Diabetes Self-Management Care via Cell Phone: A Systematic Review

Abstract

Background

The objective of this study was to evaluate the evidence on the impact of cell phone interventions for persons with diabetes and/or obesity in improving health outcomes and/or processes of care for persons with diabetes and/or obesity.

Methods

We searched Medline (1966–2007) and reviewed reference lists from included studies and relevant reviews to identify additional studies. We extracted descriptions of the study design, sample size, patient age, duration of study, technology, educational content and delivery environment, intervention and control groups, process and outcome measures, and statistical significance.

Results

In this review, we included 20 articles, representing 18 studies, evaluating the use of a cell phone for health information for persons with diabetes or obesity. Thirteen of 18 studies measured health outcomes and the remaining 5 studies evaluated processes of care. Outcomes were grouped into learning, behavior change, clinical improvement, and improved health status. Nine out of 10 studies that measured hemoglobin A1c reported significant improvement among those receiving education and care support. Cell phone and text message interventions increased patient–provider and parent–child communication and satisfaction with care.

Conclusions

Providing care and support with cell phones and text message interventions can improve clinically relevant diabetes-related health outcomes by increasing knowledge and self-efficacy to carry out self-management behaviors.

It is an ongoing challenge to provide care and support that will produce and sustain the desired improvements in the health of persons with a chronic illness such as diabetes. Quality health care requires effective collaboration between clinicians and patients.^1,2 Finding novel ways to enhance communication and improve the health of those with chronic diseases is also a continuing part of providing care. Interventions involving automated telephone message systems have been shown to improve knowledge and health outcomes.^3–5 Telephone-based interventions have had positive results even among persons of low socioeconomic status and ethnic minorities.⁶

According to the Cellular Telecommunications and Internet Association, there are over 255 million cell phone subscribers in the United States.⁷ Although income may seem to be a major barrier in cell phone ownership, every two out of three households in the United States have a cell phone.⁸ In a survey of those owning a cell phone, 35% said that they use it for text messaging.⁹ There is a need to examine if the various functions of cell phones, including text messaging, can help with providing better health care and lead to improved health outcomes.

Education and information technology are parts of diabetes care that have been studied. There are other reviews on health care telephone technology,^10,11 automated telephone messages,^4,12 cell phone technology,¹³ diabetes Web-assisted interventions,¹⁴ and diabetes-computerized learning technologies.¹⁵ However, no systematic review or meta-analysis of cell phone-based interventions for persons with diabetes and/or obesity exists to our knowledge that analyzes the evidence on the use of cell phones and text messaging interventions to improve health outcomes, processes of care, acceptance by users, and whether it is cost-effective.

The objective of this study was to evaluate the evidence on the impact of cell phone interventions in improving health outcomes and/or processes of care for persons with diabetes and/or obesity. We systematically reviewed studies to evaluate the impact of care and support interventions via cell phone in improving health outcomes and processes of care for persons with diabetes and/or obesity.

Methods

Data Sources

We searched Medline (1966–2007) for eligible trials using combinations of the following search terms: (1) diabetes mellitus (medical subject headings (MeSH)), type 1 diabetes mellitus (MeSH), type 2 diabetes mellitus (MeSH), or obesity (MeSH); (2) telephone (MeSH), cellular phone (MeSH), handheld computers (MeSH), cell phones, mobile phones, text messages, short message service (SMS), or personal digital assistants (PDA); and (3) patient education as topic (MeSH), health education (MeSH), patient education, or health education. We also systematically searched the reference lists of included studies and relevant reviews.

Study Selection and Data Extraction

The investigators reviewed the titles and abstracts of the identified citations and identified eligible articles based on the following criteria. Inclusion criteria included any randomized controlled trial, quasi-experimental study, or pre–post study evaluating the use of cell phones for health information for persons with diabetes and/or obesity. The included studies measured health outcomes or processes of care. Studies published in a language other than English with a complete English abstract were included if they met the specified inclusion criteria. The investigators collected the following information from each article that was eligible: descriptions of the study design, country, sample size, patient age, duration of study, technology, frequency, intervention and control groups, educational content and delivery, process and outcome measures, and statistical significance.

