As a comparator control, usual diabetes camp care with the participant's own insulin pump.

Glucose readings were taken every 5 minutes by the CGM. The percentage of time that the glucose concentration was less than the following ranges were calculated: < 50 mg/dl (2.8 mmol/L), < 70 mg/dl (3.9 mmol/L), 70-180 mg/dl (3.9 to 10.0 mmol/L), > 180 mg/dL (10.0 mmol/L).

Glucose readings were taken every 5 minutes by the CGM. The percentage of time that the glucose concentration was less than the following ranges were calculated: < 50 mg/dl (2.8 mmol/L), < 70 mg/dl (3.9 mmol/L), 70-180 mg/dl (3.9 to 10.0 mmol/L), > 180 mg/dL (10.0 mmol/L).

Glucose reading were taken every 5 minutes by the CGM. The glucose readings were averaged. 154 mg/dL was the estimated average glucose corresponding to a Hemoglobin A1C of 7.0%.

Glucose reading were taken every 5 minutes by the CGM. The glucose readings were averaged. 169 mg/dL was the estimated average glucose corresponding to a Hemoglobin A1C of 7.5%.

Glucose reading were taken every 5 minutes by the CGM. The glucose readings were averaged. 183 mg/dL was the estimated average glucose corresponding to a Hemoglobin A1C of 8.0%.

A series of hypoglycemic measurements is defined as a single event until there is a break of ≥ 30 minutes between measurements below the defined thresholds of < 70, < 60, and <50 mg/dL.

Fingerstick plasma glucose measurements were obtained before meals, at bedtime, at midnight, and at about 3:45 AM (six scheduled measurements). The plasma glucose readings were averaged. The plasma glucose results on Day 1, Days 1 to 5 and Days 2 to 5 were averaged.

Fingerstick plasma glucose measurements were obtained before meals, at bedtime, at midnight, and at about 3:45 AM (six scheduled measurements). The percentage of time that the plasma glucose concentration was less than the following ranges were calculated: < 70 mg/dl (3.9 mmol/L), < 60 mg/dL (3.3 mmol/L), < 50 mg/dl (2.8 mmol/L).

Fingerstick plasma glucose measurements were obtained before meals, at bedtime, at midnight, and at about 3:45 AM (six scheduled measurements). The percentage of time that the plasma glucose concentration was less than the following ranges were calculated: < 70 mg/dL (3.9 mmol/L), < 60 mg/dL (3.3 mmol/L), < 50 mg/dL (2.8 mmol/L).

Fingerstick plasma glucose measurements were obtained before meals, at bedtime, at midnight, and at about 3:45 AM (six scheduled measurements). The percentage of time that the plasma glucose concentration was less than the following ranges were calculated: < 70 mg/dl (3.9 mmol/L), < 60 mg/dL (3.3 mmol/L), < 50 mg/dl (2.8 mmol/L).

Fingerstick plasma glucose measurements were obtained before meals, at bedtime, at midnight, and at about 3:45 AM (six scheduled measurements). The plasma glucose readings were averaged. 154 mg/dL was the estimated average glucose corresponding to a Hemoglobin A1C of 7.0%.

Fingerstick plasma glucose measurements were obtained before meals, at bedtime, at midnight, and at about 3:45 AM (six scheduled measurements). The plasma glucose readings were averaged. 169 mg/dL was the estimated average glucose corresponding to a Hemoglobin A1C of 7.5%.

Fingerstick plasma glucose measurements were obtained before meals, at bedtime, at midnight, and at about 3:45 AM (six scheduled measurements). The plasma glucose readings were averaged. 183 mg/dL was the estimated average glucose corresponding to a Hemoglobin A1C of 8.0%.

A series of hypoglycemic measurements is defined as a single event until there is a break of ≥ 30 minutes between measurements below the defined thresholds of < 70, < 60, and <50 mg/dL.

The percentage of days during the Usual Care period that participants used CGM data as part of their diabetes management.

Fingerstick plasma glucose measurements were obtained before meals, at bedtime, at midnight, and at about 3:45 AM (six scheduled measurements). Participants were given 15 grams (g) of simple carbohydrates if their plasma glucose concentration dropped below 4.4 mmol/L. These simple carbohydrates were counted as interventions for study outcomes if the plasma glucose concentration was less than 3.9 mmol/L. A second intervention of 15 g of carbohydrate was given if a repeat measurement in 15-20 min was less than 3.9 mmol/L. The total number of carbohydrate interventions are reported.

Fingerstick plasma glucose measurements were obtained before meals, at bedtime, at midnight, and at about 3:45 AM (six scheduled measurements). Participants were given 15 grams (g) of simple carbohydrates if their plasma glucose concentration dropped below 4.4 mmol/L. These simple carbohydrates were counted as interventions for study outcomes if the plasma glucose concentration was less than 3.9 mmol/L. A second intervention of 15 g of carbohydrate was given if a repeat measurement in 15-20 min was less than 3.9 mmol/L.

