Longitudinal changes in glycated haemoglobin following treatment intensification after inadequate response to two oral antidiabetic agents in patients with type 2 diabetes

Aims To identify change in glycated haemoglobin (HbA1c) for 1 year after treatment intensification in patients with HbA1c >53 mmol/mol (7.0%) while on two classes of oral antidiabetic drugs (OADs). Material and methods A retrospective cohort study was conducted using a regional health plan claims database for the period January 1, 2010 to March 31, 2017. Patients with type 2 diabetes (T2DM) whose treatment was intensified with insulin, a glucagon‐like peptide‐1 receptor agonist or a third OAD within 365 days of having HbA1c ≥53 mmol/mol (7.0%) on two OADs were included. The HbA1c trajectory for 1 year after intensification was estimated using a mixed‐effects regression model. Results The analysis included 1226 patients with a mean ± SD HbA1c at treatment intensification of 74.2 ± 18.7 mmol/mol (8.93 ± 1.7%). HbA1c was higher in the insulin group (74.2 mmol/mol) than in the non‐insulin group (70.6 mmol/mol), as was the HbA1c decrease (P < 0.01) over the 1‐year follow‐up, particularly in patients with baseline HbA1c >9%. After intensification, insulin‐ and non‐insulin‐treated patients achieved an average change by month in HbA1c of −4.7 mmol/mol and −2.6 mmol/mol points, respectively. The analysis predicted HbA1c to be the lowest at 6 to 10 months post intensification, depending on intensification treatment and HbA1c at intensification; however, on average, HbA1c remained above 64.0 mmol/mol (8.0%). Conclusion In patients with T2DM, intensification following an HbA1c value ≥53 mmol/mol (7.0%) while on two OADs was associated with a significant improvement in glycaemic control. Patients intensified with insulin had a higher baseline HbA1c but greater HbA1c reduction than those intensified with a non‐insulin agent. However, HbA1c remained above 64 mmol/mol (8.0%) overall. Additional opportunity exists to further intensify therapy to improve glycaemic control.

Aims: To identify change in glycated haemoglobin (HbA1c) for 1 year after treatment intensification in patients with HbA1c >53 mmol/mol (7.0%) while on two classes of oral antidiabetic drugs (OADs).
Material and methods: A retrospective cohort study was conducted using a regional health plan claims database for the period January 1, 2010 to March 31, 2017. Patients with type 2 diabetes (T2DM) whose treatment was intensified with insulin, a glucagon-like peptide-1 receptor agonist or a third OAD within 365 days of having HbA1c ≥53 mmol/mol (7.0%) on two OADs were included. The HbA1c trajectory for 1 year after intensification was estimated using a mixedeffects regression model.

Results:
The analysis included 1226 patients with a mean ± SD HbA1c at treatment intensification of 74.2 ± 18.7 mmol/mol (8.93 ± 1.7%). HbA1c was higher in the insulin group (74.2 mmol/mol) than in the non-insulin group (70.6 mmol/mol), as was the HbA1c decrease (P < 0.01) over the 1-year follow-up, particularly in patients with baseline HbA1c >9%. After intensification, insulin-and non-insulin-treated patients achieved an average change by month in HbA1c of −4.7 mmol/mol and −2.6 mmol/mol points, respectively. The analysis predicted HbA1c to be the lowest at 6 to 10 months post intensification, depending on intensification treatment and HbA1c at intensification; however, on average, HbA1c remained above 64.0 mmol/mol (8.0%).
Conclusion: In patients with T2DM, intensification following an HbA1c value ≥53 mmol/mol (7.0%) while on two OADs was associated with a significant improvement in glycaemic control.
Patients intensified with insulin had a higher baseline HbA1c but greater HbA1c reduction than those intensified with a non-insulin agent. However, HbA1c remained above 64 mmol/mol (8.0%) overall. Additional opportunity exists to further intensify therapy to improve glycaemic control.