Results

Comprehensive literature searches identified 51 articles. The titles and abstracts of these articles were read, and 26 articles were determined to be potentially relevant. After reading the full articles, 20 articles representing 18 studies met the eligibility criteria. Articles were excluded if they did not focus on diabetes or obesity (8 articles), a cell phone or text messaging for health information or education was not used (15 articles), or health outcomes or processes of care were not measured (8 articles) (Tables 1 and 2).

Table 1.

Cell Phone Interventions and Health Outcomes

Author/year	Study design	Sample, age	Duration in months	Clinical area	Control	Intervention	Measures	Results C vs I or pre–post
Benhamou et al., 200716	RCT, crossover	30, 41.3 years	12	Type 1 diabetes	No weekly SMS support	Weekly clinical support via SMS	HbA1c	+0.12 vs −0.14%, P < 0.10
							SMBG	+5 vs −6 mg/dl, P = 0.06
							QOL score	0.0 vs +5.6, P < 0.05
							Satisfaction with life	−0.01 vs + 8.1, P < 0.05
							Hypoglycemic episodes	79.1 vs 69.1/patient, NS
							No. of BG tests/day	−0.16 vs −0.11/day, NS
Durso et al., 200317	Pre–post	7, 78.43 years	3	Type 2 diabetes	N/A	Personalized diabetes management messages	Convenience CP vs HP	3.71 vs 4.17
							Ease of use CP vs HP	2.57 vs 4.33
							Ease of recording	3.86
							Communication aid	4.28
							pre vs post: average knowledge score	74 vs 82
							HbA1c (%)	8 vs 7
							Check BG ≥once/day	57% vs 72%
							Check feet daily	100% vs 100%
							Never miss medication	85% vs 85%
							PA ≥5–6 days/week	29% vs 57%
							Hypo symptoms: never	43% vs 14%
							Hyper symptoms: never	57% vs 43%
							BMI	30 vs 29
Franklin et al., 200618	RCT	92, 8–18 years	12	Type 1 diabetes	CIT-Grp1	CIT+ST-Grp2, IIT+ST-Grp3	HbA1c	10.3 vs 10.1 vs 9.2%, P < 0.01
							Self-efficacy	56.0 vs 62.1, P < 0.01
							Adherence	70.4 vs 77.2, P < 0.05
Hurling et al., 200719	RCT	77, 40.4 years	4	Healthy	Verbal advice during clinic visit, no phone support	Cell phone support, i.e., exercise plan, PA charts, reminders, tailored advice	Change in:
							PA overall, MET min/week	4.0 vs 12, NS
							PA leisure time, MET min/week	−5.5 vs 4.1, P < 0.05
							Hours sitting: overall	−0.17 vs −2.18, P < 0.05
							Hours sitting: weekday	1.4 vs −5.9, P < 0.05
							Hours sitting: weekends	−0.2 vs −5.2, NS
							Accelerometer epochs	208.7 vs 218.5, P < 0.05
							BMI	0.10 vs −0.24, NS
							Lost % body fat	−0.17% vs −2.18%, P < 0.05
							BP, diastolic	0.73 vs 0.69, NS
							BP, systolic	0.41 vs 0.13, NS
							Perceived control	−0.37 vs 0.57, P < 0.01
							Intention to exercise	−0.01 vs 0.45, P < 0.01
							Internal control	5.85 vs 7.24, P < 0.001
							External control	5.33 vs 6.38, P < 0.01
Kim et al., 200520	Pre–post	42, 41.5 years	3	Type 2 diabetes	N/A	Blood glucose transfer and advice	FPG	−28.6 mg/dl, P < 0.05
							2HPPG	−78.4 mg/dl, P < 0.05
