Impact of urgently initiated tele-prescription due to COVID-19 on glycemic control in patients with type 2 diabetes

Article information

Korean J Intern Med. 2021;36(4):942-948
Publication date (electronic) : 2021 June 7
doi : https://doi.org/10.3904/kjim.2020.464
1Department of Internal Medicine, Kyungpook National University Hospital, School of Medicine, Kyungpook National University, Daegu, Korea
2Department of Internal Medicine, Kyungpook National University Chilgok Hospital, School of Medicine, Kyungpook National University, Daegu, Korea
Correspondence to: Yeon-Kyung Choi, M.D., Division of Endocrinology and Metabolism, Department of Internal Medicine, Kyungpook National University Chilgok Hospital, School of Medicine, Kyungpook National University, 807 Hoguk-ro, Buk-gu, Daegu 41404, Korea, Tel: +82-53-200-2175, Fax: +82-53-200-2176, E-mail: ykchoi@knu.ac.kr, https://orcid.org/0000-0003-0996-6437
*

These authors contributed equally.

Received 2020 August 27; Revised 2020 November 26; Accepted 2021 January 25.

Abstract

Background/Aims

Coronavirus disease 2019 (COVID-19) is a global pandemic that had affected more than 13,000 people in South Korea by July 2020. To prevent spread of COVID-19, tele-prescription was permitted temporarily. This study investigated the impact of tele-prescription on glycemic control in patients with type 2 diabetes.

Methods

Glycated hemoglobin (HbA1c) concentrations were retrospectively analyzed in patients with type 2 diabetes who were treated with tele-prescription because of COVID-19 and those who were treated by face-to-face care (non-tele-prescription group) enrolled at the same period of time. Mean HbA1c concentrations and mean change in HbA1c concentration (ΔHbA1c) were compared in these two groups.

Results

The mean HbA1c levels of patients were significantly higher after than before the tele-prescription period (7.46% ± 1.24% vs. 7.27% ± 1.13%, p < 0.05). Mean ΔHbA1c was significantly higher in the tele-prescription than in the non-tele-prescription group (0.19% ± 0.68% vs. 0.04% ± 0.95%, p < 0.05). HbA1c was significantly greater in patients taking fewer oral hypoglycemic agents, no insulin, fewer comorbidities (e.g., coronary artery disease, cerebrovascular accident, and diabetic neuropathy), and higher baseline HbA1c.

Conclusions

Tele-prescription may worsen glycemic control in patients with type 2 diabetes during public health crises.

INTRODUCTION

Since the onset in December 2019 of coronavirus disease 2019 (COVID-19), caused by severe acute respiratory syndrome coronavirus-2 (SARS-CoV-2), this disease has spread rapidly to over 160 countries worldwide [1]. As of July 10, 2020, there were more than 13,000 confirmed cases and about 280 COVID-19-related deaths in South Korea [2]. Chronic diseases such as diabetes have been identified as important risk factors for severe illness and death from COVID-19 [3,4]. One meta-analysis showed that mortality rates in patients with confirmed COVID-19 were about 2-fold higher in patients with than without diabetes [5]. Therefore, to reduce mortality from COVID-19, it is important to prevent patients with diabetes from contracting this disease.

The rapid spread of COVID-19 in South Korea led to the unprecedented implementation of strict social distancing. Due to the limitation of face-to-face visits in clinics, telemedicine with remote prescriptions was permitted temporarily, beginning on February 24, 2020 [2]. Originally, telemedicine was developed to provide accessible, cost-effective, high-quality health care services to patients in remote locations using telecommunication technology [6]. It uses electronic communications, advanced computing, and telematics technology to exchange medical information [7]. Telemedicine had already shown its usefulness in disasters and public health crises [8,9], and is expected to reduce the likelihood of COVID-19 infection by avoiding face-to-face contact during the COVID-19 pandemic.

The potential benefit of telemedicine for managing diabetes has been advocated since the early 1980s. Some studies actively involved in blood sugar control using a self-reporting system show that telemedicine can be a good strategy for the treatment of patients with diabetes [10,11]. However, rigorous telemedicine was not ready to start in the current medical environment; therefore, it was started temporally with tele-prescriptions, a form of telephone consultation and remote prescription. Although telemedicine is applicable during disasters and public health crises [8,9], the effects of urgently initiated tele-prescription strategies on glycemic control in patients with type 2 diabetes during public health crises are not clear.

