Impact of a Community Health Worker–Led Diabetes Education Program on Hospital and Emergency Department Utilization and Costs

Diabetes

Kathryn L. Schmidt, MPH, Ashley W. Collinsworth, ScD, MPH, Sunni A. Barnes, PhD, Rachel M. Brown, BA, Erin P. Kane, MD, Christine A. Snead, BSN, RN, CPHQ, and Neil S. Fleming, PhD

From Baylor Scott & White Health.

 

ABSTRACT

Objective: To assess the impact of a community health worker–led diabetes education program (DEP) on hospital utilization trends and to evaluate the return on investment.

Methods: A retrospective, pre-post study design was used to examine differences in inpatient and emergency department (ED) utilization and costs for DEP patients in the year prior to and following enrollment. Patients with diabetes who received care at the same clinics but who were not enrolled in DEP served as a control group. Analysis of covariance was used to test for differences in utilization outcomes between DEP patients and controls, controlling for age, sex, and ethnicity.

Results: DEP patients had a significant reduction in mean inpatient encounters (0.08 vs. 0.18), LOS per inpatient encounter (0.28 vs. 0.67), and inpatient cost per patient ($406 vs. $902) following DEP enrollment. Patients in the control group also experienced a significant reduction in inpatient LOS and costs. Neither group experienced a significant difference in ED utilization or costs. Return on investment for DEP was –66%, as the annual cost savings generated per patient from reduced utilization ($137.19) were less than the annual DEP costs (investment) per patient ($402.80).

Conclusion: CHW-led diabetes education programs like DEP may provide additional benefits than can be gained from access to primary care alone in terms of avoidance of costly hospitalizations for diabetes-related complications. Although the DEP did not generate an overall cost savings for the health care system in the short term, additional savings may be generated in the long term through reductions of diabetes-related complications.

 

Diabetes poses a significant burden on the population of the United States in terms of morbidity, mortality, and costs of care. Currently, 29.1 million people in the United States have diabetes [1], and it is expected that 1 in 3 people could have diabetes by the year 2050 [2]. The economic impact of diabetes is also significant; as of 2012, 1 in 5 health care dollars went towards diabetes care [3], and the total cost of diabetes was estimated to be $245 billion [1]. This rapid increase in the incidence of diabetes and associated costs emphasizes the need for cost-effective strategies to prevent and manage diabetes within the U.S. population.

One strategy that has shown success in improving diabetes care and outcomes is the use of community health workers (CHWs) to deliver disease education and management programs. A CHW is a front-line public health worker who is often a trusted member of the community and has the capacity to influence patient understanding, enhance patient compliance, and promote more equitable practices in patient care management [4–6]. Several studies have shown that CHW-led interventions in diabetes management can help patients achieve control over their disease and improved health indicators such as HbA1c, blood pressure, lipid levels, and weight [7–17]. In addition, a few studies have shown that CHW-led interventions for patients with diabetes can help these patients avoid costly health care utilization in the form of emergency department (ED) visits and hospitalizations. Fedder et al found that diabetes patients on Medicaid who worked closely with CHWs experienced a decrease in ED visits (38%), hospitalizations (30%), and hospital admissions from ED (53%) [11]. This CHW intervention was associated with improved patient quality of life and a cost-savings of $2245 per patient per year for a total savings of $262,080 for 117 patients. More recently, a 2010 randomized controlled trial showed that African-American patients with diabetes who worked with a nurse case manager and a CHW were 23% less likely to make ED visits than those who just had a nurse case manager [18].

Despite the documented successes of CHW programs, the CHW model has not been widely adopted within integrated health care systems. A major barrier to adoption of CHW programs has been the lack of sustained funding for CHW services [17,19]. Historically, many CHW programs were supported by grants, as most payers were unwilling to fund these initiatives [4,17,19.] In 2008, the Centers for Medicare and Medicaid Services provided a mechanism to support CHW activities by approving a Medicaid state plan amendment authorizing payment for CHWs who worked under Medicaid-approved providers, such as physicians and nurses [19]. However, the lack of data available regarding the costs, cost-effectiveness, and potential costs savings of CHW programs continues to serve as a barrier to adoption [13,20].

