Layout 1 [Qualitative Research in Medicine & Healthcare 2019; 3:7875] [page 1] Introduction In October 2016, Instagram (IG) User 1 received an email from a nurse at the Reproductive Endocrinology and Infertility (REI) clinic where she underwent a suc- cessful fertility treatment and was monitoring an early pregnancy (user names related to images have been anonymized as User 1, 2, 3, etc.). At 2:15 p.m., User 1 read this email, captured it with a smartphone and by 2:19 p.m. posted it on her IG account without any identifying information redacted. User 1’s image caption responded to the nurse’s comment, “sorry for the typo” with “Typo… that’s a pretty big, can cause you to miscarry the baby, typo. Ack. This does not help my anxiety at all.” The comment section below the image (Figure 1) of the email and the caption included references to institu- tional health systems, encouraging messages about self- advocacy, individuals sharing their own treatment horror stories and a series of emoji. The hashtags used by User 1 made this post searchable by millions of IG followers, despite the nurse’s full name and the user’s full legal name (not the IG account name) listed in the image. It is not clear whether User 1 shared their concerns directly with the nurse and the REI practice in question, or isolated their reaction to their account on this social media platform. It is evident that a semiotic act occurred in this post, in that the discourse (e.g., through images, texts), both symboli- cally and literally (re)created meaning. Here, the images, text and hypertext – individually and taken together – re- flect the determined agency of a patient challenging the rhetoric of passivity, and signaling to the viewing audi- ence that in the age of social media, medical mistakes or misinformation can be posted and enter the public record. The poster is raising awareness of medical errors that can and do occur; to blindly trust medical staff is unwise. The post operates as a kind of threat as well – the public airing of these errors could have professional and monetary con- sequences for REI practices. Examining the post semiotically (through the study of signs and symbols and their meanings) offers ample ma- terial in the responses to this post (e.g., hashtagging and #ttc on Instagram: a multimodal discourse analysis of the treatment experience of patients pursuing in vitro fertilization Bethany Johnson,1,2 Margaret M. Quinlan,1 Nathan Pope1 1Department of Communication Studies; 2Department of History, University of North Carolina at Charlotte, Charlotte, NC, USA ABSTRACT While searching the Internet for health information is now commonplace, giving and receiving medical expertise on social media platforms such as Instagram (IG) is underexplored. In order to investigate the ways in which social media platforms are a resource for seeking medical expertise, we employed a multimodal discourse analysis; specifically, we focused on the hashtag #ttc and the engagement of the IG community it comprised. We examined three main themes around medical expertise during treatment indexed by the hashtag: treatment protocol choices, treatment side effects and pregnancy diagnosis/confirmation. Our findings suggest that social media platforms provide a new arena in which patients can challenge medical expertise by circumventing it or adopting lay expertise instead of practitioner suggestions. We close with research implications and suggestions for further research. Correspondence: Margaret M. Quinlan, Department of Communi- cation Studies, University of North Carolina at Charlotte, 9201 University City Blvd., Charlotte, NC 28223, USA. E-mail: mquinla1@uncc.edu Key words: Instagram; Multimodal discourse analysis; Imagetext; Infertility; Support groups. Acknowledgements: we would like to thank Drs. Robin Jensen and Heather Carmack for valuable feedback on this manuscript and to Rachel Ayers for her assistance in data collection. We are indebted to the feedback our supportive editor, Dr. Mariaelena Bartesaghi, Francesca Baccino and two anonymous reviewers. Contributions: the authors contributed in order of authorship. Conflict of interest: the authors declare no potential conflict of in- terest. Funding: none. Conference presentation: a preliminary version of this paper was presented in 2017 at the Organization for the Study of Communi- cation, Language, & Gender, 40th Annual Conference, Omaha, NE. Received for publication: 11 October 2018. Revision received: 14 February 2019. Accepted for publication: 16 February 2019. This work is licensed under a Creative Commons Attribution Non- Commercial 4.0 License (CC BY-NC 4.0). ©Copyright B. Johnson et al., 2019 Licensee PAGEPress, Italy Qualitative Research in Medicine & Healthcare 2019; 3:1-14 doi:10.4081/qrmh.2019.7875 Qualitative Research in Medicine & Healthcare 2019; volume 3:1-14 No n- co mm er cia l u se on ly dialogue). Multimodal discourses (textual, visual, inter- textual, hypertextual) confirm that account followers un- derstood the aims of the post, promising to increase their own attentiveness to treatment protocol, calling out health systems where they supposed this type of error would occur and encouraging other readers to be aware, with one user concluding “now you’ll be questioning everything I bet.” Infertility remains a pressing concern for individuals struggling to create a family, just as it was over a century ago. Yet today, individuals can share their experiences with hundreds, thousands or millions of strangers in real time. The result is a unique type of intimacy, one that in- fertile individuals did not have access to previously. Since 1878, estimates of infertility rates have hovered between 9%-11% for heterosexual couples, with outliers ranging from 6% to 20%.1-3 Currently, the Centers for Dis- ease Control (CDC) estimates that “12% of women aged 15-44 years”,4 will experience infertility, while the Na- tional Institute of Child Health and Human Development estimates “9% of men and 11% of women”,5 and the World Health Organization (WHO) project “10% of women” will experience this condition. The CDC and WHO define infertility as occurring after 12 months of unprotected, heterosexual sex between two cis partners. Similar to the variety of percentages available infertility is described using a wide range of definitions, framing the issue as anatomical, physiological, or a reflection of de- mography.6 Infertility remains a persistent health chal- lenge despite advances in treatment including in vitro fertilization (IVF), intra-cytoplasmic sperm injection (ICSI) and genetic testing for embryos, which emerged by the late 20th century.7,8 The Mayo Clinic defines IVF as “a complex series of procedures used to treat fertility or genetic problems and assist with the conception of a child” (n.p.). During IVF, mature eggs are retrieved from the ovaries and fertilized with sperm in the lab setting. Later (e.g., 3, 5, or 6 days), the embryos are placed into your uterus. As for ICSI (pronounced icksy), this proce- dure resulted from the successful experiments of Dr. Gi- anpiero Palermro, who manually fertilized one egg with one sperm in 1992.9 Regardless of availability, treatment (with or without medication) is expensive and many insurance companies do not offer sufficient coverage.7,10,11 Aside from the pro- cedural cost, medicines can cause uncomfortable side ef- fects, and treatment cycles dictate schedules, placing [page 2] [Qualitative Research in Medicine & Healthcare 2019; 3:7875] Article Figure 1. Reproductive endocrinology and infertility practice error. No n- co mm er cia l u se on ly social and emotional pressures on familial and other in- terpersonal relationships; patients often feel isolated, which can increase anxiety.8,11-17 This isolation can result from the difficulty of the treatments (e.g., cost, side effects), but also from a lack of agreement among researchers, medical practitioners and patients on language framing and describing diag- nosis and treatment. For example, physicians may ref- erence fertility treatments for an infertility diagnosis, even when that diagnosis is unexplained. Inversely, some patients never-self identify as infertile or never re- ceive a formal diagnosis yet undergo fertility treatments. Individuals may conceptualize infertility as a journey, a condition, a disease, or even a social construct and choose language that constitutes their perspective and experiences. Therefore, definitions of infertility are con- textual: shaped by individual and social constructs, em- bedded in systems of oppression.7,10-12 For example, it is not necessary to link motherhood to any individual of any gender identity,12 yet the inability to conceive is still framed as a female problem as cis women are assumed to be natural mothers.7,11 