Therefore, the present study investigated whether glycated hemoglobin (HbA1c) concentrations differ between patients with type 2 diabetes treated by tele-prescription and those treated by face-to-face care.

METHODS

Study design and participants

This retrospective, case-control study included patients with type 2 diabetes aged > 20 years who visited two tertiary hospitals, Kyungpook National University Hospital (KNUH) and Kyungpook National University Chilgok Hospital (KNUCH), in Daegu, South Korea. The study protocol was approved by the Institutional Review Boards of KNUH and KNUCH (IRB numbers 2020-07-020 and 2020-05-021, respectively). The ethics boards of the two hospitals waived the requirement for informed consent because of the retrospective study design.

Tele-prescription was conducted entirely at the request of the patient, who received a remote prescription after telephone counseling. The process of tele-prescription included self-monitoring of blood glucose and counseling for insulin dose titration or diet control; therefore, the time required for tele-prescription was almost the same as that for face-to-face care. The number of tele-prescriptions implemented during the entire study period was limited to one. For patients with diabetes, who need immediate blood sugar control through strong interventions (e.g., first insulin start, hyperglycemic crisis, and hypoglycemia), face-to-face care was recommended. However, patients who did not require strong intervention and were already using insulin were managed via tele-prescription. The flow chart of the study design is presented in Fig. 1. The study periods were defined according to the introduction of tele-prescription relative to the COVID-19 outbreak in Daegu. The tele-prescription period was defined as 2 months after tele-prescription was first introduced (February 26 to April 25, 2020). Period 1 (August 26, 2019 to February 25, 2020) was defined as 6 months before the start of tele-prescription and Period 2 as 4 months after the introduction of tele-prescription. HbA1c was measured during Period 1 and again during Period 2. HbA1c results for Period 1 and Period 2 were selected as values; these were measured only once each within 1 to 6 months before and after the tele-prescription period. The control group was selected during the same period using the same method.

Figure 1

Flow chart of the study design.

A total of 1,000 patients were treated by tele-prescription at the two hospitals (733 at KNUH and 267 at KNUCH) during the defined tele-prescription period, which is equivalent to 14.17% and 16.88%, respectively, of all outpatients of the department of ‘Endocrinology and metabolism’. Among these, 539 patients, 419 at KNUH and 120 at KNUCH, were treated for diabetes. Of these 539 patients, 236 were excluded, including (1) patients who were not followed-up during Period 2 (127 at KNUH, 48 at KNUCH); (2) patients who changed oral hypoglycemic agents (OHA) or insulin during Period 1 (27 at KNUH, 10 at KNUCH); (3) patients diagnosed with diabetes < 1 year before Period 1 (none at KNUH, two at KNUCH); (4) patients hospitalized < 1 year before Period 1 (eight at KNUH, two at KNUCH); (5) patients diagnosed with type 1 diabetes (two at KNUH, none at KNUCH); and (6) patients being treated for serious illnesses, including those receiving chemotherapy (four at KNUH, six at KNUCH). The case group, therefore, consisted of 303 patients.

During the same period of time, 2,852 patients (2,316 at KNUH and 536 at KNUCH) with diabetes were treated by face-to-face care. Among these, 1,821 patients (1,405 at KNUH and 416 at KNUCH) measured baseline HbA1c levels during Period 1. To effectively control for confounding factors, such as age, gender, and baseline HbA1c, 1:1 propensity score matching was performed to select 302 patients as the control group according to the same inclusion and exclusion criteria.

Statistical analysis

All statistical analyses were performed using SPSS version 26.0 software (IBM Inc., Armonk, NY, USA). Paired t tests were used to compare mean HbA1c levels within each group during Periods 1 and 2. Independent t tests were used to compare differences in ΔHbA1c level in the two groups between Periods 1 and 2. A p < 0.05 was considered statistically significant.

RESULTS

Baseline characteristics of patients with type 2 diabetes in the tele-prescription and non-tele-prescription groups

The characteristics of the patients enrolled in this study and of all patients with diabetes are presented in Table 1 and Supplementary Table 1, respectively. The mean age of all participants was 62.78 ± 11.70 years, the mean duration of diabetes was 10.36 ± 7.75 years, the mean baseline HbA1c was 7.26% ± 1.15%, the mean number of OHA per patient was 2.25 ± 0.83, and the mean number of comorbidities per patient was 2.43 ± 1.33. Of the total participants, 134 (22.1%) patients were taking insulin, 273 (45.1%) were taking anti-hypertensive drugs, and 373 (61.7%) were taking lipid-lowering drugs. There were no between-group differences in patient age, gender, duration of diabetes, baseline HbA1c concentration, number of OHAs, insulin use, and number of comorbidities. However, there were significant differences between the groups regarding diabetic neuropathy and cerebrovascular accident (CVA).

