Cultural, demographic, and other non-demographic factors associated with cancer genetic counseling patients' appointment accompaniment preferences in the United States
Abstract
Although the presence of companion(s) in a genetic counseling session can positively influence session dynamics, research has found that some patients prefer to attend their appointments alone. To date, no studies have examined patient accompaniment preferences across different cultural groups in the context of genetic counseling. This quantitative study aimed to identify factors associated with individual preferences in accompaniment at cancer genetic counseling appointments in a sample (N = 130) of Hispanic/Latine (n = 29) and non-Hispanic/Latine White (n = 101) participants at a large academic medical institution. Variables examined included demographics, horizontal and vertical collectivism, and Hispanic and American acculturation. A link to an online questionnaire was emailed to patients who met four criteria: (1) identified as either Hispanic/Latine or non-Hispanic/Latine White; (2) had attended a cancer genetic counseling appointment at UCLA Health to discuss genetic testing options between October 2020 and December 2022; (3) were at least 18 years of age at the time of their appointment; and (4) indicated they were comfortable reading in Spanish or English; responses were anonymous. Logistic regression analyses identified four significant variables in the model associated with accompaniment preferences: individuals with at least one parent born outside of the US, those who attended their appointment in-person, and those with a higher horizontal collectivism score were less likely to want to attend their cancer genetic counseling appointment alone, while the converse was true among those with a higher American acculturation score. These findings highlight cultural and demographic factors that are associated with patient accompaniment preferences unrelated to ethnicity, indicating genetic counselors should not make assumptions regarding accompaniment preferences based solely on cultural or racial/ethnic background. Genetic counselors should incorporate this understanding when assessing patients' accompaniment preferences.
What is known about this topic
Patient circumstances and motivations influence whether they actually attend their appointment alone or with others. Patient accompaniment statuses (i.e., whether individuals attended their appointment with someone or alone) may provide information on their family background and expectations for their appointment.
What this paper adds to the topic
Patient accompaniment preferences for a cancer genetic counseling session in the US are influenced by many factors, including the mode of service delivery, acculturation, collectivism, and the number of parents born outside of the US. Genetic counselors should provide individualized assessments of accompaniment preferences to all patients regardless of their ethnic or cultural background.
1 INTRODUCTION
The decision to pursue predictive genetic testing for cancer is often viewed as a highly personal choice. Making this decision alone may be challenging for some patients, especially given that genetic testing results can implicate both the patient and family members, as well as family relationships (Douglas et al., 2009). Therefore, patients may view the role of appointment accompaniment differently for a genetic counseling appointment when compared to other areas of healthcare.
The presence of a companion, either family member(s) or friend(s), in a genetic counseling session has the potential to influence the dynamics of the session. Companions may play an important role in the patient's decision-making process; they may, for example, provide emotional support to patients, act as a second pair of ears to listen in on information, and provide guidance and perspectives on decisions (Matsukawa et al., 2021). Prior studies into accompaniment in non-genetic counseling cancer appointments showed that companions often contributed by asking questions, expressing concerns, or through stating opinions or preferences (Street & Gordon, 2008). They may also provide emotional support and participate in medical decision-making which may start before and continue after an appointment with the patient's healthcare provider(s) (Laidsaar-Powell et al., 2016).
The ability for companions, especially those who are family members, to influence patient decision-making, is in line with family systems theory, whereby patients are viewed as more than just individuals in an “isolated” system. More specifically, patients often take into account the support they are receiving and would receive from family members during and after the genetic counseling process, perspectives of family members on their medical decisions, and the effect that certain actions (such as genetic testing) will have on family members, among many more (MacLeod et al., 2021; Peterson, 2005). Under family systems theory, genetic counselors should understand the patient's situation coming into genetic counseling in the context of their life histories, values, and goals as active members of social groups, and they must be mindful of the potential impact of genetic testing and diseases on the wider family (MacLeod et al., 2021). However, some patients may feel the need to attend genetic counseling sessions alone; many patients who choose to go alone may feel that the disease in question (in this case cancer) is about them and their decision to pursue testing is theirs to make alone (Matsukawa et al., 2021).
However, there is limited literature examining the role of accompaniment in genetic counseling, let alone factors that are associated with accompaniment preferences in genetic counseling. Arning et al. (2015) found that male patients who decided in favor of predictive testing for Huntington disease were more often accompanied by their partner during the pre-test counseling session than those who decided against testing. Pregnant women in prenatal genetic counseling were more likely to rate their interaction with the genetic counselor positively if their companion was more involved in the session (Aalfs et al., 2006). To the best of our knowledge, the only study that examined accompaniment in cancer genetic counseling was a recent publication by Matsukawa et al. (2021); they examined both actual accompaniment status and motivations in the context of a highly collectivist society among Japanese women coming in for hereditary breast and ovarian cancer genetic counseling. Slightly more than half of the participants attended their session alone. Matsukawa and colleagues found that accompaniment status was influenced by factors including decision-making style, personal confidence, family members' habits and time availability, considerations and conflicts with family members, and healthcare provider's suggestion.
The United States (US) has a culturally and ethnically diverse population; the most recent US census (2022) identified non-Hispanic/Latine White and Hispanic/Latine as the two largest ethnoracial groups in the country. Yet, there is currently no study to date that evaluates accompaniment preferences among these two groups in the context of genetic counseling. There is also a lack of literature exploring the effect of cultural variables, specifically as they relate to collectivism and acculturation, on genetic counseling accompaniment preferences. Examining the impact of cultural variables, such as collectivism and acculturation, on accompaniment preferences, may be more important than directly examining the impact of ethnoracial status on accompaniment preferences, as ethnoracial status often serves as a proxy for cultural variables (Sayegh & Knight, 2013).
