Perceptions of Psychosocial Factors and the Insulin Pump

We conducted five focus groups (16) to explore the experiences of adult insulin pump users. We used a purposive sample of 30 adults with type 1 diabetes currently receiving treatment at the Joslin Diabetes Center. Subjects were recruited through direct mailings to 120 patients using a pump on the basis of their laboratory A1C levels, to which ∼60 responded. Of these, 6 participated in the pilot study, and 24 could not attend at a time when the meeting of their A1C group was held or did not meet the inclusion/exclusion criteria. Inclusion criteria included type 1 diabetes, age of 18–70 years, and at least 1 year of pump use. Exclusion criteria included pregnancy, severe depression, blindness or any other condition that made survey completion difficult, and no current enrollment in any pump education programs. Groups were formed on the basis of participants’ availability and loosely on level of blood glucose control: low A1C (<7.0%), mid A1C (7.1–8.0%), and high A1C (>8.0%). Subjects received free parking and $25.00 compensation for their time. The Committee on Human Studies at the Joslin Diabetes Center, Boston, Massachusetts, approved the protocol, and all subjects provided informed written consent.

Before subject recruitment, a multidisciplinary team (endocrinologist, nurse practitioner, clinical psychologist, and research psychologist) developed focus group questions and follow-up probes derived from their clinical experiences (Table 1). Questions were broad and open-ended and addressed the experience of living with the insulin pump. Probes were used to elicit more specific information once unstructured discussion on the topic ceased. The questions and format were pilot tested and then refined before implementation.

A series of five 90-min focus groups was held over a 6-month period. A clinical psychologist (M.D.R.), experienced in insulin pump therapy and conducting focus groups, facilitated all groups. An observer (A.S.) took notes and audiotaped the discussion. During group discussion, the facilitator put forth a question, and participants responded while interacting with one another. Probes were used to gain more detailed information and personal accounts. At times, participants would give advice to one another, and the group facilitator refocused discussion on sharing of experiences. After each group, the research team met and debriefed on interactions and discussions in the groups. Laboratory data and health information were obtained from medical records and demographic questionnaires.

Audiotapes were transcribed and entered into NVivo2 qualitative software (QSR International, Victoria, Australia) for analysis. Transcripts were coded independently by five team members (M.D.R., A.S., J.L., H.W., and K.W.) for identification of initial themes. Content analysis (17,18) was conducted by marking and categorizing key words and phrases to identify the psychosocial issues of relevance to the group participants. Research team members met regularly over a 1-year period to establish group consensus regarding identification and definition of themes and selection of examples from transcripts. We achieved investigator triangulation, a process in which more than one investigator analyzes the data (19,20) through the use of a multidisciplinary team with members experienced in the care of adults using insulin pump therapy. Further triangulation was achieved through review of the findings by experienced clinicians outside of the research team.

Thirty adults (59% female, 97% white, and 76% married; aged 47.0 ± 9.5 years) with type 1 diabetes participated in five focus groups: two low A1C (6.8 ± 0.4%), one mid A1C (7.8 ± 0.3%), and two high A1C (9.1 ± 0.5%). Participants, on average, were highly educated (15.4 ± 1.5 years education), had a long duration of diabetes (27.3 ± 13.1 years), and had worn the pump for 3.6 ± 2.2 years. Focus groups did not differ on demographic characteristics.

Three major themes were identified: impact on diabetes self-care, emotional reactions to insulin pump, and body image and social acceptance. The first two themes arose in all focus groups, but the groups differed in their discussions of these themes by level of glycemic control. Clear differences were found between low and high A1C groups, whereas the discussion of mid A1C group sometimes reflected the low A1C and sometimes the high A1C group discussions. Table 2 identifies areas of overlap of the mid A1C group. In the following, we focus on the differences between the low and high groups. For the social acceptance theme, sex rather than glucose control differences existed.

Responses to questions about wearing an insulin pump and managing one’s diabetes focused mainly on self-care activities, conceptions or misconceptions of the pump in the treatment of diabetes, and whether group members described themselves as actively engaged or more passive in their diabetes self-care behaviors. All groups discussed the insulin pump as making self-care more “convenient,” expressed in terms of having greater “flexibility” and “freedom.”

