A Formative Evaluation of Two FASD Prevention Communication Strategies

A Formative Evaluation of Two FASD Prevention Communication Strategies Abstract Aims To evaluate the feasibility, acceptability and effectiveness of placing FASD prevention messages in the women’s restrooms of establishments serving alcohol in Alaska and the Yukon, regions with high rates of FASD. Methods Our team placed an FASD educational poster, and posters affixed to a pregnancy test dispenser, in women’s restrooms of bars and restaurants. We compared drinking behaviors and knowledge and beliefs about FASD among participants at baseline and at follow-up. Results Respondents consisted of 2132 women who completed a baseline survey and 1182 women who completed both a baseline and a follow-up survey. Women in both groups showed improvement in knowledge of FASD; the dispenser group scored higher than participants in the poster group on the FASD Health Belief questions at both baseline and follow-up. Forty-three women learned they were pregnant from our pregnancy tests and alcohol consumption among pregnant women was lower at follow-up than at baseline. Conclusions FASD prevention messages, particularly paired with pregnancy test dispensers, in the women’s restrooms of establishments that serve alcohol can effectively promote informed alcohol consumption decisions among women who are, or may become, pregnant. Short Summary In this FASD prevention feasibility study, we found that FASD prevention messages, particularly paired with pregnancy test dispensers, placed in the women’s restrooms of establishments that serve alcohol can effectively promote informed alcohol consumption decisions among women who are, or may become, pregnant. INTRODUCTION Alcohol use during pregnancy is the leading cause of preventable birth defects and developmental disabilities in the United States (Tan et al., 2015). Fetal alcohol syndrome (FAS) has been associated with a variety of risk factors, including binge drinking; advanced maternal age; high gravidity and parity; unstable marital status; and use of other substances including tobacco (May et al., 2005). The Institute of Medicine (Stratton et al., 1996) defines fetal alcohol spectrum disorder (FASD) as an umbrella term encompassing four discrete diagnoses that comprise a clinical spectrum from severe to mild (fetal alcohol syndrome, partial fetal alcohol syndrome, alcohol-related neurodevelopmental disorder and alcohol-related birth defects). Severity of FASD has been associated with the amount of alcohol consumed during gestation, modified by certain maternal characteristics (Maier and West, 2001; Diagnosis, 2014). The U.S. Surgeon General (Notice to readers: Surgeon General’s advisory on alcohol use in pregnancy, 2005), Centers for Disease Control and Prevention (Alcohol Use in Pregnancy, 2016), and the American Academy of Pediatrics (Williams and Smith, 2015) recommend that pregnant women and women who are considering pregnancy, abstain from alcohol consumption to eliminate the risk of FASD. Historically Alaska has had the highest rate of FASD among states that track these outcomes (Miller et al., 2002). However, more recent data suggest that Alaska, with 1.5 FAS cases per 1000 live births, falls within the national average 1–3 cases of FAS per 1000 births (Camerlin, 2012). A conservative estimate for the rate of all FASD cases in Alaska is still extremely high with 81.8 per 10,000 live births based on cases reported to the Alaska Birth Defects Registry. A substantial number of Alaskan women may have an unintended alcohol-exposed pregnancy due to lack of awareness of either their pregnancy status or the risks of alcohol consumption during pregnancy. Approximately half of all pregnancies nationally and in Alaska are unintended (Kost, 2015). Over half of all women in Alaska report consuming alcohol (State-Specific Alcohol Consumption Rates for 2013, 2015). In a national study, 22% of women reported consuming alcohol during their first month of pregnancy, 8% during the second and 5% during the third (Ethen et al., 2009). In 2010, 7% of Alaskan mothers reported that they consumed alcohol throughout their pregnancy (PRAMStat: Explore PRAMS Data by State, 2010). Response Many FASD prevention interventions take the form of health education messages using public service announcements and other forms of mass media. These interventions, which fall into the category of universal prevention strategies, are designed to reach the entire population without regard to individual risk factors (Gordon, 1983). There are few examples of FASD prevention campaigns that employ either a selective strategy focusing on subgroups at higher risk or an indicated strategy targeting individuals with a specific risk factor that puts them at higher risk for FASD. The most commonly known prevention programs are CHOICES, which focus on pregnancy prevention through contraception, and Parent-Child Assistance Program which implements prevention strategies after a woman finds out she is pregnant. Although these programs are beneficial, additional prevention efforts are needed for the highly vulnerable population of women who are consuming alcohol and are unaware of their pregnancy status. In one such strategy implemented in Mankato, Minnesota, a pregnancy test dispenser featuring information about the risks of FASD was placed in the women’s restroom of an establishment serving alcohol (Capecchi, 2012). For a minimal charge ($3.00 on a debit card), a female patron could purchase a pregnancy test and determine her pregnancy status prior to consuming alcohol. The goal was to provide both information about FASD and the means to prevent it. This prevention strategy has not been evaluated. Objectives This paper presents the results of a formative evaluation designed to assess the utility of two innovative FASD prevention strategies. The specific objectives for this study were as follows: To develop two FASD prevention strategies differentiated by the media on which the message is displayed. To assess the feasibility of implementing the FASD prevention strategies in women’s restrooms of establishments that serve alcohol, and; To assess the acceptability and compare the effectiveness of these strategies for promoting informed alcohol consumption decisions among women who are pregnant, or may become pregnant. We evaluate the relationship between the two FASD prevention strategies and women’s drinking behavior and knowledge and beliefs related to alcohol consumption at baseline and 6 months following exposure to each intervention. MATERIALS AND METHODS This formative evaluation study used a matched-site comparison design. We posted the same prevention message in women’s restrooms at establishments serving alcohol in seven communities in Alaska, and one in the Yukon Territory. Intervention communities were matched by racial and ethnic composition, number of establishments serving alcohol, and connections to rural and remote communities (the latter being not connected to any other community by road). Bar recruitment Bars in each study community were recruited by community liaisons with local knowledge of the service industry and regional context. The liaison initiated contacts with owners and managers of establishments serving alcohol in each study community. The liaison then coordinated placement of the health communication message using either a pregnancy test dispenser or a poster as the medium of communication (hereafter referred to as the interventions) and managed the intervention as necessary during data collection, which lasted approximately 1 year. Liaisons met regularly with owners, managers and staff to discuss issues or challenges as they arose throughout the year and at the conclusion of data collection they removed the interventions. Message development An expert advisory panel (EAP) of FASD prevention experts including obstetrics practitioners, certified nurse midwives and family planning counselors from study communities developed the draft educational message. The panel recommended that the message be clear, creative and personal to increase engagement. The EAP members worked over the course of 3 months to develop a draft message that met these criteria, while avoiding the appearance of stigma for mothers of children with FASD. The draft message developed by the EAP was tested and refined with input from a review panel of women aged 21–44 years of age who consume alcohol. Each intervention message included a toll-free number to a regional family planning counseling service. It also included a discrete toll-free number, web address and quick response (QR) code that allowed users to link to our electronic survey. All interventions were paired with a condom dispenser to provide a readily-accessible form of birth control. The final message consisted of three elements: any alcohol use during pregnancy can cause lifelong health problems for the child; the prospective adverse health outcomes associated with FASD are severe and difficult to address; and a woman should be aware of her potential pregnancy status prior to consuming alcohol (Fig. 1). Fig. 1. View largeDownload slide FASD Informational Poster. Fig. 1. View largeDownload slide FASD Informational Poster. Six sites featured the message in a framed poster, and eight featured the message affixed to a pregnancy test dispenser. The dispensers were programmed to delay 60-s between the dispensing of individual pregnancy tests to reduce the potential for one user to withdraw all the pregnancy tests in the unit at one time. A message on the pregnancy test kit sleeve warned recipients that a pregnancy test conducted within 1 week after a missed period could be misleading, and recommended further testing in such cases. Pregnancy tests dispensed in the Yukon Territory required a small cost of two Canadian dollars. Survey description Surveys were completed by participants via text messaging or web access to assess the acceptability of the two types of communication interventions. We asked participants in the dispenser group if they obtained a pregnancy test from the dispenser. Participants who reported obtaining a test were also asked whether they had used it, and if so, the results of that test. All participants who reported not using a pregnancy test from a study dispenser were asked if they knew their pregnancy status. Participants who reported being pregnant were asked to describe their alcohol consumption prior to and after learning their status. Participants who responded that they were either not pregnant or were unsure of their status were asked if there had been a time during the past 6 months when they might have been pregnant. These participants were asked to respond to similar questions about alcohol consumption as those who were pregnant. We assessed alcohol consumption by asking participants if they had consumed alcohol in the past 30 days, and if so, the number of drinking days per week and the amount of alcohol consumed on days when they drank. Finally, a series of eight questions were asked concerning knowledge and beliefs about FASD risk using a 5 point Likert-scale ranging from 1 Strongly Disagree to 5 Strongly Agree. Two questions were reverse coded for consistency in data analysis. Questions with the lowest corrected item-to-total correlation were removed individually until the inter-item reliability was maximized. The corrected item-to-total correlation for each question is recommended to be r> 0.30 (Windsor, 2015). Based on standard psychometric analyses we restricted our analysis to the following five questions: As long as a woman does not drink too much alcohol, it is safe to have 1–2 alcoholic drinks at one time when pregnant. (Reverse coded) Evidence shows that drinking alcohol during pregnancy affects the health of the child after he or she is born. Drinking alcohol during pregnancy will harm the mother’s health more than the unborn child’s health. (Reverse coded) Raising a child with a Fetal Alcohol Spectrum Disorder is hard. If a woman drinks alcohol before she knows she is pregnant, her unborn child could still be affected. Using responses to these five questions a cumulative score was computed to evaluate participant’s beliefs about FASD risk. A higher score indicates a more accurate understanding of FASD risks and harms. All participants ended the survey with general demographic information concerning ethnicity, age and education. Participants were asked to select all appropriate races or ethnicities which included White, Black/African American/Negro, American Indian or Alaskan Native, Asian, Native Hawaiian or Pacific Islander, Hispanic/Latino/Other Spanish and Other. Participants who selected other were able to enter in their own race or ethnicity as text. For analysis we recoded ethnicity combining participants who selected Other with participants who selected multiple races. We sent all participants an invitation to participate in a follow-up survey 6 months after completing the baseline survey. The follow-up survey differed from the baseline only with the addition of a question regarding the use of community health resources and the removal of demographic and pregnancy tests questions. Participants received a $25 gift certificate to an online retailer for completing each of the two surveys. Impact analyses The effectiveness of the two messaging strategies was evaluated by comparing differences in alcohol consumption and knowledge and beliefs about FASD between the dispenser and poster intervention groups at baseline and follow-up. We also assessed changes in alcohol consumption from baseline to follow-up for participants who used a pregnancy test from one of our test dispensers and reported a positive test result, or who reported being pregnant or possibly pregnant at baseline. To minimize duplicate survey responses, we developed a protocol for determining level of suspicion that a survey was a duplicate based on duplicate e-mail addresses or similar e-mail addresses from the same IP address. Surveys might also be deemed suspicious if they used a link that had been replaced or removed during the study period. Highly suspicious surveys were deleted; less suspicious surveys were verified by e-mail and only included if the participant confirmed having submitted only one survey. A total of 207 duplicates were excluded from the analyses. The statistical significance of group comparisons was assessed using Chi-square or non-parametric Mann Whitney tests as appropriate, with a significance level of P < 0.05. We used ordinal regression to adjust for age, education and ethnicity for differences that were statistically significant. Goodness of fit for regression models was checked using Chi-Square with alpha set at P > 0.05. The study was approved by the University of Alaska Anchorage’s Institutional Review Board. RESULTS Intervention feasibility Site selection, recruitment and intervention placement took place over the first 6 months of 2015. Community liaisons initiated contacts with establishments in eight study communities. The communities assigned to the dispenser intervention were: the Anchorage, Alaska metropolitan area, the rural towns of Homer, Alaska and Whitehorse, Yukon, and the remote town of Kodiak, Alaska. Communities assigned to the poster intervention were: the Juneau, Alaska and Fairbanks, Alaska metropolitan areas, and the remote towns of Dillingham and Nome. Fairbanks, which