Diabetes mellitus (DM) is one of the most common chronic and costly conditions in the world, which is associated with serious comorbidities. Strict glycemic control has been shown to reduce the risk of micro- and macrovascular complications [1–3]. However, despite improved understanding of the disease and a variety of new medications and technologies, many patients still fail to achieve treatment targets and remain at risk of these complications [2, 4, 5]. Particularly in Brazil, a recent cross-sectional study including 5,692 outpatients with type 2 diabetes (T2DM) showed that the rate of poor glycemic control was 73% (90% among insulin-treated subjects) [1].

As a typical chronic condition, DM requires continuous care and successful management of this disease cannot be achieved without deep patient involvement [2]. In this respect, the effect of educational interventions on the acquisition of self-care behaviors is particularly important for improving glycemic control because self-management provides a report of current treatment status, with immediate therapeutic benefits and patient empowerment [6]. However, educational interventions that are brief, infrequent, or designed solely to increase patient knowledge are unlikely to improve self-care or glycemic control as desired [7].

Self-monitoring of blood glucose (SMBG) has been recommended as a useful tool for improving glycemic control and is considered an essential component in treatment programs for patients with insulin-treated DM, favoring dietary changes, physical activity and pharmacological therapy, including titration of insulin doses [1, 4, 8]. Nevertheless, despite the availability of the method and of information, SMBG is still an underutilized strategy and most diabetic patients are unaware of the actions that must be taken in response to its results and do not adjust their treatment [9–13].

Currently, most patients with T2DM requiring insulin therapy have their doses titrated by their clinicians, which is a time-consuming process. Evidence suggests that this approach may not provide optimal glycemic management for the patients [14]. The association between SMBG and glycemic control could be strengthened as healthcare professionals improve their ability to teach patients self-management skills, to instill greater awareness of this importance, to enhance their self-confidence, and to motivate them to make behavioral changes in response to readings [15]. Patients need to understand why they are being asked to self-monitor, what their glycemic targets are, and what attitudes they should take based on the results of SMBG [4].

The aim of the present study was to evaluate the effectiveness and safety of an educational program for self-adjustment of insulin doses associated with structured SMBG for the improvement of glycemic control in poorly controlled insulin-treated patients with T2DM at a secondary care unit in Brazil.

The increasing prevalence of T2DM, together with the world’s ageing population, places an increasing burden on healthcare systems, particularly healthcare professionals [14, 16]. Thus, diabetes self-management educational programs have been considered by some authors as an essential strategy for improving the health behaviors of diabetic adults [1, 14, 17].

The success of long-term management of insulin-requiring patients with T2DM is the result of a complex interaction of different factors, including the mode of insulin and diet therapy, individual motivation and self-care behavior, and the patients’ knowledge and skills regarding the treatment of their illness [16]. It has been suggested that different or better implementation of existing approaches is needed to help patients understand and achieve glycemic targets in order to improve glycemic control and to prevent or delay the complications of DM [11].

Treatment guidelines by global organizations recommend insulin intensification to achieve A1C targets as T2DM progresses, but fewer patients are being progressed than would be indicated based on their disease status [18, 19]. Evidence suggests that in conventional regimens guided by physicians subjects remain on low doses of insulin and are seldom titrated sufficiently to achieve treatment targets [20]. Thus, a simple and safe titration regimen that could be successfully undertaken by the patients themselves would be beneficial.

We anticipated that if patients were able and willing to undergo a program of self-adjustment of insulin doses associated with structured SMBG, their metabolic control would improve. The aim of the present study was to test the effectiveness, practicability and safety of an outpatient program that could help patients with T2DM to make safe and effective intensive insulin therapy self-adjustments.

Some randomized controlled trials examining different self-titration techniques, most of them with basal or premixed insulin analogues, have found that self-adjustment of insulin is effective in helping patients with T2DM safely meet their treatment goals [9, 13, 20–25]. However, few studies are available that specifically consider patients already on insulin therapy and using NPH insulin. In a recent cross-sectional study carried out to investigate the hypothesis that self-titration of insulin would improve metabolic control, Beluchin et al. found that two thirds of patients who had undergone training for self-management practiced it, but there were no significant differences regarding A1C between patients who did or did not perform self-adjustment [26].

In the present study, patients who received training in self-titrating insulin doses according to a specific protocol achieved a significant reduction in A1C levels from 9.0 ± 0.8 to 8.0 ± 1.2% (p < 0.006) after a period of 12 weeks, while this effect was not detected in the control group (A1C: 9.6 ± 1.6 to 9.0% ± 1.1%, p = 0.131). This improvement in A1C was achieved with a nonsignificant incidence of hypoglycemia or change in body weight, which could be concerns regarding the safety of a self-titration insulin regimen.

