'Teenagers are into perfect-looking things': Dating, sexual attitudes and experiences of adolescents with severe obesity.

BACKGROUND: This qualitative study explored the dating and sexual health attitudes and behaviours among adolescents with severe obesity (body mass index [BMI] > 99th%) attending a multidisciplinary weight-management programme.
METHODS: Semi-structured interviews were conducted with 19 participants (12 females, 7 males; Mage  = 16.8) and analysed through reflective thematic analysis.
RESULTS: Participants described polarized dating behaviours in which dating and sexual relationships were either avoided due to this not being a priority, lack of time, feared rejection, and/or body size as perceived barrier or in contrast, when approached, involved greater sexual risk.
CONCLUSIONS: These findings have numerous implications including the need for increased education on the romantic developmental challenges faced by adolescents with severe obesity, the importance of ongoing screening of high-risk sexual behaviours and body dissatisfaction from frontline care providers, and the ability to support referrals to psychosocial services when appropriate.

Adolescents with severe obesity may present with a different trajectory of psychosexual development.

Reasons qualifying differences in sexual development and health behaviours are less known.

This study highlights the polarized dating behaviours of adolescents with severe obesity in which dating and sexual relationships are either avoided or involve greater risk when approached.

Results further inform our understanding of vulnerabilities among adolescents with obesity that can impede normative sexual and romantic development.

Healthcare professionals need to be sensitized to both the psychological and sexual health risk factors of adolescents with severe obesity.

INTRODUCTION

Adolescence is a critical formative stage for the development of romantic relationships. According to Connolly and Goldberg's (1999) phase‐based model, normative adolescent romantic development is thought to progress from small groups of same‐sex peers in preadolescence, to larger, mixed‐gender groups by middle adolescence (i.e., ‘group dating’), to finally a single, romantic dyad and sexual relationship by late adolescence. The developmental trajectory of adolescents with obesity, however, seems to unfold at a delayed rate in which both males and females are less likely to initiate and/or engage in a romantic or sexual relationship (Cawley et al., 2006). In addition, adolescents with obesity are less likely than their healthy weight peers to have had sexual intercourse (Averett et al., 2013; Becnel et al., 2017; Lowry et al., 2014; Ratcliff et al., 2011) and for those who are sexually active, they are more likely to engage in risky sexual behaviours including casual encounters (Becnel et al., 2017; Eisenberg et al., 2003), multiple sexual partners (Lowry et al., 2014; Turay, 2013; Villers et al., 2010), inconsistent contraception use (Becnel et al., 2017; Chang et al., 2015; Eisenberg et al., 2003; Lowry et al., 2014), sex while under the influence of substances (Becnel et al., 2017; Eisenberg et al., 2003; Ratcliff et al., 2011), engaging in anal intercourse (Averett et al., 2013; Gordon et al., 2016) and oral intercourse at a younger age (Gordon et al., 2016). These behaviours can lead to potentially increased sexual health challenges, including sexually transmitted infections (STIs) and unintended pregnancies (Becnel et al., 2017).

Healthcare professionals are responsible for counselling a growing number of adolescents with obesity regarding their sexual health and contraception (Alexander et al., 2014). It is well established that adolescents prefer physicians to initiate the topic of sex during clinical encounters (Sobal et al., 1995). Unfortunately, only 28%–44% of adolescent females with severe obesity report receiving contraceptive information from physicians (Becnel et al., 2017). This is significant given the increasing prevalence of adolescent obesity and, thereby, the greater number of youths potentially at risk for unintended pregnancies and STIs (Alexander et al., 2014).

To date, although research has established that sexually active youth with obesity engage in risky sexual behaviours, the reasons behind this have not yet been fully explored. Multiple quantitative studies have shown a relationship between risky sexual behaviour and psychological factors such as stigmatization, low body esteem, negative self‐perception, weight‐based teasing and social alienation (Alice Cheng & Landale, 2011; Anderson, 2010; Aruguete et al., 2009; Chen & Brown, 2005; Farhat et al., 2015; Kann et al., 2014; Larson et al., 2012; McLaren & Gauvin, 2002; Pisk et al., 2012; Wingood et al., 2002); however, there are no qualitative studies to explain these observed phenomena. Such qualitative explorations would allow to draw from the actual lived experiences of adolescent respondents, allowing for increased depth and understanding. Our objective was to explore dating and sexual health attitudes and behaviours among adolescents with severe obesity to comprehend why they do not engage in sexual relationships, or when they do, why they engage in risky sexual behaviours.

