Research Article Vol:3,Issue:2
Waris Qidwai, Nabeea Junaid
To determine the time to first insulinization in patients with type 2 diabetes mellitus (T2DM) seeking out-patient consultation in Pakistan. In this observational multicenter cross-sectional study conducted between April 2015 and October 2016 in outpatient settings, we analyzed data collected from patients with T2DM in whom the investigator decided to initiate insulin therapy. Data collection included patients’ diabetes history, prior treatments, reasons for commencing insulin treatment, type and dose of insulin prescribed, and support provided to patients towards insulinization. Data analysis was done using descriptive statistics. Mean [standard deviation (SD)] time to insulinization in the 758 patients analyzed was 7.8 (5.2) years. Failure to attain target HbA1c and fasting blood glucose were the most common reason for commencing insulin, reported in 623 (82.2%) and 579 (76.4%) patients, respectively. A total of 510 (67.3%) patients presented diabetic complications at enrolment, mostly as sensory neuropathy (n=391, 51.6%), retinopathy (n=200, 26.4%), and microalbuminuria (n=110, 14.5%). Insulin types prescribed were premix (n=490, 64.9%), basal insulin (n=217, 28.6%), and prandial (n=24, 3.2%) for average (SD) daily doses of 29.2 (12.9) U, 15.3 (8.3) U, and 18.9 (10.8) U, respectively. Basal + Prandial insulin was prescribed in 2.6% (n=20) patients. Support towards insulinization – counselling, training, dose titration, self-monitoring of blood glucose – were largely provided by healthcare practitioners. In Pakistan, insulin therapy is substantially delayed in patients with T2DM despite consensus guidelines. Appropriate measures are required at the outpatient setting to ensure adherence to guidelines for timely insulinization in the management of T2DM.
Key Words: Insulinization, type 2 diabetes mellitus, Pakistan, outpatient.
SAJMED.2018; 3(2):91-101 PDF Download
Type 2 diabetes mellitus (T2DM) has transformed into a major global health concern, affecting an estimated 425 million people world-wide1 and is the sixth leading cause of death2. The prevalence of T2DM is growing more rapidly in low-to-middle income countries in Asia, The Middle East, and Africa than in developed countries3, thereby, affecting healthcare resources and infrastructure4,5. With approximately 7.5 million T2DM patients, a national age-standardized prevalence of 8.3%, and an estimated 8.0% of overall mortality associated with diabetes1. Pakistan is one of the top countries in Asia in terms of burden of T2DM.
Various international6-9 and local10,11 guidelines have been established to assist in the clinical management of T2DM. In general, therapeutic intervention in T2DM depends largely on the stage of the disease and severity of diabetic complications at the time of diagnosis. Patients with early-stage diabetes or pre-diabetes are usually recommended lifestyle and dietary modifications in addition to metformin, sometimes, as the first line of treatment. However, T2DM is a progressive disease and the increasing deterioration of b-cell function in many patients eventually needs correction by administration of exogenous insulin. Besides a remarkable effectiveness in attaining glycemic control, exogenous insulin also confers other proven benefits – protection against endothelial dysfunction, delay in onset of micro- and macro-vascular diabetic complications, salvation of and improvement in b-cell function, reduction of cardiovascular risk, and so on12. Based on a vast body of evidence collected over time, almost all international guidelines for T2DM management recommend timely and intensive insulinization, especially when targeted glycemic indices of glycosylated hemoglobin (HbA1c) and/or fasting blood glucose (FBG) have not been attained following 3 months of a combination of >2 oral antidiabetic (OADs) agents at their maximally tolerated doses6,9.
Despite knowledge of these established benefits of insulin, the major challenge in real-world clinical practice is identifying the time to initiate insulin regimen. In the Asian context, various factors such as clinical inertia, patient’s concerns about hypoglycemia and weight gain, socioeconomic factors, disease awareness, and physicians’ knowledge of guidelines have an impact on time of insulinization in patients13. Results from the First Basal Insulin Evaluation (FINE Asia) study, conducted across 11 countries in Asia including Pakistan, showed that at the time of commencement of basal insulin, study patients had already been diagnosed with T2DM for an average of 9.3 years (Pakistan: 8.6 years), had mean HbA1c values of 9.8% (Pakistan: 10.1%), and had been receiving OAD treatment for an average of 8.7 years (Pakistan: 7.8 years)14. Further analysis of the FINE Asia database also showed that for countries where patients had T2DM for >9 years prior to insulin initiation, the reductions in HbA1c and the proportion of patients attaining their HbA1c and FBG targets were the lowest15. Hence, besides disease progression, a delay in insulinization can also lead to exacerbation of diabetic complications, disability, lowered quality-of-life, and mortality14.
