Assessing the contraceptive needs of homeless women in Salt Lake City, Utah

Background: Homeless women across the US have reduced access to contraception and reproductive healthcare. They are at especially high risk for sexual trauma, contracting HIV or other sexually transmitted infections, and suffering from mental and physical conditions and substance use disorders. The result of the challenges that accompany homelessness is a low prioritization of reproductive planning, putting homeless women at high risk for unintended pregnancy and poor pregnancy outcomes. Objectives: To evaluate the contraceptive needs and pregnancy desires in women experiencing housing instability or homelessness. Methods: A convenience sample of reproductive age women who utilized homeless support or healthcare services across a metropolitan area completed a self-administered survey collecting demographics, reproductive histories and contraceptive experiences. We completed descriptive analyses and a multivariate logistic regression model to explore variables associated with ideal contraceptive method use. Results: We surveyed 76 women with a median age of 29y (range 18-45). Of the total cohort, 56 (74%) had a previous livebirth, 23 (30%) were currently pregnant and 10 (14%) desired a pregnancy in the next year. Current contraceptive methods, among non-pregnant women included 18 (39%) using a Tier 1 method, 9 (20%) a Tier 2 method, and 10 (21) Tiers 3 or 4 methods and 16 (35%) using no method. Of those at risk for unintended pregnancy, 27 (58%) reported not using their ideal contraceptive method and method cost and inability to find a provider were reported as barriers. Women living in transitional housing (motels, doubling up, halfway house, etc.) were less likely to be using their ideal contraceptive method than those established in a shelter (aOR: 0.06, 95% CI: 0.00-0.92). Conclusions: Women living with housing insecurity are less visible in the homeless population than those established in shelters, yet at higher risk for an unintended pregnancy destabilizing a fragile housing situation. Missed opportunities exist for interventions to improve reproductive healthcare coordination, including comprehensive contraceptive care, within transitional housing and homeless outreach services.

ASSESSING THE CONTRACEPTIVE NEEDS OF HOMELESS WOMEN IN SALT LAKE CITY, UTAH by Zoe Kozlowski A Senior Honors Thesis Submitted to the Faculty of The University of Utah In Partial Fulfillment of the Requirements for the Honors Degree in Bachelor of Science In Health, Society, and Policy Approved: ______________________________ Lori Gawron, M.D., M.P.H. Thesis Faculty Supervisor _____________________________ Rebecca Utz Chair, Department of Health, Society, and Policy _______________________________ Rebecca Utz Honors Faculty Advisor _____________________________ Sylvia D. Torti, PhD Dean, Honors College April 2018 Copyright © 2018 All Rights Reserved ABSTRACT Background: Homeless women across the US have reduced access to contraception and reproductive healthcare. They are at especially high risk for sexual trauma, contracting HIV or other sexually transmitted infections, and suffering from mental and physical conditions and substance use disorders. The result of the challenges that accompany homelessness is a low prioritization of reproductive planning, putting homeless women at high risk for unintended pregnancy and poor pregnancy outcomes. Objectives: To evaluate the contraceptive needs and pregnancy desires in women experiencing housing instability or homelessness. Methods: A convenience sample of reproductive age women who utilized homeless support or healthcare services across a metropolitan area completed a self-administered survey collecting demographics, reproductive histories and contraceptive experiences. We completed descriptive analyses and a multivariate logistic regression model to explore variables associated with ideal contraceptive method use. Results: We surveyed 76 women with a median age of 29y (range 18-45). Of the total cohort, 56 (74%) had a previous livebirth, 23 (30%) were currently pregnant and 10 (14%) desired a pregnancy in the next year. Current contraceptive methods, among nonpregnant women included 18 (39%) using a Tier 1 method, 9 (20%) a Tier 2 method, and 10 (21) Tiers 3 or 4 methods and 16 (35%) using no method. Of those at risk for unintended pregnancy, 27 (58%) reported not using their ideal contraceptive method and method cost and inability to find a provider were reported as barriers. Women living in transitional housing (motels, doubling up, halfway house, etc.) were less likely to be ii using their ideal contraceptive method than those established in a shelter (aOR: 0.06, 95% CI: 0.00-0.92). Conclusions: Women living with housing insecurity are less visible in the homeless population than those established in shelters, yet at higher risk for an unintended pregnancy destabilizing a fragile housing situation. Missed opportunities exist for interventions to improve reproductive healthcare coordination, including comprehensive contraceptive care, within transitional housing and homeless outreach services. iii TABLE OF CONTENTS ABSTRACT ii INTRODUCTION 1 METHODS 3 RESULTS 5 DISCUSSION 7 ACKNOWLEDGEMENTS 19 REFERENCES 20 APPENDIXES 22 iv Ending homelessness is a national priority in the United States (Henry, Watt, Rosenthal, & Shivji, 2017). The point in time (PIT) count, led by the