Patient Perception and Satisfaction With Insulin Pump System: Pilot User Experience Survey

Abstract

Background:

The goal of this study was to assess patient perspectives and satisfaction with the MiniMed 670G insulin pump. Those participants who used the pump as part of a hybrid closed loop were also asked to provide their views on the automatic feature (auto mode).

Methods:

Adults with type 1 diabetes mellitus using the Medtronic™ 670G pump were asked about their experience with the device using a semi-structured survey developed by the research team. Responses were quantified to identify emergent themes.

Results:

Seventeen participants used the pump as part of a hybrid closed loop system, while four participants used the pump in combination with a nonintegrated continuous glucose monitoring system. Overall, participants indicated a high level of satisfaction with the pump (14/21) mostly because of improvements in blood glucose (BG) control (15/21). Least liked features were physical design and structure (6/21), frequency of user input (5/21), alert frequency (4/21), and difficulty of use (3/21). Those using the hybrid closed loop were satisfied with the auto mode feature (11/17), mostly because of improvements in BG control (9/17). The least liked features of the auto mode technology were that blood glucose levels remained elevated (5/17) and the frequency of alerts (4/17).

Conclusion:

Participants indicated a high level of satisfaction with the pump and its auto mode featured mostly because of improvements in BG control. They also pointed out some key aspects of the device that are of potential clinical or commercial relevance. Additional research is needed to further evaluate users' perspectives on this new device.

Patients with T1DM rely on lifelong insulin treatment to keep blood glucose (BG) levels in a desired target range. In the past, the only option for insulin delivery was injecting insulin through syringes or insulin pens. Today, due to advances in T1DM technologies, tools such as continuous subcutaneous insulin infusion devices (aka CSII or insulin pumps), and continuous glucose monitoring systems (CGMS) are available to administer insulin and aid patients in making accurate insulin dosing decisions.

The use of these technologies has increased dramatically since 1993, when the Diabetes Control and Complication Trial confirmed that remaining in a target glucose range, via intensive insulin therapy, could delay the progression of diabetes complications.^1,2 Estimates show that about 400 000 patients with T1DM in the U.S. are using CSII.³ The first FDA approved hybrid closed loop system was launched in June 2017. The hybrid closed loop system is composed of the MiniMed 670G insulin pump and the Guardian Sensor 3 CGMS. The system utilizes a control algorithm to automatically increase, decrease, and suspend insulin delivery when set to auto mode function.^4,5 The system is configured to only support two BG targets: 120 mg/dL and 150 mg/dL. When used in auto mode, patients continue to enter mealtime carbohydrates and BGs to calibrate the sensor. The system can also be placed in manual mode to function as an insulin delivery system controlled completely by the patient.

Before the pivotal trial evaluating the safety of the hybrid closed loop,⁶ several groups have demonstrated feasibility, safety and effectiveness of closed-loop technology.^7-11 The MiniMed 670G can be also used in combination with CGMS different from the Guardian Sensor 3. When the pump is not used as a hybrid closed loop and it communicates with the CGMS it called sensor augmented pump therapy. Feasibility studies using the first hybrid closed loop system showed substantial improvements in time at BG target in comparison with alternatives such as conventional or sensor-augmented pump therapy.^11-13 Recent studies have suggested that patients who used sensor-augmented pump therapy with adherence to continuous glucose monitoring had improved hemoglobin A1c (HbA1c) levels without an increased rate of hypoglycemia when compared with the use of multiple daily insulin injections^14-17 The use of CGMS has also been demonstrated to be clinically valuable by reducing HbA1c, risk of hypoglycemia and hyperglycemia, glycemic variability, and improving quality of life when tested with different patient populations.¹⁸

Farrington reviewed literature on psychosocial impact on users of hybrid closed-loop systems.¹⁹ He reported that there is a small but growing body of research on this area. He indicated that reports are generally positive in terms of user experience, including reassurance and reduced anxiety, improved sleep and confidence, and “time off” from diabetes demands. The counterpart of those outcomes has been reports on variable levels, concerns about physical bulk, technical glitches and difficulties incorporating closed-loop systems into everyday life. Most of the 13 psychosocial studies reviewed by Farrington were limited to interventions that required the use of the closed loop systems overnight, in a controlled environment, or during 24 hours but for only a few days. Most importantly, none of those studies were done on users of the Medtronic MiniMed 670G pump.

Objective

The goal of this study was to assess patient perspectives and satisfaction with the MiniMed 670G insulin pump, Those participants who used the pump as part of a hybrid closed loop were also asked to provide their views on the automatic feature (auto mode).

