Contraceptive Decision Making Among Latina Immigrants: Developing Theory-Based Survey Items

Abstract

Background:

Contraception is important for reproductive autonomy, yet many Latinas do not use contraception consistently despite research reporting a desire to do so. Factors varying in priority and value come into play during contraceptive decision making. When measuring these, relevant survey items may vary by populations.

Aim:

This study focused on developing an ethnically responsive, patient-centered, content-valid survey for measuring factors that influence contraceptive decision making among immigrant Latinas.

Method:

Nonpregnant self-identified Latinas ages 15 to 24 years in Baltimore, MD, were recruited from a family planning facility. Using the theory of planned behavior as a theoretical framework and prior formative research, initial survey items were drafted (Step 1). Content validation and cognitive interviewing procedures (Step 2 and Step 3) were used to develop final items.

Results:

Final items (27) were content-validated by the target population; items reflect important factors and relevant contexts affecting contraceptive decision making among Latinas in Baltimore.

Discussion:

These theory-based items provide an important contribution to the literature because they measure and explore factors related to contraceptive decision making in an understudied population. Providers might consider these factors during counseling to build patient-centered communication. These items might serve to measure responses to theory of planned behavior–based interventions designed to improve the contraceptive counseling of Latinas.

Introduction

For women especially, the ability to control their own reproductive health is associated with personal and community well-being (Finer & Sonfield, 2013). The adverse outcomes associated with a lack of control and choice over childbearing are more severe for low-income racial and ethnic minority groups, particularly African American women and Latinas (Finer & Zolna, 2016; Jackson, Wang, & Morse, 2017; Kost, Henshaw, & Carlin, 2010; Mosher & Jones, 2010). Latino/a/x immigrants are the fastest growing ethnic group in the United States (Martin et al., 2012; U.S. Census Bureau, 2008) and in Baltimore, Maryland (U.S. Census Bureau, 2008). In Baltimore, more than 50% of the Latinx population is low income and have less than a high school education (Office of Epidemiology and Planning at the Baltimore City Health Department, 2011), placing them and their families at high risk for inadequate receipt of health care services.

Contraception is vital for control over childbearing, yet in the United States, many Latinas do not use contraception consistently (Garcés-Palacio, Altarac, & Scarinci, 2008; Green, Oman, Vesely, Cheney, & Carroll, 2017; Masinter, Feinglass, & Simon, 2013; Sangi-Haghpeykar, Ali, Posner, & Poindexter, 2006) despite research reporting a desire to use contraception (Caal, Guzman, Berger, Ramos, & Golub, 2013; Carvajal, Gioia, Rivera, Brown, & Barnet, 2017; Galloway, Duffy, Dixon, & Fuller, 2017; Oakley, Harvey, & López-Cevallos, 2018). Inconsistent use among Latinas has been previously attributed to lack of knowledge about options (Craig, Dehlendorf, Borrero, Harper, & Rocca, 2014), male partner dominance (Schwartz, Brindis, Ralph, & Biggs, 2011; Yee & Simon, 2010), a desire for large families (Aarons & Jenkins, 2002; Craig et al., 2014), and religious objection (Sangi-Haghpeykar et al., 2006). Reasons for inconsistent contraception use have largely been studied among U.S.-born Latinas and some of the literature may be outdated. Furthermore, non-U.S.-born, immigrant Latinas may have differing perceptions about contraception and reasons for use/nonuse compared with U.S.-born Latinas because of factors related to the immigration experience (Tapales, Douglas-Hall, & Whitehead, 2018). Immigration processes and experiences of perceived discrimination can adversely affect health, health care, and well-being among Latinx immigrants (Abraido-Lanza, Cespedes, Daya, Florez, & White, 2011; Ayón & Becerra, 2013; Becerra, Androff, Messing, Castillo, & Cimino, 2015; Martinez Tyson, Arriola, & Corvin, 2016; Tapales et al., 2018). Immigrant Latinas are less able to access reproductive health care services than their U.S.-born counterparts (Hasstedt, Desai, & Ansari-Thomas, 2018), and their experiences such as fleeing unstable governments, gang violence, and/or poverty may influence and inform personal goals and childbearing plans.