Characteristics of patients with type 2 diabetes in the tele-prescription and non-tele-prescription groups

Mean HbA1c and ΔHbA1c of patients with type 2 diabetes in the tele-prescription and non-tele-prescription groups

In the tele-prescription group, the mean HbA1c levels were significantly higher during Period 2 than Period 1 (7.46% ± 1.24% vs. 7.27% ± 1.13%, p < 0.05) (Fig. 2). By contrast, mean HbA1c levels were similar during Periods 1 and 2 in the non-tele-prescription group (7.24% ± 1.16% vs. 7.28% ± 1.24%, p = 0.48). Regarding the difference in HbA1c between Periods 1 and 2, ΔHbA1c was significantly greater in the tele-prescription than in the non-tele-prescription group (0.19% ± 0.68% vs. 0.04% ± 0.95%, p < 0.05) (Fig. 3).

Figure 2

Mean glycated hemoglobin (HbA1c) of patients with diabetes in the tele-prescription and non-tele-prescription groups. Tele-prescription group: 7.27% ± 1.13% (Period 1), 7.46% ± 1.24% (Period 2); non-tele-prescription group: 7.24% ± 1.16% (Period 1), 7.28% ± 1.24% (Period 2). Results are reported as the mean ± standard deviation HbA1c concentration. Data were analyzed using a paired t test. NS, not significant. ap < 0.05.

Figure 3

Glycated hemoglobin concentration (ΔHbA1c) of patients with diabetes in the tele-prescription and non-tele-prescription groups. Tele-prescription group: 0.19 ± 0.68; non-tele-prescription group: 0.04 ± 0.95. Results are reported as the mean ± standard deviation ΔHbA1c concentration. Data were analyzed using an independent t test. ap < 0.05.

Subgroup analysis of delta glycated hemoglobin in the tele-prescription and non-tele-prescription groups

Subgroup analysis was performed to assess the relationships of ΔHbA1c with gender, age, duration of diabetes, number of OHAs, insulin use, number of comorbidities, and baseline HbA1c in the two groups (Table 2). The HbA1c in participants with fewer OHAs (≤ 2), non-insulin users, fewer comorbidities (≤ 2), and higher baseline HbA1c (> 7.5%) was significantly higher after than before the introduction of tele-prescription (p < 0.05). In addition, HbA1c was significantly higher after than before tele-prescription in patients without coronary artery disease, CVA, and diabetic neuropathy (p < 0.05) (Supplementary Table 2).

Subgroup analysis of mean and delta glycated hemoglobin in the tele-prescription and non-tele-prescription groups

DISCUSSION

The results of this study demonstrated that urgently initiated tele-prescription did not show a better effect on glycemic control in patients with type 2 diabetes during the COVID-19 outbreak than face-to-face care. Mean HbA1c concentration showed a greater increase in patients after introduction of tele-prescription than in those treated by face-to-face care. ΔHbA1c in the tele-prescription group was especially higher in patients with fewer OHA medications, no insulin, fewer comorbidities, and higher baseline HbA1c.

Social distancing, introduced after the start of the COVID-19 pandemic, has resulted in many changes to the medical environment. The need for telemedicine has been emphasized, with several countries, including China [12], Singapore [13], and Australia [14], implementing telemedicine programs. Some studies demonstrate the benefits of telemedicine in patients with diabetes. A meta-analysis of 35 randomized controlled trials of telemedicine using video, phone, and e-mail showed a reduction in HbA1c of −0.37% (p < 0.001) in the telemedicine group compared with controls [15]. A recent study that included patients with type 1 and 2 diabetes showed an overall mean reduction in HbA1c in the telemedicine groups with type 1 (−0.12% to −0.86%) and type 2 (−0.01% to −1.13%) diabetes [16]. Although many advantages of telemedicine are known [17,18], it was difficult to implement proper telemedicine protocols in this study due to limited availability of medical technology and devices. We found that the mean HbA1c level increased significantly after introduction of tele-prescription, but was unaltered in the non-tele-prescription group. Furthermore, ΔHbA1c was significantly greater in the tele-prescription than in the non-tele-prescription group.

