Sage Journals: Discover world-class research

Abstract

Background:

This study aims to provide a better understanding of the ability of mobile health tools to offer glycemic control for patients with type 1 diabetes mellitus.

Methods:

Data gained from research articles searched in PubMed, Ovid (Medline), and CINAHL from 2005 to 2013 focused on interventions introduced to a type 1 diabetic population. Articles were screened to identify interventions that examined mobile health tools effect on glycemic control using %A1C as a proxy. Fourteen articles were included in this study. Descriptive data, %A1C difference, and statistical significance, if available, were extracted for comparison.

Results:

Five major categories were identified across the spectrum of interventions, including “Internet,” “Mobile,” “Mobile and Internet,” “Phone,” and “Videoconference and phone.” Seven of the 14 articles reported statistically significant decreases in measured outcomes. Seven studies examine a single cohort, and 7 examined a double cohort. Eleven of the 14 authors (79%) reported success with their intervention. Twelve studies reported a decrease in %A1C values in their intervention groups.

Conclusions:

Initial results for glycemic control through these tools appear promising, though inconclusive. Additional measures of mobile health tool efficacy should be assessed more directly. More rigorous study methods are also needed to improve the reliability of results.

Keywords

blood glucose mobile health type 1 diabetes mellitus self-management self-monitoring of blood glucose telehealth

Diabetes is a chronic condition that continues to permeate the health care atmosphere as one of critical importance. Management of this condition and more specifically type 1 diabetes mellitus (T1DM) has drawn considerable attention as health care systems begin to renew their focus on costs and quality improvement with the ongoing implementation of the new health care law.^1,2 At the same time telehealth, mobile health, and similar movements have all appeared in the health care sphere with increasing popularity as potential tools to aid persons with diabetes in the managing of their condition while lowering costs.³ The rationale underlying the use of these tools is three-fold. First, these tools allow for enhanced communication with a provider. Second, patients are empowered to independently assess and control their disease state.⁴ Finally, cost savings round out these benefits, demonstrating mobile health tools as one of the key technological instruments that attack chronic conditions in way that is affordable, effective, and accessible.⁵ Broadly defined, these tools can be thought of as any use of mobile communications for health information and services.⁶

Unfortunately, use of these tools as interventions in the disease states of person with type 1 diabetic persons has led to mixed results even as the incidence of T1DM in the United States remains among the highest in the world.^7-10 Further studies specifically focusing on mobile tools impact on metabolic control outcomes have produced mixed results.^11,12

T1DM merits particular attention, since it affects certain populations in much greater proportions than others. Several authors have documented the difficulty for pediatric populations to manage T1DM,^13-15 while others have noted the unique challenges facing pregnant women diagnosed with T1DM.^16,17 There is then a need to focus exclusively on the ability of these new digital tools to impact the aforementioned and general adult populations’ ability to manage T1DM with the aid of these tools. The objective then of this study was to review recent literature documenting the ability of mobile health tools to impact the glycemic control of type 1 diabetic patients.

Methods

A search of PubMed (2005-2013), CINAHL (2005-2013), and Ovid Medline (2005-2013) for research articles was conducted in October 2013 using the following search terms: self-management (Medical Subject Headings [MeSH]) and mobile health (MeSH) and type 1 diabetes mellitus (MeSH). These search terms and filters were used to find the most contemporarily relevant articles that embraced all methods of employing mobile health tools.

Inclusion and Exclusion Criteria

Inclusion criteria were any research article evaluating a diabetes self-care management technology that could be accessed in a mobile manner with a measured %A1C outcome for populations of persons diagnosed with T1DM. Studies were excluded if they did not have data on a patient population exclusively consisting of persons with T1DM, or which did not report outcomes including %A1C as a proxy for glycemic control. Previous work has identified %A1C as a viable proxy for glycemic variability, and this was continued here.¹⁸

Study Selection and Data Extraction

The titles, abstracts, results, and citations were reviewed and screened based on the inclusion and exclusion criteria described earlier. Relevant data were individually collected from each screened article including patient sample, patient sample characteristics, intervention, intervention type, and length of intervention. The %A1C difference between start and end of study, descriptive statistics and statistical significance were also collected. The patient population characteristic data were sorted and then arrayed into a table, as was the intervention type. Data were analyzed for patterns based on differences in the intervention type used and the patient population measured. These population subgroups were analyzed in accordance with the original article to retain the intent and analysis of the originating data. For this reason, pregnant and pediatric populations were analyzed separately as well as in conjunction with the general adult cohort studies.

