Evaluating the Experience of Children With Type 1 Diabetes and Their Parents Taking Part in an Artificial Pancreas Clinical Trial Over Multiple Days in a Diabetes Camp Setting

As type 1 diabetes management moves toward the development and integration of new technologies, it becomes more critical to consider the psychological factors likely to play a significant role in the use of these devices (1). Understanding psychological implications and managing expectations are particularly important in the case of artificial pancreas (AP) technology. As noted by Barnard et al. (2), from a psychosocial perspective, the AP is unique in taking responsibility for glucose concentrations, and it is this transfer of trust, along with the burden of multiple devices, novelty, and the day-to-day user requirement that makes acceptance of AP technology a challenge. For these reasons, the impact of AP technology requires a specific evaluation of related psychological implications: the opportunity to assess human factors and effects on the lived experience is the only way to reduce the risk of producing technologies that may be fit for “glycemic” purposes but unfit for “lived experience” (2,3).

Within the existing literature examining the psychosocial aspects of AP technologies, most studies have been performed on adults. In these studies, participants who directly (4,5) and indirectly (by a “paper case description”) (6,7) experienced AP systems, described the positive impact of the tested AP technology, positive attitudes, and a high likelihood of future acceptance. Similarly, the evaluation of the parental feelings of children and young people with type 1 diabetes about diabetes management through overnight closed loop insulin delivery, revealed trust and a positive attitude toward the new technology (8). Barnard et al. (9) evaluated the experiences of adolescents with type 1 diabetes (and their parents) who took part in an overnight closed-loop study in a home setting, and found that the psychological and physical benefits of the closed-loop system outweighed the practical challenges reported. The two existing studies (10,11) assessing the impact of an AP system on children under real-life conditions described reductions in worries about hypoglycemia and regimen burden, a high level of satisfaction, and an increased perceived ease of use of the new technology.

In light of the acknowledged importance of the psychosocial side of AP systems and the critical role that human factors play in the adoption and efficient use of this system, the aim of this study was to evaluate the experiences of children with type 1 diabetes (and their parents) who took part in an AP clinical trial over multiple days in a diabetes camp setting. It was assumed that the adoption and maintenance of a new health care regimen is highly dependent on the subjective perceptions of potential users, here including children and their parents. Since there are currently no existing validated and reliable measures that specifically assess the psychosocial aspects of AP systems, variables of the Technology Acceptance Model (TAM) (12,13) and diabetes treatment satisfaction were considered in the current study. The TAM, providing a general explanation of the intention to use a specific system, has been adopted as a theoretical framework that can explain and identify psychological variables thought to be key factors for understanding and identifying the human factor side of AP technology (6,12). Treatment satisfaction is regarded as an important factor in the management of type 1 diabetes and is assumed to play a crucial role in predicting adherence to a new diabetes regimen, since its absence is considered to be a barrier that is likely to interfere with the optimal use of and therapeutic benefits from this technology.

The current study included a two-feature method approach: a semistructured interview to evaluate participants’ experience and perceptions of AP technology, and psychosocial questionnaires to assess AP treatment satisfaction and acceptance.

The participants included 33 children and 37 parents (some children were accompanied by both parents) attending a diabetes camp for an AP clinical trial. They were recruited from five Italian pediatric centers in Verona, Milano, Torino, Napoli, and Roma. Inclusion criteria were as follows: age 5–9 years; a diagnosis of type 1 diabetes for at least 12 months; use of an insulin pump and sensor for ≥3 months; HbA_1c level <10%; a family proficient in continuous subcutaneous insulin infusion/sensor-augmented pump; and the willingness to send at least one parent to the camp. Exclusion criteria were diabetic ketoacidosis or severe hypoglycemia within the last month; concomitant disease; and any medication use or conditions that could influence metabolic control, compromise safety, or prevent study completion. This was an open-label, randomized, crossover trial comparing 3 days with an AP to 3 days of parent-managed sensor-augmented pump use during a 7-day camp. There was a 1-day washout period between the two treatments. Diet and physical activities (static in the morning and moderate-to-high intensity in the afternoon) were kept as constant as possible during the two study periods. The camp also hosted five pediatric diabetologists, three diabetologists who were responsible for the study, and four bioengineers. Given the young age of the children, each family was housed together. During the 3 days of AP use, the system was used continuously, except for the time needed for personal hygiene. As a result, the AP remained fully operational 97% of the time. When the AP system was active, the study team constantly monitored the participants through a telemedicine system. More details about the methodology and results of the trial have been described previously (14).

