Text-Messaging to Support Diabetes Self-Management in a Rural Health Clinic: A Quality Improvement Project

Citation

Russell, N., Vess, J., Durham, C. & Johnson, E. (Summer, 2017). Text-Messaging to Support Diabetes Self-Management in a Rural Health Clinic: A Quality Improvement Project. Online Journal of Nursing Informatics (OJNI), 21(2), Available at http://www.himss.org/ojni

Abstract

Diabetes is a chronic disease that requires lifelong treatment extending well beyond the primary care office.  While in-office visits are an essential component in disease management, ongoing patient support is requisite to impart and strengthen the knowledge, skills, and ability necessary for diabetes self-care.  Practice guidelines recommend diabetes self-management education (DSME) and support for all diabetic patients.  DSME promotes active participation in disease management through education, behavior modification, and active collaboration with healthcare personnel.  A quality improvement project involved interactive, bi-directional text-messaging interventions sent through a secure, messaging system.  Text messages included educational information, tips, or reminders related to medication, self-monitoring blood glucose, diet, and exercise.  After the intervention, a reduction in participants’ mean finger stick blood glucose (FSBG) level occurred.  The findings of this project suggest that mobile phone technology is a feasible, and sustainable means to augment current diabetes management with text-messaging interventions that promote DSME.

Clinical Issue

The prevalence of diabetes mellitus has more than doubled over the past 30 years nationwide (Centers for Disease Control and Prevention [CDC], 2014).  In fact, diabetes afflicts 29.1 million Americans and contributes to the death of more than 200,000 individuals annually (CDC, 2014).  As of 2012, the national prevalence of diabetes was 9% (South Carolina Department of Health and Environmental Control [SCDHEC], 2014) and 1.4 million Americans are diagnosed with diabetes annually.  Diabetes is the seventh leading cause of death in the U.S. and is the underlying cause of many co-morbid conditions and complications (CDC, 2014).  Consequently, primary care offices are inundated with an ever-expanding population of patients with diabetes who have multiple comorbid conditions and, often, one or more vascular complications. 

In 2010, the hospitalization rates for strokes were 1.5 times higher in adults with diabetes than in non-diabetic adults; additionally, diabetes was cited as the primary cause in 44% of new cases of renal failure (CDC, 2014).  Moreover, disease management costs an average of $245 billion each year secondary to microvascular and macrovascular complications, including stroke, myocardial infarction, diabetic retinopathy, kidney disease, and amputations, to list a few (CDC, 2014).  Notably, the statistics support that current interventions are limited as evidenced by the increase in disease prevalence and associated conditions.

Diabetes is a chronic disease that requires lifelong treatment extending beyond medicinal therapy.  Ongoing patient self-management and support are required to thwart or decrease diabetes associated complications and must be initiated immediately upon diagnosis.  Diabetes self-management education (DSME) is an evidence-based standard and is defined as the ongoing process to improve the knowledge, skills, and ability essential for diabetes self-management (American Diabetes Association [ADA], 2014).  The goal of DSME is to promote active participation in disease management through education, behavior modification, and active collaboration with healthcare personnel (ADA, 2014).  Ongoing collaboration and communication with one’s healthcare team imparts and strengthens the knowledge, skills, and ability necessary for diabetic self-care and better clinical outcomes, including a reduction in hemoglobin A1c (HbA1c) values, weight, diabetes associated complications, and healthcare cost, to list a few (ADA, 2014; Nundy et al., 2012; Siminerio, 2010). 

Face-to-face visits afford the opportunity for periodic DSME but often do not meet the DSME expectation of support that should occur between visits, as recommended by the American Diabetes Association (ADA, 2014).  The ADA has challenged clinicians to develop and employ innovative strategies that promote ongoing DSME and support to facilitate self-management (ADA, 2014). 