is on the road system, and Juneau, which is off the road system, are smaller metropolitan areas than Anchorage but larger than Homer and Whitehorse. Within the four dispenser communities, eight bars agreed to place FASD prevention messages on pregnancy test dispensers in women’s restrooms. Liaisons reported that business owners or managers who refused participation did so based on concerns regarding the sexual nature of the message. Among those who allowed us to install pregnancy test dispensers, establishment owners, managers and staff described an initial phase of the intervention in which condoms and pregnancy tests were removed from the dispensers but left at the table, employed for jokes or pranks, or otherwise wasted by patrons. Some owners placed signage alongside the dispensers to formally express support for the study and request that the tests and condoms not be wasted. Staff at other establishments employed informal social sanctions such as disapproving comments or lack of a humorous response to the patron’s behavior. Although the waste gradually ceased, staff suggested that if the condoms and pregnancy tests had some minimal cost, perhaps $1.00 each, this behavior might be avoided. Owners, managers and staff in all study communities reported that the overwhelming number of responses to the interventions were positive. These comments most often took the form of support for the goal of preventing FASD, and gratitude for the condoms and pregnancy tests. Wait staff reported that younger female patrons were most likely to be interested and engaged by the message. The rare negative comments were expressed by patrons concerned with the sexual component of the message. Owners expressed the opinion that interventions were associated with increased traffic and sales. Two business owners took steps to continue the distribution of no-cost condoms at the conclusion of the study. Three wait staff passed on personal narratives from women who stopped drinking alcohol after receiving a positive pregnancy test result at the establishment. Participants The baseline survey was completed by 2132 women of childbearing age. We sent 1975 (92.6%) follow-up invitation emails; 75 of baseline participants did not provide an e-mail address and 82 emails were returned as undeliverable. Of those remaining women, 1182 (60%) completed a follow-up survey. As shown in Table 1, women completing the baseline survey who viewed the FASD prevention messages on the dispenser were older (P = 0.01), less educated and more likely to be a minority than the poster group (P < 0.01 for both). Similar differences were observed among women completing the follow-up survey (P < 0.01 for each). Table 1. Demographics of baseline participants, follow-up participants and follow-up nonparticipants Baseline Follow-up Follow-up nonparticipants Dispenser Poster Total Dispenser Poster Total Dispenser Poster Total Measure % (n) % (n) % (n) % (n) % (n) % (n) % (n) % (n) % (n) Age  21 to 25 34.4 (570) 43.0 (188) 36.2 (758) 30.0 (286) 45.0 (103) 32.9 (389) 40.5 (284) 40.9 (85) 40.5 (369)  26 to 30 33.2 (550) 30.9 (135) 32.7 (685) 37.5 (357) 31.0 (71) 36.2 (428) 27.5 (193) 30.8 (64) 28.2 (257)  31 to 35 20.0 (331) 16.9 (74) 19.4 (405) 21.3 (203) 16.6 (38) 20.4 (241) 18.2 (128) 17.3 (36) 18.0 (164)  36 to 40 8.2 (135) 5.7 (25) 7.6 (160) 7.2 (69) 5.7 (13) 6.9 (82) 9.4 (66) 5.8 (12) 8.6 (78)  41 to 44 4.2 (69) 3.4 (15) 4.0 (84) 4.0 (38) 1.7 (4) 3.6 (42) 4.4 (31) 5.3 (11) 4.6 (42) Education  Some High School 3.9 (64) 3.9 (17) 3.9 (81) 2.7 (26) 2.6 (6) 2.7 (32) 5.4 (38) 5.4 (11) 5.4 (49)  Diploma or GED 18.6 (308) 10.7 (46) 17.0 (354) 16.3 (155) 8.3 (19) 14.7 (174) 21.8 (153) 13.4 (27) 19.9 (180)  Some College 42.1 (696) 34.8 (150) 40.6 (846) 42.9 (409) 28.4 (65) 40.1 (474) 40.9 (287) 42.1 (85) 41.2 (372)  Bachelors 25.1 (415) 32.5 (140) 26.6 (555) 27.0 (257) 38.4 (88) 29.2 (345) 22.5 (158) 25.7 (52) 23.3 (210)  Graduate 10.3 (171) 18.1 (78) 11.9 (249) 11.1 (106) 22.3 (51) 13.3 (157) 9.3 (65) 13.4 (27) 10.2 (92) Race or Ethnicity  White 58.5 (966) 70.5 (304) 61.0 (1270) 61.1 (581) 74.1 (169) 63.6 (750) 55.1 (385) 66.5 (135) 57.6 (520)  Blacka 5.2 (86) 6.3 (27) 5.4 (113) 4.9 (47) 5.3 (12) 5.0 (59) 5.6 (39) 7.4 (15) 6.0 (54)  AIANb 6.8 (113) 5.6 (24) 6.6 (137) 5.8 (55) 5.7 (13) 5.8 (68) 8.3 (58) 5.4 (11) 7.6 (69)  Asian 8.7 (143) 6.3 (27) 8.2 (170) 8.8 (84) 5.3 (12) 8.1 (96) 8.4 (59) 7.4 (15) 8.2 (74)  NHOPIc 2.8 (46) 1.4 (6) 2.5 (52) 2.0 (19) 2.6 (6) 2.1 (25) 3.9 (27) 0.0 (0) 3.0 (27)  Hispanicd 5.5 (91) 3.7 (16) 5.1 (107) 5.8 (55) 2.6 (6) 5.2 (61) 5.2 (36) 4.9 (10) 5.1 (46)  Other or Multiple 12.4 (205) 6.3 (27) 11.1 (232) 11.6 (110) 4.4 (10) 10.2 (120) 13.6 (95) 8.4 (17) 12.4 (112) Baseline Follow-up Follow-up nonparticipants Dispenser Poster Total Dispenser Poster Total Dispenser Poster Total Measure % (n) % (n) % (n) % (n) % (n) % (n) % (n) % (n) % (n) Age  21 to 25 34.4 (570) 43.0 (188) 36.2 (758) 30.0 (286) 45.0 (103) 32.9 (389) 40.5 (284) 40.9 (85) 40.5 (369)  26 to 30 33.2 (550) 30.9 (135) 32.7 (685) 37.5 (357) 31.0 (71) 36.2 (428) 27.5 (193) 30.8 (64) 28.2 (257)  31 to 35 20.0 (331) 16.9 (74) 19.4 (405) 21.3 (203) 16.6 (38) 20.4 (241) 18.2 (128) 17.3 (36) 18.0 (164)  36 to 40 8.2 (135) 5.7 (25) 7.6 (160) 7.2 (69) 5.7 (13) 6.9 (82) 9.4 (66) 5.8 (12) 8.6 (78)  41 to 44 4.2 (69) 3.4 (15) 4.0 (84) 4.0 (38) 1.7 (4) 3.6 (42) 4.4 (31) 5.3 (11) 4.6 (42) Education  Some High School 3.9 (64) 3.9 (17) 3.9 (81) 2.7 (26) 2.6 (6) 2.7 (32) 5.4 (38) 5.4 (11) 5.4 (49)  Diploma or GED 18.6 (308) 10.7 (46) 17.0 (354) 16.3 (155) 8.3 (19) 14.7 (174) 21.8 (153) 13.4 (27) 19.9 (180)  Some College 42.1 (696) 34.8 (150) 40.6 (846) 42.9 (409) 28.4 (65) 40.1 (474) 40.9 (287) 42.1 (85) 41.2 (372)  Bachelors 25.1 (415) 32.5 (140) 26.6 (555) 27.0 (257) 38.4 (88) 29.2 (345) 22.5 (158) 25.7 (52) 23.3 (210)  Graduate 10.3 (171) 18.1 (78) 11.9 (249) 11.1 (106) 22.3 (51) 13.3 (157) 9.3 (65) 13.4 (27) 10.2 (92) Race or Ethnicity  White 58.5 (966) 70.5 (304) 61.0 (1270) 61.1 (581) 74.1 (169) 63.6 (750) 55.1 (385) 66.5 (135) 57.6 (520)  Blacka 5.2 (86) 6.3 (27) 5.4 (113) 4.9 (47) 5.3 (12) 5.0 (59) 5.6 (39) 7.4 (15) 6.0 (54)  AIANb 6.8 (113) 5.6 (24) 6.6 (137) 5.8 (55) 5.7 (13) 5.8 (68) 8.3 (58) 5.4 (11) 7.6 (69)  Asian 8.7 (143) 6.3 (27) 8.2 (170) 8.8 (84) 5.3 (12) 8.1 (96) 8.4 (59) 7.4 (15) 8.2 (74)  NHOPIc 2.8 (46) 1.4 (6) 2.5 (52) 2.0 (19) 2.6 (6) 2.1 (25) 3.9 (27) 0.0 (0) 3.0 (27)  Hispanicd 5.5 (91) 3.7 (16) 5.1 (107) 5.8 (55) 2.6 (6) 5.2 (61) 5.2 (36) 4.9 (10) 5.1 (46)  Other or Multiple 12.4 (205) 6.3 (27) 11.1 (232) 11.6 (110) 4.4 (10) 10.2 (120) 13.6 (95) 8.4 (17) 12.4 (112) aBlack/African American/Negro. bAmerican Indian or Alaskan Native. cNative Hawaiian and Other Pacific Islander. dHispanic/Latino/Other Spanish. Table 1. Demographics of baseline participants, follow-up participants and follow-up nonparticipants Baseline Follow-up Follow-up nonparticipants Dispenser Poster Total Dispenser Poster Total Dispenser Poster Total Measure % (n) % (n) % (n) % (n) % (n) % (n) % (n) % (n) % (n) Age  21 to 25 34.4 (570) 43.0 (188) 36.2 (758) 30.0 (286) 45.0 (103) 32.9 (389) 40.5 (284) 40.9 (85) 40.5 (369)  26 to 30 33.2 (550) 30.9 (135) 32.7 (685) 37.5 (357) 31.0 (71) 36.2 (428) 27.5 (193) 30.8 (64) 28.2 (257)  31 to 35 20.0 (331) 16.9 (74) 19.4 (405) 21.3 (203) 16.6 (38) 20.4 (241) 18.2 (128) 17.3 (36) 18.0 (164)  36 to 40 8.2 (135) 5.7 (25) 7.6 (160) 7.2 (69) 5.7 (13) 6.9 (82) 9.4 (66) 5.8 (12) 8.6 (78)  41 to 44 4.2 (69) 3.4 (15) 4.0 (84) 4.0 (38) 1.7 (4) 3.6 (42) 4.4 (31) 5.3 (11) 4.6 (42) Education  Some High School 3.9 (64) 3.9 (17) 3.9 (81) 2.7 (26) 2.6 (6) 2.7 (32) 5.4 (38) 5.4 (11) 5.4 (49)  Diploma or GED 18.6 (308) 10.7 (46) 17.0 (354) 16.3 (155) 8.3 (19) 14.7 (174) 21.8 (153) 13.4 (27) 19.9 (180)  Some College 42.1 (696) 34.8 (150) 40.6 (846) 42.9 (409) 28.4 (65) 40.1 (474) 40.9 (287) 42.1 (85) 41.2 (372)  Bachelors 25.1 (415) 32.5 (140) 26.6 (555) 27.0 (257) 38.4 (88) 29.2 (345) 22.5 (158) 25.7 (52) 23.3 (210)  Graduate 10.3 (171) 18.1 (78) 11.9 (249) 11.1 (106) 22.3 (51) 13.3 (157) 9.3 (65) 13.4 (27) 10.2 (92) Race or Ethnicity  White 58.5 (966) 70.5 (304) 61.0 (1270) 61.1 (581) 74.1 (169) 63.6 (750) 55.1 (385) 66.5 (135) 57.6 (520)  Blacka 5.2 (86) 6.3 (27) 5.4 (113) 4.9 (47) 5.3 (12) 5.0 (59) 5.6 (39) 7.4 (15) 6.0 (54)  AIANb 6.8 (113) 5.6 (24) 6.6 (137) 5.8 (55) 5.7 (13) 5.8 (68) 8.3 (58) 5.4 (11) 7.6 (69)  Asian 8.7 (143) 6.3 (27) 8.2 (170) 8.8 (84) 5.3 (12) 8.1 (96) 8.4 (59) 7.4 (15) 8.2 (74)  NHOPIc 2.8 (46) 1.4 (6) 2.5 (52) 2.0 (19) 2.6 (6) 2.1 (25) 3.9 (27) 0.0 (0) 3.0 (27)  Hispanicd 5.5 (91) 3.7 (16) 5.1 (107) 5.8 (55) 2.6 (6) 5.2 (61) 5.2 (36) 4.9 (10) 5.1 (46)  Other or Multiple 12.4 (205) 6.3 (27) 11.1 (232) 11.6 (110) 4.4 (10) 10.2 (120) 13.6 (95) 8.4 (17) 12.4 (112) Baseline Follow-up Follow-up nonparticipants Dispenser Poster Total Dispenser Poster Total Dispenser Poster Total Measure % (n) % (n) % (n) % (n) % (n) % (n) % (n) % (n) % (n) Age  21 to 25 34.4 (570) 43.0 (188) 36.2 (758) 30.0 (286) 45.0 (103) 32.9 (389) 40.5 (284) 40.9 (85) 40.5 (369)  26 to 30 33.2 (550) 30.9 (135) 32.7 (685) 37.5 (357) 31.0 (71) 36.2 (428) 27.5 (193) 30.8 (64) 28.2 (257)  31 to 35 20.0 (331) 16.9 (74) 19.4 (405) 21.3 (203) 16.6 (38) 20.4 (241) 18.2 (128) 17.3 (36) 18.0 (164)  36 to 40 8.2 (135) 5.7 (25) 7.6 (160) 7.2 (69) 5.7 (13) 6.9 (82) 9.4 (66) 5.8 (12) 8.6 (78)  41 to 44 4.2 (69) 3.4 (15) 4.0 (84) 4.0 (38) 1.7 (4) 3.6 (42) 4.4 (31) 5.3 (11) 4.6 (42) Education  Some High School 3.9 (64) 3.9 (17) 3.9 (81) 2.7 (26) 2.6 (6) 2.7 (32) 5.4 (38) 5.4 (11) 5.4 (49)  Diploma or GED 18.6 (308) 10.7 (46) 17.0 (354) 16.3 (155) 8.3 (19) 14.7 (174) 21.8 (153) 13.4 (27) 19.9 (180)  Some College 42.1 (696) 34.8 (150) 40.6 (846) 42.9 (409) 28.4 (65) 40.1 (474) 40.9 (287) 42.1 (85) 41.2 (372)  Bachelors 25.1 (415) 32.5 (140) 26.6 (555) 27.0 (257) 38.4 (88) 29.2 (345) 22.5 (158) 25.7 (52) 23.3 (210)  Graduate 10.3 (171) 18.1 (78) 11.9 (249) 11.1 (106) 22.3 (51) 13.3 (157) 9.3 (65) 13.4 (27) 10.2 (92) Race or Ethnicity  White 58.5 (966) 70.5 (304) 61.0 (1270) 61.1 (581) 74.1 (169) 63.6 (750) 55.1 (385) 66.5 (135) 57.6 (520)  Blacka 5.2 (86) 6.3 (27) 5.4 (113) 4.9 (47) 5.3 (12) 5.0 (59) 5.6 (39) 7.4 (15) 6.0 (54)  AIANb 6.8 (113) 5.6 (24) 6.6 (137) 5.8 (55) 5.7 (13) 5.8 (68) 8.3 (58) 5.4 (11) 7.6 (69)  Asian 8.7 (143) 6.3 (27) 8.2 (170) 8.8 (84) 5.3 (12) 8.1 (96) 8.4 (59) 7.4 (15) 8.2 (74)  NHOPIc 2.8 (46) 1.4 (6) 2.5 (52) 2.0 (19) 2.6 (6) 2.1 (25) 3.9 (27) 0.0 (0) 3.0 (27)  Hispanicd 5.5 (91) 3.7 (16) 5.1 (107) 5.8 (55) 2.6 (6) 5.2 (61) 5.2 (36) 4.9 (10) 5.1 (46)  Other or Multiple 12.4 (205) 6.3 (27) 11.1 (232) 11.6 (110) 4.4 (10) 10.2 (120) 13.6 (95) 8.4 (17) 12.4 (112) aBlack/African American/Negro. bAmerican Indian or Alaskan Native. cNative Hawaiian and Other Pacific Islander. dHispanic/Latino/Other Spanish. Women who completed the follow-up were more educated and older than those who did not complete follow-up (P < 0.01 for education and age). There were no significant differences in ethnicity between the two groups. Survey data collection was challenged early in the process by a small number of participants who submitted multiple responses. Our team followed-up on questionable survey responses with a confirmation e-mail requesting additional information. If no responses were received, or the responses indicated a duplicate survey, they were removed from the sample. Alcohol consumption Women who completed the follow-up survey were more likely to report drinking alcohol in the past 30 days at baseline than those who did not complete the follow-up survey. Among women who reported drinking any alcohol in the past 30 days at baseline, those who completed the follow-up survey reported fewer drinks per drinking day at baseline than those who did not complete follow-up (P = 0.02). There were no statistically significant differences between follow-up participants and nonparticipants in the average number of drinking days or the number of binge drinking occasions reported at baseline. Table 2 presents the patterns of alcohol consumption among participants at both baseline and follow-up. The dispenser group was more likely to report alcohol consumption in the past 30 days both at baseline and follow-up after adjusting for age, education and ethnicity (P < 0.01 for both baseline and follow-up). After adjusting for age, education and ethnicity when the sample was restricted to only participants who completed follow-up the outcome was still statistically significant. Table 2. Alcohol consumption of all participants at baseline and follow-up Baseline Follow-up Dispenser Poster Total Dispenser Poster Total Measure % (n) % (n) % (n) % (n) % (n) % (n) Consumed alcohol in the past 30 days Pa = 0.05 Pa = 0.10  Yes 66.0 (1092) 60.8 (265) 64.9 (1357) 66.5 (630) 60.7 (139) 65.4 (769)  No 34.0 (563) 39.2 (171) 35.1 (734) 33.5 (317) 39.3 (90) 34.6 (407) Days per week Pa < 0.01 Pa = 0.25  1 or fewer 38.2 (416) 28.3 (75) 36.2 (491) 40.4 (254) 34.8 (48) 39.4 (302)  2 to 3 42.8 (467) 46.0 (122) 43.5 (589) 42.3 (266) 50.0 (69) 43.7 (335)  4 or more 19.0 (207) 25.7 (68) 20.3 (275) 17.3 (109) 15.2 (21) 16.9 (130) Average Number of drinks per occasion Pa = 0.19 Pa = 0.63  1 or fewer drinks 26.7 (281) 20.7 (53) 25.5 (334) 27.4 (165) 31.2 (43) 28.1 (208)  More than 1 to less than 4 drinks 54.4 (573) 56.3 (144) 54.7 (717) 54.9 (331) 50.0 (69) 54.0 (400)  4 to less than 5 drinks 7.4 (78) 8.6 (22) 7.6 (100) 9.5 (57) 11.6 (16) 9.9 (73)  5 or more drinks 11.6 (122) 14.5 (37) 12.1 (159) 8.3 (50) 7.2 (10) 8.1 (60) Number of occasions 4 or more drinks were consumed Pa = 0.16 Pa = 0.62  0 33.5 (366) 26.3 (70) 32.1 (436) 40.3 (253) 38.1 (53) 39.9 (306)  1 to 4 50.6 (553) 53.8 (143) 51.3 (696) 49.4 (310) 48.2 (67) 49.2 (377)  5 to 8 9.0 (98) 12.8 (34) 9.7 (132) 6.8 (43) 8.6 (12) 7.2 (55)  9 to 12 3.6 (39) 3.4 (9) 3.5 (48) 1.8 (11) 2.2 (3) 1.8 (14)  13 to 16 1.5 (16) 1.1 (3) 1.4 (19) 0.3 (2) 1.4 (2) 0.5 (4)  ≥17 1.8 (20) 2.6 (7) 2.0 (27) 1.4 (9) 1.4 (2) 1.4 (11) Baseline Follow-up Dispenser Poster Total Dispenser Poster Total Measure % (n) % (n) % (n) % (n) % (n) % (n) Consumed alcohol in the past 30 days Pa = 0.05 Pa = 0.10  Yes 66.0 (1092) 60.8 (265) 64.9 (1357) 66.5 (630) 60.7 (139) 65.4 (769)  No 34.0 (563) 39.2 (171) 35.1 (734) 33.5 (317) 39.3 (90) 34.6 (407) Days per week Pa < 0.01 Pa = 0.25  1 or fewer 38.2 (416) 28.3 (75) 36.2 (491) 40.4 (254) 34.8 (48) 39.4 (302)  2 to 3 42.8 (467) 46.0 (122) 43.5 (589) 42.3 (266) 50.0 (69) 43.7 (335)  4 or more 19.0 (207) 25.7 (68) 20.3 (275) 17.3 (109) 15.2 (21) 16.9 (130) Average Number of drinks per occasion Pa = 0.19 Pa = 0.63  1 or fewer drinks 26.7 (281) 20.7 (53) 25.5 (334) 27.4 (165) 31.2 (43) 28.1 (208)  More than 1 to less than 4 drinks 54.4 (573) 56.3 (144) 54.7 (717) 54.9 (331) 50.0 (69) 54.0 (400)  4 to less than 5 drinks 7.4 (78) 8.6 (22) 7.6 (100) 9.5 (57) 11.6 (16) 9.9 (73)  5 or more drinks 11.6 (122) 14.5 (37) 12.1 (159) 8.3 (50) 7.2 (10) 8.1 (60) Number of occasions 4 or more drinks were consumed Pa = 0.16 Pa = 0.62  0 33.5 (366) 26.3 (70) 32.1 (436) 40.3 (253) 38.1 (53) 39.9 (306)  1 to 4 50.6 (553) 53.8 (143) 51.3 (696) 49.4 (310) 48.2 (67) 49.2 (377)  5 to 8 9.0 (98) 12.8 (34) 9.7 (132) 6.8 (43) 8.6 (12) 7.2 (55)  9 to 12 3.6 (39) 3.4 (9) 3.5 (48) 1.8 (11) 2.2 (3) 1.8 (14)  13 to 16 1.5 (16) 1.1 (3) 1.4 (19) 0.3 (2) 1.4 (2) 0.5 (4)  ≥17 1.8 (20) 2.6 (7) 2.0 (27) 1.4 (9) 1.4 (2) 1.4 (11) aP-value for difference between dispenser and poster groups based on Chi-Square analysis adjusted