To explore participants’ experience and perceptions of AP technology, a semistructured interview was designed based on the TAM (12,13). In particular, by making reference to the questions used by van Bon et al. (6), for which there is no validated translation for the Italian population, a survey with 14 items adapted for both adults and children was constructed. An overview of the methods used to develop the semistructured interview and its contents is given in Supplementary Appendix 1. After 3 days of AP treatment, a psychologist conducted the interviews individually, face to face; the interviews were audio recorded and then transcribed literally. When both parents were present, only one parent was interviewed.

At the end of the AP treatment periods, the participants’ parents were asked to individually complete the Diabetes Treatment Satisfaction Questionnaire for parents (DTSQ-parent) (15) and the AP Acceptance Questionnaire (7).

The DTSQ-parent version (15) is based on the widely used eight-item DTSQ for adults (16,17) that was developed and expanded using extended interview work with parents of individuals with type 1 diabetes. The DTSQ-parent is a 14-item measure that enables self-reporting by parents of the satisfaction about the current treatment of their children. Treatment satisfaction is the combined score of the 10 items (items 1, 5–9, and 11–14) evaluating satisfaction, ease, flexibility, school day, medical support, and continued treatment. The remaining questions relate to perceived frequencies of hyperglycemia or hypoglycemia (items 2–4) and the effects of the current treatment on the parents’ life (item 10). All items are rated from 0 (very unsatisfied) to 6 (very satisfied), and the range of the total score is 0 to 84, with higher scores indicating better satisfaction. The author was initially contacted for permission to use the DTSQ-parent and for a copy of the instrument itself, and then the Italian version was constructed. It should be specified that a trial-specific version of DTQS-parent was used, with specific amendments suggested and authorized by the author herself to make the questionnaire appropriate for the aims of the current study. For example, the clarification wording “past 3 days” has been added to replace “current treatment” in some items of the original DTSQ-parent (Supplementary Appendix 2).

The AP Acceptance Questionnaire, based on the TAM, was developed to examine the future acceptance of an AP (7). It included 15 items measuring intention to use (items 1 and 2), perceived usefulness (items 3–10), perceived ease of use (items 11–13), and trust in AP (items 14 and 15). The questionnaire was translated into Italian from a previous study (18) using a forward-back translation procedure. For the current study, the questionnaire, which usually addressed patients, was modified to also fit the parents. Answers were given on a 7-point Likert scale, where 0 = strongly disagree, 1 = moderately disagree, 2 = somewhat disagree, 3 = neutral, 4 = somewhat agree, 5 = moderately agree, and 6 = strongly agree. A higher total score indicates higher AP acceptance (range 0–90). Further information on items of the AP Acceptance Questionnaire can be found in Supplementary Appendix 3.

The statistical analysis was performed with Statistical Package for the Social Sciences (SPSS) version 21.0 for Macintosh. Frequencies or descriptive statistics were run for demographic information and data from questionnaires.

To quantitatively analyze the interviews, computerized textual analysis was applied to the textual corpus obtained from the verbatim transcript (both parents’ and children’s answers) of the semistructured interviews. The participants’ answers were copied and corrected for grammatical errors and pasted into a text file. The names and surnames of the participants were deleted. This remaining text constitutes the text corpus. Textual analysis of the participants’ answers was conducted using a content analysis software package called T-Lab 5.1 (19). Two of the functions of T-Lab were used in this research: co-occurrence analysis (computation of word association and sequence analysis of key words); and thematic analyses of elementary contexts (ECs). To explore participants’ subjective perceptions of AP, a computation of word association was first used to analyze the relationship between the word “AP” and each other word in the corpus, identifying which concepts were most frequently associated with “AP” in the entire text, both in parents’ and in children’s interviews. An association chart of the word “AP” was constructed with T-Lab to give an impression of the contextual use of that word in terms of which other words the term under consideration often co-occurred. The relative closeness of one word to another word in the chart was assessed according to a parameter called the cosine coefficient. The sequence analysis identified the key words that are closer to the word “AP,” selecting the words with the highest probability of preceding and succeeding the word “AP” in both parents’ and children’s interviews.

