Pivotal Study #1:

Minor subjects: Subjects aged 15 or younger were excluded from 2 sites (Phoenix and Houston) because parental consent for this age was required. The other clinical sites did not impose age restrictions. No specific measures were taken to recruit minors at the sites so that the population would be representative of that seeking emergency contraception.

Assessment of Self-Selection/De-Selection

Self-selection: Subjects were asked to review the drug package, to complete the Screening Form and to determine whether or not the product was appropriate for them. The validity of the self-selection decision was determined based on the reasons that the subject requested and used Plan B.

Self-deselection: Contraindicated uses were evaluated for assessment of self-deselection, as follows:

Determination of Pre-existing pregnancy: Pre-existing pregnancy was evaluated at screening time (last menstrual period and usual menstrual cycle length), at follow-up contact (menstrual profile, pregnancy test results since last menstrual period (LMP), other reasons to suspect pregnancy), and by estimated fertilization date (pregnancy at least 14 days after fertilization)

Unexplained vaginal bleeding: At the earliest follow-up contact, the medical monitor determined the onset of “any unusual” vaginal bleeding, and how it was unusual, before subject took the first pill.

Allergic to the product: Information on allergy to all medications, foods and Plan B was collected during both follow-up contacts to avoid influencing behavior.

The reasons to deselect Plan B in those subjects who did not request or who received Plan B but did not use it were not evaluated in the study.

Assessment of Correct/Incorrect use

The date and time of coitus and subsequent ingestion of each pill were collected. The following questions were asked to collect data about intentional and unintentional incorrect use:

Did you use Plan B according to all the instructions on the box? If no, what instruction did you not follow?

Did you know you were not following the instructions when you took the Plan B pills or did you realize that only after you took the pills?

Primary analyses of incorrect use: the subjects did not follow the dosing schedule: “Take the first tablet as soon as possible within 3 days (72 hours) after unprotected sex; Take the second tablet 12 hours after you take the first tablet”.

Secondary analyses of incorrect use: First tablet up to 72 hours after intercourse and second tablet up to 16 hours after the first; First tablet up to 72 hours after intercourse and second tablet up to 6-18 hours (the rationale for this timing was not provided in the report) after the first; First tablet up to 120 hours after intercourse and second tablet up to 24 hours after the first.

Subjects with unclassifiable use patterns were excluded in the analyses.

Safety Assessment

Adverse events were defined as medical problems that started or worsened after Plan B use

Adverse events were recorded in the Study Data Cards and were collected during the follow-up contacts (week 1, week 4 and thereafter for some subjects), and recorded in the “Adverse Events Notes page” by study staff, reviewed and then transferred to the Adverse Events data form by the clinician. The frequency, severity (mild, moderate or severe), and seriousness, and the relevance to the drug treatment were analyzed.

Site staffs did not investigate discrepancies of AE reports between the cards and the follow-up contacts unless the cards indicated the possibility of an unusual, serious, or severe event.

Adverse events were coded in COSTART terminology. The 95% confidence limits around frequencies within body system class were calculated by the exact binomial method.

Concomitant Medications: Medications used by enrolled subjects during the week before enrollment and during the follow-up period were coded according to the WHO Anatomical Therapeutic Chemical Classification system.

Efficacy Assessment

At the follow-up contacts, information about menstrual history, pregnancy test results since LMP, and any suspected pregnancy were collected and evaluated by medical monitors.

Criteria to determine pregnancy (any of the following):

a. Did not have menses following product use

b. Had a positive pregnancy test

c. Had any other evidence of pregnancy, such as ultrasound, abortion.

d. Had no evidence ruling out pregnancy as assessed by site staff.

Estimate Fertilization Date

Pregnancy occurring before Plan B use was determined by the estimated fertilization date (based on data collected from all pregnant subjects) or date of sexual intercourse. There was no diagnostic examination (ultrasound and/or blood/urine test) to confirm the fertilization date.

Criteria to exclude pregnancy

a. Had a menstrual period after Plan B use

b. Had a negative pregnancy test at least 2 weeks after the products

c. Other information (not specified in the report), confirmed by medical monitor individually

Undetermined pregnancy:

Subjects who did not meet the above inclusion and exclusion criteria to determine pregnancy were classified as “undermined pregnancy” and were followed up until pregnancy status was clarified.

