Sexual and Contraceptive Behavior studies on Plan B

A Literature Review

 

SUMMARY

 

The sponsor submitted five published and three unpublished studies of varying quality that assess the effects of advance provision of emergency contraception on sexual and contraceptive behaviors. When reviewing these studies as a body of evidence, the quality or applicability of each study must be considered.  While most of the studies state that they were randomized, only one study (Ellertson) used a random number generator as the method of randomization. The other studies used systems such as date of clinic visit or other less desirable systems. The two unpublished studies would not have received peer review (although Gold et. al was recently accepted for publication by the Journal Of Pediatric and Adolescent Gynecology).  Three of the studies were performed in foreign countries and may have limited generalizability to the United States. The overall summary of eight behavioral studies is presented below and in Table 1 followed by the individual reviews. 

 

 

STUDY DESIGN

 

1.          Study Location: Five studies were conducted in the USA and one each was conducted in the UK, Africa and India.

 

2.          Subjects: All subjects in the eight studies were recruited from family planning clinics (the purposes for visiting the clinics were EC consultation, post-abortion follow-up and postpartum evaluation). The age range was 15-45 years with most enrollees being around 20 years old.

 

3.          Study Groups: Subjects were randomly (in most studies) assigned to the following 2 or 3 groups. All subjects received education regarding emergency contraception use (this is in contradistinction to the actual use study submitted by the sponsor where education was not given to the subjects enrolled in the study, thereby more closely mimicking an OTC environment).

 

Advance EC Group: Subjects received in advance one course of EC pills in six studies and three courses of EC pills in two studies (one in US and one on India);

Control Group: Subjects received EC education only (including advice on where to get and how to use EC) except for one study where EC education was not given in the control group;

Pharmacy EC Access Group: Subjects received EC when needed from pharmacy in one US study (in California).

 

4.          Sample Size: Number of subjects ranged from 160 to 1020 in the five US studies and 210-1083 in three studies outside US.

 

5.          Follow-up Period: Subjects were followed from 8 weeks up to 1 year after admission to the studies.

 

RESULTS

 

1.          EC Use: All studies suggest that the advance EC provision increase EC use.  This supported the hypothesis of the studies that easier access would translate into increased use.

 

2.          Unprotected Sex: In these studies, unprotected sex was defined as lack of use of a contraceptive.  All studies demonstrated that compared to baseline, the advance EC group and control group had decreased frequency of unprotected sex.  In some studies, the decrease in unprotected sex was greater in the control group.

 

3.          Condom Use: One US study (sponsored by Women’s Capital Corporation) suggests that the advance, pharmacy and standard EC access groups plus EC education had an increase in more effective methods of contraception with a corresponding decrease in condom use. The other 6 studies either demonstrated no significant decrease in condom use with advance EC provision or in education alone (control groups) or demonstrated that “used condoms every time” increased in treatment and control groups when compared to baseline.

 

4.          Consistent Use of Regular Contraception: Most of the studies demonstrate that women in both the treatment and control groups increase their use of a regular contraception compared to baseline.  One US study suggest that women with advance EC access are more likely to use less-effective contraception (although they had less unprotected sex compared to baseline and increased “condom use every time” from 12% at baseline to 47% at study completion), and another US study showed higher frequency of missing oral contraceptive pills in subjects provided with advance EC than those in control.

 

 

COMMENTS

 

1.          These studies were not conducted in a simulated OTC setting.  However, several of the studies would have recruited a similar subject population as that used in the actual use study.  The main difference in design would be that subjects in the literature review would have received education compared to the subjects in the actual use study and would have received an advanced provision of EC.

 

2.          Subjects were recruited exclusively from clinic sites and received EC education during enrollment (except one study in which control subjects did not receive EC education), which can not be generalizable to the OTC population.

 

3.          Study population in each study was a subset of general population and was heterogeneous among all these studies. This diversity is desirable reflecting many subgroups and capturing the many aspects of an OTC population.  The similarity of results is also reassuring in that the different subsets tend to exhibit the same behavioral trends.

 

4.          Most studies provided only one course of advance EC. In those studies, after the one course of EC pills were used, subjects in the advance EC group would have the same accessibility to EC as the control group.

 

5.          Subjects recruited and studies conducted outside US may not be extrapolated to US population.

 

6.          There were limitations of design and/or methodology of the studies and conclusions should be considered in that context.

 

 

CONCLUSION

 

1.          The literature review studies suggests that the advance access of emergency contraception did not increase the likelihood of unprotected sex in women populations who visited family planning clinics. The study duration’s ranged from 4-12 months in follow-up.  The results may provide certain supportive evidence to resolve some issues raised from the actual use study (such as whether consistent use of routine contraception persist greater than the one month of observation in the actual use study data submitted by the sponsors).

 

2.          The studies did not simulate an OTC setting although some of the studies have recruited similar subject populations as those enrolled in the actual use study.

 

3.          Most of the studies demonstrate that women in both the treatment and control groups increase their use of a routine contraception (less unprotected sex) compared to baseline.