Collectivism describes the tendency to value community and group unity over individuality (APA Dictionary of Psychology, 2018a, 2018b). There are two forms of collectivism, namely horizontal and vertical collectivism, horizontal collectivism emphasizes interdependence and equality in status among members of society (Moon et al., 2018). Like horizontal collectivism, vertical collectivism also emphasizes interdependence. However, vertical collectivism endorses a hierarchy in status among the members of a society (Moon et al., 2018). Collectivism may be associated with accompaniment preferences because Yates and de Oliveira (2016) suggest that when it comes to general decision-making, those from collectivist cultures often weigh input from others more than those who come from more individualistic cultures. There is also existing literature showing that Hispanic/Latine persons tend to score higher on measures of collectivism (on average) than those of Northern and Western European background (e.g., Marin & Triandis, 1985; Oyserman et al., 2002; Shkodriani & Gibbons, 1995). However, whether this trend holds true today in the US needs to be examined.
Acculturation is used to describe the process of assimilation to a new culture or cultural environment (APA Dictionary of Psychology, 2018a, 2018b). Acculturation may also play an important role in patient accompaniment preferences. Past studies have demonstrated that some values and behaviors among individuals of Hispanic/Latine identity differ based on an individual's level of acculturation to American culture; these include values related to “inner harmony” (i.e., freedom from internal conflict), parenting style, and health-related behaviors (Abraído-Lanza et al., 2016; Gonzalez & Méndez-Pounds, 2018; Reardon et al., 1997). Therefore, it will be important to investigate the impact of acculturation on accompaniment preferences in the context of genetic counseling.
The primary purpose of this study is to identify, in a sample of US Hispanic/Latine and non-Hispanic/Latine White patients, cultural and demographic factors that are associated with their accompaniment preference (with someone or alone) when attending cancer genetic counseling session. Investigation into these research questions is important, as it can help genetic counselors in the US better understand the preferences of a diverse patient population and how these preferences may differ across the two largest ethnoracial groups in the US. This study may also serve as a framework for understanding accompaniment preferences among patient populations outside the US. Ultimately, cultural competency and humility are key in the successful delivery of personalized medicine. The overarching aim of this research study is to contribute to genetics providers' knowledge and understanding of patients' accompaniment preferences and the potential role of culture and other factors in these preferences, and how providers can utilize this information to better serve their individual patients.
2 METHODS
2.1 Study design
This was an observational and retrospective cross-sectional study investigating the following research question:
What cultural and demographic factors are associated with accompaniment preferences to a cancer genetic counseling appointment among Hispanic/Latine persons and non-Hispanic/Latine White persons in the US?
2.2 Participants and procedures
Participants were initially identified through the UCLA Health electronic health record system by UCLA's Clinical and Translational Science Institute based on the following criteria: (1) self-identified as either Hispanic/Latine (any race) or non-Hispanic/Latine White in their UCLA medical record, (2) at least 18 years of age at the time of their cancer genetic counseling appointment, and (3) attended a cancer genetic counseling appointment at UCLA Health (a large academic medical institution) to discuss testing options between October 2020 and December 2022.
A total of 1689 participants met these inclusion criteria and were sent a recruitment email containing a link to a one-time online anonymous questionnaire administered through REDCap, which took approximately 10–15 min to complete.
Participants were further screened through the eligibility screening portion of the questionnaire and were eligible for the study if they met the following inclusion criteria: (1) self-identified as either Hispanic/Latine or non-Hispanic/Latine White, (2) at least 18 years of age at the time of their cancer genetic counseling appointment, (3) attended a cancer genetic counseling appointment at UCLA Health to discuss genetic testing options in the past 2 years, and (4) indicated being comfortable with reading in Spanish or English. All participants who completed the questionnaire were given the option to participate in a raffle prize for one of four $25 USD gift cards; participants interested in enrolling in the raffle were directed to a separate REDCap survey through a link provided at the end of the questionnaire. Both English and Spanish translations of the recruitment email and online questionnaire were made available to participants (Data S1 and S2). A credible third-party translation service was utilized for translation of the questionnaire and recruitment email. All data for the questionnaire were collected between October 2022 and December 2022. This study was approved by the UCLA Institutional Review Board (#22-001115).
2.3 Instrumentation
The questionnaire consisted of five parts: (1) eligibility screening, (2) demographic-related questions, (3) two yes/no questions related to accompaniment (i.e., whether the patient actually attended the cancer genetic counseling appointment with someone; and, if they wanted to attend their appointment with someone), (4) questions that measured horizontal and vertical collectivism, and (5) questions that measured acculturation. The survey is included as a supplemental file. Please note that the study questionnaire employs the term “Latino” instead of “Latine” since “Latino” is likely a more familiar and/or recognized term among the general public.
Demographic variables included race/ethnicity (i.e., Hispanic/Latine versus non-Hispanic/Latine White), age, sex assigned at birth, place of birth, the number of parents born outside of the US, education, and marital status. Non-demographic variables included mode of genetic counseling appointment delivery, time since appointment, horizontal collectivism score, vertical collectivism score, American acculturation score, and Hispanic acculturation score. Mode of service delivery for cancer genetic counseling appointments at UCLA Health are in-person or via video visits (telehealth) where the patient uses their personal device from a private location of their choosing to connect with the genetic counselor's device.
Horizontal collectivism and vertical collectivism were measured using the Individualism and Collectivism Scale developed by Triandis and Gelfand (1998). The scale was shown to have good convergent and divergent validity (Triandis & Gelfand, 1998). The original scale contained 16 items designed to capture four dimensions of collectivism and individualism. For this study, only the vertical collectivism and horizontal collectivism measures were included, as the focus of the research questions was on the effects of collectivism on accompaniment preferences, and we aimed to reduce the length of time needed to complete the online questionnaire to maximize participation and completion. All items were answered using a 9-point Likert scale, ranging from 1 = never or definitely no to 9 = always or definitely yes.