Further, the technical defini- tions presented by many organizations maintain a het- eronormative, cis-gender focus that does not address possible infertility experiences of LGBTQQIA+ couples and/or individuals.12,18 Given the difficulties in describ- ing and defining infertility and the experience of fertility treatment(s), individuals who are already using social media are turning to applications (apps) like IG and Facebook to discuss treatments, ask questions about doc- tors and supportive therapies and offer and receive emo- tional support.12 In our book, You’re Doing it Wrong! Mothering, Media, and Medical Expertise,12 we examined the impact of social media platforms on the conversation around in- fertility diagnosis and treatment; specifically, the long his- tory of the phrase just relax, including how medical language shifted throughout the last century to maintain the notion that female patients could mentally control their fertility. We traced this pattern into the present, where we found it populating the methods of alternative practitioners, social media influencers and even traditional doctors on blogs, YouTube, social media platforms and products like eCourses for fertility. Responding to our own call for future research, we investigated communi- cation around infertility treatment on social media. In this study, we observed the unique ways medical expertise was exchanged on IG before, during and/or after IVF treatments. The multimodal capacity of social media plat- forms like IG allows users to index multiple layers of in- formation and experience by way of a single post, availing themselves of multiple discursive resources at once. In turn, these semiotic practices are recontextualized as the spoken discourse dynamics of healthcare, providing pa- tients new strategies and vocabularies to introduce into their practitioner-patient relationships. Social media use for health information While the earliest conceptualizations, frameworks, and algorithms for what became today’s social media landscape were well underway by 1965, the coding for the World Wide Web was not available until 1991. Be- tween 1991 and today, the majority of individuals in America gained access to the Internet, web search engines and social media platforms, mainly through hand-held de- vices known as smartphones.13-21 After the advent of the Internet and before the popularity of social media, schol- ars began investigating movements of individual users navigating health information collected by search engines as they sought advice and expertise online.22-24 Initially, Internet use focused on desktop computers, and then more portable laptops.19 Today, individuals favor smartphones; as of January 2014, 56% of American adults own a smart- phone.25 The prevalence of smartphone use drastically in- creased mobile, instant, interactive communication through social networking. Today, smartphones redirect health information gathering to smartphone apps and so- cial media platforms.13,26,27 Instagram IG, one of many available social media content-shar- ing platforms, is now the second most used, with over 700 million active users.25 Instagram launched a mere six years ago and quickly became popular because of its photo-forward character and easy editing features.28 In its first three years, IG amassed more than 150 million fol- lowers and was purchased for more than $1 billion by Facebook.28 Users post more than 95 million photographs per day and “like” 4.2 billion times each day.25 To use IG, individuals or groups represented by an account (i.e., Users) can upload (i.e., post) photographs, videos, com- ments, hashtags, links, and other materials (i.e., Usercon- tent). Instagram is designed for smartphone use, since photographs, videos, text and hashtags can only be posted and hyperlinked via the mobile application. While early Internet adopters preferred search engines and WebMD,29,30 millennial parents (and those seeking to be parents), “spent their formative years steeped in per- sonal technology,” according to Bruce Feiler of the New York Times.31 As adults, this demographic favors smart- phones and other portable devices (Apple watch, etc.) to access social media platforms and a range of smartphone apps.26,31,32 Recent research supports the idea that parents today seek information specific to parenting, including medical expertise for their babies and young chil- dren.26,27,32 Deborah Lupton2 addressed smartphone apps used specifically by mothers to monitor their pregnancies and the behavioral patterns and health data of their young infants (e.g., eating and sleeping times, weight). Gender scholar Sophia Johnson26 referred to the device-ification of motherhood, an intrusion by and integration of social media platforms in the everyday lives of mother and child [Qualitative Research in Medicine & Healthcare 2019; 3:7875] [page 3] Article No n- co mm er cia l u se on ly through smartphone apps, tablets and mobile devices.26 In agreement with Lupton,2 we also observed the increasing dataveillance integral to mothering with young children and the shift from Dr. Google and WebMD and toward social media and interactive apps. In our book, we inves- tigated the ways motherhood and mothering begin earlier and earlier, now including pre-conception. With that, ex- pectations for behavior and approaches to conception re- quire engagement in cultural scripts surrounding mothering.12 Not surprisingly, the research underscores the perva- sive influence of social media platforms such as Face- book, IG, Twitter and others on day-to-day interactions, including seeking medical expertise.26,27,33-35 Surprisingly, IG also provides a space for sharing or resisting medical expertise. During infertility treatments, individuals hoping to conceive or partner with a surrogate engage IG as a space to interrogate medical protocols, seek lay expertise on medical decisions, exchange medicines and engage in a semiotic community, bounded by unique acronyms, lan- guage constructions and hashtagging patterns. The inter- actions popular on IG (e.g., hashtagging, tagging accounts, etc.) are affordancesor aspects of usage and de- sign that assist users in finding popular posts and content with which to engage. Experts and those who seek medical expertise and health information continue to assess the utility of hashtags, a word or short phrase (without spaces) beginning with a #, which creates a digital tag associated with a user’s post; this is a type of affordance. Hashtags can be used across social media platforms, including Facebook and Twitter. On IG, servers use hashtag metadata to group similar posts, make posts searchable, and track popular themes. For users, hashtags encourage the creation of niche groups within IG, referred to as a hashtag public.36,37 For example, in this study, hashtag use on IG guided members of the #ttc (trying to conceive) community to medical expertise on IVF. Hash- tag publics assign a connotative, relational context to cer- tain words, phrases or acronyms that represent meanings shared among members, which are then signified by the hashtag.37,38 Hashtag publics may also use hashtags in a way that differs from the denotative meaning of the term or phrase such as #fail representing a social mishap or embar- rassing moment rather than unsatisfactory performance.39,40 The social impact of hashtags is significant, as they now appear in everyday language, when a person might say aloud, “hashtag tired!” Hashtag publics can also act as counterpublics for users who resist normative ideologies and narratives (e.g., trans individuals using hashtags such as #trans and #transgender with uses in the millions).30,37,38 Irrespective of the narrative, hashtags are integrated within the text posted with images, so through these embedded hashtags, members are connected to the information they are seeking. Currently, scholars are studying the use of hashtags to understand better how hashtag users find or dis- tribute health information and medical expertise.41-44 Originally purposed as topical markers, the first hash- tag appeared in an August 2007 tweet by Twitter co- founder Chris Messina.45 Hashtags are now used to represent a range of meanings in various social media messages, those these meanings vary and remain con- tested. Scholars analyzing semiotics on social media plat- forms refer to hashtagged text as hypertextual and view affordances either as intertextual elements emergent from or built within platform designs. In online domains, inter- textuality could be seen as the connection between texts and other pages of the same website through hypertextu- ality (e.g., hyperlinks, icons, or buttons), where this hy- pertextuality is site-internal.38-40 Hashtag publicscoalesce around the use of these hyptertexual