 

4.          Most of the studies either demonstrated no significant decrease in condom use with advance EC provision and control groups or demonstrated that “used condoms every time” increased in treatment and control groups when compared to baseline

 

Table 1. Effects of Emergency Contraception under Advance Provision on Sexual and Contraceptive Behavior

 

Author & Publication	Study Design	Study Location	Subjects	Follow-up Periods	Advance EC Access caused changes in
					Sexual Behavior	Contraceptive Behavior
Raine et al: Obstet Gynecol 2000 Literature #1	Non-randomized 2 groups: Advance EC (one EC course) & Control (EC education)	USA, Family planning clinics	263 women age 16-24 (64% adolescents); 32% Latina & 29% AA; Excluded subjects presenting for EC	4 months	Decrease in unprotected sex in both groups vs. baseline (Control>Tx)	More likely to use less effective contraception (increased condom use) Increased EC use;
Raine et al: UCSF Study (NDA: vol 13, p134) Unpublished Literature #2	Randomized 3 groups: Advance EC Provision (3 EC courses), Pharmacy EC Access; Standard EC Access	USA, Family planning clinics	1020 women age 15-24 years (20±3 yrs); 20% Latina & 17% AA Excluded subjects presenting for EC	6 months	Decrease in unprotected sex in all 3 groups vs. baseline (SA>PA>AP(p<0.05 in PA & SA groups) No increase in incidence of STDs compared to Stnd EC Access	Increase in OC use in all 3 groups with an offset decrease in condom use in all 3 groups Decrease in condom use greatest in AP & PA groups
Jackson et al, Obstet Gynecol 2003 Literature #3	Randomized 2 groups by date of hospital admin: Advance EC (one EC course) & Control (but no EC education)	USA, Inner-city hospital	370 Postpartum women age 26±6 yrs 72% Latina; 43% Married	6 months &   12 months	Increased consistent use contraception and more effective method in both groups.  No increase in report of unprotected sex	No change in routine contraception and condom use; Increase in EC use.
Belzer et al: J Adol Health (Abstract), 2003 Literature #4	Randomized 2 groups: Advance EC (one EC course) & Control	USA, Inner-city (unknown site)	160 adolescent mothers age 14-20 yrs; 83% Latina & 16% AA	6 months	No increase in unprotected sex (but limited data available)	No decrease in condom use and primary contraception between groups.  No data provided on within group changes; (limited data available) Increase in EC use
Gold: Unpublished Manuscript Literature #5	Randomized 2 groups: Advance EC (one EC course) & Control	USA, an urban hospital-based adolescent clinic	301 adolescent women age 15-20 (17±2); 58% AA	8 months	No increase in unprotected intercourse No increase in STDs compared to control	No decrease in condom use; Other info not available
Glasier & Baird: New Eng J Med 1998 Literature #6	Randomized 2 groups by birth date: Advance EC (one EC course) & Control (EC education)	UK, Family planning clinics	1083 women age 16-44 (23% age 16-20), 20%>30 y/o Post EC or Therapeutic abortion	1-year	Decrease in unprotected sex in both groups vs. baseline.	Increase in OC use in both groups with decrease in condom use similar changes between 2 groups. Increase EC use.
Lovvorn et al: Contraception 2000 Literature #7	Non-randomized 2 groups: Advance EC (one EC course) & Control (EC education)	Africa, Family planning clinics	211 women (spermicide users) age 18-45 yrs	8 weeks	Decrease in  unprotected sex compared to baseline in both groups (Control>AEC) Significant limitations in study design.	Increase EC use; Other info not reported. Significant flaws in study design.
Ellertson et al: Obetet Gynecol 2001 Literature #8	Randomized 2 group: Advance EC (3 EC course) & Control	India, family planning clinics	411 women (condom users); age 25±4 yrs (83% 20-29 yr); Barrier method users	12 months (38% 12-month; 90% 3-month); pts off study if switched to more reliable method (23%)	Similar proportion having unprotected sex vs. the control	Increase EC use.

Information in the table is extracted and summarized from the individual literature reviews as attached in the following pages.

The Advance EC (AEC) or the Advance EC Provision (AP) or Treatment(Tx) group : Subjects received EC pills in advance plus EC education at the enrollment.

The Control  or Standard EC access (SA) group: Subjects received only EC education (except the Jackson’s study, Literature #3) and were advised to request EC pills from the clinics (the same sites as the advance group) by prescription when needed.

The Pharmacy EC Access  (PA) group: Subjects received EC pills from pharmacy without prescription.

OC: Oral Birth Control Pills;  AA: Africa American; EC: Emergency Contraception; STDs: sexually transmitted diseases;

 

 

 

Table 1 (Cont). Effects of Emergency Contraception under Advance Provision on Sexual and Contraceptive Behavior

Literature #1   (vol. 13, page 068)

 

Emergency Contraception: Advance Provision in a Young, High-Risk Clinic Population

 

Author:                       Tina Raine, Cynthia Harper, Kathleen Leon, and Philip Darney

 

Affiliate:                     Department of Obstetrics, Gynecology, and Reproductive Sciences

Center for Reproductive Health Research and Policy

University of California, San Francisco, California.

 

Sponsor:                     Compton Foundation, Menlo Park, California

Fred Gellert Family Foundation, San Francisco, California.