							TC	−13.5, NS
							HDL	+27.3, NS
							Satisfaction with care	+10.9, P < 0.05
Kim, 200521	RCT	34 adults	3	Type 2 diabetes and obesity	N/A	Educational messages	HbA1c	−120.1 mg/dl, P < 0.05
							2HPPG	1.2% vs 0%, P < 0.05
							Total cholesterol	NS
							TG	NS
							HDL	NS
Kim et al., 200622	Pre–post	45, 43.5 years	3	Type 2 diabetes	N/A	Educational messages	HbA1c	−1.1%, P < 0.01
							Diabetic diet	−0.8, days/week, NS
							Exercise	0.9 days/week, P < 0.05
							Medication	1.1 days/week, P < 0.05
							Foot care	1.1 days/week, P < 0.05
Kim, 200724	RCT	51, 47 years	3	Type 2 diabetes	Standard care during clinic visit	Weekly BG-based optimal recommedations via SMS	Group 1: <7%, pre–post:
							HbA1c	0.53 NS vs −0.21, P < 0.05
							FPG levels mg/dl	−5.8 NS vs −13.4, P < 0.05
							2HPMG	−3.1 NS vs −56.0, P < 0.05
							Group 2: ≥7%:
							HbA1c	0.22 NS vs −2.15, P < 0.05
							FPG levels mg/dl	14.5 NS vs −3.3 NS
							2HPMG	24.8, NS vs −115.2, NS
Kubota et al., 200426	Pre–post	136 adults	3	Obesity	N/A	i-exerM health education program	Weight reduction–M	−2.1 kg, P < 0.01
							Weight reduction–F	−1.2 kg, P < 0.01
							% perceived it effective	32%
Kwon et al., 200427	Pre–post	185, 42.4 years	3	Type 1 and type 2 diabetes	N/A	Advice on, dose adjustment, diet, exercise, and diabetes information	HbA1c, all patients	−0.5%, P < 0.001
							HbA1c, >7.0%	−0.9%, P < 0.001
							FPG levels	+6.6 mg/dl, NS
							Total cholesterol	−0.9 mg/dl, NS
							Triglycerides	−24.4 mg/dl, P < 0.01
							HDL–cholesterol	5.7 mg/dl, P < 0.05
Liu et al., 200528	Pre–post	274, 63 years	8	Type 2 diabetes	Usual care	POEM system and reminders	FPG	F = 7.898, P < 0.05
							HbA1c	F = 7.345, P < 0.05
							Total cholesterol	F = 4.139, P < 0.05
							Log-ins, 1–3 months	9.6/mo per patient
							Log-ins, 4–6 months	8.5/mo per patient
Rami et al., 200629	RCT	36, 15.3 years	6, 3-month cross-over	Type 1 diabetes	Conven-tional support and paper diary	Monitoring and support by SMS	HbA1c change 3 months	+1.0 vs −0.15
							HbA1c change 6 months	+0.15 vs −0.05
							3 months: HbA1c	0.07 vs −1.15%, P < 0.05
							FPG levels mg/dl	5.4 vs −8.0, NS
							2HPMG	14.7 vs −85.1 mg/dl, P < 0.05
							6 months: HbA1c	0.11 vs −1.05%, P < 0.05
Kim, 200723			3				FPG levels mg/dl	7.3 vs −5.8, NS
							2HPMG	13.8 vs −63.6 mg/dl, P < 0.05
Kim and Jeong, 200725	RCT	51, 47 years	6	Type 2 diabetes	Usual care and support	Weekly patient input of SMBG, medication details, diet, and exercise and optimal advice from a nurse via SMS or the Internet	9 months: HbA1c	0.33 vs −1.31, P < 0.05
							FPG levels mg/dl	12.2 vs −10.5, NS
							2HPMG	−17.4 vs −66.8, P < 0.05
Yoon and Kim, 200730			12				12 months: HbA1c	0.81 vs −1.32, P < 0.05
							FPG levels mg/dl	27.7 vs −10.7, NS
							2HPMG	18.1 vs −100, P < 0.05
							3−, 6−, 9−, 12-month change in:
							total cholesterol	NS
							triglycerides	NS
							HDL	NS