Categorical data concerning type of intervention were separated into categories, “Mobile,” “Mobile and Internet,” “Internet,” and “Videoconferencing and phone,” as was appropriate based on the description of the intervention. A second interventional tag described the service offered by the device; these included “message service,” “diary,” and “phone service.”

Population samples were adjusted based on the author’s description of those patients who fully participated in the study, for the entire duration of the study. For example, if the author stated that 30 patients were included at the start of the study but 5 patients dropped out during the study, the sample population (n) would be listed as 25. The mean duration of the intervention was recorded if given and rounded appropriately to give an integer.

Results

Comprehensive literature searches identified 54 articles (Figure 1). The titles and abstracts of these articles were read and 14 articles were ultimately determined to be potentially relevant. The distribution of articles included a predominance of cohort studies, with 3 prospective, or pilot studies, 1 randomized controlled trial, and 1 case study (Table 1). Six studies^7-10,19-24 focused on a pediatric population and 1 study focused on a population of pregnant women.²⁵ The median study duration was 6 months (range, 1-9 months). It is worth noting that only 7 of these studies employed a methodology that compared an interventional group to a control group (Table 1). By each article’s author’s own determination of intervention success, 11 of the 14 authors (79%) considered their interventions successful. The studies deemed unsuccessful by the author’s determination were for failing to produce a statistically significant effect or the desired effect at all.^21,26,27 Seven of the 14 studies reported statistically significant declining %A1C values.^{19,20,23,25,26,28,29}

Figure 1.

Search strategy flow diagram.

Table 1.

Types of Articles Included in Review, Including # of 2-Cohort Studies.

Study type	#	%	# 2-cohort studies
Case study	1	7	0
Prospective cohort	9	64	6
Pilot study	3	21	0
Randomized controlled trial	1	7	1

Five intervention approaches were observed in these studies: Internet based,^23,26,30 Mobile based,^25,28,31,32 Mobile and Internet based,^{20,22,27,33,34} a telephone service,^21,24 and an optional videoconference and phone service (Table 2).¹⁹ The type of service offered further categorized these interventions into 3 broad categories of “message service,” “phone service,” and “diary” (Table 2). The videoconference with phone service article was excluded from this second categorization since its service is self-descriptive.

Table 2.

Intervention Descriptions, Subdescriptions, and Distribution.

Intervention tag	#	%
Internet	3	21
Diary	1	7
Message service	2	14
Mobile	4	29
Diary	1	7
Message service	3	21
Mobile and Internet	4	29
Diary	1	7
Message service	2	14
Phone service	1	7
Phone (phone service)	2	14
Videoconference and phone	1	7

Single-Cohort Studies With %A1C Outcome

There were 127 subjects involved in single-cohort studies using %A1C as an outcome. Among these 7 articles, 4 reported statistically significant results. All of these reported a decrease in the mean %A1C value (Table 3). The highest of these changes was −3.2%, the lowest was −0.33%, and the median was −0.82%.

Table 3.

Single Cohort Studies With %A1C Change and Population Characteristics.

Single-cohort studies with %A1C change and population characteristics	Intervention description	N	Mean duration of diabetes (years)	Length of intervention (months)	Intervention difference in mean %A1C	P value
Adult
Everett et al³⁰	Internet	16	30	6	−1.10	N/R
Gonzalez-Molero et al²⁸	Mobile	15	21.9	6	−0.53	.011
Ładyżyński et al²⁹	Mobile	13	8.7	1	−3.20	.0002
Rossi et al³³	M & I	41	9.5	N/R	−0.33	.27
Pediatric
Adkins et al²⁴	Phone	2	4	3	−2.40	N/R
Peña et al²⁰	M & I	13	3.8	3	−0.46	.012
Heidgerken et al¹⁹	Videoconference & phone	27	14	6	−0.07	S

M & I, mobile and Internet; N/R, not reported; S, significant.