tags, which further bound, define and redefine these groups as they create meaning together.33,35-37 With consideration to Barthes’ traditional explanation of semiotics – where the signified and signifier develop signs – hashtags act as a signifier where the signified is the underlying meaning of the word or phrase.46 A hashtag is a semiotic act, and as such, illustrates that hashtags hold socially constructed meanings.38,47,48 A hashtag acts as a marker, guide and record on social media platforms, sig- nifying a distinct, constructed meaning that is understood and used by a hashtag public, bounding that public and signaling community and social similarity to potential members.42-44,47 For example, the hashtag #ttc could rep- resent several different acronyms. However, when it is paired with #infertility, it is understood to mean trying to conceive; in that usage, those trying to conceive through infertility treatments will find accounts to follow by searching or engaging with the hypertext of that usage. To guide our study, we explored how medical expert- ise was communicated, negotiated and exchanged within the infertility/#ttc community. We considered the impor- tance and increasing use of social media platforms to seek medical expertise and how these exchanges (both inter- textual and hypertextual) function within the particular hashtag public of the infertility/#ttc community. Materials and Methods In our study of Instagram, we utilized Kress and van Leeuwen’s49 multimodal discourse analysis framework50 because it allowed us to explore the complex interrela- tions of social media discourse. With the rise of social media platforms as sources of news and information, in- cluding health and medical information, a multimodal discourse analysis allowed the authors to examine the text, image as intertext, and hypertext (e.g., hashtags), both separately and together.37,47 As Berger51 discussed well before the advent of the Internet, images are as im- portant as texts since they also contain multiple mean- ings, communicate with and impact the viewer in varying ways. Mitchell52 urged scholars to recognize the ways im- [page 4] [Qualitative Research in Medicine & Healthcare 2019; 3:7875] Article No n- co mm er cia l u se on ly ages can be understood as a kind of language. Later, he referred to imagetexts;53 Jensen54 explained imagetexts in her work on blogs, noting that “visual images are often read using text, associative images are generally accom- panied by text, and more often than not, text and images are presented together (p. 4). As Jensen54 argued, websites provide the rich material with which to examine hyper- textual arguments (p. 4) given that images and text appear together in new and revolutionary ways. When perform- ing a multimodal discourse analysis on an image-forward social media platform like IG, the notion of imagetexts is useful. And while imagetext allows for the investigation of both text and image, multimodal discourse analysis goes further, by engaging intertextual and hypertextual re- lationships as well, including text that is live or linked to other locations or accounts on social media platforms. As Miller47 argued, multimodality closely parallels hypertex- tuality and intertextuality which were both integral to our analysis. We were interested in how texts develop throughout the analysis period (e.g., two weeks) and, as comments collect under images, references to other texts and digital hyperlinks emerged. As such, we monitored IG’s affordances in the design and performance functions (e.g., hashtags, comments, images, links in bio, formatting of text, and filters).55 Data collection After Institutional Review Board approval, the authors and another research assistant, Rachel Ayers, collected 199 IG posts between October 10th and 24th of 2016. The study’s sample consisted of IG users and followers who posted comments/info and visual media content on their own or others’ IG feeds related to a diagnosis of infertility and/or fertility treatment (Johnson received numerous in- fertility treatments over a period of four years; however, she did not utilize IG during this period). The search term infertilitywas entered into IG’s hash- tag search engine to identify posts related to infertility, which revealed the most popular infertility-related hash- tags and formed our sample of public posts to be analyzed. We gathered individual posts through hashtag use. The re- search team selected hashtags with the largest number of uses that referenced REI treatment. In our first exploratory views, it became evident that a majority of the posts fo- cused on IVF treatment cycles. As such, we narrowed the treatment focus to IVF given that it requires a serious commitment of time and monetary resources – the proce- dure happens throughout several months and even if there is insurance coverage, costs many thousands of dollars.8,15 Furthermore, patients perform much of the treatment on their own, out of the office (e.g., injections).8,15 The first two hashtags we chose to appear as a setbecause they are often paired in posts. We selected: i) #ivfsisters/ #ivfcom- munity; ii) #IVFfail/#infertilitysucks; ii) #ivfjourney. The research team also selected what we termed out- lier hashtags, meaning hashtags that frequently accompa- nied IVF posts, but do not always address the IVF cycle (e.g., shots of a patient on vacation). These were: i) #ttc; ii) #ttcsisters; iii) #infertility. Selected posts used at least 1 of the total 7 hashtags, and a majority used 2 or more. We collected 199 images and analyzed the first 25 comments (or all if there were fewer) for types of discourse (e.g., discourse around med- ical expertise) and themes emerging from that overarching discourse. Any IG user who used our searched hashtags may have appeared in this study’s sample. The specific demo- graphics of these individuals remains unknown, given that users can reveal as little or much about themselves as they choose (e.g., age, real name, geographic location). How- ever, individuals sometimes included specific personal in- formation in their feeds, such as demographic location, medical status, insurance information, and doctors’ names, REI practice names, and their bioinformatics.56 We were vigilant about privacy concerns, and thus this sam- pling procedure allowed us to collect the most popular public posts.47,57-61 We did not interact directly with par- ticipants; to further protect our participants, identifying information was blurred within images. Once the team collected images, screenshots were taken with a smartphone and then uploaded to a secure site. Once uploaded, those images were arranged in lin- ear rows, with the singular image and caption, then the first 25 comments (not all posts had 25 comments). De- pending on the number of comments, some rows in- cluded seven images and others had as few as three. To label the images, we called each spreadsheet INSTA with a letter for each sheet, which resulted in INSTA A through INSTA J, with each row numbered and images were identified by sheet and row. For example, B-7. Our files were stored on Johnson and Quinlan’s password protected computers. Due to the high quality of the im- ages and the number of images (posts and comments) the files were too large to email and had to be stored in a locked folder on our secure Dropbox. Once we col- lected enough images to populate ten spreadsheets, with comments for each image, we reached saturation. For in- stance, comments and posts began to repeat and no new information was gathered. Throughout, we reviewed the data collection process to ensure we could approximate the IG experience of scrollingwithin a spreadsheet. As with other social media platforms, IG images and resulting discourse are best viewed in situ.47 However, images and accounts can be deleted at any time, and the hypertextual hashtag feed changes every few seconds; to get a snapshot of the dis- course during the two weeks of our study, we captured the images and place them in a more permanent location (e.g., a spreadsheet) for analysis. Data analysis As qualitative researchers, we proceeded both induc- [Qualitative Research in Medicine & Healthcare 2019; 3:7875] [page 5] Article No n- co mm er cia l u se on ly tively and deductively. We immersed ourselves in the data and analyzed and re-analyzed our dataset inductively, searching for commonalities and repetitions in the lan- guage, images, and intertextual and hypertextual dis- course.47 The team met on three occasions to create a system for notes and observations, to discuss themes emerging the images, text and hypertext and to compare personal research notes and observations for reliability re- garding emerging themes. The team used a series of spreadsheets from which to make notes on