Sexual and Contraceptive behaviors

Information about sexual behavior and contraception methods before (one month) and after (4 weeks) Plan B use were collected using multiple questions containing the following parameters:

Had sex: Subjects had sex with or without contraception during specific interval.

Had at least one sex act without contraception: included any subject who was classified as having had sex in the specified interval but indicated that she did not use any contraception at least once.

Used a “more effective” method: oral contraceptive pills, Depo-Provera, Lunelle, vasectomy, or intrauterine device at any time during the interval.

Used a “less effective” method: other than “more effective” methods, including those who did not indicate use of any method.

Used a specific contraceptive method: included any subject who indicated that she had used that method at any time in the specified interval.

Used no condoms: included any subject classified as having had sex in the specified interval who was not classified as ever having used condoms in that interval.

Subjects’ reports of abstinence (“yes” answers to “have not had sex: in the past month, since receiving Plan B, or since One Week contact”) were not included in these classifications because some subjects gave contradictory responses.

Data Collection

Data were collected by the investigators using the following forms at the study sites. Data recorded on all forms received at FHI were entered into the computer with Clintrial 4.3 software; the data entry system was validated by FHI data management staff.

Plan B Study Screening Form

Plan B Study Background Questionnaire

Plan B Study Disposition Form

Plan B Study One Week Contact Form

Plan B Study Four Week Contact form

Plan B Study Data Card

Plan B Study Data Card Transcription Form

Plan B Supplemental Contact Form

Study Data Card: Information about the LMP, the sex act prompting use of the product, timing of ingestion of each tablet, and pregnancy test were collected. Information was transcribed onto Study Data Card Transcription Form during site visits by FHI.

Site staffs were instructed not to investigate discrepancies between the information obtained at the follow-up contacts and on the cards.

Data Audit: There was computer hardware problem (p029, vol. 27) during the study, which may have compromised some data entries. It was/is unclear how much of the study data were affected by this problem. FHI conducted an audit analysis, showing the error rate was 0.011-0.036% of 95% CI. The sponsor stated that this was less than the pre-defined 0.05% (audited 61,336 data points from key data fields for the primary outcomes).

Data Analyses

Statistical analyses: Data were summarized in tabular forms with SAS (version 8.0). The mean, median, minimum-maximum, and standard deviation (SD) were calculated for continuous variables and frequency tables were used for categorical data. For proportions, exact binomial 97.5% confidence intervals were calculated.

[See statistical review for certain statistical issues].

Definition of analysis populations:

Screened Population: All subjects screened in the study and no one was excluded.

Enrolled Population: All subjects who enrolled in this study (i.e., received study product).

Per-Protocol Population: The Enrolled Population excluding subjects enrolled with violations of any of the study admission protocol criteria.

Incomplete Follow-up Population: All subjects who did not complete follow-up procedures (i.e., did not complete both scheduled contacts and all required supplemental contacts or did not mail in the Study Data Card).

Lost to Follow-up Population: All subjects who provided no follow-up data.

Missing Data: Analyses of contraindication use, incorrect use, and pregnancy included only subjects with sufficient data to allow classification of the status of the outcome. The sponsor assumed that subjects with missing data had the same outcomes as subjects who provided data.

Data obtained at visits that were outside the “per protocol” time windows were included in the primary outcome analysis. In all analyses, the data from the Study Data Cards were used only if the corresponding data from the contact visits were missing (except where otherwise noted).

Deviations from the Study Protocol

1. Informed Consents: Two versions of the Informed Consent forms were used, one (amendment #1) at 4 clinical sites and pharmacy sites and another at the Boston site (using the final version, amendment #2), as listed in Table 2. The IRB at each site approved the consent form that the site actually used. (Reviewer: the discrepancies were minor and may not affect either subject welfare or the integrity of the data).

2. Missing package: Study drug packages (dispensed or returned) were accounted for all sites except at the Houston site where 2 packages were missing.

3. Evaluable population: In the protocol, this was all screened subjects who met all eligibility criteria, who used the product, and who completed both follow-up contacts. In the actual analysis, the evaluable population included those subjects who received and used the study product and completed one or both follow-up contacts.

