This is an open-access article distributed under the terms of the Creative Commons Attribution License (http://creativecommons.org/licenses/by/2.0/), which permits unrestricted use, distribution, and reproduction in any medium, provided the original work, first published in the Journal of Medical Internet Research, is properly cited. The complete bibliographic information, a link to the original publication on http://www.jmir.org/, as well as this copyright and license information must be included.
Information technology–based interventions are increasingly being used to manage health care. However, there is conflicting evidence regarding whether these interventions improve outcomes in people with type 2 diabetes.
The objective of this study was to conduct a systematic review and meta-analysis of clinical trials, assessing the impact of information technology on changes in the levels of hemoglobin A1c (HbA1c) and mapping the interventions with chronic care model (CCM) elements.
Electronic databases PubMed and EMBASE were searched to identify relevant studies that were published up until July 2016, a method that was supplemented by identifying articles from the references of the articles already selected using the electronic search tools. The study search and selection were performed by independent reviewers. Of the 1082 articles retrieved, 32 trials (focusing on a total of 40,454 patients) were included. A random-effects model was applied to estimate the pooled results.
Information technology–based interventions were associated with a statistically significant reduction in HbA1c levels (mean difference −0.33%, 95% CI −0.40 to −0.26,
Information technology strategies combined with the other elements of chronic care models are associated with improved glycemic control in people with diabetes. No clinically relevant impact was observed on low-density lipoprotein levels and blood pressure, but there was evidence that the cost of care was lower.
Chronic diseases such as diabetes can be managed better by implementing system-wide practices such as the chronic care model (CCM). This model identifies 6 components as essential for chronic disease management: health system organization, delivery system design, self-management support, community resources, decision support, and clinical information systems [
Several systematic reviews evaluated the potential benefits of information technology–based diabetes management interventions, and all concluded that information technology–based interventions could improve diabetes management for adult care [
A comprehensive literature search was conducted using PubMed and EMBASE for articles focusing on information technology–based diabetes interventions, which were published up until July 2016. A search strategy that combined keywords and Medical Subject Headings (MeSH) using the terms “diabetes,” “diabetes mellitus,” “non-insulin-dependent,” “diabetes type 2,” and “informatics” was used. In addition, international journals were searched manually and the reference lists from retrieved articles were reviewed in order to identify additional, relevant papers (
Titles and abstracts of all studies identified were independently reviewed by 2 reviewers (NSA and NA) from February to July 2016. Any discrepancies between the choices of the 2 reviewers were resolved by another reviewer (SDL). The inclusion and exclusion criteria for the study are presented in the
Search strategies.
Database | Search terms | Number of studies |
PubMed | 1: “Diabetes Mellitus”[Mesh] | 22,247 |
2: “Medical Informatics Applications”[Mesh] | 37,851 | |
1 and 2 | 425 | |
EMBASE | 2: 'diabetes'/exp AND 'mellitus'/exp | 537,195 |
1: 'information'/exp AND 'technology'/exp | 28,774 | |
1 and 2 | 557 |
The study design specifically evaluated the use of information technology–based interventions for the management of diabetes mellitus or T2DM, but the authors also included studies where information technology was part of a comprehensive intervention in which the impact of the information technology element was reported separately
The study focused on T2DM or both type 1 and type 2 diabetes mellitus, because T2DM accounts for more than 90% of all diabetes cases [
The study reported glycated hemoglobin (hemoglobin A1c or HbA1c) as an outcome measure
The study had one of the following study designs: randomized controlled trial, nonrandomized controlled trial, and before-after trial
Reviews lacking original study data
Studies that evaluated information technology–based interventions in other chronic diseases
Studies published in languages other than English or Arabic
Studies of children with diabetes, as very few have T2DM, or studies of pregnant women with gestational diabetes, as this is not T2DM (even though people with gestational diabetes are at an increased risk)
Papers using the same data as those already selected for use in the review
Two reviewers (NH-NS) independently reviewed the title, the abstract, and the article. Discrepancies were resolved by consensus or determined by other reviewers (SDL). Information was taken from each study using a predesigned collection form: authors, date of the study, technology type, country, study site, duration of the intervention, type of diabetes, study design, communication type, main user, number of participants, and outcome measures. Relevant missing data were obtained from authors. A qualitative review was performed to extract information about the clinical and process outcome measures: body weight, systolic blood pressure, diastolic blood pressure, low-density lipoprotein (LDL) cholesterol, high-density lipoprotein, process of care, cost of care, patients’ satisfaction, smoking levels, and medication adherence. As part of data collection, quality assessment for each included study was conducted using PRISMA (Preferred Reporting Items for Systematic Reviews and Meta-Analyses) guidelines [
Whether the study design was randomized
Whether the study described criteria for selection of participant
Whether both groups had similar baseline
Whether the study described the intervention methods
Whether the study evaluated the interventions after 6 months or more
Whether the study used intention-to-treat analysis
Whether the study reported method of blinding
The outcome measure was the changes in HbA1c levels from baseline to follow-up. HbA1c is recognized as a significant indicator of information technology–based intervention effectiveness in patients with T2DM because it reflects average glycemia over 8 weeks and is strongly associated with diabetes complications [
The data search produced 982 studies and a further 100 studies were identified by manual searching and from the references of included articles, giving a total of 1082 studies. A flow diagram of the search and selection process is shown in
All 32 studies selected for the review were published in English. Included studies had a total of 40,454 patients, more than half of them with both type 1 diabetes mellitus (T1DM) and T2DM, the others suffering from T2DM alone. Most of the included studies were conducted in the United States, while the 5 remaining studies were carried out in the United Kingdom [
Study selection process.