When considering the role of health care providers in contraceptive care, research shows that patient–provider communication affects contraceptive use and adherence (Isaacs & Creinin, 2003; Stevenson, Cox, Britten, & Dundar, 2004). For example, low patient-reported communication with providers has been associated with lower likelihood of continued or consistent use of contraception (Isaacs & Creinin, 2003; Peipert et al., 2011). A qualitative study in Baltimore demonstrated that Latina participants overwhelmingly expressed the importance of good communication and trusting relationships with health care providers vis-è-vis contraceptive counseling (Carvajal et al., 2017). Congruent with trusting and communicative relationships is the notion of patient-centered communication—defined by the Institute of Medicine (now the National Academy of Medicine) as care that is guided by patient values and is responsive to patient preferences and needs (Corrigan, Donaldson, Kohn, Maguire, & Pike, 2001). Patient-centered communication is associated with improved use of preventive health services (Doyle, Lennox, & Bell, 2013; Rathert, Wyrwich, & Boren, 2013; Stevenson et al., 2004) and with contraceptive satisfaction and continuation (Dehlendorf et al., 2016; Isaacs & Creinin, 2003; Peipert et al., 2011).

Social and personal factors informed by life experiences matter when women make decisions about contraception. This is also true of immigrant Latinas given their experiences of migration—often in search of educational, career, and life opportunities that are more readily available in the U.S. Within the context of patient-centered care, the theory of planned behavior (TPB; Glanz, Rimer, & Viswanath, 2008) can provide a framework to understand the interplay among factors that influence contraceptive decision making. The TPB posits that decisions/intentions to carryout health behaviors (e.g., use contraceptives) are influenced by a person’s behavioral beliefs and attitudes, normative beliefs and beliefs about norms, as well as a perception of control over the relevant behavior. Intentions are in turn, predictive of health behaviors (Ajzen, 2002; Ajzen & Fishbein, 1972; Glanz et al., 2008). Studies suggest the TPB accounts for approximately one third of the variance in intention and behavior (Armitage & Conner, 2001).

Distinct racial/ethnic groups may differ in their experiences, beliefs, and attitudes (perceived advantages and disadvantages), normative beliefs and subjective norms (recognized individuals/groups who support/oppose contraceptive use and whether those opinions influence participants), and control beliefs and perceived behavioral control about contraception (factors that facilitate or impede use; Ajzen, 2002; Francis et al., 2004; Glanz et al., 2008). Therefore, a single set of survey items that assess beliefs/attitudes, norms, and perceived control may not work across diverse populations. When using the TPB as a guide for understanding contraceptive decision making among immigrant Latinas, it is important to first identify population-specific, patient-oriented factors germane to health-related decisions.

The essential first step is therefore to create an instrument that accurately identifies the factors that can influence the relevant behavior of the population of interest. While standard protocols for health decision survey items using the TPB as a theoretical framework exist (Francis et al., 2004), the specific content, structure, and wording of items must be assessed and understood by our population of interest—immigrant Latinas. Therefore, the purpose of this study was to develop ethnically responsive, content-valid, theory-based question items for measuring the personal and health care factors that influence contraceptive decision making among immigrant Latinas—an understudied, vulnerable group. In future studies these items may be used as measures in TPB-based interventions.

Method

Study Setting, Participants, and Recruitment Procedures

Participants were recruited from the city health department family planning clinic (see Figure 1 for enrollment flow diagrams) that serves a predominantly low-income, uninsured population of minority patients who are charged a sliding scale fee. Eligibility criteria included self-identified Latinas (potential participants were asked: “Are you Latina?” followed by a list of options for countries of birth if the participant answered “yes”), ages 15 to 24 years who were not pregnant or intending to become pregnant at the time of survey administration. This age range was chosen because at the time of study recruitment, adolescents ages 15 to 17 and women ages 18 to 24 years old represented the groups with the highest rates of unintended pregnancy and poorest outcomes (Finer & Zolna, 2016; Masinter et al., 2013). Potential participants were approached in the facility waiting room by a bilingual research assistant. Patient consent (>18 years old) or parental consent and patient assent (<18 years old) was obtained. Each session lasted approximately 30 to 60 minutes; survey answers were manually recorded by the research assistant while survey administration was also audio-recorded. Each participant received a gift card as compensation for her time. This study was approved by the appropriate institutional review boards.