Subgroup analysis showed that HbA1c after tele-prescription care was significantly higher in patients with fewer OHA medications, and comorbidities. In addition, tele-prescription care was associated with a higher ΔHbA1c in patients not taking insulin, which may be explained, at least in part, by their inability to adjust their medications according to glucose levels (unlike insulin users). Furthermore, ΔHbA1c was higher in patients with baseline HbA1c > 7.5% than in those with baseline levels ≤ 7.5%, suggesting that tele-prescription may be deleterious in patients who require immediate and active strategies for glucose control. Interestingly, especially in patients without comorbidities such as coronary artery disease, CVA, and diabetic neuropathy, HbA1c was significantly higher after than before the tele-prescription period. It is assumed that the changes in lifestyle interventions may have had an impact on glucose control these groups due to the COVID-19 pandemic, but further research is required to prove a causal relationship. These findings suggest that if remote diagnosis is not conducted, glycemic control may worsen; thus tele-prescription should be used carefully, especially for patients with relatively lesser comorbidities during the public health crisis.

Since this is a retrospective study, it has several limitations. First, the baseline characteristics of the patient group, such as height and weight, were not investigated, and relatively different HbA1c values (depending on gender) may introduce selection bias that affects the results in subgroup analysis. Second, the glucose control was assessed only by measuring HbA1c levels; we did not take into account additional factors such as physical activity, diet, or adherence to medication, which could affect glycemic control. Third, patients who request telephone prescriptions are more likely to show relatively low compliance with treatment; however, this variable was not reflected in our study. Fourth, the observation period was too short to fully assess the impact of tele-prescription care on glucose control. To overcome this limitation, a preemptive and well-organized system is required. Objective indicators of diet, physical activity, and drug compliance, which are factors that can affect blood sugar control, should be identified. Finally, a discussion about how to implement this for patients who need appropriate intervention through remote counseling is needed.

Despite these limitations, our study has generated meaningful findings. To the best of our knowledge, this report is the first to show the impact of tele-prescription on glycemic control in patients with type 2 diabetes during the COVID-19 pandemic. These findings suggest that care should be exercised when introducing a tele-prescription strategy for treatment of diabetes during a public health crisis, especially for patients with relatively well controlled blood sugar. In addition, such strategies should be accompanied by appropriate intervention and remote diagnosis, which includes self-monitoring of glucose. Additional studies are needed to determine the impact of telemedicine on clinical outcomes during acute health crises.

KEY MESSAGE

1. Tele-prescription, which was urgently initiated during the coronavirus disease 2019 (COVID-19) outbreak, did not show a better effect on glycemic control in patients with diabetes than face-to face care.

2. In particular, the following patients with diabetes need more careful monitoring during urgently initiated tele-prescription: less oral hypoglycemic agent (≤ 2), non-insulin users, fewer comorbidities (≤ 2), and higher baseline glycated hemoglobin (> 7.5%).

3. During a public health crisis, tele-prescription is expected to provide many benefits for patients with diabetes, but this should be accompanied by a well-organized system and remote diagnosis.

Supplementary Materials

Notes

No potential conflict of interest relevant to this article was reported.

Acknowledgements

This work was supported by a Biomedical Research Institute grant, Kyungpook National University Hospital (2020).

We thank all the patients for participating in this study.

