Currently Viewing:
The American Journal of Managed Care January 2013
Closing the Personalized Medicine Information Gap: HER2 Test Documentation Practice
Ilia L. Ferrusi, PhD; Craig C. Earle, MD; Maureen Trudeau, MD; Natasha B. Leighl, MD; Eleanor Pullenayegum, PhD; Hoa Khong, MD; Jeffrey S. Hoch, PhD; and Deborah A. Marshall, PhD
Predicting Asthma Outcomes in Commercially Insured and Medicaid Populations
Richard H. Stanford, PharmD, MS; Manan B. Shah, PharmD, PhD; Anna O. D'Souza, BPharm, PhD; and Michael Schatz, MD, MS
Inside Out: Reversing the Focus on Emergency Departments to Enhance Efficiency
Adam Sharp, MD, MS; and A. Mark Fendrick, MD
Questioning the Widely Publicized Savings
Robert Cosway, FSA, MAAA; L. Allen Dobson, Jr, MD
Providers' Perspective on Diabetes Case Management: A Descriptive Study
Nacide Ercan-Fang, MD; Kiranjot Gujral, MD; Nancy Greer, PhD; and Areef Ishani, MD, MS
Can Targeted Messaging Encourage PCP Contact Before ED Visits?
Maria C. Raven, MD, MPH, MSc; Scott M. Kotchko, MA; and David A. Gould, PhD
Emergency Department Visits for Nonurgent Conditions: Systematic Literature Review
Lori Uscher-Pines, PhD, MSc; Jesse Pines, MD, MBA; Arthur Kellermann, MD, MPH; Emily Gillen, MA; and Ateev Mehrotra, MD, MS
Process, Cost, and Clinical Quality: The Initial Oral Contraceptive Visit
Michael J. McMullen, MD; Samuel W. Woolford, PhD; Charles L. Moore, MBA; and Barry M. Berger, MD
Emergency Department Nonurgent Visits in Iran: Prevalence and Associated Factors
Mohammad Jalili, MD; Farzaneh Shirani, MD; Mohamad Hosseininejad, MD; and Hossein Asl-e-Soleimani, MD
Embracing a Diversified Future for US Primary Care
Timothy Hoff, PhD
Currently Reading
Full Coverage for Hypertension Drugs in Rural Communities in China
Baorong Yu, PhD; Xiaojuan Zhang, MS; and Guijing Wang, PhD

Full Coverage for Hypertension Drugs in Rural Communities in China

Baorong Yu, PhD; Xiaojuan Zhang, MS; and Guijing Wang, PhD
Low-income rural residents receiving hypertension drugs at no charge had enhanced medication adherence and reduced medical costs in China.
Background: The control rate for hypertension is unacceptably low worldwide, and poor adherence to medication is a primary reason.

Objectives: To evaluate the impact of full coverage for hypertension drugs on adherence to medication, medical costs, and hypertension control in Shandong Province, China.

Methods: In November 2009, we interviewed 110 hypertensive patients who had been participating in a free medication program since May 2008 and 241 hypertensive patients who were not participating. We used a 1:1 propensity-score matching technique to obtain matched samples of 102 program participants (intervention) and 102 nonparticipants (control). We used univariate analysis to compare patient drug-taking behaviors, medical costs, and hypertension control between the 2 groups.

Results: All intervention patients took >1 drugs for hypertension control and 93% of them took >3 such drugs, 15 control patients (15%) did not take any, and only 39% took 3 or more (P <.001). Three-fourths (75%) of the intervention patients took the prescribed drugs regularly, whereas 66% of the control group (P = .034) did so. Participation in the program was associated with lower annual out-of-pocket medical costs both overall and for outpatient services (P <.001 for both).

Conclusions: Low-income rural residents in China receiving free drugs had enhanced medication adherence and reduced total medical costs. Providing hypertension drugs at no charge may be a promising strategy for preventing costly cardiovascular events associated with hypertension in China and other parts of the world with growing rates of cardiovascular disease.

(Am J Manag Care. 2013;19(1):e22-e29)
In both developed and developing countries, the control rate for hypertension is unacceptably low, and poor adherence to medication is a primary reason.

  •  A free medication program is effective in improving adherence and reducing total medical costs in rural communities in China.

