Decision making about Pap test use among Korean immigrant women: A qualitative study

Abstract Background Understanding how individuals make decisions about Pap tests concerning their personal values helps health‐care providers offer tailored approaches to guide patients’ decision making. Yet research has largely ignored decision making about Pap tests among immigrant women who experience increased risk of cervical cancer. Objective To explore decision making about Pap tests among Korean immigrant women. Methods We conducted a qualitative descriptive study using 32 semi‐structured, in‐depth interviews with Korean immigrant women residing in a north‐eastern metropolitan area. Data were audio‐recorded, transcribed verbatim and analysed using inductive coding. Results Although most women with positive decisions made their own decisions, some women deferred to their providers, and others made decisions in collaboration with their providers and significant others. While women making positive decisions tended to consider both barriers to and facilitators of having Pap tests, women making negative decisions predominantly discussed the barriers to having Pap tests, such as modesty and differences between the South Korean and US health‐care systems. The women's reflections on their decisions differed regarding their Pap test decisions. Conclusions Women's desired roles in the decision‐making process and reflection on their decision outcome appeared to vary, although most participants with positive decisions made their own decisions and were satisfied with their decisions. Future research should conduct longitudinal, quantitative studies to test our findings regarding decision‐making processes and outcomes about Pap tests. Implications The findings should be incorporated into cervical cancer screening practices to fulfil the unmet needs of immigrant women in patient‐provider communication and to facilitate women's decision making about Pap tests.

which more than one possible option exist, such as prostate cancer screening. 1,3,4 Studies of shared decision-making in breast, colorectal and prostate cancer screening revealed that patient/provider communications are often ineffective. [5][6][7][8] Elston Lafata et al. 5 found that health-care providers generally acknowledged the importance of shared decision making in cancer prevention and endorsed the importance of discussing risks (64%) and benefits (79%) of tests, yet only about half of respondents endorsed eliciting patients' preferences. Hoffman et al. 6 also found that providers did not inquire about patients' screening preferences and tended not to invite patients to engage in decision making. Patients who discussed cancer screening with their health-care providers frequently felt uninformed about the screening including drawbacks of testing. 6 Gaps in communications between providers and patients could negotiate genuine sharedness in the decision-making process. Hoffman et al. 7 found that while 68%-85% of providers expressed opinions about the screening (recommendation), many participants (45%-69%) considered patients the final decision makers. In some cases, patients made decisions with the aid of their provider (27%-38%); only a few patients relied on health-care providers' opinion alone. Yet, these studies showed that health-care providers usually failed to encourage a balanced discussion of screening and patients' preferences. 7,8 This failing might have reduced the quality of the cancer screening decision and may have precluded patients from making an autonomous decision.
Despite progress in US cancer control through regular cervical cancer screening (ie a Pap test, also called a Pap smear), Korean immigrant women (KIW) suffer from considerable disparities in cervical cancer.
They have the second highest cervical cancer incidence rate. [9][10][11] Only 63% to 68% of KIW had a triennial Pap test (vs 89% of non-Hispanic whites). [11][12][13] Studies have examined the correlates of KIW's Pap tests, including English proficiency, physician's recommendations and lower perceived barriers to Pap tests. [14][15][16][17] However, there has been little discussion of decision-making to undergo Pap tests. Compared to screenings for prostate cancer for which a screening recommendation is controversial, the benefits of regular Pap tests are well-known.
Nevertheless, updates on screening methods and intervals for cervical cancer over the last few decades may create confusion; the present guidelines recommend that average-risk women aged 21-65 years receive a Pap test every 3-5 years in the USA. 18,19 Understanding how KIW make a decision about Pap tests in relation to their cultural and personal values can help health-care providers offer a tailored approach to facilitate their decision making. No known study, however, has explored what contributes to their decision to undergo or not undergo a Pap test, KIW's role in the decision-making process or how these women reflect on their decision outcome. Our qualitative study was designed to explore KIW's decision making about whether or not to have a Pap test.