observations. To the far left of each spreadsheet, we used a column which listed the main and outlier hashtags for each post, as well as the number of user followers and notes. These notations and observations assisted us in distilling our themes and providing examples for each. After the team agreed on initial themes we selected representative images and blurred them to obscure personal information and po- tential identifiers. We also proceeded deductively, by recording frequen- cies by which patterns occurred, and situating those patterns within an interpretive context for discussing our overarch- ing themes. The major themes were classified through pri- oritizing, synthesizing, and abstracting.62 In the last reading, the researchers analyzed the intersemiotic relationships conveyed by hashtag publics such as the interaction of vi- sual, textual and hypertextual elements, including both ex- plicit content and implicit connotations.63 To explore intertextuality and hypertextuality in our data set, we made notes on the affordances we observed (e.g., intertextual el- ements such as likes and emoji in response to images and/or comments) and hypertextual elements such as hashtags and tagging practices and how these elements impacted dis- course around medical expertise.47 After analyzing images, hashtags and other hypertext through up to 25 comments, the research team coded the discourse for medical expertise such as information on re- actions to (in)fertility treatment and evidence of preg- nancy/pregnancy symptoms. We found that the overarching discourse centered around patient perceptions of (and most often resistance to) medical expertise. Our final three themes included: treatment protocol choices, treatment side effects, and pregnancy diagnosis/confirma- tion. In the findings, intertextual and hypertextual data were included to provide evidence for the researchers’ in- terpretations of semiotic signaling.47,64 Results and Discussion As a result of the availability of social media, many IG users ask for and provide medical expertise before, during and after IVF treatment and in response to posted images and accompanying text. At times, we perceived the personal struggle of individuals seeking medical ex- pertise, projected into the public space of the social media world, and these posts were far more evocative, even dis- comforting. For example, the image an account holder posted of her cervical mucus, wondering if its viscosity hinted at pregnancy. Users offered followers hypertext with hashtags, #ivfsuccess, #ivfmeds, #ivfsucks while also tagging other account holders to come back to the discourse for further engagement, or as a marker of infor- mation they may need later (e.g., after the tagged account holder started her own IVF cycle). Importantly, each of these inclusions represents particular choices of self-dis- closure, and engagement with the discourse around the power of medical expertise during the IVF process. These divulgences and the paring of intertextual and hypertex- tual affordances (e.g., hashtags, tagging, acronym use) comprise a multimodal form of communication, which can deepen bonds between users and obscure information for outsiders not practiced in the language of IVF treat- ments in the #ttc community. Treatment protocol choices During infertility treatment, particularly IVF, different practices create and rely on particular treatment protocols, including pre-treatment tests.65 For example, some proto- cols privilege certain drugs (e.g., Gonal-F®, Menopur®), some prefer freezing embryos prior to transfer, though age limits vary (e.g., some require an egg donor for cis female patients over 40), some practices prefer progesterone in- jections (via an injection) to creams or suppositories. Practices rely on their protocols to produce superior re- sults (i.e., confirmed pregnancies), which they can adver- tise publicly to attract clientele and as a result, can be resistant to patient requests to adjust or alter the preferred protocol.11,65 Against this backdrop, #ttc and IVF patients use social media platforms to ask about the protocols of others in their hashtag public, and what to discuss with their doc- tors, “Any suggestions for questions to ask or tests to try?” And below a post about a failed cycle a commenter of- fered, “…Maybe you could also do a repeat pregnancy loss panel?” Another commenter disclosed a miscarriage they experienced and the poster provided medical expert- ise in response, including urging specific testing and ask- ing about hormone levels: “I am sorry to hear about your miscarriage [broken heart] did you PGS test your em- bryos… Don’t go into your 3rd cycle without knowing of you have endometriosis…have you been tested for NK cells? APA panel? Immune issues in general? What is your TSH now?” These types of comments suggested medical expertise but also bound the community through lan- guage, acronyms and procedural references that would be confusing or opaque to outsiders, even medical practition- ers outside the field of reproductive endocrinology. The implicit message is if you know, you know. But posts also addressed when and how to resist doc- tor-mandated treatment choices. For example, one account seeks to answer protocol-based and other medical questions and in one post, focused on an endometrial receptivity array test or ERA. Commenters questioned whether they [page 6] [Qualitative Research in Medicine & Healthcare 2019; 3:7875] Article No n- co mm er cia l u se on ly would be goodcandidates for the test. However, other com- menters entered the discourse to remind IG users what this test was reallyabout: “Don’t forget to add that it’s basically a mock cycle so you are pumping your body with all the synthetic hormones you are just not getting the benefit of retrieving your eggs.” In these instances, individuals with unknown training and expertise sought to draw the conver- sation back to what could be a hurdle for many women un- dergoing the test: all the treatment, and no egg retrieval. The implicit connotation is if you are going through a cycle you’ll want something to show for it. Items in images included medical instruments, medi- cines, supplies, and even bruised and swollen abdomens and marked injection points. Multiple images were paired with extensive commentary using the terminology of tech- nical, medical expertise. One post included a woman pic- tured in her kitchen, hand on hip, head angled up at the camera with a slight smile – here, the user signaled to the viewer that she is competent, ready to perform her injection at home and armed, literally and figuratively, with the ca- pacity and information to complete this treatment. She also complicated stereotypical ideations around infertile bodies; the user appears to be white and cis female, but she also young (early 20s?), with a full arm of tattoos and a lip ring, she signals to the viewer that she understands this tension with the hashtag #thisiswhatinfertilitylookslike. While the image might inspire confidence and is suggestive of self- sufficiency, the text confirmed this inference, while utiliz- ing the language of medical expertise: A lot of people have asked me what that nasal spray is for and I’m not sure my explanation was very clear. Here’s a better one: Nafarelin is a gonadotropin-releasing hormone agonist (GnRH agonist). Its proposed mecha- nism of action is the desensitization of pituitary GnRH re- ceptors leading to a decrease in gonadotropin release... Instagrammers may encounter this image, the accom- panying text, and the hypertext hashtagged beneath the picture (e.g., #infertilitysucks, #ttcjourney) by searching any of these hashtags and/or through other accounts fol- lowing this user. While informative, the image was also empowering, illustrating the courage and confidence one can achieve in an otherwise isolating and difficult process. Some of the posts operated similarly to a Yelp review for fertility practices. One image has a purple background with a text bubble: “Has anyone been to Seattle IVF?” The text underneath the image read: “…I would love to hear your experiences if you’ve been there!” One com- menter offered, “My cousin did and had a great experi- ence.” A former employee from a local, competing REI practice discussed their former employer and then stated, “I hope it works out with Seattle IVF! They seem like a great company who cares about the patients more than just the $$$ [heart emoji]!” Alongside this recommenda- tion to visit Seattle IVF the poster attempted to soften the comment with symbolism – the commenter never said money, but the dollar sign had explicit connotations. The heart emoji seems to offer good luck and well wishes, but it is paired with a direct critique of the way some REI em- ployees saw their practices and