 

Study Location:         USA, Family Planning Clinics, San Francisco, California

From June to November 1998

 

Publication:                Obstet Gynecol 96:1–7, 2000

 

Design:                       Single-center, non-randomized, clinical trial;

                                    4-month follow-up

                                    Single course of Advance EC Provision

 

 

METHODS

 

Subject

 

A total of 263 female subjects were recruited and enrolled from a family planning clinic of San Francisco General Hospital between June and November 1998.

 

Inclusion criteria:

Women age 16–24 years

Able to speak English or Spanish

Available for follow-up in 4 months.

 

Exclusion criteria:

Pregnancy

Using contraceptive implants

Using intrauterine devices

Presentation for emergency contraception

Contraindications to oral contraceptive (OC) pills.

 

The subjects were assigned on an alternating basis into the following 2 groups:

 

Treatment groups: 130 subjects received EC education and one course of EC pills (comprised 8 OC pills; each contained 0.15 mg of levonorgestrel and 30 ug of ethinyl estradiol).

 

Control group: 133 subjects received EC education alone.

 

Data Collection

 

Research assistants interviewed subjects at enrollment and at follow-up (at 4 month by telephone or clinic visit) using a questionnaire to obtain demographic information and to measure outcomes, including contraceptive methods and patterns of use.

 

Data Analysis

 

All analyses were conducted using the intent-to-treat population, with all study subjects analyzed according to their initial group assignment. Differences between Treatment and Control were analyzed with a Chi-square test for categorical variables and t tests for continuous variables. A multiple logistic regression analysis was used to determine the effect of advance provision of emergency contraception on use at follow-up.

 

RESULTS

 

Subject Demographics

 

Only age and race/ethnicity were reported in the article, as summarized in Table 1. The mean age was 19 years (64% adolescents). Most subjects were minorities. The demographic distribution between 2 groups was similar.

 

Table 1. Demographics of Subjects

(% of enrolled subjects)

 

Demographic	Treatment n=130	Control n=133	Total n=263
Mean age (years)	19.2	18.8	19.0
Race or ethnicity
Hispanic	33.1	30.1	31.6
Black	26.2	31.6	28.9
White	16.9	12.8	14.8
Asian	14.6	16.5	15.6
Other (biracial)	9.2	9.0	9.1
Primary language Spanish	16.2	14.3	15.2

Data were extracted from the author’s Table 1.

 

Baseline Characteristics

 

At enrollment the sexual activity, contraception, pregnancy history and reasons for clinic visit were comparable between treatment and control groups (Table 2), except that the history of unprotected sex was lower in the treatment group than in control group (15% vs. 24%). The most common contraception method used by the study population was condoms, and a higher proportion of subjects in the treatment arm reported at baseline that they used condoms for contraception than in the control arm (47% vs. 39%).  At baseline a higher proportion of the subjects in the control arm reported that their use of either condoms or oral contraceptives was consistent (used condoms every time, never missed pills) than subjects in the treatment arm.  Consistent condom use was reported in 24% of control subjects who used condoms compared to 12% on the treatment arm.  Consistent use of oral contraceptives was reported in 42% of the control subjects who used oral contraceptives, compared to 25% on the treatment arm. 

 

The table below demonstrates that the subjects in the treatment arm reported higher frequency of sexual acts, higher proportion of condom use as a method of contraception, lower rate of unprotected sex, higher proportion with a history of elective abortion, higher proportion with a history of pregnancy and more births.  More subjects in the treatment arm presented to the clinic visit for an “infection check”. 

 

Table 2. Baseline Traits of Subjects at enrollment

(% of enrolled subjects)

 

Baseline Characteristics	Treatment n=130	Control n=133	Total n=263
Reason for clinic visit*
Papanicoloau smear or check-up	16.2	19.5	17.9
Contraception	39.2	32.3	35.8
Follow-up abortion	7.7	6.8	7.2
Pregnancy test	37.7	44.4	41.4
Infection check	20.8	13.6	17.1
First visit to clinic	38.8	40.6	39.7
Pregnancy History
Ever pregnant	56.2	47.4	51.7
Ever gave birth	20.8	16.5	18.6
Ever had an elective abortion	40.8	34.6	37.6
History of sexually transmitted disease	18.6	18.0	18.3
Sexual Acts in past 4 months
None	3.8	5.3	4.6
Sporadic†	33.9	39.8	36.9
Once a week	25.4	27.8	26.6
More than once a week	36.9	27.1	31.9
New sexual partner	23.1	21.0	22.0
Current contraception‡
Condoms	46.9	39.1	43.0
Oral contraceptive	27.7	24.8	26.2
Depot medroxyprogesterone acetate	10.0	11.3	10.6
Other	0.7	0.8	0.8
None (unprotected sex?)	14.6	24.1	19.4
Dual use (hormonal with condoms)	16.9	17.3	17.1

* Participant might have had more than one reason for clinic visit.

† Sporadic: once or twice in past 4 months or once to twice a month.

‡ Current contraception: most effective method reported if more than one used.

 

Changes in Sexual and Contraceptive Behavior (Table 5)

 

EC Use:

 

·          Women in the treatment group were significantly more likely to use emergency contraceptives than those in control groups (20% vs. 7%, p=0.006).  This difference between treatment arms remained statistically significant in multiple logistic regression analyses that evaluated the impact of contraceptive method, pattern of contraceptive use at enrollment and frequency of unprotected sex reported at enrollment.