Insulin total daily dose is reported in units per kilogram per day (U/kg/day).

Glucagon dose level is reported in micrograms per kilogram of body mass per day (µg/kg/day).

The bionic pancreas automatically adapted insulin dosing to each individual's needs. When CGM data were not available (because of sensor failure or during the warm-up time after sensor replacement), the bionic pancreas automatically delivered a dose of basal insulin based on the mean basal dosing it had calculated at that time on previous days. Daily basal insulin dose is reported in units per kilogram (kg) per day (U/kg).

The first time the bionic pancreas was used in each participant, a partial meal-priming bolus based on the participant's body mass (0.05 units/kg) was delivered. After the first use, the size of the meal-priming bolus was adapted by the bionic pancreas to 75% of the 4-hour prandial insulin used for that meal type and size. Daily bolus insulin dose is reported in units per kilogram (kg) per day (U/kg).

Carbohydrate intake included meals, snacks and unscheduled carbohydrates administered when a participant's blood glucose was <80 mg/dl (or for symptoms at any glucose level). Carbohydrate intake per day was averaged and is reported in grams (g) per kilogram (kg) per day (g/kg/day).

Infusion sets for administering insulin and glucagon were placed under the skin the in the abdomen, buttocks, arms, or legs. Infusion set changes due to pain, infusion set falling out or infusion set failure are reported. Camp policy was to suspect failure of the infusion set whenever ketonemia occurred, so failures may not have actually been set failures, but rather failure to deliver enough insulin.

Infusion sets for administering insulin and glucagon were placed under the skin the in the abdomen, buttocks, arms, or legs. Infusion set changes due to pain, infusion set falling out or infusion set failure are reported. Camp policy was to suspect failure of the infusion set whenever ketonemia occurred, so failures may not have actually been set failures, but rather failure to deliver enough insulin.

Infusion sets for administering insulin and glucagon were placed under the skin the in the abdomen, buttocks, arms, or legs. Infusion set changes due to pain, infusion set falling out or infusion set failure are reported. Camp policy was to suspect failure of the infusion set whenever ketonemia occurred, so failures may not have actually been set failures, but rather failure to deliver enough insulin.

Local infusion site reactions are defined as pain at the infusion site of the Bionic Pancreas. Itching and redness may have also been present.

Participants rated their nausea using a 0 to 10 centimeter (cm) VAS where 0=least severe nausea to 10=most severe nausea. The average nausea index scores from Day 1 to Day 5 were calculated

A severe hypoglycemic event is an event where the participant is unable to self-treat and requires the assistance of another person.

Percentage of time that the Bionic pancreas was not functioning properly due to loss of wireless connectivity.

Percentage of time without CGM monitoring data is the time when the participant's CGM device lost its CGM signal.

The change in body weight collected at Day 5 of each period relative to Baseline. A negative change from Baseline indicates a reduction in body weight and a positive change from Baseline indicates an increase in body weight.

Reliability index was calculated as the percentage of time values were actually recorded by CGM.

The number of time a CGM sensor was replaced due to falling out or failing to report a glucose value.

The percentage of participants who used a GLP-1 agonist to manage their diabetes during the Usual Care period.

The percentage of participants who used pramlintide to manage their diabetes during the Usual Care period.

Glucose reading were taken every 5 minutes by the CGM. The percentage of time that the glucose concentration was less than 60 mg/dL [3.3 millimoles/liter (mmol/L)] during Days 2 through 5 was calculated.

All randomized participants who completed both periods of the study.

Glucose reading were taken every 5 minutes by the CGM. The glucose results for each time frame were averaged.

All randomized participants who completed both periods of the study.

Glucose readings were taken every 5 minutes by the CGM. The percentage of time that the glucose concentration was less than the following ranges were calculated: < 50 mg/dL(2.8 mmol/L), < 70 mg/dL (3.9 mmol/L), 70-180 mg/dL (3.9 to 10.0 mmol/L), > 180 mg/dL (10.0 mmol/L).

All randomized participants who completed both periods of the study.

Glucose readings were taken every 5 minutes by the CGM. The percentage of time that the glucose concentration was less than the following ranges were calculated: < 50 mg/dl (2.8 mmol/L), < 70 mg/dl (3.9 mmol/L), 70-180 mg/dl (3.9 to 10.0 mmol/L), > 180 mg/dL (10.0 mmol/L).

All randomized participants who completed both periods of the study.

Glucose readings were taken every 5 minutes by the CGM. The percentage of time that the glucose concentration was less than the following ranges were calculated: < 50 mg/dl (2.8 mmol/L), < 70 mg/dl (3.9 mmol/L), 70-180 mg/dl (3.9 to 10.0 mmol/L), > 180 mg/dL (10.0 mmol/L).