U.S. Department of Housing and Urban Development in 2017, found an estimated 553,742 individuals were homeless on a single night nationally and identified approximately 1 in 1,000 people living in the state of Utah as homeless (Comprehensive Report on Homelessness State of Utah, 2017). While men are typically the face of homelessness, single women and those with children in custody make up 40% of the homeless population (Henry, Watt, Rosenthal, & Shivji, 2017). The HUD definition of homelessness, includes both individuals sleeping on the street and those experiencing housing insecurity (including temporary housing, shelters, sharing-rooms, living on couches, etc.), but not those at imminent risk of losing previously stable housing (Henry, Watt, Rosenthal, & Shivji, 2017). Pathways to homelessness are different for women than in men. The leading cause of homelessness in women and families is sexual and intimate partner violence (Silver, Panares). From 2016 to 2017, the state of Utah saw a 20% increase in the percent of individuals who were displaced due to intimate partner violence (Comprehensive Report on Homelessness State of Utah, 2017). Women also have high risk health needs and limited access to care. Approximately one third of the care provided at homeless clinics to women is for chronic disease management (Health Care for Homeless Women, 2013). Many homeless people do not seek treatment and use alcohol and drugs as a way to self-manage chronic conditions (Silver & Panares, 2000). This leads to a reliance on emergency medical services due to reduced access to primary and preventative care (Health Care for Homeless Women, 2013). As a result, women do not prioritize reproductive health needs which puts them at high risk for unintended pregnancy (Kennedy, Grewal, Roberts, Steinauer, & Dehlendorf, 2014). Greater access and utilization of contraceptives could improve pregnancy planning and subsequent pregnancy outcomes. In Utah, thirty-six percent of all pregnancies were unintended in 2010 with disproportionately high rates for low-income women (Kost, 2015). Unintended pregnancy increases risk of adverse pregnancy outcomes (Kost & Lindberg, 2015). Homeless women are 2.9 times more likely to have a preterm birth than the general population, when controlling for age, previous pregnancy outcomes, and smoking status (Health Care for Homeless Women, 2013). Additionally, a review of state vital records found homeless women were less likely to breastfeed or obtain adequate preconception care, such as prenatal vitamins and regular prenatal visits. Their infants had increased risks of low birth weight, prolonged hospital stays and need for neonatal intensive care (Richards, Merrill, & Baksh, 2011). Pregnancies are healthiest when they are planned and contraception is the most effective way to prevent unintended pregnancy. While there is limited research available on the pregnancy intentions of homeless women, previous studies indicate that having children is important to them (Gelberg, Browner, & Arangua, 2008). A qualitative study interviewing 22 homeless women with children found that while most women desired a future pregnancy (or expressed ambivalence toward becoming pregnant), none of the participants desired a pregnancy while homeless. They also found that few women were using contraception and for those who were, utilization was erratic (Kennedy, Grewal, Roberts, Steinauer, & Dehlendorf, 2014). While there are many forms of contraception, method efficacy varies. The most effective forms are long-acting reversible contraceptives (LARC) which includes intrauterine devices (IUD) and the contraceptive implant (small rod inserted into the arm) (Effectiveness of Family Planning Methods, n.d.). While these methods are the most effective, they are costly which limits 2 access to low-income women. Homeless healthcare clinics often have limited contraceptive methods available and may only offer barrier or short acting methods such as condoms or the birth control pill (Dasari et al., n.d.). Unintended pregnancy is a challenge for any woman, but it can cause undue stress for someone already struggling with homelessness. With the added uncertainties surrounding basic shelter, food, safety, and sexual health, having the autonomy and the resources to prevent an unintended pregnancy can make a drastic positive impact for an individual. Currently, there is little research that explores reproductive health needs in homeless women in Utah. Thus, we sought to explore the contraceptive needs and pregnancy desires of homeless women in Salt Lake City to support future clinical interventions and policy changes to better meet this need, and ultimately reduce rates of unintended pregnancy. Methods Study design and eligibility. This is a prospective, cross sectional study using convenience sampling to include women age 18-45y who utilized homeless services and reported housing instability or homelessness in Salt Lake City, Utah. We excluded women outside the designated age range and those who reported they were post-menopausal or had a hysterectomy. The University of Utah Institutional Review Board approved this study. Survey design. We designed a self-administered anonymous survey with an estimated sixth grade literacy level with both quantitative and multiple-choice answers and space for free-text entries. Questions assessed medical and reproductive histories, past, current and ideal contraceptive method use, desire