·          Overall EC use increased from enrollment to follow-up (4% vs. 14%) in both groups, with more increase in the treatment group.

 

Routine Contraception:

 

·          Women in the treatment group were more likely to have switched to a less-effective  contraception method  than those in the control groups at the time of follow-up (28% vs. 17%, p=0.05).  (Level of effectiveness was ordered from most effective to least effective for this analysis as follows:  depot,  oral contraceptive, barrier, none.)  The proportion that didn’t change method or continued to use no method at all was similar between arms at the time of follow-up.

 

·          Women in the treatment group tended to be less likely to use more effective contraception than those in the control groups (20% vs. 29%, p=0.1).

 

·          The proportion of women in the treatment group who reported consistent oral contraceptive (OC) use was  less than in the control group at baseline (25% vs. 42%, p=0.08)   Although the proportion reporting consistent oral contraceptive use remained lower on the treatment arm relative to the control arm at the time of follow-up (32% vs. 58%, p=0.03), the proportion of subjects who reported consistent use increased in both groups at the time of follow-up.

 

Unprotected Sex:

 

·          Overall “never had unprotected sex” (had protected sex) increased at the follow-up as compared to that at the enrollment (33% vs. 56%). As compared to the baseline, increase in protected sex was 18% (from 32% to 50%) in the treatment group and 28% (from 34% to 62%) in the control group (no statistical tests were available).

 

Condom Use:

 

·          Condom use increased in both groups at follow-up as compared to  at enrollment (Treatment group: 12% vs. 47.4%, Control group: 24.3% vs 50%).

 

·          There were no significant difference at the time of follow-up between treatment and control groups in the proportion of condom use.

 

·          Since there was less condom use at baseline in the treatment group than in control group, the proportionate increase in condom use in the treatment group was greater than in control group (4x increase vs. 2x increase).

 

 

Table 5. Contraceptive Behavior during the Study Period Compared to Baseline

Between Treatment and Control Groups

(% of enrolled subjects)

 

Contraceptive Behavior	Treatment % (n)	Control % (n)	Total % (n)	P
Initial (at Enrollment)
Never had unprotected sex	32.3 (42)	33.8 (45)	33.1 (87)	0.92
Used condoms every time	12.0 (10)	24.3 (18)	17.8 (28)	0.08
OC users who never missed pills	25.0 (11)	42.2 (19)	33.7 (30)	0.08
Used emergency contraception	4.6 (6)	3.0 (4)	3.8 (10)	0.75
Follow-up
Never had unprotected sex	50.4 (56)	61.8 (63)	55.9 (119)	0.42
Used condoms every time	47.4 (18)	50.0 (19)	48.7 (37)	0.71
OC users who never missed pills	31.7 (13)	57.8 (26)	45.4 (39)	0.03
Used emergency contraception	19.8 (22)	6.9 (7)	13.6 (29)	0.006
More effective method†	19.8 (22)	29.4 (30)	24.4 (52)	0.10
Less effective method‡	27.9 (31)	16.7 (17)	22.5 (48)	0.05
No method at enrollment and follow-up	7.2 (8)	6.9 (7)	7.0 (15)	0.92
Same method at enrollment and follow-up	45.0 (50)	47.1 (48)	46.0 (98)	0.77

Data are extracted from the author’s Tables 2, 4 and 5.

† More effective methods: Depot medroxyprogesterone acetate and OC; and ‡ less effective methods: spermicides, diaphragms, and withdrawal.

 

 

COMMENTS

 

1.     The subjects were not randomly assigned. This created an imbalance at baseline (unprotected sex, condom use, missed OC pills and EC use) and could have introduced bias into the study. This is a major flaw and limits conclusions.

 

2.     Only a single course of EC was provided to the treatment group (advance provision), so the study observation may not truly reflect changes in sexual and contraceptive behaviors that may occur in the OTC setting.

 

3.     Information on education, literacy level, and income of subjects were not provided.  Given the non-randomized design there are no assurances of an even distribution of these demographics.

 

4.     Sample size was small (n=130 in the Advance group and n=133 in the Control group).

 

5.     Subjects were recruited from clinical sites and were high risk, which may not be generalizable to OTC population.

 

 

CONCLUSION

 

This study demonstrated the following: Compared to their baseline, women age 16-24 with advance EC provision are::

 

·       Less likely to have unprotected sex

·       More likely to use condoms (every time)

·       More likely to use EC pills

·       More consistent with their use of  OC pills

·       Compared to the control group, the women provided with Advance EC were more likely to switch to a less effective routine contraception method and have a higher proportional increase in condom use

 

When compared to the treatment group, the control group was more likely to have a greater absolute change in “Never had unprotected sex” and “Never missed pills”. However, it should also be noted that at baseline, the treatment group appeared to potentially be a higher risk group compared to the control group, with a greater percentage of subjects that were presenting to the clinic for contraception, infection checks, had been pregnant, had given birth, had an elective abortion, had a new sexual partner in the past 4 months and had an elective abortion.  Because of these imbalances at baseline, between group comparisons should be made with caution.

