table VI for MTOP). Most obviously, women in all three groups, regardless of the method of abortion, were more likely to choose oral contraception and less likely to choose barrier methods for use after the abortion (P < 0.001 for all comparisons). When compared with women in the control group, women in the intervention group were more likely to choose a contraceptive implant {P < 0.001 for both STOP [95% confidence interval (CI) 2.7–13.2] and for MTOP (95% CI 7.9–15.4)}. They were also less likely to leave hospital with no method [P < 0.001 (–14.8 to –4.2) following STOP and P < 0.01 (–14.2 to –0.7) for MTOP]. No woman in the intervention group chose to be sterilized following termination of pregnancy.
|
|
Contraceptive use at 16-week follow-up
There was no difference in the response rate to the follow-up questionnaire—63% of women in the intervention group (n = 199) compared with 60% of controls (178 women). Women who responded were more likely to be nulliparous (86%, 324 women versus 77%, 181 women), less likely to have completed their education by age 16 years (34 versus 63%) and more likely to have completed tertiary education (31%, 53 women versus 11%, 35 women), and less likely to have had a previous abortion (16%, 60 women versus 31%, 64 women) than women who did not respond. All these differences were statistically significant (P < 0.001). There were no significant differences between responders and non-responders in age, deprivation status or method of abortion (medical or surgical) chosen.
There was no difference between the control and intervention groups in the number of women at 16 weeks after abortion reporting attendance at some sort of follow-up (79% in each group).
Of those attending for follow-up, 88% of women (161) in the intervention group and 89% (140) in the control group claimed to be using a method of contraception 16 weeks after abortion. In both groups, only 3% of women in a sexual relationship were using no contraception. The distribution of contraceptive methods used at 16 weeks after the abortion for women who attended the follow-up is summarized in Table VII.
|
At 16 weeks after the abortion, women in the intervention group were significantly more likely to be using a long-acting method of contraception (implants, injectables and IUD/IUS) than women in the control group [63 women (37%) versus 38 women (26%); P = 0.04]. Thirty-two women in the intervention group (19%) were using a contraceptive implant compared with only six women (5%) in the control group (P < 0.001).
For 142 women in the intervention group and 127 in the control group, data were available for both contraceptive method provided at discharge from hospital and the method being used at the 16-week follow-up (Table VIII). There were no statistically significant differences in continuation rates of any method of contraception between women receiving expert individualized contraceptive advice and those receiving standard care. Implanon was not available to women in the control group in the abortion clinic; women wanting to use Implanon had to arrange insertion after they had left hospital. Only one woman in the control group said that she wanted to use an implant for contraception after the abortion; in fact at 16 weeks, she was using the combined pill. The continuation rate for Implanon among women in the intervention group was 95% at 16 weeks.
|
The case notes of 570 women were retrieved 2 years after the abortion, 302 from the intervention group (95.6%) and 268 from the control group (90%). Forty-four women (14.6%) in the intervention group and 27 women (10%) in the control group had experienced at least one further unintended pregnancy which ended in therapeutic termination within 2 years of their earlier abortion. The difference between the two groups was not statistically significant (Chi-squared tests for trend P = 0.267, linear-by-linear association P = 0.122).
|
Discussion |
|---|
|
TopAbstractIntroductionMaterials and methodsResultsDiscussionAcknowledgementsReferences |
|---|
This was a complicated trial with a number of obvious shortcomings. Baseline data were collected from women in the intervention group at interview and so are much more accurate and complete than those collected from the control group and from women unwilling to participate (which came from scrutiny of the case notes). We believe that our method of randomization has overcome any major bias. Follow-up was limited to 16 weeks, and almost 40% of women in both groups did not respond to the 16-week questionnaire. Moreover, there were demographic differences between responders and non-responders, which might suggest that the responders were more likely to use contraception effectively. Case-note review was used to obtain data on repeat abortions 2 years after the study, and the results are likely to be an underestimate because we only identified women who had undergone repeat abortion in the same hospital and some may have had another pregnancy terminated elsewhere.