for pregnancy, LARC interest, barriers to accessing care, history of sex exchange, and basic demographic information (See Appendix A). We sought input 3 from family planning and healthcare providers for homeless services on survey content. We piloted the survey for content understanding and question flow prior to finalizing. Recruitment. We sought key stakeholders within homeless services and shelters to ensure sensitivity, best approach in each setting and to ensure site commitment to the study. We recruited women during meal times or other group encounters between October 2017 and March 2018 at the following locations; the YWCA (a shelter for women and children experiencing domestic violence), Fourth Street Clinic (a healthcare clinic for homeless individuals), mobile outreach through Fourth Street Clinic, Youth Resource Center (operated by the VOA; serving youth age 15-23y who are experiencing homelessness), the Midvale Road Home (serving families), Palmer Court Road Home (permanent supportive housing for men, women, and children with a history of chronic homelessness), and the University of Utah Department of Obstetrics and Gynecology Substance Use in Pregnancy Recovery, Addiction and Dependence (SUPeRAD) Prenatal Specialty Clinic (high risk pregnancy clinic for women with drug dependence). At each location, the site coordinators played a vital role in letting the women know that participating in the study was an option for them and the study team was available during survey completion to help with literacy questions as needed. We did not provide compensation for survey completion, although some shelters offered standard participation incentives, such as socks or gloves, as available. Analyses. We stratified participants by those who were using their ideal contraceptive method combination, those who were not, and those who were currently pregnant to compare sociodemographic, medical or mental health and reproductive characteristics. Medical diagnoses included those reported by the US Medical Eligibility Criteria for Contraceptive Use as conditions which increase the risk of adverse pregnancy outcomes in the setting of an unintended 4 pregnancy. (Curtis, et al. 2016) Within each group, we calculated frequencies and proportions for each descriptive variable and compared groups by Chi square tests. Adverse pregnancy outcomes included history of miscarriage, preterm birth and fetal demise (“stillbirth”). (Preterm (Premature) Labor and Birth, 2016). On the question regarding pregnancy intention, we collapsed the responses to allow for three options; desires a pregnancy in the next year (which includes “Yes” and “I’m okay either way”), does not desire a pregnancy (which includes “No, I don’t want any future pregnancies” and “No, I don’t want to become pregnant in the next year, but maybe in the future”), and those not sure of pregnancy desires. We defined housing options as those established in a homeless shelter, those “doubling up” by staying with family or friends, and those in temporary or transitional housing, such as rehabilitation centers, halfway houses, or motels. We then calculated the proportion of participants who reported past or current contraceptive method use and the ideal method they desired for all participants. We then excluded participants who reported a current pregnancy and compared the current method use to the ideal method desired to calculate the proportion of participants at risk for pregnancy who were not using their ideal method. Finally, we compared barriers to contraceptive access, prior contraceptive care and preferred healthcare access points for future care. We then used a multivariate logistic regression model to explore factors associated with ideal contraceptive method use. Results A total of seventy-six women completed self-administered surveys. The participants had a median age of 29 years (range 18-45), 24 women (32%) reported a current pregnancy, and 41 women (54%) reported a history of adverse pregnancy outcomes. They had a mean number of 3 previous pregnancies with a range of 0-20. As shown in table 1, 68% of the participants reported 5 having sex with a man in the past 3 months. We also found that 47% had a history of being forced to have sex and 31% had at some point exchanged sex for drugs, shelter, food, or money. Differences in those using their ideal method or not and those currently pregnant emerged across groups. Only 65% of those currently pregnant were cohabitating or married. Insurance coverage varied, with 96% of pregnant women, 64% of ideal method users and only 48% of those not using their ideal method reporting coverage. (P<0.001) While not significant between groups, only 59% of those not using their ideal method reported high school completion or greater compared to 77% of ideal method users. In comparing housing status across groups, 59% of women using their ideal method were established in a shelter, while 45% of women not using their ideal method reported temporary or transitional housing. (P<0.001) (Table 1) Women reported a range of past and current contraceptive method use and combinations of methods. The most common past methods included male partner-dependent methods, such as male condoms (51%) and withdrawal (36%), birth control pills (43%), and depot medroxyprogesterone acetate (DMPA) (26%). The most common current methods reported were no method (30%), male condoms (17%), and IUDs or tubal ligations (each 