Baylor Scott & White Health North (BSWH), formerly Baylor Health Care System in Dallas, Texas, created the CHW-led Diabetes Equity Project, a 5-year program supported with funding from Merck Foundation’s Alliance to Reduce Disparities in Diabetes, with the goal of reducing observed disparities in diabetes care and outcomes in the medically underserved, predominantly Hispanic communities surrounding BSWH hospitals [16,21,22]. The program featured specially trained, bilingual CHWs who served as members of primary care teams in 5 community clinics and delivered a culturally relevant diabetes self-management and education curriculum (DSME) targeting barriers to diabetes management commonly experienced by Hispanics. The objective of this study was to assess the impact of a CHW-led diabetes educucation program (DEP) on hospital utilization trends and to evaluate the return on investment of the program.

 

METHODS

Setting

BSWH is one of the largest nonprofit health care systems in the United States and includes 46 hospitals, > 800 patient care sites, > 6000 affiliated physicians, 35,000 employees, and an accountable care organization. This study was conducted in 5 community clinics located in the Dallas metroplex surrounding BSWH North hospitals. The community clinics serve low-income, uninsured, and chronically ill patients. The study was approved by the Baylor Research Institute institutional review board.

 

DEP Intervention

The DEP program consisted of 2 initial 60-minute educational sessions and quarterly clinical assessments scheduled for 30 to 60 minutes for a maximum of 6 patient-contact hours over 12 consecutive months. The DSME curriculum for DEP was adapted from CoDE, a pilot program implemented in a Dallas clinic serving a largely uninsured Mexican American population [23]. Patients who participated in CoDE for 12 months experienced a significant reduction in HbA1c [23,24]. During the 2 educational sessions, the CHWs educated DEP participants about diabetes and the importance of blood glucose control, medication adherence, diet, and exercise. In addition to the educational sessions, CHWs performed quarterly clinical assessments of HbA1c, blood pressure, weight, and foot condition (visual and monofilament assessment). They also assessed self-management behaviors and facilitated goal setting at each visit. The CHWs documented patient visits in the electronic health record and contacted the patient’s primary care provider immediately if the patient was symptomatic or had critical blood glucose or blood pressure measurements as defined by program protocol.

 

Participants

Study participants were recruited from the clinics or referred by Baylor care coordinators following hospital visits related to uncontrolled diabetes from September 2009 to July 2013. Participants had to be 18 years or older with a diagnosis of type 2 diabetes and be uninsured or underinsured. Although the program targeted Hispanic patients, all patients who met the inclusion criteria were eligible to participate. To control for internal threats to validity such as history and maturation biases, we created a control group consisting of clinic patients who had a diagnosis of diabetes and met the DEP inclusion criteria but who did not enroll in DEP.

 

Assessment

We used a retrospective pre-post study design to assess the impact of the DEP on hospital utilization trends and the program’s return on investment (ROI). The primary outcomes were number of hospital encounters, length of stay (LOS) per encounter, and direct cost per patient. The pre-enrollment period was defined as the year priod to the day of the initial DEP visit (the date of enrollment for DEP patients) or the initial clinic visit (the date of “enrollment” for control patients). All study participants were included in the analysis regardless of their number of inpatient or ED encounters. If a participant did not have a documented encounter during the analysis time period, encounters, LOS, and medical costs were set to zero. Patients with at least 2 HbA1c measurements taken during the first year post enrollment were included in the analysis.

 

Data Sources

Inpatient and emergency department (ED) utilization data were obtained from the Dallas Fort-Worth Hospital Council (DFWHC) database. The DFWHC captures administrative data from over 80 participating hospital systems and 9 million unique patients in North Texas. The DFWHC applied a matching algorithm using first and last name and date of birth to match study participants with encounter and length of stay detail for all hospitalizations across all DFWHC member hospitals. We obtained direct medical costs for patients treated at BSWH facilities from the BSWH Trendstar administrative database. Direct medical cost was not available for encounters at non-BSWH facilities so cost was estimated for these patients based on BSWH costs using a prediction model that accounted for LOS, primary ICD-9 diagnosis, patient age, sex, and race. The model explained approximately 66% of the variation in direct costs (R2 = 0.6581).

 

Statistical Analysis

All analyses were performed in SAS V9.3 using an α level of 0.05. A 2-tailed independent samples t test was used to test the mean differences in utilization outcomes one year prior to and post program enrollment. An analysis of covariance (ANCOVA) was used to assess whether the mean change in utilization outcomes was greater for DEP patients than the control group, , after controlling for age, sex, and ethnicity. The gamma distribution with a log link function was used to model direct cost and the negative binomial distribution was used to model hospital encounters and LOS.

