Efficacy and Safety of Insulin Therapy in Patients with Type 2 Diabetes Treated at Different Grades of Hospitals in China: Subgroup Analysis of the Real-World SEAS Study

Abstract

Backgrounds:

As patients attending hospitals of different grades in China may receive different medical care, we investigated the clinical efficacy and safety of routine insulin therapy in patients with type 2 diabetes who were treated at grade 2 and grade 3 (highest grade) hospitals in China.

Methods:

2683 patients with type 2 diabetes were enrolled in a multicenter, nonrandomized, open-label, noninterventional, 12-week clinical trial performed at 62 Chinese hospitals. Patients were divided into two groups according to the hospitals' grading. Data were analyzed for efficacy (changes and normalization of glycated hemoglobin [HbA1c] and changes in fasting plasma glucose and 2-h postprandial blood glucose [PBG] levels from baseline to the final visit) and for safety (hypoglycemia).

Results:

After 12 weeks of routine human insulin (SciLin) therapy, decreases in mean HbA1c and PBG levels were significantly greater in patients treated at second-grade hospitals (all P < 0.001 vs. third-grade hospitals), and the HbA1c success rate (<7%) was significantly higher (46.94% vs. 38.85%; P = 0.0002). However, patients treated at second-grade hospitals had more weight gain (0.29 kg vs. 0.04 kg; P < 0.0001) and a higher incidence of total hypoglycemic events (21.82% vs. 16.79%; P = 0.0002).

Conclusions:

Routine insulin treatment of patients with type 2 diabetes in China demonstrates acceptable safety and effectiveness, improving blood glucose control with a low incidence of severe hypoglycemia. Patients treated at second-grade hospitals had a greater HbA1c success rate than those treated at third-grade hospitals, but with more weight gain and more hypoglycemic events.

Introduction

Type 2 diabetes mellitus is a progressive disease with high morbidity and mortality. The prevalence of type 2 diabetes is reported to be 9% among Caucasians and around 10% to 20% among Asian Indians, Arabs, Chinese, Africans, and Hispanics.¹ As in other countries, a rising trend in the prevalence of type 2 diabetes has been observed in China.² A recent comprehensive epidemiological survey of diabetes revealed that 11.6% of Chinese adults, or 114 million people, have type 2 diabetes, which represents 1 in 3 patients with the condition globally.³ Thus, preventing and controlling the progress of diabetes in China is regarded as an important public health objective.

Professional organizations in China have produced national treatment guidelines recommending that patients with type 2 diabetes should have their glycated hemoglobin (HbA1c) level maintained below <7.0% (<53 mmol/mol) to minimize the risk of developing macrovascular and microvascular complications such as cardiovascular events, retinopathy, and nephropathy.⁴ In the United States, the American Diabetes Association also recommends a HbA1c goal of <7% (<53 mmol/mol).⁵ Since type 2 diabetes cannot be adequately controlled with diet or oral antidiabetic drugs in the long term, the majority of patients eventually require insulin therapy, which is regarded as the most effective measure to improve glycemic control.⁶ However, a study conducted in China by Lu et al.⁷ reported that even with effective insulin treatment, only 27.04% of patients reach the HbA1c target of <7.0% (<53 mmol/mol). The reason for this has not been fully explained.

Interest in real-world outcomes in patients with diabetes has been growing. In China, hospitals are divided into three grades according to their size and the range of medical services they provide, with the highest grade being grade 3 and the lowest grade 1. The higher the grade of a hospital, the more expensive its fees. In addition, clinicians receive different training at the different grades of hospitals, and those trained at higher grade hospitals are more often regarded as specialists. Consequently, patients attending hospitals of different grades often receive different medical care.

To obtain a better understanding of the effectiveness of antidiabetic treatment in routine clinical practice in China, a real-world, observational study known as the SciLin Efficacy And Safety (SEAS) study was performed. In this subgroup analysis of data from the SEAS study, the efficacy and safety of routine insulin therapy were investigated in Chinese patients with type 2 diabetes treated at two different grades of hospitals.