Notes: 2HPPG, 2-hour postprandial blood glucose; 2HPMG, 2-hour postmeal glucose, BG, blood glucose; BMI, body mass index; BP, blood pressure; CIT, conventional insulin therapy; CP, cell phone; F, female; FPG, fasting plasma glucose; HbA1c, hemoglobin A1c; HDL, high-density lipoprotein; HP, home phone; I, Internet; IIT, intensive insulin therapy; M, male; MET, metabolic equivalent; N/A, not available; NS, not significant; PA, physical activity; POEM, patient-oriented education management; QOL, quality of life; RCT, randomized controlled trial; SMBG, self-monitored blood glucose; SMS, short message service; ST, Sweet Talk.

Table 2.

Cell Phone Interventions and Process of Care

Author/year	Study design	Sample/age	Duration	Clinical area	Control	Intervention	Measures	ResultsC vs I
Aoki et al., 200531	Pre–post	30, 12–24 years	N/A	Type 1 diabetes	N/A	Educational game INSULOT	Entertainment score (1–7)	5.57
							Usability score (1–7)	5.44
							Recommend to others	>80%
Ferrer-Roca et al., 200432	Pre–post	23, 18–75 years	8 months	Type 2 diabetes	N/A	Diabetes management messages	Satisfaction	3.4 (range 1–5)
							Running costs	€3.0 to patient €3.75 to system
							System usage	33 messages/month
Gammon et al., 200533	Pre–post	15, 9–15 years	4 months	Type 1 diabetes	N/A	Data transfer 3 times/day	Parent vs child:
							Receiving BG	93% vs 80%
							Easier mgmt.	40% vs 13%
							Manual/auto BG transfer	93% vs 53%
Tasker et al., 200734	RCT	37, 7–18 years	12 months	Type 1 diabetes	Paper diary during clinic visit	Daily text message requesting response to questions	Frequency of hypos	5.2/month
							Hypos response rate	65% vs 89% vs 95%
							(Paper diary, CBI, CP)	CBI 54%
							Preference over diary	CP 65%
Wangberg et al., 200635	Pre–post	11, 9–15 years and parents	3 months	Type 1 diabetes	N/A	Diabetes education messages	User satisfaction	Satisfied
							Perceived pros	Parent–child communication
								Easily made part of daily lives
							Perceived cons	Unable to store or print all messages

Notes: BG, blood glucose; CBI, computer-based interviewing; CP, cell phone; hypos, hypoglycemic events; N/A, not available.

The final set of 18 studies included 1176 participants, with 6 studies involving 221 children and 12 involving 955 adults. Sample sizes ranged from 7 to 274 participants for adult studies and 11 to 92 participants for studies involving children. Three studies enrolled more than 130 participants.

Study duration ranged from 3 to 12 months, with the exception of one study lasting for the duration of a diabetes summer camp.³¹ Eight studies were of 3 months duration,^{17,20–22,25–27,35} two of 8 months duration,^28,32 one of 4 months duration,³³ and two of 6 months duration.^23,29 Four of the selected studies were of 12 months' duration.^16,18,30,34 In studies lasting more than 3 months, return visits for diabetes patients were scheduled for regular checkups and to obtain data for laboratory measurements.

Studies took place in nine different countries, including one in Austria,²⁹ two in Japan,^26,31 one in France,¹⁶ six in Korea,^{20–25,27,30} two in Norway,^33,35 one in Spain,³² one in Taiwan,²⁸ two in the United Kingdom,^18,34 and one in the United States.¹⁷