METHOD

Study design

This qualitative study used a reflective thematic analysis (TA) approach (Braun & Clarke, 2006). Reflective TA allows to identify patterns of meaning across a data set, from familiarization with interview content to theme generation and review (Clarke et al., 2019). This study was approved by the Research Ethics Board (REB).

Participants

Participants comprised of patients from a multidisciplinary weight‐management programme located in a large paediatric hospital. Approximately 150 adolescents participate in the 24‐month obesity management programme at any given time, with approximately 70 new intake consultations per year. Study inclusion criteria included the following: (1) being able to provide informed consent, (2) aged ≥ 16 years, (3) having severe obesity as defined by a BMI ≥ 99th% for age and sex and (4) having attended the weight‐management programme for ≥3 months. Purposeful sampling was used to identify individuals from the clinic that provided variable information across gender, age and length of time in the programme. This form of sampling was selected to include participants that have particular knowledge that can help inform the in‐depth exploration of this subject area. In total, 27 adolescents were initially approached to participate in this study. Eight participants (three males and five females) declined, such that the final study sample comprised 19 participants.

Data collection

Participation in this study was voluntary. Following written informed consent, one‐on‐one semi‐structured interviews were conducted privately and in‐person by a female clinician‐researcher (Ó. W.) with experience in qualitative research methods and who was not involved in the delivery of clinical care. The interviewer did not have an established relationship with the participants, nor did she have any biases or assumptions prior to the study. The one‐time interview lasted approximately 1 h and was audio‐recorded with permission. Participants were informed that interview excerpts may be used with their sex and gender identified and consented to this use. A semi‐structured interview guide was used to maintain enough flexibility to allow participants to explore issues throughout the discussion. The interview guide included a broad range of topics including (1) general information (e.g., ‘How do you get along with your parents/siblings?’), (2) psychosocial functioning (e.g., ‘How do you feel in social situations?’, ‘How do you think other people see you?’ and ‘How would you describe your mood in the past month?’) and (3) dating and sex (e.g., ‘What does it mean to be dating someone?’, ‘What does a healthy relationship look like?’ and ‘What are the good/bad things about having sex?’).

Participants were given a $25 gift card as compensation for their time and contribution to the study. Baseline demographic information, comorbid mental health diagnoses and anthropometry on all participants were obtained from their electronic medical record.

Data analysis

Interviews were digitally audio‐recorded for verbatim transcription. Transcriptions were checked against the recordings for accuracy and uploaded into software supporting qualitative data analysis (NVivo 10). Data collection continued until saturation was reached; this refers to the point in data collection when only variations on established themes rather than substantially new themes emerge (Rotermann, 2012).

Five sequential steps were used to conduct the data analysis: (1) data familiarization including a repeated reading of each transcript; (2) careful documentation of overall interview impressions and code assignment involving line‐by‐line analyses; (3) examination and grouping of assigned codes into broader themes; (4) theme review through comparisons across, between and within participants' transcripts to delineate divergent and similar patterns; and (5) detailed analysis of each theme including thematic naming and mapping. Two members of the research team (Ó.W. and J.C.) independently reviewed the transcripts and identified codes that were discussed among members of the research team to ensure consensus, including theme saturation. A journal was kept documenting the context of each interview and reflections, including an audit trail of key analytic decisions.

Demographic, mental health diagnoses and anthropometric data were analysed descriptively using Microsoft Excel (version 16.19).

RESULTS

Table 1 outlines participants' socio‐demographic characteristics. In total, 19 interviews were conducted with adolescents aged 16 to 19 years, the majority of which (73%) were biological females. Table 2 outlines participants' comorbid mental health diagnoses. From the overall sample, 16 participants (84%) had at least one mental health diagnosis, and eight participants (42%) had at least three comorbid mental health conditions.