 

Literature #2: UCSF Study #H9738-18501-02

(Vol. 32/p134; EDR dated 2003-09-08)

 

Provision of Emergency Contraception to Women enrolled in the study prior to December 31, 2001: Pharmacy Access and Advance Distribution Evaluation

 

Investigator:              UCSF (by Tina Raine, et al)

 

Sponsor:                     Women’s Capital Corporation

 

Study Location:         USA, Family Planning clinics four clinical (San Francisco)

July 9, 2001 to December 31, 2001

 

Report Date:              January 30, 2003 (prepared by Pinney Associates)

 

Published:                   Not

 

Study Design:             Randomized 3-arm clinical trial

6-12-month follow-up

                                    3 courses of advance EC.

 

Primary objectives:

 

To compare the rates of unintended pregnancy and sexually transmitted disease (STD) among three different distributions (advance provision, pharmacy access and standard access) for emergency contraception.

 

Secondary objectives:

 

To assess the effects of the three different emergency contraception distribution methods on sexual and contraceptive behaviors, such as the frequency of unprotected sex, and use of condoms, oral contraceptives, and emergency contraception use.

 

 

METHODS

 

Subjects

 

Subjects were recruited from four family planning clinic sites (Table 1) in the San Francisco bay area (CA) with the following inclusion and exclusion criteria.

 

Inclusion Criteria

·        Women age 15-24 at high risk for unintended pregnancies

·        Speak either Spanish or English

·        Be available in 6 months for a follow-up visit.

 

Exclusion Criteria

·        Women were currently pregnant;

·        Actively trying to get pregnant;

·        Sterilized; using Depo-Provera, IUD, Norplant or Lunelle;

·        Reported having had unprotected sexual intercourse in the past 3 days.

·        Women who presented to the clinic specifically requesting emergency contraception (the rationale was not specified)

 

Table 1. Enrolled subjects and follow-up compliance from 4 clinical sites

 

Clinical Sites (San Francisco, CA)	Pharmacy Access (N = 343)	Advance Provision (N = 340)	Standard Access (N = 337)	Total (N = 1020)
City College	30 (96.8)	29 (90.6)	31 (96.9)	90 (94.7)
Planned Parenthood: Dale City	44 (97.8)	42 (95.5)	40 (88.9)	126 (94.0)
New Generations	136 (91.9)	141 (95.9)	128 (90.8)	405 (92.9)
Planned Parenthood: San Francisco	104 (87.4)	104 (88.9)	107 (89.9)	315 (88.7)

Data were adapted from the author’s Table 4 (p6).

 

Subject Disposition

 

Of 2012 screened women, 1,020 were enrolled (see below) and randomly assigned into 3 groups.

 

Total approached women:          2,012    (100%)

Total screened women:              1,804    (89.7%)

Ineligible women:                       992       (49.3%)

Eligible women:             1,024    (56.7%)

Enrolled subjects:                       1,020    (56.4%)

 

Pharmacy Access (PA): Subjects in this group obtained Plan B at the local pharmacy without a prescription through a collaborative agreement between clinics and pharmacies;

 

Advance Provision (AP): Subjects were given Plan B (3 complete packages) to take home and use as needed;

 

Standard Access (SA): Subjects returned to the clinic to obtain supplies. 

 

All subjects received information and counseling on emergency contraception, and were reimbursed $10 (during the visit). They were also reimbursed $20 for completing the follow-up visit procedure.

 

Data Collection

 

Baseline data: urine tests (for pregnancy, Chlamydia and gonorrhea) and blood test (for HSV-2 antibody) and interview (for demographics, sexual history, knowledge of emergency contraception).

 

Follow-up visit: occurred 6 months or more (up to 1 year) after enrollment. Data were collected on sexual history, use of emergency contraception, urine test (for pregnancy, Chlamydia, and gonorrhea), and blood test (for HSV-2 antibody).

 

Data Analysis

 

One-way analysis of variance, contingency table analyses, and a Chi-square test were used for different variables. In cases of small numbers, when cells had fewer than 5 observations, the Fisher’s exact test was conducted. All analyses were evaluated at the two-tailed probability level of p<0.05 and no adjustments were made for the number of analyses or pair-wise comparison.

 

RESULTS

 

Subject Demographics

 

Demographic characteristics of the enrolled subjects are summarized in Table 2. Overall they were comparable among the 3 study groups. Races (white, black, Latina, and Asian) were evenly distributed among 3 groups. The following were the major characteristics:

 

Mean age:                                                            20±2.6 (15-24) years

Marital status:                                                       86% single

Active sex (within 6 months):                                 100%

Unprotected intercourse (within 6 months): 50%

Currently using condoms:                                      67%

Currently using oral contraceptives:                        41%

Previous emergency contraception:             35%

Education and literacy level:                                   unknown (not reported).

 

A third of participants reported having been pregnant previously, with 9% reporting ever given birth.

 

Follow-up Compliance

 

Approximately 92% of subjects (936 of 1,020) in each group completed follow-up assessment within one year (211±39 days) (Table 3).