One innovative strategy that has shown to be effective is the use of mobile technology and text-messaging.  Mobile phone technology provides a means to expand upon the current management of diabetes that has proven to be insufficient in primary care practices across the U.S.  Studies reflect a synergistic relationship between the implementation of  text-messaging interventions in primary care and positive behavior modification among diabetic patients (Chang et al., 2014; Fischer et al., 2012; Liang et al., 2011; Nundy et al., 2014; Nundy, Dick, Solomon, & Peek, 2013; Yeager & Menachemi, 2011).  Frequent communication through text messages has been shown to foster the patient-provider relationship and, ultimately, leads to improved glycemic control (Chang et al., 2014; Fischer et al., 2012; Liang et al., 2011; Nundy et al., 2014; Nundy, Dick, Solomon, & Peek, 2013; Yeager & Menachemi, 2011).  Delivery of DSME via short messaging service (SMS) can improve communication and facilitate self-care behaviors including compliance with medication, diet, exercise, and blood glucose monitoring. 

Given the complexity and cost associated with diabetes management in the rural setting and the need for improved DSME utilization, technology is a proven means to meet the needs of these patients.  According to Chang et al. (2014), 91% of Americans own a cell phone and 79% use phones to send and receive text messages.  Furthermore, low-income households texted an average of 59 times a day compared to an average of 32 times a day in higher-income households (Smith, 2011).  Therefore, mobile phone technology is an accessible and promising means to augment current diabetic management and practice.  Text-messaging interventions may afford the opportunity to promote diabetes self-care, provide personal support, and empower patients to take charge of their personal health to achieve optimal glycemic control. 

The use of text-messaging DSME was implemented in a quality improvement (QI) project with diabetes patients in an impoverished community.  The specific clinical question was: In adult patients with diabetes seen at a rural primary care clinic, will text-messaging interventions that promote self-management lead to a decrease in mean finger stick blood glucose (FSBG) level over a 12-week period?

Literature Review

In a meta-analysis of 22 randomized trials, Liang et al. (2011) examined the impact of mobile phone interventions on diabetes self-management and ultimately, glycemic control.  Collectively, 1657 participants, with either Type I or Type II diabetes mellitus, received lifestyle modification and disease management support, including glucose monitoring, diet, exercise, and medication adjustment through mobile phone short message service (SMS) alone or in conjunction to the internet or clinical counseling sessions with a diabetes nurse educator.  In each trial, the researchers analyzed pre and post-intervention HbA1c values and compared the intervention to the control group.  Spanning a period of six months, mobile phone intervention groups depicted a 0.5% reduction in HbA1c levels as opposed to control groups, particularly in patients with Type II diabetes (Liang et al., 2011). 

In a quasi-experimental pilot study, Fisher et al. (2012) sought to evaluate the role of mobile phone technology in promoting diabetes self-management behaviors by using SMS intervention.  A total of 47 adult patients with Type II diabetes received text-messaging prompts regarding glucose measurement, appointment reminders, and healthy self-care behaviors to facilitate optimal disease outcome.  Results revealed that 79% of participants responded to more than 50% of the 1585 text messages, an indicator of ease of use.  Moreover, 66.4% of study participants provided glucometer readings on follow-up visits when prompted via text messaging as opposed to only 12% at pre-intervention office visits.  The participants affirmed that a cell-phone based text messaging program offers personal support, promotes accountability, and cultivates a bipartisan approach to diabetes management.  The researchers also concluded that text-messaging interventions improve glycemic control, prevent diabetes associated complications, and enhance quality of life (Fisher et al., 2012).

In a quasi-experimental study, Nundy et al. (2014) explored the use of mobile messaging to promote behavior modification in adults with type I or II diabetes mellitus.  A total of 348 participants were enrolled in the six month study, 74 in the intervention group and 274 in the control group.  Researchers implemented a CareSmart software program designed to provide self-management support and education through automated text messaging.  Behavioral interventions were measured including healthy eating, foot care, blood glucose monitoring, as well as medication adherence; the results depicted improvement among the CareSmart participants.  In a seven-day period, healthy eating increased from 4.5 days to 5.2 days; blood glucose monitoring climbed from 4.3 days to 4.9 days, and foot care rose from 3.6 days to 4.3 days; additionally, medication adherence improved from 83% to 91%.  Finally, HbA1c levels decreased in the CareSmart group from 7.9% pre-intervention to 7.2% following intervention (Nundy et al., 2014).

Yeager and Menachemi (2011) conducted a comprehensive review of 61 studies to assess the impact of text messaging interventions on healthcare outcomes.  Researchers concluded that SMS interventions significantly improved quality of life, enhanced diabetes self-management skills, decreased daily blood glucose levels, and reduced HbA1c values.  In fact, 50 of the 61 studies or 81% found that SMS interventions positively impact primary healthcare outcomes by enhancing DSM and improving glycemic control.