References

1. Del Rio C, Malani PN. COVID-19: new insights on a rapidly changing epidemic. JAMA 2020;323:1339–1340.
2. Central Disaster Management Headquarters. COVID-19 [Internet] Cheongju (KR): Korea Center for Disease Control and Prevention; 2020. [cited 2021 May 12]. Available from: http://ncov.mohw.go.kr .
3. Zhou F, Yu T, Du R, et al. Clinical course and risk factors for mortality of adult inpatients with COVID-19 in Wuhan, China: a retrospective cohort study. Lancet 2020;395:1054–1062.
4. Liu H, Chen S, Liu M, Nie H, Lu H. Comorbid chronic diseases are strongly correlated with disease severity among COVID-19 patients: a systematic review and meta-analysis. Aging Dis 2020;11:668–678.
5. Kumar A, Arora A, Sharma P, et al. Is diabetes mellitus associated with mortality and severity of COVID-19? A meta-analysis. Diabetes Metab Syndr 2020;14:535–545.
6. Wootton R. Recent advances: telemedicine. BMJ 2001;323:557–560.
7. Krumm M. A review of contemporary tele-audiology literature. J Hear Sci 2016;6:9–21.
8. Lurie N, Carr BG. The role of telehealth in the medical response to disasters. JAMA Intern Med 2018;178:745–746.
9. Garshnek V, Burkle FM Jr. Applications of telemedicine and telecommunications to disaster medicine: historical and future perspectives. J Am Med Inform Assoc 1999;6:26–37.
10. Bian Y, Xiang Y, Tong B, Feng B, Weng X. Artificial intelligence-assisted system in postoperative follow-up of orthopedic patients: exploratory quantitative and qualitative study. J Med Internet Res 2020;22:e16896.
11. Piette JD, Weinberger M, McPhee SJ. The effect of automated calls with telephone nurse follow-up on patient-centered outcomes of diabetes care: a randomized, controlled trial. Med Care 2000;38:218–230.
12. Song X, Liu X, Wang C. The role of telemedicine during the COVID-19 epidemic in China-experience from Shandong province. Crit Care 2020;24:178.
13. Ngoh CLY, Wong WK, Leo CCH, Choo TT, Khan BA. Rapid transition to a telemedicine service at Singapore community dialysis centers during COVID-19. NEJM Catal Innov Care Deliv 2020. May. 28. Available from: https://catalyst.nejm.org/doi/full/10.1056/cat.20.0145 .
14. Fisk M, Livingstone A, Pit SW. Telehealth in the context of COVID-19: changing perspectives in Australia, the United Kingdom, and the United States. J Med Internet Res 2020;22:e19264.
15. Zhai YK, Zhu WJ, Cai YL, Sun DX, Zhao J. Clinical- and cost-effectiveness of telemedicine in type 2 diabetes mellitus: a systematic review and meta-analysis. Medicine (Baltimore) 2014;93:e312.
16. Timpel P, Oswald S, Schwarz PEH, Harst L. Mapping the evidence on the effectiveness of telemedicine interventions in diabetes, dyslipidemia, and hypertension: an umbrella review of systematic reviews and meta-analyses. J Med Internet Res 2020;22:e16791.
17. Flodgren G, Rachas A, Farmer AJ, Inzitari M, Shepperd S. Interactive telemedicine: effects on professional practice and health care outcomes. Cochrane Database Syst Rev 2015;(9):CD002098.
18. Xu T, Pujara S, Sutton S, Rhee M. Telemedicine in the management of type 1 diabetes. Prev Chronic Dis 2018;15:E13.

Article information Continued

Figure 1

Flow chart of the study design.

Figure 2

Mean glycated hemoglobin (HbA1c) of patients with diabetes in the tele-prescription and non-tele-prescription groups. Tele-prescription group: 7.27% ± 1.13% (Period 1), 7.46% ± 1.24% (Period 2); non-tele-prescription group: 7.24% ± 1.16% (Period 1), 7.28% ± 1.24% (Period 2). Results are reported as the mean ± standard deviation HbA1c concentration. Data were analyzed using a paired t test. NS, not significant. ap < 0.05.

Figure 3

Glycated hemoglobin concentration (ΔHbA1c) of patients with diabetes in the tele-prescription and non-tele-prescription groups. Tele-prescription group: 0.19 ± 0.68; non-tele-prescription group: 0.04 ± 0.95. Results are reported as the mean ± standard deviation ΔHbA1c concentration. Data were analyzed using an independent t test. ap < 0.05.

Table 1

Characteristics of patients with type 2 diabetes in the tele-prescription and non-tele-prescription groups

Characteristic Total Tele-prescription (n = 303) Non-tele-prescription (n = 302) p value
Age, yr 62.78 ± 11.70 63.02 ± 12.16 62.53 ± 11.24 0.61
Male sex 295 (48.8) 146 (48.2) 149 (49.3) 0.78
Duration of diabetes, yr 10.36 ± 7.75 10.57 ± 7.80 10.15 ± 7.71 0.50
Baseline HbA1c, % 7.26 ± 1.15 7.27 ± 1.13 7.24 ± 1.16 0.75
No. of OHAs 2.25 ± 0.83 2.25 ± 0.84 2.26 ± 0.82 0.91
Insulin use 134 (22.1) 61 (20.1) 73 (24.2) 0.23
Comorbidities
 Total number 2.43 ± 1.33 2.42 ± 1.35 2.44 ± 1.30 0.82
 HTN 273 (45.1) 132 (43.6) 141 (46.7) 0.44
 Dyslipidemia 373 (61.7) 168 (55.4) 205 (67.9) < 0.01
 Others 423 (69.9) 220 (72.6) 203 (67.2) 0.23
Micro-vascular complications
 Diabetic retinopathy 46 (7.6) 23 (7.6) 23 (7.6) 0.99
 Diabetic nephropathy 69 (11.4) 34 (11.2) 35 (11.6) 0.89
 Diabetic neuropathy 70 (11.6) 43 (14.2) 27 (8.9) < 0.05
Macro-vascular complications
 Coronary artery disease 99 (16.4) 56 (17.5) 46 (15.2) 0.45
 CVA 66 (10.9) 42 (13.9) 24 (7.9) < 0.05
 Peripheral vascular disease 16 (2.6) 11 (3.6) 5 (1.7) 0.13

