Investigating the association between experiencing discrimination in healthcare settings and avoidance of healthcare services among Pacific Rainbow+ in Aotearoa New Zealand

Pacific peoples are estimated to make up 8.9% of the Aotearoa New Zealand population.¹ While Pacific peoples are an integral part of New Zealand society, alongside Māori, they consistently experience poorer health and wellbeing outcomes than other ethnic groups such as Pākehā (New Zealand European) and Asian New Zealanders.² Much research identifies multiple factors underpinning persistent disparities in health and wellbeing among Pacific peoples and Māori in Aotearoa New Zealand, with both structural and interpersonal racism predicting adverse outcomes, which negatively impact healthcare service use.^3,4

Discrimination (in multiple forms) negatively impacts health outcomes and healthcare avoidance among different minority groups.^5–7 Rainbow+, an inclusive term used in Aotearoa New Zealand to encompass gender- and sexuality-diverse individuals who identify as, or a combination of, (L)esbian, (G)ay, (B)isexual, (T)ransgender, (Q)ueer, (I)ntersex, (A)sexual, and in the Pacific context (M)āhū, (V)akasalewalewa, (P)alopa, (F)aʻafafine/faʻatama, (A)kavaine, (F)akafifine, (F)akaleiti/leiti (LGBTQIA+ MVPFAFF+) consistently experience multiple forms of discrimination in schools, workplaces, places of worship, at home and when seeking healthcare services.^8–12 Experiencing overt and subtle forms of discrimination (specifically homophobia/transphobia and heterosexism) has negative impacts on the health and healthcare-seeking behaviours of Rainbow+ individuals.^13–15 In Aotearoa New Zealand, one in 10 Rainbow+ young people (14–26 years old) reported they had been treated unfairly by a healthcare professional because of their Rainbow+ identity, with gender-diverse Rainbow+ young people reporting a rate 3.5 times higher than cisgender sexuality-diverse respondents.⁸ For gender-diverse Rainbow+ individuals, data from Counting Ourselves report over a third of respondents (36%) avoided seeking out healthcare services due to worry about disrespect or mistreatment as a gender-diverse (transgender or non-binary) person.¹² Furthermore, Pacific Rainbow+ youth who completed the Youth 19 survey were more likely to forego healthcare than their Pākehā counterparts,¹⁰ and nearly half of gender-diverse youth who completed the Identify survey reported facing difficulties in accessing healthcare services.⁸ Sixty percent of Pacific Rainbow+ individuals that completed the Manalagi Survey indicated they had experienced discrimination in a healthcare setting in Aotearoa New Zealand, with discrimination associated with race or ethnicity (racism) more highly reported than discrimination associated with their Rainbow+ identity.¹⁶

International literature demonstrates how gender-diverse individuals avoid healthcare due to perceived discrimination, seek medication without doctor’s supervision, find it difficult to access relevant medications or do not disclose their identity to their doctors.^17–19 In the United States, some Rainbow+ individuals avoid healthcare due to anticipated discrimination related to their gender identity.²⁰ This pattern of avoidance is also highlighted in studies that investigate healthcare interactions experienced by gender-diverse individuals, where such negative interactions compel them to avoid seeking healthcare altogether.²¹ This avoidance leads to delays in necessary care and can have adverse outcomes for patients and families.²² The mental health of Rainbow+ people, in places like North Macedonia as an example, is hindered by societal stigmatisation and discrimination, which exacerbates barriers to accessing mental health services.²³

Considering local and global research findings highlighting the negative impact discrimination has on Rainbow+ health outcomes and healthcare-seeking behaviours, and given that local research shows that both Pacific and Rainbow+ individuals in Aotearoa New Zealand may experience discrimination in healthcare settings, we hypothesise an association between reported experiences of discrimination in healthcare settings and healthcare avoidance. In that, higher levels of reported discrimination experienced in healthcare settings by Pacific Rainbow+ individuals in Aotearoa New Zealand are likely to be associated with healthcare service avoidance.