Table 2. Demographics of subjects

 

Demographics	Pharmacy Access (N = 314)	Advance Provision (N = 316)	Standard Access (N = 306)	Total (N = 936)
Age (years)
Mean ± SD	19.7 ± 2.6	19.7 ± 2.6	19.9 ± 2.6	19.7 ± 2.6
Race
Latina	66 (21.0)	64 (20.3)	60 (19.6)	190 (20.3)
Black	53 (16.9)	54 (17.1)	52 (17.0)	159 (17.0)
White	79 (25.2)	100 (31.7)	83 (27.1)	262 (28.0)
Asian/Pacific Islander	57 (18.2)	62 (19.6)	69 (22.6)	188 (20.1)
Multiracial	48 (15.3)	29 (9.2)	35 (11.4)	122 (12.0)
Other	11 (3.5)	7 (2.2)	7 (2.3)	25 (2.7)
Marital Status
Single	263 (83.8)	273 (86.4)	271 (88.6)	807 (86.2)
Cohabiting	42 (13.4)	31 (9.8)	28 (9.2)	101 (10.8)
Married	7 (2.2)	10 (3.2)	6 (2.0)	23 (2.5)
Married, but separated	2 (0.6)	-	-	2 (0.2)
Divorced	-	2 (0.6)	1 (0.3)	3 (0.3)
Widowed	-	-	-	-

Data were extracted from the author’s Table 5 (p7) and presented as “No. (%).”

 

 

 

Table 3. Subject Disposition at Follow-up

 

Disposition	Pharmacy Access No. (%)	Advance Provision No. (%)	Standard Access No. (%)	Total No. (%)
Enrolled Subjects	343	340	337	1020
Lost to Follow-up	23 (6.7)	23 (6.8)	27 (8.0)	73 (7.2)
Refused Follow-up	6 (1.8)	1 (0.3)	4 (1.2)	11 (1.1)
Completed Follow-up*	314 (91.6)	316 (92.9)	306 (90.8)	936 (91.8)

* Mean follow-up days (post-baseline) was 211±39 days, median follow-up days: 195 days. There were no statistical differences in demographics of subjects between Lost-to-Follow-up and Completed-Follow-up.

 

Sexual and Contraceptive Behavior

 

Pregnancy:

 

The overall pregnancy rate (Table 4) in the 936 subjects who completed the follow-up interview was 8%. There were no differences in pregnancy rate among 3 groups including when analysis was controlled for baseline history of pregnancy (p<0.89, Chi-square test).

 

Sexually Transmitted Diseases (STDs):

 

Subjects were considered to have acquired an STD during the follow-up period if they were positive for herpes (new diagnosis), chlamydia, gonorrhea (self-reported or by laboratory tests), trichomonas, PID (self-reported). There were no differences in STD acquisition among 3 groups when controlling for baseline history of STDs (p<0.427, Chi-square test).

 

Table 4. Pregnancy and STD during study

 

Outcome		Pharmacy Access (N = 314)	Advance Provision (N = 316)	Standard Access (N = 306)	Total (N = 936)
Pregnancy	Previous	108 (34.4)	97 (30.7)	99 (32.5)	304 (32.5)
	Follow-up	24 (7.6)	26 (8.2)	25 (8.2)	75 (8.0)
STDs*	History	62 (19.8)	69 (21.8)	77 (25.4)	208 (22.3)
	Acquired during Study	58 (18.5)	47 (14.9)	51 (16.7)	156 (16.7)

Data were extracted from author’s Tables 6, 7 & 8 (p8-9) and presented as “No. (%).”

* STDs at follow-up were newly acquired (not included baseline).

 

Emergency Contraception (Table 5):

 

Overall 29% (269 of 936) of subjects used emergency contraception during the study period.

 

1.     Subjects were more likely to use emergency contraception in the Advance Provision group (39.2%) than in the Pharmacy Access group (26.5%) or in the Standard Access group (20.3%); they were also more likely to report convenience of emergency contraception compared to those in the Pharmacy Access group or the Standard Access group (96%, 87% and 87%, respectively).

 

2.     There were no differences in the time to take the first pill among the 3 groups or in overall proper use of emergency contraception (p>0.05). Subjects in the Standard Access group tended to have higher correct use than other 2 groups, 97% (SA), 92% (AP) and 90% (PA), and were more likely (100%) to take the second pill than the Advance Provision group (93%) and the Pharmacy Access group (90%).  It should be noted that the time interval for use of the second pill was not defined in this study  as it was in the actual use study (12 hours after first pill).

 

3.     The proportion of repeat use was highest in the Advance Provision group.  The baseline EC use was 35% for the entire study population.

 

Table 5. Emergency Contraception Usage during Study

 

EC Use	Pharmacy Access (N = 314)	Advance Provision (N = 316)	Standard Access (N = 306)	All Subjects (N = 936)
Total*	83 (26.5)	124 (39.2)	62 (20.3)	269 (28.8)
Never	230 (73.5)	192 (60.8)	244 (79.7)	666 (71.2)
One time	52 (16.6)	75 (23.7)	45 (14.7)	172 (18.4)
Two times	20 (6.4)	28 (8.9)	13 (4.3)	61 (6.5)
> 3 times	11 (3.5)	21 (6.7)	4 (1.3)	36 (3.8)

Data were extracted from the author’s Table 10 (p13) and presented as “No. (%).