Figure 1.

Enrollment flow diagram, content validation, and cognitive interviewing.

We developed the survey tool in three steps. Briefly, and specified in more detail later, in Step 1, we drafted initial survey items. For Step 2, we conducted content validation (CV) procedures with the survey items. Finally, in Step 3, we conducted cognitive interviewing with revised items. The three steps of our survey development and validation are described below.

Step 1: Initial Item Development

We used data from a prior qualitative study of immigrant Latinas in Baltimore (Carvajal et al., 2017) to identify the TPB domains that influence contraceptive decision making and use (detailed results are reported elsewhere by Carvajal et al., 2017). The larger, meta-domain was identified as contraceptive motivations. Three TPB subdomains were identified: (1) beliefs and attitudes about contraceptive use (advantages and disadvantages), (2) normative beliefs and subjective norms (individuals/groups who support/oppose contraceptive use and whether their opinions influence use/nonuse), and (3) control beliefs and perceived behavioral control about contraception (factors that facilitate/impede use including how health care providers might influence decision making). Initial item content was developed for the survey based on these subdomains yielding a set of 32 questions drafted in both English and Spanish using Constructing Questionnaires Based on Theory of Planned Behavior (Francis et al., 2004) to determine initial structure and wording of questions. Within the subdomains, important concepts related to contraceptive decision making were the following: (1) advantages and disadvantages of contraception use: pregnancy prevention (advantage), ability to attain educational/career goals (advantage), side effects (disadvantage), and concern for future infertility (disadvantage); (2) important individuals/groups who support/oppose contraceptive use: family members, partners, friends, schools/teachers, and church/religion; and (3) factors that facilitate/impede use: positive patient–provider communication and trust (facilitator), shared decision making with a provider (facilitator), provider accessibility (facilitator), patient–provider language and gender concordance (facilitators), contraceptive availability and cost (barriers), and provider judgement/racial-ethnic discrimination (barrier). Because there was no certainty about the relevance of each of these concepts with respect to contraception use in our population of interest, we used CV (detailed below) to determine relevancy (Aday & Cornelius, 2006).

Step 2: Content Validation Process and Assessment

CV refers to how well the empirical measures represent the theoretical domains they presumably reflect (Aday & Cornelius, 2006). The initial draft was examined for content validity with a group of Latinas (20 participants) ranging in age from 15 to 24 years from July 2016 through November 2016. It is generally suggested that when using CV, there to five participants from each group of interest should be consulted (Aday & Cornelius, 2006; Netemeyer, Bearden, & Sharma, 2003). In this case, the authors aimed to collect data from five participants within each of four groups (in both English and Spanish for age groups 15–19 and 20–24 years). CV relies on experts (i.e., Latina participants) to make judgments about whether the a priori theoretical items generated represent the concepts that they are intended to reflect and whether or not the concepts are relevant to the reference action (Aday & Cornelius, 2006). CV has been used to help validate surveys in obstetrics (Ruhl, Scheich, Onokpise, & Bingham, 2015), family medicine (Phillips, Prunuske, Fitzpatrick, & Mavis, 2018), and rheumatology (Oh, Han, Kim, & Seo, 2018). In this study, the CV approach was formalized by asking participants to rate specific items on whether or not they are relevant (scale: not all important to extremely important) to the decision to use contraception. Special attention was paid to items that represented themes that were consistently identified as not at all important or slightly important. Items that were identified as having little relevance to contraceptive decision making or noted by participants to be redundant were eliminated. Subsequently, we assessed the second set of items using cognitive interviewing (Presser et al., 2004) to determine whether they were clearly written, presented, and understood by study participants.

