Do patients with mental health and substance use conditions experience discrimination and diagnostic overshadowing in primary care in Aotearoa New Zealand? Results from a national online survey

Content and development

The survey consisted of four main sections: mental health and addiction service use; physical health service use (including primary care and other health services); stigma and discrimination; and demographics (including MHSUC diagnoses).

Respondents were asked if they had used primary care services for physical health issues in the last 5 years. This timeframe was chosen to provide a sizeable period for which recall of experiences should still be robust. In addition, health practice has not changed considerably over this period.

In their interactions with primary care services, respondents were asked how often they experienced the following aspects of service quality:

Responses were gathered through a five-point Likert scale – always, most of the time, sometimes, never, unsure.

Questions about relationship quality and discrimination were adapted from New Zealand’s Adult Primary Care Patient Experience Survey,³¹ a large national survey of patients attending primary care services, which has been extensively tested and reviewed.³² The question on diagnostic overshadowing was developed in consultation with a study advisory group, as no existing validated questions were found.

Demographic questions included age, gender, ethnicity and sexual orientation. Respondents were asked to indicate MHSUC diagnoses they had received from a list (depression, anxiety, schizophrenia, bipolar disorder, personality disorder and addiction). Multiple responses were allowed, with an ‘Other’ option for diagnoses not listed. ‘Other’ options were manually coded into new or existing categories.

The survey questionnaire was reviewed by the advisory group, which included clinicians and people with lived experience of MHSUC, and pre-tested with a Māori male with lived experience of MHSUC.

Recruitment and eligibility

The survey was fielded from 31 January to 1 April 2022. It was distributed through email, Facebook and Twitter by the researcher team, the research advisory group, Equally Well (a network of organisations and individuals working to improve physical health outcomes for people with experience of MHSUC),³³ and the University of Waikato student association. Paid advertising on Facebook was used from 17 to 31 March 2022. Information and invitations to participate were included in health sector newsletters, and sent to email distribution lists through Housing First, non-governmental addiction services and Māori health providers.

Respondents were eligible if they had accessed primary or secondary health care services for a MHSUC in the past 5 years, had engaged with any healthcare service for a physical issue in the past 5 years and were >18 years. From 488 visitors to the online survey, 408 agreed to participate. People who only partially completed the survey and did not answer any questions about provision of physical health services were excluded, as was one duplicate response. The final dataset comprised 354 eligible individuals. The analysis sample for this paper included only those who had used primary care services for physical health (n = 335).

Survey responses were anonymous unless respondents provided contact details for a follow-up interview. This personal information was stored separately from the survey data, in secure folders protected by the University of Otago digital system, which only researchers could access.

Data analysis

As data were from a convenience, self-selected sample, only descriptive and unweighted statistics (sample proportions) were calculated using Microsoft Excel version 2205 (Microsoft Corporation, Redmond, WA, USA). Relationships between variables were analysed using chi-squared tests because of the categorical nature of the data. Respondents were not forced to answer any question so there were variables with missing data due to incomplete or partial responses.

For demographic questions, chi-squared tests were performed for the demographic and service quality questions answered. For diagnosis and discrimination questions, comparison groups were those not reporting that diagnosis or that type of discrimination. Only groups with at least 20 people were tested for differences. The significance threshold was P < 0.05. Questions about demographic characteristics and MHSUC diagnoses were asked at the end of the survey, and around 12% of respondents did not complete these sections, although they contributed to earlier survey questions.

For service quality questions, the ‘always’ and ‘most of the time’ responses were combined into one category, as were the ‘sometimes’ and ‘never’ responses. ‘Unsure’ responses were excluded from analyses. Number of diagnoses was created by a count of all indicated diagnoses from each respondent. Schizophrenia (n = 15) was combined with bipolar disorder (n = 51) and reported diagnoses of n < 20 in total were not included (eg obsessivecompulsive disorder, Asperger’s syndrome, autism, attention deficit hyperactivity disorder, eating disorder, dissociative disorder).