Next, a thematic analysis of ECs was conducted to identify the content of the main corpus through a small number of significant thematic clusters emerging from the text. The clusters divided the text into a number of themes that represent the content of the narrative. Each cluster was labeled based on the qualitative interpretation performed by analyzing the ECs grouped in each theme and the words connected to each cluster. Thematic clusters were internally homogeneous and externally heterogeneous. Because of the shortage of text provided by the children’s interviews, this analysis was performed only on parents’ answers. For further information, see Lancia (19–21) and http://www.tlab.it.

Of 33 enrolled children, one discontinued participation because of a febrile illness, one because of the illness of the parent attending the camp, and one because of poor acceptance of camp lifestyle. In total, 30 children (19 males) and their parents (N = 34) completed the study (mean age 7.6 ± 1.2 years, mean body weight 26.0 ± 6.1 kg, mean height 123 ± 8 cm, mean BMI 16.9 ± 2.1 kg/m², mean BMI z-score −0.09 ± 0.91, mean HbA_1c 7.3 ± 0.9% [57 ± 10 mmol/mol], mean duration of type 1 diabetes 4.7 ± 1.6 years, mean pump use for 3.3 ± 1.9 years, and mean total daily insulin use of 20.3 ± 6.2 units/day (0.78 ± 0.16 units/kg/day)] (14).

Altogether, 27 parents (22 mothers) and 27 children were interviewed. Seven parents were not interviewed because they were not available during the interview period (e.g., they were resting). Only three children refused to be interviewed because they were unwilling to leave their current activity.

The concepts most frequently associated with the word “AP” in the entire text, both in parents’ interviews (Fig. 1) and in children’s interviews (Fig. 2), were identified. The higher the cosine coefficient of a word with respect to the word “AP,” the closer that word is positioned to the center of the figure.

Table 1 displays a sorted list of predecessors and successors of the word “AP” in parents’ and children’s interviews. The list is shown in descending order according to the probability values (Table 1, Prob.). For example, Table 1 shows that the probability that the parents’ answers included the word “son” after the word “AP” was 0.033.

The output of the thematic analyses of the ECs resulted in a structure of the following four thematic clusters: check, using, glycemic control, and change. Table 2 shows the four thematic clusters, listing the χ² test values of the most important lexical units in each cluster. A high χ² value means that the corresponding lexical unit was central to a cluster.

The four clusters contain words underlying the parents’ main attitude toward the use of AP concerning the properties of the device. Especially in the “change” cluster, some positive aspects and advantages of the new technology were emphasized (e.g., “curiosity,” “new,” “intriguing,” and “positive”). The remaining clusters showed other positive views concerning general type 1 diabetes management, such as medical examination, frequency of physician office visits, and diabetes check data (cluster “check”: “for me/ok,” “reducing,” “improving,” and “better”); AP management and related issues and figures involved in this process (cluster “using”: “learning,” “engineer,” and “easy”); and all contents related to the blood glucose monitoring, such as nocturnal glucose control and glucose level variation control (e.g., resulting from food intake) (cluster “glycemic control”: “peace of mind,” “to be glad,” and “calm”).

Focusing on words referring to key troublesome aspects of the AP, the main concerns included the psychological burden related to the constant effort to adequately manage diabetes (cluster “check”: “thinking,” “being alarmed,” and “continuously”); all emotional correlates associated with glucose control, such as fear of hypoglycemia (especially during sleep), the desire to maintain autonomous and controlled regulation not affecting children’s autonomy, and the desire to minimize stress related to the risk of poor glycemic control (cluster “glycemic control”: “night,” “hypoglycemia,” “autonomy,” and “intervening”); and the risks and difficulties of getting used to the new system (cluster “using”: “being alarmed” and “help”; cluster “change”: “disadvantages,” “difficult,” and “tired”).

Based on the overall number of ECs, the largest cluster was “glycemic control” (31.6%), followed by “using” (27.7%), “change” (26.5%), and “check” (14.2%).