for age, education and ethnicity. Table 2. Alcohol consumption of all participants at baseline and follow-up Baseline Follow-up Dispenser Poster Total Dispenser Poster Total Measure % (n) % (n) % (n) % (n) % (n) % (n) Consumed alcohol in the past 30 days Pa = 0.05 Pa = 0.10  Yes 66.0 (1092) 60.8 (265) 64.9 (1357) 66.5 (630) 60.7 (139) 65.4 (769)  No 34.0 (563) 39.2 (171) 35.1 (734) 33.5 (317) 39.3 (90) 34.6 (407) Days per week Pa < 0.01 Pa = 0.25  1 or fewer 38.2 (416) 28.3 (75) 36.2 (491) 40.4 (254) 34.8 (48) 39.4 (302)  2 to 3 42.8 (467) 46.0 (122) 43.5 (589) 42.3 (266) 50.0 (69) 43.7 (335)  4 or more 19.0 (207) 25.7 (68) 20.3 (275) 17.3 (109) 15.2 (21) 16.9 (130) Average Number of drinks per occasion Pa = 0.19 Pa = 0.63  1 or fewer drinks 26.7 (281) 20.7 (53) 25.5 (334) 27.4 (165) 31.2 (43) 28.1 (208)  More than 1 to less than 4 drinks 54.4 (573) 56.3 (144) 54.7 (717) 54.9 (331) 50.0 (69) 54.0 (400)  4 to less than 5 drinks 7.4 (78) 8.6 (22) 7.6 (100) 9.5 (57) 11.6 (16) 9.9 (73)  5 or more drinks 11.6 (122) 14.5 (37) 12.1 (159) 8.3 (50) 7.2 (10) 8.1 (60) Number of occasions 4 or more drinks were consumed Pa = 0.16 Pa = 0.62  0 33.5 (366) 26.3 (70) 32.1 (436) 40.3 (253) 38.1 (53) 39.9 (306)  1 to 4 50.6 (553) 53.8 (143) 51.3 (696) 49.4 (310) 48.2 (67) 49.2 (377)  5 to 8 9.0 (98) 12.8 (34) 9.7 (132) 6.8 (43) 8.6 (12) 7.2 (55)  9 to 12 3.6 (39) 3.4 (9) 3.5 (48) 1.8 (11) 2.2 (3) 1.8 (14)  13 to 16 1.5 (16) 1.1 (3) 1.4 (19) 0.3 (2) 1.4 (2) 0.5 (4)  ≥17 1.8 (20) 2.6 (7) 2.0 (27) 1.4 (9) 1.4 (2) 1.4 (11) Baseline Follow-up Dispenser Poster Total Dispenser Poster Total Measure % (n) % (n) % (n) % (n) % (n) % (n) Consumed alcohol in the past 30 days Pa = 0.05 Pa = 0.10  Yes 66.0 (1092) 60.8 (265) 64.9 (1357) 66.5 (630) 60.7 (139) 65.4 (769)  No 34.0 (563) 39.2 (171) 35.1 (734) 33.5 (317) 39.3 (90) 34.6 (407) Days per week Pa < 0.01 Pa = 0.25  1 or fewer 38.2 (416) 28.3 (75) 36.2 (491) 40.4 (254) 34.8 (48) 39.4 (302)  2 to 3 42.8 (467) 46.0 (122) 43.5 (589) 42.3 (266) 50.0 (69) 43.7 (335)  4 or more 19.0 (207) 25.7 (68) 20.3 (275) 17.3 (109) 15.2 (21) 16.9 (130) Average Number of drinks per occasion Pa = 0.19 Pa = 0.63  1 or fewer drinks 26.7 (281) 20.7 (53) 25.5 (334) 27.4 (165) 31.2 (43) 28.1 (208)  More than 1 to less than 4 drinks 54.4 (573) 56.3 (144) 54.7 (717) 54.9 (331) 50.0 (69) 54.0 (400)  4 to less than 5 drinks 7.4 (78) 8.6 (22) 7.6 (100) 9.5 (57) 11.6 (16) 9.9 (73)  5 or more drinks 11.6 (122) 14.5 (37) 12.1 (159) 8.3 (50) 7.2 (10) 8.1 (60) Number of occasions 4 or more drinks were consumed Pa = 0.16 Pa = 0.62  0 33.5 (366) 26.3 (70) 32.1 (436) 40.3 (253) 38.1 (53) 39.9 (306)  1 to 4 50.6 (553) 53.8 (143) 51.3 (696) 49.4 (310) 48.2 (67) 49.2 (377)  5 to 8 9.0 (98) 12.8 (34) 9.7 (132) 6.8 (43) 8.6 (12) 7.2 (55)  9 to 12 3.6 (39) 3.4 (9) 3.5 (48) 1.8 (11) 2.2 (3) 1.8 (14)  13 to 16 1.5 (16) 1.1 (3) 1.4 (19) 0.3 (2) 1.4 (2) 0.5 (4)  ≥17 1.8 (20) 2.6 (7) 2.0 (27) 1.4 (9) 1.4 (2) 1.4 (11) aP-value for difference between dispenser and poster groups based on Chi-Square analysis adjusted for age, education and ethnicity. Among women who reported any alcohol consumption in the past 30 days at baseline, women who saw the FASD prevention message on the pregnancy test dispensers reported fewer drinking days per week then women who saw the message on a poster (P < 0.01). These differences remained statistically significant after adjusting for age, education and ethnicity (P < 0.01). Differences in drinking behavior between intervention groups were not statistically significant when restricted to participants who completed both baseline and follow-up. All other measures of drinking behavior were similar between intervention groups both at baseline and follow-up. Pregnancy tests dispensed Throughout the course of the study, we distributed over 12,000 pregnancy tests in 5 dispenser communities. In those communities 1688 participants completed a baseline survey, and 413 reported taking a pregnancy test from the dispenser. Of these 413 women, 265 reported using it and 43 (16.2%) had a positive test result. Yukon participants, who incurred a small charge for the pregnancy tests, were less likely to obtain a pregnancy test from the dispenser than Alaska participants who could obtain them for free (Alaska: 25.2%; Yukon: 10.0%; P = 0.06). Of the 43 women who had a positive pregnancy test result, 12 (27.9%) reported consuming alcohol during the past 30 days. Seventeen completed a follow-up survey with six again reporting being pregnant. Pregnancies A total of 192 participants reported being or becoming pregnant during the study. Of those participants, 94 reported knowing they were pregnant at baseline when they first viewed the FASD prevention message. An additional 43 women found out they were pregnant by using a pregnancy test they obtained from a dispenser. A total of 71 participants reported being pregnant at follow-up. Sixteen of these women had also reported being pregnant at baseline. Figure 2 illustrates the attrition and pregnancy status of participants at baseline and follow-up. Fig. 2. View largeDownload slide Attrition and pregnancy status of participants. Fig. 2. View largeDownload slide Attrition and pregnancy status of participants. Table 3 presents drinking habits for the 94 women who knew they were pregnant before participating in the study and the 71 participants who reported they were pregnant at follow-up. Almost 40% of pregnant participants both at baseline (36.2%) and follow-up (39.4%) reported no alcohol consumption between their last period and when they found out they were pregnant. Alcohol consumption among pregnant women was lower at follow-up than at baseline. Less than 20% reported drinking any alcohol once they knew they were pregnant, 18.1% at baseline and 12.7% at follow-up. Table 3. Alcohol consumption of pregnant participants and participants who reported they may have been pregnant during the 6 months prior to completing their survey Measure Pregnant Possibly pregnant % (n) % (n) % (n) % (n) Alcohol intake between last period to recognition of pregnancy  No Alcohol 36.2 (34) 39.4 (28) –a – – –  1 to 3 drinks at least once 30.9 (29) 46.5 (33) – – – –  1 to 3 drinks at a time, 1 or 2 times a week 11.7 (11) 8.5 (6) – – – –  1 to 3 drinks at a time, 3 or more times a week 6.4 (6) 2.8 (2) – – – –  4 or more drinks at one time, at least once 14.9 (14) 2.8 (2) – – – – Alcohol use since pregnancy recognized  No Alcohol 81.9 (77) 87.3 (62) – – – –  1 to 3 drinks at least once 0.0 (0) 12.7 (9) – – – –  1 to 3 drinks at a time, 1 or 2 times a week 5.3 (5) 0.0 (0) – – – –  1 to 3 drinks at a time, 3 or more times a week 6.4 (6) 0.0 (0) – – – –  4 or more drinks at one time, at least once 6.4 (6) 0.0 (0) – – – – Alcohol consumption during 6 months while pregnancy status was unknown  No Alcohol – – – – 32.5 (188) 35.0 (104)  1 to 3 drinks at least once – – – – 25.2 (146) 33.0 (98)  1 to 3 drinks at a time, 1 or 2 times a week – – – – 20.4 (118) 13.5 (40)  1 to 3 drinks at a time, 3 or more times a week – – – – 8.3 (48) 9.4 (28)  4 or more drinks at one time, at least once – – – – 13.6 (79) 9.1 (27) Consumed alcohol in the past 30 days  Yes 17.0 (16) 9.9 (7) 71.8 (416) 72.6 (215)  No 83.0 (78) 90.1 (64) 28.2 (163) 27.4 (81)  Days per week  1 or fewer 62.5 (10) 100 (7) 38.1 (158) 41.4 (89)  2 to 3 25.0 (4) 0.0 (0) 40.7 (169) 44.7 (96)  4 or more 12.5 (2) 0.0 (0) 21.2 (88) 14.0 (30) Average Number of drinks per occasion  1 or fewer drinks 28.6 (4) 85.7 (6) 27.9 (113) 28.5 (59)  More than 1 to less than 4 drinks 50.0 (7) 14.3 (1) 52.1 (211) 49.3 (102)  4 to less than 5 drinks 14.3 (2) 0.0 (0) 6.2 (25) 12.6 (26)  5 or more drinks 7.1 (1) 0.0 (0) 13.8 (56) 9.7 (20) Number of occasions 4 or more drinks were consumed  0 56.3 (9) 100 (7) 33.2 (138) 39.1 (84)  1 to 4 31.3 (5) 0.0 (0) 45.2 (188) 51.6 (111)  5 to 8 6.3 (1) 0.0 (0) 11.8 (49) 4.7 (10)  9 to 12 6.3 (1) 0.0 (0) 4.8 (20) 1.9 (4)  13 to 16 0.0 (0) 0.0 (0) 2.4 (10) 0.5 (1)  ≥17 0.0 (0) 0.0 (0) 2.6 (11) 2.3 (5) Measure Pregnant Possibly pregnant % (n) % (n) % (n) % (n) Alcohol intake between last period to recognition of pregnancy  No Alcohol 36.2 (34) 39.4 (28) –a – – –  1 to 3 drinks at least once 30.9 (29) 46.5 (33) – – – –  1 to 3 drinks at a time, 1 or 2 times a week 11.7 (11) 8.5 (6) – – – –  1 to 3 drinks at a time, 3 or more times a week 6.4 (6) 2.8 (2) – – – –  4 or more drinks at one time, at least once 14.9 (14) 2.8 (2) – – – – Alcohol use since pregnancy recognized  No Alcohol 81.9 (77) 87.3 (62) – – – –  1 to 3 drinks at least once 0.0 (0) 12.7 (9) – – – –  1 to 3 drinks at a time, 1 or 2 times a week 5.3 (5) 0.0 (0) – – – –  1 to 3 drinks at a time, 3 or more times a week 6.4 (6) 0.0 (0) – – – –  4 or more drinks at one time, at least once 6.4 (6) 0.0 (0) – – – – Alcohol consumption during 6 months while pregnancy status was unknown  No Alcohol – – – – 32.5 (188) 35.0 (104)  1 to 3 drinks at least once – – – – 25.2 (146) 33.0 (98)  1 to 3 drinks at a time, 1 or 2 times a week – – – – 20.4 (118) 13.5 (40)  1 to 3 drinks at a time, 3 or more times a week – – – – 8.3 (48) 9.4 (28)  4 or more drinks at one time, at least once – – – – 13.6 (79) 9.1 (27) Consumed alcohol in the past 30 days  Yes 17.0 (16) 9.9 (7) 71.8 (416) 72.6 (215)  No 83.0 (78) 90.1 (64) 28.2 (163) 27.4 (81)  Days per week  1 or fewer 62.5 (10) 100 (7) 38.1 (158) 41.4 (89)  2 to 3 25.0 (4) 0.0 (0) 40.7 (169) 44.7 (96)  4 or more 12.5 (2) 0.0 (0) 21.2 (88) 14.0 (30) Average Number of drinks per occasion  1 or fewer drinks 28.6 (4) 85.7 (6) 27.9 (113) 28.5 (59)  More than 1 to less than 4 drinks 50.0 (7) 14.3 (1) 52.1 (211) 49.3 (102)  4 to less than 5 drinks 14.3 (2) 0.0 (0) 6.2 (25) 12.6 (26)  5 or more drinks 7.1 (1) 0.0 (0) 13.8 (56) 9.7 (20) Number of occasions 4 or more drinks were consumed  0 56.3 (9) 100 (7) 33.2 (138) 39.1 (84)  1 to 4 31.3 (5) 0.0 (0) 45.2 (188) 51.6 (111)  5 to 8 6.3 (1) 0.0 (0) 11.8 (49) 4.7 (10)  9 to 12 6.3 (1) 0.0 (0) 4.8 (20) 1.9 (4)  13 to 16 0.0 (0) 0.0 (0) 2.4 (10) 0.5 (1)  ≥17 0.0 (0) 0.0 (0) 2.6 (11) 2.3 (5) aSubsample was not measured on the listed variable. Table 3. Alcohol consumption of pregnant participants and participants who reported they may have been pregnant during the 6 months prior to completing their survey Measure Pregnant Possibly pregnant % (n) % (n) % (n) % (n) Alcohol intake between last period to recognition of pregnancy  No Alcohol 36.2 (34) 39.4 (28) –a – – –  1 to 3 drinks at least once 30.9 (29) 46.5 (33) – – – –  1 to 3 drinks at a time, 1 or 2 times a week 11.7 (11) 8.5 (6) – – – –  1 to 3 drinks at a time, 3 or more times a week 6.4 (6) 2.8 (2) – – – –  4 or more drinks at one time, at least once 14.9 (14) 2.8 (2) – – – – Alcohol use since pregnancy recognized  No Alcohol 81.9 (77) 87.3 (62) – – – –  1 to 3 drinks at least once 0.0 (0) 12.7 (9) – – – –  1 to 3 drinks at a time, 1 or 2 times a week 5.3 (5) 0.0 (0) – – – –  1 to 3 drinks at a time, 3 or more times a week 6.4 (6) 0.0 (0) – – – –  4 or more drinks at one time, at least once 6.4 (6) 0.0 (0) – – – – Alcohol consumption during 6 months while pregnancy status was unknown  No Alcohol – – – – 32.5 (188) 35.0 (104)  1 to 3 drinks at least once – – – – 25.2 (146) 33.0 (98)  1 to 3 drinks at a time, 1 or 2 times a week – – – – 20.4 (118) 13.5 (40)  1 to 3 drinks at a time, 3 or more times a week – – – – 8.3 (48) 9.4 (28)  4 or more drinks at one time, at least once – – – – 13.6 (79) 9.1 (27) Consumed alcohol in the past 30 days  Yes 17.0 (16) 9.9 (7) 71.8 (416) 72.6 (215)  No 83.0 (78) 90.1 (64) 28.2 (163) 27.4 (81)  Days per week  1 or fewer 62.5 (10) 100 (7) 38.1 (158) 41.4 (89)  2 to 3 25.0 (4) 0.0 (0) 40.7 (169) 44.7 (96)  4 or more 12.5 (2) 0.0 (0) 21.2 (88) 14.0 (30) Average Number of drinks per occasion  1 or fewer drinks 28.6 (4) 85.7 (6) 27.9 (113) 28.5 (59)  More than 1 to less than 4 drinks 50.0 (7) 14.3 (1) 52.1 (211) 49.3 (102)  4 to less than 5 drinks 14.3 (2) 0.0 (0) 6.2 (25) 12.6 (26)  5 or more drinks 7.1 (1) 0.0 (0) 13.8 (56) 9.7 (20) Number of occasions 4 or more drinks were consumed  0 56.3 (9) 100 (7) 33.2 (138) 39.1 (84)  1 to 4 31.3 (5) 0.0 (0) 45.2 (188) 51.6 (111)  5 to 8 6.3 (1) 0.0 (0) 11.8 (49) 4.7 (10)  9 to 12 6.3 (1) 0.0 (0) 4.8 (20) 1.9 (4)  13 to 16 0.0 (0) 0.0 (0) 2.4 (10) 0.5 (1)  ≥17 0.0 (0) 0.0 (0) 2.6 (11) 2.3 (5) Measure Pregnant Possibly pregnant % (n) % (n) % (n) % (n) Alcohol intake between last period to recognition of pregnancy  No Alcohol 36.2 (34) 39.4 (28) –a – – –  1 to 3 drinks at least once 30.9 (29) 46.5 (33) – – – –  1 to 3 drinks at a time, 1 or 2 times a week 11.7 (11) 8.5 (6) – – – –  1 to 3 drinks at a time, 3 or more times a week 6.4 (6) 2.8 (2) – – – –  4 or more drinks at one time, at least once 14.9 (14) 2.8 (2) – – – – Alcohol use since pregnancy recognized  No Alcohol 81.9 (77) 87.3 (62) – – – –  1 to 3 drinks at least once 0.0 (0) 12.7 (9) – – – –  1 to 3 drinks at a time, 1 or 2 times a week 5.3 (5) 0.0 (0) – – – –  1 to 3 drinks at a time, 3 or more times a week 6.4 (6) 0.0 (0) – – – –  4 or more drinks at one time, at least once 6.4 (6) 0.0 (0) – – – – Alcohol consumption during 6 months while pregnancy status was unknown  No Alcohol – – – – 32.5 (188) 35.0 (104)  1 to 3 drinks at least once – – – – 25.2 (146) 33.0 (98)  1 to 3 drinks at a time, 1 or 2 times a week – – – – 20.4 (118) 13.5 (40)  1 to 3 drinks at a time, 3 or more times a week – – – – 8.3 (48) 9.4 (28)  4 or more drinks at one time, at least once – – – – 13.6 (79) 9.1 (27) Consumed alcohol in the past 30 days  Yes 17.0 (16) 9.9 (7) 71.8 (416) 72.6 (215)  No 83.0 (78) 90.1 (64) 28.2 (163) 27.4 (81)  Days per week  1 or fewer 62.5 (10) 100 (7) 38.1 (158) 41.4 (89)  2 to 3 25.0 (4) 0.0 (0) 40.7 (169) 44.7 (96)  4 or more 12.5 (2) 0.0 (0) 21.2 (88) 14.0 (30) Average Number of drinks per occasion  1 or fewer drinks 28.6 (4) 85.7 (6) 27.9 (113) 28.5 (59)  More than 1 to less than 4 drinks 50.0 (7) 14.3 (1) 52.1 (211) 49.3 (102)  4 to less than 5 drinks 14.3 (2) 0.0 (0) 6.2 (25) 12.6 (26)  5 or more drinks 7.1 (1) 0.0 (0) 13.8 (56) 9.7 (20) Number of occasions 4 or more drinks were consumed  0 56.3 (9) 100 (7) 33.2 (138) 39.1 (84)  1 to 4 31.3 (5) 0.0 (0) 45.2 (188) 51.6 (111)  5 to 8 6.3 (1) 0.0 (0) 11.8 (49) 4.7 (10)  9 to 12 6.3 (1) 0.0 (0) 4.8 (20) 1.9 (4)  13 to 16 0.0 (0) 0.0 (0) 2.4 (10) 0.5 (1)  ≥17 0.0 (0) 0.0 (0) 2.6 (11) 2.3 (5) aSubsample was not measured on the listed variable. Possible pregnancy At baseline 579 participants reported there was a time they may have been pregnant in the past 6 months, as did 297 participants at follow-up. Table 3 shows alcohol consumption among these participants at baseline and follow-up. At both time points, approximately 65% reported consuming alcohol while knowing that they could be pregnant. Nearly 72% of these participants reported consuming alcohol within the past 30 days and approximately 20% of participants reported consuming an average of four or more drinks on the days they consumed alcohol. More than 60% reported one or more occasions of binge drinking in the past 30 days. Knowledge and beliefs of FASD The dispenser group scored significantly higher on the questions about their knowledge and beliefs than the poster group at both baseline (Mean Dispenser score: 4.3; Mean Poster score: 4.1; P < 0.01) and follow-up (Dispenser: 4.4; Poster: 4.1; P < 0.01), indicating a more accurate understanding of FASD risks and harms. These differences remained significant even after adjusting for age, education and ethnicity (P < 0.01 for both baseline and follow-up). The mean of the health belief questions was higher for those who consumed alcohol in the past 30 days than those who reported not drinking at both baseline (Mean of those who drank: 4.4, Mean of those who did not drink: 4.1; P < 0.01) and follow-up (Mean of those who drank: 4.5, Mean of those who did not drink: 4.1; P < 0.01). The Cronbach’s alpha reliability coefficient for the FASD health belief questions was r = 0.65 at baseline and r = 0.67 at follow-up. Corrected item-to-total correlation for all items were ≥0.35 for both baseline and follow-up. The five items provided positive data indicating all five questions contributed to the overall reliability of the FASD health belief questions. Future assessment of health beliefs about FASD should expand the types of questions to achieve an Internal Consistency coefficient of r = 0.80. Table 4 shows the means, standard deviations and corrected item-to-total correlation for each FASD health belief questions. Table 4. FASD health belief question analysis outcomes at baseline and follow-up Baseline Follow-up Dispenser Poster Total Dispenser Poster Total Question M (SD) M (SD) M (SD) CI-TC M (SD) M (SD) M (SD) CI-TC As long as a woman does not drink too much alcohol, it is safe to have 1–2 alcoholic drinks at one time when pregnant.a 1.6 (1.2) 2.0 (1.3) 1.7 (1.2) 0.42 1.5 (1.0) 2.0 (1.2) 1.6 (1.0) 0.43 Evidence shows that drinking alcohol during pregnancy affects the health of the child after he or she is born. 4.5 (0.9) 4.4 (1.1) 4.5 (1.0) 0.49 4.6 (0.8) 4.2 (1.2) 4.5 (0.9) 0.57 Drinking alcohol during pregnancy will harm the mother’s health more than the unborn child’s health.a 2.1 (1.3) 2.2 (1.3) 2.1 (1.3) 0.35 1.9 (1.2) 1.9 (1.1) 1.9 (1.2) 0.37 Raising a child with a Fetal Alcohol Spectrum Disorder is hard. 4.5 (0.9) 4.3 (1.1) 4.4 (1.0) 0.43 4.6 (0.8) 4.3 (1.1) 4.6 (0.9) 0.42 If a woman drinks alcohol before she knows she is pregnant, her unborn child could still be affected. 