their bosses. The commen- tary was a warning to newcomers too; at some practices, it was not altruism, it was business. Also connected to protocols was the practice of post- ing pictures of medication that the IG user intends to gift, or has received from another member in the community; these posts are also hyptertextual (e.g., #ivfmeds, #menopur). Regarding the use of these medications, prac- titioner opinions vary: some refused to participate in the practice, others understand the difficulties in paying for medicine, particularly without insurance and allow their patients leeway; others take unused medication from their patients and keep it for other patients. As User 2 reported: “I am so happy I had a doctor who bucked the rules and took all my unused meds for patients without IVF cover- age” (Figure 2). Some posters offered to ship medicine: “[lists medi- cines] these need to be used in the next three weeks – so if you need them, I’d love to help you out & donate them to you!! DM me so we can figure out shipment!” In the comments, some wondered about the legality of shipping medications through the mail: “Question is it legal to ship medication…I think the post office throws them out if it isn’t from a pharmacy.” The sale of medication between individuals through the mail is illegal. The sale and ex- change of medications is also risky – drugs can be tam- pered with and sent or stored in improper temperatures. However, that is not what is happening here. Individuals are donating medications to one another, which is an im- portant distinction. Still, shipping medications through the mail can lead to a charge of mail fraud.66,67 Despite the risks, the most common response to these posts are others providing offers of left-over drugs or further requests for medication: “I have some that expire next month, inter- ested?” “I have so many left over!! Not sure what to do with them.” There was an even a partner advocating for medications: “Hi, my wife and I are starting our IVF jour- ney and none of our meds are covered. Does anyone have any they are willing to donate?” We cannot ascertain the actual incidence of medicine exchange, but the number of posts showing gifted drugs suggested that IG posts can facilitate the practice. Regarding sharing/shipping medi- cines, many posters showed what they received, discussed the market value of fertility drugs, and celebrated their ability to return to treatment. The economic pressure of Fertility, Inc., the biomedical system driving reproductive endocrinology today is well studied; what we discovered is that social media presents a place to explicitly and im- plicitly work around some of these economic constraints to maintain REI protocols, regardless of cost.16,68 While the text focused on gratefulness for the medica- tions or desires to share it with someone else, the images and hashtags do specific work here. Using hashtags such as #ivfmeds, #menopur could potentially help users track [Qualitative Research in Medicine & Healthcare 2019; 3:7875] [page 7] Article No n- co mm er cia l u se on ly medication donors through IG’s search engine. The im- ages are intertextual too – they illustrated medications with valid names and packaging and are shown in decent lighting and offer users close-ups to prove they are intact, unadulterated and safe, whether or not they are. But when medicine is proferred, it is always pictured. Another rea- son for including a picture is that the same drug can be packaged in pens or vials and with varying doses, so a pa- tient has to see the particular doses offered again to un- derstand if or how it overlaps is there parallel with a patient particular protocol. Treatment side effects Infertility drugs can cause many side effects including weight gain and hot flashes, bloating and achiness. During IVF treatment patients take a host of these drugs simulta- neously to facilitate ova growth and prepare for egg re- trieval. Patients often turned to their hashtag public for expertise on side effects related to the medicine specifically (e.g., how to prepare and draw up meds for injection) and then to discuss their reactions to these drugs. Throughout this theme, the research team noted that within the com- ments section, users relied on symbolic imagetext like emoji and hypertext like hashtags and account tagging, helping other users come back to the post and add to the discourse or use the information during their own treatment cycles. Altogether, these elements create an imagetext tap- estry, which included numerous perspectives, sources of in- formation and forms of knowledge, within and because of intertextual and hypertexual discourse. One user posted a picture of their medicine in the background with a sizable needle in the foreground and asked, “anybody else had blood SPRAY out after this in- jection?” While the poster did not provide context, the treatment issue was implicit and commenters responded without requests for further clarification: “I hope the next one is easy peasy!” Others added their own experience and suggested varying angles or positions to lower the chance of drawing blood. There is explicit solidarity here through the expertise proposed, but also emotional sup- port through encouragement and understanding from oth- ers that used the drug, knew what it was for and already used it safely. Another user posted a picture of the drug [Merional®] ampoule: “so…any tips on snapping the glass so it doesn’t go everywhere?” Commenters recommended a range of [page 8] [Qualitative Research in Medicine & Healthcare 2019; 3:7875] Article Figure 2. Medicine distribution, Menopur®. No n- co mm er cia l u se on ly methods and identified themselves as both patients and technical experts. One directed: “as a nurse we fold gauze (or a napkin if you don’t have gauze) and pinch the glass with the gauze over it and snap it at the line. This allowed any broken pieces to end up in the gauze and not every- where else. Hope that made sense (blushing smiley face emoji).” The nurse cancellare gave detailed expertise but then ends with an emotive tone, symbolizing friendliness and perhaps even humility – almost an apology for her ex- pertise – with the emoji. One IG poster bemoaned the “revenge of teenage acne/IVF problem spots” and shared her journey from great skin to acne as a result of IVF treatment. In the image, she displayed before and after pictures, but show- ing regression, with what appeared to be flawless skin to skin with obvious blemishes. The user linked 20 hashtags in her post, among them: ivfsupport, ivfmeds, ttc, ivfwith- icsi and shareyourstory. The poster uses the term IVF SKIN in all capital letters twice in the caption, suggesting that this may be a common experience with others who have received treatment. The number of hashtags used at the bottom also implies the user is trying to draw people to her post, potentially to engage in dialogue, invite the stories of others, commiserate or find expertise on battling hormonal acne. The other hypertext in this is the user’s location, which is marked as her REI practice, so it is also possible IVF skin is a term she discussed with her medical team, though she does not intimate there is a cure. There are four follower comments, but all remarked on the strength of IVF patients. While the poster did not explic- itly ask for support, the use of a series of selfies illustrat- ing a treatment side effect such as acne, which can foster pubescent angst, embarrassment, anxiety and shame, rep- resented an intertexual attempt to situate personal experi- ence and boost confidence while seeking affirmation. Pregnant diagnosis/confirmation In our previous qualitative research on infertility treat- ment support participants disclosed practices REI practi- tioners advised them on what to avoid.15,16 These include but are not limited to: sharing medications and taking pregnancy tests before the beta test (e.g., a blood test at an office). REI practitioners we interviewed realize pa- tients often take pregnancy tests anyway, but they con- tinue to discourage the practice, which can cause anxiety and emotional upset from false negatives or excitement and then devastation from false positives. There are posts discussing when to start home pregnancy tests after em- bryo implantation, all communicating that patients know better but then providing detailed images of test results, asking for feedback on symptoms, and querying other IG users for positive stories and experiences. All of the dis- course around pregnancy testing reflects resistance to medical expertise on testing early. While testing is emo- tionally wrought, it reflects the individual desire to rein- state some control over the process, to know what is happening in the body after treatment and before the beta. Some