All randomized participants who completed both periods of the study.

Glucose reading were taken every 5 minutes by the CGM. The glucose readings were averaged. 154 mg/dL was the estimated average glucose corresponding to a Hemoglobin A1C of 7.0%.

All randomized participants who completed both periods of the study.

Glucose reading were taken every 5 minutes by the CGM. The glucose readings were averaged. 169 mg/dL was the estimated average glucose corresponding to a Hemoglobin A1C of 7.5%.

All randomized participants who completed both periods of the study.

Glucose reading were taken every 5 minutes by the CGM. The glucose readings were averaged. 183 mg/dL was the estimated average glucose corresponding to a Hemoglobin A1C of 8.0%.

All randomized participants who completed both periods of the study.

A series of hypoglycemic measurements is defined as a single event until there is a break of ≥ 30 minutes between measurements below the defined thresholds of < 70, < 60, and <50 mg/dL.

Fingerstick plasma glucose measurements were obtained before meals, at bedtime, at midnight, and at about 3:45 AM (six scheduled measurements). The plasma glucose readings were averaged. The plasma glucose results on Day 1, Days 1 to 5 and Days 2 to 5 were averaged.

All randomized participants who completed both periods of the study.

Fingerstick plasma glucose measurements were obtained before meals, at bedtime, at midnight, and at about 3:45 AM (six scheduled measurements). The percentage of time that the plasma glucose concentration was less than the following ranges were calculated: < 70 mg/dl (3.9 mmol/L), < 60 mg/dL (3.3 mmol/L), < 50 mg/dl (2.8 mmol/L).

All randomized participants who completed both periods of the study.

Fingerstick plasma glucose measurements were obtained before meals, at bedtime, at midnight, and at about 3:45 AM (six scheduled measurements). The percentage of time that the plasma glucose concentration was less than the following ranges were calculated: < 70 mg/dL (3.9 mmol/L), < 60 mg/dL (3.3 mmol/L), < 50 mg/dL (2.8 mmol/L).

All randomized participants who completed both periods of the study.

Fingerstick plasma glucose measurements were obtained before meals, at bedtime, at midnight, and at about 3:45 AM (six scheduled measurements). The percentage of time that the plasma glucose concentration was less than the following ranges were calculated: < 70 mg/dl (3.9 mmol/L), < 60 mg/dL (3.3 mmol/L), < 50 mg/dl (2.8 mmol/L).

All randomized participants who completed both periods of the study.

Fingerstick plasma glucose measurements were obtained before meals, at bedtime, at midnight, and at about 3:45 AM (six scheduled measurements). The plasma glucose readings were averaged. 154 mg/dL was the estimated average glucose corresponding to a Hemoglobin A1C of 7.0%.

All randomized participants who completed both periods of the study.

Fingerstick plasma glucose measurements were obtained before meals, at bedtime, at midnight, and at about 3:45 AM (six scheduled measurements). The plasma glucose readings were averaged. 169 mg/dL was the estimated average glucose corresponding to a Hemoglobin A1C of 7.5%.

All randomized participants who completed both periods of the study.

Fingerstick plasma glucose measurements were obtained before meals, at bedtime, at midnight, and at about 3:45 AM (six scheduled measurements). The plasma glucose readings were averaged. 183 mg/dL was the estimated average glucose corresponding to a Hemoglobin A1C of 8.0%.

All randomized participants who completed both periods of the study.

A series of hypoglycemic measurements is defined as a single event until there is a break of ≥ 30 minutes between measurements below the defined thresholds of < 70, < 60, and <50 mg/dL.

The percentage of days during the Usual Care period that participants used CGM data as part of their diabetes management.

All randomized participants who completed both periods of the study.

Fingerstick plasma glucose measurements were obtained before meals, at bedtime, at midnight, and at about 3:45 AM (six scheduled measurements). Participants were given 15 grams (g) of simple carbohydrates if their plasma glucose concentration dropped below 4.4 mmol/L. These simple carbohydrates were counted as interventions for study outcomes if the plasma glucose concentration was less than 3.9 mmol/L. A second intervention of 15 g of carbohydrate was given if a repeat measurement in 15-20 min was less than 3.9 mmol/L. The total number of carbohydrate interventions are reported.

All randomized participants who completed both periods of the study.

Fingerstick plasma glucose measurements were obtained before meals, at bedtime, at midnight, and at about 3:45 AM (six scheduled measurements). Participants were given 15 grams (g) of simple carbohydrates if their plasma glucose concentration dropped below 4.4 mmol/L. These simple carbohydrates were counted as interventions for study outcomes if the plasma glucose concentration was less than 3.9 mmol/L. A second intervention of 15 g of carbohydrate was given if a repeat measurement in 15-20 min was less than 3.9 mmol/L.