Materials and Methods

The SEAS study was a multicenter, nonrandomized, open-label, noninterventional, observational, 12-week clinical trial conducted at 62 medical centers in China (shown in Appendix 1) between August 10, 2012, and January 15, 2014. A total of 2683 patients with type 2 diabetes received routine treatment with SciLin human insulin (Bayer Schering Pharma, China) at the various centers.

The original study protocol for the SEAS study was approved by the institutional review boards and/or ethics committees of all participating hospitals. Eligible patients participated in the study voluntarily, and all provided written informed consent to do so. All study procedures performed were in accordance with the Declaration of Helsinki and Good Clinical Practice principles. Patients were permitted to withdraw from the study at any time.

The study was registered with the ClinicalTrials.gov database (No. NCT01588639: “To Evaluate Clinical Outcome and Injection Compliance of Scilin” [SEAS]).

Study design

Patients who were treated with SciLin human insulin were considered eligible for the study based on their physicians' clinical judgment. All SciLin insulins and oral antidiabetic drugs administered in the study were purchased by the patients from their local pharmacies. All patients were diagnosed with type 2 diabetes according to the 1999 World Health Organization (WHO) criteria. Patients were excluded if they were pregnant, breastfeeding, or wished to conceive within the next 3 months, were receiving concomitant therapy with other types of insulin, or were participating in other clinical trials of antidiabetic medications. Patients visited their medical centers twice during the trial—at baseline and after 12 weeks of insulin therapy (final visit).

All procedures performed in this real-world, observational study were consistent with routine clinical practice. Patient data collected at the baseline visit included demographic and anthropometric data (gender, age, height, weight, and body mass index [BMI]), medical history (type and duration of diabetes and insulin dosage), and clinical data (blood pressure within 4 weeks before recruitment, glycated hemoglobin (HbA1c) at 4 weeks before recruitment, and fasting plasma glucose [FPG] and postprandial blood glucose [PBG] levels at 1 week before recruitment). Data collected at the final visit after 12 weeks of insulin treatment included body weight, current insulin therapy, the most recent HbA1c measurement during the previous 4 weeks, the most recent FPG measurement during the previous week, blood pressure during the last 4 weeks, and all serious adverse events, adverse drug reactions (ADRs), and pregnancies. Hypoglycemia was defined as blood glucose levels <70 mg/dL (3.9 mmol/L). Severe hypoglycemia was defined as confirmed symptoms consistent with hypoglycemia, the subjects cannot manage low blood sugar themselves.

The 2683 patients enrolled in the study were divided into two groups based on the grade of hospital they were treated at: 812 were treated at second-grade hospitals and 1871 at third-grade hospitals. In China, hospitals were divided into three grades according to the beds, staffs, and the abilities in clinical work and laboratory research of the hospital. First-grade hospitals refer to the clinics in the town and provide basic health services. Second-grade hospitals refer to the city and county district hospital, providing comprehensive health services to a number of communities, performing some teaching, and undertaking some scientific research. Third-grade hospitals refers to the national, provincial hospital and medical school-affiliated hospitals, providing a high level of specialized medical and health services to several regions and implementing high levels of education and scientific research. Doctors in first-grade hospitals are general practitioners. They seldom prescribe insulin and usually refer the diabetic patients who have to inject insulin to second-grade hospital or above. So we did not enroll the first-grade hospitals in this study.

Statistical analyses

The evaluation of efficacy outcomes was undertaken in the efficacy analysis set (i.e., patients who had not used prohibited medications during the study period, whose interval between the first and the final visit ranged between 56 and 112 days, and who had valid HbA1c data at the first and last visits) and included changes from baseline in HbA1c, the percentage of patients achieving a HbA1c level <7% (<53 mmol/mol) [or 6.5% (48 mmol/mol)], and changes in fasting blood glucose (FBG) and PBG levels, insulin dosages, body weight, systolic blood pressure (SBP), diastolic blood pressure (DBP), and serum lipid concentrations. Safety outcomes, which were evaluated in the safety analysis set (i.e., all patients who received the study medications at least once during the 12-week trial period), included the incidence of severe ADRs (including severe hypoglycemic events), the overall incidence of hypoglycemia, and the incidences of daytime and nocturnal hypoglycemia. No method of imputation was applied for missing data.