Materials and Methods

An electronic survey was designed by endocrinologists on the team to get insights into patient’s perspectives on the pump and its auto mode and its auto mode feature (see Supplementary Material). The survey was divided into structured and unstructured questions. For the structured questions, Likert-type scales were used. The unstructured questions aimed at understanding the reasons behind participants’ satisfaction with the device and its auto mode functionality.

Following institutional review board approval, adults with T1DM using the Medtronic™ 670G pump paired with any brand of continuous glucose monitor (CGM) were recruited from an outpatient endocrinology clinic. It was anticipated that there would be some subjects who had adopted the hybrid system, and others who would use the 670G with some other type of CGMS, so both were included. Potential participants were identified through chart reviews by medical providers and were approached about the study during their regular outpatient visits. The survey was administered to participants during a scheduled appointment.

Survey responses were quantified to analyze associations and identify key themes. To quantify responses to the unstructured survey questions two coders manually and independently categorized them into emerging themes. Inconsistencies were resolved by consensus.

Results

Twenty-one participants completed the survey. Nineteen participants provided demographic information (Table 1).

Table 1.

Demographic Information for 19 of the 21 Participants.

Demographic	Count (%), or mean (standard deviation)
Age (years)	44.3 (18)
HbA1c	6.7% (0.7%)
Length of diagnosis (years)	26 (19)
Insulin pump usage (years)	13 (10)
Hybrid closed loop system usage (days)	154 (122)
# female	14 (74%)
# Caucasian	18 (95%)

Seventeen participants used the pump as part of a hybrid closed loop system. For the remaining participants, one paired the pump with the Enlite® CGM, and three participants paired the pump with Dexcom® CGM.

All 21 study participants were asked questions related to the pump. For those who did not use the hybrid closed loop system, opinions were divided between very satisfied (1), neither satisfied or dissatisfied (1), dissatisfied (1), and very dissatisfied (1). For those who used the pump as part of the hybrid closed loop, the views were mostly positive: very satisfied (7), satisfied (6), dissatisfied (1), and very dissatisfied (3).

When all the participants were asked what they liked most (Table 2), better control and more consistent BG levels were the most liked featured (15), followed by equally rated general ease of use (3) and technical features (3). The screen/physical display (2) was also liked. One participant indicated not liking anything about the pump. When discussing least like features of the pump (Table 3), physical and structural issues came first in the list (6). It was followed by increased user input (5), high alert frequency (4), difficulty of use (3), BG levels kept too high (3), and settings not flexible enough (2).

Table 2.

Most Liked Features of the Pump.

Category of comment	Count	Pump use	Quotes
Better control/more consistent BG levels	15	Pump & CGM	“It will help me to obtain more consistent BG” (P1, 1-week user)
		Hybrid closed loop	“Allowing the auto function has eliminated drastic low blood sugars” (P3) “No need for basals, micro-boluses” (P5) “Keeps me within levels more than other systems” (P7) “Better sensor accuracy, intelligent algorithm compensating for highs and lows” (P8) “Better control” (P9) “CGM and auto mode functionality” (P10) “Smart guard feature” (P11) “It adjusts to me and my life” (P12) “Auto mode helps me a lot” (P13) “Suspend on and before low. It’s nice if my BG isn’t super low and I’m not able to eat something immediately” (P14) “The ability to control my levels better” (P16) “CGM and basal adjustment” (P18) “Better BG control, easier adjustments” (P20) “It alerts me to high and lows. It helps keep my blood sugars in my target range” (P21)
General ease of use	3	Pump & CGM	—
		Hybrid closed loop	“I like that my CGM and pump are on one device” (P13) “Talks with sensor” (P17) “Easy to use. Has proved to be intuitive technology and easy to use” (P19)
Technical features	3	Pump & CGM	—
		Hybrid closed loop	“It calibrates faster” (P2) “Waterproof” (P5, P11)
Screen/physical display	2	Pump & CGM	“The screen is more aesthetically pleasing compared to the older Medtronic pumps” (P6) “Color screen” (P15)
		Hybrid closed loop	—
Nothing	1	Pump & CGM	“I don’t like it” (P4)
		Hybrid closed loop	—

Table 3.

Least Liked Features of the Pump.

