**5. Research limitations and implications**

The current research on the effect of HITs in diabetes management has several limitations. First of all, the published trials often do not provide protocols for studies [45]. There is also lack of information on the theoretical bases of the

**143**

**forward**

*Health Information Technologies in Diabetes Management*

interventions, and whether the HIT interventions are accompanied by other pharmaceutical or lifestyle therapies in their publications. As these HIT interventions are main therapeutic agents, it would be beneficial to explicitly prescribe interventions for trials and state the active components (behavior-change techniques), dose (frequency and intensity of interactions), route (mode of delivery), and duration of treatment [45]. There is also a need to clarify other ingredients in the intervention such as medication, standard care from health professionals, so that the major role of the HITs to the effectiveness of the interventions can be estimated, separating the

Additionally, intervention periods in published trials are short (most trials under 1 year) [33] and few systematic reviews provided effect estimation by length of follow-up. Studies by Tao et al. and Heitkemper et al. showed that HITs' effect on glycemic control was diminishing as the interventions proceeded [18, 21]. It is not clear whether intervention effect and compliance with the HIT interventions would sustain in the long term. Misuse or nonuse of technological support is a common problem in disease management, which greatly affects patient's outcomes. There is also lack of focus on cardiovascular health assessments in HIT interventions for diabetes management. We only found two systematic reviews that discussed CVD outcomes in addition to glycemic control. Because very few trials included cardiovascular risk factor evaluations, the synthesized findings were modest (**Table 1**). As we discussed earlier, because CVD causes major morbidity and mortality among T2D patients, designing and evaluating HITs for diabetes management should include cardiovascular health indicators. Further, many review studies only reported standardized difference in means [18, 21], which may be less intuitive to patients who care the absolute changes (i.e., mean difference) in outcomes (e.g., HbA1c) due to an intervention. Moreover, it remains unclear whether there are harms associated with the intervention. It has been reported that people may suffer from negative consequences of excessive self-monitoring by finding it uncomfortable, intrusive, and unpleasant [46, 47]. Studies found patients with diabetes who self-monitor their own blood glucose concentration did not benefit from increased glycemic control but rather found their disease more intrusive [48]. The interaction between a HIT device and a patient can be complex, and further studies need to consider these in more detail. Further, whether the interventions would be cost-effective if it required significant health professional support in a long-run has not been documented well in the literature [33, 49]. Additional research with more time points of follow-up is warranted to maximize data to inform the compliance with the HITs, long-term impact on health outcomes, to look for evidence of harms and to determine the cost-effectiveness in the intervention [49]. Studies with CVD risk factor assessments and absolute outcome measurement are also needed. Moreover, it is unknown which populations will benefit the most from the HIT intervention as the current research in HITs has not always directly engaged diverse end users. There are also many questions surrounding the "digital divide" in HITs use, where the access, usability, and effectiveness of diabetes technologies are divided by users' age, education, computer literacy, culture, and affluence [49]. These issues highlight the importance of engaging more research to design, test,

*DOI: http://dx.doi.org/10.5772/intechopen.83693*

effects from usual care and treatment [33].

and implement HITs for diverse patients with diabetes.

**6. Barriers of using HITs in the real-world context and steps to move** 

While features of HITs can expand patients' ability in diabetes management and the results from the existing research showed their positive effects on outcomes of

*Type 2 Diabetes - From Pathophysiology to Modern Management*

of a pharmacological therapy or of behavioral interventions.

management among patients with T2D.

**5. Research limitations and implications**

usual care [33]. This result suggests that HITs are the key to the effectiveness rather than tools or components of these trials. Additionally, pharmacotherapies often use motivated patients' sample and they cannot generate their full effects without patients' adherence to treatment and persistence in usage [33]. In this sense, HITs may add additional value in the effectiveness by addressing challenges in adherence

**4. HITs in managing cardiovascular risks among patients with T2D**

diabetes have a 77–87% prevalence of hypertension, a 74–81% prevalence of elevated low-density lipoprotein cholesterol (LDL), and a 62–67% prevalence of obesity [37]. Cardiovascular disease (CVD) is recognized as the most frequent cause of morbidity and mortality in patients with diabetes, causing up to 70% of all deaths in this patient group [2]. Type 2 diabetes (T2D) confers an approximate twofold elevation of CVD risk, equivalent to that of a previous myocardial infarction [3, 38]. In light of CVD burden in those with diabetes, the management of modifiable CVD risk factors, including hypertension, dyslipidemia, and obesity, is critical to minimizing the risk of macrovascular complications as well as death of diabetes. Yet, the implementation of preventive strategies to CVD among individuals with T2D is often not adequate [39–41] and less than half of patients who visit their care provider meet recommended levels for blood pressure (BP) and lipids [42]. Innovative approaches such as HITs are needed to facilitate CVD risk factor