High A1C groups focused mainly on the “convenience” of the pump. Comments were more passive about their experience with the pump, e.g., describing little interest in learning more about their carbohydrate-to-insulin ratio. They seemed to have a “magical” notion that the pump would make diabetes disappear. For example, one participant described how she thought the pump would be a “miracle” and “… you can do anything you want now.” High A1C group participants also expressed a reluctance for performing expected self-care behaviors. They reported not valuing food records or blood glucose diaries, and they admitted to sometimes forgetting to use bolus insulin. They agreed that using the pump was more difficult than they expected. For example, when one participant stated that the pump is “still a lot of hard work,” another responded, “people don’t realize that and it’s frustrating ….” Further, they seemed to overrate their level of knowledge. As one participant noted, “You know yourself better than what’s written on paper … figure I have a PhD and an MD by now.”

High A1C groups also described fearing technology at the start of pump use and reported feeling less comfortable using it, particularly the advanced features. Furthermore, these participants expressed difficulty grasping how they could improve glucose control with the insulin pump. For example, one participant stated,

Low A1C group members discussed the pump as a tool that required their engagement in checking glucose levels, counting carbohydrates, and using bolus insulin. They described themselves as actively involved in their diabetes self-care. Group members exchanged ways to examine blood glucose readings for trends and discussed whether, when, and how they checked their ketones. They were aware and stated that beyond “convenience” the true benefit of the pump was improved glycemic control: “… why it’s [the pump] so good is just the clinical benefits. You know, less hypoglycemia and … more stable blood sugars.” They also appeared to be keenly aware that the pump was not a “magic pill.” “Yeah, it’s a necessary evil… . It’s the best hassle out of all the ones I could have. It extended my life the best. So as far as it goes, I honor it.”

The discussion among low A1C participants suggested that they were self-reflecting and had insight about the limitations of the freedom and flexibility of the pump. Their discussion also suggested that they noticed their self-care behaviors and attempted to understand them. They reported that they could not do whatever they wanted and were aware that unrestrained eating behavior was not ideal for optimal weight or glycemic control. In addition, they reported losing weight as a result of not having to treat as many hypoglycemic events, e.g., “I lost a lot of weight. I was really surprised… . I think before I was eating all kinds of snacks to keep from getting low or just because I was afraid of getting low, and now … I don’t eat so much… .” The following discussion about eating and insulin illustrates the low A1C group’s self-reflection about their behaviors:

On the other hand, high A1C participants had difficulty realizing the need for limiting their behaviors. Some discussed how they had gained weight as a result of their increased freedom. One woman stated: “I’ve put on weight, but also I have the flexibility of being … normal, and I’m diabetic … so I’ll do what I want to do, eat what I want to eat.” Another responded to this statement by saying, “It’s not a pump issue, it’s a brain issue.” Furthermore, some participants discussed how they lost weight as a result of skipping meals or not using boluses of extra insulin to cover carbohydrates. Thus, the high A1C groups acknowledged that self-monitoring and restraint were needed but were not able to utilize them to their advantage. For example, one participant responded to several group members’ admissions of forgetting to administer a bolus of extra insulin by saying, “I think we all know what to do. It’s just a matter of doing it.”

All groups described emotional responses to the insulin pump in relation to their diabetes. Participants in the low A1C groups described increased feelings of normalcy and greater ease talking about their diabetes since using the pump. However, high A1C group members described their pump experiences in more negative terms. For example, only one female participant in the high A1C groups mentioned feeling more normal with pump use, but she discussed “how feeling normal” contributed to excessive eating. Both low A1C groups reported emotional reactions that included the perception of initiating pump therapy as reminiscent of how they felt when initially diagnosed with diabetes. This response did not arise in the high A1C groups.

Low A1C participants described their experiences dealing with hypoglycemia before and after the insulin pump. Before the pump, the necessity of having to pay close attention to the timing of insulin injections and to eat when not hungry to avoid hypoglycemia was viewed by the group as making them feel less “normal.” They discussed the pump as contributing to feelings of being “regular, normal, and real” because of the freedom to eat at different times of the day or to skip or delay a meal without fear of hypoglycemia. For example, one woman reported that “being able to eat whenever you want to eat instead of having to eat at this hour” was a normalizing benefit of pump use. Another woman described how she “started to feel hungry again.”