Methods

The Manalagi Survey instrument

Data used to test our hypothesis are derived from the Manalagi Survey.¹⁶ The Manalagi Survey design, research procedure and associated documents were reviewed and given ethical approval by the Southern Health and Disability Ethics Committee under approval number 2021 EXP 10986. The Manalagi Survey—structured around pillars of Pacific wellbeing (family as foundation, physical health, mental health, spiritual health, other identity factors, with culture as the roof engaging with context, time and space) articulated in the Fonofale Model of Pacific Wellbeing²⁴—was co-designed with Pacific Rainbow+ individuals across Aotearoa New Zealand through 11 community consultation meetings held in eight cities.^16,25,26 The survey was administered online and was available for Pacific Rainbow+ individuals and allies to complete from February 2022 to August 2022. Those who indicated that they were allies were branched to a separate survey for friends, family and allies. Criteria for inclusion in the Manalagi Pacific Rainbow+ sample required respondents to be able to give informed consent, be at least 15 years of age, be a resident in Aotearoa New Zealand at the time, and identify as both a Pacific person and as sexuality- or gender-diverse.

Participants

The survey collected 757 preliminary responses. The final sample for this study comprised 396 respondents after excluding those who did not complete beyond the demographic section (n=91); did not indicate their sexual or gender identity (n=7); identified as allies (n=234); were 14 years old or younger (n=3); were not currently residing in Aotearoa New Zealand (n=10); and did not have Pacific whakapapa (genealogy) (n=16). For the purposes of this study, Māori were not included. While Māori are part of the Pacific family, they hold tangata whenua and mana whenua status as the Indigenous people of Aotearoa New Zealand; thus, research pertaining to Māori Rainbow+ communities is regarded as a distinct domain that should be led by Māori. As non-Māori, it is not appropriate nor ethical for this research and research team to claim, use or publish Māori data in the Aotearoa New Zealand context.

Table 1 presents the demographic details of the Pacific Rainbow+ respondents categorised by sexuality-diverse respondents labelled as 1) cisgender sexual minorities (n=239), and gender-diverse respondents labelled as 2) transgender (n=60) and non-binary (n=66). All respondents were counted only once within these categories. Most participants fall within the young adult and adult age ranges, come from large cities like Auckland and Wellington and earn below NZ$59,999.

View Table 1–4.

Measures

Demographics

Rainbow+. Participants were asked, “Do you identify as part of the Pacific Rainbow+ community?” Response options provided were “yes” or “no”. We then asked those who selected "no" if they were “questioning”, “an ally” or “neither”. Participants who were questioning their identity were grouped as part of the Rainbow+ sample, while those who identified as “an ally” or “neither” were excluded.

Gender. Participants were asked two sets of questions about their gender. First, a multi-select, single-item question: “Which of these statements apply to your gender identity?” Response options included “I am cisgender”, “I am transgender”, “I am non-binary” and “None of these apply”. Participants were categorised into three categories: cisgender, transgender or non-binary. A total of 77 participants selected “None of these apply”. Speculatively, it is possible that this high number is due to respondents not understanding the cisgender/transgender terminology or not wanting to conform to a prescribed Western gender classification, as much has been written on the incongruence between Indigenous Pacific Rainbow+ gender expressions (cultural identities) and the LGBTQIA+ classification (sexuality and gender diverse).^25,26,28

In cases where respondents selected “None of these apply” or it was unclear if a respondent should be categorised into one of our pre-determined categories, we referred to the response for the open-ended question “How do you describe your gender?” For example, respondents who selected “None of these apply” but wrote responses such as “Woman mostly, but gender fluid” and “Fa’afafine” were classified as non-binary. Respondents who were cisgender but categorised as Rainbow+ due to being sexuality diverse (LGB) were grouped as cisgender sexual minorities.

Pacific whakapapa (genealogy). Participants were asked, “Which Pacific Island group(s) do you identify with/whakapapa to? Select as many as apply.” The options provided were: Cook Islands, Fiji, Kanaka Maoli (Hawaii), Kiribati, Niue, Papua New Guinea, Rotuma, Samoa, Solomon Islands, Tahiti, Tokelau, Tonga, Tuvalu, Vanuatu and not listed. Examples of non-listed Pacific Islands include Pitcairn, Marquesas and Norfolk. These responses were collated into a nominal variable to be used as a covariate in the regression models. A “multi-ethnic” category was created to encompass those who selected more than one Pacific ethnicity.

Region. We asked participants, “Where in Aotearoa New Zealand do you live?”