TABLE 1

Socio‐demographic characteristics of study population (n = 19)

Variable	Mean (SD) or frequency (%)
Age	16.8 (1.0)
Body mass index (kg/m 2 )	43.9 (8.6)
Ethnicity
White	9 (47%)
Black	3 (16%)
Mixed race/other	7 (37%)
Biological sex
Male	7 (37%)
Female	12 (63%)
Sexuality
Heterosexual	15 (80%)
Other (homo/pan/bi/asexual)	4 (20%)
Duration in obesity management programme (months)	17.9 (15.8)
Family household income
>$100 K	5 (26%)
$50–100 K	7 (37%)
<$50 K	7 (37%)

TABLE 2

Comorbid mental health diagnoses among youth with severe obesity

Variable	Frequency
Number of comorbid mental health diagnoses
0	3 (15.8%)
1	5 (26.3%)
2	3 (15.8%)
≥3	8 (42.1%)
Types of mental health disorders a
Anxiety and/or depressive disorder	23
Self‐harm and/or suicidality	5
Binge eating disorder	5
Attention deficit and hyperactivity disorder	4
Other (oppositional defiant, personality or post‐traumatic stress disorder)	5

Total types of mental health disorders greater than sample size of 19 as participants can have more than one co‐occurring mental health condition.

Table 3 outlines participants' dating and sexual experiences including risky behaviours. Descriptions were categorized as either ‘developmentally appropriate’ or ‘delayed’ romantic development. In line with the expectations of late adolescent dating (Connolly & Goldberg, 1999), a developmentally appropriate romantic development was defined by the research team as an in‐person, mutually beneficial dating experience with a peer. In contrast, a delayed romantic development was defined as dating and sexual experiences that were typical of earlier stages, such as the emergent romantic activities described in preadolescence and middle adolescence (Connolly & Goldberg, 1999). Based on this operationalization, four participants (21%) had dated in a developmentally appropriate way, whereas the majority of the sample reported a delayed romantic development, avoiding romantic relationships for reasons including dating not being deemed a high priority and fearing rejection due to their body size. Six participants (31.6%) reported having engaged in sexual activity. Of the six who were sexually active, five participants (83.3%) endorsed risky behaviours.

TABLE 3

Summary of dating and sexual experiences among youth with severe obesity

Types of experiences	Frequency (%)
Dating
None	5 (26.3%)
Online only	4 (21%)
Developmentally appropriate patterns	4 (21%)
Delayed patterns	6 (31.6%)
Sexual activity a
None	13 (68.4%)
Not including sexual intercourse	3 (15.8%)
Including sexual intercourse	3 (15.8%)
Risky behaviour among those sexually active (n = 6)	5 (83.3%)

Types of experiences represent mutually exclusive, hierarchical categories.

Reflective thematic analysis

As displayed in the thematic map (Figure 1), several themes and subthemes capturing the contrast between the relationship ideals of adolescents with severe obesity and their actual lived experiences of dating and sexual behaviours were identified.

FIGURE 1

Thematic map depicting the dating, romantic and sexual experiences of adolescents with severe obesity

The idealized relationship

Although most participants had not engaged in romantic relationships, they described the qualities they would seek and value most in an ideal romantic partner. The most frequent theme was that of having a ‘good personality’, with examples including being funny, nice or kind. Participants often noted that they chose not to date because of a perceived societal focus on appearance. Notably, none of the participants listed physical attractiveness as a necessary quality in a romantic partner:

People can just see someone and be attracted to them by their looks and not really know much … like, who they really are … Looks are important but I feel like what's inside is more important than what's on the outside. (Participant #19, female)

A second theme that emerged within the idealized relationship was that of prioritizing connection over appearance:

I do not think bodies and body image should be a huge part of a relationship. I think it's more of the connection between the people and about being happy. (Participant #15, gender fluid)

In general, sexual activity, in all forms, was not highlighted as an important quality for romantic relationships. Instead, one participant described wanting to slow down sexual activity when dating a previous partner:

People just want to be sexual right away and I'm like, now pump the brakes a little bit, I'm not going to do that and then they are like, “Well have you ever done this before” and I'm like yes, but that does not mean I'm going to do it right away with you, like relax. (Participant #7, female)

The theme of trust and respect came up numerous times as necessary qualities in a partner:

I think when they are like best friends to you—that you can talk to each other about other stuff and they respect each other and trust. (Participant #18, female)

The foundation of friendship in romantic relationships was also important to many:

… That's my thing, is that if you at the end of the day can still call your partner your best friend that's a healthy relationship. (Participant #6, female)

The dating experience

When it came to participants' descriptions of their actual experiences of dating, relationships and sexual behaviours, two main yet divergent themes of avoidance and risky behaviours emerged.