The DTSQ-parent was completed by 29 parents (11 fathers and 18 mothers). Five parents did not complete the questionnaire because they were not available during the administration period. The mean scores of the Treatment Satisfaction (44.24 ± 5.99, range 32–53) and Perceived Diabetes Control (7.8 ± 2.2, range 3–12) subscales indicated a general satisfaction and trustful views of type 1 diabetes control provided by the new system among parents. The mean scores of the Perceived Frequency of Hypoglycemia (2.59 ± 1.7, range 0–5) and Hyperglycemia (2.83 ± 1.56, range 0–6) subscales suggested a low level of perceived risk of poor glycemic control. The mean score of the item concerning the effects of the treatment on the parent’s life indicated moderate perceived positive consequences of using the new system (4.2 ± 1.18, range 2–6).

The AP Acceptance Questionnaire was completed by 34 parents. Most parents (94.1%) noted the intention for their child to use the AP in the long term (85.3% strongly, 8.8% moderately). Furthermore, 67.6% felt strongly and 23.7% moderately that the AP would improve their child’s glucose control; 70.6% believed strongly and 26.5% moderately that it may reduce the frequency of hypoglycemia in their child; and 52.9% believed strongly and 26.5% moderately that it may prevent hyperglycemia. Of the parents, 70.5% considered AP easy to use (52.9% strongly and 17.6% moderately) and 82.4% (47.1% strongly and 35.3% moderately) thought that their children could handle it. Parents had trust in the system, with 64.7% noting that the AP will administer the right amount of insulin and 76.5% believing that the blood glucose level measured by sensor was correct.

This study investigated the psychological aspects of AP technology during day-and-night use. Assessing participants attending diabetes summer camp, where they were able to experience aspects of how an AP would work in their daily life setting, rather than tightly controlled experimental conditions, allowed for direct and more realistic analysis of their subjective experience. In addition, since there are currently no existing validated and reliable measures that specifically address the needs of AP as a novel technology, a computerized text analysis was performed for the first time on participants’ interview answers to better understand and identify the psychological aspects of the new technology. Automatic text analysis has already proven to be a powerful tool for helping chronic disease researchers understand the typology and nature of the responses, and the needs of those who are directly affected by a severe illness (22,23).

In line with evidence from studies evaluating the impact of the AP tested technology (4–7,10,11), the current results suggest a substantial positive experience and approach of participants toward the AP. In particular, the analysis of participant interviews showed that the parents are mainly focused on understanding the mechanisms of the new device, the risks and benefits of using it in daily diabetes care, and the associated need to obtain assurances regarding the safety of using the AP in managing hypoglycemia. This is clear from an analysis of the most relevant words surrounding the word “AP” (“insulin,” “hypoglycemia,” and “meal” followed by “immediately,” “basal,” “working,” “sensor,” “child,” “night,” “seeing,” “managing to,” and “school”), and by the words identified as its main predecessors (“hypoglycemia,” “using,” “safe,” and “management”) and successors (“managing to,” “seeing,” and “intervening”). The children’s perspective was characterized by curiosity about the new system as revealed by the most relevant words surrounding the word “AP” in children’s answers (“seeing/monitoring,” “asking,” “telephone,” “beautiful,” “experience,” “glycemia,” “touching,” “parent,” “teacher,” “new,” and “increasing”), and by its predecessors (“seeing/monitoring,” “trying,” “new,” “easy,” and “beautiful”) and successors (“beautiful,” “telephone,” “new,” and “easy”).