4. Repeat users: The number of repeat uses was low and thus no separate analysis was conducted in the report.

5. An analysis of “deliberate incorrect and contraindicated uses” was proposed in the protocol, but the final report focused on “unintentional incorrect and contraindicated use”.

Table 2. Differences between the informed consents used in different study sites

Deviations	Boston Site	All others
Version of Consent Form used	Amendment #2 (final)	Amendment #1
Consult health care professionals	“Please let us know if you would like to speak with a pharmacist/clinician at any time during the research study."	"You may speak with a pharmacist/clinician at any time during the research study."
Ask additional contact	verbally inform each subject at the four-week contact (at all sites)	Grand Rapids, Seattle, and the pharmacies, but others, did not specifically state that subjects might have additional contacts after the one-month contact.
State special warning	No special warnings were used in the informed consent.	“Make sure that you understand the risks and side effects of the package insert instructions before you take the pill. If you have any questions, call [name] or the clinic/pharmacy”.

RESULTS

Subject

Enrollment: A total of 665 women were screened at least once in the 5 clinical sites and 5 pharmacy sites; 585 (88%) subjects met all eligibility criteria and 80 (12%) were ineligible. The distribution of screened subjects and their eligibility were summarized in Table 3. Of the 80 ineligible subjects, 38 women who indicated eligibility did not sign the informed consent and thus became ineligible; the other 42 failed to meet one or more criteria listed in Table 4.

Table 3. Subjects Screened and Enrolled in Each Study Site

	Clinical Site						Pharmacy Sites	Total
	Boston	Grand Rapid	Houston	Phoenix	Seattle	Total	Pharmacy Sites	Total
Screened Subjects	129	125	170	123	79	626	39	665
Eligible Subjects	111	117	138	111	72	549	36	585
Eligible Subjects	86.0%	93.6%	81.2%	90.2%	91.1%	87.7%	92.3%	88.0%
Ineligible Subjects	18	8	32	12	7	77	3	80
Ineligible Subjects	14.0%	6.4%	18.8%	9.8%	8.9%	12.3%	7.7%	12.0%

Table 4. Reasons that screened subjects did not receive Plan B

	Clinical Sites N=626	Pharmacy Sites N=39	Total N=665
Ineligible subjects	77	3	80
Did not sign the consent form (but eligible)	ND	ND	38
Did not think they should receive Plan B	42	0	42
Change mind	2	0	2
Indication	10	0	10
Instruction	1	0	1
Side effect	7	0	7
Want more info	17	0	17
Other options	1	0	1
No reason recorded	4	0	4

Data were extracted from text and Table 1.3 (vol. 28, p011). ND: no data available.

Subject Disposition: The disposition of subjects is summarized in Figure 2 and Table 5. Of the 585 eligible subjects, 576 (98%) were enrolled at first screening into the study (received the study product). Nine eligible subjects declined to participate in the study but 6 of them ultimately received prescription Plan B. Of the 80 ineligible subjects, 60 received prescription Plan B, 9 received the study product, 9 received no treatment and 2 received oral contraceptive pills.

Table 5. Subjects Who Received Treatment

Eligibility	Subjects Screened	Subjects Received Study Plan B	Subjects Received Rx Plan B	Subjects not Received Plan B
Eligible	585	576	6	3
Ineligible	80	9	60	11
Total	665	585	66	14

Demographics: The demographic characteristics of screened and enrolled populations are summarized in Tables 6 and 7. The mean age of the enrolled subjects was 22.1±5.0 (14-44) years old. Seventy-four percent of subjects were 17-25 years old and 5% were ages 14-16. Approximately 74% of the enrolled subjects had at least some college education and 0.3% had less than a 9^th grade education.

The demographic characteristics between the screened and enrolled populations (Table 6), and between the incomplete follow-up/lost to follow-up populations and the screened population seemed to be comparable.

All subjects ages 14-16 were in middle school or high school. Nine percent of subjects age 17-44 had less than an 11^th grade education; 0.2% had less than an 8^th grade education and 8.6% had completed 9^th-11^th grade (Table 7).