15%). The most common ideal methods included pills (20%), IUDs (18%), and implants (17%). (Table 2) In comparing the current method use to the ideal method desires of non-pregnant participants, we found 27 (58%) were not using their ideal method. Seven of the seventy-four respondents (9%) indicated that they were currently using a highly effective method only, which aligned with their ideal response. An additional 7(9%) indicated current moderately effective method use with their ideal being highly effective methods only. (Table 3) While no significant difference was found across groups in barriers to ideal methods or preferences for care, some patterns emerged. Women who were currently pregnant and 6 established in care were less likely to report the ability to find a provider as a barrier to method access and more likely to desire future contraceptive care in a physician’s office. Those using their ideal method were more likely to have used Planned Parenthood in the past and desire care in the future. Those not using their ideal method identified more barriers to care, including cost and feeling judged by a provider than those who used their ideal method. (Table 4) We allowed for free-text responses on whether women would be interested in LARC options and barriers or concerns related to LARC. A number of myths regarding infertility and infections related to IUDs emerged. One woman commented, “I want my tubes tide [sic]. No implants or IUD's please. But I'm poor so I'll take what I can get!”. Some of the comments we received related to LARC concerns were, “That it [an IUD] could cause infection or get gross,” “I don’t want it to give me pain on a daily basis or any other side effects [sic],”and “2 People I know had IUD's and could not get pregnant after removal [sic].” Multiple participants commented with requests for more information such as, “[Just] would like more information first, I don't know much about the actual implant,” “What are the risks?” and “[How] does it work? Will it bend inside?” We explored factors associated with ideal method use using logistic regression models. We found no significant association between age, race/ethnicity, relationship status, education, insurance, sexual activity, pregnancy history or recent incarceration. Controlling for all factors in the multivariable models, housing status remained significant with those reporting transitional or temporary housing less likely to be using their ideal method compared to those established in a shelter (aOR: 0.06, 95% CI: 0.00-0.92). (Table 5) Discussion This study explored the contraceptive needs and reproductive desires in women experiencing homelessness or housing insecurity in the Salt Lake Valley, Utah. We found high 7 rates of adverse pregnancy outcomes, as shown by the proportion of women who had experienced a preterm birth. Participants expressed interest in a wide range of contraceptive methods and 58% were not using their ideal method. We found that 23% percent of women desired a pregnancy. This illustrates the complex relationship that exists with their homeless status and pregnancy desire as was highlighted in previous studies (Gelberg, Browner, & Arangua, 2008). Ensuring access to adequate preconception care is critical for this population, especially for women with low socioeconomic status and multiple comorbidities, which includes the majority of the women we surveyed. The participants in this study expressed interest in a wide range of contraceptive methods, including LARC methods. This need for expanded services was also reflected in Kennedy’s study specifically looking at homeless women with children (Kennedy, Grewal, Roberts, Steinauer, & Dehlendorf, 2014). Due to limited access to a range of methods at homeless healthcare access points, an expansion of available methods is necessary to best meet this need. Currently, there is a widespread shortage of access not only to the methods themselves, but also a lack of information available on the options that exist. In Salt Lake City, women experiencing homeless or housing insecurity are often uninsured or underinsured, limiting where they can seek healthcare services. The 4th Street clinic offers short acting hormonal methods, such as pills or the vaginal ring. An IUD would require applications for a patient assistance program and multiple visits including a gynecology clinic visit that only occurs at monthly intervals. Planned Parenthood of Utah does provide comprehensive contraceptive services on a sliding fee scale so, although it may be reduced, women are still required to pay a fee. This makes their services inaccessible for some women, especially when considering the high costs of LARC methods. 