The key obstacle to customizing clinical management of T2DM in Pakistan is the lack of relevant data, especially with regards to real-world translation of updated guidelines, on insulinization in Pakistani patients; hence, the main goal of our study was to fill this knowledge gap. The primary objective of our study was to document the time to first insulinization since diagnosis in patients with type 2 diabetes attending outpatient clinics in Pakistan. We also aimed to ascertain the choice and starting dose of insulin as well as the reasons for initiating insulin. Additionally, we also sought to determine the strategies employed for insulinization, including patient education, dose titration, and blood glucose monitoring.
In this national, multicenter, cross-sectional, observational study, data was compiled from out-patient settings in Pakistan between April 2015 and October 2016. Study investigators were randomly selected from a list of qualified physicians available in the database of the diabetes portfolio of sanofi aventis Pakistan limited and invited to participate in the study. Qualified physicians were either general practitioners or family physicians with a private practice who treated patients with T2DM and had the potential to initiate insulin therapy. Being cross-sectional by nature, our study involved a single visit at recruitment with no follow-up visit. This study was performed according to the principles laid down by the 18th World Medical Assembly (Helsinki, 1964) and in compliance with the guidelines for Good Epidemiological Practice and approval from IRBs/IECs. All study patients provided informed consent to participate in the study.
Study patients were recruited consecutively, five at each study site, after they had fulfilled the eligibility criteria and were willing to provide informed consent. In order to be included in the study, patients had to be ≥18 years of age, with pre-existent T2DM. The investigator decided to initiate insulin therapy in agreement with the respective patient. Patients who were diagnosed with T1DM, had reservations to initiate insulin therapy or were on prior insulinization, hospitalized, or pregnant, were excluded from the study. Patient data was collected for anthropometrics; socioeconomic profile; history of diabetes, including time since diagnosis and previous treatments; prior hypoglycemic events and prevalence of micro- and macrovascular complications; reasons for commencing insulin treatment and the starting dose of insulin; parameters of insulinization including patient education and support; and other treatments prescribed at enrolment.
We calculated the sample size assuming that 44.0% of the patients would have experienced secondary failure to oral therapy six years after diagnosis of T2DM and would hence be considered as candidates for commencing insulin. After factoring in a confidence level of 95% and a margin of error of 3.6%, the sample size required to address the primary objective of this study was 731 to account for missing information and misplaced forms, we planned to recruit a total of 800 patients from 160 sites across Pakistan. We analyzed data from all patients who met the eligibility criteria and for whom all responses were documented in the case report forms and used descriptive statistics to characterize the study patients. Continuous variables such as time to insulinization, starting dose of insulin, demographics, etc. were described using mean and standard deviation (SD) while categorical variables such as HbA1c levels, reasons for insulin initiation, and parameters of insulinization were described using frequencies and percentages.
A total of 760 patients from 137 sites were recruited, and data from 758 patients were available for the final analysis. The summary of patient characteristics at enrolment is presented in Table 1. Overall, there was near equal representation of men and women (49.2% males and 50.8% females) in the study with an average (SD) age of 50.6 (10.6) years. The majority of patients (n=636, 83.9%) were urban residents and almost half (n=341, 45.0%) were either illiterate or had only primary education. Mean (SD) height, weight, body mass index (BMI), and waist circumference (WC) were 162.1 (11.7) cm, 75.4 (15.3) kg, 28.8 (6.2) kg/m2, 95.7 (12.1) cm, respectively. Obesity (defined as BMI >25 mg/kg2) was reported in 555 (73.2%) patients and the proportion of males with WC >90 cm and females with WC > 80 cm was 73.2% (n=273) and 92.9% (n=358), respectively (Table 1). Mean (SD) systolic and diastolic blood pressure were 135.4 (15.8) mmHg and 85.6 (8.9) mmHg, respectively. Mean (SD) time since diagnosis of T2DM in the study patients was 7.8 (5.2) years (Table 1). A total of 473 (67.4%) patients reported a family history of T2DM. Mean (SD) HbA1c and FBG at enrolment were 9.8% (2.1%) and 208.3 (79.9) mg/dL, respectively. A total of 722 (95.3%) and 547 (72.2%) patients had HbA1c ≥6.5% and FBG ≥ 126 mg/dL, respectively (Table 1). Diabetic complications were reported in 510 (67.3%) patients. Among microvascular complications, the prevalence was the highest for sensory neuropathy (n=391, 51.6%), followed by retinopathy (n=200, 26.4%), and microalbuminuria (n=110, 14.5%). Macrovascular complications – angina, peripheral vascular disease, and myocardial infarction/acute coronary syndrome – were reported in 59 (7.8%), 45 (5.9%), and 24 (3.2%) patients, respectively (Table 1).