Figure 2. Disposition of Subjects

(Adapted from the sponsor’s Figure 1 in page 070 of vol. 27)

Table 6. Demographics of Screened and Enrolled Populations

(Adapted from the sponsor’s Table F in page 44 of vol. 27)

Table 7. Demographics of Enrolled Subjects by Age

(% of enrolled subjects in parentheses)

Characteristics	Age (years)		Total N=585
Characteristics	14-16 N=29	17-44 N=556	Total N=585
*Education*
≤ 8^th Grade	1 (3.4)	1 (0.2)	2 (0.3)
9^th-11^th Grade	28 (96.6)	48 (8.6)	76 (13.0)
High school/Graduated	0	77 (13.8)	77 (13.2)
Vocational/Technical School	0	9 (1.6)	9 (1.5)
Some college	0	290 (52.2)	290 (49.6)
Finished college	0	95 (17.1)	95 (16.2)
Graduate School	0	36 (6.5)	36 (6.2)
Race*
Asian	2 (6.9)	35 (6.3)	37 (6.3)
American Indian/Alaskan Native	2 (6.9)	8 (1.4)	10 (1.7)
Black	3 (10.3)	55 (9.9)	58 (9.9)
Native Hawaiian or pacific Islander	0	10 (1.8)	10 (1.7)
White	23 (79.3)	424 (79.3)	447 (76.4)
Missing	1 (3.4)	38 (6.8)	39 (6.7)
*Prior EC Use*
Ever	8 (27.6)	226 (40.6)	234 (40)
Never	21 (72.4)	330 (59.4)	351 (60)

Data were extracted from Table 2.2c and g (p027 and p035 of vol. 28). * Some subject listed more than 1 race.

Compliance of Follow-up Contacts (Figure 3, Table 8)

Of the 585 enrolled subjects (576 eligible and 9 ineligible), 42 subjects (7.4%) did not provide follow-up data, constituting the Lost to Follow-up Population; 501 subjects (85%) completed both contacts, but only 262 subjects (45%) provided contact information within the per-protocol time windows (5-8 days for the first contact and ≥ 28 days for the second contact). Contacts were by phone (98%) and in person (2%).

Compliance with the follow-up schedule was less complete for subjects age ≤ 16 years and subjects with less than a high school education (Table 9). There were no significant differences in the follow-up compliance among those of different races, ethnicity and history of emergency contraceptive pill (ECP) use.

The compliance for the 1^st contact was much better at the clinics than at the pharmacies.

A total 322 subjects (55% of enrolled population) completed follow-up contacts (2 contacts and required supplemental contacts) and mailed the Study Data Card.

Figure 3. Incomplete or Lost Follow-up Subjects

The figure was adapted from the sponsor’s Figure 2 in page 071 of vol. 27. “Complete follow-up” was referred to those subjects who completed ≥ 2 follow-up contacts as well as mailed in the study data cards.

Table 8. Compliance of Follow-up Contacts

(% of enrolled population)

Follow-up Contact	Clinical Sites N=550	Pharmacy Sites N=35	All sites N=585
0 Contact	7.1%	11.4%	7.4%
1 Contact	5.3%	0	5.0%
≥ 2 Contacts (Total)	85.5%	88.6%	85.6%
≥ 2 Contacts (per protocol)†	46.5%	17.1%	44.8%
Timing of 1^st Contact
5-8 days	52.5%	17.1%	50.4%
Mean days (Min-Max)	11.0±7.9 (4-81)	21.3±21.6 (6-92)	11.6±9.5 (4-92)
Timing of any 2^nd Contact
≥ 28 days	80.7%	88.6%	81.2%
Mean Days (Min-Max)	32.5±8.0 (12-94)	38.5±14.2 (28-92)	32.9±8.6 (12-94)
Study Data Card Received	57.8%	51.4%	57.4%

Data are extracted from the sponsor’s Table 1.4a (vol. 28, p012).