Acculturation was measured using a modified version of the Multidimensional Acculturation Scale-II (MAS-II; Rodriguez et al., 2007); the word “Mexican” was replaced with “Latino/Hispanic” throughout the entire scale to examine Hispanic/Latine acculturation at large. One question on the original version of the MAS-II, which asked, “How much do you enjoy celebrating Mexican holidays or traditions (e.g., Dia de los Meurtos, Quinceañeras)” was changed to “How much do you enjoy celebrating Latino/Hispanic holidays or traditions,” removing examples of Mexican-specific holidays. The MAS-II is a 22-item, validated, and reliable measure that assesses one's level of acculturation. Items were originally constructed to capture the degree of involvement in and identification with Mexican culture and American culture separately. This measure contains four subscales: English proficiency (Cronbach's α = 0.93), Spanish proficiency (Cronbach's α = 0.92), Mexican cultural identity (Cronbach's α = 0.81), and American cultural identity (Cronbach's α = 0.78; Rodriguez et al., 2007). All items were answered using a 6-point Likert scale, ranging from 0 = does not apply/not well/not at all to 5 = very well/very much. Items from English proficiency and American cultural identity were summed and averaged to generate an American acculturation score. Similarly, items from Spanish proficiency and Hispanic cultural identity were summed and averaged to generate a Hispanic acculturation score. The remaining questionnaire items were developed by the research team.
2.4 Data analysis
Variables were summarized using averages (standard deviations) for quantitative variables and frequencies for categorical variables. Statistical analysis was carried out using SPSS Statistical Software (Version 28.0.0.0).
To address the research aim, we first performed univariate associations of demographic and non-demographic factors with accompaniment preference using t-test, Chi-square, or Fisher's exact test. Where t-tests and Fisher's exact tests were utilized in the analyses, a significance threshold of 0.002 was used to correct for multiple univariate testing (i.e., Bonferroni correction). This threshold was calculated by dividing the significance threshold of 0.05 by the total number of variables examined in the univariate analysis.
We then performed a backward stepwise binary logistic regression using accompaniment preference to the cancer genetic counseling session (wanted to go alone/wanted to go with someone) as the outcome variable. Demographic and non-demographic factors (complete-case data) were used as variables for this regression analyses. Actual accompaniment status was excluded as a variable due to issues related to complete separation when this variable was included in the logistic regression model. To maximize the complete-case data, we analyzed sex identified at birth for which there were no missing responses instead of gender identity which had two missing responses. We note that there was concordance between these two variables for 127/128 respondents with data on both variables. The levels of significance used to include and remove variables in the stepwise regression was 0.05 for entry and 0.10 for removal.
3 RESULTS
3.1 Participant demographics
A total of 154 anonymous responses were obtained from the 1689 eligible participants who were sent recruitment emails containing the online questionnaire. Responses were excluded if respondents (1) did not complete the eligibility screening or did not meet eligibility criteria and/or (2) did not answer demographic-related questions as well as questions related to accompaniment on the questionnaire. Ultimately, a total of 130 responses were analyzed for a response rate of 7.7%.
Table 1 provides results on participants' demographics based on the total sample and across the Hispanic/Latine and non-Hispanic/Latine White participants. Among these 130 participants, most were assigned female sex at birth (87%) and identified as women (86.7%). The majority of participants identified as non-Hispanic/Latine White (78%) while 22% identified as Hispanic/Latine. Among Hispanic/Latine participants, most (79%) also identified as White and the remainder (21%) identified with other race categories. The average age of participants was 51.0 years (SD = 15.4 years; range min = 21, max = 83). Most participants in the sample (73%) earned a bachelor's degree or higher. Approximately two-thirds of participants (86/129) indicated that they had a partner (i.e., domestic partner or were married). Among the 17 participants who brought others with them to their appointment, 11 indicated that they brought family member(s), and six indicated that they brought partner(s). Most participants were born within the US (113/128, 88%).
| All participants (n = 130) | Hispanic/Latine (n = 29) | Non-Hispanic/Latine White (n = 101) | p a | |
|---|---|---|---|---|
| n (%)/Average | ||||
| Age (SD) (n = 130) | 51.0 (15.4) | 46.2 (16.2) | 52.3 (14.9) | ns |
| Sex assigned at birth (n = 130) | ||||
| Male | 17 (13) | 4 (14) | 13 (13) | ns |
| Female | 113 (87) | 25 (86) | 88 (87) | |
| Place of birth (n = 128) | ||||
| US | 113 (88) | 24 (86) | 89 (89) | ns |
| Outside of the US | 15 (12) | 4 (14) | 11 (11) | |
| Number of parents born outside of the US (n = 128) | ||||
| None | 82 (64) | 7 (25) | 75 (75) | <0.001 |
| One or more | 46 (36) | 21 (75) | 25 (25) | |
| Educationb (n = 129) | ||||
| Less than Bachelor's degree | 35 (27) | 8 (29) | 27 (27) | ns |
| Bachelor's degree or higher | 94 (73) | 20 (71) | 74 (73) | |
| Marital status (n = 129) | ||||
| Single (single, divorced, widowed) | 43 (33) | 13 (46) | 30 (30) | ns |
| Partner (domestic partner, married) | 86 (67) | 15 (54) | 71 (70) | |
- Abbreviation: SD, standard deviation.
- a p-Values from Fisher's exact test for categorical variables, and independent t-tests for equality of averages (two-sided, equal variances assumed) for continuous variables comparing demographic variables between Hispanic/Latine participants and non-Hispanic/Latine White participants. A significance threshold of 0.002 was utilized to correct for multiple testing. “ns” indicates that the p-value was non-significant (>0.002).
- b Less than Bachelor's degree includes: less than high school diploma, high school diploma, some college but no degree, and Associate's degree. Bachelor's degree or higher includes: Bachelor's degree, Master's degree, Professional Degree (e.g., MD, DDS, JD), and Doctorate (e.g., PhD, EdD).
As seen in Table 1, there were no statistically significant differences in average age, sex assigned at birth, place of birth, education, and marital status between individuals who self-reported as Hispanic/Latine versus non-Hispanic/Latine White. However, there was a statistically significant difference (p < 0.001) between the two groups in terms of the number of parents born outside of the US, with Hispanic/Latine respondents being more likely to indicate that they have at least one parent born outside of the US.