*Pair-wise comparisons: AP vs. PA (p<0.001), AP vs. SA (p<0.001), and PA vs. SA (p<0.067).

 

 

Sexual behavior (Table 6):

 

Overall 96.6% (903 of 936) were sexually active during the study period.

 

1.     There were no statistically significant differences in the rates of unprotected sex among the 3 groups. Subjects in the Advance Provision group tended to have higher frequency of unprotected intercourse (47% in AP, 41% in PA and SA).

 

2.     As compared to the baseline, the frequency of unprotected sex decreased in all three groups.  This change was statistically significant in the SA and PA groups but not in the AP group (Decrease from 50% to 41% in PA, and from 53% to 41% in SA, p< 0.01 by McNemar’s test; but from 49% to 47% in AP, p=NS).

Table 6. Sexual and Contraceptive Behaviors

 

Behavior	Pharmacy Access N=314	Advance Provision N=316	Standard Access N=306	Total N=936
Unprotected Intercourse
At Baseline (past 6 month)
Total	156 (49.7)	153 (48.6)	162 (52.9)	471 (50.4)
Every time	12 (3.8)	13 (4.1)	12 (3.9)	37 (4.0)
Most of the time	18 (5.7)	32 (10.2)	25 (8.2)	75 (8.0)
Some of the time	126 (40.1)	108 (34.3)	125 (40.9)	359 (38.4)
Never	158 (50.3)	162 (51.4)	144 (47.1)	464 (49.6)
At last sex	26 (8.3)	32 (10.1)	35 (11.4)	93 (9.9)
At Follow-up(≥ 6 month)
Total	127 (40.6)	147 (46.7)	124 (40.5)	398 (42.6)
Every time	9 (2.9)	11 (3.5)	7 (2.3)	27 (2.9)
Most of the time	16 (5.1)	24 (7.6)	22 (7.2)	62 (6.6)
Some of the time	102 (32.6)	112 (35.6)	95 (31.1)	309 (33.1)
Never	186 (59.4)	168 (53.3)	182 (59.5)	536 (57.4)
At last sex	30 (9.6)	37 (11.8)	24 (7.9)	91 (9.7)
Condom Use
At Baseline (past 6 month)
Total	257 (81.8)	254 (80.4)	248 (81.3)	759 (81.2)
Every time	87 (27.7)	83 (26.3)	75 (24.6)	245 (26.2)
Most of the time	86 (27.4)	83 (26.3)	86 (28.2)	255 (27.3)
Some of the time	84 (26.8)	88 (27.9)	87 (28.5)	259 (27.7)
Never	57 (18.2)	62 (19.6)	57 (18.7)	176 (18.8)
Use at last sex	190 (60.5)	184 (58.2)	174 (56.9)	548 (58.6)
Currently using	219 (69.8)	214 (67.7)	192 (62.8)	625 (66.8)
At Follow-up(≥ 6 month)
Total	226 (72.7)	232 (74.4)	224 (73.4)	682 (73.5)
Every time	84 (27.0)	68 (21.8)	78 (25.6)	230 (24.8)
Most of the time	77 (24.8)	82 (26.3)	75 (24.6)	234 (25.2)
Some of the time	65 (20.9)	82 (26.3)	71 (23.3)	218 (23.5)
Never	85 (27.3)	80 (25.6)	81 (26.6)	246 (26.5)
Use at last sex	159 (50.6)	154 (48.9)	170 (55.7)	483 (51.7)
Currently using	178 (56.7)	179 (56.7)	186 (60.8)	543 (58.0)
Oral Contraceptive Use
At Baseline (past 6 month)
Missing pill (per pack)	243 (63.6)	74 (58.3)	90 (68.7)	79 (63.7)
Use at last sex	117 (37.3)	125 (39.6)	115 (37.6)	357 (38.1)
Currently using	129 (41.1)	132 (41.8)	125 (40.9)	386 (41.2)
At Follow-up(≥ 6 month)
Missing pill (per pack)	276 (68.0)	96 (65.7)	97 (70.3)	83 (68.0)
Use at last sex	159 (50.6)	150 (47.8)	141 (46.2)	450 (48.2)
Currently using	159 (50.6)	150 (47.5)	139 (45.4)	448 (47.9)

Data were extracted and summarized from author’s Table 6 (p8) and Table 9 (p11), presented as “No. (%)

Contraception Methods (Table 6):

 

There were no statistically significant differences in condom and OC uses among the 3 groups during the study period.

 

1.     Overall condom use decreased in all 3 groups as compared with the baseline; condom use “at last sex” significantly decreased in the Advance Provision and the Pharmacy Access groups (P< 0.01), but remained relatively stable in the Standard Access group (p< 0.65).

 

2.     During the same period that condom use decreased, there were increase in “currently using” OC in all groups compared to the baseline.  The “Use at last sex” increased in all groups (PA change=37% to 51%, AP change=40% to 48% and SA change=38% to 46%).

 

Age difference:

 

There were no significant differences in observed parameters (pregnancy, condom use, unprotected sex, routine OC use) between adolescents (15-17 years old) and adults (18-24 years old).