8 Kennedy’s study surmised three key reasons for a lack of seeking and utilizing contraceptive services, including (1) an inability to prioritize health due to competing demands, (2) shelter-related obstacles and restrictive provider practices that impede access to reproductive health care services and the use of contraception, and (3) change in the power dynamics of sexual relationships while homeless, making women more vulnerable to sexual exploitation (Kennedy, Grewal, Roberts, Steinauer, & Dehlendorf, 2014). These results show the barriers to positive sexual health within the homeless population and the complicated relationship between homelessness and the desire to become pregnant for some women (Kennedy, Grewal, Roberts, Steinauer, & Dehlendorf, 2014). In other cases, the struggle lies in how to prevent pregnancy. No matter the desire of the woman, it is important to know what a woman’s pregnancy intentions are and make care available that supports their reproductive goals. When considering the percent of women who have either been forced to have sex or have a history of sex exchange, we see a part of the population that has limited control over whether they have sex. Utilization of short acting and LARC methods would allow women in this situation control over pregnancy risk, but the methods must be accessible. Not only does unintended pregnancy have a significant impact on any woman and but for those who are experiencing homelessness this could be the difference between finding stable housing or remaining in a shelter. There was a higher likelihood for women to use their ideal method combination when living in shelters as compared to women living in transitional housing. This suggests that housing situations with increased instability like rehabilitation facilities, halfway houses, motels, and similar situations impact access to reproductive healthcare. Additionally, this may suggest that shelters and services for those facing housing instability that offer increased long-term support are better equipped to connect residents with a broad range of resources in the 9 community. Integrating contraceptive education and programs that link residents in transitional housing with family planning resources could help to mitigate this effect. As few health providers assess reproductive planning and contraceptive needs outside of reproductive health visits, there is a missed opportunity to close the gap, including in homeless services. One study showed that when primary care clinicians provided contraceptive counseling there was an increase in utilization of contraceptives (Lee et al., 2011). The One Key Question was developed to assist primary care providers in assessing this preventive health measure. The One Key Question- “Would you like to become pregnant in the next year?”- was created by the Oregon Foundation for Reproductive Health and is used as a guide for contraceptive counseling to help evaluate a woman’s needs regarding contraceptive and/or preconception care (Bellanca & Stranger Hunter, 2016). Training and dissemination of the One Key Question with providers of homeless services could be one way to improve pregnancy risk screening and connect women with resources. This study evaluated homeless health care access points and shelters from across the geographic catchment area in an effort to provide a comprehensive picture of the need experienced by homeless women in Salt Lake Valley, Utah. Additionally, the data were obtained directly from participants and could contain reporting bias. The results may not be generalizable to a larger homeless population due to small sample size. Although we administered surveys at numerous locations, there are many women staying in motels, on friends’ couches, and in situations that were not captured in this data, leaving a gap in the need reflected here. We had 6 women indicate that they are renting an apartment on the survey. In some of the outreach settings women were asked to self-select for housing instability or homelessness and these individuals 10 opted to participate in the survey reflecting a broader population in need of services who would not meet traditional definitions for homelessness. An important first step in better meeting this need is increasing the capacity of homeless health care access points regarding contraceptive provision. Not only does this involve increasing the range of methods available in clinics, but also training providers to allow for same day LARC insertion and incorporating contraceptive counseling into standard intake procedures. This one on one conversation about contraception would also help to increase the amount of information women have about their options. Our data suggests a high prevalence of misinformation regarding LARC methods. By providing women with the opportunity to ask questions and learn about the range of methods available, it may help to dispel some of the LARC myths that were expressed in the survey responses and increase contraceptive utilization overall. Women experiencing homelessness in Salt Lake City, Utah do not have access to the range of contraceptive methods that they desire or the preconception care necessary to support healthy pregnancies. While we accessed many homeless resources in the valley, there are many women who may not be comfortable engaging with these services, thus interventions need to engage women living in poverty across non-traditional housing situations. Future research could explore the needs and barriers to accessing care for women who are experiencing housing instability and are not accessing homeless services. 