4.2 (1.1) 4.0 (1.1) 4.2 (1.1) 0.39 4.3 (1.0) 4.1 (1.1) 4.2 (1.1) 0.37 Total FASD HB Score 4.3 (0.7) 4.1 (0.8) 4.3 (0.7) r = 0.65 4.4 (0.6)b 4.1 (0.8)c 4.4 (0.7)c r = 0.67 Baseline Follow-up Dispenser Poster Total Dispenser Poster Total Question M (SD) M (SD) M (SD) CI-TC M (SD) M (SD) M (SD) CI-TC As long as a woman does not drink too much alcohol, it is safe to have 1–2 alcoholic drinks at one time when pregnant.a 1.6 (1.2) 2.0 (1.3) 1.7 (1.2) 0.42 1.5 (1.0) 2.0 (1.2) 1.6 (1.0) 0.43 Evidence shows that drinking alcohol during pregnancy affects the health of the child after he or she is born. 4.5 (0.9) 4.4 (1.1) 4.5 (1.0) 0.49 4.6 (0.8) 4.2 (1.2) 4.5 (0.9) 0.57 Drinking alcohol during pregnancy will harm the mother’s health more than the unborn child’s health.a 2.1 (1.3) 2.2 (1.3) 2.1 (1.3) 0.35 1.9 (1.2) 1.9 (1.1) 1.9 (1.2) 0.37 Raising a child with a Fetal Alcohol Spectrum Disorder is hard. 4.5 (0.9) 4.3 (1.1) 4.4 (1.0) 0.43 4.6 (0.8) 4.3 (1.1) 4.6 (0.9) 0.42 If a woman drinks alcohol before she knows she is pregnant, her unborn child could still be affected. 4.2 (1.1) 4.0 (1.1) 4.2 (1.1) 0.39 4.3 (1.0) 4.1 (1.1) 4.2 (1.1) 0.37 Total FASD HB Score 4.3 (0.7) 4.1 (0.8) 4.3 (0.7) r = 0.65 4.4 (0.6)b 4.1 (0.8)c 4.4 (0.7)c r = 0.67 M: mean. SD: standard deviation. CI-TC: corrected item-total correlation. r: Cronbach's alpha. aItem reverse coded when summed into Total FASD HB score. bP < 0.05 for difference between baseline and follow-up cP < 0.01 for difference between baseline and follow-up Table 4. FASD health belief question analysis outcomes at baseline and follow-up Baseline Follow-up Dispenser Poster Total Dispenser Poster Total Question M (SD) M (SD) M (SD) CI-TC M (SD) M (SD) M (SD) CI-TC As long as a woman does not drink too much alcohol, it is safe to have 1–2 alcoholic drinks at one time when pregnant.a 1.6 (1.2) 2.0 (1.3) 1.7 (1.2) 0.42 1.5 (1.0) 2.0 (1.2) 1.6 (1.0) 0.43 Evidence shows that drinking alcohol during pregnancy affects the health of the child after he or she is born. 4.5 (0.9) 4.4 (1.1) 4.5 (1.0) 0.49 4.6 (0.8) 4.2 (1.2) 4.5 (0.9) 0.57 Drinking alcohol during pregnancy will harm the mother’s health more than the unborn child’s health.a 2.1 (1.3) 2.2 (1.3) 2.1 (1.3) 0.35 1.9 (1.2) 1.9 (1.1) 1.9 (1.2) 0.37 Raising a child with a Fetal Alcohol Spectrum Disorder is hard. 4.5 (0.9) 4.3 (1.1) 4.4 (1.0) 0.43 4.6 (0.8) 4.3 (1.1) 4.6 (0.9) 0.42 If a woman drinks alcohol before she knows she is pregnant, her unborn child could still be affected. 4.2 (1.1) 4.0 (1.1) 4.2 (1.1) 0.39 4.3 (1.0) 4.1 (1.1) 4.2 (1.1) 0.37 Total FASD HB Score 4.3 (0.7) 4.1 (0.8) 4.3 (0.7) r = 0.65 4.4 (0.6)b 4.1 (0.8)c 4.4 (0.7)c r = 0.67 Baseline Follow-up Dispenser Poster Total Dispenser Poster Total Question M (SD) M (SD) M (SD) CI-TC M (SD) M (SD) M (SD) CI-TC As long as a woman does not drink too much alcohol, it is safe to have 1–2 alcoholic drinks at one time when pregnant.a 1.6 (1.2) 2.0 (1.3) 1.7 (1.2) 0.42 1.5 (1.0) 2.0 (1.2) 1.6 (1.0) 0.43 Evidence shows that drinking alcohol during pregnancy affects the health of the child after he or she is born. 4.5 (0.9) 4.4 (1.1) 4.5 (1.0) 0.49 4.6 (0.8) 4.2 (1.2) 4.5 (0.9) 0.57 Drinking alcohol during pregnancy will harm the mother’s health more than the unborn child’s health.a 2.1 (1.3) 2.2 (1.3) 2.1 (1.3) 0.35 1.9 (1.2) 1.9 (1.1) 1.9 (1.2) 0.37 Raising a child with a Fetal Alcohol Spectrum Disorder is hard. 4.5 (0.9) 4.3 (1.1) 4.4 (1.0) 0.43 4.6 (0.8) 4.3 (1.1) 4.6 (0.9) 0.42 If a woman drinks alcohol before she knows she is pregnant, her unborn child could still be affected. 4.2 (1.1) 4.0 (1.1) 4.2 (1.1) 0.39 4.3 (1.0) 4.1 (1.1) 4.2 (1.1) 0.37 Total FASD HB Score 4.3 (0.7) 4.1 (0.8) 4.3 (0.7) r = 0.65 4.4 (0.6)b 4.1 (0.8)c 4.4 (0.7)c r = 0.67 M: mean. SD: standard deviation. CI-TC: corrected item-total correlation. r: Cronbach's alpha. aItem reverse coded when summed into Total FASD HB score. bP < 0.05 for difference between baseline and follow-up cP < 0.01 for difference between baseline and follow-up DISCUSSION This is the first rigorous feasibility study of an FASD prevention strategy employing pregnancy tests in establishments serving alcohol. Interventions were either an FASD informational poster affixed to a pregnancy test dispenser or an FASD informational poster alone. The goal of this study was to assess the feasibility of deploying FASD prevention messages on either a poster or a pregnancy test dispenser in women’s restrooms and collect pilot data to measure the utility of both interventions at promoting informed alcohol consumption behavior. Participating establishments reported generally positive comments from patrons. No establishments requested that the messages be removed prior to the end of the study. Local staff reported anecdotal evidence that patrons at pregnancy test dispenser sites were employing those tests on-site prior to consuming alcohol. Both interventions were effective at engaging women with FASD prevention messages and increasing knowledge of risks associated with alcohol consumption during pregnancy. Participants who viewed the poster on a pregnancy test dispenser had better knowledge and beliefs than their poster only counterparts at both baseline (average score of 4.3 and 4.1, respectively) and follow-up (average score of 4.4 and 4.1, respectively), even after adjusting for education and other demographics. This finding is consistent with the dispenser having a greater immediate impact than the poster alone. A smaller proportion of pregnant participants reported drinking alcohol within the past 30 days at follow-up (10%) than at baseline (17%) which may indicate the intervention influenced pregnant participants to stop drinking. Limitations We did not collect data before the interventions were implemented to determine the participants’ knowledge and belief regarding FASD and the risks of consuming alcohol while pregnant or possibly pregnant. Resource constraints precluded the creation of a control group of multiple sites of women frequenting bars with neither intervention. Thus, it was not possible to determine how much of the difference between the two groups was due to the interventions and how much was due to differences in the two study groups. However, the similarity in the self-reported drinking behavior between the two groups for the 30 days before baseline strengthens our conclusion that the difference in the FASD prevention health belief scores was due in part to the differential effect of the two interventions and not just to pre-existing differences in the two study populations. Because the baseline survey was accessible to anyone who had access to the survey link, it was possible for individuals to respond to the survey more than once or to share the survey link with individuals who had not actually viewed the interventions. We employed a protocol to identify duplicate and suspect responses for the baseline survey which allowed us to eliminate many, but not necessarily all, of these responses. We were also able to better control access to the follow-up surveys by sending an e-mail with the survey link 6 months after the baseline survey was completed. The cost of obtaining a pregnancy test may have affected the number of participants who obtained a test from the Yukon dispensers. However, the sample size for the Yukon dispenser site is very small (n = 36) precluding the ability to accurately determine the effect of cost on the use of the test dispenser. Next steps Future impact evaluations need to increase the number of sites, sample size and participant diversity to increase statistical power and to have a sufficient number of sites in areas with known demographic characteristics. Demographic matching at baseline of a sufficient number of bars with large, diverse samples of women should enable future evaluations to establish ‘Control and Experimental Sites’ to improve the methodological quality and generalizability of assessments of intervention effectiveness (Murray, 1998; Windsor, 2015). We also recommend that future studies consider methological variations such as assessing effects of charging for pregnancy tests and the differential effects of different costs per test, e.g. $1 – $2 – $3, and that the basic variations in intervention types be placed in culturally diverse locations to measure differences in effectiveness between subsamples. The FASD health belief questions provided useful data and insight about perceptions of risk of women of childbearing age. Future studies, however, should consider implementing a process to expand the type and number of health belief questions to enable additional validity and reliability analyses to refine message content and graphics. Public health implications Our team distributed over 12,000 pregnancy tests in eight communities. Interviews with bar managers and staff suggest that tests may have been inappropriately dispensed early in the intervention when they were still perceived as a novelty. However, the overall demand indicates a substantial service gap for accessible and affordable pregnancy tests for women at risk for an inadvertent alcohol-exposed pregnancy. Overcoming barriers to accessible pregnancy tests is especially important in scenarios where the cost of pregnancy tests are prohibitive or personal anonymity is jeopardized. CONCLUSIONS Given the relatively high unintended pregnancy rate, empowering sexually active women to know their pregnancy status before consuming alcohol is a sound public health strategy. Because accessible contraception also remains essential to FASD prevention, our study distributed condoms at all intervention sites. However, while considerable focus in public health has been placed on the importance of contraception, the valuable role of affordable and accessible pregnancy tests appears to be somewhat precluded and should receive more attention. ACKNOWLEDGEMENTS This study was funded by the Alaska State Legislature: 2015–17. We particularly wish to thank the members of our Expert Advisory Group, and Message Testing Group, for their support. We thank Mr. Jody Allen Crowe for support with the placement and maintenance of pregnancy test dispensers in establishments that serve alcohol. The surveys used open-source materials from the University of California, San Diego. They were developed in collaboration with Dr. Christina Chambers of the University of California. We would also like to thank Julie Holden, Sarah Shimer, Ginger Cooley and Rebecca Van Wyck who assisted with the study. FUNDING This project was supported by the Alaska State Legislature. CONFLICT OF INTEREST STATEMENT None declared. REFERENCES Alcohol Use in Pregnancy . Fetal Alcohol Spectr Disord 2016 . Camerlin A Alaska Maternal and Child Health Data Book 2012: Birth Defects Surveillance Edition. Anchorage, AK, 2012 . Capecchi C Would You Take a Pregnancy Test in a Bar? New York Times 2012 . Diagnosis . Fetal Alcohol Spectr Disord 2014 . Ethen M , Ramadhani T , Scheuerle A , et al. . ( 2009 ) Alcohol consumption by women before and during pregnancy . Matern Child Health J 13 : 274 – 85 . Google Scholar CrossRef Search ADS PubMed Gordon R . ( 1983 ) An operational classification of disease prevention . Public Health Rep 98 : 107 – 9 . Google Scholar PubMed Kost K Unintended Pregnancy Rates at the State Level: Estimates for 2010 and Trends Since 2002. New York, 2015 . Maier S , West J . ( 2001 ) Drinking patterns and alcohol-related birth defects . Alcohol Res Heal 25 : 168 – 74 . May PA , Gossage JP , Brooke LE , et al. . ( 2005 ) Maternal risk factors for fetal alcohol syndrome in the Western Cape Province of South Africa: a population-based study . Am J Public Heal J Public Heal 9595 : 1190 – 9 . Google Scholar CrossRef Search ADS Miller L , Tolliver R , Druschel C , et al. . ( 2002 ) Fetal alcohol syndrome: Alaska, Arizona, Colorado, and New York, 1995–1997 . Morb Mortal Wkly Rep 51 : 433 – 5 . Murray DM . ( 1998 ) Design and Analysis of Group-Randomized Trials . New York, NY : Oxford University Press . Notice to readers . ( 2005 ) Surgeon general’s advisory on alcohol use in pregnancy . Morb Mortal Wkly Rep 54 : 229 . PRAMStat: Explore PRAMS Data by State . Centers Dis Control Prev 2010 . State-Specific Alcohol Consumption Rates for 2013 . Fetal Alcohol Spectr Disord 2015 . Stratton K , Howe C , Battaglia F (eds) . ( 1996 ) Fetal Alcohol Syndrome: Diagnosis, Epidemiology, Prevention, and Treatment . Washington, DC : National Academy Press . Tan CH , Denny CH , Cheal NE , et al. . ( 2015 ) Alcohol use and binge drinking among women of childbearing age—United States, 2011–2013 . MMWR Morb Mortal Wkly Rep 64 : 1042 – 6 . Google Scholar CrossRef Search ADS PubMed Williams J , Smith V . ( 2015 ) Fetal alcohol spectrum disorder . Pediatrics 136 : 1395 – 406 . Google Scholar CrossRef Search ADS Windsor R . ( 2015 ) Evaluation of Health Promotion, and Disease Prevention and Management Programs: Improving Population Health through Evidence-Based Practices , 5th ed . Silver Spring, MD : Oxford University Press . Google Scholar CrossRef Search ADS © The Author(s) 2018. Medical Council on Alcohol and Oxford University Press. All rights reserved. This article is published and distributed under the terms of the Oxford University Press, Standard Journals Publication Model (https://academic.oup.com/journals/pages/about_us/legal/notices) http://www.deepdyve.com/assets/images/DeepDyve-Logo-lg.png Alcohol and Alcoholism Oxford University Press