Values are presented as mean ± standard deviation or number (%). Data were analyzed using a paired t test.

HbA1c, glycated hemoglobin; OHA, oral hypoglycemic agent; HTN, hypertension; CVA, cerebrovascular accident.

Table 2

Subgroup analysis of mean and delta glycated hemoglobin in the tele-prescription and non-tele-prescription groups

Variable Tele-prescription Non-tele-prescription p value


Period 1 Period 2 Δ HbA1c, % Period 1 Period 2 Δ HbA1c, %
Male 146 149

 HbA1c, % 7.21 ± 1.01 7.34 ± 1.10 0.128 6.94 ± 1.04 6.92 ± 0.91 −0.011 0.172

Female 157 153

 HbA1c, % 7.33 ± 1.20 7.58 ± 1.36 0.252 7.54 ± 1.20 7.63 ± 1.42 0.088 0.061

Age ≤ 65 years 179 175

 HbA1c, % 7.31 ± 1.09 7.48 ± 1.21 0.170 7.20 ± 1.19 7.20 ± 1.16 0.001 0.052

Age > 65 years 124 127

 HbA1c, % 7.22 ± 1.18 7.45 ± 1.29 0.225 7.30 ± 1.13 7.39 ± 1.34 0.091 0.202

Duration ≤ 15 years 243 243

 HbA1c, % 7.24 ± 1.18 7.41 ± 1.24 0.163 7.14 ± 1.17 7.16 ± 1.18 0.024 0.053

Duration > 15 years 60 59

 HbA1c, % 7.39 ± 0.87 7.70 ± 1.24 0.310 7.66 ± 1.05 7.76 ± 1.37 0.100 0.221

No. of OHAs ≤ 2 195 192

 HbA1c, % 6.99 ± 1.00 7.21 ± 1.16 0.220 7.11 ± 1.21 7.15 ± 1.30 0.042 0.030

No. of OHAs > 2 108 110

 HbA1c, % 7.77 ± 1.18 7.92 ± 1.27 0.143 7.47 ± 1.04 7.50 ± 1.11 0.034 0.346

Insulin use 61 73

 HbA1c, % 8.17 ± 1.24 8.54 ± 1.43 0.374 7.74 ± 1.51 7.92 ± 1.75 0.175 0.310

Insulin non-use 242 229

 HbA1c, % 7.05 ± 0.98 7.19 ± 1.03 0.147 7.08 ± 0.98 7.08 ± 0.95 −0.005 0.021

No. of comorbidities ≤ 2 169 172

 HbA1c, % 7.16 ± 1.01 7.39 ± 1.15 0.232 7.19 ± 1.13 7.15 ± 1.10 −0.035 0.002

No. of comorbidities > 2 134 130

 HbA1c, % 7.41 ± 1.25 7.55 ± 1.35 0.143 7.31 ± 1.21 7.45 ± 1.40 0.136 0.953

Baseline HbA1c ≤ 7.5 200 214

 HbA1c, % 6.61 ± 0.51 6.82 ± 0.64 0.208 6.66 ± 0.51 6.83 ± 0.76 0.171 0.522

Baseline HbA1c > 7.5 103 88

 HbA1c, % 8.55 ± 0.87 8.72 ± 1.18 0.163 8.65 ± 1.11 8.37 ± 1.50 −0.282 0.009

Values are presented as number or mean ± standard deviation. Data were analyzed using a paired t test and an independent t test.

ΔHbA1c, glycated hemoglobin concentration; HbA1c, glycated hemoglobin; OHA, oral hypoglycemic agent.