The ROI calculation included 2 components: DEP investment cost and risk-adjusted direct medical cost savings 1 year post program enrollment. DEP investment cost included the average yearly costs per community health worker, the fixed one-time start-up costs distributed across the length of the program (4 years), and the salaries of the 5 community health workers employed for the duration of the program. These costs were divided by the number of patients who enrolled in the DEP to calculate the per patient cost (investment). Direct medical cost savings were calculated as the mean reduction in hospitalization and ED costs per patient adjusted for age, sex, and ethnicity.

To account for DEP participants without encounter data, we estimated average cost savings in a 4-step process using a propensity score adjustment approach. We stratified encounters based on admission type (inpatient vs. ED) and combined the average cost savings calculated for both admission types to calculate mean total medical cost savings.

In the first step, direct cost was modeled on DEP participants with an encounter to obtain parameter estimates for pre- and post- program enrollment, age, sex, and ethnicity. The model was based on the log-link function with the gamma distribution. In the second step, we ran a propensity score logistic regression model to estimate the probability of any encounter being identified in the DFWHC database, adjusted for age, sex, ethnicity, and DEP clinic. In the third step, we generated 10,000 bootstrap samples, with replacement, to estimate the DEP population’s median expected change in cost. In the bootstrap sample, the parameter estimates from step 1 were used to predict the cost pre- and post- program enrollment for each patient. The expected change in cost per participant was then calculated, while adjusting for the propensity to have an encounter, estimated in step 2. The formula for calculating change in cost was Expected change in cost = [pre probability*pre cost – post probability*post cost].

In the final step, we retained the median cost savings from each bootstrap sample. Inpatient and ED cost savings were combined to produce a total cost savings across the patient population. The ROI formula was Financial gain (utilization cost savings – DEP investment)/DEP investment.

 

OR_Diabetes_Table1RESULTS

Participant Demographics

Of the 1140 study participants, 878 were DEP patients and 262 were controls (Table 1). The majority of DEP patients were Hispanic females older than 40 years while the majority of the the control group was non-Hispanic males older than 40. Table 2 and Table 3 display the top 5 primary ICD-9 diagnoses by intervention group and pre/post enrollment status. Type 2 diabetes was the top diagnosis among DEP participants and controls with a pre-enrollment encounter. OR_Diabetes_Table2Acute pancreatitis was the top diagnosis among DEP participants and controls with a post-enrollment encounter.

 

Inpatient Encounters

Fourteen percent of the DEP participants and 37% of control group patients matched to an inpatient record during the pre-enrollment time period. The number of mean inpatient encounters for DEP participants decreased from 0.18 to 0.08 (P < 0.001) in the post period OR_Diabetes_Table3(Table 4). Patients in the control group also experienced a reduction in mean inpatient encounters (0.66 to 0.52), but this reduction was not significant. Both DEP and control group patients had a significant reduction in mean LOS in the post period. LOS decreased from 0.67 days to 0.28 days (P < 0.001) per encounter for DEP patients and from 2.32 days to 1.13 days (P = 0.001) for controls. However, the mean percentage change in LOS was not significantly different between the 2 study groups (P = 0.90). DEP patients and control patients also OR_Diabetes_Table4had a significant reduction in mean inpatient cost per patient, decreasing from $902 to $406 (P < 0.001) for DEP patients and from $3054 to $1688 for controls (P = 0.006). Observed reductions in cost were not significantly different between the 2 groups (P = 0.93).

 

Emergency Department Encounters

Eight percent of the DEP participants and 30% of controls matched to an ED record for the year prior to program enrollment. Neither DEP patients nor patients in the control group had a significant reduction in mean ED encounters (P = 0.88 and 0.74, respectively) or costs (P = 0.76 and 0.12, respectively) post enrollment.

 

OR_Diabetes_Table5ROI

The cost savings per DEP patient are shown in Table 5. The annual cost for a CHW to educate 1 DEP patient was $403. The combined inpatient and ED cost savings for DEP patients post-program enrollment was $137 per patient. ROI for the DEP was –66%, indicating that DEP investment costs were greater than the savings achieved through the reduction of inpatient and ED costs for DEP patients.