Prior to insulinization, majority of the patients (n=738, 97.4%) with T2DM were being managed by pharmacotherapy through the use of oral antidiabetics (OADs). A total of 584 (77.0%) patients reported taking single agent OAD in the past. Likewise, a total of 535 (70.4%) and 168 (22.2%) patients had been prescribed 2 OADs or >2OADs, respectively. The average (SD) duration of taking single-agent OAD, 2 OADs, and >2 OADs was 37.5 (34.9), 45.9 (38.7), and 27.9 (26.7) months, respectively. The most common single-agent OADs prescribed were biguanides (n=415, 54.7%) and sulfonylureas (n=156, 20.6%); the most common 2 OAD regimens were biguanides + sulfonylureas (n=466, 61.5%) and biguanides + dipeptidyl peptidase 4 (DPP-4) inhibitors (n=34, 4.5%); the most common >2 OAD regimens were biguanides + sulfonylureas + DPP-4 inhibitors (n=137, 18.1%) and biguanides + sulfonylureas + thiazolidinediones (n=26, 3.4%). Dietary restrictions and physical exercise had been prescribed in 646 (85.2%) and 524 (69.1%) patients, respectively, and both had been prescribed in 513 (67.7%) patients. A total of 105 (13.9%) patients had been taking alternative medicines for T2DM management. Prior hypoglycemic experience in the month preceding enrolment was reported by 100 (13.2%) patients. Of these the majority had experienced either one (n=50, 50.0%) or two (n=30, 30.0%) hypoglycemic episodes. In the opinion of study investigators, the principal reason for initiation of insulin was patients’ inability to achieve target HbA1c, reported for 623 (82.2%) patients. The inability to achieve target FBG (n=579, 76.4%), as well as target postprandial plasma glucose level (n=425, 56.1%) were also some of the main reasons for insulinization of study patients (Table 2).
Premix insulin was the most commonly prescribed insulin (n=490, 64.9%) and the premix insulin of choice was premixed human insulin (n=432, 56.9%) (Table 3). Mean (SD) starting dose of premix insulin was 29.2 (12.9) U administered through an average of 2.2 injections per day. Basal insulin was prescribed in 217 (28.6%) patients of whom a majority received long-acting insulin analog (n=161, 21.2%). Mean (SD) starting dose of basal insulin was 15.3 (8.3) U administered via an average of 1.2 injections per day. Prandial insulin was commenced in 24 (3.2%) patients with a mean (SD) dose of 18.9 (10.8) U administered through an average of 2.5 injections per day (Table 3). The most common insulin combination reported was basal + prandial (n=20, 2.6%). Other treatments prescribed concomitant to insulin were diet (n=792, 92.6%), exercise (n=654, 86.3%), and OADs (n=522, 68.8%). Basal insulin + biguanides was the most frequently prescribed insulin + OAD combination (n=54, 7.1%).
For the majority of patients, the medical professional responsible for initiating insulin; providing counselling; imparting training on insulin delivery devices, insulin dose titration, self-monitoring of blood glucose and other factors of T2DM management was the healthcare practitioner (HCP). To a lesser extent, diabetes educators and nurses were also reported to be involved in these procedures. Likewise, for a large proportion of patients, clinics were the medical facility where insulinization support was most comprehensively provided. The most common parameters for patient education were T2DM and insulin (n=729, 96.2%), counselling on initiation of insulin (n=713, 94.1%), and diet modification (n=694, 91.6%) (Table 4).
In this study, our goal was to get an overview of the real-world practice of diabetes management in Pakistan by assessing the time to insulinization in patients with T2DM consulting in outpatient settings. We noticed a slight reduction in time to commencing insulin treatment after first diagnosis in this study than the Pakistan cohort of the FINE Asia study (7.8 years vs. 8.6 years).15 However, this was not an optimal timeframe since average values of HbA1c and FBG were substantially high at the time of insulinization (9.8% and 208.3 mg/dL, respectively). Moreover, a large proportion of patients in this study had already developed diabetic complications, indicating that disease progression was well underway. We also noticed that premix insulin was the choice of insulin for investigators while initiating insulin treatment and the most common reasons for insulinization were failure to achieve glycemic targets for HbA1c, FBG, and prandial plasma glucose levels.