Two-Cohort Studies With %A1C Outcome

Among the comparative studies reporting %A1C values there were 295 subjects involved. There was no study population that had more than 50 people in either the control or the interventional group. Among these, only 3 reported a statistically significant finding for comparing the 2 groups %A1C differences. It is also worth noting that only 1 of all these studies had a randomized subject base.²⁷ Amongst the control population, the range of %A1C difference from start to end was −5.10% to 0.98%, with a median value of −0.04%. The intervention groups’ range of %A1C difference values is −1.40% to 0.40%, with a median value of −0.68% (Table 4).

Table 4.

Two-Cohort Studies With %A1C Change, Population Characteristics, and P Values.

	Control population						Intervention population
Study	Intervention	Study length (months)	N	Mean duration of diabetes (years)	Difference in %A1C	1-sample P value	N	Mean duration of diabetes (years)	Difference in mean %A1C	1-sample P value	2-samples P value
Adult
Bujnowska-Fedak²⁶	Internet	6	15	N/R	−5.10	N/R	15	N/R	−1.20	N/R	S
Farmer et al^27 a	M & I	9	46	13.3	−0.34	<.04	47	11.6	−0.62	<.001	.33
Skrøvseth et al³¹	Mobile	3	12	32.9	0.02	N/S	18	43.2	0.00	N/S	N/S
Pediatric
Lehmkuhl et al²¹	Phone	3	14	N/R	−0.09	N/R	18	N/R	−0.74	N/R	N/S
Mulvaney et al²²	M & I	3	23	15.9	0.98	N/R	23	7.9	−0.10	N/R	.42
Pinsker et al^23 b	Internet	6	7	N/R	0.01	.013	9	N/R	−1.40	.06	.03
Pinsker et al^23 c	Internet	6	7	N/R	0.10	.01	9	N/R	0.40	.14	.45
Pregnant
Dalfra et al²⁵	Mobile	9	15	17.4	−0.60	N/R	17	16.1	−0.80	N/R	.008

M & I, mobile and Internet; N/R, not reported; N/S, reported as not significant; S, significant.

Randomized subjects.

Cohort of patients within ADA age-recommended guidelines.

Cohort of patients outside ADA age-recommended guidelines.

Pediatric Population Subgroup Analysis

There were several studies that analyzed an exclusively pediatric population.^19-24 Three of these studies found a statistically significant decrease.^19,20,23 The range of %A1C difference in the pediatric intervention populations was −2.80% to 0.40%, with a median of −0.60%.

Pregnant Population Subgroup Analysis

One study examined the effect of a mobile health management intervention in a population of type 1 diabetic pregnant women.²⁵ Using a mobile intervention, Dalfra et al’s²⁵ intervention population experienced a statistically significant 0.8% decrease in %A1C from start to end of the study, compared to the control group’s 0.6% decrease in %A1C (Table 4).

Discussion

This systematic review focused on the ability of mobile health tools to grant patients with T1DM greater glycemic control. These tools took a variety of forms, and provided a number of different services to aid diabetic patients. The majority of authors determined their intervention to be successful. This coincides with the majority of studies that demonstrated a decrease in %A1C. However, not all of these values were statistically significant. In addition, the predominant use of prospective studies in place of randomized trials fails to demonstrate with any meaningful certainty the efficacy of these tools. It is worth noting that the overall impression of these tools’ ability to aid diabetic patients with their conditions remains positive.

While certain studies focused on the effect of these interventions to aid specific subpopulations within the type 1 diabetic population, very little evidence of success was found. Thus, the muddied picture of how these tools significantly affect persons with T1DM is not made any clearer by examining 1 subpopulation in comparison to another. Further work in this area is needed to remedy this.

Certain limitations must be kept in mind when considering this study. First, a concerning number of results with nonsignificant P values^{21-23,27,31,33} and P values that weren’t reported^{21,22,24,26,30} suggests a need for stronger study planning and more randomized controlled trials. This is especially important, as a concern over reproducibility in studies has begun to draw attention to the statistical significance standards, and the need for stricter statistical measuring.³⁵ This should be a prime focus for further research as it will add greater weight to study results, and more precisely measure these tools’ efficacy. In addition, the heterogeneity of studies, methods, interventions, and populations prevented a meta-analysis of these data, which would have allowed for further interpretation of results. Finally, it should be noted that a single author conducted this review, which may allow for a greater potential of missed studies and errors in transcription.