Participants in the low A1C groups stated that the pump helped them feel more accepting of their diabetes, less ashamed of it, and able to speak about it to others for the first time. One person said that the pump helped “lift the stigma” of diabetes, whereas another described the pump as making the “psychological burden of diabetes lighter somehow.” A female mathematics teacher described how she incorporates her insulin-to-carbohydrate ratio when teaching mathematics concepts: “… so I teach them and hopefully I’m changing the world.” Another participant described the pump as a vehicle for talking about having diabetes:

In contrast, high A1C group participants spoke of being “tired” of the pump. They described feeling discouraged and “frustrated” that the pump did not “fix everything” and that “it’s still a lot of work.” They described feeling burdened by having to deal with the pump on a daily basis. As one woman stated: “I just get tired of it … wearing it all the time gets old. Sometimes I’d like to take it off and just forget about it.”

Interestingly, members of the low A1C groups described their initial experiences with the pump as reminiscent of feelings they experienced when first diagnosed with diabetes. For example, one woman stated that it recalled the trauma of her initial diagnosis: “… when I decided to go on the pump… . I felt … almost like a trauma that I felt when I was first diagnosed.” Starting the pump apparently recalled the earlier experience of learning about diabetes self-care:

All groups discussed the ways that wearing a noticeable mechanical device could affect one’s feelings about his or her body image and sense of social acceptance. Men and women differed in their discussion of this issue.

Female group members verbalized more about the ways the pump affected their body images than did male group members. They expressed concern about how to integrate this mechanical device with their body image, and they varied in their views of the pump regarding this concern. Their conflicts were voiced in terms of whether they saw the insulin pump as “inside or outside” of their bodies. They all were aware that the pump was attached to them, but some described it almost as a part of themselves, whereas others had more trouble integrating the pump with their body integrity. For example, one woman stated: “I don’t even think of it as separate. I don’t want to say it’s part of me, but it is.” Another woman said: “I’m so against something on the outside of me… You just feel different.”

Some female participants discussed feeling an increased sense of self-consciousness from wearing the pump in public. They expressed how the pump could make them feel different in relation to their bodies, wearing clothing, and social comfort in public settings. They seemed more concerned than the men about being able to wear the pump comfortably, stylishly, and unobtrusively, particularly when wearing dresses and bathing suits. One woman described the pump as a “fashion challenge”:

On the other hand, most of the men appeared less bothered by social acceptance in relation to the pump. They stated that the changing times and greater use of technological devices made them less self-conscious and more able to wear the pump without fearing social ostracism or discomfort. They felt more at ease with the pump as just another device, “no different than a pager,” which could be worn in public. However, not all men felt this way, and one man also expressed his feeling of being different in public:

In this study, focus groups provided a qualitative method for obtaining everyday experiences of living with an insulin pump for individuals with type 1 diabetes. From these discussions, three themes were identified: impact on diabetes self-care, emotional reactions to the insulin pump, and body image and social acceptance. Differences were observed among glycemic control groups regarding pump expectations, approaches to pump use, and emotional reactions to the pump. Sex differences were found regarding body image and social acceptance in insulin pump use.

Participants who expected the pump to serve as a tool rather than a panacea had a more active approach to their diabetes care and better glycemic control. Their active participation included an understanding that the pump was a technological device making life with diabetes more flexible and convenient but not a vehicle for total freedom from diabetes. In contrast, the high A1C group expressed unrealistic expectations and a more passive model for self-care. They expected the pump to be a “miracle” that would not require close attention to self-care behaviors and that allowed them to do whatever they wanted, particularly with regard to eating. This finding on expectations of the pump is consistent with that reported in adolescents (21). Before initiating pump therapy, health care providers need to assess patients’ expectations of the pump, dispel magical beliefs and attributions, and review educational materials to ensure that the pump is not portrayed as allowing total dietary freedom. Further, pump companies should monitor marketing materials to prevent 1) the promotion of unrealistic expectations of the pump and 2) the notion of total dietary freedom. Initial and ongoing pump treatment should reinforce the idea of the pump as a tool that one must learn to use and master.