Evaluation of the most frequent topics in parents’ interviews confirms the crucial role that the efficacy of glycemic control provided by the new system played in the parents’ perceptions. The analysis of the lexical units connected to the cluster “glycemic control” shows, in fact, that, along with the hope of better quality of life (“autonomy,” “to be glad,” “sleeping,” “advantage,” and “calm”), there were some concerns in adopting the new system, already described in the existing literature (4,6,9), related to the quality of glucose control that the new system would provide, especially during nocturnal sleep (“night,” “hypoglycemia,” “peace of mind,” “intervening,” and “constant”). These doubts and uncertainties were most likely caused by several factors, such as the fear of changing the usual (and well-known) diabetes regimen, the lack of experience with the new technology, and the challenges in using a new therapy with related loss or passage of treatment responsibility from the parent to a technological device. The weights of the “using” and “change” clusters provide further information regarding the relevant components shaping parents’ perception, such as the commitment and effort requested of the child and family in learning and growing accustomed to the functioning of a new system (“new,” “difficult,” “to get used,” “tired,” and “changing” in the “change” cluster; “help,” “learning,” “easy,” and “time” in the “using” cluster), with related issues (i.e., inconvenience of the size of the device: “bringing” in the “change” cluster; risks: “to be alarmed,” “relying on” in the “using” cluster; and “disadvantages” in the “change” cluster), the potential need to involve specific figures to the aim of a proper use of AP (“help,” “engineer,” “relying on,” and “teacher” in the “using” cluster), and curiosity toward the new technology (“curiosity,” “new,” “hoping,” and “enthusiastic” in the “change” cluster). The “check” cluster provides further information about some of the possible effects of AP on general disease management, especially in terms of changes of commitment and tasks (e.g., how and when to undergo medical examination, frequency of physician office visits: “medical examination,” “month,” “doctor,” “check up,” “times,” and “treatment”) and the related psychological and emotional burden resulting from the novelties in diabetes regimen (e.g., the need to think and check, the risk of being alarmed: “thinking,” “being alarmed,” “continuously” “checking,” and “managing to”).

Data from the questionnaires indicated a general parental satisfaction and trustful views of diabetes control provided by the new system. Indeed, most parents believed it was easy to use and capable of providing good glucose control. Most parents noted their intention to use the AP and expressed a low level of the perceived frequency of hypoglycemia and hyperglycemia.

One limitation of this study was the small number of participants and the short duration of the AP period: 3 days of AP use may not provide a full evaluation of the psychological aspects and acceptance of a new device created for long-term use. Another limitation was that two of the used measures, the AP Acceptance Questionnaire and the 14-item semistructured interview survey, have only preliminary empirical validation supporting their use in younger patients, parents, and the Italian language. Furthermore, the subjects involved in the current study tended to be highly motivated, subsequently making them less representative of the general type 1 diabetes population to provide their perspective on an AP system. These limitations must be addressed in future AP trials.

In conclusion, these findings identify the perceptions and emotions of participants about using the technology, and their ability to trust the new system. These aspects have been shown to be critical to the use and adherence of diabetes technology (1,2,9), and contribute to a greater understanding of the psychological impact of AP technology. Although the parents’ concerns do not seem to affect the generally positive approach, they must be considered to better understand how to increase adherence to new diabetes treatments and to prevent potential psychological barriers from reducing the likelihood of AP technology adoption.

Further studies are required to extensively explore participants’ perceptions early in the AP development to individualize the new treatment as much as possible and tailor it to respond to their needs and values.

Acknowledgments. The authors thank the children and their families for participating in the study.

Funding. This study was partially supported by private donors and associations of patients with diabetes. The Italian research foundation “Società Italiana Diabetologia” provided administrative support. Y.L. is funded by the Agency for Management of University and Research Grants of the Government of Catalonia, Spain (Beatriu de Pinós grant BP-DGR 2013).

Duality of Interest. No potential conflicts of interest relevant to this article were reported.

Author Contributions. All authors reviewed and provided feedback on drafts of the manuscript. A.T. was chief psychologist of the Naples site; and contributed to the protocol design, data collection, data analysis, and drafting of the manuscript. R.B., I.R., A.S., R.S., A.G., M. Marigliano, N.R., A.R., D.T., V.V., A.Z., F.B., S.G., and D.C. were camp physicians and contributed to data collection. D.I. was chief pediatrician of the Naples site and camp physician, and contributed to protocol design and data collection. S.D.F. was senior camp engineer and contributed to data analysis. G.L., M. Messori, F.D.P., and L.M. were engineers responsible for the design and implementation of the control algorithm and telemedicine. R.V., R.C., Y.L., and E.L. were camp engineers. S.Q. was chief engineer of the Telemedicine Unit in Pavia. C.C. was chief of the AP project and camp engineer, and contributed to the protocol design. D.B. was study coordinator and camp chief physician; and contributed to the protocol design, data analysis, and drafting of the manuscript. A.T. and D.I. are the guarantors of this work and, as such, had full access to all the data in the study and take responsibility for the integrity of the data and the accuracy of the data analysis.

Prior Presentation. Parts of this study were presented in abstract form at the 9th International Conference on Advanced Technologies & Treatments for Diabetes, Milan, Italy, 3–6 February 2016.