Randomized trials of interventions designed to improve contraceptive use are scarce. In a systematic review of the US literature, Moos et al. (2003) examined 354 articles on counselling in a clinical setting to prevent unintended pregnancy and concluded that ‘no robust evidence exists to determine effective counselling approaches for changing knowledge, attitudes or behaviour’. Only 13 articles reporting experimental or cohort designs, only one of them a randomized trial, were included. The authors of the review had many concerns about the design of most of the studies. Interventions reported in the literature often have short periods of follow-up, and sustainability of any measurable effect is not addressed; outcomes, whether contraceptive use or pregnancy, are often self-reported; and if long-term follow-up is attempted, loss to follow-up is high. It is difficult to follow-up young people from a mobile population for long, particularly if recruited at the time of an abortion. In our clinics, 50% of women given a routine follow-up appointment within 4 weeks after abortion fail to attend. We tried to address the problem by combining an achievable prospective follow-up period with later case-note review. We chose 16 weeks as duration of this study because women given 3 months’ supply of oral contraceptives or an injection of Depo Provera would have needed to attend a doctor before then to continue the method. Our intervention was successful on two counts; we increased the uptake of contraception (any method) immediately following the abortion, and we increased the uptake of long-acting methods [among the most effective methods (Steiner et al., 2003)]. By 16 weeks, however, equal numbers of women in the control group were using a method of contraception, and our intervention had made no difference, compared with the standard treatment, to either attendance for follow-up or continuation rate of the contraceptive method chosen. Women in the intervention group were more likely to be using long-acting contraception; however, implant use accounted for most of this difference, and because few providers outside community FPCs in Edinburgh remove implants (Fleming et al., 1998), the discontinuation of the method involves the user in more effort, so high continuation rates after only four months should not be surprising. Overall, then, the primary outcome of the trial was encouraging because the intervention did result in improved uptake of a highly effective long-acting method of contraception.
However, while not the primary end-point of the study and while we were unable to review the case notes of all the women, the intervention appeared to have no effect on the rate of repeated unwanted pregnancy. The number of women having a repeat abortion within 2 years, with or without the intervention, is disappointing. Intervention trials with pregnancy rates as an end-point are even harder to find. Three studies in the Moos et al. review (2003) included pregnancy as an end-point (Namerow et al., 1989; Adam et al., 1990; Winter and Breckenmaker, 1991); none was randomized, and only one showed a statistically significant reduction in pregnancy rates and that was not sustained at 1 year of follow-up (Namerow et al., 1989). Pregnancy is a relatively rare event, and sample sizes must be large, follow-up long and rates of loss low. These studies are prohibitively expensive. We doubt whether adequate follow-up would be achievable in a cohort of women undergoing abortion, and case-note review, with all its shortcomings and biases, was the best that we could do to measure an effect of the intervention on abortion rates. We have been able to find only one other intervention that has recently been systematically tested with pregnancy rates as an end-point. Five trials testing advanced provision of emergency contraception included pregnancy as an end-point (Glasier and Baird, 1998; Glasier et al., 2004; Lo et al., 2004; Raine et al., 2005; Hu et al., 2005). In only two was the demonstration of a reduction in pregnancy a primary aim (Glasier et al., 2004; Hu et al., 2005), although in retrospect both were probably underpowered. None demonstrated an effect of advanced provision of emergency contraception on pregnancy rates.
Many factors influence sexual behaviour which clearly changes with time and has tended recently to become more ‘risky’ (Wellings et al., 2001). Factors well beyond the influence of health professionals determine when couples want to have children and how readily they will decide to have a pregnancy terminated rather than continue it. Pregnancy is commonly unintended among young women with few career aspirations from communities that are socially or economically excluded (UNICEF, 2003). The average woman in the UK will start to have sex while she is still a teenager, delay her first pregnancy until the age of 28, have fewer than two children, will continue to work while bringing them up and not reach the menopause until the age of 51 (Wellings and Diamond, 2005). Even a woman who is highly motivated to avoid unwanted pregnancy will have to use contraception for more than 30 years—allowing plenty of time for her concentration to lapse. Numerous factors determine compliance with, and continuation of, contraception, including reservations about the method and its side effects, partner support of the method, women’s beliefs about their own fertility and ambivalence about becoming pregnant. We know nothing about how women in our trial used contraception beyond 16 weeks after the abortion whether they went on to have another abortion or not. The lack of such data from UK women has been highlighted as a research priority in the NICE Guideline of Long-acting Methods of Contraception (NICE, 2005).