All randomized participants who completed both periods of the study.

Insulin total daily dose is reported in units per kilogram per day (U/kg/day).

All randomized participants who completed both periods of the study.

Glucagon dose level is reported in micrograms per kilogram of body mass per day (µg/kg/day).

All participants who completed both periods of the study. Results are reported for the Bionic Pancreas period only.

The bionic pancreas automatically adapted insulin dosing to each individual's needs. When CGM data were not available (because of sensor failure or during the warm-up time after sensor replacement), the bionic pancreas automatically delivered a dose of basal insulin based on the mean basal dosing it had calculated at that time on previous days. Daily basal insulin dose is reported in units per kilogram (kg) per day (U/kg).

All participants who completed both periods of the study. Results are reported for the Bionic Pancreas period only.

The first time the bionic pancreas was used in each participant, a partial meal-priming bolus based on the participant's body mass (0.05 units/kg) was delivered. After the first use, the size of the meal-priming bolus was adapted by the bionic pancreas to 75% of the 4-hour prandial insulin used for that meal type and size. Daily bolus insulin dose is reported in units per kilogram (kg) per day (U/kg).

All participants who completed both periods of the study. Results are reported for the Bionic Pancreas period only.

Carbohydrate intake included meals, snacks and unscheduled carbohydrates administered when a participant's blood glucose was <80 mg/dl (or for symptoms at any glucose level). Carbohydrate intake per day was averaged and is reported in grams (g) per kilogram (kg) per day (g/kg/day).

All randomized participants who completed both periods of the study.

Infusion sets for administering insulin and glucagon were placed under the skin the in the abdomen, buttocks, arms, or legs. Infusion set changes due to pain, infusion set falling out or infusion set failure are reported. Camp policy was to suspect failure of the infusion set whenever ketonemia occurred, so failures may not have actually been set failures, but rather failure to deliver enough insulin.

All randomized participants who completed both periods of the study.

Infusion sets for administering insulin and glucagon were placed under the skin the in the abdomen, buttocks, arms, or legs. Infusion set changes due to pain, infusion set falling out or infusion set failure are reported. Camp policy was to suspect failure of the infusion set whenever ketonemia occurred, so failures may not have actually been set failures, but rather failure to deliver enough insulin.

All randomized participants who completed both periods of the study.

Infusion sets for administering insulin and glucagon were placed under the skin the in the abdomen, buttocks, arms, or legs. Infusion set changes due to pain, infusion set falling out or infusion set failure are reported. Camp policy was to suspect failure of the infusion set whenever ketonemia occurred, so failures may not have actually been set failures, but rather failure to deliver enough insulin.

All randomized participants who completed both periods of the study.

Local infusion site reactions are defined as pain at the infusion site of the Bionic Pancreas. Itching and redness may have also been present.

All randomized participants who completed both periods of the study. Results are reported for the Bionic Pancreas period only.

Participants rated their nausea using a 0 to 10 centimeter (cm) VAS where 0=least severe nausea to 10=most severe nausea. The average nausea index scores from Day 1 to Day 5 were calculated

All randomized participants who completed both periods of the study.

A severe hypoglycemic event is an event where the participant is unable to self-treat and requires the assistance of another person.

All randomized participants who completed both periods of the study.

Percentage of time that the Bionic pancreas was not functioning properly due to loss of wireless connectivity.

All randomized participants who completed both periods of the study. Reported for the Bionic Pancreas period only.

Percentage of time without CGM monitoring data is the time when the participant's CGM device lost its CGM signal.

All randomized participants who completed both periods of the study.

The change in body weight collected at Day 5 of each period relative to Baseline. A negative change from Baseline indicates a reduction in body weight and a positive change from Baseline indicates an increase in body weight.

All randomized participants who completed both periods of the study.

Reliability index was calculated as the percentage of time values were actually recorded by CGM.

All randomized participants who completed both periods of the study.

Data was not collected and analyzed to this level of detail.

The number of time a CGM sensor was replaced due to falling out or failing to report a glucose value.

All randomized participants who completed both periods of the study.

The percentage of participants who used a GLP-1 agonist to manage their diabetes during the Usual Care period.

All participants who completed both periods of the study. Results are reported for the Usual Care period only.

The percentage of participants who used pramlintide to manage their diabetes during the Usual Care period.

All participants who completed both periods of the study. Results are reported for the Usual Care period only.

Usual Care diabetes management in a diabetes camp environment including a nurse or nursing student assigned to each cabin and review and adjustment of the insulin regimen daily by a physician or nurse practitioner, all participants using the participant's own insulin pump and a continuous glucose monitor if they use one as part of their usual care, for 5 days.