Nundy, Dick, Solomon, & Peek (2013) conducted a qualitative, post-controlled pilot study aimed to evaluate the feelings and beliefs of the individuals who participated in a text message-based diabetes self-management program.  Researchers conducted audiotaped interviews lasting approximately 60 minutes.  The interviews were transcribed with precision and entered into software designed to detect characteristic patterns.  Researchers concluded that SMS intervention enhanced awareness of the seriousness of DM, increased feeling of support, and promoted accountability for better self-management among participants.  Moreover, provider interaction and feedback fortified positive self-management behavior and redirected poor behavior (Nundy et al., 2013).

Findings from these studies suggest that SMS interventions have a positive impact on numerous aspects of health care, particularly chronic diseases such as diabetes mellitus.  The implementation of text-messaging communication programs have enhanced clinical outcomes among diabetic patients by cultivating self-management behavior modification, promoting compliance and adherence to the treatment plan, improving glycemic control, and in due course, preventing or reducing diabetes associated complications.  Moreover, enhancing communication among the patient, provider, and healthcare team fosters trust and commitment to improving the skills and knowledge essential for optimal management of diabetes mellitus. 

Methods

The purpose of this Quality Improvement (QI) project was to use text-messaging interventions to support diabetes self-management and to improve glycemic control.  The goal was to maximize the use of mobile phone technology in order to positively impact participants’ self-care knowledge and behaviors related to medication, self-monitoring blood glucose (SMBG), diet, and exercise as demonstrated by a decrease in the post-intervention mean FSBG obtained on follow-up visits in April, May, and June of 2016 compared to FSBG results in the same time period in 2015.

Framework

The Chronic Care Model (CCM) for improving diabetic management was used as the framework for this project.  The CCM focuses on chronic disease management and contains key elements essential to the success of this project, including health system, community, self-management support, decision support, clinical information systems, and delivery system design (ADA, 2014).  The Plan, Do, Study, Act model is a standard QI framework that allows for the rapid implementation and evaluation of the CCM QI strategies prior to implementation on an entire population (Gregg, Jirjis, & Garriss, 2007; Nundy et al., 2012).

Setting/Participants

This QI study was completed at a free clinic in the southeastern U.S.  The clinic provides free primary care to the uninsured and indigent adult residents of two counties.  Clinic personnel involved in the project included the clinic director, two nurse practitioners, and the medical coordinator.  The QI project participants included all patients with diabetes managed at the clinic that chose to participate in the text message program.  According to the clinic director, this patient population inconsistently self-monitored blood glucose levels and poorly adhered to their medication regimen with an average HbA1c of more than 9%.  In addition, a review of patient charts revealed that only 10% of diabetes patients had received education regarding diet and exercise.  Although most patients with diabetes treated in the clinic had a personal cell phone, any diabetic patient without a cell phone was provided with one through a prior clinic grant.  The mean age of participants was 50 years old.  Of the 49 participants, 65% were African American, 33% Caucasian, and 2% Hispanic.  Twenty-four of the participants were female; 67% of female participants were African American, 29% were Caucasian, and 4% were Hispanic.  Twenty-five participants were male, with 64% being African American and 9% Caucasian. 

Intervention

A staff educational presentation was provided to all clinic personnel prior to implementation of the text-messaging intervention.  The presentation included information on DSME and the use of text-messaging interventions to support and enhance current diabetes management within the clinic.  The content and frequency of each text message was shared with the clinic staff.  Furthermore, personnel were advised to contact the project director with any questions or concerns via email or phone for the duration of the project.

Text messages were sent through a secure, messaging system called CareMessage for a period of 12 weeks beginning April 1 and ending July 1, 2016.  The CareMessage software system contains prepopulated messages that contain educational information and specific tips related to exercise and diet for patients with diabetes; however, messages within the system may be altered or created to better serve a specific population.  Most text messages for this QI project were developed by the project director.  Prior to project implementation, CareMessage consent forms were obtained from all participants to corroborate permission to send text messages to the participants’ personal cell phones.