 

DISCUSSION

In our examination of the impact of the DEP on inpatient and ED utilization, we found that DEP patients experienced a significant decrease in inpatient visits, LOS, and direct costs in the year following DEP enrollment. In comparison, a control group of patients who were treated at the same clinics as DEP patients also experienced significant decreases in inpatient LOS and direct costs. No significant differences in ED visits, LOS, or direct costs were observed for either DEP patients or control patients in the post period.

The reduction in inpatient visits and LOS for DEP patients in the year following DEP enrollment indicates that the DEP helped patients achieve improved health and avoid costly hospitalizations. The control group did experience greater reductions in inpatient utilization, LOS, and direct costs. However, because this was not designed as a controlled trial, we utilized a nonequivalent control group and inherent differences between the DEP and control patients and utilization patterns made it difficult to draw unbiased comparisons between the groups. For instance, the number of inpatient encounters, LOS, and costs were 2 to 3 times higher in the pre-period for the control group. The mean number of inpatient encounters for DEP patients prior to DEP participation was 0.18, and the smaller change observed in utilization for DEP patients is likely due to a floor effect.

Despite the observed reductions for both DEP patients and the control group in inpatient utilization, neither group had significant reductions in ED visits or costs per patient in the post- period. The majority of ED use in the pre-period may have been due to diabetes-related complications that are difficult to prevent even with improved diabetes care or for emergencies not related to diabetes, as we included all ED admissions regardless of admitting diagnosis. In addition, similar to observed trends in inpatient utilization, ED use in the pre-period was relatively low for DEP patients (0.16), and the lack of observed changes in ED utilization for DEP patients was also likely due to a floor effect.

The DEP generated a negative ROI (–66%) in the short term as the annual cost savings generated per patient from reduced utilization ($137) were less than the annual DEP costs (investment) per patient ($403). This finding is not surprising, as it is difficult to achieve cost savings for interventions designed to increase access to health care in underserved populations.[15] Although Fedder et al. observed an average savings of $2245 per patient per year in Medicaid reimbursements for a CHW-led outreach program for patients on Medicaid with diabetes, patients who participated in this program had much higher inpatient encounters (0.95 vs. 0.18) and ED utilization (1.49 vs. 0.16) at baseline compared to DEP patients. DEP patients may not have been as sick as these patients or have been more reluctant to seek medical care due to their lack of insurance. In addition, Fedder et al did not factor in the costs of the CHW program in the cost savings calculation. However, these costs were likely to be much lower than the costs of the DEP, as the program relied on volunteer CHWs instead of paid CHWs who were also certified medical assistants.

Several studies have evaluated the cost-effectiveness of CHW-led diabetes management programs rather than the ROI as these types of programs are often associated with improved health outcomes but also increased health care costs as the result of expanding health care access and services to underserved populations [15,25–29]. Most of these evaluations modeled the long-term cost-effectiveness, as the majority of cost savings from diabetes management programs are likely to accrue in the long run as a result of the prevention of diabetes-related complications such as amputations, blindness, kidney failure, coronary heart disease, and stroke [15,25]. Although these CHW-led interventions for diabetes differed in scope, the incremental cost effectiveness ratios for these interventions were less than the common willingness-to-pay threshold of $50,000 per quality-adjusted life year (QALY). These studies may have underestimated the societal benefits of these programs as they did not incorporate non-medical cost savings such as those that may be attributed to gains in productivity [26].

A recent cost-effectiveness analysis of the DEP and the 4 other programs that were part of the Alliance to Reduce Disparities in Diabetes found that these programs spent an average of $975 per patient in the first year and additional $520 per patient in subsequent years to improve care for diabetes patients [15]. Based on improvements in health indicators such as HbA1c, systolic blood pressure, and total cholesterol observed for participants in Alliance programs, the incremental cost-effectiveness ratio of these programs was $23,161 per QALY given the optimistic assumption that the observed improvements in health indicators were all attributable to the interventions. DEP patients achieved a 1% average reduction in mean HbA1c and this improvement was sustained over the course of the program. The UK Prospective Diabetes Study Group found that every 1% reduction in HbA1C reduces a patient’s risk of developing eye, kidney, and nerve disease by 40% and the risk of heart attack by 14% [30]. Thus, if DEP patients sustain improvements gained as a result of program participation, they may avoid serious and expensive medical complications in the future.