The rigorous demands of evidence-based medicine—and the desire to publish results in high-impact journals—lead us all towards testing one tightly controlled intervention which, because of the difficulties discussed earlier, is necessarily limited. Some interventions which work in some settings (Hubacher et al., 2000) do not work in others (Smith et al., 2002). Some do not work at all (Keder et al., 1995). It is unlikely that any single health-related intervention, and probably not even a package of interventions, will make a measurable difference to unintended pregnancy rates. UNICEF (2003) recommends improving ‘enabling’ sex education as well as contraceptive availability if teenage pregnancy is to be reduced. However, two studies of sex education packages tested in randomized controlled trials in UK schools (Graham et al., 2002; Wight et al., 2002) changed attitudes and improved knowledge but had no measurable effect on sexual behaviour, whereas a third study changed only one small element of behaviour (Stephenson et al., 2004). It is probably unrealistic to expect that a 20-min discussion in a clinic or a handful of lessons at school will change behaviour, given the myriad of other far greater influences in our lives.
The results of this study may not, of course, be representative of what might happen if a similar intervention was made in another country, particularly a less-developed country. However, a randomized trial of pre-abortion counselling and its effect on post-abortion contraceptive use was undertaken in Sweden recently (Bender and Geirsson, 2004), and in that setting, pre-abortion counselling about contraception made no difference to patterns of contraceptive use.
Though we should not give up trying to find ways to change sexual behaviour and to show that they work, we will probably not have much effect without major changes in society.
|
Acknowledgements |
|---|
|
TopAbstractIntroductionMaterials and methodsResultsDiscussionAcknowledgementsReferences |
|---|
The authors particularly thank the staff of the abortion unit of the Royal Infirmary of Edinburgh for help in recruiting the subjects, Dr Rob Elton for help with the analysis, Fiona Cowan for administrative support and Mrs Anne Mayo for help with data entry. The study was funded by a grant from the Scottish Executive for the Scottish Health Demonstration Project Healthy Respect.
|
References |
|---|
|
TopAbstractIntroductionMaterials and methodsResultsDiscussionAcknowledgementsReferences |
|---|
Adam BN, McAnarney ER, Panzarine S, Tuttle JI. (1990) Successful contraceptive behaviour among adolescent mothers: are there predictors? J Adolesc Health Care 11:319–325.[CrossRef][Medline]
Bender SS and Geirsson RT. (2004) Effectivenesss of preabortion counseling on postabortion contraceptive use. Contraception 69:481–487.[CrossRef][Web of Science][Medline]
Carstairs V and Morris R. (1991) Deprivation and Health in Scotland(Aberdeen University Press, Aberdeen).
Dawe F and Rainford L. (2004) Contraception and sexual health 2003. A report on research using the ONS Omnibus Survey produced by the Office for National Statistics on behalf of the Department of Health (Office for National Statistics, London).
Department of Health. (2004) Abortion Statistics, England and Wales: 2004 (http://www.dh.gov.uk/publications).
Fleming DJ, Davie J, Glasier A. (1998) Continuation rates of long-acting methods of contraception. Contraception 57:19–21.[CrossRef][Web of Science][Medline]
Garg M, Singh M, Mansour D. (2001) Peri-abortion contraceptive care: can we reduce the incidence of repeat abortions? J Fam Plann Reprod Health Care 27:77–80.[CrossRef][Web of Science][Medline]
Glasier A and Baird DT. (1998) The effects of self-administering emergency contraception. N Engl J Med 339:1–4.
Glasier A, Fairhurst K, Wyke S, Ziebland S, Seaman P, Walker J, Lakha F. (2004) Advanced provision of emergency contraception has not reduced abortion rates in Lothian. Contraception 69:361–366.[CrossRef][Web of Science][Medline]
Graham A, Moore L, Sharp D, Diamond I. (2002) Improving teenagers knowledge of emergency contraception; cluster randomised trial of a teacher led intervention. BMJ 324:1179–1185.
Grigorieva V, Gallo MF, Grimes DA. (2005) Strategies to improve compliance and acceptability of hormonal methods of contraception. Cochrane Database Syst Rev The Cochrane Library volume 3.
Hu X, Cheng L, Hua X, Glasier A. (2005) Advanced provision of emergency contraception to postnatal women in China makes no difference to abortion rates. A randomised controlled trial. Contraception 72:111–116.[CrossRef][Web of Science][Medline]
Hubacher D, Goco N, Gonzalez B, Taylor D. (2000) Factors affecting continuation rates of DMPA. Contraception 60:345–351.[CrossRef]
Information and Statistics Division, Scottish Health Service. (2004) Scottish Health Statistics, Abortions (http://www.isdscotland.org).
