Patient Communication practices in conversations about sexual health in medical healthcare settings: A systematic review that

Objective: Many healthcare professionals experience difficulties in discussing sexual health with their pa- tients. The aim of this review was to synthesize results of studies on communication practices in interactions about sexual health in medical settings, to offer healthcare professionals suggestions on how to communicate about this topic. Methods: We searched for studies using five databases. Reference lists and specialist bibliographies were searched to identify additional studies. We included discourse analytic studies that used recordings of medical consultations. Results: We identified five studies that met the inclusion criteria. Findings were synthesized into seven categories of practices deployed by patients and healthcare professionals when talking about sexual health: avoiding delicate terms (1), delaying potentially delicate words and issues (2), using assumptive talk (3), generalized advice-giving (4), deploying patients’ talk (5), depersonalization (6), and patient-initiated advice (7). Conclusion: Practices indicate the delicacy associated with discussing sexual health issues, but results also shed light on practices that can help professionals to deal with this delicacy, and to be responsive to patients’ needs and concerns. Practice implications: Findings will assist healthcare professionals in broaching topics related to sexual health so they can help patients deal


Introduction
According to the World Health Organization, sexual health is 'a state of physical, emotional, mental, and social well-being in relation to sexuality; it is not merely the absence of disease, dysfunction or infirmity'. Thus, sexual health is an essential part of overall health and general well-being [1]. In line with this definition, van Lunsen and Laan [2] define sexual health as 'the ability to have a pleasurable sex life, to adapt and self-manage it in the face of social, physical, and emotional challenges in different phases of life' (p 181). Particular diseases, medical conditions and its treatments, such as cancer [3], Parkinson's disease [4], and other chronic conditions [5], constitute such challenges. To provide patients with the necessary care to help them deal with these challenges and consequently contribute to their general wellbeing, it is essential that the discussion of sexual health is seen as a routine part of healthcare [2].
Although healthcare professionals acknowledge the importance of having conversations with patients about sexual health, discussing this can be problematic for them [6,7]. Healthcare professionals treating patients in various settings cite barriers to communicating about sexual health [7][8][9][10]. For example, general practitioners and practice nurses in primary care felt limited in their ability to address sexual issues, assuming this would require specialist knowledge and skills [8]. Professionals treating patients with chronic illness reported that discomfort, lack of confidence, and patient factors such as cultural norms and values were important barriers in discussing sexual health [10]. Another salient obstacle is a lack of insight into ways in which such a delicate theme may be handled in talk [7,8].
Research has shown that language use in discussing sexual health deserves attention because it can affect the quality of this conversation [11]. Over the last few years, efforts were made to provide recommendations on discussing sexual health in medical settings and developing communication training interventions. However, these recommendations and interventions are often based on self-reported attitudes and experiences of patients and professionals with regard to consultations [11][12][13]. Such self-reports do not demonstrate how the topic of sexual health was actually handled in consultations.
Discourse analysis enables us to gain a more profound understanding of what happens during these conversations. Its objective is not to identify underlying mental states, perceptions and motivations, but to analyse text and talk in order to uncover patterns in how discourse constructs our social world [14]. Although various forms of discourse analysis exist, our focus is on those where the sequential analysis is a core element (e.g. Conversation Analysis and Discursive Psychology). This methodological approach treats discourse as action-oriented, meaning that people often perform a range of actions through discourse [15]. Conversations between healthcare professionals and patients can also be conceptualized within this action framework. Patients, for example, may present a problem or justify the decision to seek medical attention. Healthcare professionals may give advice or recommend a treatment. In order to identify these actions, it is important to look at the practices for building turns (e.g. choice of words, intonation, and pauses). Practices are those features of a conversational turn that are distinctive, specifically located within a turn or sequence of turns [16]. These different (communication) practices are mobilized to accomplish diverse actions or fulfil various functions in talk [16,17]. By empirically studying the way in which participants treat each other's utterances, discourse analysts can capture the nature of these actions. For instance, by responding with "Oh, really" to the deliverance of a diagnosis by a practitioner, a patient treats the diagnosis as unexpected [18]. So, by studying talk in relation to sexual health on a micro-level, an understanding of the functions of that talk can be obtained [19].
To this date, systematic reviews of discourse analytic studies in healthcare settings [20][21][22][23][24] are scarce, and reviews in the context of sexual health are lacking altogether. A review of discourse analytic studies is especially useful for making recommendations on ways to carry out conversations in healthcare. In the present review, we gather evidence from discourse analytic studies of actual interactions in healthcare settings to provide professionals with suggestions on how to communicate about sexual health. This review is guided by a method specifically developed for systematically reviewing evidence from conversation analytic and similar discourse analytic studies [25]. With this aggregative approach, we aim to (1) identify communication practices that occur in interactions about sexual health in medical settings, (2) identify the effects of these practices, (3) provide healthcare professionals with suggestions on how to communicate about topics related to sexual health, and (4) identify gaps in current evidence.