Step 3: Cognitive Interviewing Process and Assessment

Cognitive Interviewing (CI) tests psychometric properties of survey items by examining the cognitive processes that a participant experiences when answering questions. Participants are asked to deconstruct and reconstruct the meaning of questions (Presser et al., 2004) to identify potential problems with survey items and inform revisions (Howlett, McKinstry, & Lannin, 2017). The process of CI has also been used in cultural and language-based adaptations of existing surveys and to determine the suitability of items for those experiencing the survey (Beck et al., 2017; Lee, Lee, & Aranda, 2018). For the CI process in this study, we used the second iteration of items and recruited 12 new participants ages 16 to 24 years from November 2016 through February 2017. Conventional statistical sampling methods are not normally used in CI, and the literature does not provide established guidelines for determining an appropriate sample size other than collecting data from participants across a range of ages and other important characteristics as is applicable for the study (Beatty & Willis, 2007; Blair, Johnny, & Conrad, 2011). However, it is recommended that sample sizes commonly range between 5 and 15 interviews (Beatty & Willis, 2007). CI attempts to reduce the cognitive burden on respondents (i.e., to decrease the amount of information retrieved from memory) through the use of verbal probes and can reveal inconsistencies between patients’ and researchers’ understanding of item content (Lavoie Smith et al., 2017). During each interview, a guide with standard verbal probes was used to prompt participant questions, concerns, and proposed solutions for each item. Examples of verbal probes that were presented to participants after an item was read are the following:

• Can you repeat the question in your own words? (Tests how well the subject comprehends the question.)

• I’d like you to think about your answer to the question. Can you tell me what you had to think about in order to answer the question? How easy or hard was that to do? What made it easy or hard?

• Is there anything about how we asked this question that is confusing? If so, what?

• How can we clarify this question?

• Are there any specific words in this question that you did not understand?

• Does the response scale make sense or is it presented in a way that is confusing? How would you change it so that it easier to understand?

Analysis and Interpretation

Survey administration for both CV and CI was audio-recorded. Researchers reviewed written data and field notes and then simultaneously listened to transcripts to decide how changes to the structure and wording of questions would proceed.

Content Validation Analysis

Data were independently analyzed by two researchers. Items were considered relevant if they were labeled as either quite important or extremely important by more than 50% of participants. Items that were labeled as slightly important or not important at all by 75% or more of participants were considered not relevant. An example of a “very important” survey item is the following: “When it comes to making decisions about birth control, how important is it to young women like you to trust their doctor?” Participants were also asked to comment on the redundancy of items.

Cognitive Interviewing Analysis

Initial analysis included one researcher listening to recordings while simultaneously reviewing the written data. Field notes were considered by the research team to ensure that participants’ thoughts and understanding of the questions were clearly recorded and recognized. Participants’ difficulties in understanding the question as intended (along with proposed solutions) were noted and recorded in an electronic database. Results were summarized by one researcher and then discussed with the rest of the research team; researchers then agreed upon the necessary amendments. After every two to three interviews, items were rewritten to improve the clarity of the questions based on participants’ responses. The final 24-item survey was created after all 12 interviews were completed. One researcher (DNC) had the main responsibility for the analysis, but the analysis took place in close collaboration with the research team.

Results

Content Validation Results

CV included 20 participants ranging in age from 15 to 24 years with a median age of 19. Twelve participants were under age 20; and the remainder was between the ages of 20 and 24. Ninety percent (18 of 20) were born outside of the United States, representing Mexico (5), Honduras (5), El Salvador (5), Ecuador (2), and the Dominican Republic (1). The two U.S.-born participants were included in the sample even though they would not usually be considered immigrants because they reported identifying more as immigrants than as U.S.-born citizens; while both were born in the United States, each has spent a significant amount of time in their parents’ native countries. For participants born outside of the United States, the average number of years in the United States was 4.5 years (see Table 1).

Table 1.

Participant Demographics, Content Validation.