Continuous variables are presented as mean and standard deviation. Categorical variables presented as numbers (%). T-tests were performed to analyze continuous variables, and paired t-tests were performed to compare changes in efficacy parameters from baseline. Chi-square or Fisher's exact probability tests were used to compare categorical variables. Univariate analysis of covariance model was used to adjust the baseline dates of PBG, DBP, low-density lipoprotein cholesterol (LDL-C), insulin dosage, and body weight. All statistical analyses were performed using SAS^® software, version 9.2 (Cary, NC). A significance level of 0.05 was applied for all tests.

Results

Baseline characteristics of the patients

The baseline characteristics of the 2683 patients treated are shown in Table 1. These patients comprised the study's safety analysis set. Clinical characteristics such as age, gender, weight, HbA1c, FBG level, SBP, and oral antidiabetic drugs at baseline were not significantly different between the groups of patients treated at second-grade and third-grade hospitals. However, as shown in Table 1, there were some statistically significant differences between the groups.

Table 1.

Baseline Characteristics of the Study Population by Hospital Group

	Hospital group
	Second-grade hospitals (n = 812)	Third-grade hospitals (n = 1871)	P
Age (years)	56.08 (11.91)	55.15 (12.25)	0.0671
Male sex	432 (53.20%)	1037 (55.42%)	0.2879
Body weight (kg)	68.19 (10.91)	67.45 (11.56)	0.1197
BMI (kg/m²)	24.89 (3.28)	24.53 (3.36)	0.0099^a
Diabetes diagnosed before	672 (82.86%)	1479 (79.05%)	0.0229^a
Duration of diabetes (months)	69.56 (60.63)	77.49 (71.71)	0.0130^a
Oral antidiabetic drugs
Metformin	487 (59.98%)	1186 (63.39%)	0.0937
α-Glucosidase inhibitors	325 (40.02%)	702 (37.52%)	0.2201
Sulfonylureas	193 (23.77%)	314 (16.78%)	0.1188
HbA1c (%) [mmol/mol]	9.75 (2.46) [83 (26.9)]	9.66 (2.05) [82 (22.4)]	0.3508
Blood glucose (mmol/L)
FBG	11.76 (4.22)	11.52 (4.03)	0.1841
PBG-breakfast	16.47 (4.94)	15.40 (4.79)	<0.0001^a
PBG-lunch	15.77 (4.65)	14.81 (4.53)	0.0001^a
PBG-dinner	14.99 (4.47)	14.35 (4.60)	0.0123^a
SBP (mmHg)	129.58 (13.66)	130.39 (14.72)	0.1839
DBP (mmHg)	79.06 (8.13)	80.78 (9.77)	<0.0001^a
LDL-C (mmol/L)	2.94 (0.99)	3.07 (1.03)	0.0051^a
HDL-C (mmol/L)	1.25 (0.46)	1.29 (0.56)	0.0775
Total cholesterol (mmol/L)	5.14 (1.27)	5.16 (1.33)	0.7335
Triglycerides (mmol/L)	2.18 (1.77)	2.30 (2.17)	0.1686

Data are presented as mean (±SD) or as numbers of patients (with percentages), and were tested by paired t-tests.

Indicates a statistically significant difference between patients treated at second-grade and third-grade hospitals (P < 0.05).

BMI, body mass index; DBP, diastolic blood pressure; FBG, fasting blood glucose; HbA1c, glycated hemoglobin; HDL-C, high-density lipoprotein cholesterol; LDL-C, low-density lipoprotein cholesterol; PBG, postprandial blood glucose; PBG-breakfast, 2-h post-breakfast blood glucose; PBG-lunch, 2-h post-lunch blood glucose; PBG-dinner, 2-h post-dinner blood glucose; SBP, systolic blood pressure.

Efficacy evaluation

HbA1c

The proportion of patients achieving the target HbA1c level of <7.0% (<53 mmol/mol) was significantly higher in patients treated at second-grade hospitals than in those treated at third-grade hospitals (46.94% vs. 38.85%, respectively; P < 0.001) (Table 2). The mean HbA1c level decreased from 9.75% (83 mmol/mol) at baseline to 7.07% (54 mmol/mol) at week 12 in patients treated at second-grade hospital and from 9.66% (82 mmol/mol) at baseline to 7.48% (58 mmol/mol) at week 12 in patients treated at third-grade hospitals. The mean change in HbA1c was significantly greater in the former group (−2.68% [−29.3 mmol/mol] vs. −2.18% [−23.8 mmol/mol], respectively; P < 0.001) (Table 3).