Jones RK, Darroch JE, Henshaw SK. (2002) Contraceptive use among U.S. women having abortions in 2000–2001. Perspect Sex Reprod Health 34:294–303.[CrossRef][Web of Science][Medline]
Keder LM, Rulin MC, Gruss J. (1995) Compliance with depot medroxyprogesterone acetate: a randomized controlled trial of intensive reminders. Am J Obstet Gynecol 169:583–585.
Kumar U, Baraister P, Morton S, Massil H. (2004) Peri-abortion contraception: a qualitative study of user’s experiences. J Fam Plann Reprod Health Care 30:55–57.[CrossRef][Web of Science][Medline]
Lo SST, Fan SYS, Ho PC, Glasier AF. (2004) Effect of advanced provision of emergency contraception on women’s contraceptive behaviour: a randomized controlled trial. Hum Reprod 19:2404–2410.
Moos M-K, Bartholomew NE, Lohr KN. (2003) Counseling in the clinical setting to prevent unintended pregnancy: an evidence based research agenda. Contraception 67:115–132.[CrossRef][Web of Science][Medline]
Namerow PB, Weatherby N, Williams-Kaye J. (1989) The effectiveness of contingency-planning counselling. Fam Plann Perspect 21:115–119.[CrossRef][Web of Science][Medline]
National Institute for Health and Clinical Excellence (NICE). (2005) Long-acting reversible methods of contraception (http://www.nice.org.uk).
Penney GC, Templeton AA, Glasier A. (1994) Patients’ views on abortion care in Scottish Hospitals. Health Bull 6:431–437.
Raine TR, Harper CC, Rocca CH, et al. (2005) Direct access to emergency contraception through pharmacies and effect on unintended pregnancies and STIs. A randomized controlled trial. JAMA 293:54–62.
Rosenberg MJ, Waugh MS, Burnhill MS. (1995) Unintended pregnancies and use, misuse, and discontinuation of oral contraceptives. J Reprod Med 40:355–360.[Web of Science][Medline]
Smith KB, van der Spuy ZM, Cheng L, Elton R, Glasier A. (2002) Is postpartum contraceptive advice given antenatally of value? Contraception 65:237–243.[CrossRef][Web of Science][Medline]
Steiner MJ, Dalebout S, Condon S, Dominik R, Trussel J. (2003) Understanding risk: a randomised controlled trial of communicating contraceptive effectiveness. Obstet Gynecol 102:709–717.[CrossRef][Web of Science][Medline]
Stephenson JM, Strange V, Forrest S. the RIPPLE study team. (2004) Pupil-led sex education in England (Ripple study): cluster-randomised intervention trial. Lancet 364:338–346.[CrossRef][Web of Science][Medline]
Trussell J. (2004) Contraceptive failure. In Hatcher RA, Trussell J, Stewart F (Eds.), et al. Contraceptive Technology 18th edn. (Ardent Media Inc, New York, USA) pp. 226.
UNICEF. (2003) A league table of teenage births in rich nations. Innocenti Report Card No 3, July 2003 (Innocenti Research Foundation, Florence, Italy).
Wellings K and Diamond I. (2005) Agenda setting: from population control to reproductive health. In Glasier A, Wellings K, Critchley HOD (Eds.). Contraception and Contraceptive Use(RCOG Press, London, UK) Report of a Study Group.
Wellings K, Nanchahal K, McManus S, et al. (2001) Sexual behaviour in Britain: early heterosexual experience. Lancet 358:1843–1850.[CrossRef][Web of Science][Medline]
Wight D, Raab GM, Henderson M, Abraham C, Buston K, Hart G, Scott S. (2002) Limits of teacher delivered sex education: interim behavioural outcomes from a randomised trial. BMJ 324:1430–1436.
Winter L and Breckenmaker LC. (1991) Tailoring family planning services to the special needs of adolescents. Fam Plann Perspect 23:24–30.[CrossRef][Web of Science][Medline]
Submitted on October 19, 2005; resubmitted on March 16, 2006; resubmitted on April 12, 2006; accepted on April 21, 2006.
CiteULike 
Connotea 
Del.icio.us What's this?
This article has been cited by other articles:
|
|
 |
|
  Minerva BMJ, September 23, 2006; 333(7569): 662 - 662. [Full Text] [PDF]
|
|
| ||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||