| Study design
We conducted a qualitative descriptive study using 32 semistructured, in-depth individual interviews with KIW residing in the Baltimore-Washington metropolitan area to explore their decision making about a Pap test. The individual interview facilitates in-depth exploration of a respondent's perspectives on health behaviours such as cervical cancer screening to construct meaning and is especially useful when the topic is sensitive. 20,21

| Setting and sample
The inclusion criteria were as follows: (i) KIW 21-65 years of age, (ii) able to read and write in English or Korean; and (iii) had not undergone a hysterectomy. This age range was determined based on the US national cervical cancer screening guidelines. 18 Participants were recruited from a pool of KIW in the control group from a community-based randomized controlled trial to promote breast and cervical cancer screenings; 22 faith-based organizations; an outpatient obstetrics and gynaecology (OB/GYN) clinic; and by word-of-mouth in the Baltimore-Washington metropolitan area. Recruiting the sample for interviews ended when informational redundancy was achieved. 23 Figure 1 describes the recruitment process. Using a standardized phone script to achieve a heterogeneous sample, the principal investigator called 53 control group participants in the community-based trial (n=280), who were selected based on their age, educational level, years of residency in the USA, health insurance status and physician's recommendations. 24 The principle investigator received 30 unique contacts through distributing flyers in faith-based organizations and an outpatient OB/GYN clinic, and by word-of-mouth. Of 83 potential participants, 21 expressed their lack of interest in the study as a reason for refusal, and 30 were unreachable.

| Procedures
The interview guide was prepared with emphasis on KIW's experience with and culture-specific perceptions of having a Pap test, decision making about Pap tests including their roles in decision making, and confidence and satisfaction with the decisions (see Table 1). In collaboration with community members whose characteristics were similar to study participants, two nurses including the principal investigator developed and finalized the interview guide. The interview guide was prepared in English and was then translated into Korean. The team reviewed all changes.
The study team obtained approval from the institutional review board. After ensuring an understanding of the purpose and potential risks and benefits of the study as well as the voluntary participation of participants, the interviewer obtained informed consent. All interviews were conducted in Korean by a PhD candidate in nursing who was a native Korean speaker and fluent in English. The interviews were conducted in a private place selected for the participants' convenience, such as their home and a room at a public library. Each interview lasted on average 1 hour (range 30 minutes-2 hours), and each woman received $30 as a token of appreciation. If needed, a telephone follow-up was requested and completed to clarify any necessary data.
During the interview process, input from participants was added following original discussion by comparing results against prior interviews, which informed the modification of questions in the guide. The confidentiality of participants was maintained throughout the process.
The interviewer ensured that all participants were aware of Pap tests prior to the interview. During the interviews, the interviewer took field notes. The interviews were audio-recorded and transcribed verbatim.
Approximately 72% of the participants were married and middle-aged (mean[SD]=48.7 [11.8] years). More than half of the participants had some college education and more than 10 years of residency in the USA (mean[SD]=14.4 [2.7]), yet had difficulty with English (66%).
Nine participants lacked health insurance, and 22% reported receiving a physician's recommendation to obtain Pap tests. Fifteen had received a triennial Pap test, and six had never received one.

| Data analysis
The interviews were analysed using QSR International's NVivo 10 qualitative data analysis software. The analysis drew on borrowed components from the grounded theory methodology: 20 (i) simultaneous involvement in data collection and reflection on interviews, (ii) open, inductive coding and (iii) memo writing, rather than developing a theory using theoretical sampling and using a constant comparative method. Two bilingual coders read the interview transcripts and field notes several times to develop a general understanding of the interviews and highlight possible categories to explore; the two coders performed the coding process independently. Open, inductive coding was conducted to develop a comprehensive codebook using the first three interviews. Coder agreement rates for each code ranged from 89% to 100%. Discrepancies between the two coders were resolved during team discussions, which is a strategy used to reach a deeper understanding of the data and incorporate diverse perspectives by facilitating discussions between the two coders. Subsequently, the codebook was applied to three interviews to determine whether it fit the data. The codebook was revised based on the identified discrepancies with the data. All transcripts were coded using the finalized codebook and field notes as well as memos, although newly emergent concepts were added to the finalized codebook and applied to the data. The team discussed emerging categories during regular meetings. Memos kept the coders involved in the analysis of the data and helped raise the level of abstraction of the primary coder.

| Methodological rigor
To mitigate methodological concerns related to repeated translation, the final results (ie categories, subcategories, relevant quotes) were

| RESULTS
Three main categories were as follows: women's preferred role in the decision-making process; weighing barriers to and facilitators of a positive decision (vs negative decisions, eg decisions not to have Pap tests); and women's reflections on their decision-making outcome.
Each category included several subcategories explaining the process of women's decision making.

