A content analysis on the perceptions of LGBTQ+ (centred) health care on Twitter

Abstract Background LGBTQ+ individuals have experienced many barriers to receiving quality health care, but the worldwide implementation of person‐centred care should make a positive change. However, as forthright disclosures are difficult to find using traditional methods, novel approaches should be utilized to uncover opinions and experiences on LGBTQ+ health care. Twitter could be a place where people post on this topic. Aim This study aimed to explore tweets mentioning LGBTQ+ (centred) health care. Methods The methods consisted of an explorative qualitative content analysis of tweets. The tweets were collected between 26 February and 30 March 2021, resulting in 2524 tweets of which 659 were relevant for content analysis. Results The results showed an excess of political tweets involving LGBTQ+ health care. Many tweets included general statements on the need for LGBTQ+ health care. The few tweets on personal experiences in LGBTQ+ health care showed the overwhelming need for quality care that has been made difficult by political developments. Conclusion Most tweets were made to inform others of the necessity of quality health care for LGBTQ+ individuals, but the utilization of person‐centred care is hardly noticeable. Public Contribution This study was conducted with the involvement of a public partner (second author) who contributed to the design, data analyses and writing of the paper. Moreover, this study involves the analysis of data provided by the public and published on social media.

much higher. 4 A large part of the problem is the lack of competency from healthcare professionals to understand personal beliefs, specific issues, and the importance of an inclusive environment. 3 Two reviews concluded that healthcare professionals were aware of their limitations in providing appropriate care to LGBTQ+ individuals and expressed that a person-centred approach to healthcare would increase the quality of care for these patients. 4,5 In recent decades, person-centred care (PCC) has become an increasingly discussed concept in health care. Healthcare professionals, patient organizations and policymakers increasingly aspire to PCC to optimize cost containment and quality of care. 6 PCC is directed at improving the health and recovery process of patients by forming a partnership between the patient (often with relatives) and the healthcare professionals. A central challenge of PCC is how to incorporate the needs of everyone, including the underrepresented voices. When a PCC approach is applied in care for LGBTQ+ individuals, the healthcare professional can understand LGBTQ+ identity development from the perspective of the individual. 7 PCC therapy includes a focus on interpersonal and intrapersonal difficulties faced in health care such as the experience of stigma. 8 PCC is an umbrella term to cover an array of different concepts and terms that are often used interchangeably: patient-centred care, client-centred care, and family-centred care. [9][10][11] This plurality indicates that researchers, healthcare professionals, and patients may not have the same understanding of PCC.
Tweets can hold unexpected potential for exploring frank opinions on contested issues. Twitter is a supply-based platform on which people offer information and present their opinions to connect with like-minded people. 12,13 LGBTQ+ individuals and their allies use Twitter to discuss health and social needs relevant to the population. 14 This social network platform is used to find news feeds and content relevant to the LGBTQ+ identity that can be difficult to find in offline settings. 15 The unfiltered nature of Twitter can be a valuable source of insight into the honest opinions of individuals who tend to feel left out when it comes to health. 16 Twitter contains vast amounts of freely available, user-generated microblogs and provides real-time monitoring of public health topics in an efficient and automated manner. Eysenbach defines this monitoring as 'Infoveillance' which is 'the science of distribution and determinants of information in an electronic medium, specifically the Internet, or in a population, with the ultimate aim to inform public health and public policy'. 17, p.2 Infoveillance may have many benefits for the field of LGBTQ+ health care as it could provide new insights from voices that would otherwise go unheard. The analysis of the LGBTQ+ health care contents of the tweets can offer both a better understanding of how it is expressed and an indication of why the tweet was made.
Tweeting can be considered a performative platform to selfidentify. 18,19 By interacting or simply spreading posts, a performance of self is provided for an imagined audience. 13 Self-disclosures of personal emotional messages receive the most positive response from the imagined audience. 20 A study on the #meToo movement in tweets has shown that individuals tweet about their experiences with as much detail as the microblogs of 280 characters will allow. 21 An empirical study on the Nigerian Queer community showed expressions of their sexualities, as well as purposive identity formations, manifest in both in-and out-group advocacies. 22 Based on these studies, presenting experiences of health care from the LGBTQ+ community could be found on Twitter. Moreover, the literature has shown how important the experiences of healthcare professionals with LGBTQ+ individuals are and therefore more perspectives need to be considered in this study to portray the perceptions of LGBTQ+ health care.
To get a better perspective on what people are posting about LGBTQ+ (centred) health care, only the content of the tweet is used.
This approach allows for the exploration of this topic within tweets. The purpose of this study is to explore the tweets mentioning LGBTQ+ health care on Twitter and uncover if person-centredness appears in any of these tweets.