K E Y W O R D S
antidiabetic drug, glycaemic control, observational study 1 | INTRODUCTION Type 2 diabetes mellitus (T2DM) affects 8.5% to 9% of the US population. 1 Of the necessary assessments for comprehensive diabetes management, glycated haemoglobin (HbA1c) has been a primary indicator for the performance of antidiabetic care and a key predictor of long-term outcomes. 2 Studies have shown that wellcontrolled glycaemic level (HbA1c <53 mmol/mol or 7.0%) reduces the risk of complications. 3 Conversely, inadequately controlled HbA1c is associated with a significantly higher risk of vascular complications and leads to blindness or premature death. 4,5 Consequently, patients with poorly controlled diabetes are expected to have worse outcomes. Despite this evidence,~50% of patients with T2DM fail to attain optimal glycaemic control. 6 Pharmacotherapy intensification is generally recommended when glycaemic level is inadequately controlled despite acceptable medication adherence. 7 Supporting the clinical guidelines, observational studies have consistently shown a positive association between the decrease in HbA1c and treatment intensification in patients who did not achieve glycaemic control with metformin (MET)-based antidiabetic care. [8][9][10] Time in clinical inertia, that is, the failure to initiate or intensify treatment when indicated, needs to be shortened to increase the likelihood of attaining glycaemic control and lower HbA1c levels. 11 However, 52% of patients with insufficient response to MET, alone or along with another oral antidiabetic drug (OAD), experienced therapeutic inertia for ≥1 year. 8 A better understanding of the changes in HbA1c over time following treatment intensification could provide healthcare providers with insights into what to expect in terms of a trajectory and duration of improvement. However, we are unaware of clinical trials that assess outcomes after treatment intensification that are not drug-specific.
Clinical trials also often provide HbA1c data captured during regular follow-up visits. Administrative claims and laboratory data captured during the course of patient care can help to fill this gap. Evaluation of HbA1c trajectories in administrative data is complex, however, because of factors such as sub-optimal adherence and inconsistent follow-up and monitoring. Therefore, observational studies often assess the impact of therapeutic changes on glycaemic control based on a single follow-up HbA1c reading closest to a target follow-up date (e.g. 6 or 12 months after treatment change) or based on an average HbA1c over a defined follow-up period. intensifying treatment in patients who have failed to maintain glycaemic control while on OAD therapy with two agents has not been assessed. These patients with more progressed disease may have a different response to add-on therapy than those who are treatment-naïve or have been on monotherapy. 15 The first objective of the present real-world study was to compare glycaemic control as measured by HbA1c before and after the treatment intensification in patients whose HbA1c was ≥53 mmol/mol (7.0%) while on two OADs. The second objective was to project the HbA1c trajectory after the treatment intensification therapy stratified by insulin and non-insulin medications, which included glucagon-like peptide-1 receptor antagonist (GLP-1RA) agents or a third OAD.

| Data
This was a retrospective observational study of commercially insured patients with T2DM from January 2010 to March 2017. The study was based on medical and pharmacy claims data obtained from SelectHealth, a health plan provider in the intermountain region with 800 000 enrollees, of whom most are in a commercial plan (84%) and a small proportion in a Medicare plan (3%). The study population was, therefore, relatively young compared to the general T2DM population. The SelectHealth claims data were augmented with HbA1c values obtained through a provider-incentivized quality improvement programme. Thus, this study has the strength of a claims dataset in terms of access to comprehensive medication use data plus laboratory data to assess clinical outcomes. While other claims datasets are augmented with laboratory data, SelectHealth's quality initiative has resulted in HbA1c data being relatively well documented for this population, reducing the risk of measurement bias attributable to missing data.

| Study cohort
The analytical cohort was extracted from the claims for medical and Index (DCSI) was calculated based on diagnoses codes captured during the 365-day baseline period. 16

| Outcomes
The outcome was the difference in HbA1c before and after the date of treatment intensification.

| Covariates
Baseline characteristics were described by treatment intensification group categorized as insulin versus non-insulin and included age, gender, geographic location by state, type of insurance, HbA1c at baseline and at intensification, baseline OAD treatment, and DCSI. Average proportion of days covered over the baseline period while patients were on two OADs was calculated using Choudhry's prescriptionbased method to address adherence to an antidiabetic treatment regimen. 17

| Statistical analyses
Descriptive statistics were used to present baseline characteristics.
Statistical significance between groups was tested using Student's t-

| Institutional review board approval
This study protocol was reviewed and deemed exempt from University of Utah institutional review (#00098483) and was approved by the Intermountain Healthcare institutional review board (#1050483).
The age at treatment intensification was similar in the insulin and the non-insulin groups, with the mean ± SD age being 54 ± 11 years and 55 ± 10 years, respectively (P = 0. 19

| HbA1c trajectory: mixed-effect regression model
The results of the regression analyses estimated that HbA1c at inten-  These data indicate that a greater HbA1c reduction was observed in patients with higher HbA1c at intensification, overall. Insulin was associated with a higher HbA1c at intensification and a greater reduction over the follow-up period than intensification with a non-insulin agent. For all patients with HbA1c ≥75 mmol/mol (9.0%) at intensification and non-insulin patients with HbA1c <75 mmol/mol (9.0%) at intensification, HbA1c reduction levelled off between months 6 and 8, then began to rise. Insulin patients with HbA1c <75 mmol/mol (9.0%) at intensification experienced a modest drop in HbA1c until month 9, then HbA1c stabilized. No significant difference in HbA1c change was identified between patients whose therapy was intensified with GLP-1RAs versus OADs (Table S1 and Figure 2).