†The 1^st contact at 5-8 days and the 2^nd contact at ≥28 days

Table 9. Follow-up Compliance in Subjects by Ages and Education Levels

(% of enrolled population)

No. of Follow-up Contacts	Age (years)		Education Levels		Prior EC Use		Total N=585
	14-16	17-44	< HS	≥ HS	Ever	Never
	N=29	N=556	N=78	N=507	N=234	N=351
0	24.1	6.5	19.2	5.5	9	6.3	7.4
1	20.7	4.1	11.5	3.9	4.7	5.1	5.0
2	55.2	87.2	66.7	88.6	85.0	86.0	85.6
≥ 3	0	2.2	2.6	2.0	1.3	2.6	2.1

Data are extracted from Tables 1.4c and 1.4f (vol. 28, p012).

Disagreements between Follow-up Contact and the Study Data Card

Approximately 57% of the enrolled subjects (336/585) returned the Study Data Card. Most of the returned cards (80%) contained at least one discrepancy compared to the follow-up contact. Approximately 50% of the returned cards included adverse event information that differed from the follow-up contact information; however, all AEs from the cards and the follow-up contacts were included in analysis (as seen in Safety section of this review). The overall discrepancy of data (excluding AEs) obtained from the Study Data Card was 41% (138/336) compared to the data from follow-up contacts (Table 10).

Table 10. Discrepancies between data (excluding AEs) obtained from

the Study Data Card and the Follow-up Contacts

(% of 336 subjects who returned the Study Data Card)

Parameters	No Discrepancy	Discrepancy	Data from Card only	Data from Contact only
Onset data of menstrual period before Plan B use	61.9	33.6	1.5	3.0
Date of sex act that caused subject to want Plan B	80.1	13.1	6.3	0.6
Date and time 1^st Plan B pill taken	76.5	19.0	0.9	3.6
Date and time 2^nd Plan B pill taken	69.3	25.0	1.8	3.9
Had menstrual period after taking Plan B	69.3	18.8	0.3	11.6
Onset date of menstrual period after taking Plan B	73.8	19.3	1.5	5.4
Date and results of pregnancy test	81.3	3.6	3.6	11.6
All information combined	20.2	41.1	12.5	26.2

Data were extracted from Table 4.1a in page 094 of vol. 28.

Due to the low return rate of the Study Data Cards, there was greater available data from follow-up contacts for the final analyses. Considering this may have introduced interview bias into the results, the Agency sent a request to the sponsor on Oct 1, 2003 to provide data analyses based on data from the Study Card only. As per the sponsor’s response on Oct 17, 2003, the incorrect use analysis using data recorded on the Study Card was no higher than that from the follow-up contact data as stratified by age, educational levels and Ever/Never EC experience.

Reproductive History

Approximately 68% of the enrolled subjects had no history of pregnancy before the first screening, and 84% had no living children (Table 11).

Table 11. Gravidity History of Enrolled Subjects

(% of enrolled population)

	Clinical Sites N=550	Pharmacy Sites N=35	Total N=585
*Previous Pregnancy*
0	68.4	65.7	68.2
1	18.4	11.4	17.9
≥ 2	12.4	22.9	13.0
Missing	0.9	0.0	0.9
*Living Children*
0	84.7	71.4	83.9
1	8.5	14.3	8.9
≥ 2	5.5	14.3	6.0
Missing	1.3	0	1.2

Data are extracted from the sponsor’s Table 2.6a in page 047 of vol. 28.

Emergency Contraceptive History

About 60% of subjects had no previous experience using an emergency contraception (EC) (Tables 12 and 13). The demographic characteristics of Ever and Never EC users were comparable.

Of those with previous EC experience, 2% of subjects used it within the past month and 8% within the last 1-3 months (Table 14).

Subjects with prior emergency contraceptive use were more likely to have had at least one act of sexual intercourse without contraception in the past month (p<0.05) and to have a positive pregnancy history (p< 0.01). There were no significant differences in condom use between ever EC users and never EC users (Table 15).