As seen in Table 2, of the 130 participants in the sample, the majority (67%) indicated that they wanted to attend their cancer genetic counseling appointment alone, with most (87%) indicating that they did attend their appointment alone. Most participants (80%) had a virtual (telehealth) appointment and 20% had an in-person appointment. The majority (57%) of participants had their cancer genetic counseling appointment over 1 year ago and the remainder had their appointment less than 1 year ago.
| All participants (n = 130) | Hispanic/Latine (n = 29) | Non-Hispanic/Latine White (n = 101) | p a | |
|---|---|---|---|---|
| n (%)/Average | ||||
| Actual accompaniment to cancer genetic counseling appointment (n = 130) | ||||
| Went with someone | 17 (13) | 1 (3) | 16 (16) | ns |
| Went alone | 113 (87) | 28 (97) | 85 (84) | |
| Accompaniment preference to cancer genetic counseling appointment (n = 130) | ||||
| Preferred to have someone with them | 43 (33) | 10 (34) | 33 (33) | ns |
| Preferred to go alone | 87 (67) | 19 (66) | 68 (67) | |
| Mode of genetic counseling delivery (n = 130) | ||||
| Virtual/Telehealth | 104 (80) | 23 (79) | 81 (80) | ns |
| In-person | 26 (20) | 6 (21) | 20 (20) | |
| Time since cancer genetic counseling appointment (n = 130) | ||||
| Less than 1 year ago | 56 (43) | 15 (52) | 41 (41) | ns |
| More than 1 year ago | 74 (57) | 14 (48) | 60 (59) | |
| HC score (SD, n = 125) | – | 7.58 (1.09) | 7.67 (1.21) | ns |
| VC score (SD, n = 125) | – | 6.73 (1.72) | 6.73 (1.57) | ns |
| American acculturation score (SD, n = 112) | – | 4.34 (0.59) | 4.13 (0.63) | ns |
| Hispanic acculturation score (SD, n = 112) | – | 3.43 (1.36) | 1.33 (0.89) | <0.001 |
- Abbreviations: HC, horizonal collectivism; SD, standard deviation; VC, vertical collectivism.
- a p-Values from Fisher's exact test for categorical variables, and independent t-tests for equality of averages (two-sided, equal variances assumed) for continuous variables comparing non-demographic variables between Hispanic/Latine participants and non-Hispanic/Latine White participants. A significance threshold of 0.002 was utilized to correct for multiple testing. “ns” indicates that the p-value was non-significant (>0.002).
No statistically significant differences were identified in actual accompaniment status, accompaniment preferences, mode of genetic counseling delivery (virtual versus in-person), or time since genetic counseling appointment (<1 or >1 year ago) between individuals of Hispanic/Latine and non-Hispanic/Latine White identity. The average Hispanic acculturation score (n = 112, t(df) = 110, p < 0.001) was significantly higher in the Hispanic/Latine group than in the group who identified as non-Hispanic/Latine White. No statistically significant differences were identified in average horizontal collectivism score (n = 125), average vertical collectivism score (n = 125), and average American acculturation score (n = 112) between individuals of Hispanic/Latine and non-Hispanic/Latine White identity.
3.2 Univariate analysis of factors associated with preference for attending a cancer genetic counseling appointment with someone or alone
Individuals who indicated that someone accompanied them to their cancer genetic counseling appointment were more likely to prefer to have someone attend the appointment with them (p < 0.001; Table 3).
| Preferred to have someone accompany them to cancer genetic counseling appointment (n = 43) | Preferred not to have someone accompany them to cancer genetic counseling appointment (n = 87) | p a | |
|---|---|---|---|
| n (%)/Average | |||
| Actual accompaniment to cancer genetic counseling appointment (n = 130) | |||
| Went with someone (n = 17) | 15 (88) | 2 (12) | <0.001 |
| Went alone (n = 113) | 28 (25) | 85 (75) | |
| Age (SD) (n = 130) | 50.3 (16.9) | 51.3 (14.6) | ns |
| Sex assigned at birth (n = 130) | |||
| Male (n = 17) | 3 (18) | 14 (82) | ns |
| Female (n = 113) | 40 (35) | 73 (65) | |
| Place of birth (n = 128) | |||
| US (n = 113) | 35 (31) | 78 (69) | ns |
| Outside of the US (n = 15) | 7 (47) | 8 (53) | |
| Ethnicity (n = 130) | |||
| Hispanic/Latine (n = 29) | 10 (35) | 19 (65) | ns |
| Non-Hispanic White/Non-Latine White (n = 101) | 33 (33) | 68 (67) | |
| Number of parents born outside of the US (n = 128) | |||
| None (n = 82) | 20 (24) | 62 (76) | ns* |
| One or more (n = 46) | 22 (48) | 24 (52) | |
| Educationb (n = 129) | |||
| Less than Bachelor's degree (n = 35) | 17 (49) | 18 (51) | ns* |
| Bachelor's degree or higher (n = 94) | 26 (28) | 68 (72) | |
| Marital status (n = 129) | |||
| Single (single, divorced, widowed) (n = 43) | 20 (46) | 23 (54) | ns* |
| Partner (domestic partner, married) (n = 86) | 23 (27) | 63 (73) | |
| Mode of genetic counseling delivery (n = 130) | |||
| Virtual/Telehealth (n = 104) | 29 (28) | 75 (72) | ns* |
| In-person (n = 26) | 14 (54) | 12 (46) | |
| Time since cancer genetic counseling appointment (n = 130) | |||
| Less than 1 year ago (n = 56) | 19 (34) | 37 (66) | ns |
| More than 1 year ago (n = 74) | 24 (32) | 50 (68) | |
| HC score (SD, n = 125) | 7.89 (0.83) | 7.53 (1.31) | ns |
| VC score (SD, n = 125) | 6.71 (1.47) | 6.74 (1.66) | ns |
| American acculturation score (SD, n = 112) | 4.05 (0.67) | 4.24 (0.61) | ns |
| Hispanic acculturation score (SD, n = 112) | 1.75 (1.40) | 1.86 (1.33) | ns |
- Abbreviations: HC, horizonal collectivism; SD, standard deviation; VC, vertical collectivism.