Category of comment	Count	Pump use	Quotes
Physical design/structural issues	6	Pump & CGM	“Orientation on the body for ease of reviewing” (P1) “The way its shaped and performs too many options” (P4) “The pump battery dies very quickly” (P6)
		Hybrid closed loop	“Wired. Used to have Omnipod wireless and loved it” (P17) “Certain parts seem to break compromising the integrity of the device” (P19) “Inserting the infusion sets, because sometimes they bend” (P21)
Need for increased user input	5	Pump & CGM	“The sensor is very high maintenance. It asks for blood glucose readings constantly. It does not give any data if you do not calibrate every time it asks” (P6) “Multiple clicking to deliver bolus/change settings” (P15)
		Hybrid closed loop	“Calibration and BG input requirements sometimes seem to step on each other” (P8) “Number of calculations when inserting new sensor” (P9) “Requires a lot of diagnosis BG reading (sometimes 3 in a row when I’m sleeping) and that it stops reading my BG when I need to calibrate” (P13)
High alert frequency	4	Pump & CGM	—
		Hybrid closed loop	“The rate change alerts could be better” (P2) “Alarms and frequent BG notifications” (P5) “Alerts” (P11) “Buzzes a lot with reminders” (P12)
Difficulty of use	3	Pump & CGM	—
		Hybrid closed loop	“I don’t like wearing it” (P16) “When you are in auto mode and it goes to manual mode it doesn’t remember or help with lows until the new blood glucose” (P10) “Difficulty of use in the beginning, I made a lot of mistakes” (P20)
Blood glucose levels kept too high	3	Pump & CGM	“Auto-mode did not control my blood sugars. My BG was much higher than I am comfortable with it being” (P6) “I don’t like auto mode. Correction to 120 mg/dL is too high, and there is no reasonable way to give a correction bolus without entering in carbs that I’m not eating” (P14)
		Hybrid closed loop	“Running higher than I like” (P3)
Settings not flexible enough	2	Pump & CGM	—
		Hybrid closed loop	“Inability to adjust bolus calculations” (P7) “Inflexible target range” (P18)

The 17 participants who used the hybrid closed loop were asked to answer questions related to the auto mode feature of the pump. Survey responses revealed that participants were mostly satisfied with the auto mode feature in the pump: very satisfied (3), satisfied (8), neither satisfied nor dissatisfied (3), dissatisfied (1), and very dissatisfied (2).

Participants’ responses to what they liked the most and least about the auto mode feature of the pump and why are presented in Tables 4 and 5. As most liked features, better control and more consistent BG level (9) was followed by ease of use (5). Few participants indicated that they liked nothing about the auto mode (2). Satisfaction with insulin delivery was also mentioned (1). Between the least liked features, pump allowing BG to remain elevated was top rated (5), followed by complains about high alert frequency (4), dissatisfaction with insulin administration (3), and settings not flexible enough (2). Few participants indicated that there was nothing about the auto mode that they disliked (2). One participant indicating not using the auto mode feature.

Table 4.

Most Liked Features of Auto Mode of the Pump.

Category of comment	Count	Quotes
Better control/more consistent BG levels	9	“Correcting low problems” (P3) “Keeps me in acceptable range for longer periods of time” (P7) “More uniform stable BG” (P8) “Safety” (P9) “Helps with low blood sugars” (P10) “Better control” (P11) “It has helped keep me from going low, which I used to go low very often (at least once or twice a day)” (P13) “Tighter control. Less variation” (P16) “It helps keep my blood sugars in target range” (P21)
Ease of use	5	“The pump takes over based on CGM readings” (P5) “Does everything for you” (P17) “Works most of the time” (P18) “Self-sufficient system” (P19) “Easy to use” (P20)
Nothing	2	“Nothing. It is not ready for Prime Time” (P2) “Literally nothing” (P14)
Satisfaction with insulin delivery	1	“It still delivers insulin” (P12)

Table 5.

Least Liked Features of Auto Mode of the Pump.

Category of comment	Count	Quotes
Blood glucose levels kept too high	5	“Running higher than I prefer” (P3) “Can’t adjust basal without “cheating” on carb intake” (P7) “Correction to 120 mg/dL is too high, and there is no reasonable way to give a correction bolus without entering in carbs that I’m not eating” (P14) “Tends to keep me at higher BG levels than when I’m manual mode” (P19) “The auto mode does not calculate adequate boluses for high blood sugars” (P21)
Alerts	4	“Frequent BG alarms” (P5) “Alarms for calibration input” (P8) “Beeping sound” (P9) “Lots of alerts” (P20)
Dissatisfaction with insulin administration	3	“Insulin dosages are extremely small” (P2) “It kicks on too quick if I haven’t adjusted or calibrated” (P12) “Doesn’t back off or give more insulin based on where levels are headed” (P16)
Nothing	2	“Nothing” (P10, P17)
Settings not flexible enough	2	“That it has a target BG sent and I cannot change it. Also, that it has a max and min amount of insulin it can give for a certain period of time” (P13) “Inflexible target range” (P18)
Do not use auto mode	1	“N/A” (P11)

Participants who used the hybrid closed loop were asked if they used the manual mode feature of the pump. Frequency of usability varied: always (3), very often (3), often (5), rarely (5), and never (1). Participants were also asked why they would switch from auto to manual mode, if they do so (Table 6). Reasons for switching to manual mode were being recommended by the pump (4), having to change sensors (4), correcting for high BG (2), calibration issues (2), and others (1). This question was not applicable to four participants, who indicated not using the auto or manual mode.