Ultimately, the goal of the DEP was to provide patients with improved access to diabetes care and to help them achieve improved glucose control. Improving access to chronic disease care is costly. However, employment of CHWs is a less costly alternative to employing additional clinicians, and CHWs may be more effective in assisting patients with chronic disease management particularly those from underserved and vulnerable populations. Although we did not generate a positive ROI for the DEP in this short-term analysis, the program is likely cost-effective when the low cost of the program is compared with the improvements in health outcomes we have observed.

This study had several limitations. There were observed differences in gender and ethnicity between the intervention and control groups and it is likely that risk-adjustment (including propensity scoring) did not fully accountfor underlying differences between the populations. The observed reductions in inpatient utilization and costs in the intervention group may have been due to other factors besides the DEP as the control group also achieved reductions in inpatient LOS and costs. The control group did contain a few outliers in terms of high pre-period medical costs which were retained in the analysis and these outliers may have caused reductions in costs for this group to be overstated. The observed outcomes may also be biased due to intervention contamination. Patients in the control group attended the same clinics as DEP patients and were treated by the same primary care physicians. The primary care physicians likely applied the knowledge gained from working with DEP patients and the CHWs, including how to identify and help patients address the specific barriers to diabetes self-management faced by clinic patients, to the treatment of patients who were not enrolled in the DEP. The fact that health care utilization and costs were much higher for patients in the control group in the pre-period indicates that these patients had greater severity of illness and additional comorbidities, and the observed reductions in utilization and costs for these patients may have been the result of obtaining access to a regular source of primary care through the clinics. However, the observed decrease in inpatient encounters from .18 to .08 in the post period for DEP patients as well as the observed decreases in inpatient LOS and direct costs indicate that the DEP may provide additional benefits compared to access to primary care alone in terms of avoidance costly hospitalizations for diabetes-related complications.

 

Conclusion

From the health care system perspective, CHW-led diabetes education programs like the DEP may provide additional benefits than can be gained from access to primary care alone in terms of avoidance of costly hospitalizations for diabetes-related complications. Although the costs of the DEP were greater than the savings it generated through reduced inpatient utilization and costs in the short term, additional savings to the health care system or society may be generated in the long term through reductions of diabetes-related complications in patients who were able to achieve improved glycemic control through program participation. More importantly, the improvements in glycemic control achieved by DEP patients can lead to both short and long term gains in overall health and quality of life.

 

Corresponding author: Ashley Collinsworth, ScD, MPH, Scott & White Health, Center for Clinical Effectiveness, Dallas, TX 75206, Ashley.Collinsworth@baylorhealth.edu.

Funding/support: This program/initiative was supported by a grant from the Merck Company Foundation through its Merck Alliance to Reduce Disparities in Diabetes program.

 

References

1. Centers for Disease Control and Prevention, National Diabetes Statistics Report: estimates of diabetes and its burden in the United States, 2014. US Department of Health and Human Services: Atlanta, GA.

2. Centers for Disease Control and Prevention, Diabetes Report Card 2012. US Department of Health and Human Services: Atlanta, GA.

3. American Diabetes Association. Economic costs of diabetes in the U.S. in 2012. Diabetes Care 2013;36:1033–46.

4. Balcazar H, Rosenthal EL, Brownstein JN, et al. Community health workers can be a public health force for change in the United States: three actions for a new paradigm. Am J Public Health 2011;101:2199–203.

5. Rosenthal EL, Brownstein JN, Rush CH, et al. Community health workers: part of the solution. Health Aff (Millwood) 2010;29:1338–42.

6. Viswanathan M, Kraschnewski JL, Nishikawa B, et al. Outcomes and costs of community health worker interventions: a systematic review. Med Care 2010;48:792–808.

7. Babamoto KS, Sey KA, Camilleri AJ, et al. Improving diabetes care and health measures among hispanics using community health workers: results from a randomized controlled trial. Health Educ Behav 2009;36:113–26.

8. Brown SA, Garcia AA, Kouzekanani K, Hanis CL. Culturally competent diabetes self-management education for Mexican Americans: the Starr County border health initiative. Diabetes Care 2002;25:259–68.

9. Prezio EA, Cheng D, Balasubramanian BA, et al. Community Diabetes Education (CoDE) for uninsured Mexican Americans: a randomized controlled trial of a culturally tailored diabetes education and management program led by a community health worker. Diabetes Res Clin Pract 2013;100:19–28.