- a p-Values from Fisher's exact test for categorical variables, and independent t-tests for equality of averages (two-sided, equal variances assumed) for continuous variables comparing demographic and non-demographic variables between participants who wanted to have someone attend their appointment with them compared to those who wanted to attend their appointment alone. A significance threshold of 0.002 was utilized to correct for multiple testing. “ns*” indicates that the p-value was non-significant, but less than 0.05 (0.05 > p > 0.002). “ns” indicates that the p-value was non-significant (p > 0.05).
- b Less than Bachelor's degree includes: less than high school diploma, high school diploma, some college but no degree, and Associate's degree. Bachelor's degree or higher includes: Bachelor's degree, Master's degree, and Professional Degree (e.g., MD, DDS, JD), Doctorate (e.g., PhD, EdD).
No statistically significant differences were found in average age, sex assigned at birth, place of birth, the number of parents born outside of the US, education level, marital status, race/ethnicity (i.e., Hispanic/Latine persons and non-Hispanic/Latine White participants), time since genetic counseling appointment, mode of genetic counseling delivery, average horizontal collectivism score, average vertical collectivism score, average American acculturation score, and average Hispanic acculturation score between participants who indicated a preference to attend their appointment with someone versus those who preferred to attend their appointment alone (Table 3).
3.3 Multivariable analysis of factors associated with preference for attending a cancer genetic counseling appointment with someone or alone
Multivariable analysis through backward stepwise binary logistic regression was performed to examine the association between demographic and non-demographic variables with accompaniment preference as the outcome variable (see Table 4). A total of 110 participant responses were included in this analysis and 20 were omitted due to missing data on at least one variable examined in the model.
| Variable | B | Exp(B) | CI (lower) | CI (upper) | SE | df | Sig. |
|---|---|---|---|---|---|---|---|
| Mode of genetic counseling delivery | −1.22 | 0.30 | 0.10 | 0.85 | 0.54 | 1 | 0.02 |
| HC score | −0.61 | 0.55 | 0.33 | 0.90 | 0.25 | 1 | 0.02 |
| American acculturation score | 1.12 | 3.08 | 1.35 | 7.00 | 0.42 | 1 | 0.01 |
| Number of parents born outside of the US | −1.19 | 0.31 | 0.12 | 0.78 | 0.48 | 1 | 0.01 |
- Abbreviations: CI, confidence interval for Exp(B); df, degrees of freedom; HC, horizonal collectivism; SE, standard error; sig, significance.
Multivariable analysis identified mode of genetic counseling appointment delivery, horizontal collectivism score, American acculturation score, and the number of parents born outside of the US as being significantly associated with accompaniment preference. Individuals who attended their appointment in-person were also significantly less likely to prefer to attend their cancer genetic counseling appointment alone. Additionally, those with a higher horizontal collectivism score were significantly less likely to prefer to attend their cancer genetic counseling appointment alone. Participants with a higher American acculturation score were more likely to prefer to attend their cancer genetic counseling appointment alone. Finally, individuals with at least one parent born outside of the US were significantly less likely to prefer to attend their cancer genetic counseling appointment alone.
4 DISCUSSION
The primary purpose of this observational and retrospective cross-sectional study was to identify factors that are associated with cancer genetic counseling accompaniment preferences among a sample of Hispanic/Latine and non-Hispanic/Latine White patients in the US. Multivariable analysis revealed that mode of genetic counseling appointment delivery, horizontal collectivism score, American acculturation score, and the number of parents born outside of the US are significantly associated with accompaniment preferences. As anticipated, there was a significant difference in the average Hispanic acculturation score, such that individuals of Hispanic/Latine identity scored higher on average on this measure of acculturation, indicative of greater affiliation with Hispanic culture, compared to individuals of non-Hispanic/Latine identity in our sample. However, no significant group differences were found in the average horizontal collectivism score, the average vertical collectivism score, the average American acculturation score, and accompaniment preferences. In total, our findings highlight that cultural and demographic factors unrelated to ethnicity are associated with patient accompaniment preferences.
4.1 Participant characteristics
The participants in our sample predominantly identified as non-Hispanic/Latine White and most participants were assigned female sex at birth and identified as a woman. The ethnic composition of participants is fairly consistent with the West Los Angeles (LA), California population; approximately 67% of the population in West LA identifies as non-Hispanic/Latine White and 12% of the population identifies as Hispanic or Latine (Los Angeles City Planning, 2021). The cancer genetic counseling clinics where participants were seen in this study are all located in or near West LA; however, it is not expected that all patients who receive care from these clinics reside in the West LA area, which may be especially true with the increase in uptake of and support for telehealth services since the earlier stages of the COVID-19 pandemic in the US (i.e., Spring 2020) at various medical institutions including UCLA Health. The large proportion of female participants in our sample is also fairly consistent with the population of cancer genetic counseling patients in the US. Using data from the 2015 National Health Interview Survey (NHIS), Stamp et al. (2019) found that females were significantly more likely to uptake cancer genetic counseling compared to males regardless of whether they had a cancer diagnosis (OR = 1.70, 95% CI [1.30, 2.21]) versus not (OR = 1.78, 95% CI [1.18, 2.67]). The proportion of individuals in our sample (73%) who indicated that they had a bachelor's degree or higher is greater than what is observed in the population in West LA (55%; Los Angeles City Planning, 2021). Therefore, our findings are more likely to be generalizable to individuals of these two backgrounds (i.e., non-Hispanic/Latine White and Hispanic/Latine) who have obtained a higher level of education (i.e., bachelor's degree or higher). Furthermore, these findings may be more likely to be generalizable to individuals of Hispanic/Latine background of Mexican ancestry as compared to other Hispanic/Latine ancestries as the Hispanic/Latine population in Los Angeles is predominantly (75%) comprised of individuals of Mexican ancestry (U.S. Census Bureau, 2021).