Ethics approval was granted by the Southern Health and Disability Ethics Committee (reference: 21/STH/216). Consent was assumed by engagement with the online survey; information about the survey, details about privacy and confidentiality and contact details for support was provided in the introduction to the survey.

Comparison with other literature

Māori respondents comprised 17% of the sample, comparable to 2018 Census data (16.5%).³⁹ The proportions of respondents identifying as LGBQA+ and gender diverse were higher than in population-based surveys,^40,41 which could be due to significantly higher rates of MHSUC in these groups.^42,43

From New Zealand’s Adult Primary Care Patient Experience Survey (APCPES) in February 2022,³¹ 95% of patients reported that their primary care professional (Yes, definitely) treated them with respect (94% for Māori, 95% for female) and 92% reported being listened to (Yes, definitely) (91% for Māori, 91% for female).⁴⁴ Although we cannot test for significant differences between the APCPES and our survey, people with MHSUC reported worse experiences and sizeable differences for some diagnoses. For example, in our survey, only 61% of those with bipolar/schizophrenia felt listened to. From other surveys, 17–31% of people with MHSUC have reported discrimination in physical health settings, which is comparable to our study.⁴⁵

The importance of relationships for people with MHSUC attending primary care has been explored in other research. In qualitative interviews with people with MHSUC, lack of knowledge and stigma from providers were identified as barriers to accessing care, whereas positive relationships, characterised as empathetic, non-judgmental and person-centred, enabled access.⁴⁶ Having a regular GP, with the associated benefits of continuity of care and building a trusted relationship, leads to greater satisfaction with primary care services.⁴⁷

Strengths and limitations

This survey fills a gap in experiences of physical health care in general practice settings from the perspective of people with MHSUC in NZ. However, as for most web-based surveys, we were unable to define our target population and calculate a response rate. The sample size was relatively small, meaning that sub-group comparisons for Pacific, Asian and gender-diverse people were not feasible. Respondents reported a range of MHSUC diagnoses, but were not necessarily a representative sample and those without internet access would be excluded. Self-selection bias, where people interested in the topic are more likely to participate, could have affected results, particularly if those with negative experiences of primary care responded at a higher rate. We were unable to distinguish the impact of other types of discrimination, such as racism, sexism, bias against LGBQA+ people, on experiences of care.

Implications for practice

Experiences of discrimination from clinicians, being ignored and disrespected and having physical symptoms overlooked deter people with MHSUC from seeking health care and thus contribute both directly and indirectly to poorer health outcomes.^10,48 Interventions to improve health outcomes in people with MHSUC often focus on patient factors, such as lifestyle interventions and optimised management of both mental and physical health. However, such strategies need to be multi-pronged and incorporate health service delivery, workforce development and training, national and local policies and social support.^10,49

In primary care, ensuring that all practitioners are trained, skilled and comfortable with interacting and supporting people with MHSUC is not only foundational for improving health outcomes, but also for reducing stigma and discrimination towards people with MHSUC.^11,14,45 For clinicians (including trainees), implicit bias tests can be useful in raising awareness, but concrete actions are needed to effect changes in discriminatory practice.^11,50 Programmes are more likely to be successful if they cater to specific workforce groups, and include features such as social contact with people living well with MHSUC and an emphasis on recovery, to counter therapeutic pessimism.^11,51

The APCPES provides an opportunity for ongoing measurement of the experiences of people with MHSUC, although it currently does not report on quality measures separately for people with MHSUC. At a practice level, systematically gathering both quantitative and qualitative patient feedback can be used to drive service improvement, but requires leadership, active engagement of patients and commitment by staff to enacting change.⁵²

The high level of diagnostic overshadowing reported in this survey requires more research into how to best quantify this experience and how it can be reduced. A future publication will analyse qualitative data collected in this survey, to further understand how people with MHSUC experience diagnostic overshadowing and discrimination.