Methods
With this systematic review we synthesized results from discourse analytic studies following a guide developed by Parry and Land [25]. This guide draws on already existing methods for conducting systematic reviews but is tailored to work with evidence from discourse analytic studies. For data extraction and quality appraisal we made use of templates specifically designed for this type of evidence [25]. We elaborated the template for quality appraisal (see appendix A) by adding components relating to characteristics of conversation analytic research [26], since these components allowed us to get a more extensive understanding of the quality of the included studies. In the following sections, we will outline our approach. We report in accordance with PRISMA guidelines [27].

Study selection
A search strategy was designed in cooperation with two information specialists of the HU University of Applied Sciences Utrecht. This search strategy consisted of search terms in the following four categories: sexual health, communication, healthcare settings, and audio and/or video recordings. In December 2019, we searched the following databases: PubMed, CINAHL Plus with Full Text, PsycINFO, Web of Science, and Communication and Mass Media Complete. The date of publications was not restricted. After searching the databases, one author (IK) removed duplicates and carried out a screening of titles and abstracts for potentially relevant articles. The remaining publications and additional publications were read in full and independently assessed by two authors (IK and PS). In case of doubt, discussions took place in order to reach consensus.
In addition, we made use of existing knowledge within our team to add potentially relevant publications. Online specialist bibliographies and reference lists of publications that were already found, were also screened to search for additional studies.
Studies were included when they: 1. use analysis of audio or audio-visual recordings; 2. use a discourse analytic approach with a focus on sequential analysis (Conversation Analysis, Discursive Psychology and related approaches); 3. rely on naturally occurring interactions in healthcare settings; 4. focus on the discussion of sexual health in relation to the medical condition of patients; 5. are peer-reviewed papers or book chapters written in English.

Data extraction
Data were extracted from publications independently by two authors (IK and PS), based on the data extraction template by Parry and Land [25]. Because two components of the template did not apply to our study, these were deleted. This led to the following information being extracted from each paper: publication details (1), country in which data were collected (2), applied methodological approach (3), setting (4), number and characteristics of participants (5), number of speakers in an interaction (6), size of dataset (7), type of recording (i.e., video and/or audio) (8), and characteristics of conversations studied (i.e., detailed information about communication practices) (9).

Data synthesis
We started our synthesis with comparing a finding reported in one study to similar findings in other studies. This process was repeated until all findings were synthesized into categories of practices that shared similarities in their functioning. Given the limited number of five studies included in this review, we decided to include all findings in the synthesis, even those practices that were unique to just one study.

Quality appraisal
The distinctive characteristics of discourse analytic studies, being non-numerical and non-statistical, require a different approach to quality assessment of studies than quantitative studies. Furthermore, findings are also not compatible with standard qualitative results as the objective of discourse analysis is not to identify underlying mental states, perceptions and motivations, but to uncover patterns of communication. Our adaptation of the quality assessment template recommended by Parry and Land [25] consisted of an assessment of (1) type and amount of data and of (2) detail and depth of analysis (see Appendix A). Criteria for the detail and depth of analysis involve 1) the use of principles of Conversation Analysis (e.g. verification that analysis includes examination of more than one party's conversational turns); 2) a deviant cases analysis, based on the principle that parts of the conversation that ostensibly deviate from established patterns may show other characteristics that indicate that a speaker interprets the regularities as normative; 3) a judgement of the degree to which findings show coherence with previously established findings; 4) inclusion of controllable authentic data such as transcripts; 5) inclusion of high ratio of analysis to description of the extracts; 6) the extent to which analysis is grounded in the data; 7) and a judgement of the extent to which the analysis is fine-grained. Based on these criteria, the quality of each study was individually assessed by two authors (IK and PS). Any discrepancies were discussed until agreement was reached.

