Diabetes Mellitus
Conditions
Keywords
Insulin, Telemedicine
Brief summary
The purpose of this pilot study is to determine whether text message (and phone) communication can be effectively utilized to adjust long-acting insulin, compared to standard practice.
Detailed description
The current practice of insulin titration for diabetics requires multiple in-person clinic visits, during which a patient's long-acting insulin dose is adjusted until the optimal dose to control glycemia is reached. Finding this optimal dose can take weeks in an ideal setting, but often takes much longer in a busy urban clinic such as Bellevue Hospital Center. Relaying titration instructions to patients via phone and text message has the potential to decrease the titration timeline, thus reducing the number of clinic visits and the time it takes patients to reach their target blood glucose levels. For this pilot project, study staff will recruit patients who are initiating long-acting insulin treatment or initiating the titration of their existing long acting insulin treatment at Bellevue Hospital Center's Adult Primary Care Center. Patients who volunteer to enroll and provide informed consent will be randomized to one of two arms (MITI or current best practice arm) at the time of enrollment and stratified by whether the patient is initiating insulin treatment or initiating the titration of his/her existing insulin dose. The study staff will provide a cell phone (for temporary use) to any patients who are randomized to the MITI arm and don't own a personal cell phone (or whose personal cell phone is not able to receive the Sense Health text messages). Patients will use the cell phone free of cost to participate in the intervention (receive Sense Health text messages, send their fasting blood glucose levels, and speak with the diabetes nurse and study staff.) Patients in the MITI arm will receive automated text messages 5 weekdays per week, for up to 12 weeks, from Sense Health. These text messages will request the patient's fasting blood glucose level. The patient will reply with his/her fasting blood glucose value, which will be logged in password-protected accounts on www.sensehealth.com. The clinic's diabetes nurses will check each patient's fasting blood glucose level on www.sensehealth.com each weekday and call any patient with fasting blood glucose values \< 80 mg/dL and \> 400 mg/dL. Each Thursday, patients will receive a phone call from a nurse, who will adjust the patient's insulin dose according to the study titration protocol. Each patient will continue to receive daily text messages and weekly phone calls until the first of three events occur: the patient reaches his/her optimal insulin dose for achieving glycemic control, 12 weeks elapse, or the patient withdraws from the study. Patients in the current best practice arm (CBP) will attend scheduled clinic appointments during which a provider will review the patient's fasting blood glucose log and titrate the insulin dose according to current best practice. Patients in both arms will continue receiving routine care, including HbA1c values every 3 months and other routine labs and measures as per standard of care.
Interventions
OTHERMobile Insulin Titration Intervention
Patients send their fasting blood glucose levels to the clinic diabetes nurses via text message each weekday. The diabetes nurses call each patient once a week to give insulin titration instructions to replace in-person clinic visits for insulin titration.
Sponsors
New York City Health and Hospitals Corporation
NYU Langone Health
Study design
Allocation
RANDOMIZED
Intervention model
PARALLEL
Primary purpose
TREATMENT
Masking
NONE
Eligibility
Sex/Gender
ALL
Age
18 Years to 70 Years
Healthy volunteers
No
Inclusion criteria
* Initiating long-acting insulin treatment or initiating the titration of long-acting insulin treatment * Speaks English or Spanish * Hemoglobin A1c \> or = 8% * Able and willing to inject insulin * Able and willing to provide informed consent
Exclusion criteria
* Short-acting insulin treatment * Systemic glucocorticoids * Sustained elevated serum creatinine \> or = 1.5 mg/dL for men and \> or = 1.4 mg/dL for women * Hypoglycemia unawareness * Type 1 diabetes
Design outcomes
Primary
| Measure | Time frame |
|---|---|
| Percentage of Subjects Who Reach Optimal Long-acting Insulin Dose | 12 weeks |
Secondary
| Measure | Time frame | Description |
|---|---|---|
| Hemoglobin A1c | baseline, 12 weeks (approximately 3 months) | Change in hemoglobin A1c |
| Baseline Treatment Satisfaction | baseline | The Diabetes Treatment Satisfaction Questionnaire standard (DTSQs) will be used to measure the patient's satisfaction with diabetes treatment received prior to study participation. Scores on questionnaire range from 0 to 6: 0 = very dissatisfied, 6 = very satisfied. |
| Time to Reach Optimal Long-acting Insulin Dose | 12 weeks | The time it takes a patient to reach his/her optimal long-acting insulin dose will be measured for both study arms. |