Education Content and Method

In addition to general information on diabetes^{17,18,27,31,32,35} or weight reduction,^19,26 educational intervention and support in the studies were personalized to an individual care plan of the participants. Patients were requested to send information via voice mail or text message or at the Web site, which became the basis of personalized advice and support from the diabetes care team.^{16,17,19,26,27,29,32,33} Participants in diabetes studies, for example, were provided monitoring and advice based on individual blood sugar measurements, medications, insulin dose information, diet, weight, physical activities, and other information input by the participants as required by the study protocol. In providing personalized advice, health care providers also took into consideration patient history, such as age, years with diabetes, comorbidities, family history of diabetes, and laboratory results. Participants in studies focusing on weight reduction and management received personalized tips on diet and nutrition, how to overcome barriers, and how to maintain a regular physical activity program leading to weight reduction.^19,26

As part of the intervention, patients also received reminders to perform diabetes self-management activities (e.g., check their blood sugar and feet,^17,33 take medication,¹⁷ log-in and input information,^25,30,34 complete assessment questionnaires,^17,18,34 or review information on the next follow-up visit²⁸). In two studies, motivational messages were sent to the intervention group participants for maintaining their exercise schedule and overcoming barriers to maintaining a healthy life style.^19,26 The personalized messages were generated by algorithms based on data the patients input.

Technology of Intervention

Technology used to provide education and information interventions in all studies included cell phones and SMS. While the majority of studies utilized the text messaging feature of cell phones to provide tailored advice and support, in some studies,^22–26,30 patients had the option of using wireless or wired Internet to input their glucose measurements. One study sent reminders about the next appointment to the cell phone and via email.²⁸ Another study also utilized PDAs.¹⁶ Only one study performed all interactions with study participants through technology-based communication and did not provide any face-to-face meeting opportunity during the study duration.²⁶ Regardless of the mode of communication, all studies provided advice tailored to individual needs and the treatment regimen. Tables 3 and 4 provide the details of specific education and information provided.

Table 3.

Technology in Studies with Outcomes of Care

Author/year	Technology/functionality	Frequency	Educational content and delivery
Benhamou et al.,200716	Cell phone	Weekly	Participants performed weekly transfer of SMBG using a palm PDA that communicated with the glucometer and cell phone, and 3-monthly transfer of Diabetes Quality of Life survey responses plus any comments. They received weekly SMS treatment advice from their health care providers based on glucose values.
	SMS
	PDA
	Internet
Durso et al.,200317	Cell phone	Daily	Patients used a cell phone to input blood sugar, weight, any diabetic symptoms, medication side effects, or any problems with the care plan. They received daily personalized SMS messages from the nurse practitioner reinforcing knowledge and self-care behaviors and a phone call if any health problems needing immediate attention were identified from the data input.
	SMS
	Voice mail
Franklin et al.,200618	Cell phone	Daily	Goals set during clinic visits were reinforced by daily text messages from the Sweet Talk software system, containing personalized goal-specific prompts and messages tailored to patients' age, sex and insulin regimen.
	SMS
Hurling et al., 200719	Cell phone	Weekly	Internet, email, and mobile phone behavior change system, which included a Web-based interactive system for users to input perceived barriers, report exercise levels during the previous week, and plan physical activity for the next 7 days. Participants received personalized advice and motivational tips on how to overcome barriers and maintain an appropriate level of activity, as well as information about various types of physical activities. Charts were displayed with daily, weekly, and overall activity levels.
	Internet
	Voice mail
	email
Kim et al., 200520; Kim et al., 200622; Kim, 200724; Yoon and Kim, 200730	Cell phone	Weekly	Patient input their SMBG, medications, and insulin with doses, diet, and exercise level via cell phone, SMS, or the Internet and viewed their electronic chart, including lab results and nurse sent weekly recommendations (e.g., increase insulin by xx units, add another tablet of …, etc.) via cell phone, SMS, or the Internet. Also, patients received reminders if did not input information at least once a week.
	SMS
	Internet
Kubota et al., 200426	Cell phone	Daily	The i-exerM health education program was used to send participants informational message once a day to their mobile phone on weight reduction. They were asked to input their weight via the Internet from time to time. Information for each individual at the start and the end of the i-exerM monitoring session was collected with a questionnaire covering physical conditions, lifestyle, and program evaluation, without any face-to-face meeting during the study period.
	Internet
	email
Kwon et al., 200427	Cell phone	Variable	Participants sent their self-measured blood glucose, medication and its dosages, hypoglycemic events, amount of meals, and degree of exercise and questions. Their dietitians, endocrinologists, and nurses sent individualized recommendations for adjustment to dose, diet, and exercise for diabetes management based on data inputs.
	Internet
Liu et al., 200528	Cell phone	Following each visit throughout the study period	The POEM system was used to present on the Web each patient's education materials, medication data, and laboratory test results from each visit for patients to review. The system also sent reminders for the next follow-up via emails and short messages via cell phone.
	Internet
	email
Rami et al., 200629	Cell phone	Daily	Participants sent their data (date, time, blood glucose, carbohydrate intake, and insulin dosage, divided into short- and long-acting insulin) every time they measured a blood glucose value or at least once a day. For safety reasons, they continued their diary notes during the telemedical support phase. Patients received either an automatically generated SMS message if no treatment changes were needed or a personalized message with more specific advice on insulin dose adjustment.
	SMS