A Formative Evaluation of Two FASD Prevention Communication Strategies

Loading next page...
 
/lp/ou_press/a-formative-evaluation-of-two-fasd-prevention-communication-strategies-SsIeU1JLNh
Publisher
Oxford University Press
Copyright
© The Author(s) 2018. Medical Council on Alcohol and Oxford University Press. All rights reserved.
ISSN
0735-0414
eISSN
1464-3502
D.O.I.
10.1093/alcalc/agx122
Publisher site
See Article on Publisher Site

Abstract

Abstract Aims To evaluate the feasibility, acceptability and effectiveness of placing FASD prevention messages in the women’s restrooms of establishments serving alcohol in Alaska and the Yukon, regions with high rates of FASD. Methods Our team placed an FASD educational poster, and posters affixed to a pregnancy test dispenser, in women’s restrooms of bars and restaurants. We compared drinking behaviors and knowledge and beliefs about FASD among participants at baseline and at follow-up. Results Respondents consisted of 2132 women who completed a baseline survey and 1182 women who completed both a baseline and a follow-up survey. Women in both groups showed improvement in knowledge of FASD; the dispenser group scored higher than participants in the poster group on the FASD Health Belief questions at both baseline and follow-up. Forty-three women learned they were pregnant from our pregnancy tests and alcohol consumption among pregnant women was lower at follow-up than at baseline. Conclusions FASD prevention messages, particularly paired with pregnancy test dispensers, in the women’s restrooms of establishments that serve alcohol can effectively promote informed alcohol consumption decisions among women who are, or may become, pregnant. Short Summary In this FASD prevention feasibility study, we found that FASD prevention messages, particularly paired with pregnancy test dispensers, placed in the women’s restrooms of establishments that serve alcohol can effectively promote informed alcohol consumption decisions among women who are, or may become, pregnant. INTRODUCTION Alcohol use during pregnancy is the leading cause of preventable birth defects and developmental disabilities in the United States (Tan et al., 2015). Fetal alcohol syndrome (FAS) has been associated with a variety of risk factors, including binge drinking; advanced maternal age; high gravidity and parity; unstable marital status; and use of other substances including tobacco (May et al., 2005). The Institute of Medicine (Stratton et al., 1996) defines fetal alcohol spectrum disorder (FASD) as an umbrella term encompassing four discrete diagnoses that comprise a clinical spectrum from severe to mild (fetal alcohol syndrome, partial fetal alcohol syndrome, alcohol-related neurodevelopmental disorder and alcohol-related birth defects). Severity of FASD has been associated with the amount of alcohol consumed during gestation, modified by certain maternal characteristics (Maier and West, 2001; Diagnosis, 2014). The U.S. Surgeon General (Notice to readers: Surgeon General’s advisory on alcohol use in pregnancy, 2005), Centers for Disease Control and Prevention (Alcohol Use in Pregnancy, 2016), and the American Academy of Pediatrics (Williams and Smith, 2015) recommend that pregnant women and women who are considering pregnancy, abstain from alcohol consumption to eliminate the risk of FASD. Historically Alaska has had the highest rate of FASD among states that track these outcomes (Miller et al., 2002). However, more recent data suggest that Alaska, with 1.5 FAS cases per 1000 live births, falls within the national average 1–3 cases of FAS per 1000 births (Camerlin, 2012). A conservative estimate for the rate of all FASD cases in Alaska is still extremely high with 81.8 per 10,000 live births based on cases reported to the Alaska Birth Defects Registry. A substantial number of Alaskan women may have an unintended alcohol-exposed pregnancy due to lack of awareness of either their pregnancy status or the risks of alcohol consumption during pregnancy. Approximately half of all pregnancies nationally and in Alaska are unintended (Kost, 2015). Over half of all women in Alaska report consuming alcohol (State-Specific Alcohol Consumption Rates for 2013, 2015). In a national study, 22% of women reported consuming alcohol during their first month of pregnancy, 8% during the second and 5% during the third (Ethen et al., 2009). In 2010, 7% of Alaskan mothers reported that they consumed alcohol throughout their pregnancy (PRAMStat: Explore PRAMS Data by State, 2010). Response Many FASD prevention interventions take the form of health education messages using public service announcements and other forms of mass media. These interventions, which fall into the category of universal prevention strategies, are designed to reach the entire population without regard to individual risk factors (Gordon, 1983). There are few examples of FASD prevention campaigns that employ either a selective strategy focusing on subgroups at higher risk or an indicated strategy targeting individuals with a specific risk factor that puts them at higher risk for FASD. The most commonly known prevention programs are CHOICES, which focus on pregnancy prevention through contraception, and Parent-Child Assistance Program which implements prevention strategies after a woman finds out she is pregnant. Although these programs are beneficial, additional prevention efforts are needed for the highly vulnerable population of women who are consuming alcohol and are unaware of their pregnancy status. In one such strategy implemented in Mankato, Minnesota, a pregnancy test dispenser featuring information about the risks of FASD was placed in the women’s restroom of an establishment serving alcohol (Capecchi, 2012). For a minimal charge ($3.00 on a debit card), a female patron could purchase a pregnancy test and determine her pregnancy status prior to consuming alcohol. The goal was to provide both information about FASD and the means to prevent it. This prevention strategy has not been evaluated. Objectives This paper presents the results of a formative evaluation designed to assess the utility of two innovative FASD prevention strategies. The specific objectives for this study were as follows: To develop two FASD prevention strategies differentiated by the media on which the message is displayed. To assess the feasibility of implementing the FASD prevention strategies in women’s restrooms of establishments that serve alcohol, and; To assess the acceptability and compare the effectiveness of these strategies for promoting informed alcohol consumption decisions among women who are pregnant, or may become pregnant. We evaluate the relationship between the two FASD prevention strategies and women’s drinking behavior and knowledge and beliefs related to alcohol consumption at baseline and 6 months following exposure to each intervention. MATERIALS AND METHODS This formative evaluation study used a matched-site comparison design. We posted the same prevention message in women’s restrooms at establishments serving alcohol in seven communities in Alaska, and one in the Yukon Territory. Intervention communities were matched by racial and ethnic composition, number of establishments serving alcohol, and connections to rural and remote communities (the latter being not connected to any other community by road). Bar recruitment Bars in each study community were recruited by community liaisons with local knowledge of the service industry and regional context. The liaison initiated contacts with owners and managers of establishments serving alcohol in each study community. The liaison then coordinated placement of the health communication message using either a pregnancy test dispenser or a poster as the medium of communication (hereafter referred to as the interventions) and managed the intervention as necessary during data collection, which lasted approximately 1 year. Liaisons met regularly with owners, managers and staff to discuss issues or challenges as they arose throughout the year and at the conclusion of data collection they removed the interventions. Message development An expert advisory panel (EAP) of FASD prevention experts including obstetrics practitioners, certified nurse midwives and family planning counselors from study communities developed the draft educational message. The panel recommended that the message be clear, creative and personal to increase engagement. The EAP members worked over the course of 3 months to develop a draft message that met these criteria, while avoiding the appearance of stigma for mothers of children with FASD. The draft message developed by the EAP was tested and refined with input from a review panel of women aged 21–44 years of age who consume alcohol. Each intervention message included a toll-free number to a regional family planning counseling service. It also included a discrete toll-free number, web address and quick response (QR) code that allowed users to link to our electronic survey. All interventions were paired with a condom dispenser to provide a readily-accessible form of birth control. The final message consisted of three elements: any alcohol use during pregnancy can cause lifelong health problems for the child; the prospective adverse health outcomes associated with FASD are severe and difficult to address; and a woman should be aware of her potential pregnancy status prior to consuming alcohol (Fig. 1). Fig. 1. View largeDownload slide FASD Informational Poster. Fig. 1. View largeDownload slide FASD Informational Poster. Six sites featured the message in a framed poster, and eight featured the message affixed to a pregnancy test dispenser. The dispensers were programmed to delay 60-s between the dispensing of individual pregnancy tests to reduce the potential for one user to withdraw all the pregnancy tests in the unit at one time. A message on the pregnancy test kit sleeve warned recipients that a pregnancy test conducted within 1 week after a missed period could be misleading, and recommended further testing in such cases. Pregnancy tests dispensed in the Yukon Territory required a small cost of two Canadian dollars. Survey description Surveys were completed by participants via text messaging or web access to assess the acceptability of the two types of communication interventions. We asked participants in the dispenser group if they obtained a pregnancy test from the dispenser. Participants who reported obtaining a test were also asked whether they had used it, and if so, the results of that test. All participants who reported not using a pregnancy test from a study dispenser were asked if they knew their pregnancy status. Participants who reported being pregnant were asked to describe their alcohol consumption prior to and after learning their status. Participants who responded that they were either not pregnant or were unsure of their status were asked if there had been a time during the past 6 months when they might have been pregnant. These participants were asked to respond to similar questions about alcohol consumption as those who were pregnant. We assessed alcohol consumption by asking participants if they had consumed alcohol in the past 30 days, and if so, the number of drinking days per week and the amount of alcohol consumed on days when they drank. Finally, a series of eight questions were asked concerning knowledge and beliefs about FASD risk using a 5 point Likert-scale ranging from 1 Strongly Disagree to 5 Strongly Agree. Two questions were reverse coded for consistency in data analysis. Questions with the lowest corrected item-to-total correlation were removed individually until the inter-item reliability was maximized. The corrected item-to-total correlation for each question is recommended to be r> 0.30 (Windsor, 2015). Based on standard psychometric analyses we restricted our analysis to the following five questions: As long as a woman does not drink too much alcohol, it is safe to have 1–2 alcoholic drinks at one time when pregnant. (Reverse coded) Evidence shows that drinking alcohol during pregnancy affects the health of the child after he or she is born. Drinking alcohol during pregnancy will harm the mother’s health more than the unborn child’s health. (Reverse coded) Raising a child with a Fetal Alcohol Spectrum Disorder is hard. If a woman drinks alcohol before she knows she is pregnant, her unborn child could still be affected. Using responses to these five questions a cumulative score was computed to evaluate participant’s beliefs about FASD risk. A higher score indicates a more accurate understanding of FASD risks and harms. All participants ended the survey with general demographic information concerning ethnicity, age and education. Participants were asked to select all appropriate races or ethnicities which included White, Black/African American/Negro, American Indian or Alaskan Native, Asian, Native Hawaiian or Pacific Islander, Hispanic/Latino/Other Spanish and Other. Participants who selected other were able to enter in their own race or ethnicity as text. For analysis we recoded ethnicity combining participants who selected Other with participants who selected multiple races. We sent all participants an invitation to participate in a follow-up survey 6 months after completing the baseline survey. The follow-up survey differed from the baseline only with the addition of a question regarding the use of community health resources and the removal of demographic and pregnancy tests questions. Participants received a $25 gift certificate to an online retailer for completing each of the two surveys. Impact analyses The effectiveness of the two messaging strategies was evaluated by comparing differences in alcohol consumption and knowledge and beliefs about FASD between the dispenser and poster intervention groups at baseline and follow-up. We also assessed changes in alcohol consumption from baseline to follow-up for participants who used a pregnancy test from one of our test dispensers and reported a positive test result, or who reported being pregnant or possibly pregnant at baseline. To minimize duplicate survey responses, we developed a protocol for determining level of suspicion that a survey was a duplicate based on duplicate e-mail addresses or similar e-mail addresses from the same IP address. Surveys might also be deemed suspicious if they used a link that had been replaced or removed during the study period. Highly suspicious surveys were deleted; less suspicious surveys were verified by e-mail and only included if the participant confirmed having submitted only one survey. A total of 207 duplicates were excluded from the analyses. The statistical significance of group comparisons was assessed using Chi-square or non-parametric Mann Whitney tests as appropriate, with a significance level of P < 0.05. We used ordinal regression to adjust for age, education and ethnicity for differences that were statistically significant. Goodness of fit for regression models was checked using Chi-Square with alpha set at P > 0.05. The study was approved by the University of Alaska Anchorage’s Institutional Review Board. RESULTS Intervention feasibility Site selection, recruitment and intervention placement took place over the first 6 months of 2015. Community liaisons initiated contacts with establishments in eight study communities. The communities assigned to the dispenser intervention were: the Anchorage, Alaska metropolitan area, the rural towns of Homer, Alaska and Whitehorse, Yukon, and the remote town of Kodiak, Alaska. Communities assigned to the poster intervention were: the Juneau, Alaska and Fairbanks, Alaska metropolitan areas, and the remote towns of Dillingham and Nome. Fairbanks, which is on the road system, and Juneau, which is off the road system, are smaller metropolitan