pregnancy-related posts showed pictures of pregnancy tests with results or discussed if positive lines were growing darker and more apparent, while others dis- cussed testing out or monitoring the drop of synthetic HCG from the trigger shot (e.g., Ovidrel®), which prompts ovulation before egg retrieval. Once that level disappears, some continue testing to see if their body pro- duces its own HCG and thus a positive pregnancy test, in- dicating the IVF cycle was successful. However, using pregnancy tests from the drugstore, with dye-based lines instead of digital readings can produce confusing results, so some turn to IG for feedback. There is also an eco- nomic constraint here – low-cost pregnancy tests are ac- cessible (e.g., at the Dollar Tree, where tests are $1 if you buy 4 or more) and attractive for patients struggling to cover treatment costs, but these tests are notoriously un- reliable. This tension is well understood in the #ttc com- munity, especially among those on self-pay for meds and treatment, so, IG users present visual and textual supports when posts seek clarification on test results. User 3 posted an image of three pregnancy tests, each from successive days with a positive line becoming slightly more noticeable across the three days. In the picture the tests are labeled as 5DP5DT, 6DP6DT, 7DP7DT, meaning 5, 6, or 7 days post-five-day transfer, etc. The text says, “Last Ivf cycle, I didn’t get a line UNTIL day 7 (and it was a squinter to say the least)…However, each day the line continues to get darker and my confidence grows…[little plant emoji; #ivfsuccess].” Commenters were encouraging: “Looking good!!” and “You’ve got a sticky one in there! Keep testing for as long as it makes you feel good” Post viewers understood both the explicit and implicit question the image posed, and responded by confirming the tests be- came more positive over time. There is no contextual in- formation about the picture, including the lighting for the image, the filter used or other information addressing the impact of setting on the appearance of the result lines. Some users asked the question directly through their posts: “I KNOW everyone is different…I just wanted to hear from you guys when your trigger was gone if you tested it out? How many days after the shot were your tests negative?” The user goes on to discuss the progression in the com- ments and is second-guessing her results. User 3 concluded: “If it’s still there I’ll ask when we have our WTF appoint- ment next Friday” (Figure 3). Perhaps the most visually provocative image we en- countered showed the user’s cervical mucus. The poster realized their post was unusual, though the hashtagging within the post made the image discoverable to hundreds of thousands of IG users. The caption read: “This is very TMI, ladies!!! Can you tell me what kind of cm [cervical mucus] this is? I’m having a hard time understanding this cm. [Anxious face with sweat]…Sticky? School glue? Btw, my breasts are still sore. [Face with medical mask emoji] P.S. I’m so sorry if this picture is disturbing you. I [Qualitative Research in Medicine & Healthcare 2019; 3:7875] [page 9] Article No n- co mm er cia l u se on ly might be deleting this pic after I know the answer.” The commenters did not seem deterred by this and instead re- sponded very positively: “I’d call it somewhere between sticky and creamy” and “Not to get your hopes up or any- thing but my body produced stuff like that when I was pregnant,” or “Creamy/school glue I think!” This image is certainly explicit in its signaling – this is clearly a body fluid. Yet the apologetic tone and use of TMI (which means too much information and flags information some users might find too personal) with the image are inter- textual, asking for forgiveness with a parallel demand for confirmation. The responses were effusive and mimic the intimacy of the image by pairing adjectives common in pregnancy manuals and doctor’s pamphlets with encour- aging words, positive emoji and assurances based on an image in an unknown setting. Again affordances matter here – we wondered where the picture was taken (con- text), in what lighting, at what time of day, and posted with what IG filter? All of these contextual elements could change the appearance of the opacity and viscosity of the material on the poster’s fingers and as a result, the actual answer to her question. Research implications Based on our findings, patients before, during and after IVF treatments are using IG to give and exchange medical expertise, but also to build community and sup- port one another (e.g., sharing words of encouragement, donating surplus drugs to each other). The intertextual and hypertextual nature of these posts (i.e., the use of image- text to communicate) makes them accessible to potentially millions of IG users and creates a time-specific record of a dialogue between members of this hashtag public. Given the public nature of this discourse and the personal infor- mation and data that is often shared, there are legal issues to consider. Legally, problems may arise from medicine exchanges, particularly if sending donated drugs is con- sidered mail fraud. Is there liability in situations where IG users make personal health decisions based on IG dis- course, and if so, who would be responsible? Ethical guidelines and legal frameworks are far behind the poten- tialities of social media interactions, and so much of what we observed reflects this gap. On the other hand, the lack of oversight and regulation create meaningful spaces where individuals can build community and impact each other in tangible ways, which is vital given the isolation commonly experienced by fertility patients.11,13,14,17 Considerations of legality also intersect with privacy issues. In our data set, posters included images that re- vealed their full names, birthdates and social security numbers, as well as screenshots of private emails received from their REI practices, with the first and last names of their practitioners publicized. Interestingly, many of these were posted through accounts in which the individual’s IVF-specific account is obscured or hidden so strangers were potentially more likely to view this information than close friends or family. In the email instance, the nurse in question was clarifying a treatment error and commenters revealed the particular medical system and location of the practice. We are unaware if anyone from the practice was aware of this post, and the nurse may or may not have had anything to do with the clinical error in the message. [page 10] [Qualitative Research in Medicine & Healthcare 2019; 3:7875] Article Figure 3. Blur of pregnancy tests, 5DP5DT. No n- co mm er cia l u se on ly Furthermore, it is unclear if IG users understand the searchability of their images based on their use of hash- tags; do users realize that anyone that uses the same hash- tag or searches that hashtag can view their information and potentially use it in ways the poster did not envision or would not approve (e.g., identity theft)? One protection available to users is to set their account to private but in the #ttc community, accounts are often public so users can follow the hypertextual trail to find and support each other.69 And if you have a private profile but leave a com- ment on a public account, your profile is visible and other users can click on your username.69 Because IG delineates the privacy policy in their user agreement,69 it is unclear whether or how Instagram supports users who claim in- vasion of privacy or identity theft; hypertext is an affor- dance, so it leaves a trail by design. Lastly, for some IG users, IG represents a safe space to share their fertility journey because they are not discussing it openly with friends and family. The popularity of hypertext in these posts complicates the idea that simply obscuring one’s identity with a username provides the sort of anonymity expected or desired. For REI practices, it may be valuable for doctors and nurses to understand where patients are seeking and re- ceiving information to better prepare patients to ascertain what is accurate advice. As we have seen from our previ- ous research, telling patients don’t Google is not effective,15 and during IVF cycles, treatments (such as in- jections) happen in the home and generally not during business hours. Patients understandably have questions or concerns but due to long wait times and/or lack of avail- ability, talking to a practitioner immediately might not be possible. Given how quickly some posters turned to IG to gather information or share their frustration with medical experts (4 minutes in the case discussed), it is likely some users will find expertise and support on IG prior to mak- ing contact with a practitioner. On