There were no significant differences in average vertical and horizontal collectivism scores across the two ethnic groups. This contrasts with past literature that has found that individuals of Hispanic/Latine identity tend to score higher on measures of collectivism (on average) than those of Northern and Western European backgrounds (Marin & Triandis, 1985; Oyserman et al., 2002; Shkodriani & Gibbons, 1995). However, many of these studies that examined trends in collectivism are at least a decade old. We found no differences in average American acculturation scores across Hispanic/Latine persons and non-Hispanic/Latine White persons in our sample, indicating that individuals of Hispanic/Latine identity in our sample are similarly acculturated to the mainstream American culture as measured by the MAS-II. Past studies have found that Hispanic/Latine individuals appeared to lose the strength of their collectivistic values (i.e., score lower on measures of collectivism) the more acculturated they were to the US culture (Gomez, 2003; Sawicki & Chapa, 2018). Gomez (2003) also highlighted that the elevated level of acculturation seen in their sample of Hispanic/Latine participants was, on average, elevated given their high level of education. This could explain why there were no significant differences in collectivism scores across the two ethnic groups, given that both Hispanic/Latine and non-Hispanic/Latine White persons in our sample were, on average, similarly acculturated to the US culture and had similar levels of education.
4.2 Variables associated with accompaniment preferences
Through multivariable analysis, using data from all participants, we identified four variables significantly associated with accompaniment preferences: (1) the mode of genetic counseling appointment delivery, (2) horizontal collectivism score, (3) American acculturation score, and (4) the number of parents born outside of the US. First, participants who attended their appointment in-person were also less likely to want to attend their appointment alone. While this study did not explore reasons for this, they may be related to transportation needs or psychosocial concerns that can be better addressed with the visual and physical cues afforded by in-person interactions. This is an area in need of additional research.
Horizontal collectivism score was significantly associated with accompaniment preferences. Those with a higher score on horizontal collectivism, indicating higher agreement with collectivism, were less likely to want to attend their cancer genetic counseling appointment alone. This is consistent with the idea that members of more collectivistic societies are highly interdependent, which may translate to a greater likelihood to seek the support and attendance of other members in healthcare appointments. Although the reason why horizontal collectivism score had a significant association but not vertical collectivism score remains unknown, a study by Kazarian (2005) found that horizontal collectivism specifically was a positive predictor of healthy family functioning. It is hypothesized that a healthy family functioning translates to individuals feeling more comfortable with bringing family members with them to healthcare appointments, such as cancer genetic counseling appointments.
Horizontal-collectivistic and vertical-collectivistic societies also differ in values of equality; horizontal-collectivistic societies tend to hold egalitarian views among group members while vertical-collectivistic societies adopt a sense of hierarchy among group members (e.g., elders are seen as higher up on this “hierarchy” than younger members) (Čavojová et al., 2022). Therefore, it is possible that one's sense of verticality within a society may result in a sense of separation of oneself from other members (Schermer et al., 2023). In line with this, Schermer et al. (2023) found that horizontal and vertical aspects of collectivism were significantly associated with loneliness such that horizontal collectivism was significantly associated with lower loneliness scores while vertical collectivism was associated with higher loneliness scores among participants. Based on this information, it is theorized that a higher horizontal collectivism score would be more positively associated with intergroup connectivity than vertical collectivism, which may translate to greater likelihood to attend their genetic counseling appointments with others.
Additionally, participants with a higher average American acculturation score were more likely to want to attend their cancer genetic counseling appointment alone. Although individualism was not measured in this study, it is hypothesized that values of individualism would be positively correlated with acculturation to the American mainstream culture based on prior research (Tamimy et al., 2022; Yates & de Oliveira, 2016). Individualism is often characterized by values of independence and self-reliance; previous studies have suggested that when it comes to general decision-making, those from collectivist cultures often weigh input from others more than those who come from more individualistic cultures, which may translate to a greater likelihood of wanting to attend medical appointments alone (e.g., Yates & de Oliveira, 2016).
In addition, those with at least one parent born outside of the US were less likely to want to attend their appointment alone. Interestingly, place of birth (i.e., born in the US versus outside of the US) was not significantly associated with accompaniment preferences in the multivariable analysis. However, these two variables are significantly correlated and it may be that the number of parents born outside of the US had a greater association with accompaniment preferences than place of birth. Pang et al. (2003) suggest that positive family relationships may aid and encourage an older immigrant's use of healthcare through facilitating the communication of healthcare information, bridging language/cultural gaps, and providing transportation to clinics. Given that patient decisions to pursue genetic testing can implicate family members as well, it is hypothesized that patients with immigrant parents are more likely to bring these family members with them to their appointment for reasons such as those stated above.
Finally, although there were discrepancies in which variables were significant in relation to accompaniment preferences between the univariate and multivariable analyses (e.g., horizontal collectivism and American acculturation were not significant in the univariate analysis but were significant in the multivariable analysis), the multivariable analysis was able to account for and/or control for the effects of multiple variables on the outcome variable as part of the stepwise analysis. Stepwise logistic regression identifies variables associated with the outcome variable sequentially, and new variables can become significant as other variables are added to the model. This may likely explain why variables that were non-significant in the univariate analysis became significant in the multivariable analysis.
There is limited literature exploring accompaniment preferences in genetic counseling, including cancer. A study examining accompaniment statuses among Japanese patients who received cancer genetic counseling due to concerns over hereditary breast and ovarian cancer found that most individuals (59%) attended their appointment by themselves (Matsukawa et al., 2021). In our sample, most individuals also attended their appointment by themselves (87%), though at a much higher proportion, with no significant differences between individuals of Hispanic/Latine and non-Hispanic/Latine White identity. One factor that could explain this difference is the fact that, as previously noted, American culture as a whole is significantly more individualistic than collectivistic (Tamimy et al., 2022) in comparison to Japanese culture, which is often viewed as highly collectivistic (Matsukawa et al., 2021). This may have influenced more individuals in our sample (all residing in the US) to attend their appointment alone (Uhlmann et al., 2009). Temporal effects may have also played a role in these differences; data from the Japanese study were collected prior to the COVID-19 pandemic (between September 2016 and April 2017), whereas participants in our study had their appointments during the pandemic (between October 2020 and December 2022). The pandemic limited the availability of in-person services, with most participants in our sample indicating that they had their appointment conducted virtually (80%), a trend that is consistent with the genetic counseling literature during this general time frame (Mills et al., 2021). A general decrease in tolerance for social gatherings and a decrease in tolerance for others to join patients' in-person appointments during the height of the pandemic may be contributing factors as to why many patients chose to attend their appointment alone. However, many hospitals had exceptions to their visitation policy when it came to outpatient services to allow for one or more family members to accompany them to their appointment (Jaswaney et al., 2022; Weiner et al., 2021).