The interactive bi-directional text-messaging interventions were implemented for a period of 12 weeks, with two daily reminders and two weekly educational messages or tips.  The two daily messages aimed at gathering self-reported behavior data asked these questions: Did you remember to take your diabetes medication today?  Did you remember to check your blood sugar today?  Patients were asked to reply yes or no to each question daily.  Patients who replied no to either question received a text message response that reminded or reinforced the importance of medication compliance and SMBG.  Patients who replied yes to either question received a text message response that encouraged them to continue the good work (Table 1).

The bi-weekly messages contained educational information and tips pertaining to diet and exercise.  The dietary text messages were distributed on Tuesday of each week, and the exercise text messages were distributed on Friday of each week.  Each weekly message differed, as the following examples indicate: A healthy plate: ½ vegetables, ¼ protein, and ¼ carbs/starches (bread, rice, potatoes).  You can have a piece of fruit to complete your meal.  Try to exercise 30 minutes a day.  If you cannot exercise 30 minutes, start with 10 minutes and slowly add more minutes (Table 2).

Data Collection

The aim of this project was to identify changes in glycemic control, as measured by the mean FSBG prior to and after text-messaging interventions were utilized.  Patients’ aggregate mean FSBG for the three months of the project in 2016 were compared with patients’ mean aggregate FSBG for the corresponding three months in 2015.  Graphs were created to illustrate the pre and post-intervention mean FSBG for each month during the project as compared to the mean FSBG for the corresponding three months in 2015.  A frequency analysis via the CareMessage software was completed, and graphs were created to depict the following monthly data: the percentage of yes responses, no responses, or lack of patient responses to each daily question regarding SMBG and medication intake.  A biostatistician was consulted prior to and during the QI project, as well as during data analysis.  Microsoft Excel was utilized to calculate statistics and to produce graphs.  All patient data was de-identified prior to inputting into the Excel worksheet. 

Results

Participants’ aggregate mean FSBG in the pre-intervention period (npre=98) was 574.07, compared to 509.02 in the post-intervention timeframe (npost=90), representing an 11.3% reduction.  Figure 1 displays monthly results.  Following intervention, the mean FSBG decreased by 11.2% in April, 8.8% in May, and 14.3% in June 2016 from the pre-intervention months.

Discussion

The findings of this QI project reflect a reduction in participants’ mean FSBG level following text-messaging interventions that promoted DSME and support.  The project timeline was followed as intended, and the goal or desired outcome was met.  The text-messaging intervention for the proposed clinical change was consistently accepted as part of the usual care for all participants with diabetes.  Clinic staff employed bi-monthly phone calls to program participants in an effort to remind them to respond to each question daily as well as to read and apply the bi-weekly educational messages and tips.  Moreover, the importance of DSME and support via text message was re-emphasized on follow-up visits to facilitate patient commitment to self-management.  The percentage of yes responses to the daily medication and blood sugar questions increased monthly throughout the intervention period (Figure 2 and 3).

While the project was successful, some limitations did exist.  The biggest limitation was patient participation; however, despite the phone calls and encouragement during office visits, consistency remained an issue throughout the program, with four patients often failing to respond.  Additional limitations were technological issues with the CareMessage software whereby the system failed to deliver the daily text messages several times throughout the project; notably, this occurred more frequently during the initial weeks and decreased significantly with help from software engineers.  Still in June, the system failed to deliver the daily questions to 42 patients for two consecutive days.   Furthermore, data entry errors regarding mobile phone numbers during the early weeks of the project resulted in a lack of self-reported data for five participants: three incorrect mobile numbers and two entered incorrectly in the system.  Another project limitation is that diabetes control and management is best measured by HbA1c level.  This blood test is routinely obtained every three to six months for all patients as part of their diabetes care at the clinic and is considered a standard of care for diabetic patients (ADA, 2014).  However, the 12 week project period did not allow a viable time frame for this measurement. 

Recommendations

Although this QI project was successful, further research is necessary to determine if the effects were sustainable following the conclusion of the intervention.  Furthermore, HbA1c reduction is the marker for diabetes control; thus, it is vital that mobile phone interventions be implemented for a minimum of 16 weeks in order to correlate DSME and support via text-messaging to a statistically significant reduction in HbA1c.  Finally, studies to identify the best program design including time-frame or length of intervention, SMS frequency, communication direction, and text message content should be conducted being certain to compare each aspect to HbA1c pre and post intervention values. 