11 Table 1: Sociodemographic, health and housing characteristics of women experiencing housing instability or homelessness Sociodemographic Characteristics Not Using Ideal Using Ideal n(%) Method Method Currently pCombination Combination Pregnant value Age 18-26 9(29) 7(32) 8(35) 0.904 27+ 22(71) 15(68) 15(65) Education No high school diploma/GED 12(41) 5(23) 6(26) 0.298 Has high/school diploma/GED 17(59) 17(77) 17(74) Relationship Status Not married/not cohabitating 15(50) 15(71) 8(35) 0.051 Married/Cohabitating 15(50) 6(29) 15(65) Insurance 0.001 Insurance 15(48) 14(64) 22(96) No insurance 16(52) 8(36) 1(4) Race/Ethnicity White, Non-Hispanic 16(52) 14(64) 16(70) 0.634 Hispanic, Non-White 7(23) 5(23) 3(13) Non-Hispanic, Other 8(26) 3(14) 4(17) Housing 0.001 Homeless shelter 9(29) 13(59) 2(9) Living with friends or family 4(13) 2(9) 9(39) Temporary/transitional housing 14(45) 5(23) 5(22) Other 4(13) 2(9) 7(30) Sex with a man in the past 3 months Yes 21(68) 13(59) 18(78) 0.485 No 9(29) 9(41) 5(22) Prefer not to answer 1(3) 0(0) 0(0) History of forced sex Yes 15(48) 11(50) 10(43) 0.739 No 11(35) 10(45) 10(43) Prefer not to answer 5(16) 1(5) 3(13) History of sex exchange Yes 10(32) 6(29) 7(30) 0.815 No 19(61) 15(71) 15(65) Prefer not to answer 2(6) 0(0) 1(4) Desires a pregnancy in the next year 12 Yes No Don't know Previous pregnancies Yes No Number of lives births Yes No Predictors for high risk pregnancy Tobacco dependence One or more chronic medical conditions Substance use disorder Mental health condition History of adverse pregnancy outcomes Jail or prison Yes No Total 7(23) 21(68) 3(10) 4(19) 17(81) 0(0) 6(27) 12(55) 4(18) 0.269 26(87) 4(13) 18(86) 3(14) 22(96) 1(4) 0.483 25(81) 6(19) 18(82) 4(18) 13(57) 10(43) 0.081 20(67) 11(55) 14(61) 0.705 20(65) 12(55) 10(43) 23(74) 22(71) 14(64) 15(68) 14(61) 11(50) 0.306 0.54 0.26 18(58) 10(45) 13(57) 0.634 10(34) 19(66) 31(100) 5(23) 17(77) 22(100) 8(35) 14(61) 23(100) 0.494 13 Table 2: Contraceptive method use and preferences in women experiencing homelessness or housing insecurity by ideal method use and pregnancy status * Past use Current use Ideal method n=76 (100%) n=46 (100%)** n=76 (100%) N=76 (100%) 7 (9) 14 (30) 10 (13) None 11 (14) 6 (13) 6 (8) Abstinence 15 (20) 0 4 (5) Emergency contraception 39 (51) 8 (17) 12 (16) Male condoms 3 (4) 0 0 Female condoms 27 (36) 5 (11) 7 (9) Withdrawal 10 (13) 1 (2) 10 (13) Rhythm method 5 (7) 1 (2) 0 Spermicide 1 (1) 0 0 Diaphragm 33 (43) 4 (9) 15 (20) Pills 9 (12) 2 (4) 8 (11) Ring 26 (34) 4 (9) 9 (12) DMPA 6 (8) 4 (9) 13 (17) Implant 18 (24) 7 (15) 14 (18) IUD 1 (1) 0 3 (4) Vasectomy 2 (3) 7 (15) 6 (8) Tubal ligation 12 (16) 0 2 (3) Pregnancy/ Breastfeeding *Participants may select more than one method in each column **Current use included only non-pregnant women (n=46) DMPA= depot medroxyprogesterone acetate; IUD= intrauterine device 14 Table 3: Current contraceptive method use compared to ideal method desire in non-pregnant women experiencing housing instability or homelessness Current Method Combination Ideal Method LEM MEM HEM HEM + HEM + Currently Combination None Total only only only LEM MEM Pregnant None 7 0 2 2 1 0 LEM only 2 2 1 1 0 0 MEM only 3 1 2 2 0 0 MEM + LEM 1 2 0 0 0 0 HEM only 3 0 7 7 2 1 HEM + LEM 0 0 0 0 1 0 HEM + MEM 1 3 0 0 1 0 HEM + LEM + MEM 1 0 0 0 0 0 Total 18 8 9 12 5 1 *LEM (less effective method) includes emergency contraceptives, withdrawal, condoms, and fertility awareness. MEM (moderately effective method) includes the pill, patch, ring, and DMPA. HEM (highly effective method) includes the IUD or implants. 15 4 3 3 2 4 4 1 15 9 14 7 17 5 6 2 23 3 74 Table 4: Barriers to ideal contraceptive method use and historical and preferred reproductive healthcare access points for women experiencing housing insecurity or homelessness by ideal method use and pregnancy status Not using ideal method n= 31 (100%) Using ideal method n=22 (100%) Currently pregnant n=23 (100%) P value 13 (50) 8 (32) 8 (31) 4 (17) 4 (18) 9 (43) 6 (30) 7 (33) 2 (10) 1 (5) 8 (40) 4 (20) 4 (20) 4 (21) 0 0.76 0.76 0.8 0.99 0.17 2 (6) 7 (23) 18 (58) 7 (23) 0 9 (41) 13 (59) 4 (18) 1 (4) 8 (35) 12 (52) 3 (13) 0.49 0.34 0.88 0.8 9 (29) 8 (26) 15 (48) 1 (3) 5 (23) 8 (36) 10 (45) 3 (14) 2 (9) 3 (13) 17 (74) 2 (9) 0.19 0.19 0.09 0.38 Reported barriers Method cost Ability to find a provider Unsure which facility to go to Partner does not like method Feeling of judgement by provider Past contraceptive healthcare Homeless Healthcare Clinic Planned Parenthood Clinic Physician's office Other Preferred future contraceptive healthcare Homeless Healthcare Clinic Planned Parenthood Clinic Physician's office Other 16 Table 5: Multivariate logistic regression models exploring factors associated with ideal contraceptive method use in women experiencing housing instability or homelessness. Unadjusted models Adjusted models Age >26 years ref ref no 1.14 [0.35, 3.73] 0.73 [0.06, 9.04] yes Cohabitating/married ref ref no 0.40 [0.12, 1.31] 0.27 [0.02, 3.12] yes High school education or higher ref ref no 2.40 [0.69, 8.30] 3.41 [0.49, 23.49] yes Racial/ethnic identity ref ref White, non-Hispanic 0.82 [0.21, 3.16] 0.65 [0.08, 5.42] Black, non-Hispanic 0.43 [0.09, 1.94] 0.14 [0.01, 2.26] Hispanic Housing status ref ref Stable housing/ shelter 0.35 [0.05, 2.31] 0.35 [0.02, 5.64] Staying with family/ friends * 0.25 [0.07, 0.93] 0.06* [0.01, 0.92] Transitional/ temporary housing 0.35 [0.05, 2.31] 0.0822 [0.01, 1.67] Other Sex with a man in past 3 months ref ref no 0.62 [0.19, 1.96] 0.48 [0.04, 6.54] yes History adverse pregnancy outcomes ref ref No 0.60 [0.20, 1.81] 0.21 [0.03, 1.81] Yes History Planned Parenthood clinic use ref ref No 2.37 [0.72, 7.85] 2.59 [0.35, 18.9] Yes History of incarceration ref ref No 0.56 [0.16, 1.97] 1.14 [0.09, 14.34] Yes Chronic medical condition (s) ref ref No 0.66 [0.22, 2.02] 1.14 [0.14, 9.44] Yes History substance abuse ref ref No 0.61 [0.19, 1.99] 0.35 [0.04, 3.09] Yes 17 Prior pregnancies ref 0.92 [0.18, 4.63] ref 0.26 [0.04, 1.75] ref No 0.54 [0.18, 1.64] Yes Exponentiated coefficients; 95% confidence intervals in brackets * p < 0.05 ref 0.47 [0.06, 3.77] None Yes Insurance coverage 18 Acknowledgements We would like to thank Lori M. Gawron, MD, MPH, Jessica Sanders, PhD, MPH, David Turok, MD, MPH, Morgan Millar, PhD, Kyl Myers, PhD, Matt Pierce and the Undergraduate Research Opportunities Program, the Lawrence T. & Janet T. Dee Foundation, and NICHD K12 HD085816 for their financial support in this research. 