The low A1C groups described greater comfort with using advanced pump features, the ability to observe and learn from their behaviors, and an awareness of the need for self-discipline and self-regulation. Self-efficacy and self-regulation may be useful psychological constructs for understanding the ways one addresses self-care (22,23). However, only two studies have investigated self-efficacy and the insulin pump with adolescents (24,25), and no studies on self-regulation and the insulin pump have been reported. Important differences in self-efficacy and other psychological constructs may exist, and future research needs to quantitatively differentiate active versus passive approaches to diabetes self-care to better understand the observed qualitative differences.

The description of how starting the insulin pump was reminiscent of the feelings experienced at initial diagnosis of diabetes has not previously been reported in the literature. This experience of the recall of diabetes diagnosis may be partially explained by study participants’ long duration of diabetes and many years of insulin injection use. If so, transition to an insulin pump might be more stressful for patients such as these compared with those with shorter diabetes duration. However, no differences in diabetes duration existed among our focus groups, and it is noteworthy that those in the high A1C group did not report this experience. One explanation may be that the high A1C group’s passive approach entailed continuing to think about diabetes in the same fashion as when they were originally diagnosed. Starting the pump was not a relearning experience and did not recall the feelings of having to face diabetes in learning to take care of oneself. Further research needs to address whether recall of diabetes diagnosis during insulin pump initiation interacts with self-care and whether discussion of coping with these memories can contribute to more effective insulin pump use.

Women in our study were more concerned than men about perceived body images and social acceptance with regard to using the insulin pump. Prior research has identified increased self-consciousness (26), greater body dissatisfaction (27), and decreased body esteem (28) for women without diabetes and body image concerns for younger women with diabetes using the insulin pump (29). It is not surprising that women in our study were more self-conscious about how they looked in public when wearing an insulin pump. What was unexpected was the existence of both feelings of normalcy and feelings of difference in these subjects in relation to the pump. Low A1C group participants discussed their increased sense of normalcy with regard to diabetes as a result of pump use in terms of having more freedom about their eating schedule and a decreased fear of hypoglycemia. However, this feeling of normalcy did not extend to their body image and public appearance. The public image of being noticed with an insulin pump caused feelings of being different and less acceptable. This finding differs from those from a previous study that found adolescents were not more self-conscious as a result of wearing the pump (21). However, our results are consistent with that study’s finding that female adolescents reported concerns about fashion inconveniences. Further research is needed to address the impact of self-consciousness and body image on diabetes self-care for people managing the pump and to investigate the effects of age and sex on these factors.

There are several limitations to this study. Focus groups may inhibit expression of differing viewpoints, implying greater consensus of viewpoints than is actual. Intergroup comparisons may be related to differences in group dynamics rather than real differences in attitudes or experiences (30,31). Although our subjects varied in age, the sample was mainly white, educated insulin pump users with type 1 diabetes. Thus, these results can be generalized to similar populations but need confirmation in other ethnic and socioeconomic groups.

Further research is necessary to understand factors associated with active engagement in diabetes self-care. This understanding is particularly needed as more and newer technological devices for the treatment of diabetes emerge. There is a danger that when given new technologies, patients may not develop a sense of their active role and responsibility in their own self-care. This can only lead to negative feelings and disappointments rather than further empowerment of the patient and optimal diabetes management. Findings of this qualitative study suggest that interventions to improve diabetes management with pump use should include evaluation and discussion of active versus passive approaches to self-care, recall of diabetes diagnosis, expectations of the pump, and pump-related self-consciousness and body image concerns. The roles of these factors in optimal diabetes management warrant further investigation.

This work was supported by National Institutes of Health (NIH) Grant R01 DK60115 (to K.W.), in part by NIH Training Grant DK07260 (to A.S.), by Diabetes and Endocrinology Research Core NIH Grant P30 DK36836, and by an unrestricted educational grant from Animas Corporation (to H.W.).