Income. We asked participants, “What is your approximate personal annual income?”

Predictors

Healthcare discrimination: Participants were asked, “Have you ever experienced any of these types of discriminatory behaviours at your general practitioner’s (GP’s) office, a medical clinic or hospital in Aotearoa New Zealand before? Check all that apply.” A definition was provided for “microaggressions” to indicate “the everyday, subtle, intentional—and oftentimes unintentional—interactions or behaviours that communicate some sort of bias toward historically marginalised groups.” Refer to Table 2 for the classified response options for different forms of discrimination: ethnic/racial, gender and other.

We generated an “Overall discrimination” variable by summing all reported experiences of discrimination in a healthcare setting. The index ranges from zero to 15. A response of one indicates that participants reported experiencing one of the discriminatory behaviours assessed. In Table 2, we report the average number of discriminatory experiences for each gender group.

Outcomes

Healthcare avoidance. We asked participants, “Have you ever avoided seeing your GP, or going to a medical clinic or hospital because of fear for your safety or repercussions in any way (e.g., discrimination, violence or visa status, etc.)?” This question was adapted from the Counting Ourselves¹² and Identify⁸ surveys, with final wording confirmed by Pacific Rainbow+ communities in survey testing.¹⁶

Seeking mental health support. We asked participants, “In the last 12 months how many times have you sought help from mental health support services (i.e., counsellors, helplines, etc.)?” We adapted this question from Honour Project Aotearoa²⁷ and the Counting Ourselves Survey,¹² with final wording approved through community consultation and testing.¹⁷ The responses were binarised as “yes” or “no”.

Data analysis

All statistical analyses were conducted in IBM SPSS Statistics v30. First, we conducted descriptive analyses of the predictor and outcome variables for the two respective gender groups. We strategically limited our analysis to two gender groups in this study due to the low number of affirmative responses observed in preliminary analyses for predictor variables specific to transgender and non-binary sub-groups. This decision was made to avoid overestimating odds ratios, which could introduce bias resulting from small sample sizes.²⁹ Subsequently, multiple sets of multivariate logistic regression analyses were performed to ascertain the extent of association between each predictor on various types of discrimination, and the two outcome variables: healthcare avoidance and mental health support. These analyses were conducted using generalised linear models and adjusted for the effects of demographic variables (age, Pacific whakapapa, region and income). Regression analyses were undertaken only for variables with at least 20 participants reporting a discriminatory experience, to guarantee sufficient statistical power for identifying a true difference. Statistical significance was determined at an alpha level of p<.05.

Results

Table 2 outlines the patterns of health and mental health care utilisation and experiences. Nearly one in five cisgender sexual minorities, and more than a quarter of transgender and non-binary respondents, have avoided accessing healthcare due to fears for their safety or potential repercussions. Multivariate findings in Table 3 indicate that Rainbow+ participants (both cisgender sexual minorities; transgender and non-binary) who had experienced ethnic/race-based discrimination including being treated differently from their Pākehā counterparts (p<.001) and experiencing ethnic-based microaggressions (p<.01) had increased odds of avoiding health services.

Transgender and non-binary respondents with prior exposure to misgendering in healthcare settings were significantly more likely to avoid seeking healthcare services (p<.05). Cisgender sexual minority respondents who had experienced microaggressions based on their presenting characteristics (p<.01), or had their symptoms minimised by healthcare staff (p<.01), also reported elevated odds of healthcare services avoidance. Transgender and non-binary respondents who felt they were treated differently from their cisgender counterparts (p<.001) or had their symptoms minimised (p<.001) were significantly more likely to avoid healthcare services.

Cisgender sexual minorities respondents reported experiencing 1.15 instances of the listed forms of discrimination, while transgender and non-binary respondents reported nearly double with 2.18. When the overall number of discriminatory exposures increased by one point, it was estimated that both cisgender sexuality minorities as well as transgender and non-binary respondents had an approximate 60% increased likelihood of avoiding healthcare.