areas than Anchorage but larger than Homer and Whitehorse. Within the four dispenser communities, eight bars agreed to place FASD prevention messages on pregnancy test dispensers in women’s restrooms. Liaisons reported that business owners or managers who refused participation did so based on concerns regarding the sexual nature of the message. Among those who allowed us to install pregnancy test dispensers, establishment owners, managers and staff described an initial phase of the intervention in which condoms and pregnancy tests were removed from the dispensers but left at the table, employed for jokes or pranks, or otherwise wasted by patrons. Some owners placed signage alongside the dispensers to formally express support for the study and request that the tests and condoms not be wasted. Staff at other establishments employed informal social sanctions such as disapproving comments or lack of a humorous response to the patron’s behavior. Although the waste gradually ceased, staff suggested that if the condoms and pregnancy tests had some minimal cost, perhaps $1.00 each, this behavior might be avoided. Owners, managers and staff in all study communities reported that the overwhelming number of responses to the interventions were positive. These comments most often took the form of support for the goal of preventing FASD, and gratitude for the condoms and pregnancy tests. Wait staff reported that younger female patrons were most likely to be interested and engaged by the message. The rare negative comments were expressed by patrons concerned with the sexual component of the message. Owners expressed the opinion that interventions were associated with increased traffic and sales. Two business owners took steps to continue the distribution of no-cost condoms at the conclusion of the study. Three wait staff passed on personal narratives from women who stopped drinking alcohol after receiving a positive pregnancy test result at the establishment. Participants The baseline survey was completed by 2132 women of childbearing age. We sent 1975 (92.6%) follow-up invitation emails; 75 of baseline participants did not provide an e-mail address and 82 emails were returned as undeliverable. Of those remaining women, 1182 (60%) completed a follow-up survey. As shown in Table 1, women completing the baseline survey who viewed the FASD prevention messages on the dispenser were older (P = 0.01), less educated and more likely to be a minority than the poster group (P < 0.01 for both). Similar differences were observed among women completing the follow-up survey (P < 0.01 for each). Table 1. Demographics of baseline participants, follow-up participants and follow-up nonparticipants Baseline Follow-up Follow-up nonparticipants Dispenser Poster Total Dispenser Poster Total Dispenser Poster Total Measure % (n) % (n) % (n) % (n) % (n) % (n) % (n) % (n) % (n) Age  21 to 25 34.4 (570) 43.0 (188) 36.2 (758) 30.0 (286) 45.0 (103) 32.9 (389) 40.5 (284) 40.9 (85) 40.5 (369)  26 to 30 33.2 (550) 30.9 (135) 32.7 (685) 37.5 (357) 31.0 (71) 36.2 (428) 27.5 (193) 30.8 (64) 28.2 (257)  31 to 35 20.0 (331) 16.9 (74) 19.4 (405) 21.3 (203) 16.6 (38) 20.4 (241) 18.2 (128) 17.3 (36) 18.0 (164)  36 to 40 8.2 (135) 5.7 (25) 7.6 (160) 7.2 (69) 5.7 (13) 6.9 (82) 9.4 (66) 5.8 (12) 8.6 (78)  41 to 44 4.2 (69) 3.4 (15) 4.0 (84) 4.0 (38) 1.7 (4) 3.6 (42) 4.4 (31) 5.3 (11) 4.6 (42) Education  Some High School 3.9 (64) 3.9 (17) 3.9 (81) 2.7 (26) 2.6 (6) 2.7 (32) 5.4 (38) 5.4 (11) 5.4 (49)  Diploma or GED 18.6 (308) 10.7 (46) 17.0 (354) 16.3 (155) 8.3 (19) 14.7 (174) 21.8 (153) 13.4 (27) 19.9 (180)  Some College 42.1 (696) 34.8 (150) 40.6 (846) 42.9 (409) 28.4 (65) 40.1 (474) 40.9 (287) 42.1 (85) 41.2 (372)  Bachelors 25.1 (415) 32.5 (140) 26.6 (555) 27.0 (257) 38.4 (88) 29.2 (345) 22.5 (158) 25.7 (52) 23.3 (210)  Graduate 10.3 (171) 18.1 (78) 11.9 (249) 11.1 (106) 22.3 (51) 13.3 (157) 9.3 (65) 13.4 (27) 10.2 (92) Race or Ethnicity  White 58.5 (966) 70.5 (304) 61.0 (1270) 61.1 (581) 74.1 (169) 63.6 (750) 55.1 (385) 66.5 (135) 57.6 (520)  Blacka 5.2 (86) 6.3 (27) 5.4 (113) 4.9 (47) 5.3 (12) 5.0 (59) 5.6 (39) 7.4 (15) 6.0 (54)  AIANb 6.8 (113) 5.6 (24) 6.6 (137) 5.8 (55) 5.7 (13) 5.8 (68) 8.3 (58) 5.4 (11) 7.6 (69)  Asian 8.7 (143) 6.3 (27) 8.2 (170) 8.8 (84) 5.3 (12) 8.1 (96) 8.4 (59) 7.4 (15) 8.2 (74)  NHOPIc 2.8 (46) 1.4 (6) 2.5 (52) 2.0 (19) 2.6 (6) 2.1 (25) 3.9 (27) 0.0 (0) 3.0 (27)  Hispanicd 5.5 (91) 3.7 (16) 5.1 (107) 5.8 (55) 2.6 (6) 5.2 (61) 5.2 (36) 4.9 (10) 5.1 (46)  Other or Multiple 12.4 (205) 6.3 (27) 11.1 (232) 11.6 (110) 4.4 (10) 10.2 (120) 13.6 (95) 8.4 (17) 12.4 (112) Baseline Follow-up Follow-up nonparticipants Dispenser Poster Total Dispenser Poster Total Dispenser Poster Total Measure % (n) % (n) % (n) % (n) % (n) % (n) % (n) % (n) % (n) Age  21 to 25 34.4 (570) 43.0 (188) 36.2 (758) 30.0 (286) 45.0 (103) 32.9 (389) 40.5 (284) 40.9 (85) 40.5 (369)  26 to 30 33.2 (550) 30.9 (135) 32.7 (685) 37.5 (357) 31.0 (71) 36.2 (428) 27.5 (193) 30.8 (64) 28.2 (257)  31 to 35 20.0 (331) 16.9 (74) 19.4 (405) 21.3 (203) 16.6 (38) 20.4 (241) 18.2 (128) 17.3 (36) 18.0 (164)  36 to 40 8.2 (135) 5.7 (25) 7.6 (160) 7.2 (69) 5.7 (13) 6.9 (82) 9.4 (66) 5.8 (12) 8.6 (78)  41 to 44 4.2 (69) 3.4 (15) 4.0 (84) 4.0 (38) 1.7 (4) 3.6 (42) 4.4 (31) 5.3 (11) 4.6 (42) Education  Some High School 3.9 (64) 3.9 (17) 3.9 (81) 2.7 (26) 2.6 (6) 2.7 (32) 5.4 (38) 5.4 (11) 5.4 (49)  Diploma or GED 18.6 (308) 10.7 (46) 17.0 (354) 16.3 (155) 8.3 (19) 14.7 (174) 21.8 (153) 13.4 (27) 19.9 (180)  Some College 42.1 (696) 34.8 (150) 40.6 (846) 42.9 (409) 28.4 (65) 40.1 (474) 40.9 (287) 42.1 (85) 41.2 (372)  Bachelors 25.1 (415) 32.5 (140) 26.6 (555) 27.0 (257) 38.4 (88) 29.2 (345) 22.5 (158) 25.7 (52) 23.3 (210)  Graduate 10.3 (171) 18.1 (78) 11.9 (249) 11.1 (106) 22.3 (51) 13.3 (157) 9.3 (65) 13.4 (27) 10.2 (92) Race or Ethnicity  White 58.5 (966) 70.5 (304) 61.0 (1270) 61.1 (581) 74.1 (169) 63.6 (750) 55.1 (385) 66.5 (135) 57.6 (520)  Blacka 5.2 (86) 6.3 (27) 5.4 (113) 4.9 (47) 5.3 (12) 5.0 (59) 5.6 (39) 7.4 (15) 6.0 (54)  AIANb 6.8 (113) 5.6 (24) 6.6 (137) 5.8 (55) 5.7 (13) 5.8 (68) 8.3 (58) 5.4 (11) 7.6 (69)  Asian 8.7 (143) 6.3 (27) 8.2 (170) 8.8 (84) 5.3 (12) 8.1 (96) 8.4 (59) 7.4 (15) 8.2 (74)  NHOPIc 2.8 (46) 1.4 (6) 2.5 (52) 2.0 (19) 2.6 (6) 2.1 (25) 3.9 (27) 0.0 (0) 3.0 (27)  Hispanicd 5.5 (91) 3.7 (16) 5.1 (107) 5.8 (55) 2.6 (6) 5.2 (61) 5.2 (36) 4.9 (10) 5.1 (46)  Other or Multiple 12.4 (205) 6.3 (27) 11.1 (232) 11.6 (110) 4.4 (10) 10.2 (120) 13.6 (95) 8.4 (17) 12.4 (112) aBlack/African American/Negro. bAmerican Indian or Alaskan Native. cNative Hawaiian and Other Pacific Islander. dHispanic/Latino/Other Spanish. Table 1. Demographics of baseline participants, follow-up participants and follow-up nonparticipants Baseline Follow-up Follow-up nonparticipants Dispenser Poster Total Dispenser Poster Total Dispenser Poster Total Measure % (n) % (n) % (n) % (n) % (n) % (n) % (n) % (n) % (n) Age  21 to 25 34.4 (570) 43.0 (188) 36.2 (758) 30.0 (286) 45.0 (103) 32.9 (389) 40.5 (284) 40.9 (85) 40.5 (369)  26 to 30 33.2 (550) 30.9 (135) 32.7 (685) 37.5 (357) 31.0 (71) 36.2 (428) 27.5 (193) 30.8 (64) 28.2 (257)  31 to 35 20.0 (331) 16.9 (74) 19.4 (405) 21.3 (203) 16.6 (38) 20.4 (241) 18.2 (128) 17.3 (36) 18.0 (164)  36 to 40 8.2 (135) 5.7 (25) 7.6 (160) 7.2 (69) 5.7 (13) 6.9 (82) 9.4 (66) 5.8 (12) 8.6 (78)  41 to 44 4.2 (69) 3.4 (15) 4.0 (84) 4.0 (38) 1.7 (4) 3.6 (42) 4.4 (31) 5.3 (11) 4.6 (42) Education  Some High School 3.9 (64) 3.9 (17) 3.9 (81) 2.7 (26) 2.6 (6) 2.7 (32) 5.4 (38) 5.4 (11) 5.4 (49)  Diploma or GED 18.6 (308) 10.7 (46) 17.0 (354) 16.3 (155) 8.3 (19) 14.7 (174) 21.8 (153) 13.4 (27) 19.9 (180)  Some College 42.1 (696) 34.8 (150) 40.6 (846) 42.9 (409) 28.4 (65) 40.1 (474) 40.9 (287) 42.1 (85) 41.2 (372)  Bachelors 25.1 (415) 32.5 (140) 26.6 (555) 27.0 (257) 38.4 (88) 29.2 (345) 22.5 (158) 25.7 (52) 23.3 (210)  Graduate 10.3 (171) 18.1 (78) 11.9 (249) 11.1 (106) 22.3 (51) 13.3 (157) 9.3 (65) 13.4 (27) 10.2 (92) Race or Ethnicity  White 58.5 (966) 70.5 (304) 61.0 (1270) 61.1 (581) 74.1 (169) 63.6 (750) 55.1 (385) 66.5 (135) 57.6 (520)  Blacka 5.2 (86) 6.3 (27) 5.4 (113) 4.9 (47) 5.3 (12) 5.0 (59) 5.6 (39) 7.4 (15) 6.0 (54)  AIANb 6.8 (113) 5.6 (24) 6.6 (137) 5.8 (55) 5.7 (13) 5.8 (68) 8.3 (58) 5.4 (11) 7.6 (69)  Asian 8.7 (143) 6.3 (27) 8.2 (170) 8.8 (84) 5.3 (12) 8.1 (96) 8.4 (59) 7.4 (15) 8.2 (74)  NHOPIc 2.8 (46) 1.4 (6) 2.5 (52) 2.0 (19) 2.6 (6) 2.1 (25) 3.9 (27) 0.0 (0) 3.0 (27)  Hispanicd 5.5 (91) 3.7 (16) 5.1 (107) 5.8 (55) 2.6 (6) 5.2 (61) 5.2 (36) 4.9 (10) 5.1 (46)  Other or Multiple 12.4 (205) 6.3 (27) 11.1 (232) 11.6 (110) 4.4 (10) 10.2 (120) 13.6 (95) 8.4 (17) 12.4 (112) Baseline Follow-up Follow-up nonparticipants Dispenser Poster Total Dispenser Poster Total Dispenser Poster Total Measure % (n) % (n) % (n) % (n) % (n) % (n) % (n) % (n) % (n) Age  21 to 25 34.4 (570) 43.0 (188) 36.2 (758) 30.0 (286) 45.0 (103) 32.9 (389) 40.5 (284) 40.9 (85) 40.5 (369)  26 to 30 33.2 (550) 30.9 (135) 32.7 (685) 37.5 (357) 31.0 (71) 36.2 (428) 27.5 (193) 30.8 (64) 28.2 (257)  31 to 35 20.0 (331) 16.9 (74) 19.4 (405) 21.3 (203) 16.6 (38) 20.4 (241) 18.2 (128) 17.3 (36) 18.0 (164)  36 to 40 8.2 (135) 5.7 (25) 7.6 (160) 7.2 (69) 5.7 (13) 6.9 (82) 9.4 (66) 5.8 (12) 8.6 (78)  41 to 44 4.2 (69) 3.4 (15) 4.0 (84) 4.0 (38) 1.7 (4) 3.6 (42) 4.4 (31) 5.3 (11) 4.6 (42) Education  Some High School 3.9 (64) 3.9 (17) 3.9 (81) 2.7 (26) 2.6 (6) 2.7 (32) 5.4 (38) 5.4 (11) 5.4 (49)  Diploma or GED 18.6 (308) 10.7 (46) 17.0 (354) 16.3 (155) 8.3 (19) 14.7 (174) 21.8 (153) 13.4 (27) 19.9 (180)  Some College 42.1 (696) 34.8 (150) 40.6 (846) 42.9 (409) 28.4 (65) 40.1 (474) 40.9 (287) 42.1 (85) 41.2 (372)  Bachelors 25.1 (415) 32.5 (140) 26.6 (555) 27.0 (257) 38.4 (88) 29.2 (345) 22.5 (158) 25.7 (52) 23.3 (210)  Graduate 10.3 (171) 18.1 (78) 11.9 (249) 11.1 (106) 22.3 (51) 13.3 (157) 9.3 (65) 13.4 (27) 10.2 (92) Race or Ethnicity  White 58.5 (966) 70.5 (304) 61.0 (1270) 61.1 (581) 74.1 (169) 63.6 (750) 55.1 (385) 66.5 (135) 57.6 (520)  Blacka 5.2 (86) 6.3 (27) 5.4 (113) 4.9 (47) 5.3 (12) 5.0 (59) 5.6 (39) 7.4 (15) 6.0 (54)  AIANb 6.8 (113) 5.6 (24) 6.6 (137) 5.8 (55) 5.7 (13) 5.8 (68) 8.3 (58) 5.4 (11) 7.6 (69)  Asian 8.7 (143) 6.3 (27) 8.2 (170) 8.8 (84) 5.3 (12) 8.1 (96) 8.4 (59) 7.4 (15) 8.2 (74)  NHOPIc 2.8 (46) 1.4 (6) 2.5 (52) 2.0 (19) 2.6 (6) 2.1 (25) 3.9 (27) 0.0 (0) 3.0 (27)  Hispanicd 5.5 (91) 3.7 (16) 5.1 (107) 5.8 (55) 2.6 (6) 5.2 (61) 5.2 (36) 4.9 (10) 5.1 (46)  Other or Multiple 12.4 (205) 6.3 (27) 11.1 (232) 11.6 (110) 4.4 (10) 10.2 (120) 13.6 (95) 8.4 (17) 12.4 (112) aBlack/African American/Negro. bAmerican Indian or Alaskan Native. cNative Hawaiian and Other Pacific Islander. dHispanic/Latino/Other Spanish. Women who completed the follow-up were more educated and older than those who did not complete follow-up (P < 0.01 for education and age). There were no significant differences in ethnicity between the two groups. Survey data collection was challenged early in the process by a small number of participants who submitted multiple responses. Our team followed-up on questionable survey responses with a confirmation e-mail requesting additional information. If no responses were received, or the responses indicated a duplicate survey, they were removed from the sample. Alcohol consumption Women who completed the follow-up survey were more likely to report drinking alcohol in the past 30 days at baseline than those who did not complete the follow-up survey. Among women who reported drinking any alcohol in the past 30 days at baseline, those who completed the follow-up survey reported fewer drinks per drinking day at baseline than those who did not complete follow-up (P = 0.02). There were no statistically significant differences between follow-up participants and nonparticipants in the average number of drinking days or the number of binge drinking occasions reported at baseline. Table 2 presents the patterns of alcohol consumption among participants at both baseline and follow-up. The dispenser group was more likely to report alcohol consumption in the past 30 days both at baseline and follow-up after adjusting for age, education and ethnicity (P < 0.01 for both baseline and follow-up). After adjusting for age, education and ethnicity when the sample was restricted to only participants who completed follow-up the outcome was still statistically significant. Table 2. Alcohol consumption of all participants at baseline and follow-up Baseline Follow-up Dispenser Poster Total Dispenser Poster Total Measure % (n) % (n) % (n) % (n) % (n) % (n) Consumed alcohol in the past 30 days Pa = 0.05 Pa = 0.10  Yes 66.0 (1092) 60.8 (265) 64.9 (1357) 66.5 (630) 60.7 (139) 65.4 (769)  No 34.0 (563) 39.2 (171) 35.1 (734) 33.5 (317) 39.3 (90) 34.6 (407) Days per week Pa < 0.01 Pa = 0.25  1 or fewer 38.2 (416) 28.3 (75) 36.2 (491) 40.4 (254) 34.8 (48) 39.4 (302)  2 to 3 42.8 (467) 46.0 (122) 43.5 (589) 42.3 (266) 50.0 (69) 43.7 (335)  4 or more 19.0 (207) 25.7 (68) 20.3 (275) 17.3 (109) 15.2 (21) 16.9 (130) Average Number of drinks per occasion Pa = 0.19 Pa = 0.63  1 or fewer drinks 26.7 (281) 20.7 (53) 25.5 (334) 27.4 (165) 31.2 (43) 28.1 (208)  More than 1 to less than 4 drinks 54.4 (573) 56.3 (144) 54.7 (717) 54.9 (331) 50.0 (69) 54.0 (400)  4 to less than 5 drinks 7.4 (78) 8.6 (22) 7.6 (100) 9.5 (57) 11.6 (16) 9.9 (73)  5 or more drinks 11.6 (122) 14.5 (37) 12.1 (159) 8.3 (50) 7.2 (10) 8.1 (60) Number of occasions 4 or more drinks were consumed Pa = 0.16 Pa = 0.62  0 33.5 (366) 26.3 (70) 32.1 (436) 40.3 (253) 38.1 (53) 39.9 (306)  1 to 4 50.6 (553) 53.8 (143) 51.3 (696) 49.4 (310) 48.2 (67) 49.2 (377)  5 to 8 9.0 (98) 12.8 (34) 9.7 (132) 6.8 (43) 8.6 (12) 7.2 (55)  9 to 12 3.6 (39) 3.4 (9) 3.5 (48) 1.8 (11) 2.2 (3) 1.8 (14)  13 to 16 1.5 (16) 1.1 (3) 1.4 (19) 0.3 (2) 1.4 (2) 0.5 (4)  ≥17 1.8 (20) 2.6 (7) 2.0 (27) 1.4 (9) 1.4 (2) 1.4 (11) Baseline Follow-up Dispenser Poster Total Dispenser Poster Total Measure % (n) % (n) % (n) % (n) % (n) % (n) Consumed alcohol in the past 30 days Pa = 0.05 Pa = 0.10  Yes 66.0 (1092) 60.8 (265) 64.9 (1357) 66.5 (630) 60.7 (139) 65.4 (769)  No 34.0 (563) 39.2 (171) 35.1 (734) 33.5 (317) 39.3 (90) 34.6 (407) Days per week Pa < 0.01 Pa = 0.25  1 or fewer 38.2 (416) 28.3 (75) 36.2 (491) 40.4 (254) 34.8 (48) 39.4 (302)  2 to 3 42.8 (467) 46.0 (122) 43.5 (589) 42.3 (266) 50.0 (69) 43.7 (335)  4 or more 19.0 (207) 25.7 (68) 20.3 (275) 17.3 (109) 15.2 (21) 16.9 (130) Average Number of drinks per occasion Pa = 0.19 Pa = 0.63  1 or fewer drinks 26.7 (281) 20.7 (53) 25.5 (334) 27.4 (165) 31.2 (43) 28.1 (208)  More than 1 to less than 4 drinks 54.4 (573) 56.3 (144) 54.7 (717) 54.9 (331) 50.0 (69) 54.0 (400)  4 to less than 5 drinks 7.4 (78) 8.6 (22) 7.6 (100) 9.5 (57) 11.6 (16) 9.9 (73)  5 or more drinks 11.6 (122) 14.5 (37) 12.1 (159) 8.3 (50) 7.2 (10) 8.1 (60) Number of occasions 4 or more drinks were consumed Pa = 0.16 Pa = 0.62  0 33.5 (366) 26.3 (70) 32.1 (436) 40.3 (253) 38.1 (53) 39.9 (306)  1 to 4 50.6 (553) 53.8 (143) 51.3 (696) 49.4 (310) 48.2 (67) 49.2 (377)  5 to 8 9.0 (98) 12.8 (34) 9.7 (132) 6.8 (43) 8.6 (12) 7.2 (55)  9 to 12 3.6 (39) 3.4 (9) 3.5 (48) 1.8 (11) 2.2 (3) 1.8 (14)  13 to 16 1.5 (16) 1.1 (3) 1.4 (19) 0.3 (2) 1.4 (2) 0.5 (4)  ≥17 1.8 (20) 2.6 (7) 2.0 (27) 1.4 (9) 1.4 (2) 1.4 (11) aP-value for difference between dispenser and poster groups based on Chi-Square analysis adjusted for age, education and ethnicity. Table 2. Alcohol consumption of all participants at baseline and follow-up Baseline Follow-up Dispenser Poster Total Dispenser Poster Total Measure % (n) % (n) % (n) % (n) % (n) % (n) Consumed alcohol in the past 30 days Pa = 0.05 Pa = 0.10  Yes 66.0 (1092) 60.8 (265) 64.9 (1357) 66.5 (630) 60.7 (139) 65.4 (769)  No 34.0 (563) 39.2 (171) 35.1 (734) 33.5 (317) 39.3 (90) 34.6 (407) Days per week Pa < 0.01 Pa = 0.25  1 or fewer 38.2 (416) 28.3 (75) 