Search results
Our search resulted in a total of 1734 studies. After removing duplicates, 1211 publications remained for the screening of title and abstract for potentially relevant articles. Eventually, 35 publications that seemed to meet the inclusion criteria were read in full. After assessing the full articles, 5 publications were included in the synthesis. The main reasons for excluding the other articles was that most of these studies did not use a discourse analytic approach or did not rely on naturally occurring interactions. Some studies that were found through our additional search did, however, use discourse analysis to examine talk. Yet, in these studies, talk about sexual health did occur but was not the main analytic focus. Therefore, these studies were not included. Fig. 1 depicts the workflow of the review process.
In all selected publications, a multi-case analysis of one-on-one interactions was carried out. The number of participants in each study was not always specified. Table 1 shows more information about the characteristics of the included studies.

Quality assessment
All studies complied with at least five of the seven quality criteria and exhibited a fine-grained analysis of interactions using substantial datasets, indicating that practices were identified based on a thorough examination of sufficient data. We therefore concluded that the overall quality of all studies is high.

Communication practices
Findings were synthesized into the following seven main categories of practices used in interactions about sexual health: avoiding delicate terms (1), delaying potentially delicate words and issues (2), using assumptive talk (3), generalized advice-giving (4), deploying patients' talk (5), depersonalization (6), and patient-initiated advice (7). In the section below, the categories are described, as well as the different ways in which the communication practices are accomplished. Additionally, their effects are described in detail, and findings are illustrated with examples. Several practices were observed in both patients' talk and healthcare professionals' talk. Others were observed exclusively in talk of either healthcare professionals or patients.

Avoiding delicate terms
Three studies [28][29][30] reported the avoidance of delicate terms relating to sexuality, such as 'vagina' [30]. One way in which healthcare professionals do this, is by discussing sexuality by using vague, general, and neutral terminology (see Table 2). It was found that when professionals asked questions in a general way (e.g. "and tell me how the sexual part of the relationship goes"), patients did not provide a conforming answer [29]. Neutral terms, however, carry few implications that can be perceived as delicate, enabling professionals to avoid terms with moral connotations, and consequently allowing patients to use their own terms [28]. Moreover, it allows professionals to only offer the amount of information necessary to convey the delicate issue, without getting specific [30]. One study [28] reported the use of neutral terms by patients when they described their sexual partners. Simultaneously, patients often talked about sexual activities indirectly, leaving it up to the professionals to draw conclusions about the type of activity patients were involved in.
Patients and professionals also avoid delicate terms with the use of pronominal reference by referring to these terms with words such as 'it' and 'that'. Since the delicate issues that are being referred to with pronouns were already introduced, albeit in a vague or general way, using pronouns oftentimes did not lead to interpretation problems [30].
The last strategy used by both professionals and patients, is the total omission of delicate terms (see Table 2). Although no detailed analysis of the effects of omissions was shown, Weijts et al. [30] mentioned that healthcare professionals do tend to adopt patients' omissions.
Generally, the avoidance of delicate terms by both parties shows the delicate character of the topic [28][29][30]. It is argued that by totally omitting delicate terms or by using vague terminology, healthcare professionals may attest to the stereotype of female sexuality being something 'dark and mysterious' [30]. These strategies are, therefore, not recommended. Yet, avoiding delicate terms can be done in ways that may not prove troublesome. Professionals may use neutral terms, and thereby invite patients to use their own terms. The use of neutral terms also allows professionals to avoid the invocation of potentially moral connotations [28]. Furthermore, using pronouns to refer to potentially delicate terms mostly does not lead to interpretation problems. However, professionals may want to be careful with an excessive use of pronominal reference, because some work is required in order to understand what is referred to [30].