Table 2.

HbA1c Success Rates After 12 Weeks of Treatment

	Second-grade hospitals	Third-grade hospitals	Chi-square	P
HbA1c			13.9	0.0002
<7% [<53 mmol/mol]	46.94%	38.85%
≥7% [≥53 mmol/mol]	53.06%	61.15%

Differences in percentages of patients with HbA1c <7% (<53 mmol/mol) between the second-grade and third-grade hospital treatment groups after 12 weeks of insulin therapy.

Table 3.

Comparison of Efficacy Parameters Before and After Treatment Between Patients Treated at Second-Grade and Third-Grade Hospitals (Efficacy Analysis Set; n = 2683)

Parameter	Second-grade hospitals	Third-grade hospitals	P
HbA1c (%) [mmol/mol]
Before treatment	9.75 (2.46) [83 (26.9)]	9.66 (2.05) [82 (22.4)]	0.3508
After treatment	7.07 (1.13) [54 (12.4)]	7.48 (1.33) [58 (14.5)]	<0.001^a
Change at week 12	−2.68 (2.34) [−29.3 (25.6)]	−2.18 (1.83) [−23.8 (20.0)]	<0.001^a
FBG (mmol/L)
Before treatment	11.76 (4.22)	11.52 (4.03)	0.1841
After treatment	7.07 (1.13)	7.48 (1.33)	<0.0001^a
Change at week 12	−4.92 (4.49)	−4.09 (3.92)	<0.0001^a
PBG-breakfast (mmol/L)
Before treatment	16.47 (4.94)	15.40 (4.79)	<0.0001^a
After treatment	8.99 (1.99)	9.62 (2.41)	<0.0001^a
Change at week 12	−7.41 (5.24)	−5.74 (4.87)	<0.0001^a,b
PBG-lunch (mmol/L)
Before treatment	15.77 (4.65)	14.81 (4.53)	0.0001^a
After treatment	8.91 (1.80)	9.66 (2.34)	<0.0001^a
Change at week 12	−6.80 (4.90)	−5.20 (4.62)	<0.0001^a,b
PBG-dinner (mmol/L)
Before treatment	14.99 (4.47)	14.35 (4.60)	0.0123^a
After treatment	8.75 (1.75)	9.63 (2.54)	<0.0001^a
Change at week 12	−5.84 (4.63)	−4.91 (4.71)	<0.0001^a,b
SBP (mmHg)
Before treatment	129.67 (13.57)	130.41 (14.37)	0.2404
After treatment	125.72 (9.47)	126.71 (9.38)	0.05
Change at week 12	−4.04 (11.56)	−3.65 (12.38)	0.4678
DBP (mmHg)
Before treatment	79.12 (8.11)	80.56 (9.56)	0.0002^a
After treatment	76.55 (6.48)	78.78 (8.06)	0.0000^a
Change at week 12	−2.58 (8.15)	−1.88 (8.81)	0.2177^b
LDL-C (mmol/L)
Before treatment	2.94 (0.99)	3.07 (1.03)	0.0051^a
After treatment	2.68 (0.82)	2.77 (0.90)	0.0325^a
Change at week 12	−0.26 (0.95)	−0.35 (0.92)	0.0827^b
HDL-C (mmol/L)
Before treatment	1.25 (0.46)	1.29 (0.56)	0.0775
After treatment	1.34 (0.44)	1.38 (0.54)	0.0934
Change at week 12	0.10 (0.46)	0.05 (0.47)	0.0308^a
Total cholesterol (mmol/L)
Before treatment	5.14 (1.27)	5.16 (1.33)	0.7335
After treatment	4.70 (0.99)	4.63 (1.05)	0.1613
Change	−0.44 (1.24)	−0.56 (1.20)	0.0484^a
Triglycerides (mmol/L)
Before treatment	2.18 (1.77)	2.30 (2.17)	0.1686
After treatment	1.58 (0.89)	1.79 (1.04)	0.0000^a
Change at week 12	−0.58 (1.71)	−0.54 (1.89)	0.6511
Insulin dosage (IU/kg/day)
Before treatment	0.47 (0.19)	0.42 (0.15)	<0.0001^a
After treatment	0.49 (0.18)	0.43 (0.15)	<0.0001^a
Change at week 12	0.01 (0.14)	0.01 (0.11)	<0.0001^a,b
Body weight (kg)
Before treatment	68.45 (10.85)	67.26 (11.45)	0.0164^a
After treatment	68.74 (10.43)	67.29 (11.11)	0.0025^a
Change at week 12	0.29 (2.16)	0.04 (2.12)	<0.0001^a,b