	Country of birth
Age range (years)	Mexico	Honduras	El Salvador	Ecuador	Dominican Republic	United States
Median age: 19
15–19	3	4	2		1	2
20–24	2	1	3	2
Total (N = 20)	5	5	5	2	1	2

Thirty-two initial items were evaluated for concept relevancy/importance to contraceptive decision making. Fifteen participants (>75%) rated the item: “When it comes to making decisions about birth control, how important is it to young women like you to think about the opinions of their church/religion?” as slightly important or not important at all. After review by the research team, the item was eliminated (Table 2). Six items (Table 2) were noted by most participants to be repetitive or redundant; these repetitive items were either eliminated or combined with other items. One item asking about the importance of teachers and school was only endorsed by adolescents ages 15 to 18 years as quite important or extremely important. Another item asking about the importance of confidentiality was only endorsed by adolescents ages 15 to 17 years as quite important or extremely important. These items were retained and used for the Step 3 process only with participants in those respective age groups. CV yielded 26 items.

Table 2.

Content Validation Results, Survey Item Changes.

Survey Item	Eliminated	Noted to be redundant; Combined with another item	Only important for adolescents/minors
“When it comes to making decisions about birth control, how important is it to young women like you to think about the opinions of their church/religion?”	✓
“When it comes to making decisions about birth control, how important is the issue of completing your education to young women like you?”		✓ (these two items were combined into one)
“When it comes to making decisions about birth control, how important is the issue of achieving your career goals to young women like you?”
“When it comes to making decisions about birth control, how important is it to young women like you to plan the best time to have a baby?”		✓ (this item was noted as redundant/similar to other items; it was eliminated)
“When it comes to making decisions about birth control, how important is it to young women like you to have a good relationship with their doctor?”		✓ (this item was noted as redundant/similar to other items; it was eliminated)
“When it comes to making decisions about birth control, how important is it to young women like you to have their doctors tell them about the advantages and disadvantages of using it?”		✓ (this item was noted as redundant/similar to other items; it was eliminated)
“When it comes to making decisions about birth control, how important is it to young women like you to have their doctors give them as much information as possible about it?”		✓ (this item was noted as redundant/similar to other items; it was eliminated)
“When it comes to making decisions about birth control, how important is it to young women like you to think about the opinions of their teachers/school?”			✓ (ages 15–18)
“When it comes to making decisions about birth control, how important is it to young women like you to have privacy and confidentiality (from your parents) with your doctor?”			✓ (ages 15–17)

Cognitive Interviewing Results

CI included 12 new participants ages 15 to 23 years (median age 18.5 years) recruited to examine their comprehension of the 26-item survey. Eighty-three percent of participants were born outside of the United States, representing countries in Mexico and Central and South America (average number of years in the United States was 6). Two participants were born in the United States; one of these was born in Puerto Rico. While the participant form Puerto Rico is not actually an immigrant as Puerto Rico is in fact, a commonwealth of the United States, given her birth outside of the 50 states, the authors considered her experiences to be most congruent with that of an immigrant. For this reason, she was included among our reported sample. The other participant who was born in the United States lived outside of the country for several years and therefore personally identified as an immigrant (see Table 3).

Table 3.

Participant Demographics, Cognitive Interviewing.

	Country of birth
Age range (years)	México	Honduras	El Salvador	Ecuador	United States including Puerto Rico
Median age: 18.5
15–19	2	3	1		2
20–23	1	1	1	1
Total (N = 12)	3	4	2	1	2

Twenty-six question items were evaluated (Table 4). Of the 26 items evaluated in during the CI process, none were eliminated, and in fact, a question was added for clarification. Twenty items had minor changes in wording and/or structure in response to study participant input. Response scales for 10 items were reworded; however, the same scale change was applied for 9 of the 10 items. The items whose wording/structure was changed were reported as unclear, confusing, or difficult to understand by a majority of participants; the items were rewritten according to participants’ suggestions and recommendations.

Table 4.

Cognitive Interviewing Results, Final Survey Items.