  • The impact of the program on blood pressure levels and disease outcomes requires further investigation.
Throughout the world, hypertension is the number one attributable risk factor for death, and its prevalence is increasing.1-4 A report for 2001 estimated that worldwide, 14% of deaths and 6% of lost disability-adjusted-life-years were due to hypertension.5 Although the developed countries account for the bulk of economic costs of hypertension, this disorder is responsible for a large and increasing economic and health burden in developing countries, and we can expect that an ever-increasing proportion of the costs is going to be borne by those countries.6,7 For example, it has been predicted that the Asia Pacific region will bear about half of the world’s cardiovascular burden, and high blood pressure is an important determinant of cardiovascular disease.8 In China, the prevalence of hypertension has also been increasing over the past decade and in recent years reached as high as 39% among adults over age 35 years. Not surprisingly, the majority (62%) of strokes and about half (49%) of the cases of ischemic heart disease were attributable to hypertension.9,10 Moreover, morbidity and mortality due to these 2 chronic conditions have been increasing rapidly in China in recent decades.11,12 Given recent trends in cardiovascular disease in China, hypertension control has become a public health priority in that nation.

Fortunately, many effective pharmacologic treatments have become available for outpatient control of high blood pressure. Inpatient care for hypertension and its complications is expensive, but many hospitalizations may be avoided with the use of antihypertensive drugs.13 The challenge is that patient adherence to antihypertensive medications remains poor in both developed and developing countries, and the lack of adherence to prescribed regimens constitutes a barrier to adequate blood pressure control and the prevention of cardiovascular events.14-18 In addition, some researchers have demonstrated that poor adherence to antihypertensive medications is associated with a higher risk of vascular events and hospitalization, and greater healthcare costs.16-18 An increased level of adherence to antihypertensive agents should result in a better health outcome for patients; therefore, providing a medication benefit may be a cost-effective policy option for improving population health.13,17,18 Many researchers have investigated strategies to improve patients’ adherence to medications in a cost-effective manner.19-23

Financial problems such as a lack of health insurance frequently affect patients’ ability to control hypertension because the medications are unaffordable for low-income, rural residents in both developed and developing countries.13,19,23 In 2 US studies, lower copayments for hypertension drugs were associated with improved adherence.20, 22 However, there is limited evidence on the relationship between reducing the cost of hypertension medications and adherence to these medications, although studies have documented the growing problem of hypertension in rural China.24-26 To fill in this gap, we evaluated the effectiveness of a medication subsidy program in rural communities in Shandong Province, China, by conducting and analyzing a face-to-face interview survey of hypertensive patients. The effectiveness was measured by patients’ medication adherence status, economic burden of disease, and blood pressure levels.


In 2007, the health department of Shandong Province developed an initiative, Rural Health Care Workers Training and Non-communicable Disease Control Programs, which was supported by the government of Luxembourg through collaborations with the World Health Organization. The initiative focused primarily on (1) training rural healthcare workers in chronic disease control and prevention and (2) improving rural residents’ awareness of health risk factors and promoting a healthy lifestyle. Eight rural counties were selected to be the project counties. A baseline survey, which elicited information from families and also included a medical examination and laboratory testing, was conducted in the project counties from April 13 to May 1, 2007. This survey revealed that many patients with hypertension did not know the adverse health impacts of high blood pressure, and even if they had hypertension, they were not taking any antihypertensive drugs. These patients were more likely to have a low income and less likely to be able to afford antihypertensive drugs or obtain standard treatment. Overall, the rates of hypertension treatment and control were unacceptably low; thus, a program to subsidize the cost of medication was implemented in May 2008 in selected villages of all 8 counties.


The subsidy program provided basic medicines at no charge for hypertensive patients in low-income families, as well as standard treatment and management as needed. The basic medication regimen, the NAH (nifedipine, atenolol, hydrochlorothiazide) program, was recommended by the China National Hypertension League. More generally, the program had 3 components: (1) At the initial clinic visit, the patient received a standard dose of medication according to blood pressure level, age, and sex (eg, nifedipine 2.5 mg 3 times daily, atenolol 6.25 mg once daily, or hydrochlorothiazide 6.25 mg once daily). The patient was asked to have a follow-up visit in 3 to 5 days. (2) At follow-up, the drug regimen was adjusted if needed according to blood pressure control status and drug complications. (3) If the 3 drugs could not bring the blood pressure under control, angiotensin-converting enzyme inhibitors, angiotensin II receptor antagonists, angiotensin II receptor blockers, and other drugs were added as appropriate. The patients received the hypertension drugs at no charge from the clinicians, and the clinicians were reimbursed 150 Chinese yuan per year for every patient served by the Rural Healthcare Workers Training and Non-communicable Disease Control Programs. (The 150 yuan was usually enough to cover the costs of hypertension drugs for each patient for a year in the rural communities.) The 3 main outcomes of interest were patient’s medication adherence status, the economic burden of disease, and the blood pressure level. Medication adherence was determined by observing the patient’s medication-taking behavior. The patient’s economic burden of disease was total medical costs of chronic diseases associated with hypertension, including inpatient, outpatient, and medication costs.