Delaying potentially delicate words and issues
Both healthcare professionals and patients delay the production of delicate words and issues, for instance, by using markers preceding potentially delicate words and issues, including pauses, hesitations, speech perturbations, self-repairs, and laugh particles [28][29][30]. The use of markers by patients was prevalent in answers to questions about sex [29], when giving descriptions of a partner or relationship [28], and in talk about sexual behaviour [28,30]. Professionals employed speech perturbations when describing a patient or a patient's partner. This delayed description was sometimes preceded by indirect talk about sexual activities, indicating the reason for delay is the delicate nature of those activities [28]. Occasionally healthcare professionals would complete patients' sentences when they contained hesitations. Weijts et al. [30] concluded that this may be a way to collaboratively manage aspects that are potentially face-threatening to the patient. Professionals did not, however, use pauses and hesitations when asking routine questions about sexuality or giving general instructions that do not necessarily apply to the patient in question [30] (see Table 3).
Two other means of delay by patients were found [30]. The first one is delaying sexual issues by only introducing these matters at the end of the conversation. Professionals do, however, pay attention to the issues presented by patients. The other way in which patients delay using delicate words, is by just not answering potentially delicate questions. Occasionally, patients eventually gave short answers. In this way, it is up to the healthcare professional whether to ask for specification (see Table 3).
Overall, delaying potentially delicate issues indicates the delicacy of the topic being discussed [28][29][30]. By using delaying strategies, healthcare professionals show they recognize and appreciate the delicacy of certain topics, by which they actually construct their professionalism [28]. Nevertheless, professionals should be aware that delay can eventually result in not or only partly addressing sexual health problems patients might have [30].

Using assumptive talk
Healthcare professionals' talk frequently contained assumptions [29][30][31]. They posed questions that displayed inaccurate presumptions about patients' situations and behaviour [30], and patients having a sex life [29] (see Table 4). Despite the potential difficulty of disagreeing with questions that show ideas about what is normal, especially when the ideas are inaccurate and pertain to delicate topics, patients did disagree with the leading questions. Still, the disagreeing responses were structured differently than agreeing answers. Patients started disagreeing answers with words such as 'well' and 'uh' [30], showing their reluctance or discomfort [32].
Moreover, it was found that professionals ask questions that elicit patients to account for their disagreeing responses [29]. This Table 1 Overview sequence of assumptive and account-seeking questioning causes misalignment between patients and healthcare professionals, and contributes to the finding that professionals display expectations considering what constitutes normal sexual behaviour. Some have argued that this may convey to patients that their behaviours are not normal, increasing stereotypes of typical sexual behaviours [29,30], and keeping patients from talking about their sexual activities and behaviours [29]. Professionals also display assumptions when giving advice [31]. They brought forward assumptions about what is relevant to patients' situations and gave advice accordingly. This abstract and assumptive advice was typically preceded by giving information, and was given in relation to this information. Thus, the advice may not align with the needs or problems of patients and may be treated as irrelevant. In reaction to this advice, patients mostly produced silences or so-called unmarked acknowledgments like 'hm hm'. From these reactions, it is unclear if patients consider the advice usable.
In conclusion, healthcare professionals displaying assumptions in either their questioning or advice-giving, do not align with patients' situations, behaviours, and needs. It is apparent from all studies that reported on the use of assumptive talk that it is, therefore, not advisable to do so.

Generalized advice-giving
Two studies showed that healthcare professionals give advice in a general way by giving advice after describing a hypothetical situation [31], and by packaging it as information [31,33]. This type of advice is not tailored to patients' needs and behaviours, and may therefore be dismissed [31,33]. Presenting a hypothetical situation with the use of an utterance such as 'if you're' can be interpreted differently (see Table 5). 'You' can be used as an indefinite pronoun, and the given advice may consequently be treated as advice that is given to every other patient. However, patients can also feel personally addressed and may, therefore, show more resistance than when receiving advice as information. Advice that is packaged as information can be heard as more general and patients are not required to respond [31].
Patients mostly respond to generalized advice by producing a minimal amount of acknowledgment or response. Unfortunately, these responses do not provide an indication of patients' interpretation [31,33]. The order in which this type of advice is given, is also of significance. When delivering a package of information early in the consultation, professionals did not ask questions for the sake of getting to know the patient. Consequently, the relevance of the advice is questionable [33].
Although advice that is given in a general way enables patients to decide for themselves whether to acknowledge its relevance, professionals run the risk of not handling in accordance with patients' needs. Thus, whether or not to deploy this practice is certainly not clear-cut.

Deploying patients' talk
In three studies, ways in which healthcare professionals deploy patients' talk to discuss sexual health are addressed. First, This example shows the counsellor using a neutral first description of a sexual partner ('person'). This description does not imply gender. Before this description, the counsellor is describing sexual activities and does this in general way ('encounter') without making the type of activity explicit. Also note the hesitations and pauses that delay the description of the activity and the sexual partner (see paragraph 3.