	N	Final relevant contexts/important factors affecting contraceptive decision making, based on the TPB
Initial survey items	26	Behavioral Beliefs and Attitudes: Pregnancy prevention to achieve educational/career goals Concern for potential side effects and future fertility
Eliminated	0
Added	1
Response scale rewording	10	Normative Beliefs and Subjective Norms: Family thoughts and advice Partner preferences School/teacher advice (adolescents 15–18 only) Friends’ advice
Minor item rewording or restructuring	20
Final survey items	27
		Control Beliefs and Perceived Behavioral Control: Perceived patient-provider communication; trust Personal contraceptive preferences Knowledge of method side effects, effectiveness Access to providers Insurance/financial means to afford methods Discussion of options in her preferred language Access to female providers Confidentiality (adolescents 15–17) Nonjudgmental/nondiscriminatory interactions with providers

Note. TPB = theory of planned behavior.

Relevant contexts and important factors endorsed by participants as being influential in contraceptive decision making are listed in Table 4. Each of these domains is represented in at least one item of the final question set, and in tandem with the TPB, include the following: (1) beliefs and attitudes about the perceived advantages/disadvantages of contraceptive use including pregnancy prevention in order to achieve educational/career goals (advantage), side effects (disadvantage), and concern for future infertility (disadvantage); (2) normative beliefs and subjective norms of important individuals/groups that support/oppose contraceptive use including family members, partners, teachers, and friends; and (3) factors that facilitate/impede use (control beliefs and perceived behavior control) including patient–provider communication and trust, provider acknowledgement of a patient’s personal preferences (consistent with shared decision making), a desire to have as much information as possible about the method of interest, confidentiality, and nonjudgmental/nondiscriminatory interactions with providers, among others.

Discussion

This study developed, piloted, and content-validated quantitative measures of contraception decision making among Baltimore Latinas. Initial, in-depth qualitative data analysis resulted in the creation of 32 preliminary question items intended to evaluate the important factors influencing contraceptive decision-making among Latinas, including patient-reported relationship factors with providers such as communication and trust. Final items totaled 27 and resulted from a rigorous process of CV followed by cognitive interviewing. Importantly, participants affirmed that religion is not an important factor associated with contraceptive decision making—which is consistent with other recent research (Green et al., 2017) but also conflicts with common assumptions and beliefs about Latinx populations (Arana, 2013; Sánchez, 2014). The other contextual elements presented in the initial survey items were, however, noted to be relevant and plausibly related to decision making.

Participants conveyed support for getting as much information as possible about contraceptives (including side effects, effectiveness) from their providers during decision making. Participants endorsed that influence and advice of important persons in their lives such as family, partners, and friends should also be considered. Importantly, participants acknowledged that nonjudgmental and nondiscriminatory interactions affect contraceptive counseling experiences and therefore decisions surrounding use. The notion of perceived discrimination or the perception of being counseled to use certain methods based on one’s race/ethnicity is consistent with recent studies reporting that women of color often feel discrimination related to contraceptive care (Becker & Tsui, 2008; Carvajal et al., 2017; Yee & Simon, 2011). Our results also support research in which participants consistently endorse the importance of having their personal preferences known, and of positive communication and trust with providers (Carvajal et al., 2017; Dehlendorf, Krajewski, & Borrero, 2014; Dehlendorf, Levy, Kelley, Grumbach, & Steinauer, 2013). Specifically, this study’s findings also reflect the importance of a patient-centered approach to contraceptive counseling and provision (Dehlendorf, Grumbach, Schmittdiel, & Steinauer, 2017).

While some of the factors confirmed as important by participants when making decisions to use contraception may be unique or related to their identities as Latina immigrants, others are not. For example, the specific desire to avoid racial/ethnically based judgement and discrimination by providers reflects the experiences of other women of color and immigrants in the United States—especially with respect to reproductive health care (Becker & Tsui, 2008; Downing, LaVeist, & Bullock, 2007; Novak et al., 2018; Yee & Simon, 2011). This important point about this vulnerable population should be noted and considered by providers during contraception counseling. Additionally, personal reasons for migration, subsequent immigration experiences, and histories of unjust reproductive practices committed against immigrants in the United States (Novak et al., 2018; Silliman, Fried, Ross, & Gutierrez, 2016) are important to note also since they are likely to inform the circumstances around control of childbearing. Just a few examples of the reproductive injustices committed against members of the Latinx population include (1) the sterilization of Mexican-origin women without their knowledge at a California hospital from 1969 to 1973; (2) contraceptive research using Mexican-origin women as subjects without their consent; and (3) a 1940s American sterilization campaign for population control in Puerto Rico during which women were consented for procedures during labor—the result of which was that by 1965, 35% of Puerto Rican women were sterilized and two thirds of them were in their 20s (Silliman et al., 2016). These previous occurrences have in the past and may continue to sew a deep mistrust of the medical system and specifically, of contraceptives.