36.2 (491) 40.4 (254) 34.8 (48) 39.4 (302)  2 to 3 42.8 (467) 46.0 (122) 43.5 (589) 42.3 (266) 50.0 (69) 43.7 (335)  4 or more 19.0 (207) 25.7 (68) 20.3 (275) 17.3 (109) 15.2 (21) 16.9 (130) Average Number of drinks per occasion Pa = 0.19 Pa = 0.63  1 or fewer drinks 26.7 (281) 20.7 (53) 25.5 (334) 27.4 (165) 31.2 (43) 28.1 (208)  More than 1 to less than 4 drinks 54.4 (573) 56.3 (144) 54.7 (717) 54.9 (331) 50.0 (69) 54.0 (400)  4 to less than 5 drinks 7.4 (78) 8.6 (22) 7.6 (100) 9.5 (57) 11.6 (16) 9.9 (73)  5 or more drinks 11.6 (122) 14.5 (37) 12.1 (159) 8.3 (50) 7.2 (10) 8.1 (60) Number of occasions 4 or more drinks were consumed Pa = 0.16 Pa = 0.62  0 33.5 (366) 26.3 (70) 32.1 (436) 40.3 (253) 38.1 (53) 39.9 (306)  1 to 4 50.6 (553) 53.8 (143) 51.3 (696) 49.4 (310) 48.2 (67) 49.2 (377)  5 to 8 9.0 (98) 12.8 (34) 9.7 (132) 6.8 (43) 8.6 (12) 7.2 (55)  9 to 12 3.6 (39) 3.4 (9) 3.5 (48) 1.8 (11) 2.2 (3) 1.8 (14)  13 to 16 1.5 (16) 1.1 (3) 1.4 (19) 0.3 (2) 1.4 (2) 0.5 (4)  ≥17 1.8 (20) 2.6 (7) 2.0 (27) 1.4 (9) 1.4 (2) 1.4 (11) Baseline Follow-up Dispenser Poster Total Dispenser Poster Total Measure % (n) % (n) % (n) % (n) % (n) % (n) Consumed alcohol in the past 30 days Pa = 0.05 Pa = 0.10  Yes 66.0 (1092) 60.8 (265) 64.9 (1357) 66.5 (630) 60.7 (139) 65.4 (769)  No 34.0 (563) 39.2 (171) 35.1 (734) 33.5 (317) 39.3 (90) 34.6 (407) Days per week Pa < 0.01 Pa = 0.25  1 or fewer 38.2 (416) 28.3 (75) 36.2 (491) 40.4 (254) 34.8 (48) 39.4 (302)  2 to 3 42.8 (467) 46.0 (122) 43.5 (589) 42.3 (266) 50.0 (69) 43.7 (335)  4 or more 19.0 (207) 25.7 (68) 20.3 (275) 17.3 (109) 15.2 (21) 16.9 (130) Average Number of drinks per occasion Pa = 0.19 Pa = 0.63  1 or fewer drinks 26.7 (281) 20.7 (53) 25.5 (334) 27.4 (165) 31.2 (43) 28.1 (208)  More than 1 to less than 4 drinks 54.4 (573) 56.3 (144) 54.7 (717) 54.9 (331) 50.0 (69) 54.0 (400)  4 to less than 5 drinks 7.4 (78) 8.6 (22) 7.6 (100) 9.5 (57) 11.6 (16) 9.9 (73)  5 or more drinks 11.6 (122) 14.5 (37) 12.1 (159) 8.3 (50) 7.2 (10) 8.1 (60) Number of occasions 4 or more drinks were consumed Pa = 0.16 Pa = 0.62  0 33.5 (366) 26.3 (70) 32.1 (436) 40.3 (253) 38.1 (53) 39.9 (306)  1 to 4 50.6 (553) 53.8 (143) 51.3 (696) 49.4 (310) 48.2 (67) 49.2 (377)  5 to 8 9.0 (98) 12.8 (34) 9.7 (132) 6.8 (43) 8.6 (12) 7.2 (55)  9 to 12 3.6 (39) 3.4 (9) 3.5 (48) 1.8 (11) 2.2 (3) 1.8 (14)  13 to 16 1.5 (16) 1.1 (3) 1.4 (19) 0.3 (2) 1.4 (2) 0.5 (4)  ≥17 1.8 (20) 2.6 (7) 2.0 (27) 1.4 (9) 1.4 (2) 1.4 (11) aP-value for difference between dispenser and poster groups based on Chi-Square analysis adjusted for age, education and ethnicity. Among women who reported any alcohol consumption in the past 30 days at baseline, women who saw the FASD prevention message on the pregnancy test dispensers reported fewer drinking days per week then women who saw the message on a poster (P < 0.01). These differences remained statistically significant after adjusting for age, education and ethnicity (P < 0.01). Differences in drinking behavior between intervention groups were not statistically significant when restricted to participants who completed both baseline and follow-up. All other measures of drinking behavior were similar between intervention groups both at baseline and follow-up. Pregnancy tests dispensed Throughout the course of the study, we distributed over 12,000 pregnancy tests in 5 dispenser communities. In those communities 1688 participants completed a baseline survey, and 413 reported taking a pregnancy test from the dispenser. Of these 413 women, 265 reported using it and 43 (16.2%) had a positive test result. Yukon participants, who incurred a small charge for the pregnancy tests, were less likely to obtain a pregnancy test from the dispenser than Alaska participants who could obtain them for free (Alaska: 25.2%; Yukon: 10.0%; P = 0.06). Of the 43 women who had a positive pregnancy test result, 12 (27.9%) reported consuming alcohol during the past 30 days. Seventeen completed a follow-up survey with six again reporting being pregnant. Pregnancies A total of 192 participants reported being or becoming pregnant during the study. Of those participants, 94 reported knowing they were pregnant at baseline when they first viewed the FASD prevention message. An additional 43 women found out they were pregnant by using a pregnancy test they obtained from a dispenser. A total of 71 participants reported being pregnant at follow-up. Sixteen of these women had also reported being pregnant at baseline. Figure 2 illustrates the attrition and pregnancy status of participants at baseline and follow-up. Fig. 2. View largeDownload slide Attrition and pregnancy status of participants. Fig. 2. View largeDownload slide Attrition and pregnancy status of participants. Table 3 presents drinking habits for the 94 women who knew they were pregnant before participating in the study and the 71 participants who reported they were pregnant at follow-up. Almost 40% of pregnant participants both at baseline (36.2%) and follow-up (39.4%) reported no alcohol consumption between their last period and when they found out they were pregnant. Alcohol consumption among pregnant women was lower at follow-up than at baseline. Less than 20% reported drinking any alcohol once they knew they were pregnant, 18.1% at baseline and 12.7% at follow-up. Table 3. Alcohol consumption of pregnant participants and participants who reported they may have been pregnant during the 6 months prior to completing their survey Measure Pregnant Possibly pregnant % (n) % (n) % (n) % (n) Alcohol intake between last period to recognition of pregnancy  No Alcohol 36.2 (34) 39.4 (28) –a – – –  1 to 3 drinks at least once 30.9 (29) 46.5 (33) – – – –  1 to 3 drinks at a time, 1 or 2 times a week 11.7 (11) 8.5 (6) – – – –  1 to 3 drinks at a time, 3 or more times a week 6.4 (6) 2.8 (2) – – – –  4 or more drinks at one time, at least once 14.9 (14) 2.8 (2) – – – – Alcohol use since pregnancy recognized  No Alcohol 81.9 (77) 87.3 (62) – – – –  1 to 3 drinks at least once 0.0 (0) 12.7 (9) – – – –  1 to 3 drinks at a time, 1 or 2 times a week 5.3 (5) 0.0 (0) – – – –  1 to 3 drinks at a time, 3 or more times a week 6.4 (6) 0.0 (0) – – – –  4 or more drinks at one time, at least once 6.4 (6) 0.0 (0) – – – – Alcohol consumption during 6 months while pregnancy status was unknown  No Alcohol – – – – 32.5 (188) 35.0 (104)  1 to 3 drinks at least once – – – – 25.2 (146) 33.0 (98)  1 to 3 drinks at a time, 1 or 2 times a week – – – – 20.4 (118) 13.5 (40)  1 to 3 drinks at a time, 3 or more times a week – – – – 8.3 (48) 9.4 (28)  4 or more drinks at one time, at least once – – – – 13.6 (79) 9.1 (27) Consumed alcohol in the past 30 days  Yes 17.0 (16) 9.9 (7) 71.8 (416) 72.6 (215)  No 83.0 (78) 90.1 (64) 28.2 (163) 27.4 (81)  Days per week  1 or fewer 62.5 (10) 100 (7) 38.1 (158) 41.4 (89)  2 to 3 25.0 (4) 0.0 (0) 40.7 (169) 44.7 (96)  4 or more 12.5 (2) 0.0 (0) 21.2 (88) 14.0 (30) Average Number of drinks per occasion  1 or fewer drinks 28.6 (4) 85.7 (6) 27.9 (113) 28.5 (59)  More than 1 to less than 4 drinks 50.0 (7) 14.3 (1) 52.1 (211) 49.3 (102)  4 to less than 5 drinks 14.3 (2) 0.0 (0) 6.2 (25) 12.6 (26)  5 or more drinks 7.1 (1) 0.0 (0) 13.8 (56) 9.7 (20) Number of occasions 4 or more drinks were consumed  0 56.3 (9) 100 (7) 33.2 (138) 39.1 (84)  1 to 4 31.3 (5) 0.0 (0) 45.2 (188) 51.6 (111)  5 to 8 6.3 (1) 0.0 (0) 11.8 (49) 4.7 (10)  9 to 12 6.3 (1) 0.0 (0) 4.8 (20) 1.9 (4)  13 to 16 0.0 (0) 0.0 (0) 2.4 (10) 0.5 (1)  ≥17 0.0 (0) 0.0 (0) 2.6 (11) 2.3 (5) Measure Pregnant Possibly pregnant % (n) % (n) % (n) % (n) Alcohol intake between last period to recognition of pregnancy  No Alcohol 36.2 (34) 39.4 (28) –a – – –  1 to 3 drinks at least once 30.9 (29) 46.5 (33) – – – –  1 to 3 drinks at a time, 1 or 2 times a week 11.7 (11) 8.5 (6) – – – –  1 to 3 drinks at a time, 3 or more times a week 6.4 (6) 2.8 (2) – – – –  4 or more drinks at one time, at least once 14.9 (14) 2.8 (2) – – – – Alcohol use since pregnancy recognized  No Alcohol 81.9 (77) 87.3 (62) – – – –  1 to 3 drinks at least once 0.0 (0) 12.7 (9) – – – –  1 to 3 drinks at a time, 1 or 2 times a week 5.3 (5) 0.0 (0) – – – –  1 to 3 drinks at a time, 3 or more times a week 6.4 (6) 0.0 (0) – – – –  4 or more drinks at one time, at least once 6.4 (6) 0.0 (0) – – – – Alcohol consumption during 6 months while pregnancy status was unknown  No Alcohol – – – – 32.5 (188) 35.0 (104)  1 to 3 drinks at least once – – – – 25.2 (146) 33.0 (98)  1 to 3 drinks at a time, 1 or 2 times a week – – – – 20.4 (118) 13.5 (40)  1 to 3 drinks at a time, 3 or more times a week – – – – 8.3 (48) 9.4 (28)  4 or more drinks at one time, at least once – – – – 13.6 (79) 9.1 (27) Consumed alcohol in the past 30 days  Yes 17.0 (16) 9.9 (7) 71.8 (416) 72.6 (215)  No 83.0 (78) 90.1 (64) 28.2 (163) 27.4 (81)  Days per week  1 or fewer 62.5 (10) 100 (7) 38.1 (158) 41.4 (89)  2 to 3 25.0 (4) 0.0 (0) 40.7 (169) 44.7 (96)  4 or more 12.5 (2) 0.0 (0) 21.2 (88) 14.0 (30) Average Number of drinks per occasion  1 or fewer drinks 28.6 (4) 85.7 (6) 27.9 (113) 28.5 (59)  More than 1 to less than 4 drinks 50.0 (7) 14.3 (1) 52.1 (211) 49.3 (102)  4 to less than 5 drinks 14.3 (2) 0.0 (0) 6.2 (25) 12.6 (26)  5 or more drinks 7.1 (1) 0.0 (0) 13.8 (56) 9.7 (20) Number of occasions 4 or more drinks were consumed  0 56.3 (9) 100 (7) 33.2 (138) 39.1 (84)  1 to 4 31.3 (5) 0.0 (0) 45.2 (188) 51.6 (111)  5 to 8 6.3 (1) 0.0 (0) 11.8 (49) 4.7 (10)  9 to 12 6.3 (1) 0.0 (0) 4.8 (20) 1.9 (4)  13 to 16 0.0 (0) 0.0 (0) 2.4 (10) 0.5 (1)  ≥17 0.0 (0) 0.0 (0) 2.6 (11) 2.3 (5) aSubsample was not measured on the listed variable. Table 3. Alcohol consumption of pregnant participants and participants who reported they may have been pregnant during the 6 months prior to completing their survey Measure Pregnant Possibly pregnant % (n) % (n) % (n) % (n) Alcohol intake between last period to recognition of pregnancy  No Alcohol 36.2 (34) 39.4 (28) –a – – –  1 to 3 drinks at least once 30.9 (29) 46.5 (33) – – – –  1 to 3 drinks at a time, 1 or 2 times a week 11.7 (11) 8.5 (6) – – – –  1 to 3 drinks at a time, 3 or more times a week 6.4 (6) 2.8 (2) – – – –  4 or more drinks at one time, at least once 14.9 (14) 2.8 (2) – – – – Alcohol use since pregnancy recognized  No Alcohol 81.9 (77) 87.3 (62) – – – –  1 to 3 drinks at least once 0.0 (0) 12.7 (9) – – – –  1 to 3 drinks at a time, 1 or 2 times a week 5.3 (5) 0.0 (0) – – – –  1 to 3 drinks at a time, 3 or more times a week 6.4 (6) 0.0 (0) – – – –  4 or more drinks at one time, at least once 6.4 (6) 0.0 (0) – – – – Alcohol consumption during 6 months while pregnancy status was unknown  No Alcohol – – – – 32.5 (188) 35.0 (104)  1 to 3 drinks at least once – – – – 25.2 (146) 33.0 (98)  1 to 3 drinks at a time, 1 or 2 times a week – – – – 20.4 (118) 13.5 (40)  1 to 3 drinks at a time, 3 or more times a week – – – – 8.3 (48) 9.4 (28)  4 or more drinks at one time, at least once – – – – 13.6 (79) 9.1 (27) Consumed alcohol in the past 30 days  Yes 17.0 (16) 9.9 (7) 71.8 (416) 72.6 (215)  No 83.0 (78) 90.1 (64) 28.2 (163) 27.4 (81)  Days per week  1 or fewer 62.5 (10) 100 (7) 38.1 (158) 41.4 (89)  2 to 3 25.0 (4) 0.0 (0) 40.7 (169) 44.7 (96)  4 or more 12.5 (2) 0.0 (0) 21.2 (88) 14.0 (30) Average Number of drinks per occasion  1 or fewer drinks 28.6 (4) 85.7 (6) 27.9 (113) 28.5 (59)  More than 1 to less than 4 drinks 50.0 (7) 14.3 (1) 52.1 (211) 49.3 (102)  4 to less than 5 drinks 14.3 (2) 0.0 (0) 6.2 (25) 12.6 (26)  5 or more drinks 7.1 (1) 0.0 (0) 13.8 (56) 9.7 (20) Number of occasions 4 or more drinks were consumed  0 56.3 (9) 100 (7) 33.2 (138) 39.1 (84)  1 to 4 31.3 (5) 0.0 (0) 45.2 (188) 51.6 (111)  5 to 8 6.3 (1) 0.0 (0) 11.8 (49) 4.7 (10)  9 to 12 6.3 (1) 0.0 (0) 4.8 (20) 1.9 (4)  13 to 16 0.0 (0) 0.0 (0) 2.4 (10) 0.5 (1)  ≥17 0.0 (0) 0.0 (0) 2.6 (11) 2.3 (5) Measure Pregnant Possibly pregnant % (n) % (n) % (n) % (n) Alcohol intake between last period to recognition of pregnancy  No Alcohol 36.2 (34) 39.4 (28) –a – – –  1 to 3 drinks at least once 30.9 (29) 46.5 (33) – – – –  1 to 3 drinks at a time, 1 or 2 times a week 11.7 (11) 8.5 (6) – – – –  1 to 3 drinks at a time, 3 or more times a week 6.4 (6) 2.8 (2) – – – –  4 or more drinks at one time, at least once 14.9 (14) 2.8 (2) – – – – Alcohol use since pregnancy recognized  No Alcohol 81.9 (77) 87.3 (62) – – – –  1 to 3 drinks at least once 0.0 (0) 12.7 (9) – – – –  1 to 3 drinks at a time, 1 or 2 times a week 5.3 (5) 0.0 (0) – – – –  1 to 3 drinks at a time, 3 or more times a week 6.4 (6) 0.0 (0) – – – –  4 or more drinks at one time, at least once 6.4 (6) 0.0 (0) – – – – Alcohol consumption during 6 months while pregnancy status was unknown  No Alcohol – – – – 32.5 (188) 35.0 (104)  1 to 3 drinks at least once – – – – 25.2 (146) 33.0 (98)  1 to 3 drinks at a time, 1 or 2 times a week – – – – 20.4 (118) 13.5 (40)  1 to 3 drinks at a time, 3 or more times a week – – – – 8.3 (48) 9.4 (28)  4 or more drinks at one time, at least once – – – – 13.6 (79) 9.1 (27) Consumed alcohol in the past 30 days  Yes 17.0 (16) 9.9 (7) 71.8 (416) 72.6 (215)  No 83.0 (78) 90.1 (64) 28.2 (163) 27.4 (81)  Days per week  1 or fewer 62.5 (10) 100 (7) 38.1 (158) 41.4 (89)  2 to 3 25.0 (4) 0.0 (0) 40.7 (169) 44.7 (96)  4 or more 12.5 (2) 0.0 (0) 21.2 (88) 14.0 (30) Average Number of drinks per occasion  1 or fewer drinks 28.6 (4) 85.7 (6) 27.9 (113) 28.5 (59)  More than 1 to less than 4 drinks 50.0 (7) 14.3 (1) 52.1 (211) 49.3 (102)  4 to less than 5 drinks 14.3 (2) 0.0 (0) 6.2 (25) 12.6 (26)  5 or more drinks 7.1 (1) 0.0 (0) 13.8 (56) 9.7 (20) Number of occasions 4 or more drinks were consumed  0 56.3 (9) 100 (7) 33.2 (138) 39.1 (84)  1 to 4 31.3 (5) 0.0 (0) 45.2 (188) 51.6 (111)  5 to 8 6.3 (1) 0.0 (0) 11.8 (49) 4.7 (10)  9 to 12 6.3 (1) 0.0 (0) 4.8 (20) 1.9 (4)  13 to 16 0.0 (0) 0.0 (0) 2.4 (10) 0.5 (1)  ≥17 0.0 (0) 0.0 (0) 2.6 (11) 2.3 (5) aSubsample was not measured on the listed variable. Possible pregnancy At baseline 579 participants reported there was a time they may have been pregnant in the past 6 months, as did 297 participants at follow-up. Table 3 shows alcohol consumption among these participants at baseline and follow-up. At both time points, approximately 65% reported consuming alcohol while knowing that they could be pregnant. Nearly 72% of these participants reported consuming alcohol within the past 30 days and approximately 20% of participants reported consuming an average of four or more drinks on the days they consumed alcohol. More than 60% reported one or more occasions of binge drinking in the past 30 days. Knowledge and beliefs of FASD The dispenser group scored significantly higher on the questions about their knowledge and beliefs than the poster group at both baseline (Mean Dispenser score: 4.3; Mean Poster score: 4.1; P < 0.01) and follow-up (Dispenser: 4.4; Poster: 4.1; P < 0.01), indicating a more accurate understanding of FASD risks and harms. These differences remained significant even after adjusting for age, education and ethnicity (P < 0.01 for both baseline and follow-up). The mean of the health belief questions was higher for those who consumed alcohol in the past 30 days than those who reported not drinking at both baseline (Mean of those who drank: 4.4, Mean of those who did not drink: 4.1; P < 0.01) and follow-up (Mean of those who drank: 4.5, Mean of those who did not drink: 4.1; P < 0.01). The Cronbach’s alpha reliability coefficient for the FASD health belief questions was r = 0.65 at baseline and r = 0.67 at follow-up. Corrected item-to-total correlation for all items were ≥0.35 for both baseline and follow-up. The five items provided positive data indicating all five questions contributed to the overall reliability of the FASD health belief questions. Future assessment of health beliefs about FASD should expand the types of questions to achieve an Internal Consistency coefficient of r = 0.80. Table 4 shows the means, standard deviations and corrected item-to-total correlation for each FASD health belief questions. Table 4. FASD health belief question analysis outcomes at baseline and follow-up Baseline Follow-up Dispenser Poster Total Dispenser Poster Total Question M (SD) M (SD) M (SD) CI-TC M (SD) M (SD) M (SD) CI-TC As long as a woman does not drink too much alcohol, it is safe to have 1–2 alcoholic drinks at one time when pregnant.a 1.6 (1.2) 2.0 (1.3) 1.7 (1.2) 0.42 1.5 (1.0) 2.0 (1.2) 1.6 (1.0) 0.43 Evidence shows that drinking alcohol during pregnancy affects the health of the child after he or she is born. 4.5 (0.9) 4.4 (1.1) 4.5 (1.0) 0.49 4.6 (0.8) 4.2 (1.2) 4.5 (0.9) 0.57 Drinking alcohol during pregnancy will harm the mother’s health more than the unborn child’s health.a 2.1 (1.3) 2.2 (1.3) 2.1 (1.3) 0.35 1.9 (1.2) 1.9 (1.1) 1.9 (1.2) 0.37 Raising a child with a Fetal Alcohol Spectrum Disorder is hard. 4.5 (0.9) 4.3 (1.1) 4.4 (1.0) 0.43 4.6 (0.8) 4.3 (1.1) 4.6 (0.9) 0.42 If a woman drinks alcohol before she knows she is pregnant, her unborn child could still be affected. 