Looking ahead, there is significant space for research of increased breadth and depth in this area. A cursory glance at the “health” section of Apple’s App Store or the shelves of a pharmacy store reveals a plethora of mobile applications and tools designed to aid persons with diabetes in the management of their condition. Yet there is little evidence to guide physicians or consumers on which tools are best. Furthermore, while existing systematic reviews have addressed the impact of mobile health tools and telemedicine on T1DM in addition to type 2 diabetes mellitus,¹¹ as well as the advantages of monitoring and communicating with patients remotely,³⁶ there is still a need for focused research exclusively on T1DM that assesses glycemic outcomes. Further research must establish this evidence and provide transparency for consumers and providers interested in these mobile health tools so a better understanding of their efficacy may be established.

Footnotes

Acknowledgements

The author wishes to thank Dr. Suzanne Boren for her advice and guidance when constructing this review.

Abbreviations

ADA, American Diabetes Association; MeSH, Medical Subject Headings, T1DM, type 1 diabetes mellitus.

Declaration of Conflicting Interests

The author declared no potential conflicts of interest with respect to the research, authorship, and/or publication of this article.

Funding

The author received no financial support for the research, authorship, and/or publication of this article.

References

Koh

Sebelius

. Promoting prevention through the Affordable Care Act. N Engl J Med. 2010;363(14):1296-1299.

Manchikanti

Caraway

Parr

Fellows

Hirsch

. Patient Protection and Affordable Care Act of 2010: reforming the health care reform for the new decade. Pain Physician. 2011;14(1):E35-E67.

Biermann

Dietrich

Rihl

Standl

. Are there time and cost savings by using telemanagement for patients on intensified insulin therapy? A randomised, controlled trial. Comput Methods Programs Biomed. 2002;69(2):137-146.

McClain

Thompson

. The use of cell phone technology provides teens more control and independence and healthcare cost savings in the management of chronic disease. Perspect Health Inf Manag. 2010;7:1g.

Costa

Fitzgerald

Jones

Dunning

AM T

. Effectiveness of IT-based diabetes management interventions: a review of the literature. BMC Fam Pract. 2009;10(1):72.

Nacinovich

. Defining mHealth. Comm Health. 2011;4(1):1-3.

Wootton

. Recent advances: telemedicine. BMJ. 2001;323(7312):557-560.

Cartwright

Hirani

Rixon

. Effect of telehealth on quality of life and psychological outcomes over 12 months (Whole Systems Demonstrator telehealth questionnaire study): nested study of patient reported outcomes in a pragmatic, cluster randomised controlled trial. BMJ. 2013;346(2):f653.

Peter

. Trends in the incidence of type I diabetes mellitus worldwide. West Indian Med J. 2007;56(3):264-269.

10.

Soltesz

Patterson

Dahlquist

, EURODIAB Study Group. Worldwide childhood type 1 diabetes incidence—what can we learn from epidemiology? Pediatr Diabetes. 2007;8(suppl 6):6-14.

11.

Farmer

Gibson

Tarassenko

Neil

. A systematic review of telemedicine interventions to support blood glucose self-monitoring in diabetes. Diabet Med. 2005;22(10):1372-1378.

12.

Scuffham

. Systematic review of cost effectiveness in telemedicine. Quality of cost effectiveness studies in systematic reviews is problematic. BMJ. 2002;325(7364):598.

13.

Leonard

Garwick

Adwan

. Adolescents’ perceptions of parental roles and involvement in diabetes management. J Pediatr Nurs. 2005;20(6):405-414.

14.

Hanna

Guthrie

. Adolescents’ perceived benefits and barriers related to diabetes self-management—part 1. Issues Compr Pediatr Nurs. 2000;23(3):165-174.

15.

Leonard

Jang

Y-P

Savik

Plumbo

Christensen

. Psychosocial factors associated with levels of metabolic control in youth with type 1 diabetes. J Pediatr Nurs. 2002;17(1):28-37.

16.

Rasmussen

Dunning

Hendrieckx

Botti

Speight

. Transition to motherhood in type 1 diabetes: design of the pregnancy and postnatal well-being in transition questionnaires. BMC Pregnancy Childbirth. 2013;13:54.

17.

Anderson

Wolpert

. A developmental perspective on the challenges of diabetes education and care during the young adult period. Patient Educ Couns. 2004;53(3):347-352.

18.