3.2)
Silverman and Bor [28]  In this extract, the patient refrains from answering (line 3) the question regarding sexuality posed by the gynaecologist (line 1-2). Note that the gynaecologist asks a leading question in line 1-2 (see paragraph 3.3.3), which shows that a no-answer is projected. After refraining from answering (the 0.4 pause), the patient gives a short answer (see line 4). Even when the gynaecologist elicits an answer in which the patient specifies the problems (line 5), the patient delays giving details, and only talks about the time she has had the problems.
Weijts et al. [30] professionals ask questions prior to initiating the topic of sexuality [29,31,33]. By inquiring about a patient's knowledge and behaviour, professionals can give advice and lead up to a discussion in a way that aligns with the needs of a patient [31,33]. It also enables professionals to expand on the knowledge patients display and, in doing so, treat patients as competent persons [31]. However, asking many questions can be time-consuming, especially when it turns out to be difficult to get patients to speak [33]. Moreover, Kinnell and Maynard [31] argue that professionals might come across as meddlesome and confrontational. Professionals additionally make use of an approach by which they first ask questions about the relationship and, subsequently, initiate sexuality as a topic [29]. By using wh-questions (e.g. 'what') and words that link the questions about sex to the relationship previously discussed, patients are invited to evaluate or assess sexual aspects of the relationship (see Table 6). In this way, the interaction flows logically through related issues, without professionals having to bluntly ask about sensitive topics. Second, professionals use patients' talk about relationships in order to initiate the topic of sexuality [29]. This particular finding shows patients bringing up the relationship and professionals consequently asking questions about sex by referencing back to the relationship with 'it' or by using a meta comment (e.g. "speaking about your wife"). This invites longer responses from patients than the responses that were seen when professionals initiated the topic of relationships. According to Speer [29], this can be a way for professionals to cater to the issues of patients. Third, healthcare professionals use patients' pronouns, omissions, and terms to collaboratively mark the delicate nature of the topic being discussed. It should be noted that patients do not reproduce the terms used by professionals [30]. In the study by Silverman and Bor [28], professionals similarly employed terms that patients previously used to talk about their sexual activities. There were some indications that this so-called 'mirroring' resulted in extensive talk by patients.
Overall, by using and reproducing patients' utterances to talk about sexual health, professionals can show their responsiveness to the needs and issues of patients. Furthermore, notwithstanding the fact that it is time-consuming to ask a series of questions, by inquiring about a patient's relationship first, professionals can initiate the topic of sexuality without being blunt.

Depersonalization
In one study the practice of depersonalization was found [30] (see Table 7). Healthcare professionals and patients used definite articles (i.e. 'the') instead of possessives (i.e. 'my' and 'your') when mentioning private parts of the body, for instance 'the vagina'. When describing private actions, professionals additionally used nouns instead of verbs. So, instead of saying 'when you're penetrated', involving the patient in the activity, professionals referred to 'the penetration'. Depersonalizing enables patients and professionals to discuss delicate matters in a discrete way, as they can keep away from referring to the person that is involved in the action. Table 4 Example 'Using assumptive talk'.

Excerpt
Analysis Study In line 3 we can see a wh-type (what) presumptive question. With this, the psychiatrist shows his presumption that the patient has a sex life. However, the patient shows that this presumption is incorrect (line 5). Moreover, in line 14-15 and 18 the psychiatrist is checking and double-checking what the patient has said, treating the response of the patient as non-normative.

Excerpt Analysis Study
However, by deploying this practice, healthcare professionals might convey to patients that their sexual problems are a delicate topic [30]. Yet, as already mentioned, this might actually be of value since Silverman and Bor [28] argue that professionalism is shown by taking account of delicacy.

Patient-initiated advice
Two studies [31,33] reported patients asking questions or asking for advice relating to sexuality on their own initiative. In a so-called 'service encounter' [33], patients are asking the questions, as opposed to the healthcare professionals (see Table 8). This is often challenging for patients, as professionals are usually the ones to hold the floor. Before patients asked a question, they did quite a lot of work by indicating having questions and asking permission to ask these questions. In this build-up, professionals were silent or only produced minimal responses such as 'hmhm'. Professionals incited patients to talk by not taking the floor. So, by staying silent professionals can make this often challenging situation less challenging.
Although patients hardly ever asked for advice [31], it was found that patients responded to the requested advice with so-called marked acknowledgements more often than they did in response to advice initiated by professionals. First, by asking questions themselves, patients show they are open to the advice. Second, by responding with marked acknowledgements such as 'oh right', they display that the given advice was informative. It might be useful for professionals to pay close attention to the type of acknowledgments patients give, because these say something about the receptiveness of the given advice.

