The 3i Conceptual Framework for Recognizing Patient Perspectives of Type 1 Diabetes During Emerging Adulthood

IMPORTANCE Early emerging adulthood, the developmental life stage roughly spanning between the ages of 18 and 24 years, can be a challenging time for individuals living with type 1 diabetes, being associated with an increased risk of acute complications, loss to follow-up, and the emergence of long-term diabetes complications. Few evidence-based interventions exist, and practice guidelines instead emphasize the importance of delivering individualized, developmentally appropriate care. OBJECTIVE To inform the provision of tailored care and education by illuminating the individual lived experiences of emerging adults with type 1 diabetes. DESIGN, SETTING, AND PARTICIPANTS This qualitative study used in-depth interviews conducted with 33 emerging adults living with type 1 diabetes between October 14, 2016, and May 16, 2017. Participants were recruited from 2 urban Young Adult Diabetes Clinics in the city of Toronto, Ontario, Canada, using a convenience sampling approach. Narrative analysis of participants’ accounts was conducted inductively to define story typology. conducted with 33 participants aged to 24 years at interview, years duration, analysis


Introduction
At its biomedical core, type 1 diabetes is a chronic disease that is characterized by insulin deficiency and elevated blood glucose. In addition to the biomedical nature of the disease, there is a human experience influenced by social interactions and relationships. 1 Early emerging adulthood (roughly between the ages of 18 and 24 years) presents a unique set of circumstances for the type 1 diabetes experience, exemplifying the intertwining between the biomedical and the social. This period of life is characterized by emotional, social, geographical, and fiscal changes 2 and its demands can conflict with the often inflexible demands of diabetes self-management, coinciding also with an expectation that the individual will transition from the stability of their familiar pediatric care professional to a new adult health care setting. 3 Against such a backdrop, it is little wonder that traditional evidencebased medicine strategies have fallen short, leaving early emerging adults with type 1 diabetes at increased risk of loss to follow-up, 4 deterioration in glycemic control, 5 acute complications, 6 and the emergence of long-term diabetes complications. 7 Indeed, in the large T1D Exchange registry, only 14% of individuals aged 18 to 25 years achieved the guideline-recommended target hemoglobin A 1c (HbA 1c ) level of less than 7%, the lowest percentage of any age group. 8 Despite the attendant risks of the early emerging adult years, research into models of transition care has mostly failed to yield the evidence necessary to inform firm guideline recommendations.
Instead, most clinical practice guidelines for transition-age patients with type 1 diabetes emphasize that care must be "individualized and developmentally appropriate," 9 yet there is no standard definition for developmentally appropriate health care. 10 To better understand the lived experiences of emerging adults with type 1 diabetes, we set out to collect and analyze the stories that they tell.
Aligned with the near-ubiquitous use of social media platforms by this population, we conjectured that the self-perceptions of emerging adults are inextricably entangled with their stories (or narratives). 11,12 Narratives contain perspectives on lived experiences as they describe characters that react to environmental cues with emotional responses. 13 Health care professionals are exposed to patients' accounts of their experiences every day. However, these accounts are rarely scrutinized in a systematic manner. Accordingly, grounded in narrative theory, [14][15][16] we set out to address the question: how does type 1 diabetes fit into an emerging adult's evolving life story?

Narrative Approach
The methodology we used was informed by Frank's 16,17 conception of narrative typology, which posits that when patients narrate illness experiences, they do so using distinct styles. By explicating the distinct styles that participants used to tell their stories, we sought to better understand how participants assigned meaning to their experiences with type 1 diabetes. 18 Not only what is said, but how it is said, is important when using this narrative approach 12 (ie, not only the facts contained within a statement but the words and expressions used to give meaning to the statement using analogy, context, or description, for instance). In this sense, the interviews that were performed can be considered a site for identity interpretation, in which participants spoke of who they were, who they are, and who they thought they could become. 19

Sampling and Recruitment
The research took place at 2 Young Adult Diabetes Clinics in Toronto, Ontario, Canada, and 1-time, in-person interviews were conducted between October 14, 2016, and May 16, 2017. Patients were eligible to participate if they were between the ages of 18 and 24 years, proficient in English, and diagnosed with type 1 diabetes for 1 year or more. Health care professionals introduced the study to eligible patients at the end of a routine clinic visit, and interested individuals met a member of the research team (B.M.), who described the study, collected contact information, and provided prospective participants with informed consent forms to review in their own time. At a later date, interested individuals were contacted (B.M.) to schedule an interview. All participants provided written informed consent. Ethics approval for the study was received from the research ethics boards of St. Michael's Hospital and Women's College Hospital. Given that the goal was to elicit a range of participant experiences and perceptions (but not all possible perspectives), a convenience sampling approach was used. 20 We approached sampling broadly and were able to include participants from a wide spectrum of sociodemographic and ethnic backgrounds. We followed the Standards for Reporting Qualitative Research (SRQR) reporting guideline in the reporting of this study.