Nearly one-third of cisgender sexual minorities respondents, and more than one-third of transgender and non-binary respondents, sought mental health support in the past year. For cisgender sexual minorities, those who had experienced race-based microaggressions (p<.05) were significantly more likely to report having sought mental health support (see Table 4). Transgender and non-binary respondents who reported ethnic (p<.01) or race-based (p<.001) microaggression had significantly higher odds of seeking mental healthcare services. Further, transgender and non-binary participants who faced symptoms being minimised reported increased odds of mental health care service utilisation (p<.001). A one-point increment in discriminatory exposure predicted a 60% heightened likelihood for transgender and non-binary respondents to seek mental health care services.

Discussion

Our multivariate regression analysis demonstrates that discrimination (racism and homophobia/transphobia and heterosexism) reported by both Pacific cisgender sexual minorities and transgender or non-binary Rainbow+ individuals in Aotearoa New Zealand within a healthcare setting is associated with increased healthcare avoidance. These results align with international literature reporting Rainbow+ groups experiencing high levels of discrimination in healthcare settings (Australia),¹⁸ as well as similar associations among Rainbow+ individuals reporting incidences of discrimination in healthcare settings and increased healthcare avoidance in Thailand (gender-affirming healthcare avoidance)¹⁷ and the United States.¹⁹ However, some measures (i.e., minimisation of symptoms) may be associated with both racism and homophobia/transphobia and have not been meaningfully disentangled in our analysis. The most significant predictors of healthcare avoidance among Pacific Rainbow+ related to being treated differently from Pākehā and experiencing microaggressions based on ethnicity and race.

For transgender and non-binary respondents, other predictors of healthcare avoidance were more significant than for cisgender sexual minorities; namely, misgendering by staff and a belief that they were treated differently than cisgender or straight-presenting patients. While having symptoms minimised could be associated with healthcare avoidance, transgender and non-binary respondents reported significantly higher odds ratios (statistically significant) of seeking mental health care based on minimisation of symptoms than cisgender sexual minorities respondents (not statistically significant).

Respondents were also more likely to access mental health services if they had experienced ethnic or race-based discrimination. This is an important finding, as it highlights the way discrimination impacts Pacific Rainbow+ individuals on two axes of difference: race/ethnicity and sexuality and/or gender diversity. Furthermore, our findings indicate that experiences of discrimination have a cumulative effect on healthcare avoidance for Pacific Rainbow+. An increment of a reported discriminatory experience—based on gender or ethnicity/race—could increase the odds of healthcare avoidance by nearly 60%.

Limitations

In the present study, we were required to collapse the diverse gender identities within Pacific Rainbow+ communities into Western categorical frameworks due to analytical constraints, including a small sample size. As a result, we were unable to meaningfully examine gender-group differences within our sample, despite existing research indicating such differences in healthcare utilisation.¹² Further, our regression findings are limited by the cross-sectional nature of the data, which restricts our ability to infer causality. While our results suggest that experiences of discrimination may contribute to healthcare avoidance and increased use of mental health services, alternative explanations remain plausible. For example, it is equally possible that individuals within Rainbow+ communities who avoid healthcare and rely more heavily on mental health services are more likely to report discrimination in healthcare settings. Nevertheless, this reverse hypothesis continues to underscore the critical importance of addressing ethnic and sexuality as well as gender diversity-based discriminatory practices experienced by Pacific Rainbow+ communities within healthcare settings.

Conclusion

Our findings affirm research that details high levels of discrimination experienced in the lives of Rainbow+ individuals both in Aotearoa New Zealand and abroad. For Pacific Rainbow+ respondents in this study, experiences of discrimination (racism, homophobia/transphobia) had a negative predictive effect on healthcare service use. Further analysis shows that discrimination related specifically to race and/or ethnicity had a stronger predictive effect than sexuality- and gender-based discrimination. For transgender and non-binary respondents, being treated differently from cisgender patients was also a strong predictor of healthcare service avoidance. Our analysis also shows that experiences of racial and ethnic based discrimination (racism) also increased the likelihood of mental health service usage among respondents. While findings of the study are limited by the cross-sectional nature of the data and the statistically strategic decision to collapse Pacific gender and sexuality diversity into narrower Western gender and sexuality categories, the intersectional complexity of the Pacific Rainbow+ experience with the Aotearoa New Zealand healthcare system needs further investigation and research. This will help provide further nuanced data and evidence to better support this multiply marginalised cohort’s comfortability with, and usage of, services provided by the Aotearoa New Zealand healthcare system.