Interestingly, when it came to accompaniment preferences, although most individuals still indicated that they wish they attended their appointment alone (with no significant differences between Hispanic/Latine and non-Hispanic/Latine White persons), the proportion of individuals who wanted to attend their appointment with someone was higher than what was observed in participants' actual accompaniment status (i.e., proportion of individuals that actually went with someone to their appointment). This discrepancy indicates that participants' accompaniment preferences were not always being met. Although this study did not explore the factors that led to this discrepancy, the study by Matsukawa et al. (2021) examined reasons why their participants attended their appointment alone or with someone. Most participants who attended their appointment with someone listed their primary reason as (1) wanting to explain the disease to him/her (25%), (2) he/she was worried about me (19%), and (3) he/she has been to genetic counseling (13%). Almost all individuals (95%) who attended their appointment alone indicated their primary reason as “it [being] about [their own] disease” (Matsukawa et al., 2021). No participants in this group indicated that they went alone because someone else was not available; however, this may have been due to a small sample size (N = 19; Matsukawa et al., 2021). Additionally, because our study asked participants to recall appointments that were conducted during the pandemic (between October 2020 and December 2022), more barriers to accompaniment (i.e., bringing others to one's appointment) were likely to be present in our sample. Some of these may include difficulties coordinating the attendance of other individuals through telehealth and overall quarantine/isolation protocols across the country that limited person-to-person contact during this time, as mentioned previously.
5 LIMITATIONS
This research study has limitations worth noting. First, only individuals who were seen by a cancer genetic counselor at a single academic medical institution in West LA were included in the study, which limits the generalizability of our findings. Participants without email addresses in the EMR, those without information about their race and/or ethnicity in the EMR, those who could not navigate completing an online survey, and those who could not read English or Spanish (who were otherwise eligible for the study) could not participate in this study. Participants were eligible if they attended a cancer genetic counseling appointment to discuss genetic testing options, but we acknowledge that participants may have had more than one visit with a genetic counselor following this appointment. Because our intent was to collect information on a single clinic visit, the recruitment information sheet included a statement at the beginning: “Please ignore this email if you have already completed this survey.” We have no mechanism for ascertaining if someone completed the survey more than once. All responses were statistically treated as independent observations.
We also modified the MAS-II by replacing the word “Mexican” with “Hispanic” to examine Hispanic acculturation at large; this modification may have impacted the validity and reliability of the measure. Additionally, eligible participants for our study were identified through a separate party (UCLA's Clinical and Translational Science Institute) and sent a recruitment email. The only data available to the researchers were anonymous responses to the questionnaire. It is possible that respondents differ from non-respondents in factor(s) relevant to the research questions; however, a comparison between these two groups is not possible. The response rate for the questionnaire was also low (7.7%) compared to other studies using Web-based questionnaires, which diminished the power to detect statistical significance (Fincham, 2008; Horevoorts et al., 2015).
Our sample size calculations indicated we would have 80% power to detect a medium effect size difference in proportions with a sample of n = 130 if both sub-samples were equal (n = 65 in each group). Unfortunately, our subsamples for ethnoracial groups were not equal (Hispanic/Latine (n = 29) and non-Hispanic/Latine White (n = 101)), thereby compromising our power and only providing 80% power to detect larger effects. Likewise, our subsamples for accompaniment preferences (preferred to have someone accompany them (n = 43) and preferred not to have someone accompany them (n = 87) were also unequal (albeit closer to being equal) than the difference seen between the ethnoracial groups). It was difficult to estimate an a priori sample size as there was limited literature to inform the values needed to calculate this estimate; however, using the results we obtained from our study to calculate the required sample size for 80% power and an alpha of 0.05 (two sided) for both the independent t test and Fisher's exact test, we found that a minority of our analyses had adequate power.
Furthermore, because this study only included participants who were seen by a cancer genetic counselor at UCLA between October 2020 and December 2022, during the height of the COVID-19 pandemic, the proportion of individuals who attended their appointment alone versus with someone may not reflect the trends that are currently seen today and in the future, as well as prior to the pandemic. Likewise, accompaniment trends may also differ among different patient populations and medical institutions. It is likely that the initial transition to telehealth and social distancing measures (adopted at the personal level, governmental level, medical/institutional level, and clinic level) and any changes in policies over time may have affected patients' ability to attend a session with others. Finally, these trends in actual accompaniment status and accompaniment preferences may not apply to other genetic counseling specialties (e.g., prenatal and pediatrics), where accompaniment may have a different level of importance or play a different role than in cancer (e.g., a parent accompanying their child for a pediatric genetics visit or a partner accompanying a patient to make a decision about a pregnancy in a prenatal visit).
6 IMPLICATIONS FOR GENETIC COUNSELING PRACTICE
These research findings demonstrate the importance of seeking competency while adopting humility in the practice of genetic counseling when it comes to working with patients of different ethnic and cultural backgrounds. Cultural competence refers to the ability for healthcare systems and providers to be cognizant of and responsive to the cultural backgrounds and perspectives of the individuals they serve (Betancourt et al., 2002). There is significant heterogeneity in the behaviors and practices of individuals within cultural groups, and it is important to consider that race, culture, and ethnicity are just a few aspects of people's identities. It is arguably the intersection of one's multiple identities and lived experiences that ultimately influences their values and behaviors, including health behaviors. Therefore, it is important to not stereotype the behaviors, values, and preferences of individuals based solely on people's cultural or ethnic background.