Data Collection
Interviews were conducted by one of us (B.M.) in a private room at St. Michael's Hospital or Women's College Hospital. The interviewer was a master of science student at the time of conducting the interviews, with no prior relationship to any of the participants. Each interview was audio recorded and transcribed verbatim. The median (range) interview length was 58 (10-125) minutes. Interviews sought to elicit in-depth descriptions about life with diabetes from the first-person perspective. 21

Narrative Analysis
Transcripts were deidentified, password protected, and encrypted. Accuracy checks were performed to ensure that the transcribed text aligned with the audio recordings. A narrative analytic approach to interview transcripts was adopted. 22 Significant features of the individual narratives, such as content, tone (language used), and form (common perspectives and discursive elements within a participant's interview), were analyzed to interpret distinct styles of storytelling exhibited by participants. 23 A strategy of multiple readings and constant comparison of transcripts was used throughout the analysis. 24 Questioning and theorizing related to the evolving interpretations was iterative and sought to connect the findings to the literature. 25 While the interviewer (B.M.) led the qualitative analysis, analytic meetings were held with all investigators throughout the study to review the analytic framework and discuss evolving interpretations. By the 22nd interview, 3 distinct types a Guiding questions were used to facilitate participant storytelling and narrative analysis of participant transcripts was used to explore the meanings participants ascribed to their experiences. Each guiding question was followed by up to 8 further probes in the interview guide. of narrative had become recognizable, and after the 33rd interview, the analytic team agreed that the narrative analysis had reached conceptual saturation, in that no further themes were identified and a justifiable narrative typology had been formulated. Following the definition of our narrative typology, a post hoc statistical analysis was performed in which participant demographic and clinical characteristics were compared according to narrative type. All participants verbally responded to a series of clinical and demographic questions asked by the interviewer during their interviews and the most recent HbA 1c values prior to the interview were recorded from the participant's medical record.

JAMA Network Open | Diabetes and Endocrinology
Participants self-reported the ethnicity with which they identified.

Statistical Analysis
Quantitative clinical and demographic data are shown as mean (standard deviation). Continuous variables were compared by Welch analysis of variance followed by Dunnett T3 multiple-comparison test using GraphPad Prism 8 for Mac OS X (GraphPad Software) and categorical variables were compared using Fisher exact test followed by Bonferroni corrected pairwise comparison. A 2-sided P value less than .05 was considered statistically significant.

Study Participants
We interviewed 33 participants (

Narrative Typologies: Ingrained, Intrusive, and Inconspicuous Lenses
Participants' accounts addressed common topics pertinent to emerging adult life circumstances including experiences with college and work, romantic relationships, drug and alcohol use, travel, and interactions with family members, friends, and health care professionals. Participants also spoke in depth about the practical tasks involved in daily diabetes self-management, their attitudes and the attitudes of others toward indicators of glycemic control (eg, HbA 1c ), and how living with diabetes intersected with their personal and social relationships. Over the course of analysis, we found that there were commonalities in the ways in which narratives were told that enabled us to define 3 broad narrative types, or lenses. Major themes and illustrative quotes for each lens type are summarized in Table 2.

An Ingrained Lens: "Diabetes Is Comfortably a Part of My Norm"
The first lens is what we came to define as an ingrained lens, conveyed by 14 of 33 participants (43%).
From an ingrained perspective, participants presented themselves as trustworthy self-managers who said that diabetes and its self-management tasks caused little disturbance to their daily lives. In ingrained narratives, participants adopted a practical, flexible approach to self-management that helped them maintain a sense of control within their daily lives. Participants recalled childhood and adolescence as times during which they had learned the foundations of self-management, and as emerging adults they felt well equipped to manage diabetes on their own.

JAMA Network Open | Diabetes and Endocrinology
Recognizing Similarly, within ingrained narratives, participants portrayed health care professionals as supportive figures. As they entered the adult care setting, these participants described taking on a new, more engaged role in health care appointments in which they felt that dialogue was more mature and built around their own individual needs.
Participant 9: "…it definitely feels like a team because everyone is involved and you talk to them as though they're, you're on equal terms in the sense that you are all just trying to make you better."

JAMA Network Open | Diabetes and Endocrinology
Recognizing Overall, from the ingrained perspective, the stories of emerging adults with type 1 diabetes characterized participants as becoming diligent self-managers who learned how to work with parents and health care professionals to integrate diabetes seamlessly into their lives.