Avoidance

A prominent avoidance subtheme was that of delayed romantic development, in which the majority of adolescents had participated at most in emergent dating activities (n = 15, 78.9%), such as ‘hanging out’ or hugging, online interactions, or no dating experiences at all. Across all interviews, there were very few instances of developmentally appropriate dating (n = 4, 21%). Many participants described having dated and enjoyed the attention, but when they were much younger and without it progressing into a relationship:

… the first person I ever dated was like my cousin's best friend, and I was young … I was just happy that someone was showing me attention … I never thought any guys would like me and then the fact that he showed me attention I was really happy (Participant #1, female)

Adolescents who had avoided in‐person romantic relationships found alternative ways to engage with others through online mechanisms such as chat platforms or gaming websites. Instant messaging and telephone calls were used to maintain these relationships:

I had one online relationship, that was my longest relationship, it lasted 6 months with somebody who was in Georgia. I met him through gaming, I had known him for a year before and then we finally decided well let us just try this, then it did not work out … we video chatted quite a lot, and just talked, just like texting each other (Participant #5, female)

Although only a few participants used online dating websites, they used these sites because they did not participate in traditional in‐person dating. However, when asked if they would meet up with these individuals in‐person, participants indicated that they would avoid these situations for fear of disappointing the other person:

I have had people be like, “Hey, do you wanna meet up?” And I'm like, “No.” (Participant #12, female)

Most participants described not having initiated romantic relationships and often articulated ‘having too much going on’ in their lives as reasons for why they had chosen to delay dating. Although school was often cited as the reason why they lacked time for relationships, most participants did not identify having hobbies and were not engaged in extracurricular activities. This suggests that being ‘too busy’ offered participants a rationale to present for their lack of dating experiences.

A third subtheme of fear of vulnerability and rejection arose in the interviews numerous times when discussing youth asking others out or sharing their interest in someone. Many chose to spend their time with their friends and described having smaller friend networks, often comprised of one or two close friends. This desire to focus on friendships was often later linked to their intentional desire to avoid pursuing romantic relationships and suggests that these interactions felt safer and less exposed in comparison with dating.

A fourth identified theme was that of body size being a barrier to dating, both with regards to internalized weight‐stigma, as well as anticipated physical limitations. Some participants described a lack of confidence that others would be interested in them due to their size. The fear of being excluded due to their weight was highlighted several times throughout multiple interviews:

I guess people do not like big people […] I do not see a lot of big people in relationships. (Participant #13, female)

When discussing engaging in sexual activity, others highlighted the fear of ‘exposing’ their bodies to others:

Because of my weight, I do not want to expose my body … […] … you know, there's no one that's going to like want to date [me]. (Participant #16, female)

Another participant added:

Skinny people have nothing embarrassing and they are pretty. Show it off. (Participant #18, female)

A female participant indicated that she had concerns with both the initial encounter of meeting someone and the potential physical constraints of engaging in sexual activities due to her size:

I feel like it [body size] does, definitely [matter], because it's not the most appealing of an appearance, so there's also that first impressions and whatnot too. So, there's that part, the initial barrier and then, like the actual … let us say the active intercourse (Participant #12)

Another participant described how her weight affected her confidence in pursuing a romantic relationship:

I have not done anything yet … I think definitely my weight … I have some confidence in myself, but I feel like my weight brings it down a little bit. (Participant #19, female)

Risky behaviours

The majority of participants (n = 13, 68.4%) had not engaged in any type of sexual activity. Among the six participants who had engaged in sexual relationships, five described negative experiences. All of these situations involved risky behaviours.

A first subtheme of underage substance use was reported in most situations in which adolescents shared engaging in sexual behaviours. One participant noted that she was unable to communicate her desire to slow down the sexual activity due to her state of intoxication:

I wanted to go slower than that, but I ended up not going slow … So, at that point [being intoxicated], it was hard setting boundaries for myself. (Participant #12, female)

Unfamiliar partners and inconsistent contraceptive use also emerged a second risk behaviour. One participant described being in an unsafe situation with an older man and not using condoms. The one male participant who previously had sexual intercourse was under the influence of alcohol at the time and did not use contraception. This participant indicated that he now only engages in sexual activity online and describes preferring this as he can manipulate the camera angle so as not to show his body:

We're friends, but we do not meet up in person for anything … There's also like the self‐esteem type thing … there's different angles, camera angles, in‐person there's no different camera angles. So, that's kind of the thing … […] … Size, like body size … weight (Participant #17, male)

Some participants preferred to seek relationships online, such that online dating and sexual activities were framed as a way to avoid in‐person interactions. In one instance, the fear of meeting face‐to‐face led to the end of the online interaction:

We'd been talking online. I met him on this app called Whisper … […] We were talking for months and then he wanted to meet. And then I was like, “I'm not ready” … and then we stopped talking. (Participant # 9, male)

Despite the perceived avoidant function of online dating and sexual activity, these behaviours also represent safety and sexual risks, given the age of participants and concerns related to online sexualization and exploitation.

A final subtheme identified was that of pressure and one‐sided sexual activities. The few sexually active female participants described feeling under pressure to engage in sexual activity and shared regretting it afterwards:

I think I did it another time, and it just wasn't the experience I wanted, if I had probably waited it'd probably be better, cause I'd actually care about the person (Participant #2, female)

The same few also described performing sexual acts without desire or expectation of reciprocation:

I've done. I have not received it [oral sex]; I do not want to receive it (Participant #7, female)

One participant explained that this was because she did not want to be seen naked:

It was more just giving than, receiving at all. I do not want all my clothes to be off. I just want to be chill, because I [was] in a state of, being drunk, I was like, “No, like, no. I do not want anything going on where I'll be in a position where he can see”. (Participant #12, female)

DISCUSSION

Our findings support the existing literature that adolescents with obesity are less likely to date or engage in sexual activity compared with nonobese and age‐matched peers (Averett et al., 2013; Becnel et al., 2017; Lowry et al., 2014; Ratcliff et al., 2011). Only 10.5% of the study participants had sexual intercourse, compared with 30% of Canadian youth aged 15 to 17 and 68% of 18 to 19 year olds (Rotermann, 2012). This study, however, is the first to explore the reasons behind these choices among adolescents with severe obesity. The majority of the participants either completely avoided in‐person relationships because it was not a priority for them, stating they did not have enough time, or feared rejection due to their body size. On the other end, among all participants who had engaged in sexual activities, 86% reported risky behaviours and that they did not enjoy the sexual experience.

Greydanus and Pratt (2016) describes normal adolescent psychosexual development in three stages. In early adolescence, sexual behaviours emerge, dating commences and peer pressure becomes significant. During middle adolescence, sexual activity often begins, the meaning of relationships is explored, and increased independence can lead to risks. Finally, in late adolescence, Kjelgaard et al. (2017) describe how adolescents develop the ability to have functional nonselfish relationships. Although this outlines the typical psychosocial development, the participants in this study described different needs and approaches to their relationships. Participants did not place high value on the role of sexual intercourse and activity in romantic relationships. Instead, friendship, trust and respect were much more important components with emphasis on the importance of personality over appearances.

The importance of friendship as a steppingstone into dating and romantic relationships is in line with adolescence dating‐stage theories in which adolescents' sequence from same‐gender friendships, to mixed‐gender interactions, and then to dating and romantic relationships (Connolly et al., 2004). However, the expectation is that older adolescents, as in the current study, would be farther along this sequencing, as opposed to the described emergent activities, which are more typical of earlier adolescence. This delayed progression has also been shown to continue and persist past adolescence into emerging adulthood, suggesting a longer term impact on healthy relationship formation (Ames & Leadbeater, 2017).

Given that most participants had experienced some form of weight‐related bullying, the desire to find a romantic partner that did not judge them based on appearance was an important theme identified. Participants described wanting partners with whom they had a deep connection with and who would not make appearance‐based assumptions. The experience of weight‐based stigma has been shown to negatively impact the formation of peer and romantic relationships (Ames & Leadbeater, 2017). This is in keeping with the concept of delayed or derailed normal psychosexual development and with the finding that the majority of the current sample did not report a developmentally appropriate romantic development.

Another way in which participants avoided relationships was by engaging in online partnerships where they played video games or chatted using online platforms. In keeping with what they valued in relationships, it was more important for them to have a non‐physical connection with their peers. For some, when the option of meeting in‐person was raised, they avoided this as they feared rejection. Potential reasons for this include weight‐based stigmatization, low self‐esteem or poor body image, and fear of judgement leading to rejection. Although this was viewed by participants as an avoidance behaviour, it also represents a risk behaviour given the potential for online victimization, including adolescent sexual solicitation and exploitation (Wurtele & Kenny, 2016).