Data are presented as mean (±SD).

Indicates statistically significant differences between patients treated at second-grade and third-grade hospitals (P < 0.05).

After adjusted baseline dates.

FBG and PBG

Differences between baseline and week 12 values for FBG and PBG concentrations are shown in Table 3. FBG and PBG values decreased in both treatment groups. Mean FBG levels decreased from 11.76 ± 4.22 mmol/L at baseline to 7.07 ± 1.13 mmol/L at week 12 in patients treated at second-grade hospitals and from 11.52 ± 4.03 to 7.48 ± 1.33 mmol/L in patients treated at third-grade hospitals. The mean change in FBG was significantly greater in the former group (Table 3). Similarly, mean changes in breakfast, lunch, and dinner PBG levels were also significantly greater in patients treated at second-grade hospitals than in those treated at third-grade hospitals after adjusting the baseline level of PBG (all P < 0.001) (Table 3).

Serum lipids

Among all treated patients, the LDL-C, total cholesterol, and triglyceride levels at week 12 were significantly lower than those measured at baseline, while the high-density lipoprotein cholesterol level at week 12 was significantly higher. In the comparison of patients treated at second-grade and third-grade hospitals, most of the changes from baseline were not significantly different between the two groups (Table 3).

Blood pressure, body weight, and insulin dosages

Differences in blood pressure, body weight, and insulin dosages between baseline and week 12 in the two treatment groups are also shown in Table 3. The mean changes in SBP and DBP from baseline were not significantly different between the two groups. Although changes in body weight during the study period were small, there was a significant difference between the two groups, with patients treated at second-grade hospitals showing a significantly greater mean weight gain (0.29 kg vs. 0.04 kg, respectively; P < 0.0001) (Table 3).

After 12 weeks of treatment, patients who were treated at second-grade hospitals were receiving significantly greater mean daily dosages of insulin than those treated at third-grade hospitals (0.49 IU/kg/day vs. 0.43 IU/kg/day, respectively; P < 0.0001), and the changes from baseline were significantly different between the two groups (Table 3).

Safety evaluation (hypoglycemic events)

Differences in the incidences of hypoglycemic events between patients treated at second-grade and third-grade hospitals were statistically significant (Table 4). The total incidence of hypoglycemic events during the study was 21.82% in patients treated at second-grade hospitals and 16.79% in patients treated at third-grade hospitals. The incidences of hypoglycemic events occurring during the day and at night in the two groups were 17.02% and 12.98%, respectively, and 8.83% and 6.35%, respectively.

Table 4.

Comparison of Hypoglycemic Events Inpatients Treated at Second-Grade and Third-Grade Hospitals (Safety Analysis Set; n = 2683)

	Second-grade hospitals (%)	Third-grade hospitals (%)	Chi-square	P
Total hypoglycemia			13.9	0.0002
Yes	21.82	16.79
No	78.18	83.21
Hypoglycemia during the day			7.41	0.0065
Yes	17.02	12.98
No	82.98	87.02
Hypoglycemia at night			5.13	0.0235
Yes	8.83	6.35
No	91.17	93.65

Discussion

Although China has the largest number of patients with diabetes globally, little is known about the management of patients with diabetes in Chinese hospitals. In this study, hospitals were divided into different grades according to the “hospital classification management method” in China. Third-class hospitals are the highest level in terms of medical treatment, teaching, and scientific research ability, which equates to the highest medical care and technical and efficiency levels. Second-grade hospitals are the next level below third-grade hospitals.