In November 2009, about 18 months after the subsidy program began, a field survey in 2 townships in Cangshan County, 1 of the intervention counties, was conducted by trained researchers and graduate students from School of Public Health, Shandong University. In both townships, 2 villages were randomly selected as the project intervention sites. All hypertensive patients living in the villages were included. The survey involved holding face-to-face interviews with 110 hypertensive patients who received free hypertension drugs in the 4 villages (the intervention group) and 241 hypertensive patients of similar economic status who did not participate in the free medication program in 7 adjacent villages (the control group). A predesigned questionnaire covered major sociodemographic variables, number of family members with high blood pressure, health complications, time since diagnosis, knowledge about hypertension, behaviors (following the doctor’s advice and taking medication), and inpatient and outpatient medical expenses. The survey was designed to evaluate the effectiveness of a medication subsidy for low-income rural patients with noncommunicable diseases. The survey contained 15 forms that were developed and pilot-tested by a research team at Shandong University. We only analyzed the survey items related to hypertension (Appendix).

Blood pressure measurements were obtained from clinic technician records from 2009 (a year after the subsidy program began). Both systolic and diastolic blood pressures were measured twice in a 3- to 5-day interval. If either the systolic or diastolic measure differed by 6 or more mm Hg between the 2 measurements, a third measurement was taken. The average of the 2 or 3 measurements was recorded as the blood pressure level. Medical cost was collected from health insurance agencies and confirmed or supplemented by the face-toface interview.

We used a 1:1 propensity-score matching technique to obtain patients for the intervention and control groups based on the information from our survey.27-30 Because the intervention and control groups were not randomly selected, the distribution of characteristics among patients in the intervention group was quite different from that among patients in the control group, although we tried to select the patients with similar economic status in both groups (Table 1). It was possible that the distribution of their demographic characteristics (eg, age, sex, education) or of other factors could have biased our results when we compared the 2 groups. Thus, to minimize the potential bias in the analysis, we derived the matched samples using a logistic regression model based on patients’ age, sex, education, marital status, economic status, health status, and insurance participation.27 Then, using the matched sample populations, we compared medication-taking behavior, medical costs, and blood pressure control status.

We used the χ2 test for categorical variables and the Wilcoxon 2-sample test for continuous variables. All tests of statistical significance were 2-tailed, and because our study samples were small, we considered P <.05 to indicate a significant difference. The SAS statistical package (Version 9.1, SAS, Cary, North Carolina) was used for all of the analyses.


Before the propensity-score matching, 5 variables were found to differ significantly by group (Table 1): education, per capita income, health complications, time since patient was first diagnosed with hypertension, and knowledge of the relationship between hypertension and diet. These differences, if left uncorrected, would have biased the comparison analysis of medication-taking behaviors, medical costs, and blood pressure levels. After propensity-score matching, the 5 significant distribution differences in the population characteristics (Table 1) were greatly reduced, and only 1 of these characteristics, health complications (P <.05), differed significantly between the intervention and control groups (Table 2). Thus, the 2 groups were now comparable and could be used for the comparison analysis.

Patients who received a medication subsidy were more likely to take some prescribed drugs for hypertension control than were unsubsidized patients (P <.001; Table 3). In the intervention group, every patient took at least 1 prescribed drug for hypertension and 93.1% took 3 or more such drugs. In contrast, 14.7% of control group patients took no prescribed hypertension drugs at all and only 39.2% took 3 or more of these drugs.

Copyright AJMC 2006-2020 Clinical Care Targeted Communications Group, LLC. All Rights Reserved.
Welcome the the new and improved, the premier managed market network. Tell us about yourself so that we can serve you better.
Sign Up