Discussion
One of the aims of this review was to provide healthcare professionals with suggestions on how to communicate about sexual health topics. Our results show that practices can accomplish different effects, depending on the context in which these practices are employed. Below we discuss the advantages and disadvantages of the main practices and make recommendations for professionals who aim to discuss the topic of sexual health with their patients effectively, based on the results.

Dealing with delicacy
This review has shown that healthcare professionals can discuss a delicate matter such as sexual health in different ways. For instance, professionals can make use of neutral terms in order to bypass terms with moral implications [28]. Yet, Weijts et al. [30] argue that avoiding delicate terms in the context of gynaecology settings can have negative consequences as it could lead to a mystique around female sexuality. It is, therefore, not advised to use vague terms or omit delicate terms completely.
Professionals can also show they take account of the delicate character of the issue at hand by delaying this issue with markers such as hesitations and pauses [28]. Conversely, delaying the discussion of delicate issues entirely could lead to sexual issues of patients being unaddressed. Therefore it could be beneficial for professionals to ask questions concerning sexual health early in the conversation [30]. Another practice professionals can deploy is the use of depersonalization. This allows professionals to talk in such a manner that they do not have to mention the person that is involved in a private action [30].

Increasing responsiveness
There are several ways in which professionals can increase their responsiveness to patients. The interview format [30,33] is a useful practice for tailoring advice to patients' needs. Because the professional first obtains information about the beliefs, needs, and concerns of patients through questioning, s/he can provide information that adheres to patients' needs. Table 6 Example 'Deploying patients' talk'.

Excerpt
Analysis Study In this extract, the psychiatrist is making a reference to the previously discussed relationship. The question in line 12-13 is linked to the relationship with the use of the word 'things'. The psychiatrist then uses a 'wh-question' (how were that). With this question, the patient is invited to give an evaluation of the relationship's sexual side. In line 14 we can see that the patient indeed responds in the projected manner. By using a patient's response to consequently ask about sexuality, the psychiatrist does not have to bluntly ask about this topic.
Speer [29]  Generalized advice-giving is not particularly tailored to patients [31,33]. In addition, research investigating views of breast cancer survivors concerning conversations about sexuality emphasizes that patients prefer specific instead of general recommendations [11]. Still, advantages of generalized advice are that it is less time-consuming than asking a great extent of questions prior to advicegiving, and that it may head off the suggestion that professionals tell patients what to do in the very intimate parts of their life [34]. We, therefore, conclude that whether to use this practice is ultimately dependent upon what the professional and patient aim to achieve with the conversation.
Patients asking for advice on their own initiative is of benefit to their receptiveness to the advice. Professionals can ensure a favourable environment for asking questions by inciting patients to talk by staying silent, by inviting patients to ask questions, and by taking account of patients indicating having questions. Moreover, professionals must also take into consideration the responses to given advice. For instance, unmarked acknowledgments such as 'hmhm' indicate that the advice may not be relevant to patients [31]. Thus, professionals can, dependent upon the response of patients, adjust their advice accordingly.
In addition, when professionals aim to show responsiveness, the practice of mirroring, i.e. repetition of the terms used by patients can be deployed. It can facilitate a situation in which patients talk extensively [28]. Earlier research confirms that mirroring can be used in order to display responsiveness since this makes it possible for professionals to explore and reflect on the views and experiences of patients. Additionally, it is suggested that mirroring can contribute to a neutral vocabulary, as professionals do not need to introduce new and potentially suggestive terms [35].
Healthcare professionals can also show their responsiveness to concerns of patients by using patients' talk about relationships. Initiating the topic of sexuality in this way shows that professionals are being prompted by the utterances of patients. When patients do not talk about their relationships on their own initiative, professionals can first inquire about this before introducing sexuality as a topic [29].