social media platforms, the contingent nature of information is visible – the way expertise is framed, consumed, and resemiotized through inter- and hypertexual discourse continues to confound the expertise binary of lay or technical.12 In this setting, users validate expertise through the nature of the dis- course, by what is posted and what is liked rather than by clear standards and structures external to this discourse. For example, when patients receive support for activities against the recommendation of medical personnel (e.g., pregnancy testing during the two-week wait, or the time between embryo implantation and the beta test). Regarding medicine exchanges, is it possible that med- icine received from another IG user is no longer viable, was accidentally contaminated or damaged during shipping? As these medicines are posted, they become parts of hypertex- tual discourse, but they also lead to offline communication and interaction, including mailing packages and thank you notes. Furthermore, these posts can be removed at any time, making the success of medicine exchanges challenging to gauge and track. Practitioners might consider that patients’ resistance to or questions about protocols can result from expertise gleaned on IG (e.g., information on multiple em- bryo transfers). All of these practicalities highlight the power and complexity of imagetext on social media plat- forms, showing the richness of these spaces for semiotic discourse, as well as the complications introduced by hy- pertextuality. Conclusions Given the availability of social media-based data, there are countless entry points for other scholars to ana- lyze multidmodal discourse on social media platforms. In terms of health issues, other studies could examine hor- monal therapy, surgery, intrauterine insemination and other procedures or treatments distinct from the conver- sations we studied. Future research could query the impact of alternative REI treatments (e.g., meditation, yoga, di- etary changes). A multimodal analysis that seeks out the perspectives of women of color, queer women, trans women, women with lower socioeconomic status and other individuals in treatment would provide more infor- mation about the impact of IG’s affordances such as hash- tag use, tagging, emoticon/emoji use and help interrogate the analytics of silencing alternative voices and perspec- tives. This silencing is a practice Rachel Cargyle (@rachel.cargyle) and others refer to as being shadow banned, wherein IG blocks an individual’s hashtag from appearing in searches without their knowledge (most users find out from their followers). Reviewing a range of these discourses can also provide information on the varying experiences of communities at potential crisis points in the life-cycle of early motherhood.12 Another call for future research is to broaden the sys- tem of data collection. There is no efficient tool to per- form textual and visual analysis of IG imagetext, unless posts are individually collected, compiled and coded. Netlytics and other programs focus on likes and other metrics, but do not have tools available for rhetorical analysis. Radian6, which collects social media data, gath- ers text from IG, Twitter, blogs and Facebook but does not collect images. To analyze the imagetext in context we had to pair screenshots with comments in a linear pro- gression, a tedious effort that worked but could be cum- bersome to analyze. Despite the challenges of data collection, IG is a social media platform with rich and wide-ranging research opportunities. Future scholars might also conduct multi-discourse analysis on other stig- matized diseases (e.g., STIs, diabetes). What types of health information and medical expertise are sought and exchanged in these communities on IG? Future studies could analyze discourses available through Facebook or IG live or stories functions and analyze users’ direct mes- saging to poster content. Finally, researchers could closely examine emoji use within the #ttc community, [Qualitative Research in Medicine & Healthcare 2019; 3:7875] [page 11] Article No n- co mm er cia l u se on ly and in other hashtag publics. Emoji are a semiotic, inter- textual form unique to the social media age, and their role offers complex, fascinating material for study. Today, we engage in a mediated world, where most discourses allow for multifaceted analysis (e.g. of inter- text, hypertext, etc.). Social media is just one level of our mediated realities. On Instagram, users in the #ttc com- munity act as informal patient advocates for themselves and others experiencing IVF treatments, while creating a multimodal, digital extension of the formal, textual med- ical record. For users, the unique nature of the IG plat- form, the power of hyptertext and the utility of images, offers a unique refuge from the biomedical gaze. As one user commented: “I’ve learned one thing, gotta advocate for yourself and pay attention to every detail.” The lack of acknowledgment by individual health practitioners, practices or systems regarding the impact of multimodal discourses on social media platforms ac- tively denies the lived reality of many patients. This de- nial refuses to grasp the dynamic nature of multimodal communication in which expertise is negotiated, recon- textualized, and even disseminated in new forms (e.g., through hyptertext such as #ivfmeds).12,70 With cautious optimism we urge healthcare practitioners and adminis- trators to consider these dynamic, intertextual and inter- connected discourses to keep practitioner-patient dialogue open and evolving in tandem with social media platforms available today and in the future. References 1. Bigelow HR. The moral significance of sterility. Obstetric Gazette 1883;16:1-24. 2. Gregg RS. Sterility versus fecundity, and the divorce evil. Milwaukee Med J 1905;13:60-2. 3. Meaker S. Human sterility: causation, diagnosis, and treat- ment. a practical manual of clinical procedure. Baltimore: Williams and Wilkin; 1934. 4. Centers for Disease Control and Prevention. Infertility FAQs. Available from: https://www.cdc.gov/reproductive- health/infertility/index.htm. Accessed: September 2018. 5. Eunice Kennedy Shriver National Institute of Child Health and Human Development of the National Institutes of Health. Infertility and fertility. Available from: https://www.nichd.nih.gov/health/topics/infertility. Ac- cessed: September 2018. 6. World Health Organization (WHO). Infertility is a global public health issue. Available from: http://www.who.int/re- productivehealth/topics/infertility/perspective/en/ Accessed: September 2018. 7. Sandelowski M. With child in mind: studies of the personal encounter with infertility. Philadelphia: University of Penn- sylvania Press; 1993. 8. Marsh M, Ronner W. The empty cradle: infertility in Amer- ica from colonial times to the present. Baltimore: Johns Hopkins University Press; 1996. 9. Mayo Clinic. In vitro fertilization (IVF). 2019. Available from: https://www.mayoclinic.org/tests-procedures/in-vitro- fertilization/about/pac-20384716. 10. Bell AV. Beyond (financial) accessibility: Inequalities within the medicalization of infertility. Sociol Health Illn 2010;32: 631-46. 11. Johnson B. Personal, interpersonal, biomedical, textual: Vi- gnettes on infertility and fertility privilege. Depart Crit Qual Res 2016;5:76-100. 12. Johnson B, Margaret M. You’re doing it wrong! Mothering, media and medical expertise. Rutgers: Rutgers University Press; 2019. 13. Bute JJ. “Nobody thinks twice about asking”: Women with a fertility problem and requests for information. Health Commun 2009;24:752-63. 14. Bute JJ, Vik TA. Privacy management as unfinished busi- ness: Shifting boundaries in the context of infertility. Com- mun Stud 2010;61:1-20. 15. Johnson B, Quinlan M, Marsh JS. Telenursing and nurse- patient communication within Fertility, Inc. J Holist Nurs 2018;36:38-53. 16. Johnson B, Quinlan M, Myers J. Commerce, industry, and security: Biomedicalization theory and the use of metaphor to describe practitioner-patient communication within Fer- tility, Inc. Womens Reprod Health 2017;4:89-105. 17. Willer EK. Health-care provider compassionate love and women’s infertility stressors. Commun Monographs 2014;81:407-38. 18. Breitkopf E, Rubin L. (Re)productive disorders: The ex- panding marketplace for infertility medicine. In: McHugh MC, Chrisler JC, eds. The wrong prescription for women: How medicine and media create a “need” for treatments, drugs, and surgery. Santa Barbara: Praeger; 2015. pp 37-59. 19. Pew Research Center. Millennials: A portrait of generation next. Available from: http://www.pewsocialtrends.org/files/ 2010/10/millennials-confident-connected-open-to-change. pdf Accessed: September 2018. 