| Change in Treatment Satisfaction | 12 weeks (approximately 3 months) | The Diabetes Treatment Satisfaction Questionnaire change (DTSQc) will be used to measure the change in the patient's satisfaction with his/her diabetes treatment since initiation of long-acting insulin titration. Scores on questionnaire range from -3 to +3: -3 = much less satisfied now, +3 = much more satisfied now. |
| Incidence of Hypoglycemia | 12 weeks | The number of instances of hypoglycemia as indicated by fasting blood glucose levels or symptoms reported by patients in both study arms. |
| Treatment Satisfaction After Initiation of Insulin Titration | 12 weeks (approximately 3 months) | The Diabetes Treatment Satisfaction Questionnaire standard (DTSQs) will be used to measure the patient's satisfaction with diabetes treatment received since initiation of long-acting insulin titration. Scores on questionnaire range from 0 to 6: 0 = very dissatisfied, 6 = very satisfied. |
Other
| Measure | Time frame | Description |
|---|---|---|
| Percentage of Text Message Responses | 12 weeks | The number of text message replies from participants compared to the total number of text messages sent to participants (asking for blood glucose values). This outcome is given as a percent. |
| Qualitative Patient Satisfaction Interview | After patient reaches optimal insulin dose or at 12 weeks | The study staff will interview MITI arm patients, using free-response questions, to assess their satisfaction with the intervention. The interviews will take place in person or over the phone at the patient's convenience, after the patient has reached his/her optimal insulin dose. If the patient does not reach optimal insulin dose, the interview will take place at approximately 12 weeks. |
| Percentage of Successful Phone Calls | 12 weeks | The number of successful insulin titration phone calls compared to the total number of insulin titration phone calls assigned to the nurse. Successful phone calls are defined as when the nurse was able to reach the participant with one call attempt, two call attempts, or by voicemail. This outcome is given as a percent. |
| Patient Healthcare Utilization | 12 weeks | The number of medication refill, emergency department, and walk-in clinic visits at Bellevue Hospital (non-insulin titration visits). |
| Costs - Provider Time Spent on Insulin Titration Visits | 12 weeks | Provider time spent on insulin titration visits by phone compared to insulin titration visits in the clinic. |
| Costs - Titration Visit Information | 12 weeks | The number of insulin titration visits (whether by phone or in the clinic). |
| Costs - Patient Travel Time | 12 weeks | The time it took patients to travel to Bellevue Hospital, reported by patients in both study arms at baseline and at any subsequent clinic visits. |
| Costs - Co-pays | baseline | At baseline, participants in both study arms (MITI and CBP) reported whether they had to pay co-pays for clinic visits at Bellevue Hospital. |
Countries
United States
Participant flow
Participants by arm
| Arm | Count |
|---|---|
| Mobile Insulin Titration Intervention Mobile Insulin Titration Intervention (MITI) arm patients will relay their fasting blood glucose levels to the study staff via text message. The patient will receive insulin titration instructions through a weekly phone call with a diabetes nurse.
Mobile Insulin Titration Intervention: Patients send their fasting blood glucose levels to the clinic diabetes nurses via text message each weekday. The diabetes nurses call each patient once a week to give insulin titration instructions to replace in-person clinic visits for insulin titration. | 33 |
| Current Best Practice Current Best Practice (CBP) arm patients will be treated according to the current best practice of insulin titration. They will attend scheduled clinic visits during which the provider will review their blood glucose logs and provide insulin titration instructions. | 28 |
| Total | 61 |
Withdrawals & dropouts
| Period | Reason | FG000 | FG001 |
|---|---|---|---|
| Overall Study | Did not enroll in texting platform | 1 | 0 |
| Overall Study | Did not initiate insulin treatment | 1 | 0 |
| Overall Study | Discontinued insulin - possible allergy | 0 | 1 |
| Overall Study | Phenotypic type 1 diabetes | 0 | 1 |
| Overall Study | Phone incompatible with texting platform | 3 | 0 |
Baseline characteristics
| Characteristic | Mobile Insulin Titration Intervention | Current Best Practice | Total |
|---|---|---|---|
| Age, Continuous | 48.48 years STANDARD_DEVIATION 11.22 | 44.61 years STANDARD_DEVIATION 9.97 | 46.70 years STANDARD_DEVIATION 10.75 |
| Sex: Female, Male Female | 15 Participants | 16 Participants | 31 Participants |
| Sex: Female, Male Male | 18 Participants | 12 Participants | 30 Participants |
Adverse events
| Event type | EG000 affected / at risk | EG001 affected / at risk |
|---|---|---|
| deaths Total, all-cause mortality | — / — | — / — |
| other Total, other adverse events | 3 / 28 | 3 / 27 |
| serious Total, serious adverse events | 0 / 28 | 0 / 27 |
Outcome results
Primary
Percentage of Subjects Who Reach Optimal Long-acting Insulin Dose
Time frame: 12 weeks
Population: No primary outcome data available for one participant in Current Best Practice arm who discontinued insulin early due to a mild possible allergy.