Table 12. Demographics of subjects with and without

Previous Emergency Contraceptive Use Experience

(% of the enrolled population)

Characteristics	Ever EC Use	Never EC Use	Total
Characteristics	N=234 (40%)	N=351 (60%)	N=585 (100%)
*Age (years)*
14-16	3.4	6.0	5.0
17-25	75.2	73.8	74.4
26-30	13.7	14.2	14.0
31-35	5.6	3.1	4.1
≥ 36	2.1	2.8	2.6
*Education*
Some College	52.6	47.6	49.6
Finished College	13.7	17.9	16.2
High school	12.0	14.0	13.2
9^th-11^th Grade	12.0	13.7	13.0
Graduate School	8.1	4.8	6.2
Technical School	1.3	1.7	1.5
≤ 8^th Grade	0.4	0.3	0.3
Missing	0	0	0
Race*
White	71.8	79.5	76.4
Black	13.7	7.4	9.9
Asian	5.1	7.1	6.3
Others	3.4	3.4	3.4
Missing	7.3	6.3	6.7
*Marital Status*
Single	86.3	90.9	89.1
Married	7.3	4.6	5.6
Others	6.4	4.2	5.1
Missing	0	0.3	0.2

Data are extracted from the sponsor’s Table 2.2g (vol. 28, p035). Some subjects had more than 1 race.

Table 13. History of Previous Emergency Contraceptive Use

(% of enrolled population)

	Clinical Sites	Pharmacy Site	Total
Never Used	60.7	48.6	60.0
Ever Used	39.3	51.4	40.0
(Times) 1	25.3	31.4	25.6
2	10.4	11.4	10.4
3	2.2	5.7	2.4
≥ 4	0.9	2.9	1.0
Missing	0.5	0	0.5

Data are extracted from the sponsor’s Table 2.6a (vol. 28, p047).

Table 14. Time since the Last Emergency Contraceptive Use

(% of enrolled population)

Months Since Last Use	Clinical Sites N=550	Pharmacy Site N=35	Total N=585
< 1	2.2	2.9	2.2
1-3	7.8	14.3	8.2
> 3	29.3	34.3	29.6
Total	39.3	51.4	40.0

Data were extracted from the sponsor’s Table H (vol. 27, p046) and Table 2.6a (vol. 28, p047).

Table 15. Reproductive and Contraceptive History

of Previous Emergency Contraceptive Users

(% of enrolled population with or without previous ECP use)

Characteristics	Ever ECP Use	Never ECP Use	Total
Characteristics	N=234 (40%)	N=351 (60%)	N=585 (100%)
*Pregnancy History†*
None	62.0	72.4	68.2
1	19.7	16.8	17.9
≥ 2	18.4	9.4	13.0
Missing Data	0	1.4	0.9
*Living Children*
None	80.3	86.3	83.9
1	10.7	7.7	8.9
≥ 2	9.0	4.0	6.0
Missing Data	0	2.0	1.2
*Contraceptive Method*
Condoms	76.5	80.3	78.8
Withdrawal	29.9	26.8	28.0
Oral Contraceptive Pills	22.2	20.5	21.2
Spermicide	7.7	8.5	8.2
Emergency Contraception	5.6	0	2.2
Natural Family Planning	2.6	1.7	2.1
DepoProvera or Lunelle	2.6	1.1	1.7
Other	0.4	0	0.2
At least one sex act without Contraception during past month‡	65.8*	56.1	60.0

Data are extracted from the sponsor’s Table 2.6g (vol. 28, p054).

† p < 0.003 by Kruskal-Wallis test and ‡ p = 0.02 by Chi-square test compared between Ever ECP and Never ECP.

History of Contraception

A total of 536 subjects (92% of the enrolled population) used a contraceptive method in the previous month (Table 16); and the most common method was the condom (79% of the enrolled population). The Lost to Follow-up population and the entire enrolled population were comparable in their histories of contraception, emergency contraception, pregnancy, and sex without contraception.

Table 16. History of Contraceptive Methods in Previous Month

(% of the enrolled population)

Contraceptive Method	Clinical Sites N=550	Pharmacy Sites N=35	Total N=585
Condoms	79.6	65.7	78.8
Withdrawal	29.1	11.4	28.0
Oral Contraceptive Pills	20.9	25.7	21.2
Spermicide	8.5	2.9	8.2
Emergency Contraception	2.2	2.9	2.2
Natural Family Planning	2.2	0	2.1
DepoProvera or Lunelle	1.3	8.6	1.7
Other	0.2	0	0.2
At least one sex act without contraception	60.2	57.1	60.0

Data were extracted from the sponsor’s Table I (vol. 27, p046) and Table 2.6a (vol. 28, p048).