Conclusion

Currently, the approach to diabetes management includes face-to-face visits approximately every three to six months in which information, education, and encouragement are provided to patients concerning their diabetes treatment plan.  Often, a time-frame of approximately 20 minutes is allotted for the patient-provider interaction leaving little time to completely discuss the aforementioned or answer questions.  As such, patient-provider relations are constrained by limited or poor communication which prohibits the formation of collaborative partnerships that are fundamental for optimal diabetes management.  Face-to-face visits afford the opportunity for periodic DSME; however, these brief encounters do not meet the expectation of ongoing DSME which would require support between visits, as the ADA (2014) advocates.  Therefore, mobile phone technology is a promising means to augment current diabetic management which is commonly insufficient in primary care practices across the U.S. 

The implementation of text-messaging interventions that promote diabetes self-management was effective in lowering the mean FSBG level during a 12-week period.  Integration of an evidence-based text-messaging communication system may support diabetes self-management and promote patient adherence with daily medication regimens, SMBG, diet, and exercise in all primary care settings, leading to better patient outcomes and cost savings for diabetic care.  Such low cost technological interventions in clinical practice have the potential to positively impact diabetic care and outcomes and may significantly improve the management and outcomes of other chronic conditions. 

References

American Diabetes Association. (2014). Standards of medical care in diabetes-2014. Diabetes

Care, 37(Supplement 1): S14-S80.

Centers for Disease Control and Prevention. (2014). National diabetes statistics report, 2014:

Estimates of diabetes and its burden in the United States. Retrieved from http://www.cdc.gov/diabetes/pubs/statsreport14/national-diabetes-report-web.pdf

Chang, T., Gossa, W., Sharp, A., Rowe, Z., Kohatsu, L., Cobb, E. M., & Heisler, M. (2014).

Text messaging as a community-based survey tool: A pilot study. BMC Public Health, 14, 936-943.

Fischer, H. H., Moore, S. L., Ginosar, D., Davidson, A., Rice-Peterson, C. M., Durfee, M. J., . . .

Steele, A. W. (2012). Care by cell phone: Text messaging for chronic disease management. The American Journal of Managed Care, 18(2):e42-7.

Gregg, W., Jirjis, J., & Garriss, W. (2007). Experiences with the academic chronic care

collaborative. Clinical Diabetes, 25(2), 72-73.

Liang, X., Wang, Q., Yang, X., Cao, J, Chen, J., Mo, X., . . . GU, D. (2011). Effect of mobile

phone intervention for diabetes on glycemic control: A meta-analysis. Diabetic Medicine, 28(4), 455-63.

Nundy, S., Dick, J. J., Chou, C. H., Nocon, R. S., Chin, M. H., & Peek, M. E. (2014). Mobile

phone diabetes project led to improve glycemic control and net savings for Chicago plan participants. Health Affairs, 33(2):265-72.

Nundy, S., Dick, J. J., Goddu, A. P., Hogan, P., Lu, C. E., Solomon, M. C., . . . Peek, M. E.

(2012). Using mobile health to support the Chronic Care Model: Developing an institutional initiative. International Journal of Telemedicine and Applications, 2012, 1-8.

Nundy, S., Dick, J. J., Solomon, M. C., & Peek, M. E. (2013). Developing a behavioral model

for mobile phone-based diabetes interventions. Patient Education and Counseling, 90(1), 125- 132.

Siminerio, L. M. (2010). The role of technology and the Chronic Care Model. Journal of

Diabetes Science and Technology, 4(2), 470–475.

Smith, A. (2011). Americans and text messaging. September, 19, 2011. Pew Research Center.

Retrieved from http://www.pewinternet.org/2011/09/19/americans-and-text-messaging/

South Carolina Department of Health and Environmental Control. (2014). Coordinated chronic

disease fact sheet: Florence County. Retrieved from https://www.scdhec.gov/Health/docs/Epi/chronic/Florence.pdf

Yeager, V. A., & Menachemi, N. (2011). Text messaging in health care: A systematic review of

impact studies. Advances in Health Care Management, 11, 235-261.