One of the most crucial aspect of insulinization is to decide on the time to commence insulin therapy. All major international guidelines as well as local recommendations in Pakistan provide detailed algorithms to assist physicians in identifying the correct time for insulinization.6-11 In general, pharmacologic intervention in T2DM management depends primarily on glycemic indices at the time of diagnosis. Recent guidelines in Pakistan recommend the use of metformin and lifestyle management in patients with HbA1c <8.0% or FBG <180 mg/dL with follow-ups at three-monthly intervals to adjust treatment in line with glycemic control.11 In patients whose glycemic indices do not reduce sufficiently, additional agents are added to the treatment regimen. When this strategy fails to achieve the glycemic target, insulin therapy is commenced. In certain cases where HbA1c levels are already substantially high at diagnosis, it is recommended to commence insulin therapy promptly. In recent years, there is growing support for the evidence-based concept of ‘timely (early) insulinization’ in order to salvage ß-cell function for as long as possible and to harvest the other benefits of physiologic levels of insulin.12,16,17 However, in real-world, the conditions differ substantially from the recommendations. Initiation and intensification of insulin therapy is delayed substantially14,15 thereby exposing the patients to a higher glycemic burden and an increased risk of developing diabetic complications. Patients with T2DM in Pakistan have been reported to have suboptimal glycemic control and more severe microvascular complications than those who had optimal glycemic control.18 Our findings corroborate previous reports; at the time of insulinization, our patients had been diagnosed with T2DM for almost 8 years, had HbA1c and FBG values of 9.8% and 208.3mg/dl, respectively; and over two-thirds of the patients had diabetic complications. While the present study was not designed to identify the reasons for delay in commencing insulin therapy in the patients, we could contextualize some explanations based on previous studies. Patient and physician barriers to insulinization, based on perceptions and fears, as well as clinical inertia in initiating and intensifying insulin regimens, are known to result in either delay in insulinization or suboptimal insulinization.13, 19-23 In a socio-economically heterogenous country such as Pakistan, the availability of healthcare resources as well as healthcare seeking behavior in the general population differs from location to location. While patients in urban areas have access to larger, tertiary care centers, in rural areas, the primary care physician is the only available medical professional to manage patients with T2DM. This could result in a substantial disparity in physicians’ awareness of updated guidelines and treatments. Furthermore, patients’ education level has a large bearing on effective insulinization. A substantial proportion of the patients in our study were either illiterate or had primary schooling only and hence would probably be less willing to insulin treatment, and its education and awareness. In addition, we must also bear in mind that a substantial proportion of healthcare cost in Pakistan is borne by the patient. Hence, it would be economically impossible for most low-to-middle-class patients with T2DM to afford the latest insulinization devices that are simple to use and patient-friendly.
Another critical aspect of initiating insulin therapy in the patients is the choice of insulin. At present, physicians have a wide range of options, depending on desired duration of action (short-acting vs. long-acting), source (human insulin vs. analogs), physiological mimicry (premix vs. basal-bolus) and so on. Almost all international guidelines recommend basal insulin for initiating insulin therapy in patients with T2DM and local guidelines in numerous Asian countries conform to these.13 Various studies have also shown that basal insulin is as effective as premix insulin in reducing HbA1c levels and has the added advantage of lower incidence of hypoglycemia and weight gain.24-29 Despite this, the real-world scenario in Asia largely favors premix insulin as the agent of choice for commencing insulin therapy in Asian patients with T2DM.30 Similar trend was observed in our study; almost two-thirds (64.9%) of our patients were prescribed premix insulin. In Southeast Asia, the overall perception is that premix insulin is the agent of choice due to less number of daily injections, easy physician-led dose titration as well as patient-led self-titration.30 Another equally contributing factor could be the cost of treatment: premix insulin is much cheaper than basal insulin analogs. Given the absence of socialized healthcare systems or health insurance in a largely low-to-middle-income country such as Pakistan, the lower cost associated with premix insulin could be the single most important criterion in choice of insulin. In addition, a new school of thought has been proposed in recent years which questions the suitability of international guidelines, especially regarding the choice of insulin in the Asian context. Various arguments have been presented in support of this inquiry, primarily, ethnic variations31 and the dichotomous Asian T2DM phenotype,32 disparity in lifestyles and diet between Western and Asian populations,30 and the need for greater prandial glycemic control in Asians.15, 29 Whether the shift in paradigm is applicable to Pakistan or not is as yet unknown, largely due to paucity of real-world data. Thus, the only way to resolve this conundrum of choosing the appropriate insulin in clinical settings in Pakistan is through field trials designed to gather information on the effectiveness of various insulin regimens. However, a patient-centric approach needs to be emphasized upon when a decision is made to initiate insulin in a patient.
Our study brings to light a few key findings on initiation of insulin therapy in Pakistani patients from a real-world perspective. However, in order to build a bigger picture which would assist in developing country-specific guidelines for effective insulinization of T2DM patients, further data would need to be gathered and this was beyond the scope of our study. Being a cross-sectional investigation, we could not follow up on the elemental aspects of insulin initiation such as its effectiveness in achieving glycemic targets, dose titration, side effects such as hypoglycemic episodes and weight gain, and patient perception and treatment adherence. Neither could we ascertain the reasons for delay in insulinization in our study patients. It is especially crucial to identify these in order to formulate not only custom-made guidelines but also effective strategies for creating physician and patient awareness.