Reasons to request Plan B (Table 17): The major reasons for the 585 enrolled subjects to request Plan B during the recruitment visit were “condom broke/slipped” (37%), “unprotected sex” (33%), “prevent pregnancy” (17%), and “Oral Contraceptive Pills (OCP) problem” (4%). The sponsor stated that subjects with the reason “prevent pregnancy” did not intend to use Plan B before sexual intercourse.

Ninety-seven percent had a correct reason to request Plan B (after excluding those subjects who provided “unspecified” reasons). See Table 17.

Reasons to use Plan B (Table 18): Approximately 98% of 540 subjects who used Plan B after enrollment had single (91%) or multiple sex acts (7%). The main factors to prompt the subjects to take Plan B were “condom broke/slipped”, “used no contraception”, and “missed OCPs”. Correct self-selection calculated by exclusion of the reasons “Other” and “Doesn’t remember/missing data” was 95% (514 of 540 users).

Overall, the factors prompting subjects to seek and use Plan B were similar among those of different ages, races, ethnicity, educational levels, or ever/never EC use experience (Tables 17 and 18).

Procedure	Screening	Week 1 (5-8 days)	Week 4 (28 days or later	Weekly thereafter (if necessary)
Collect baseline and background data	X
Determine eligibility	X
Informed consent	X
Provision of Plan B if eligible	X
Provision of Study Data Card	X
Collect information on use of product		X	X
Collect information on adverse events		X	X	X
Collect information on pregnancy		X	X	X

Reason to Request Plan B	Study Sites		Eligibility		Age (year)		Education		Prior EC Use		Total N=585
Reason to Request Plan B	Clinics N=550	Pharmacy N=35	Ineligible N=9	Eligible N=576	≤ 16 N=29	≥ 17 N=556	< HS N=78	≥ HS N=507	Ever N=234	Never N=351	Total N=585
Condom broke or slipped	37.1	42.9	44.4	37.3	31.0	37.8	34.6	37.9	32.5	40.7	37.4
Unprotected sex	33.3	31.4	11.1	33.5	27.6	33.5	23.1	34.7	37.6	30.2	33.2
Prevent pregnancy	17.1	14.3	11.1	17.0	27.6	16.4	28.2	15.2	18.8	15.7	16.9
OCP problem	3.8	5.7	11.1	3.8	3.4	4.0	5.1	3.7	3.8	4.0	3.9
Mistake/accident	2.4	0.0	11.1	2.1	6.9	2.0	3.8	2.0	2.1	2.3	2.2
Contraceptive failure (unspecified)	1.8	0.0	0.0	1.7	0.0	1.8	0.0	2.0	1.7	1.7	1.7
Withdrawal	1.5	0.0	11.1	1.2	0.0	1.4	2.6	1.2	0.0	2.3	1.4

Table 3. Subjects Screened and Enrolled in Each Study Site

Clinical Site

Pharmacy Sites

Total

Boston

Grand Rapid

Houston

Phoenix

Seattle

Total

Screened Subjects

129

125

170

123

79

626

39

665

Eligible Subjects

111

117

138

111

72

549

36

585

86.0%

93.6%

81.2%

90.2%

91.1%

87.7%

92.3%

88.0%

Ineligible Subjects

18

8

32

12

7

77

3

80

14.0%

6.4%

18.8%

9.8%

8.9%

12.3%

7.7%

12.0%

Figure 2. Disposition of Subjects

(Adapted from the sponsor’s Figure 1 in page 070 of vol. 27)

(Adapted from the sponsor’s Table F in page 44 of vol. 27)

Figure 3. Incomplete or Lost Follow-up Subjects

Follow-up Contact

Clinical Sites

Pharmacy Sites

All sites

0 Contact

7.1%

11.4%

7.4%

1 Contact

5.3%

0

5.0%

≥ 2 Contacts (Total)

85.5%

88.6%

85.6%

≥ 2 Contacts

(per protocol)†

46.5%

17.1%

44.8%

Timing of 1st Contact

5-8 days

Timing of 1^st Contact

Timing of any 2^nd Contact

†The 1^st contact at 5-8 days and the 2^nd contact at ≥28 days