Also noteworthy is that some factors previously attributed in the literature as likely affecting the contraceptive decisions of Latinas—religious ideology, for example, should not be widely generalized to all Latinas. Moreover, other important factors related to contraceptive decision making include trust, communication, and patient-centered care. These are not unique to this population as non-Latina women have reported desire to have trusting and communicative relationships with their providers regarding contraceptive care (Dehlendorf et al., 2013). Research suggests members of this study’s vulnerable population and other Latina/o/x populations do not always feel they receive such care when it comes to reproductive health (Becker & Tsui, 2008; Carvajal et al., 2017; Geronimus, 2003; Yee & Simon, 2011).

Limitations and Strengths

The generalizability of this study is rather limited in its scope, including in its generalizability to other Latinx populations, given the existing diversity in the United States and the many different factors that influence contraceptive decision making among immigrant Latinas. This survey’s utility may be sampledependent, and other Latinas may identify additional factors that influence contraceptive decision making. It is important to note that Latinx people whether born in the United States or not, hail from a variety of countries in Latin America, Europe, and the Caribbean and are characterized by a wide range of sociodemographic variables and migration experiences. Yet, there may also be certain factors that immigrating Latinx populations do have in common based on a shared immigration experience. In the future, studying a variety of Latinx populations from across Latin America, may add replicable data or further insight to our findings.

Another potential limitation is that specific data regarding income, education, or other measures of socioeconomic status were not collected. However, at the time of data collection, a conscious decision was made not to collect this data out of concern that such questions might have been perceived by participants as intrusive—potentially discouraging Latinx immigrants from participating during uncertain political times and unclear immigration policies and ramifications. Still, as previously noted, participants were recruited from a local city health department facility that typically serves a predominantly low-income, uninsured population. Patients are charged for the services they receive based on a sliding scale fee—which itself may be considered a proxy for low-socioeconomic status. Despite these limitations, this study provides an important contribution to the literature in that it has resulted in survey items that measure important factors affecting contraceptive decision-making for immigrant Latinas—to date, an understudied population. Items were vetted by the target population and have been structured to maximize participant understanding of the concepts intended for measurement. The items have resulted from in-depth qualitative research and well-accepted methods of validation; therefore, the study has succeeded in creating items of measurement for contraceptive decision making that capture parameters of the TPB.

Implications for Theory Policy and/or Practice

Patient-centered care is a key component of contraception counseling and provision (Dehlendorf, Henderson, Vittinghoff, Steinauer, & Hessler, 2016), because it focuses on the personal attributes of each patient. In 2014, the Centers for Disease Control and Prevention and the U.S. Office of Population Affairs published recommendations on quality family planning services that explicitly cited patient-centered care as a vital component of quality care (Gavin et al., 2014). Within this vein, the authors believe the items resulting from this study accurately reflect factors that are specifically germane to the contraceptive decision making of the participants. These items have the potential to measure responses to a TPB-based intervention for the population of interest—precisely because the TPB allows us to focus on the perspectives of patients regarding which factors influence their decision making while also recognizing the associated complexity of these decisions.

To adequately test the associations of important decision-making factors with outcomes such as patient satisfaction with counseling, chosen method, and continuation of methods, valid measures are needed. Survey items will be used in a subsequent study to assess how providers can facilitate contraceptive decision making and improve counseling, method satisfaction, and method continuation through patient-centered communication and trust building. These measures will help test the impact of appropriate interventions to improve the contraceptive care of Latinas.