 

 

COMMENTS

 

1.          The proportion of 15-17 year olds and the literacy level of subjects were not provided.

 

2.          “Baseline” STDs were reported, but the exposure period over which an infection was acquired was not captured. However, STDs acquired during the study among the 3 groups were comparable and were lower than the “baseline” history.

 

3.          Overall unprotected intercourse decreased in all three groups compared to baseline. This change was statistically significant in the SA and PA groups but not in the AP group (from 50% to 41% in PA, and from 53% to 41% in SA, p< 0.01 by McNemar’s test; but from 49% to 47% in PA, p=NS). 

SUMMARY

 

1.          Advance Provision did not increase STDs as compared with the Pharmacy Access and the Standard Access to EC.

 

2.          Subjects in the Advance Provision group were more likely to use EC pills as compared to the Standard and Pharmacy Accesses.

 

3.          All three groups had less unprotected intercourse during the follow-up as compared to baseline. When compared to baseline, the PA and SA groups decreased more (P<0.01) than the advance EC provision group (p=NS).

 

4.          All three EC accesses were associated with a decrease in condom use, with statistically significant decreased differences in “use at last use” among the Advance Provision and the Pharmacy Access groups. However, the  decrease in condom use was offset with increased oral contraceptive use.

 

5.          Although there was greater OC use at study end compared to baseline for “use at last sex” and “currently using” in all 3 groups the proportion in all three groups who reported “missing OC pills” at study end compared to baseline increased.

 

 

CONCLUSION

 

Advance Provision of emergency contraception was not found to be associated with a difference in pregnancy rates or acquired STDs compared to Pharmacy Access or Standard Access.  Advance Provision decreased unprotected intercourse compared to baseline, but to less of an extent that the Pharmacy Access or Standard Access groups.  All three groups had increased OC use and decreased condom use.  All three groups had increased “missing oral contraceptive pills” compared to baseline at study end.  The highest rate of unprotected sex was in the AP group (49% vs. 41% in the PA & SA groups).

Literature #3   (sNDA 21-045, Serial No. 105, p5162)

 

Advance Supply of Emergency Contraception: Effect on Use and Usual Contraception—A Randomized Trial

 

Author:                       Rebecca A. Jackson, Eleanor Bimla Schwarz, Lori Freedman, and

Philip Darney

 

Affiliate:                     Center for Reproductive Health Research and Policy and Department of Obstetrics, Gynecology, and Reproductive Sciences, and Division of General Internal Medicine

Department of Medicine, UCSF, and San Francisco General Hospital, San Francisco, California.

 

Sponsor:                     Partially funded by an unrestricted grant from the Packard Foundation. The Packard Foundation is a nonprofit organization.

They provided funds for supplies and oral contraceptive pills.

 

Study Location:         USA, a public inner-city hospital in San Francisco

From September 1998 through March 1999

 

Publication:                Obetet Gynecol 102: 8-16, 2003

 

Design:                       Randomized (by date of discharge) controlled clinical trial

1 year observation

Single course of advance EC provision

 

 

METHODS

 

Subject: A total of 370 postpartum women were enrolled from a public inner-city hospital (San Francisco, CA) from September 1998 through March 1999, with the following eligibility criteria:

 

Age: (was not specified in the Method)

Postpartum women (had a live birth)

Spoke English or Spanish

Available for follow-up in 1 year

Had not undergone a postpartum tubal ligation

 

Subjects were randomly assigned to the following 2 groups:

 

Advance Provision Group: 184 subjects received one course of EC pills and EC education. The one course EC contained 8 oral contraceptive pills containing 0.15 mg of levonorgestrel and 30 ug of ethinyl estradiol. The educational session was a 5-minute intervention and included instructions for obtaining additional emergency contraception pills if needed.

 

Control group: 186 subjects received only routine contraceptive counseling, and this did not usually include EC education.

 

To prevent interference of the difference education’s that provided to each group, the investigators enrolled all women on a given day to the same group.

 

Data Collection: A Kaiser Family Foundation Questionnaire (survey) was the data collection instrument. The questionnaire was administered in person at enrollment and by phone at 6 and 12 months. The primary outcome was self-reported use of emergency contraception. Secondary outcomes included change in use of other contraceptive methods and knowledge about emergency contraception. Contraceptive and sexual behaviors were assessed by asking about types of contraception used and consistency of use.

 

Data Analysis: The individual subject was used as the unit of analysis. Differences between groups and differences within each group over time were analyzed using the Fisher exact test, Student t test, or the McNemar test.

 

 

RESULTS

 

Subject Demographics

 

Of 721 screened subjects, 370 were enrolled and randomized to the Advance EC group (184) and Control group (186). The demographics and baseline characteristics of the enrolled subjects are summarized in Table 1; there were no statistically significant differences between the two groups. Approximately 18% were teens; 72% were Latina; 43% married. About half had a high school education.

 

Follow-up Compliance

 

At 6 months after enrollment, follow-up was available for 78% and at 1 year, 69%. Overall, 85% were available for at least one follow-up session. There were no differences between groups in the proportion lost to follow-up; nor were there differences in baseline traits between those lost to follow-up and those who completed the study (Table 1).