4.2 (1.1) 4.0 (1.1) 4.2 (1.1) 0.39 4.3 (1.0) 4.1 (1.1) 4.2 (1.1) 0.37 Total FASD HB Score 4.3 (0.7) 4.1 (0.8) 4.3 (0.7) r = 0.65 4.4 (0.6)b 4.1 (0.8)c 4.4 (0.7)c r = 0.67 Baseline Follow-up Dispenser Poster Total Dispenser Poster Total Question M (SD) M (SD) M (SD) CI-TC M (SD) M (SD) M (SD) CI-TC As long as a woman does not drink too much alcohol, it is safe to have 1–2 alcoholic drinks at one time when pregnant.a 1.6 (1.2) 2.0 (1.3) 1.7 (1.2) 0.42 1.5 (1.0) 2.0 (1.2) 1.6 (1.0) 0.43 Evidence shows that drinking alcohol during pregnancy affects the health of the child after he or she is born. 4.5 (0.9) 4.4 (1.1) 4.5 (1.0) 0.49 4.6 (0.8) 4.2 (1.2) 4.5 (0.9) 0.57 Drinking alcohol during pregnancy will harm the mother’s health more than the unborn child’s health.a 2.1 (1.3) 2.2 (1.3) 2.1 (1.3) 0.35 1.9 (1.2) 1.9 (1.1) 1.9 (1.2) 0.37 Raising a child with a Fetal Alcohol Spectrum Disorder is hard. 4.5 (0.9) 4.3 (1.1) 4.4 (1.0) 0.43 4.6 (0.8) 4.3 (1.1) 4.6 (0.9) 0.42 If a woman drinks alcohol before she knows she is pregnant, her unborn child could still be affected. 4.2 (1.1) 4.0 (1.1) 4.2 (1.1) 0.39 4.3 (1.0) 4.1 (1.1) 4.2 (1.1) 0.37 Total FASD HB Score 4.3 (0.7) 4.1 (0.8) 4.3 (0.7) r = 0.65 4.4 (0.6)b 4.1 (0.8)c 4.4 (0.7)c r = 0.67 M: mean. SD: standard deviation. CI-TC: corrected item-total correlation. r: Cronbach's alpha. aItem reverse coded when summed into Total FASD HB score. bP < 0.05 for difference between baseline and follow-up cP < 0.01 for difference between baseline and follow-up Table 4. FASD health belief question analysis outcomes at baseline and follow-up Baseline Follow-up Dispenser Poster Total Dispenser Poster Total Question M (SD) M (SD) M (SD) CI-TC M (SD) M (SD) M (SD) CI-TC As long as a woman does not drink too much alcohol, it is safe to have 1–2 alcoholic drinks at one time when pregnant.a 1.6 (1.2) 2.0 (1.3) 1.7 (1.2) 0.42 1.5 (1.0) 2.0 (1.2) 1.6 (1.0) 0.43 Evidence shows that drinking alcohol during pregnancy affects the health of the child after he or she is born. 4.5 (0.9) 4.4 (1.1) 4.5 (1.0) 0.49 4.6 (0.8) 4.2 (1.2) 4.5 (0.9) 0.57 Drinking alcohol during pregnancy will harm the mother’s health more than the unborn child’s health.a 2.1 (1.3) 2.2 (1.3) 2.1 (1.3) 0.35 1.9 (1.2) 1.9 (1.1) 1.9 (1.2) 0.37 Raising a child with a Fetal Alcohol Spectrum Disorder is hard. 4.5 (0.9) 4.3 (1.1) 4.4 (1.0) 0.43 4.6 (0.8) 4.3 (1.1) 4.6 (0.9) 0.42 If a woman drinks alcohol before she knows she is pregnant, her unborn child could still be affected. 4.2 (1.1) 4.0 (1.1) 4.2 (1.1) 0.39 4.3 (1.0) 4.1 (1.1) 4.2 (1.1) 0.37 Total FASD HB Score 4.3 (0.7) 4.1 (0.8) 4.3 (0.7) r = 0.65 4.4 (0.6)b 4.1 (0.8)c 4.4 (0.7)c r = 0.67 Baseline Follow-up Dispenser Poster Total Dispenser Poster Total Question M (SD) M (SD) M (SD) CI-TC M (SD) M (SD) M (SD) CI-TC As long as a woman does not drink too much alcohol, it is safe to have 1–2 alcoholic drinks at one time when pregnant.a 1.6 (1.2) 2.0 (1.3) 1.7 (1.2) 0.42 1.5 (1.0) 2.0 (1.2) 1.6 (1.0) 0.43 Evidence shows that drinking alcohol during pregnancy affects the health of the child after he or she is born. 4.5 (0.9) 4.4 (1.1) 4.5 (1.0) 0.49 4.6 (0.8) 4.2 (1.2) 4.5 (0.9) 0.57 Drinking alcohol during pregnancy will harm the mother’s health more than the unborn child’s health.a 2.1 (1.3) 2.2 (1.3) 2.1 (1.3) 0.35 1.9 (1.2) 1.9 (1.1) 1.9 (1.2) 0.37 Raising a child with a Fetal Alcohol Spectrum Disorder is hard. 4.5 (0.9) 4.3 (1.1) 4.4 (1.0) 0.43 4.6 (0.8) 4.3 (1.1) 4.6 (0.9) 0.42 If a woman drinks alcohol before she knows she is pregnant, her unborn child could still be affected. 4.2 (1.1) 4.0 (1.1) 4.2 (1.1) 0.39 4.3 (1.0) 4.1 (1.1) 4.2 (1.1) 0.37 Total FASD HB Score 4.3 (0.7) 4.1 (0.8) 4.3 (0.7) r = 0.65 4.4 (0.6)b 4.1 (0.8)c 4.4 (0.7)c r = 0.67 M: mean. SD: standard deviation. CI-TC: corrected item-total correlation. r: Cronbach's alpha. aItem reverse coded when summed into Total FASD HB score. bP < 0.05 for difference between baseline and follow-up cP < 0.01 for difference between baseline and follow-up DISCUSSION This is the first rigorous feasibility study of an FASD prevention strategy employing pregnancy tests in establishments serving alcohol. Interventions were either an FASD informational poster affixed to a pregnancy test dispenser or an FASD informational poster alone. The goal of this study was to assess the feasibility of deploying FASD prevention messages on either a poster or a pregnancy test dispenser in women’s restrooms and collect pilot data to measure the utility of both interventions at promoting informed alcohol consumption behavior. Participating establishments reported generally positive comments from patrons. No establishments requested that the messages be removed prior to the end of the study. Local staff reported anecdotal evidence that patrons at pregnancy test dispenser sites were employing those tests on-site prior to consuming alcohol. Both interventions were effective at engaging women with FASD prevention messages and increasing knowledge of risks associated with alcohol consumption during pregnancy. Participants who viewed the poster on a pregnancy test dispenser had better knowledge and beliefs than their poster only counterparts at both baseline (average score of 4.3 and 4.1, respectively) and follow-up (average score of 4.4 and 4.1, respectively), even after adjusting for education and other demographics. This finding is consistent with the dispenser having a greater immediate impact than the poster alone. A smaller proportion of pregnant participants reported drinking alcohol within the past 30 days at follow-up (10%) than at baseline (17%) which may indicate the intervention influenced pregnant participants to stop drinking. Limitations We did not collect data before the interventions were implemented to determine the participants’ knowledge and belief regarding FASD and the risks of consuming alcohol while pregnant or possibly pregnant. Resource constraints precluded the creation of a control group of multiple sites of women frequenting bars with neither intervention. Thus, it was not possible to determine how much of the difference between the two groups was due to the interventions and how much was due to differences in the two study groups. However, the similarity in the self-reported drinking behavior between the two groups for the 30 days before baseline strengthens our conclusion that the difference in the FASD prevention health belief scores was due in part to the differential effect of the two interventions and not just to pre-existing differences in the two study populations. Because the baseline survey was accessible to anyone who had access to the survey link, it was possible for individuals to respond to the survey more than once or to share the survey link with individuals who had not actually viewed the interventions. We employed a protocol to identify duplicate and suspect responses for the baseline survey which allowed us to eliminate many, but not necessarily all, of these responses. We were also able to better control access to the follow-up surveys by sending an e-mail with the survey link 6 months after the baseline survey was completed. The cost of obtaining a pregnancy test may have affected the number of participants who obtained a test from the Yukon dispensers. However, the sample size for the Yukon dispenser site is very small (n = 36) precluding the ability to accurately determine the effect of cost on the use of the test dispenser. Next steps Future impact evaluations need to increase the number of sites, sample size and participant diversity to increase statistical power and to have a sufficient number of sites in areas with known demographic characteristics. Demographic matching at baseline of a sufficient number of bars with large, diverse samples of women should enable future evaluations to establish ‘Control and Experimental Sites’ to improve the methodological quality and generalizability of assessments of intervention effectiveness (Murray, 1998; Windsor, 2015). We also recommend that future studies consider methological variations such as assessing effects of charging for pregnancy tests and the differential effects of different costs per test, e.g. $1 – $2 – $3, and that the basic variations in intervention types be placed in culturally diverse locations to measure differences in effectiveness between subsamples. The FASD health belief questions provided useful data and insight about perceptions of risk of women of childbearing age. Future studies, however, should consider implementing a process to expand the type and number of health belief questions to enable additional validity and reliability analyses to refine message content and graphics. Public health implications Our team distributed over 12,000 pregnancy tests in eight communities. Interviews with bar managers and staff suggest that tests may have been inappropriately dispensed early in the intervention when they were still perceived as a novelty. However, the overall demand indicates a substantial service gap for accessible and affordable pregnancy tests for women at risk for an inadvertent alcohol-exposed pregnancy. Overcoming barriers to accessible pregnancy tests is especially important in scenarios where the cost of pregnancy tests are prohibitive or personal anonymity is jeopardized. CONCLUSIONS Given the relatively high unintended pregnancy rate, empowering sexually active women to know their pregnancy status before consuming alcohol is a sound public health strategy. Because accessible contraception also remains essential to FASD prevention, our study distributed condoms at all intervention sites. However, while considerable focus in public health has been placed on the importance of contraception, the valuable role of affordable and accessible pregnancy tests appears to be somewhat precluded and should receive more attention. ACKNOWLEDGEMENTS This study was funded by the Alaska State Legislature: 2015–17. We particularly wish to thank the members of our Expert Advisory Group, and Message Testing Group, for their support. We thank Mr. Jody Allen Crowe for support with the placement and maintenance of pregnancy test dispensers in establishments that serve alcohol. The surveys used open-source materials from the University of California, San Diego. They were developed in collaboration with Dr. Christina Chambers of the University of California. We would also like to thank Julie Holden, Sarah Shimer, Ginger Cooley and Rebecca Van Wyck who assisted with the study. FUNDING This project was supported by the Alaska State Legislature. CONFLICT OF INTEREST STATEMENT None declared. REFERENCES Alcohol Use in Pregnancy . Fetal Alcohol Spectr Disord 2016 . Camerlin A Alaska Maternal and Child Health Data Book 2012: Birth Defects Surveillance Edition. Anchorage, AK, 2012 . Capecchi C Would You Take a Pregnancy Test in a Bar? New York Times 2012 . Diagnosis . Fetal Alcohol Spectr Disord 2014 . Ethen M , Ramadhani T , Scheuerle A , et al. . ( 2009 ) Alcohol consumption by women before and during pregnancy . Matern Child Health J 13 : 274 – 85 . Google Scholar CrossRef Search ADS PubMed Gordon R . ( 1983 ) An operational classification of disease prevention . Public Health Rep 98 : 107 – 9 . Google Scholar PubMed Kost K Unintended Pregnancy Rates at the State Level: Estimates for 2010 and Trends Since 2002. New York, 2015 . Maier S , West J . ( 2001 ) Drinking patterns and alcohol-related birth defects . Alcohol Res Heal 25 : 168 – 74 . May PA , Gossage JP , Brooke LE , et al. . ( 2005 ) Maternal risk factors for fetal alcohol syndrome in the Western Cape Province of South Africa: a population-based study . Am J Public Heal J Public Heal 9595 : 1190 – 9 . Google Scholar CrossRef Search ADS Miller L , Tolliver R , Druschel C , et al. . ( 2002 ) Fetal alcohol syndrome: Alaska, Arizona, Colorado, and New York, 1995–1997 . Morb Mortal Wkly Rep 51 : 433 – 5 . Murray DM . ( 1998 ) Design and Analysis of Group-Randomized Trials . New York, NY : Oxford University Press . Notice to readers . ( 2005 ) Surgeon general’s advisory on alcohol use in pregnancy . Morb Mortal Wkly Rep 54 : 229 . PRAMStat: Explore PRAMS Data by State . Centers Dis Control Prev 2010 . State-Specific Alcohol Consumption Rates for 2013 . Fetal Alcohol Spectr Disord 2015 . Stratton K , Howe C , Battaglia F (eds) . ( 1996 ) Fetal Alcohol Syndrome: Diagnosis, Epidemiology, Prevention, and Treatment . Washington, DC : National Academy Press . Tan CH , Denny CH , Cheal NE , et al. . ( 2015 ) Alcohol use and binge drinking among women of childbearing age—United States, 2011–2013 . MMWR Morb Mortal Wkly Rep 64 : 1042 – 6 . Google Scholar CrossRef Search ADS PubMed Williams J , Smith V . ( 2015 ) Fetal alcohol spectrum disorder . Pediatrics 136 : 1395 – 406 . Google Scholar CrossRef Search ADS Windsor R . ( 2015 ) Evaluation of Health Promotion, and Disease Prevention and Management Programs: Improving Population Health through Evidence-Based Practices , 5th ed . Silver Spring, MD : Oxford University Press . Google Scholar CrossRef Search ADS © The Author(s) 2018. Medical Council on Alcohol and Oxford University Press. All rights reserved. This article is published and distributed under the terms of the Oxford University Press, Standard Journals Publication Model (https://academic.oup.com/journals/pages/about_us/legal/notices)

Journal

Alcohol and AlcoholismOxford University Press

Published: Jan 10, 2018

There are no references for this article.

You’re reading a free preview. Subscribe to read the entire article.


DeepDyve is your
personal research library

It’s your single place to instantly
discover and read the research
that matters to you.

Enjoy affordable access to
over 18 million articles from more than
15,000 peer-reviewed journals.

All for just $49/month

Explore the DeepDyve Library

Search

Query the DeepDyve database, plus search all of PubMed and Google Scholar seamlessly

Organize

Save any article or search result from DeepDyve, PubMed, and Google Scholar... all in one place.

Access

Get unlimited, online access to over 18 million full-text articles from more than 15,000 scientific journals.

Your journals are on DeepDyve

Read from thousands of the leading scholarly journals from SpringerNature, Elsevier, Wiley-Blackwell, Oxford University Press and more.

All the latest content is available, no embargo periods.

See the journals in your area

DeepDyve

Freelancer

DeepDyve

Pro

Price

FREE

$49/month
$360/year

Save searches from
Google Scholar,
PubMed

Create lists to
organize your research

Export lists, citations

Read DeepDyve articles

Abstract access only

Unlimited access to over
18 million full-text articles

Print

20 pages / month

PDF Discount

20% off