An Intrusive Lens: "My Diabetes Is Like a Weight, Constantly Weighing Down on Me"
The second narrative type we delineated was intrusive, which was the dominant narrative in the interviews of 12 of 33 participants (36%). These participants' narratives articulated a tone of distress that contrasted starkly with the more positive accounts portrayed from an ingrained perspective.
Participants spoke as though diabetes was a heavy burden and they voiced struggles integrating selfmanagement tasks into their daily lives. In general, rather than gradually gaining confidence in their self-management competency, these participants tended to look back on their childhood as a time of better diabetes care and they lamented that they were no longer able to attain the level of control that they once had. Minimizing the moral context of diabetes Expressed awareness of, but comfort with, suboptimal diabetes control or self-management (ie, they described being content managing diabetes "my way," rather than overburden themselves with trying to manage the so-called "right way") "…I don't really know because my A 1c for the past couple of times that I've been there has been really good. But, yeah, again, I find myself really in control, at times, like obviously there have been times where I've had my own seizures and stuff, but again, they're learning experiences."

An Inconspicuous Lens: "I Try Not to Notice Diabetes or Make It Noticeable"
We called the third narrative type an inconspicuous lens because of the distinct minimizing language commonly used in the narratives of 7 of 33 participants (21%). At first, these narrative accounts had a superficial resemblance to those of the intrusive type, in that participants recognized that they were not managing their diabetes according to expectations, for instance, describing occasions of insulin omission or infrequent self-monitoring of blood glucose. However, unlike the intrusive lens narratives, these accounts were typified by the overarching theme of minimization rather than intrusion. shouldn't say that, I don't like to prescribe moral value to like health stuff, but like a high A 1c , you know, I would just say it was lower." In contrast to intrusive narratives, however, these participants portrayed themselves as not especially burdened by their diabetes and they seemed comfortable intentionally diverting attention away from it.

Comparison of the Clinical Characteristics of Participants According to Narrative Lens
Having defined our 3i concepts, we compared the clinical characteristics of the study participants according to the different narrative lens types ( Table 3). There were no overall differences in age, duration of diabetes, or stage of education or work between participants according to narrative lens (Table 3). In contrast, there were more women than men in the intrusive lens group (10 vs 2, respectively), and fewer participants in this group (42%) lived with a parent or guardian.
Furthermore, HbA 1c levels were significantly higher for participants within the intrusive lens group HbA 1c , 8.6% [1.5%]) and did not differ significantly from them (Table 3).

Discussion
An appreciation of the psychosocial context in which health decisions are made by patients is at the cornerstone of the modern movement toward interpersonal medicine. 26 This is especially the case

JAMA Network Open | Diabetes and Endocrinology
Recognizing Patient Perspectives of Type Although there may be overlap between lenses for any individual patient and there may be fluidity between lenses over time, we identified biomedical and sociodemographic differences between participants according to narrative typology. For instance, HbA 1c levels were lowest in participants who adopted ingrained narratives and highest in those with intrusive narratives.
Although HbA 1c levels differed, we cannot determine the causality of the association between HbA 1c level and lens type from this data set. For example, because frequency of self-monitoring of blood glucose is associated with lower HbA 1c levels, 34 participants who are more engaged in their selfmanagement, and thus of an ingrained lens type, may be expected to have lower HbA 1c levels than participants who are less engaged. Conversely, higher HbA 1c levels have been linked to avoidant coping behavior and decreases in diabetes integration. 35 Alternatively, residual pancreatic β-cell function is associated with lower HbA 1c levels 36 and it is also possible that the narratives of individuals who have residual β-cell function may contain more positive language surrounding life with type 1 diabetes and thus be categorized as being of an ingrained lens type. Aside from HbA 1c , we also observed that there was a preponderance of women in the intrusive lens group, which is consistent with a previous report 37 that symptoms of diabetes distress are more common in female than male adolescents and emerging adults with type 1 diabetes. This finding is also aligned with existing research that has found gender differences in the embodiment of diabetes, 38 along with other psychological constructs, such as coping 39 or feelings of engulfment. 40 However, whether actual diabetes distress differs between genders or whether the manner in which diabetes distress is articulated during interviews differs between genders cannot be resolved from the present data set.

Limitations
Our study has several limitations. For instance, interviews were performed at a single point in time and thus reflect the perspectives of participants at the time of their interview. Participants may have presented themselves differently at another time or if interviewed by another interviewer. The findings are also dependent on context in that a different sample of emerging adults with type 1 diabetes (eg, those attending a general adult diabetes clinic or patients in a rural setting) may have identified different issues. Accordingly, further research is needed to ascertain to what extent the 3i framework described applies more generally to other patient populations or to other emerging adults living with type 1 diabetes. Also, although we observed differences in some clinical parameters between participants, according to lens typology (eg, HbA 1c levels), it should be borne in mind that the study design was qualitative in nature, and caution should be exercised in assigning biological meaning to any of these differences. The purpose of presenting our framework is not to seek strategies to move patients between lens types, but rather to provide a heuristic for health care practitioners to understand the different perspectives emerging adults may have on living with type 1 diabetes so that care can be individualized accordingly.