Sacks

Bruns

Goldstein

MacLaren

McDonald

Parrott

. Guidelines and recommendations for laboratory analysis in the diagnosis and management of diabetes mellitus. Diabetes Care. 2002;25(4):750-786.

19.

Heidgerken

Storch

. Telehealth intervention for adolescents with type I diabetes. J Pediatr. 2006;148(5):706-707.

20.

Peña

Torres

Cardona

JAC

Iniesta

. Impact of telemedicine assessment on glycemic variability in children with type 1 diabetes mellitus. Diabetes Technol Ther. 2013;15(2):136-142.

21.

Lehmkuhl

Storch

Cammarata

. Telehealth behavior therapy for the management of type 1 diabetes in adolescents. J Diabetes Sci Technol. 2010;4(1):199-208.

22.

Mulvaney

Anders

Smith

Pittel

Johnson

. A pilot test of a tailored mobile and web-based diabetes messaging system for adolescents. J Telemed Telecare. 2012;18(2):115-118.

23.

Pinsker

Nguyen

Young

Fredericks

Chan

. A pilot project for improving paediatric diabetes outcomes using a website: the Pediatric Diabetes Education Portal. J Telemed Telecare. 2011;17(5):226-230.

24.

Adkins

Storch

Lewin

. Home-based behavioral health intervention: use of a telehealth model to address poor adherence to type-1 diabetes medical regimens. Telemed J E Health. 2006;12(3):370-372.

25.

Dalfra

Nicolucci

Lapolla

, on behalf of the TISG. The effect of telemedicine on outcome and quality of life in pregnant women with diabetes. J Telemed Telecare. 2009;15(5):238-242.

26.

Bujnowska-Fedak

Puchała

Steciwko

. Telemedicine for diabetes support in family doctors’ practices: a pilot project. J Telemed Telecare. 2006;12(5):8-10.

27.

Farmer

Gibson

Dudley

. A randomized controlled trial of the effect of real-time telemedicine support on glycemic control in young adults with type 1 diabetes (ISRCTN 46889446). Diabetes Care. 2005;28(11):2697-2702.

28.

Gonzalez-Molero

Dominguez-Lopez

Guerrero

. Use of telemedicine in subjects with type 1 diabetes equipped with an insulin pump and real-time continuous glucose monitoring. J Telemed Telecare. 2012;18(6):328-332.

29.

Ładyżyński

Wójcicki

. Home telecare during intensive insulin treatment—metabolic control does not improve as much as expected. J Telemed Telecare. 2007;13(1):44-47.

30.

Everett

Kerr

. Telehealth as adjunctive therapy in insulin pump treated patients: a pilot study. Practical Diabetes Int. 2010;27(1):9-10i.

31.

Skrøvseth

Årsand

Godtliebsen

Hartvigsen

. Mobile phone-based pattern recognition and data analysis for patients with type 1 diabetes. Diabetes Technol Ther. 2012;14(12):1098-1104.

32.

Ładyzyński

Wojcicki

Krzymień

. Mobile telecare system for intensive insulin treatment and patient education. First applications for newly diagnosed type 1 diabetic patients. Int J Artif Organs. 2006;29(11):1074-1081.

33.

Rossi

Lipworth

Polesel

. Dietary glycemic index and glycemic load and risk of pancreatic cancer: a case-control study. Ann Epidemiol. 2010;20(6):460-465.

34.

Martínez-Sarriegui

García-Sáez

Rigla

. How continuous monitoring changes the interaction of patients with a mobile telemedicine system. J Diabetes Sci Technol. 2011;5(1):5-12.

35.

Johnson

. Revised standards for statistical evidence [published online ahead of print November 11, 2013]. Proc Natl Acad Sci USA.

36.

Krishna

Balas

Boren

Maglaveras

. Patient acceptance of educational voice messages: a review of controlled clinical studies. Methods Arch. 2002;41(5):360-369.

Improving Type 1 Diabetes Management With Mobile Tools

Abstract

Background:

Methods:

Results:

Conclusions:

Keywords

Methods

Inclusion and Exclusion Criteria

Study Selection and Data Extraction

Results

Single-Cohort Studies With %A1C Outcome

Two-Cohort Studies With %A1C Outcome

Pediatric Population Subgroup Analysis

Pregnant Population Subgroup Analysis

Discussion

Footnotes

Acknowledgements

Abbreviations

Declaration of Conflicting Interests

Funding

References