| Arm | Measure | Value (NUMBER) |
|---|---|---|
| Mobile Insulin Titration Intervention | Percentage of Subjects Who Reach Optimal Long-acting Insulin Dose | 88 percentage of participants |
| Current Best Practice | Percentage of Subjects Who Reach Optimal Long-acting Insulin Dose | 37 percentage of participants |
p-value: <0.0001Chi-squared
Secondary
Baseline Treatment Satisfaction
The Diabetes Treatment Satisfaction Questionnaire standard (DTSQs) will be used to measure the patient's satisfaction with diabetes treatment received prior to study participation. Scores on questionnaire range from 0 to 6: 0 = very dissatisfied, 6 = very satisfied.
Time frame: baseline
Population: All participants who completed the treatment satisfaction questionnaire at baseline.
| Arm | Measure | Value (MEAN) | Dispersion |
|---|---|---|---|
| Mobile Insulin Titration Intervention | Baseline Treatment Satisfaction | 4.99 score on satisfaction scale | Standard Deviation 1.14 |
| Current Best Practice | Baseline Treatment Satisfaction | 5.20 score on satisfaction scale | Standard Deviation 0.61 |
p-value: 0.78Wilcoxon rank-sum test
Secondary
Change in Treatment Satisfaction
The Diabetes Treatment Satisfaction Questionnaire change (DTSQc) will be used to measure the change in the patient's satisfaction with his/her diabetes treatment since initiation of long-acting insulin titration. Scores on questionnaire range from -3 to +3: -3 = much less satisfied now, +3 = much more satisfied now.
Time frame: 12 weeks (approximately 3 months)
Population: All participants who completed the treatment satisfaction questionnaire at 12 weeks.
| Arm | Measure | Value (MEAN) | Dispersion |
|---|---|---|---|
| Mobile Insulin Titration Intervention | Change in Treatment Satisfaction | 2.71 score on satisfaction scale | Standard Deviation 0.71 |
| Current Best Practice | Change in Treatment Satisfaction | 2.42 score on satisfaction scale | Standard Deviation 0.95 |
p-value: 0.13Wilcoxon rank-sum test
Secondary
Hemoglobin A1c
Change in hemoglobin A1c
Time frame: baseline, 12 weeks (approximately 3 months)
Population: All participants with hemoglobin A1c measurements recorded at baseline and 12 weeks.
| Arm | Measure | Value (MEAN) | Dispersion |
|---|---|---|---|
| Mobile Insulin Titration Intervention | Hemoglobin A1c | -1.90 mg/dL | Standard Deviation 2.64 |
| Current Best Practice | Hemoglobin A1c | -1.81 mg/dL | Standard Deviation 2.63 |
p-value: 0.99Wilcoxon rank-sum test
Secondary
Incidence of Hypoglycemia
The number of instances of hypoglycemia as indicated by fasting blood glucose levels or symptoms reported by patients in both study arms.
Time frame: 12 weeks
Population: This outcome was analyzed for participants who completed the allocated intervention (and the participant who discontinued insulin early due to a mild possible allergy). Five participants reported hypoglycemia: 3 in the MITI arm and 2 in the CBP arm. All cases were mild.
| Arm | Measure | Value (NUMBER) |
|---|---|---|
| Mobile Insulin Titration Intervention | Incidence of Hypoglycemia | 5 instances of hypoglycemia |
| Current Best Practice | Incidence of Hypoglycemia | 3 instances of hypoglycemia |
Secondary
Time to Reach Optimal Long-acting Insulin Dose
The time it takes a patient to reach his/her optimal long-acting insulin dose will be measured for both study arms.
Time frame: 12 weeks
Population: Participants who reached optimal long-acting insulin dose.
| Arm | Measure | Value (MEDIAN) |
|---|---|---|
| Mobile Insulin Titration Intervention | Time to Reach Optimal Long-acting Insulin Dose | 3.00 weeks |
| Current Best Practice | Time to Reach Optimal Long-acting Insulin Dose | 7.07 weeks |
p-value: 0.007Interval-censoring survival analysis
Secondary
Treatment Satisfaction After Initiation of Insulin Titration
The Diabetes Treatment Satisfaction Questionnaire standard (DTSQs) will be used to measure the patient's satisfaction with diabetes treatment received since initiation of long-acting insulin titration. Scores on questionnaire range from 0 to 6: 0 = very dissatisfied, 6 = very satisfied.