On the other side of avoidance, when participants were sexually active, they often engaged in risky sexual behaviours, which increased their likelihood for STIs and pregnancy. Our findings are consistent with the literature that states that sexually active adolescents with obesity are more likely to engage in high‐risk sexual behaviours (Becnel et al., 2017; Chang et al., 2015; Eisenberg et al., 2003; Gordon et al., 2016; Lowry et al., 2014; Villers et al., 2010), with all of our participants who had sexual intercourse engaging in one‐night stands, concurrent substance use, inconsistent contraceptive use, and a case of unintended pregnancy. Of the three others who engaged in sexual activities but not intercourse, two were also involved in risky sexual behaviours including unfamiliar partners and associated substance use.

Our study also supports the literature that adolescent females with overweight and obesity are more likely to perform oral sex without reciprocation (Gordon et al., 2016) with two out of three participants experiencing this. Reasons for this included feeling pressured and also a fear of exposing their own bodies. Another participant described how online sexual activity was used to avoid pregnancy but also enabled him to hide his body size.

The avoidance of romantic relationships is supported by other qualitative research on weight‐based victimization. Griffiths and Page (2008) examined how weight‐based victimization affected peer relationships among female adolescents and found that low self‐confidence, isolation and peer anxiety were direct consequences of bullying and affected an adolescent's ability to develop peer relationships. In a similar study, Yufe et al. (2017) found that young adult bariatric patients had compromised peer and intimate relationships as a result of weight‐based bullying. Jacobs et al. (2020) lend further support for this association by demonstrating that overweight adolescents are less likely to be nominated as either social or romantic interests, compared with their normal‐weight peers. Given the importance of peer relationships in supporting normative adolescent romantic development (Connolly et al., 2000), this social rejection is an important barrier.

Physical changes during adolescence make appearance a prominent focus while at the same time social relationships are increasingly important. This combination leaves adolescents especially vulnerable to peer stigmatization (Pearce et al., 2002). The literature is unequivocal regarding the association with obesity and stigmatization with subsequent psychosocial maladjustment, echoed in our findings of 84.2% of participants having at least one and 42.1% having three or more comorbid mental health diagnoses. Kershaw et al. (2011) have demonstrated the link between psychological distress and increased sexual risk behaviour in adolescents and emerging adults with obesity, further supporting the sexual health vulnerability of this particular population.

While the results of qualitative research can be influenced by the researcher completing the interviews and performing the analysis, the in‐depth interview technique followed by sequential independent data analysis by two separate researchers provides a valuable insight into the dating and sexual experiences of these participants. The study is limited in its generalizability as it only included older adolescent participants with severe obesity who were seeking health care in a tertiary obesity management programme.

CONCLUSION

Our findings indicate that although adolescents with severe obesity value personality, trust and respect, as well as friendship and connection over appearance in their idealized romantic relationships, their actual lived experiences are divergent of these values. Adolescents are either avoiding dating and romantic interactions due to other life priorities and perceived limited time, internalized weight‐stigmatization, fear of rejection, or delayed dating activities or, in contrast, when sexually active, are engaging in risky behaviours. Of note, none of the reported avoidant or high‐risk behaviours are in line with the described values, needs and wants of their idealized relationships. These results help better inform our understanding of the multiple vulnerabilities of adolescents with severe obesity and how the continued exposure to factors such as weight discrimination and body dissatisfaction can impede on normative romantic development.

Key implications from these findings include the importance of sensitizing healthcare professionals to both the psychological and sexual health risk factors of adolescents with severe obesity, the need for early identification of developmental and sexual health risks, the importance of ongoing sexual health screening and the value of referring to appropriate psychosocial supports when needed. Effective supports should include increased risk‐awareness and education, as well as counselling adolescents on healthy relationship formation, body image and self‐esteem.

CONFLICTING OF INTERESTS

The authors have no potential conflicts of interest to disclose.

AUTHOR CONTRIBUTIONS

Study conception, design, data acquisition: Walsh, Dettmer, Regina, Christian and Toulany. Analysis and data interpretation: Walsh, Dettmer, Regina, Dentakos, Christian, Hamilton and Toulany. Drafting and critical revision: Walsh, Dettmer, Regina, Dentakos, Christian, Hamilton and Toulany.