The results of the SEAS study reflect the true status of insulin therapy for type 2 diabetes in routine clinical practice in China. Human insulin therapy significantly improved blood glucose control in Chinese patients with type 2 diabetes. After 12 weeks of insulin therapy, between 38.85% and 46.94% of patients treated at second-grade and third-grade hospitals achieved the HbA1c goal level [<7% (<53 mmol/mol)]. However, we were surprised to find that blood glucose was better controlled in second-grade hospitals than in third-grade hospitals. Consistent with the use of higher insulin dosages, the incidence of hypoglycemia was higher in patients treated at second-grade hospitals than in those treated at third-grade hospitals.

Timely initiation of insulin therapy relieves the load on islet β cells, rapidly improves hyperglycemia, attenuates hyperglycemia-induced toxicity, improves insulin resistance, and protects or even reverses residual β cell function.⁸ The 2013 Chinese Guideline for the Prevention and Therapy of Type 2 Diabetes Mellitus recommends initiation of insulin therapy when the HbA1c remains ≥7.0% (≥53 mmol/mol) after combination therapy with oral antidiabetic drugs at relatively high dosages.⁴ After 12 weeks of insulin therapy with various SciLin regimens (i.e., SciLin N, SciLin R, and SciLinM30 alone or in combination) in this study, the mean HbA1c level decreased from 9.75% (83 mmol/mol) at baseline to 7.07% (54 mmol/mol) at week 12 in patients treated at second-grade hospitals, and from 9.66% (82 mmol/mol) to 7.48% (58 mmol/mol) in patients treated at third-grade hospitals. The mean change in HbA1c was −2.68% (−29.3 mmol/mol) in the former group and −2.18% (−23.8 mmol/mol) in the latter. This HbA1c result was comparable to that reported in the multicenter A₁chieve observational study, which evaluated the therapeutic efficacy of insulin analogs in patients with type 2 diabetes, who began or were changed to insulin therapy.⁹ After 24 weeks of treatment, the mean HbA1c level of patients in this study was reduced from 9.5% (80 mmol/mol) at baseline to 7.4% (57 mmol/mol), representing a −2.2% (−24.0 mmol/mol) decrease (P < 0.0001).

The finding that patients treated at second-grade hospitals had better HbA1c control in comparison with those treated at third-grade hospitals is contrary to what we expected. Possible reasons for this finding are, first, that physicians in third-grade hospitals, who may have a better understanding of diabetes, might be more reluctant to raise the dosage of insulin to avoid adverse effects such as hypoglycemia and weight gain. It is well known that the higher the degree of control of blood glucose, the more likely it is that hypoglycemia will occur. As physicians in third-grade hospitals may be more likely to take the view that hypoglycemia is more dangerous than hyperglycemia, they may prescribe lower dosages of insulin to avoid hypoglycemia. On the contrary, doctors in second-grade hospitals, most of whom are internal medicine physicians, may not be as well informed about the harms of hypoglycemia. In this study, mean insulin dosages after 12 weeks of treatment were higher in patients treated at second-grade hospitals than in those treated at third-grade hospitals (0.49 IU/kg/day vs. 0.43 IU/kg/day, respectively) and, as a result, the incidence of hypoglycemia was higher in the former group (21.82% vs. 16.79%, respectively). Similarly, the mean weight gain in patients treated at second-grade hospitals was also higher (0.29 kg vs. 0.04 kg, respectively). Although we did not conduct a survey of the views of physicians at the different grades of hospitals, a study of physicians' attitudes and beliefs toward insulin therapy in patients with type 2 diabetes in Middle Eastern countries found that 41.8% of physicians were reluctant to initiate insulin therapy for patients ≥75 years of age because of the risk of hypoglycemia, and 22.1% were concerned about the risks associated with insulin therapy (hypoglycemia, weight gain).¹⁰

A second possible reason for the finding of better HbA1c control in patients treated at second-grade hospitals in comparison with those treated in third-grade hospitals is that patient adherence to treatment is related to glycemic control. In a recent study, it was found that there is an association between treatment adherence and HbA1c. Even after adjusting for baseline HbA1c, each 1-point increase in the baseline Morisky medication adherence total score (a higher score for which equates to less adherence) was associated with a 0.16% increase in the HbA1c level measured 6 months later. In addition, baseline acceptance of forgetting to take medication was associated with a 0.43% (4.7 mmol/mol) increase in the 6-month HbA1c level.¹¹ Patients in third-grade hospitals mostly live in big cities and may have high-pressure lives, which may lower medication adherence. The true reason needs to be explored.