Notes: PDA, personal digital assistant; POEM, patient-oriented education management; SMBG, self-monitored blood glucose; SMS, short message service.

Table 4.

Technology in Studies with Process of Care

Author/year	Technology/functionality	Frequency	Educational content and delivery
Aokiet al., 200531	Cell phone	Not reported; camp activity	Cell phone-based game “INSULOT” was used to teach children during a summer camp for those with type 1 diabetes relationships among plasma glucose level, food (carbohydrate grams), and insulin dosage.
Ferrer-Roca et al., 200432	Cell phone	Variable	Patient input biological measures via SMS, with immediate display of patient data on the Web available for review to both patients and physicians. Patients received automated diabetes warning messages from the system based on data input and preset ranges by physicians. Patients accessed the Web site on the average every 2 days, while physicians reviewed patient data on average every 4 days.
	SMS
Gammon et al., 200533	Cell phone	Daily	Patients were to send their blood glucose readings at least three times per day, ask questions, and report technical difficulties. Both parents and patients received text messages with response to their questions, information, decision support, and social support.
	SMS
Tasker et al., 200734	Cell phone	Daily	The mobile group received a SMS message every day that asked to respond to questions if they had a hypoglycemic event that day; the computer-based interviewing group received an email to log in and complete the same questions on the Web site.
	SMS
	Internet
	email
Wangberg et al., 200635	Cell phone	Daily	Children sent their blood glucose data to parents' cell phones via SMS, which were sent to the electronic patient record. Diabetes educational messages were sent up to three times per day, on themes such as diabetes definitions, blood glucose, insulin, nutrition, physical activity, illness, and rights at school.
	SMS

Note: SMS, short message service.

Studies with Health Outcomes

Outcomes of care were defined as changes in diabetes-related health outcomes because of an educational or informational intervention delivered through cell phones using voice or short message service. We grouped the outcome measures according to the diabetes self-management education core outcome measures continuum: learning, behavior change, clinical improvement, and improved health status.³⁶ Table 1 lists studies that evaluated the impact of diabetes educational interventions on the control of blood sugar levels and other related health outcomes.

Learning

Two studies measured outcomes related to knowledge and skills.^17,19 These outcomes included average knowledge scores as measured by the Diabetes Knowledge Test from The Michigan Diabetes Research and Training Center,¹⁷ as well as perceived control, internal control, external control, and intention related to exercise.¹⁹

Behavior Change

Behavior change was measured in five of the studies.^16–19,22 These measures included general measures of self-efficacy and adherence to behavioral change,¹⁸ as well as self-management behaviors such as blood glucose monitoring,^16,17 taking medication,^17,22 eating appropriately,²² physical activity,^17,19,22 and foot care.^17,22

Clinical Improvement

Thirteen studies measured clinical changes.^16–30 The clinical measures were related to hemoglobin A1c (10 studies),^{16–18,21–25,27–30} blood glucose (7 studies),^{16,20,21,23–25,27,28,30} cholesterol (5 studies),^{20,21,23,24,27,28,30} weight (3 studies),^17,19,26 and blood pressure (1 study).¹⁹ Nine of the 10 studies that measured hemoglobin A1c observed a significant improvement. The results regarding blood glucose, cholesterol, and weight were mixed, while the results for blood pressure were nonsignificant.