We originally hypothesized that one's ethnoracial background would be associated with one's actual accompaniment status and one's accompaniment preference, which was not the case in this sample. Therefore, in light of cultural humility, individuals, including genetic counselors, should not assume that patients have differing preferences in accompaniment or aspects of counseling based on ethnoracial background alone. Culture, race, and ethnicity are only a few aspects among many factors that may influence accompaniment status and preferences, and in fact, they often represent proxy variables for more specific factors (e.g., acculturation and other specific cultural values) that can be present to similar or varying degrees both within and across ethnoracial groups (Sayegh & Knight, 2013). Genetic counselors should incorporate an understanding of this into their practice when assessing the accompaniment preferences of each patient.
7 FUTURE RESEARCH
Given the discrepancies between actual accompaniment statuses and accompaniment preferences seen in our sample, future research should explore methods genetic counselors and clinics can adopt, especially in the context of telehealth, to accommodate patients who would like to bring others with them to their genetic counseling appointment. Further exploration should also be pursued to examine barriers to accompaniment across different telehealth platforms and how these platforms can better accommodate patients who wish to attend their virtual appointment with others. The impact of other cultural factors on accompaniment statuses and preferences, such as individualism, should also be examined. Future studies should also pay special attention to individuals whose actual accompaniment status differs from their accompaniment preferences. Comparisons between this group and the group whose actual accompaniment status was the same as their indicated accompaniment preferences should also be performed.
Future studies should strive to include larger and more generalizable samples to achieve greater statistical power. Furthermore, they should also explore the reasons for individuals' actual accompaniment status and accompaniment preferences (e.g., through qualitative methods), similar to what was conducted in the study by Matsukawa et al. (2021). Finally, subsequent research should also explore strategies for assessing the accompaniment preferences of patients and strategies to accommodate these preferences to provide better services for patients.
8 CONCLUSION
This study identified four variables significantly associated with accompaniment preferences in a combined sample of Hispanic/Latine and non-Hispanic/Latine White participants in the US: the mode of genetic counseling appointment delivery, horizontal collectivism score, American acculturation score, and the number of parents born outside of the US.
Our research findings highlight the importance of practicing humility in the genetic counseling profession through the avoidance of group stereotyping; no assumptions should be made about the accompaniment preferences of our patients based solely on cultural or ethnic background. Finally, our findings demonstrate a clear need to better understand the reasons behind the gap between actual accompaniment statuses and preferences, especially in the age of telehealth, and evidence-based strategies that genetic counselors and clinics can employ to bridge these disparities.
AUTHOR CONTRIBUTIONS
JL was involved in the conception, design, data collection, data analysis, writing, and editing of the manuscript. CGSP was involved in the design, data analysis, and editing of the manuscript. RMC was involved in the data analysis and editing of the manuscript. MN-S was involved in initial formulation of data access, data interpretation, and editing of the manuscript; PS was involved in the design of the research and editing of the manuscript. JL confirms that they had full access to all the data in the study and takes responsibility for the integrity of the data and the accuracy of the data analysis. All of the authors gave final approval of this version to be published and agree to be accountable for all aspects of the work in ensuring that questions related to the accuracy or integrity of any part of the work are appropriately investigated and resolved.
ACKNOWLEDGMENTS
This research study was completed as part of the first author's capstone research project during his training at UCLA's M.S. Genetic Counseling Program. We deeply appreciate all the participants for their willingness and time to take part in this study. We also thank the members of UCLA's Department of Medicine Statistics Core (DOMStat), Dr. Myung-Shin Sim and Yen-Lin Lai, for their statistical expertise. The research reported in this publication was supported by the Biomedical Informatics Program (BIP) at UCLA CTSI and funded by the National Institutes of Health (NIH) through the National Center for Advancing Translational Sciences (NCATS) Grant (#UL1TR001881). Funding for this research was provided by the UCLA M.S. Genetic Counseling Program. Positionality Statements: The authors would like to acknowledge that their views are skewed by their own personal, cultural, and psychosocial identities, which may result in biases in their writing and interpretation of the current literature and of their results. Jonathan Lin, MS, is an Asian Canadian, cisgender male who uses he/him pronouns and was raised in Ottawa, Canada. At the initial time of writing this manuscript, he was a Masters' student at UCLA's genetic counseling graduate program. He was highly involved in teaching/education, as he served as a teaching assistant for almost 2 years. He has strong interests in cancer and neurologic genetic counseling, which he currently has a career in. Christina Palmer, PhD, MS, is a cisgender White woman raised in a US Midwestern state. She is a licensed genetic counselor involved in teaching/education and research. She has a strong interest in issues pertaining to diversity, equity, and inclusion in healthcare contexts. Rita Cantor, Ph.D., is a non-Latine White, English-speaking, cisgender female statistical geneticist who provided guidance on selecting and interpreting the statistical methods used to analyze these data. Her immersion in the non-Latine White culture in Los Angeles, California, may introduce bias into her interpretation of the results. Mariana Niell-Swiller, MS, is a licensed genetic counselor providing direction in cancer genetics clinical care at UCLA and is of Hispanic ancestry. Philip Sayegh, Ph.D., MPH is an Arab American, cisgender male who uses he/him pronouns raised in the San Francisco Bay Area. He is a licensed psychologist and clinical neuropsychologist involved in direct clinical service, teaching/education, clinical supervision, and research. He has a strong interest in issues pertaining to racial and cultural bias, prejudice, and social injustice in healthcare contexts.
CONFLICT OF INTEREST STATEMENT
The authors declare that they have no conflict of interest.
ETHICS STATEMENT
Human studies and informed consent: The study was approved by the UCLA Medical Institutional Review Board 3 (IRB#22-001115) with a HIPAA general waiver for extraction of eligible participant emails from the UCLA Health electronic medical record (EMR) by the UCLA Clinical and Translational Science Institute and waiver of signed informed consent for the study survey. A study information sheet was included with the recruitment email and consent was implied for survey participation. All procedures followed were in accordance with US Federal Policy for the Protection of Human Subjects.
Animal studies: No non-human animal studies were carried out by the authors for this article.
Open Research
DATA AVAILABILITY STATEMENT
Data collected for this study are not shared.