Blood pressure and serum lipid concentrations showed decreases in patients treated at both grades of hospitals. Those treated at third-grade hospitals had greater decreases of LDL-C, which is regarded as an index of arterial sclerosis. This suggests that the physicians at third-grade hospitals may be more aware that dyslipidemia is more important than glucose control for prevention of macrovascular complications.

Overall, the findings of this real-world study indicate that insulin therapy significantly improves glycemic control and patient care in Chinese patients with type 2 diabetes, and that clinical practice in different grades of hospital in China may be different due to the heterogeneity of culture, economics, and physician education levels. As it is difficult to balance the benefits of antidiabetic therapy with its adverse effects (notably hypoglycemia and weight gain), the individualized treatment approach for patients advocated by the American Diabetes Association may be helpful. Based on parameters such as BMI, HbA1c, and insulin dosages, a specific treatment regimen can be determined for each individual patient.⁵ In doing so, attention also needs to be paid to improving patients' adherence to treatment. In this regard, physicians and patients both play important roles in determining the treatment regimen. The results of this large-scale, observational, real-world study reflect the status of insulin therapy in Chinese patients with type 2 diabetes and provide valuable evidence of the status of routine clinical therapy for the disease.

Limitations

As this was a multicenter, real-world, observational study, there was no control patient population, which does not provide the same robust evidence as a randomized, controlled trial. However, data from this study provide a basis for investigating routine insulin therapy in Chinese patients with type 2 diabetes in controlled clinical trials.

Footnotes

Acknowledgements

This study was sponsored by Bayer Healthcare Co. Ltd. The sponsor's involvement was restricted to development of the protocol, data collection and analysis, and medical writing support.

Author Disclosure Statement

No competing financial interests exist.

Appendix 1. SEAS Site List

References

Bhattarai

: Three patterns of rising type 2 diabetes prevalence in the world: need to widen the concept of prevention in individuals into control in the community. J Nepal Med Assoc, 2009; 48:173–179.

Yang

, Lu

, Weng

, et al.: Prevalence of diabetes among men and women in China. N Engl J Med, 2010; 362:1090–1101.

, Wang

, He

, et al.: Prevalence and control of diabetes in Chinese adults. JAMA, 2013; 310:948–959.

Chinese Diabetes Society: Guideline for the prevention and therapy of type 2 diabetes mellitus in China (2013). Chin J Diabetes Mellitus, 2014; 6:447–498.

American Diabetes Association. Standards of medical care in diabetes—2006. Diabetes Care, 2006; 29(Suppl 1):S4–S42.

Yang

, Gao

, Liu

, et al.: Biphasic insulin aspart 30 as insulin initial or replacement therapy: the China cohort of the IMPROVE study. Curr Med Res Opin, 2010; 26:101–107.

, Ji

, Guo

, Gu

: Glycemic and HbA1c controls of T2DM patients treated with human insulin in China. Chin J Diabetes, 2013; 9(21):803–806.

Owens

. Clinical evidence for the earlier initiation of insulin therapy in type 2 diabetes. Diabetes Technol Ther, 2013; 15:776–785.

Home

, Naggar

, Kamseh

, et al.: An observational non-interventional study of people with diabetes beginning or changed to insulin analogue therapy in non-Western countries: the A₁chieve study. Diabetes Res Clin Pract, 2011; 94:352–363.

10.

Lakkis

, Maalouf

, Mahmassani

, Hamadeh

: Insulin therapy attitudes and beliefs of physicians in Middle Eastern Arab countries. Fam Pract, 2013; 30:560–567.

11.

Aikens

, Piette

: Longitudinal association between medication adherence and glycaemic control in type 2 diabetes. Diabet Med, 2013; 30:338–344.