Time frame: 12 weeks (approximately 3 months)
Population: All participants who completed the treatment satisfaction questionnaire at 12 weeks.
| Arm | Measure | Value (MEAN) | Dispersion |
|---|---|---|---|
| Mobile Insulin Titration Intervention | Treatment Satisfaction After Initiation of Insulin Titration | 5.74 score on satisfaction scale | Standard Deviation 0.54 |
| Current Best Practice | Treatment Satisfaction After Initiation of Insulin Titration | 5.53 score on satisfaction scale | Standard Deviation 0.52 |
p-value: 0.04Wilcoxon rank-sum test
Other Pre-specified
Costs - Co-pays
At baseline, participants in both study arms (MITI and CBP) reported whether they had to pay co-pays for clinic visits at Bellevue Hospital.
Time frame: baseline
| Arm | Measure | Value (NUMBER) |
|---|---|---|
| Mobile Insulin Titration Intervention | Costs - Co-pays | 37 participants |
Other Pre-specified
Costs - Patient Travel Time
The time it took patients to travel to Bellevue Hospital, reported by patients in both study arms at baseline and at any subsequent clinic visits.
Time frame: 12 weeks
| Arm | Measure | Value (MEDIAN) |
|---|---|---|
| Mobile Insulin Titration Intervention | Costs - Patient Travel Time | 45 minutes |
Other Pre-specified
Costs - Provider Time Spent on Insulin Titration Visits
Provider time spent on insulin titration visits by phone compared to insulin titration visits in the clinic.
Time frame: 12 weeks
Population: Insulin titration visits with a duration recorded.
| Arm | Measure | Value (MEDIAN) |
|---|---|---|
| Mobile Insulin Titration Intervention | Costs - Provider Time Spent on Insulin Titration Visits | 6 minutes |
| Current Best Practice | Costs - Provider Time Spent on Insulin Titration Visits | 30 minutes |
p-value: 0.008Generalized estimating equation modeling
Other Pre-specified
Costs - Titration Visit Information
The number of insulin titration visits (whether by phone or in the clinic).
Time frame: 12 weeks
Population: Participants who completed the allocated intervention (and the participant who discontinued insulin early).
| Arm | Measure | Value (MEDIAN) |
|---|---|---|
| Mobile Insulin Titration Intervention | Costs - Titration Visit Information | 3.5 insulin titration visits |
| Current Best Practice | Costs - Titration Visit Information | 2 insulin titration visits |
p-value: 0.003Wilcoxon rank-sum test
Other Pre-specified
Patient Healthcare Utilization
The number of medication refill, emergency department, and walk-in clinic visits at Bellevue Hospital (non-insulin titration visits).
Time frame: 12 weeks
Population: All participants.
| Arm | Measure | Value (NUMBER) |
|---|---|---|
| Mobile Insulin Titration Intervention | Patient Healthcare Utilization | 6 hospital visits |
| Current Best Practice | Patient Healthcare Utilization | 11 hospital visits |
Other Pre-specified
Percentage of Successful Phone Calls
The number of successful insulin titration phone calls compared to the total number of insulin titration phone calls assigned to the nurse. Successful phone calls are defined as when the nurse was able to reach the participant with one call attempt, two call attempts, or by voicemail. This outcome is given as a percent.
Time frame: 12 weeks
Population: Participants who completed the allocated intervention.
| Arm | Measure | Value (NUMBER) |
|---|---|---|
| Mobile Insulin Titration Intervention | Percentage of Successful Phone Calls | 91 percentage of phone calls |
Other Pre-specified
Percentage of Text Message Responses
The number of text message replies from participants compared to the total number of text messages sent to participants (asking for blood glucose values). This outcome is given as a percent.
Time frame: 12 weeks
Population: Participants who completed the allocated intervention.
| Arm | Measure | Value (NUMBER) |
|---|---|---|
| Mobile Insulin Titration Intervention | Percentage of Text Message Responses | 84 percentage of text messages |
Other Pre-specified
Qualitative Patient Satisfaction Interview
The study staff will interview MITI arm patients, using free-response questions, to assess their satisfaction with the intervention. The interviews will take place in person or over the phone at the patient's convenience, after the patient has reached his/her optimal insulin dose. If the patient does not reach optimal insulin dose, the interview will take place at approximately 12 weeks.
Time frame: After patient reaches optimal insulin dose or at 12 weeks