19 References Bellanca, H. K., Stranger Hunter, M. (2016). One Key Question. In Screening Women For Pregnancy Intentions As A Critical Reproductive Health Strategy: Oregon Foundation for Reproductive Health. Comprehensive Report on Homelessness State of Utah 2017. (2017). Retrieved from The Utah Department of Workforce Services, Housing and Community Development Division Curtis, K., Tepper, N., Jatlaoui, T., et al. (2016). U.S. Medical Eligibility Criteria for Contracpetive Use. MMWR Recomm Rep 2016;65(No.RR-3):1-104. Dasari M, Borrero S, Akers AY, et al. Barriers to Long-Acting Reversible Contraceptive Uptake Among Homeless Young Women. J Pediatr Adolesc Gynecol 2016;29:104-10. Effectiveness of Family Planning Methods. In: Division of Reproductive Health, National Center for Chronic Disease Prevention and Health Promotion. Gelberg MD and HSPH, L., Browner PhD and MPH, C. H., Lejano MD, E., & Arangua MPP, L. (2008). Access to Women's Health Care: A Qualitative Study of Barriers Perceived by Homeless Women. In (Vol. 40, pp. 87-100). Women & Health. Health Care for Homeless Women. (2013). The American College of Obstetricians and Gynecologists. Retrieved from www.acog.org website: 20 Henry, M., Watt, R., Rosenthal, L., Shivji, A. (2017). The 2017 Annual Homeless Assessment Report (AHAR) to Congress. Retrieved from U.S. Department of Housing and Urban Development Kennedy, S., Grewal, M., Roberts, E. M., Steinauer, J., & Dehlendorf, C. (2014). A Qualitative Study of Pregnancy Intention and the Use of Contraception among Homeless Women with Children. In (Vol. 25, pp. 757-770). Journal of Health Care for the Poor and Underserved. Kost, K. (2015). Unintended Pregnancy Rates at the State Level: Estimates for 2010 and Trends Since 2002. Kost, K. Lindberg, L. (2015). Pregnancy intentions, maternal behaviors, and infant health: investigating relationships with new measures and propensity score analysis. Demography. 2015;52(1):83-111. Lee, J. K., Parisi, S. M., Akers, A. Y., Borrero, S., Borrerro, S., & Schwarz, E. B. (2011). The impact of contraceptive counseling in primary care on contraceptive use. J Gen Intern Med, 26(7), 731-736. doi:10.1007/s11606-011-1647-3 Preterm (Premature) Labor and Birth. (2016). ACOG. Richards, R., Merrill, R. M., & Baksh, L. (2011). Health behaviors and infant health outcomes in homeless pregnant women in the United States. In (Vol. 128, pp. 438-446). Pediatrics. Silver, G., & Panares, R. (2000). The Health of Homeless Women: Information for State Maternal and Child Health Programs. Retrieved from www.jhsph.edu. 21 Appendix A Reproductive Care Survey We would like you to take a few minutes to complete the following survey about your birth control and pregnancy desires and history. First, we’d like to ask a few questions about your health. 1) Have you ever been told by a doctor that you have any of the following? Medical condition Yes No High blood pressure Blood clot Heart disease Breast cancer Mental health issues (For example: depression or PTSD) Addiction to alcohol Addiction to drugs Stroke Liver disease (For example: liver tumors, liver cancer, or cirrhosis) HIV or AIDS Diabetes Migraines Lupus Seizure disorder (For example: epilepsy) None of the above Other conditions (please list them here): Don’t know 2) Have you ever had a hysterectomy (had your uterus removed)? o Yes  STOP HERE. Thank you for your help. o No 3) Are you now going through, or have you ever been through menopause (the point when a woman stops having monthly periods- usually near middle age)? o Yes  STOP HERE. Thank you for your help. o No 4) Do you currently smoke cigarettes or use tobacco? o Yes o No 22 5) Have you ever been pregnant (either now or in the past)? o Yes o No  SKIP TO QUESTION 9 AT THE BOTTOM OF THE NEXT PAGE. 6) Are you currently pregnant? o Yes o No o I don’t know 7) How many times have you been pregnant? __________ 8) In the table below, please select the result or outcome of each pregnancy you have had. Pregnancy Result or # outcome: Ectopic pregnancy Baby Baby (pregnancy delivered delivered in your on time early Miscarriage Abortion Stillbirth tubes) 1 2 3 4 5 6 7 Now we would like to ask you a few questions about your reproductive history. 