20. Press G. A very short history of the internet and the web. Forbes 2015. Available from: https://www.forbes.com/ sites/gilpress/2015/01/02/a-very-short-history-of-the-internet- and-the-web-2/#7f0d5bad7a4e Accessed: September 2018. 21. Zimmermann KA, Emspak J. Internet history timeline: ARPANET to the world wide web. Live Science 2017. Available from: https://www.livescience.com/20727-inter- net-history.html Accessed: September 2018. 22. Eastin MS. Credibility assessments of online health infor- mation: The effects of source expertise and knowledge of content. J Comput Mediat Commun 2001;6. 23. Kata A. A postmodern Pandora’s box: Anti-vaccination mis- information on the Internet. Vaccine 2010;28:1709-16. 24. Sillence E, Briggs P, Harris P, Fishwick L. Going online for health advice: Changes in usage and trust practices over the last five years. Interact Comput 2006;19:397-406. 25. Pew Research Center. Social media update. Available from: http://www.pewinternet.org/2016/11/11/social-media-up- date-2016/pi_2016-11-11_social-media-update_0-03/ Ac- cessed September 2018. 26. Johnson SA. “Maternal devices,” social media and the self- management of pregnancy, mothering and child health. So- cieties 2014;4:330-50. 27. Lupton D. Caring dataveillance: Women’s use of apps to monitor pregnancy and children. In: Green L, Holloway D, Stevenson K, et al. (eds.) The Routledge Companion to Dig- ital Media and Children. Abingdon-on-Thames: Routledge; 2019. 28. Riley-Smith B. How Instagram took over the world in just [page 12] [Qualitative Research in Medicine & Healthcare 2019; 3:7875] Article No n- co mm er cia l u se on ly three years. Telegraph 2013. Available from: https://www.t elegraph.co.uk/technology/10355980/How-Instagram-took- over-the-world-in-just-three-years.html Accessed: Septem- ber 2018. 29. Escoffery C, Miner KR, Adame DD, et al. Internet use for health information among college students. H Am Coll Health 2005;53:183-88. 30. Morahan-Martin JM. How internet users find, evaluate, and use online health information: A cross-cultural review. Cy- berPsychol Behav 2004;7:497-510. 31. Feiler B. App time for nap time: The Parennials are here. The New York Times 2017; Available from: https://nyti. ms/2hDJK7y Accessed: September 2018. 32. Edison Research. Moms and media. 2013. Available from: http://www.edisonresearch.com/wp-content/uploads/2013/ 05/Moms-and-Media-2013-by-Edison-Research.pdf Ac- cessed: September 2018. 33. Khouri JS, McCheyne MJ, Morrison CS. #Cleft: The use of social media amongst parents of infants with clefts. Cleft Palate Craniofac J 2017;55:974-6. 34. Landry T. How social media has changed us: The good and the bad. Business 2 Community. Weblog. 2018. Available from: https://www.business2community.com/social- media/ social-media-changed-us-good-bad-01000104# 8k65A08r 0MVI4XyD.97 Accessed: September 2018. 35. Ridgway JL, Clayton RB. Instagram unfiltered: Exploring association of body image satisfaction, Instagram #Selfie posting, and negative romantic relationship outcomes. Cyberpsychol Behav Soc Net 2016;19:2-7. 36. Olszanowski M. Feminist self-imaging and Instagram: Tac- tics of circumventing censorship. Vis Commun Q 2014; 21:83-95. 37. Olszanowski M. The 1x1 common: The role of Instagram’s hashtag in the development and maintenance of feminist ex- change. In: Rambukkana N, ed. Hashtag Publics. New York: Peter Lang Publishing; 2015. pp 229-242. 38. Zappavigna M. Searchable talk: the linguistic functions of hashtags. Soc Semiotics 2015;25:274-91. 39. Adami E. What’s in a click? A social semiotic framework for the multimodal analysis of website interactivity. Vis Commun 2014;14:133-53. 40. Hiscott R. The beginner’s guide to the hashtag. Mashable [online curated blog] 2013 Available from: https://mash- able.com/2013/10/08/what-is-hashtag/#6XoZi1AZpPqO. Accessed: 2019 Jan 25. 41. Tiidenberg K. Odes to heteronomativity: Presentations of femininity in Russian-speaking pregnant women’s Insta- gram accounts. Int J Commun 2015;9. 42. Ahmed W. Public health implications of #ShoutYourAbor- tion. Public Health. 2018;163:35-41. 43. Dorfman RG, Vaca EE, Mahmood E, et al. Plastic surgery- related hashtag utilization on Instagram: implications for ed- ucation and marketing. Aesth Surg J 2017;38:332-38. 44. Ucar T, Culpan M, Caskurlu T, et al. The activity and dis- cussion points of #Circumcision through Twitter; a mi- croblogging platform. Int J Impot Res 2018;30:249-52. 45. Doctor V. Hashtag history: When and what started it? Hash- tags.org [online curated blog]. 2013 Available from: https://www.hashtags.org/featured/hashtag-history-when- and-what-started-it/. Accessed: 2019 Jan 25. 46. Barthes R. Elements of semiology. London: Jonathan Cape; 1967. 47. Miller AF. Redneckaissance: Honey Boo Boo, Tumblr, and the stereotype of poor white trash [dissertation]. Columbia, South Carolina: University of South Carolina; 2015. 48. de Saussure F. Course in general linguistics. London: Fontana; 1974. 49. Van Leeuwen TV, Kress G. Multimodal discourse: the modes and media of contemporary communication. London: Hodder Arnold Publishers; 2001. 50. Plastina, AF. Patient (mis)understanding of prescription drug ads in social media: Multimodal discourse analysis of eDTCA. In: Maci S, Sala M, eds. Insights into medical com- munication. Bern, Switzerland: Peter Lang Publishers; 2015. pp. 189-212. 51. Berger J. Ways of Seeing. London: British Broadcasting Corporation; 1972. 52. Mitchell WT. Iconology: image, text, ideology. Chicago: University of Chicago Press; 1986. 53. Mitchell WT. Picture theory: essays on verbal and visual representation. Chicago: University of Chicago Press; 1994. 54. Jensen RE. The eating disordered lifestyle: Imagetexts and the performance of similitude. J Am Forensic Assoc 2005;42:1-18. 55. Bucher T, Helmond A. The affordances of social media plat- forms. In: Burgess J, Poell T, Marwick A, eds. The Sage hand- book of social media. London: Sage; 2017. pp. 223-253. 56. Fallon K. Streams of the self: The Instagram feed as narrative autobiography. In: Proceedings of the interactive narratives, new media & social engagement international conference. Toronto: University of Toronto Press; 2014. pp. 4-60. 57. Borzekowski DL, Schenk S, Wilson JL, Peebles R. e-Ana and e-Mia: A content analysis of pro-eating disorder web sites. Am J Public Health 2010;100:209-38. 58. Dylko IB, Beam MA, Landreville KD, Geidner N. Filtering 2008 presidential election news on YouTube by elites and novelties: An examination of the democratizing potential of the internet. New Media Soc 2012;14:832-49. 59. Hwang KO, Fahren K, Johnson CW, et al. Quality of weight loss advice on internet forums. Am J Med 2007;120:604-9. 60. Hwang KO, Ottenbacher AJ, Green AP, et al. Social support in an internet weight loss community. Int J Med Inform 2010;79:5-13. 61. Van Hoof W, Provoost V, Pennings G. Reflections of Dutch patients on IVF treatment in Belgium: A qualitative analysis of internet forums. Hum Rep 2013;28:1013-22. 62. Saldaña J. The Coding Manual for Qualitative Researchers. Thousand Oaks, CA: Sage Publishing; 2009. 63. Rambukanna N. FCJ-194 From #racefail to #Ferguson: The digital intimacies of race-activist hashtag publics. Fibrecul- ture J 2015;26:159-88. 64. Potter J, Wetherell M. Discourse and Social Psychology: Be- yond Attitudes and Behaviour. Thousand Oaks, CA: Sage Publishing; 1987. 65. Jungheim ES, Meyer M, Broughton DE. Best practices for controlled ovarian stimulation in IVF. Sem in Rep Med 2015;33:77. 66. CBSNEWS.com. Buyer beware: Fertility drugs online. CBS News Online [online news source] 2000. Available from: https://www.cbsnews.com/news/buyer-beware-fertility- drugs-online/ Accessed: 2019 Jan 25. 67. Levy L. What this “army of infertility warriors” did for complete strangers going through IVF. PopSugar [online curated blog] 2017. Available from: https://www.pop- sugar.com/family/Strangers-Donating-IVF-Medicine. Ac- cessed: 2019 Jan 25. [Qualitative Research in Medicine & Healthcare 2019; 3:7875] [page 13] Article No n- co mm er cia l u se on ly 68. Mamo L. Fertility, Inc.: Consumption and subjectification of lesbian reproductive practices. In: Clarke AE, Mamo L, Focscket JR, et al., eds. Biomedicalization: Technoscience, health and illness in the US. Durham, NC: Duke University Press; 2010. pp 173-196. 69. Instagram. Controlling your visibility. Available from: https://help.instagram.com/116024195217477/?helpref=hc_ fnav&bc[0]=Instagram%20Help&bc[1]=Privacy%20and% 20Safety%20Center Accessed: September 2018. 70. Quinlan M, Johnson B. #motherhoodishard: Narrating our re- search and mothering in the postpartum stage through texting and social media. Health Commun 2019; available from: https://www.tandfonline.com/doi/full/10.1080/10410236.201 9.1587694 [page 14] [Qualitative Research in Medicine & Healthcare 2019; 3:7875] Article No n- co mm er cia l u se on ly