Table 6.

Reasons for Switching from Auto Mode to Manual Mode in the Pump.

Category of comment	Count	Example quotes
Pump recommendation	4	“Automatically done by pump” (P5) “The pump automatically switches because of the max or min micro bolus amount” (P13) “The pump recommends it to be done” (P20) “High blood sugar, directed by pump” (P21)
Changing sensors	4	“Don’t have sensor or chose not to put sensor back in right away” (P7) “Changing sensors, unable to supply BG reading” (P8) “Taking a break from sensor” (P9) “Sensor” (P16)
Do not use auto mode	3	“N/A” (P11) “By mistake” (P12) “I don’t use auto mode” (P14)
Correction of high blood sugar	2	“To get a larger correction bolus” (P2) “Typically see lower BG levels in manual mode than in auto mode” (P19)
Calibration issues	2	“Unsure of the reason why the pump is not getting the BG” (P10) “Not able to calibrate” (P17)
Do not use manual mode	1	“N/A” (P3)
Other	1	“Lack of supplies due to insurance” (P18)

Discussion

It is important to evaluate technology usability concerns. Existing research on psychosocial impacts of hybrid closed-loops systems in the management of the diabetes indicate generally positive user experience and the identification of challenges to be address to improve usability and user experience.¹⁹

This pilot user experience survey provides initial insights into patient perceptions on the pump that is the first FDA approved hybrid closed loop system available to patients. Preliminary studies on the pump showed overall general improvement in patient satisfaction according to preset quality of life measures such as anxiety and worry.²⁰

While our survey indicated that most participants felt this system allowed them to achieve better control and more consistent BG levels, others indicated that the pump kept their BG higher than they were used to and that administered insulin dosages were very small. Consistent with what was reported by Weaver and Hirsch, participants reported entering fake carbohydrate consumption into the pump to elicit a correction bolus, even when not eating.²¹ Patients compensated for this by administering “phantom carbohydrates” (“fake carbohydrates”) by informing the insulin pump they would eat carbohydrates without actually doing so. These findings are also consistent with patients’ feedback available online or through public forums that indicate that workarounds (fake carbohydrate input, exit into manual mode to manually bolus) are being used to address these dissatisfactions.^22,23

In addition, while some participants noted that they appreciated the reduced need for user input while in auto mode, a greater number reported that they felt there was actually an increased need for user input, which includes high alert frequency. This would appear counterintuitive for an auto mode and could be an important perception for the manufacturer to understand. If patients feel the increased mental burden of adjusting their pump settings or responding to frequent notifications and alarms is too much, it could be a barrier to adoption of this new technology.

One limitation of this pilot study is that the pump had been available for just about a year at the time of the survey, and the study participants using it were still new to the system. It would be of interest to do a follow-up assessment to determine changes in perception with longer patient experience. Furthermore, a larger sample size is needed. Because the data for this study are self-reported there is no observation or validation of actual patient conduct included. In addition, the average glucose control for this population was good in general (average HbA1c of 6.7%) indicating that these patients were accustomed to keeping BG in a tight range. The technology may be perceived differently by patients who are not accustomed to maintaining BG in such a tight range. Another limitation is that this user perception survey does not include the views of providers. The provider’s involvement in learning the technology, interpreting the downloaded data, and troubleshooting questions are open questions to be addressed.²¹

This is a pilot study on patient device satisfaction. Further information is needed on patients’ perceptions of this new technology, and how patient use of and confidence in the device changes over time. Addressing usability concerns and dissatisfaction, while maintaining the features that patients expressed satisfaction with, should be a consideration in future product design. Future work will include the use and adaptation of validated diabetes insulin delivery device satisfaction questionnaires.^24,25

Conclusion

In this pilot study, patients indicated a high level of satisfaction with the pump and its auto mode feature mostly because of improvements in BG control. They also pointed out some key aspects of the device that are of potential clinical or commercial relevance. Additional research is needed to further evaluate users’ perspectives on this new device.