Avoiding assumptions
It is important that professionals keep away from using assumptive talk in their questioning, since this only has negative consequences. Assumptive questioning might carry ideas about what is normal and abnormal sexual behaviour. Patients may, therefore, be hesitant to open up about their actual behaviours [29]. Based on recommendations by Speer [29], we argue that, what is likely to be more successful in getting alignment, is for professionals to first establish whether or not patients have a relationship and whether or not it is a sexual one.
When professionals aim to give advice to patients, it is also inadvisable to display assumptions or to present advice in a general way. This type of advice is abstract and might be irrelevant to patients [31].

Strengths and limitations
The main strength of this review is that it is the first to systematically explore discourse analytic studies about sexual health in medical settings. A key advantage is the use of naturally occurring interactions in medical settings, allowing us to more fully understand how interactions about sexual health unfold. However, the review has some limitations. First, most of the included studies were carried out more than twenty years ago. In recent years, however, there has been an increasing focus on patient centred care and patient centred communication (PCC) [36][37][38]. This may or may not have influenced the way in which sexuality is discussed. Since one of the key concepts of PCC is that healthcare professionals should take into account patients' individual psychosocial and cultural contexts [36], current communication about sexual health might be more individually tailored and more responsive to patients.
Another weakness is that our review only encompassed five studies, which resulted in reporting practices that sometimes only occurred in one study. This implies that there could be practices that we did not find. In a study on self-reported views of patients and professionals on sexual heath communication in oncology settings, professionals normalized sexual health issues to let patients know others are confronted with similar issues [11]. This practice was not reported in one of the included studies in this review. New research is needed to address these gaps.
The limited number of studies also indicates there is a gap in current literature. While discourse analytic studies about sexual health in medical settings are scarce, it should be mentioned that conversations on sexual health in medical settings have been studied from other perspectives as well. For example, research has focused on observations of oncology follow-up consultations [39], and audiorecordings of sexual health discussions with older adults [40] and adolescent patients [41]. Still, despite the fact that these studies contribute to our understanding of the occurrence, context, and content of talk about sexual health, little to no attention is paid to how the topic is discussed, and how participants respond accordingly. Conversely, the action-oriented, sequential focus of the studies incorporated in this review allows for claims about the functioning of communication practices.
The studies included in this systematic review were limited to only three settings in which it may be evident that sexual health is a topic. As previously mentioned, sexual health is, however, a fundamental part of overall health. Moreover, in consultations in other settings, such as conversations with patients with breast cancer [12] and other diseases [10], patients and professionals report having difficulties with discussing sexual health. It is therefore crucial to explore patterns of communication in sexual health interactions in other settings as well. Moreover, all five studies used Conversation Analysis to analyse the interactions between healthcare professionals and patients. There are, however, more forms of Discourse Analysis that are used in social sciences to analyse discourse [14]. For instance, it would be interesting to use Discursive Psychology as an approach to focus more on how psychological constructs, such as identities, are made relevant in talk in order to use these findings as strategies for talking about sexual health [14].

Conclusion
Our review has shown that professionals interact with their patients about sexual health in multiple ways. We distinguished seven communication practices and identified their effects. Many of these practices illustrate the delicate character of discussing sexual health since issues relating to this are avoided, delayed, and discussed carefully. Yet, results also give insight into practices that professionals can use to deal with this delicacy without mystifying the matter. Additionally, some practices that were found can help professionals in stimulating patients to talk about their sexual health by being responsive to patients' beliefs, needs, and concerns, without being vague or making assumptions that communicate certain moral values.

Practice implications
This review has provided healthcare professionals with strategies to broach topics relating to sexual health. In talking about sexual health, it is crucial to be aware of the advantages and disadvantages of the various communication practices that can be deployed. It is sometimes dependent upon the setting and the goals of the interaction which practices can be considered as best practice. Even though many communication practices were found to occur in different settings, our results cannot simply be generalized to all other settings. Healthcare professionals should therefore be trained in being attentive to the different outcomes subtle differences in communication can have. These trainings can, for instance, involve looking at professionals' own interactions in medical practice. Implementing these skills may contribute to the abilities of professionals to provide patients with the needed care when faced with challenges that affect sexual health, and consequently their general well-being.

Funding
This research did not receive any specific grant from funding agencies in the public, commercial, or not-for-profit sectors.

Declaration of Competing Interest
The authors declare that they have no known competing financial interests or personal relationships that could have appeared to influence the work reported in this paper.