T2D is commonly accompanied by cardiovascular complications. Adults with

In the context of cardiovascular care among general populations, HITs were documented to offer numerous benefits and have been associated with improvements in the measurement and monitoring of heart health, including risk factors such as BP, arrhythmia, cholesterol, and weight, as well as the implementation of guideline-based decision support for providers [43]. However, CVD outcomes are usually secondary and less described compared to glycemic status in T2D management trials [26, 44]. Furthermore, many review studies examining HITs' effect in diabetes management often overlooked CVD outcomes [26, 44] or include insufficient sample size or limited CVD parameters for analysis [22, 24]. In the study by Marcolino et al., only 13 studies were included in the final analysis, within which 8 studies assessed the effect on SBP, 7 on DBP, and 5 on LDL [22]. No effects of telecommunication and information technologies were seen on SBP and DBP. They did, however, find a statistically significant reduction on LDL (−6.6 mg/dL, 95% CI −8.3, −4.9 mg/dL) associated with the technologies evaluated. They were not able to perform analysis on weight outcome, because only two studies assessed the effect of HITs on weight and both studies demonstrated a nonsignificant reduction on weight. In the systematic review by Pal et al., among 11 RCTs included in their final analysis, 5 studies looked into changes in BP (only 1 showed improvement in BP), 7 reported changes in BMI or weight (5 were combined in a meta-analysis), and 10 measured serum lipids (7 were combined in a meta-analysis) [24]. The overall pooled effect did not reach statistical significance for all of these outcomes [24].

The current research on the effect of HITs in diabetes management has several limitations. First of all, the published trials often do not provide protocols for studies [45]. There is also lack of information on the theoretical bases of the

**142**

interventions, and whether the HIT interventions are accompanied by other pharmaceutical or lifestyle therapies in their publications. As these HIT interventions are main therapeutic agents, it would be beneficial to explicitly prescribe interventions for trials and state the active components (behavior-change techniques), dose (frequency and intensity of interactions), route (mode of delivery), and duration of treatment [45]. There is also a need to clarify other ingredients in the intervention such as medication, standard care from health professionals, so that the major role of the HITs to the effectiveness of the interventions can be estimated, separating the effects from usual care and treatment [33].

Additionally, intervention periods in published trials are short (most trials under 1 year) [33] and few systematic reviews provided effect estimation by length of follow-up. Studies by Tao et al. and Heitkemper et al. showed that HITs' effect on glycemic control was diminishing as the interventions proceeded [18, 21]. It is not clear whether intervention effect and compliance with the HIT interventions would sustain in the long term. Misuse or nonuse of technological support is a common problem in disease management, which greatly affects patient's outcomes. There is also lack of focus on cardiovascular health assessments in HIT interventions for diabetes management. We only found two systematic reviews that discussed CVD outcomes in addition to glycemic control. Because very few trials included cardiovascular risk factor evaluations, the synthesized findings were modest (**Table 1**). As we discussed earlier, because CVD causes major morbidity and mortality among T2D patients, designing and evaluating HITs for diabetes management should include cardiovascular health indicators. Further, many review studies only reported standardized difference in means [18, 21], which may be less intuitive to patients who care the absolute changes (i.e., mean difference) in outcomes (e.g., HbA1c) due to an intervention. Moreover, it remains unclear whether there are harms associated with the intervention. It has been reported that people may suffer from negative consequences of excessive self-monitoring by finding it uncomfortable, intrusive, and unpleasant [46, 47]. Studies found patients with diabetes who self-monitor their own blood glucose concentration did not benefit from increased glycemic control but rather found their disease more intrusive [48]. The interaction between a HIT device and a patient can be complex, and further studies need to consider these in more detail. Further, whether the interventions would be cost-effective if it required significant health professional support in a long-run has not been documented well in the literature [33, 49]. Additional research with more time points of follow-up is warranted to maximize data to inform the compliance with the HITs, long-term impact on health outcomes, to look for evidence of harms and to determine the cost-effectiveness in the intervention [49]. Studies with CVD risk factor assessments and absolute outcome measurement are also needed.

Moreover, it is unknown which populations will benefit the most from the HIT intervention as the current research in HITs has not always directly engaged diverse end users. There are also many questions surrounding the "digital divide" in HITs use, where the access, usability, and effectiveness of diabetes technologies are divided by users' age, education, computer literacy, culture, and affluence [49]. These issues highlight the importance of engaging more research to design, test, and implement HITs for diverse patients with diabetes.