However, when the two groups were compared regarding baseline to endpoint A1C, there was no statistically significant difference (p = 0.051), although this borderline p value could indicate a strong trend in favor of self-adjustment and a result of clinical significance. This is corroborated by the fact that A1C improved in 90% of patients of the intervention group and in 50% of the control patients and that a larger number of individuals in group B achieved an A1C near the treatment target (<7.5%), with the difference being significant (50 vs. 8.3%, p < 0.029). This finding could be considered an important short-term therapeutic response.

Basal, bolus and total insulin doses did not differ between groups after 12 weeks. A possible explanation for the better results in the intervention group could be the fact that these patients were instructed on how to perform insulin bolus corrections for high BG measures based on an individual sensitivity factor. These boluses were not accounted for in the final insulin doses and were difficult to quantify during the 12-week treatment period, but all group B patients reported to have used this technique. Similar findings have been reported by Pieber et al., who described significant improvement in A1C without a change of insulin doses after an outpatient education program designed to intensify insulin therapy [27].

Few publications have described the relationship between SMBG and glycemic control beyond the frequency of testing to determine whether patients clearly understand their glycemic targets and how they respond to the information obtained from monitoring [10, 15, 28]. SMBG is an essential part of management for patients who properly self-adjust their insulin doses and patients need to know and understand their BG goals and what steps to take in response to a high or low reading, such as diet changes, exercise, and/or medication [4, 29].

In the present study, the number of BG tests per week and the compliance with the expected SMBG rate were significantly higher in the intervention group. Despite the small sample size, these results may have contributed to the achievement of a significant reduction of A1C in group B. Some authors have proposed that educational programs focusing on enhanced SMBG seem to be a stimulus for behavioral change on the part of the patients, empowering and giving them the confidence to become more involved in their treatment and resulting in improved glycemic control. This also applies to those who do not self-adjust insulin doses, with the information provided by the BG being used to promote lifestyle changes [14, 15, 30].

Thus, these findings could suggest that it is possible to use a titration regimen applied by the subjects themselves to their treatment management with positive results in glycemic control, but some limitations should be highlighted. The main limitation is the small sample size given our limited availability of BG testing strips to perform the protocol, which may have impacted the results and have conferred a strength of 70% to this study. A larger sample could increase the power of the study and demonstrate a significant difference between the strategy of self-titration and conventional treatment, favoring the diffusion of the former [31]. Another limitation was the selection of an outpatient population treated in a specialized medical center, so that whether or not or to what extent these results are applicable to other patient populations is unknown. A possible contamination effect should be pointed out in view of the infeasibility of blinding the participants, with the awareness of the group being included in a study possibly contributing to the effect of such study. As done in most studies, we have compared a more intensive intervention to basic care and education, since it is generally considered unethical to randomize a group to receive no education, which could have minimized the measured effects of the intervention [32].

Currently, most insulin-requiring patients with T2DM have their treatment titrated by their clinicians at intervals of three months, which can be a time-consuming and wearing process that may not provide optimal glycemic management for the patients [14]. The present study provides treatment optimization with insulin titration performed by elderly and middle-aged diabetic patients with longstanding disease, allowing them to safely and effectively participate in the management of their treatment. This approach has the potential to significantly improve their glycemic control and to reduce the burden of care for healthcare professionals. Empowering patients to take up a more active role in their therapy through self-titration of insulin dosing may, in some cases, be more effective than physician-directed titration in achieving glycemic control, and may also take some strain off overstretched primary care physicians through reduced patient visits [14].

The present findings may have important implications for educational program planning in DM treatment. The feasibility of this kind of teaching program for this patient profile is evidenced by the high rate of compliance in the intervention group versus the control group, the practicality of the treatment algorithm, the correction of hyperglycemia, and the lack of increase in the frequency of hypoglycemic reactions. The program also involves aspects of treatment that, if strengthened, would probably further improve metabolic control and offset any reluctance on the part of physicians to progress insulin therapy at the time of regular appointments.

The present pilot study suggests that a relatively inexpensive educational program with insulin self-titration interventions based on structured SMBG significantly reduces A1C during a follow-up of 12 weeks and shows a trend towards greater effectiveness in improving glycemic control than conventional treatment, with no increase in incidence of hypoglycemia or body weight gain. Larger randomized, controlled studies are needed to definitively assess the effectiveness of diabetes education programs focusing on insulin self-adjustment for patients with T2DM.

Daniel Dutra Romualdo Silva is an endocrinologist and Master in Diabetes Education; Adriana Aparecida Bosco is an endocrinologist, PhD and an Associate Professor at the Postgraduate Program of Santa Casa de Belo Horizonte.