Four types of technological applications were identified as constituting the information technology–based intervention: electronic self-management system, electronic decision support system, diabetes registry, and EMRs. In some studies a combination of 2 technologies was identified. However, we categorized the types based on the main technology used in such cases.
Out of 32 articles, 18 used electronic self-management tools [
In this group, the best weighted mean change in HbA1c level, −1.86%, was reported in the study by Smith et al [
Out of 32 articles, 7 used a decision support system [
Diabetes registry was the primary intervention in 12% (4/32) of the included studies [
In one study, a pragmatic, cluster randomized controlled trial was conducted over a period of 15 months, with 3608 adult patients with T2DM, older than 35 years, and clients of 58 general practices from 3 localities in England. The intervention was a computerized diabetes register that incorporated the diabetes recall and management system. The registers were based on structured datasets completed on paper forms and laboratory reports. The results revealed that the intervention group demonstrated a decline in the mean level of HbA1c, down to 7.32%. In addition to the improvement of the clinical outcome, the study also demonstrated improvements in the clinical process, including foot examinations, 67.3% (
Among the studies, 2 randomized controlled trials did not show a significant improvement in the levels of HbA1c [
Only 3 out of 32 studies utilized EMR as the primary technological equipment [
This systematic review has identified 4 broad categories of T2DM management technologies. Electronic self-management technologies were a major component of studies targeting patients. These technologies may be placed broadly into 4 categories. The first category is the Web-based intervention that is based on interactive websites. Patients upload their data and receive feedback at a time most convenient for them and are not limited to clinic office hours [
EMRs and disease registries facilitate care providers to conduct clinical audits, provide them with reports for analyzing a patient’s key diabetes-related measures, and assist in tracking the patient’s progress. Registries are a central component of the CCM within both the public and private health sectors. Previous studies have suggested that their use correlates with improved outcomes for patients with diabetes [
The overall effect of different information technology–based interventions on the mean reduction in HbA1c level was 0.33% (95% CI −0.40 to −0.26,
The reduction in HbA1c values by the type of information technology–based intervention. EMR: electronic medical record. Horizontal lines: confidence intervals, squares: means, diamonds: pooled estimated measures. SMD: standardized mean difference.
This study reviewed clinical trials that assessed the effect of information technology on glycemic control of patients with T2DM. This systematic review (32 studies, 40,454 patients) shows that information technologies achieved a significant reduction in glycated hemoglobin in patients with T2DM. Significant positive effects on HbA1c levels were found in 30 studies. The subgroup analysis demonstrated that electronic self-management technology had the greatest impact on the health of patients with T2DM, while the diabetes registry had the least effect.
The impact of diabetes registries on improving care was difficult to quantify because the registries performed many different functions: it was unclear if the improvements had been driven by the functioning of the basic diabetes registry or other interventions. In the same way, being certain about the effectiveness of electronic health record systems is challenging because there cannot be a certain relationship with any presumed dependent variable; there is at best an association between technology use and quality and satisfaction [
These days, information technologies are advancing rapidly and are ubiquitously available worldwide. There is widespread belief that information technology may reduce care costs for patients with diabetes. However, relatively few studies have evaluated the effect of information technology on costs. The secondary outcome measures were summarized qualitatively because they were measured with various instruments. We found that a number of information technology studies reported improvements in the process of care and patient satisfaction, which suggests that information technology may be an effective strategy for changing patient behaviors. Additionally, our review demonstrates that there was no clinically relevant effect on LDL and no effect on blood pressure. This finding confirms those from a previous systematic review [
For diabetes care to be successfully supported by information technology–based interventions, their use should be embedded in the CCM. This review was able to map these technologies onto the CCM. It found that the most common CCM components used in trials besides the clinical information system were self-management support, delivery system design, and decision support. Health care organization and community resources were not reported. Most of the studies reported using multiple components in their interventions. It was difficult to determine which elements of the CCM benefit diabetic patients the most. However, interventions using self-management support reported the largest improvements in HbA1c levels. Four components of the CCM have a stronger effect on HbA1c levels than do 2 or 3 elements.
Several systematic reviews related to health information technology have been undertaken, but they have limited their scope to specific systems such as telemedicine [
There is evidence to suggest that electronic self-management systems may improve glycemic control in patients with T2DM: this meta-analysis indicated that this type of technology significantly reduced HbA1c levels compared with the control group (pooled mean difference 0.50%,
This review and meta-analysis has several advantages over most, previous systematic reviews of the impact of information technology on diabetes care. We reviewed a large body of literature, assessed the quality of included trials, and contacted authors of some studies to collect missing data. To our knowledge, this systematic review presents the first pooled analysis results of varied information technology types on HbA1c levels among patients with T2DM. Nevertheless, this review also has limitations. We used HbA1c level as the primary outcome measure because of its long-established association with adverse cardiovascular outcomes in diabetes [
The findings of this review suggest that, in general, information technology interventions improve glycemic control. Patient self-management support appears most promising; EMRs and clinical decision support system appear to confer benefits, but disease registries by themselves do not appear to improve quality. In addition, the results conform to presumptions surrounding the CCM that changes must be made in multiple areas in order to considerably improve the outcomes of diabetes care. However, further investigation is still required to increase our understanding of how, why, and when information technology can improve the care of patients with T2DM. This includes a cost-benefit analysis of using information technology and the other secondary outcomes.
Summary of information technology–based interventions for type 2 diabetes.
chronic care model
electronic medical record
hemoglobin A1c
low-density lipoprotein
Preferred Reporting Items for Systematic Reviews and Meta-Analyses
type 1 diabetes mellitus
type 2 diabetes mellitus
This work was part of a PhD dissertation funded by King Saud University.
None declared.