10. Spencer MS, Rosland AM, Kieffer EC, et al. Effectiveness of a community health worker intervention among African American and Latino adults with type 2 diabetes: a randomized controlled trial. Am J Public Health 2011;101:2253–60.

11. Fedder DO, Chang RJ, Curry S, Nichols G. The effectiveness of a community health worker outreach program on healthcare utilization of west Baltimore City Medicaid patients with diabetes, with or without hypertension. Ethn Dis 2003;13:22–7.

12. Lorig KR, Ritter P, Stewart AL, et al. Chronic disease self-management program: 2-year health status and health care utilization outcomes. Med Care 2001;39:1217–23.

13. Whitley EM, Everhart RM, Wright RA. Measuring return on investment of outreach by community health workers. J Health Care Poor Underserved 2006;17(1 Suppl):6–15.

14. Skelly AH. Culturally tailored intervention for African Americans with type 2 diabetes administered by a nurse case manager and community health worker reduces emergency room visits. Evid Based Nurs 2010;13:51–2.

15. Lewis MA, Bann CM, Karns SA, et al. Cross-site evaluation of the Alliance to Reduce Disparities in Diabetes: clinical and patient-report
ed outcomes. Health Promot Pract 2014;15(2 Suppl):92S–102S.

16. Collinsworth AW, Vulimiri M, Schmidt KL, Snead CA. Effectiveness of a community health worker-led diabetes self-management education program and implications for CHW involvement in care coordination strategies. Diabetes Educ 2013;39:792–9.

17. Shah M, Kaselitz E, Heisler M. The role of community health workers in diabetes: update on current literature. Curr Diab Rep 2013;13:163–71.

18. Gary TL, Batts-Turner M, Yeh HC, et al. The effects of a nurse case manager and a community health worker team on diabetic control, emergency department visits, and hospitalizations among urban African Americans with type 2 diabetes mellitus: a randomized controlled trial. Arch Intern Med 2009;169:1788–94.

19. Martinez J, Ro M, Villa NW, et al. Transforming the delivery of care in the post-health reform era: what role will community health workers play? Am J Public Health 2011;101:e1–5.

20. Rush CH. Return on investment from employment of community health workers. J Ambul Care Manage 2012;35:133–7.

21. Collinsworth A, Vulimiri M, Snead C, Walton J. Community health workers in primary care practice: redesigning health care delivery systems to extend and improve diabetes care in underserved populations. Health Promot Pract 2014;15(2 Suppl):51S–61S.

22. Walton JW, Snead CA, Collinsworth AW, Schmidt KL. Reducing diabetes disparities through the implementation of a community health worker-led diabetes self-management education program. Fam Community Health 2012;35:161–71.

23. Culica D, Walton JW, Prezio EA. CoDE: Community Diabetes Education for uninsured Mexican Americans. Proc (Bayl Univ Med Cent) 2007;20:111–7.

24. Culica D, Walton JW, Harker K, Prezio EA. Effectiveness of a community health worker as sole diabetes educator: comparison of CoDE with similar culturally appropriate interventions. J Health Care Poor Underserved 2008;19:1076–95.

25. Brownson CA, Hoerger TJ, Fisher EB, Kilpatrick KE. Cost-effectiveness of diabetes self-management programs in community primary care settings. Diabetes Educ 2009;35:761–9.

26. Gilmer TP, Roze S, Valentine WJ, et al. Cost-effectiveness of diabetes case management for low-income populations. Health Serv Res 2007;42:1943–59.

27. Brown HS 3rd, Wilson KJ, Pagán JA, et al. Cost-effectiveness analysis of a community health worker intervention for low-income Hispanic adults with diabetes. Prev Chronic Dis 2012;9:E140.

28. Ryabov I. Cost-effectiveness of community health workers in controlling diabetes epidemic on the US-Mexico border. Public Health 2014;128:636–42.

29. Prezio EA, Pagán JA, Shuval K, Culica D. The Community Diabetes Education (CoDE) program: cost-effectiveness and health outcomes. Am J Prev Med 2014;47:771–9.

30. Intensive blood-glucose control with sulphonylureas or insulin compared with conventional treatment and risk of complications in patients with type 2 diabetes (UKPDS 33). UK Prospective Diabetes Study (UKPDS) Group. Lancet 1998;352:837–53.

Comments are closed.

« »