| METHODS
This is an explorative qualitative content analysis study on Twitter.

| Data collection
The Twitter search was conducted in the Twitter Application Programming Interface (API) called Mozdeh. 23 The search terms The API generated 2524 tweets consisting of 1578 original tweets (although many copies with the same texts), 500 original comments (these start with @####), 266 replies, and 180 retweets.
The retweets were removed from the sample as the content was made by someone else. These remaining tweets tallied a sample of 2344 tweets for the content analysis. This sample contained 167 tweets in the Spanish language which were translated by the second author who is a native speaker. 59 tweets only contained a reference to pages or usernames and were eliminated. The remaining 2285 tweets could be further analysed. The flowchart ( Figure 1) shows the steps in the data collection and analysis. VAN

| Content analysis
The content analysis was performed following the guidelines described by Graneheim and Lundman. 24  The tweets were then read through again based on their classification by both authors. The tweets need to be presented in a logical order which was achieved through categories going from universal (LGBTQ+ health beyond health care) to specific (personcentred care). Both authors discussed the analysis, and disagreements were settled through negotiated consensus. The confirmability was improved by using tweets to demonstrate the grounding of the findings in the data. Table 1 presents an example of the coding scheme.

| Ethical considerations
Twitter data is supply-based in that the researchers can only access what others have publicly presented on the platform. When people create a F I G U R E 1 Exclusion diagram Twitter account, they must agree to the terms and conditions that state that their output can be used by third parties. As Twitter users are often unaware that they agreed to this, certain steps are taken to prevent the Twitter user from harm. There has been much debate and different approaches to the portrayal of tweets in research. Some guidelines dictate that researchers should always ask for consent to quote the tweets and paraphrase all other tweets. 25 However, there is a selection bias when applying consent 26 and paraphrasing would contain a level of interpretation that takes away from the word choices of theTwitter users.
Therefore, the authors of this study decided to copy the tweets directly from the API to preserve the expression of Twitter users. However, this paper follows the general guidelines for preventing direct identification as most Twitter users preferred. 26 The focus needs to remain on the perceptions of LGBTQ+ health care which is achieved by replacing all usernames and in-text references with hashtags. Another good practice is not connecting the tweet to a specific profile. 26 The Twitter user could (possibly) be found when the tweet is searched for online, but this would only lead to the public profile of the Twitter users. The online and offline persona do not need to be the same. 18,19,27

| RESULTS
This exploratory supply-based study was organized into the following categories that displayed perceptions of LGBTQ+ (centred) health care in the Twitter sample.

| LGBTQ+ health beyond healthcare
The initial read-through of the tweets showed an overrepresentation of political tweets, spearheaded by tweets on the Equality Act

| Health care for LGBTQ+ individuals
Many of the tweets concern trans health in particular. The bills, such as the one in Arkansas, are also specific to providing treatment for gender dysmorphia. Taking this treatment out of the health insurance is a direct attack and lack of acknowledgement of the necessity of quality health care and as this tweets exemplified here:

| Mental health
Something that is touched upon in these and many other tweets is the mental health care necessity of LGBTQ+ individuals as the suicide rate among this group is up to four times higher than their peers. 28 Specifically, access to the right care is mentioned: @##### @###### mental health care is already so taboo here but for people from lgbtq + communities….
yeah. it's really scary to not be able to access available resources because they might not be accepting for you and. I am still way more privileged then so many. Yeah I fear for my daughter and her LGBT+ friends and their access to appropriate medical and mental health care that is responsive to their needs. Add to it she's mixed race and it's even more complex.

| Person-centred healthcare
The last section of the results goes into the secondary aim of uncovering Twitter after the implementation of the Equality Act in the USA. The results showed that quality health care for LGBTQ+ individuals is still far from sufficient, and this challenge needs to be addressed first before person-centredness can be found in the care. Optimistically, the increased access to quality care will coincide with increases in the implementation of more PCC in common practice to ensure that all people will receive care that is appropriate for their situation.
LGBTQ+ health-related tweets are mostly used to express a standpoint in the political debate and argue their perception through general understandings such as human rights and the Hippocratic oath.
The few examples that do disclose more personal content show how important access to quality health care and specifically PCC is for LGBTQ+ individuals.
With this study, the challenges came to light that LGBTQ+ health care is mentioned more to make a (political) standpoint than effectively addressing the inequality of care.