Health Status

Two studies measured health status outcomes.^16,17 Quality of life was determined using the Diabetes Quality of Life questionnaire and satisfaction with life was also measured.¹⁶ Diabetes complications were measured as hypoglycemic symptoms^16,17 and hyperglycemic symptoms.¹⁷

Studies with Processes of Care

Table 2 lists studies that evaluated the impact of educational intervention on the process of diabetes care.^31–35 In addition, three studies that measured diabetes-related health outcomes also measured process of care.^17,26,28 The processes of care commonly measured included system convenience or ease of use,^{17,26,31,33,35} system usage,^28,32 satisfaction with the system,^31,32,34,35 time to use the system,¹⁷ or running costs.³² The processes of care measures were generally favorable to cell phone use. Unfortunately, the diversity of process measures used made it impossible to provide a summary of significant results from process measures beyond what is shown in Table 2.

Discussion

The purpose of this review was to evaluate the contribution of cell phone interventions in the control and management of diabetes. In this review, most studies provided standard diabetes care utilizing face-to-face communication during clinic visits. In addition, the intervention group participants received care through cell phones using one or more of their functional aspects (e.g., SMS, voice mail, Internet, or email). Cell phones were used as a management tool to enable information to flow between patients and providers. In a chronic disease such as diabetes, maintaining blood sugar levels and related clinical and physiological measurements to acceptable levels requires monitoring and management through regular self-care behaviors. In some studies, cell phones and text messaging facilitated regular treatment advice and support in between clinic visits. In other studies, cell phones and text messaging proved to be good tools to deliver regular alerts and reminders to achieve desired goals. Nine out of 10 studies that used cell phone technology and measured hemoglobin A1c showed significant decreases in hemoglobin A1c values. Results of our study indicate that educational interventions providing personalized advice and support delivered through a cell phone can help avoid diabetes symptoms by providing timely treatment adjustments and can lead to improved health outcomes.

The concept of using interactive voice mail messages to deliver information or education in disease management and control is not new.³⁷ Computer and communication technology-based education and support are becoming vital components of quality diabetes care.³⁸ Diabetes education and diabetes management support through automated telephone reminders and support have been shown to increase knowledge, increase frequency of self-care behaviors among persons with diabetes, and improve health outcomes for patients who need regular care and monitoring and self-care management.³⁹ Because monitoring and support from a health care provider play important roles in achieving the desired clinical goals, the use of cell phones, especially text messaging, is a step further in achieving the health and quality of life for persons with diabetes. Using cell phones and text messaging offers great opportunities to improve patient self-management by facilitating education, monitoring, and feedback between scheduled clinic visits. The ubiquitous nature of cell phones provides the mobility and flexibility so that care can be provided wherever a patient may be.⁴⁰ A potential implication of this is licensure issues if a health care provider is licensed in one state and the patient is receiving treatment in another state. As more and more people own cell phones, text messaging may even provide cost-effective alternatives to regular phone communication when combined with other methods of education and support.^41,42 Unfortunately, only one study in this review provided information on the costs of running the system.³²

Results of this systematic review should be interpreted with limitations in mind. The studies did not provide the reliability and validity of the information that patients entered into the cell phones; however, two studies provided examples of how a automated system could be programmed to check if the value entered was within a predefined range.^16,17 The heterogeneity of the studies prevented a meta-analysis, which could have allowed for a quantitative assessment. We attempted to search comprehensively; however, we may have unknowingly left out some work that was eligible for inclusion. Regardless, the benefits offered or limitations involved in the use of cell phones in diabetes care are estimated to be the same.

Abbreviations

MeSH

Medical Subject Headings

PDA

personal digital assistants

SMS

short message service