23 9) What birth control method or methods do you CURRENTLY USE, if any? Please circle all that apply using the pictures below. No Method Male condom Injectable (DepoProvera) Ring or NuvaRing Partner’s sterilization (vasectomy) Female condom Fertility Implant that is awareness, placed in your rhythm, or arm safe period (Implanon or by calendar Nexplanon) Emergency contraceptive, Pregnant or Plan B, or breastfeeding morning after pill Tubes tied (tubal sterilization) pulling out (withdrawal) Sponge, diaphragm, or cervical cap Intrauterine device (IUD) such as Mirena, Skyla, or Paragard Spermicides Mostly have Abstinent or sex with do not have women sex 24 Birth control pills 10) What birth control method or methods have you USED IN THE PAST, if any? Please circle all that apply using the pictures below. No Method Male condom Injectable (DepoProvera) Ring or NuvaRing Partner’s sterilization (vasectomy) Female condom Fertility Implant that is awareness, placed in your rhythm, or arm safe period (Implanon or by calendar Nexplanon) Emergency contraceptive, Pregnant or Plan B, or breastfeeding morning after pill Tubes tied (tubal sterilization) pulling out (withdrawal) Sponge, diaphragm, or cervical cap Intrauterine device (IUD) such as Mirena, Skyla, or Paragard Spermicides Mostly have Abstinent or sex with do not have women sex 25 Birth control pills 11) If you could choose whatever birth control you wanted, which method or methods would be IDEAL for you, if any? Please circle all that apply using the pictures below. No Method Male condom Injectable (DepoProvera) Ring or NuvaRing Partner’s sterilization (vasectomy) Female condom Fertility Implant that is awareness, placed in your rhythm, or arm safe period (Implanon or by calendar Nexplanon) Emergency contraceptive, Pregnant or Plan B, or breastfeeding morning after pill Tubes tied (tubal sterilization) pulling out (withdrawal) Sponge, diaphragm, or cervical cap Intrauterine device (IUD) such as Mirena, Skyla, or Paragard Spermicides Mostly have Abstinent or sex with do not have women sex Birth control pills 12) Would you like to become pregnant in the next year? o No, I don’t want any future pregnancies o No, I don’t want to become pregnant in the next year, but maybe in the future o I’m not sure o I’m okay either way o Yes 13) How would you feel if you became pregnant in the next year? Circle a face on the scale below. Saddest I could possibly feel Not happy or sad 26 Happiest I could possibly feel 14) Have you had vaginal sex with a man in the past three months? o Yes o No o I prefer not to answer 15) Do any of the following make it hard for you to get the kind of birth control you want? Yes No Don’t know Cost Problems finding providers Unsure of what facilities to go to Partner does not like it Feeling of judgment from provider Other, please describe: 16) Would you be interested in getting an IUD, if it was offered to you at no cost? o Yes o No 17) Would you be interested in getting a birth control implant, if it was offered to you at no cost? o Yes o No 18) Would you have any concerns about using the IUD or implant, if it were available? o Yes o No  SKIP TO QUESTION 20 ON THE NEXT PAGE. 19) If yes, what are your concerns? ________________________________________________________________________ _ ________________________________________________________________________ _ 20) Where have you gotten birth control in the past? o 4th street clinic o Planned Parenthood o Doctor’s Office o Other, please specify:____________________________________________________ 21) Ideally, where would you like to get birth control from? o 4th street o Planned Parenthood o Doctor’s Office o Other, please specify:____________________________________________________ 27 22) Have you ever been forced to have sex with a man? o Yes o No o Prefer not to answer 23) Have you ever had sex in exchange for money, food, drugs, or a place to stay? o Yes o No o Prefer not to answer For the final section of the survey, we would like to ask a few demographic questions. 24) What best describes your current housing situation? o Homeless shelter o Living with friends or family (“doubling up”) o Temporary/transitional housing o Rent/own a house or apartment o Other, please specify: __________________ 25) Which of the following best describes your sexual orientation? o Completely heterosexual (straight) o Mostly straight o Bisexual (attracted to both men and women) o Mostly gay/lesbian o Completely homosexual (gay/lesbian) o Not sexually attracted to males or females o Other, please specify: __________________ o I don’t know o I prefer not to answer 26) What is your health insurance status? Please select all that apply: o No health insurance o Medicaid o Insurance through your job or that you buy on your own o Student health insurance o Disability or Medicare o Military or VA Coverage o Parents’ insurance o Something else, please specify:_________________ 28 27) Which of the following describes your race or ethnic background? Mark all that apply. o White o Black or African American o Asian o Pacific Islander/Native Hawaiian o American Indian/Native Alaskan o Hispanic/Latino o Other, please specify: __________________ o Prefer not to answer 28) Which of the following best describes you? o Married o Not married, but living together or in a committed relationship o Actively dating, but NOT in a relationship o Divorced/separated o Single, not actively dating o Widowed o Other, please specify: __________________ o I don’t know 29) What is your age? __________years 30) What is your current religion, if any? o Not religious o Christian (Protestant, Evangelical, Mainline, etc.) o Catholic o Jewish o Muslim o Mormon o Other, please specify: __________________ o I don’t know o I prefer not to answer 31) What is the highest grade level that you have completed? o Grade school only (Kindergarten – 6th grade) o 7th or 8th grade o Some high school, no diploma o High school diploma or GED o Trade or technical school o Some college, no degree o Four year college degree o Master’s degree or other professional degree 29 32) Have you been in jail or prison in the past year? o Yes o No o Prefer not to answer This is the end of the form. Thank you for taking the time to complete the survey! 30