| DISCUSSION
The primary goal of diabetes management is to maintain a level of blood glucose that balances the risk of complications with the risks associated with treatment, including hypoglycaemia. 7 With disease progression, glycaemic control is maintained by a stepwise addition of antidiabetic agents, after patients fail to achieve HbA1c with initial treatment, often MET monotherapy. 7 Previous research by Fu and Sheehan 8 has shown that 50% of patients who did not sufficiently respond to MET treatment as monotherapy or in combination with other oral agents required and received treatment intensification within 1 year. The authors also identified that the likelihood of better HbA1c maintenance increased when treatment inertia was shorter than 6 months. 8,9 The present study expands on the research by Fu and Sheehan 8,9 by not limiting prior therapy to MET users, and by specifically focusing on patients who had failed to maintain glycaemic control while on two OADs. We found that switching or adding a third agent significantly reduced HbA1c. The decrease in the HbA1c was greatest over the first 4 months, and HbA1c for the overall cohort remained below HbA1c at intensification for the follow-up period. HbA1c started to increase before the end of the first year after the intensification. This could be for multiple reasons, including disease progression or reduced adherence, and warrants further investigation.
Overall, glycaemic response was greatest in those with poor glycaemic control, which was similarly observed in previous observational studies. In a study by Pantalone et al including patients whose HbA1c remained >53 mmol/mol (7.0%) on MET monotherapy, the level of improvement in glycaemic control was linearly associated with an increase in the baseline HbA1c level. 11 In another recent analysis of insurance claims for patients with T2DM, where treatment failure was defined by ≥64 mmol/mol (8.0%) after a treatment with metformin monotherapy or in combination with other OADs, the change in HbA1c was greater with a higher pre-intensification HbA1c. From the same study, treatment intensification in patients having pre-index HbA1c ≥75 mmol/mol (9.0%) resulted in the HbA1c change of −23.7 mmol/mol (−2.15%), which is close to our estimate of HbA1c decrease in the poor glycaemic control group. 9 The HbA1c reduction was also 2.  HbA1c exceeds 75 or 86 mmol/mol (9.0% or 10%). 7,15,18 It also reflects the demonstrated glycaemic response to insulin. 7 The present study is important in providing data on the effectiveness of diabetes therapy intensification after failing to maintain glycaemic control on two OADs. The American Association of Clinical Endocrinologists and American College of Endocrinology consensus statement has highlighted that patients taking two or more medications will probably see less improvement with intensification or a switch to a new agent than can be expected when the same agent is used as first-or second-line therapy. 15 Observational studies have also seen a negative association between the number of prior classes used and the level of HbA1c reduction. 12 This observation may reflect the decline in insulin sensitivity and β-cell function with T2DM progression. 19  Clinical practice should be informed by the present findings.
While insulin therapy was associated with the greatest reduction in HbA1c for all patients with HbA1c ≥75 mmol/mol (9.0%), those with HbA1c ≥75 mmol/mol (9%) and receiving intensification with a GLP-1RA or third OAD also experienced significant reductions in HbA1c over the first 6 months. Despite these gains, patients on average did not attain HbA1c <64 mmol/mol (8.0%). This is a conservative goal for the study population, given their younger age and that >50%  addressed this limitation, it did not control for the effect of disease severity on follow-up HbA1c trajectories. In addition, this study captured treatment intensification based on a prescription claim for a different class of diabetes drug, but did not distinguish whether this treatment represented an add-on or switch, nor did it assess for dose escalations after intensification. Finally, the study did not control for diabetes medication compliance in the regression analyses. Baseline adherence did not differ, however, between intensification groups.
The role of post-intensification compliance on HbA1c trajectories should be addressed in future studies.
In conclusion, treatment intensification after failing to achieve glycaemic control on two OADs was associated with a significant reduction in HbA1c within 6 months that was maintained or just slightly rose after 6 to 9 months. The largest reduction in HbA1c was seen in those with HbA1c ≥75 mmol/mol (9.0%) and treated with insulin, but a significant improvement in glycaemic control was also seen with non-insulin agents. Even with improvement in glycaemic control, many patients failed to achieve an HbA1c target <64 mmol/mol (8.0%). Thus, opportunity exists for additional and/or more aggressive treatment intensification and patient engagement and management to achieve optimal glycaemic control and help reduce the risk of diabetes complications.

AUTHOR CONTRIBUTIONS
The study concept was proposed in collaboration by all authors.

SUPPORTING INFORMATION
Additional supporting information may be found online in the Supporting Information section at the end of this article.
How to cite this article: Kim K, Unni S, Brixner DI, et al.