| Women's preferred role: I would participate in making the decision but may defer to a doctor or collaborate with others
Upon being asked about their role in making a Pap test decision, many respondents paused to think, saying that they had never thought about this before. After the women were probed using the interview guide questions, they were able to explain how they participated in making a Pap test decision. KIW's preferred roles in decision making were autonomous, hierarchical, collaborative (some participants with the doctor and others with their spouse) and peer-influenced (Table 3).

| Autonomous type
Many participants claimed that, after searching for information from doctors, friends and the media, they made their own decision to have T A B L E 2 Sample characteristics of qualitative interviews (N=32)

| Hierarchical type
Some respondents who had a regular resource and hence had received provider's recommendations indicated that they followed his or her advice. Interestingly, this pattern was obvious for women who came to the clinic for noticeable health issues, such as bleeding between periods.

| Collaborative type
Respondents made the Pap test decision in collaboration with their health-care providers or significant others after reviewing the infor-

| Barriers to a positive decision to have a Pap test
The recurrent barriers were individual and systemic (Table 4).

| Facilitators of a positive decision to have a Pap test
The recurrent facilitators were perceptions about cervical cancer and the Pap test (a belief that cervical cancer will be cured if detected early, fear about cervical cancer, mother should be healthy) and peer pressure ( Table 5).

Category Sample quotes
Perceptions about cervical cancer and being a mother

| Reflecting on the decision outcome
The women's reflections on the decision outcome differed by their Pap test decisions and included the following subcategories ( OB/GYN, obstetrics and gynaecology.

| A negative decision about a Pap test: still confident or indifferent
Being confident with the decision A few women mentioned that they were confident with their decision not to get a Pap test because they did not have any noticeable symptoms, which made them think they had a low risk of developing cervical cancer. One woman mentioned,  seek medical care at all due to modesty even if they notice concerning symptoms. Nurses who are trained to see patients holistically are ideally positioned to respect patients' modesty. As a direct health-care provider and a bridge to procedures that require patients to expose their genitalia, nurses should encourage patients to express their concerns before and during the medical encounters by carefully assessing patients, building rapport and offering comfortable environments.

| DISCUSSION
In addition, there is an urgent need to help the women navigate the health-care system to gain access to culturally appropriate providers such as same-gender health-care providers. Ample evidence supports that a community-based programme including access-enhancing strategies is successful in promoting cervical cancer screening in women from diverse racial/ethnic groups including KIW. 22

| Limitations
We recruited participants from one ethnic group in a single metropolitan area, which may limit the applicability of the findings beyond the study sample. However, the purpose of this study was to understand the cultural descriptors of decision making in relation to Pap test use among KIW, who have been underrepresented in the cervical cancer screening literature. We also used a thick description strategy by presenting contextual factors and categories with sample quotes to help readers judge transferability beyond the study sample.
Another limitation has to do with potential recollection bias. The participants were asked to reflect on their experiences in relation to their decisions to receive a Pap test which happened from a few months to several years ago. Some participants had difficulty remembering or articulating their decision making about having a Pap test; these women were given the time to think about their decisions and were then probed using questions from the interview guide. This may have led the result to suggest a more rational model for KIW making decisions about Pap tests. However, a retrospective decision-making approach may have prevented social desirability bias that can be caused by a priori discussion regarding their decision about a Pap test.

| CONCLUSIONS
Understanding factors influencing decision making, women's desired role in decision-making process and reflection on their decision outcome is a first step in developing a patient-centred decision-making intervention programme salient to this population, thereby facilitating KIW's desired role in shared decision making. We made recommendations for best practice in controlling cervical cancer among KIW based on the gaps between KIW's unmet needs and current practice in relation to cervical cancer screening. Future research should also consider longitudinal, quantitative studies to examine how decisionmaking processes and outcomes influence KIW's decision to undergo (or not) a Pap test.

ACKNOWLEDGEMENT
Dr. Kim was with Johns Hopkins University when the work was conducted.

FUNDING STATEMENT
Financial support for this study was provided in part by a grant from interpreting the data, writing and publishing the report.