SEARCH

SEARCH BY CITATION

Keywords:

  • adherence;
  • antihypertensive agent;
  • chronic disease;
  • hypertension;
  • Korea

Abstract

  1. Top of page
  2. Abstract
  3. Introduction
  4. Methods
  5. Results
  6. Discussion
  7. Conclusion
  8. Acknowledgment
  9. Contributions
  10. References

Hypertension is one of the most serious health problems in Korea. The purpose of this paper is to identify factors associated with self-reported nonadherence to an antihypertensive regimen. The data were obtained from the Korean Medical Panel 2008 and 2009 database, which were surveyed by the Korea Institute for Health and Social Affairs and National Health Insurance. We analyzed 5324 patients using multivariate logistic regression models. Self-reported nonadherence was used as a dependent variable and demographic, socioeconomic, and health status were included as independent variables to investigate the associated factors. Among the patients who were diagnosed with hypertension, 13.2% did not take their medicine as prescribed. Age and education attainment level were statistically significant. Younger and lower educational attainment groups were less likely to adhere to medication regimens and showed a tendency not to take their medicine as prescribed. There were no statistically significant variables in terms of health status. Our findings suggest that nurses in clinical and community settings should pay more attention to hypertensive patients who are young and less educated.


Introduction

  1. Top of page
  2. Abstract
  3. Introduction
  4. Methods
  5. Results
  6. Discussion
  7. Conclusion
  8. Acknowledgment
  9. Contributions
  10. References

Hypertension is one of the most pressing health problems in Korea. The prevalence of hypertension is as high as 24.6% of the Korean adult population, making it the most common chronic disease (Park et al., 2010). Hypertension is a major risk factor for cardiovascular and cerebrovascular diseases, which are the second and third leading causes of mortality in Korea (Statistics Korea, 2011). People with 10 mmHg higher systolic blood pressure (BP) or 20 mmHg higher diastolic BP have approximately a 100% higher risk of stroke or coronary artery than those with normal BP (Park et al., 2010). Therefore, one of the most effective ways to prevent cardiovascular and cerebrovascular diseases could be hypertension control.

There are many ways to control BP, such as exercise, dietary, and medication therapy (Yang et al., 2010). Among them, adhering to a therapeutic regimen of medication is considered to be the most effective way to control blood pressure (Baune et al., 2005; Krousel-Wood et al., 2005; Burnier, 2006). Some evidence in Korea suggests that properly adhering to a medication regimen can prevent adverse outcomes related to hypertension. A poor adherence group showed a 2.2 times higher risk of morbidity, including instances of nephropathy, cerebrovascular disease, and cardiovascular disease (Park et al., 2010). This group was also 2.4 times more likely to be hospitalized in a year (Jang et al., 2008) compared to a group showing better adherence.

To control BP is the most widespread program implemented by nurses at community health centers in Korea. Nurses are responsible for assessing medication-taking behavior (Ben-Natan & Noselozich, 2011) and play a main role in educating hypertensive patients to take medicine appropriately. This means that nurses are required to recognize factors associated with medication adherence to help patients with hypertension.

Several studies have been conducted on the topic of antihypertensive medication adherence in Korea (Jang et al., 2008; Park et al., 2008a,b; 2010). Although there are a variety of factors related to medication adherence, numerous factors were omitted in most studies. Only one study (Park et al., 2008b) explored multiple factors through using health insurance claims data, but it did not consider socioeconomic aspects such as education and per capita income. Studies in Korea also did not identify whether patients in fact took their medicine or not due to limitations related to health insurance claims data. Therefore, this study was designed to identify factors, including demographic, socioeconomic, and health-status variables associated with self-reported nonadherence to antihypertensive regimen.

Literature review

Medication adherence refers to the extent to which a patient acts in accordance with the prescribed interval and dose of a dosing regimen. In this context, nonadherence means missing medication doses in the context of ongoing use (Cramer et al., 2008).

Adherence is determined by factors related to socioeconomic status, healthcare system, health condition, therapy, and patient-related factors (World Health Organization, 2003). Many studies consider demographic factors as distinct from socioeconomic factors (Lowry et al., 2005; Vawter et al., 2008; Braverman & Dedier, 2009; Friedman et al., 2010).

Gender, age, ethnicity, and residential area were included as demographic factors related to medication adherence in several studies. A few studies reported that it was more probable for males than females to show adherence (Degli Esposti et al., 2002; Park et al., 2008b), while females showed greater adherence to antihypertensive medication regimens in other studies (Jokisalo et al., 2002; Hyre et al., 2007; Friedman et al., 2010). Old age is associated with better adherence in studies conducted in Pakistan (Hashmi et al., 2007), Korea (Park et al., 2008b), the USA (Hyre et al., 2007) and Finland (Jokisalo et al., 2002). Persistence and adherence were both lower in urban residents compared with rural residents in Canada (Friedman et al., 2010), whereas metropolitan residents had higher adherence levels than rural residents in Korea (Park et al., 2008b).

Income, job type, insurance type, and education level as socioeconomic factors affect medication adherence as well. Adherence is increased in patients with higher incomes in Canada (Friedman et al., 2010), and groups composed of those in lower economic classes showed lower adherence rates in Portugal (Santa-Helena et al., 2010) and in the USA (Vawter et al., 2008). Persons working in the unskilled labor market were less likely to adhere to an antihypertensive medication regimen (Santa-Helena et al., 2010). Individuals with low education attainment levels tend to report unintentional nonadherence (Lowry et al., 2005; Uzun et al., 2009).

A good relationship between patient and provider as a factor of the healthcare system improves medication adherence because healthcare professionals empower patients to become involved in their treatment (Fincham, 2007). As for factors of health condition, persons with disabilities, especially mobility and communication disabilities (Park et al., 2008a), depressive symptoms (Morris et al., 2006; Krousel-Wood et al., 2011), and mental function impairments (Vawter et al., 2008) showed inappropriate medication adherence behavior compared to those without such disabilities. In contrast, several studies found that medication adherence was better in patients with comorbidity as compared to those not showing comorbidity (Lagi et al., 2006; Shaya et al., 2009; Friedman et al., 2010).

With regard to factors of therapy, patients prescribed with angiotensin-converting enzyme inhibitors showed better adherence than those taking beta-blockers or diuretics (Fitz-Simon et al., 2005; Friedman et al., 2010), and an increase in the number of pills and the required frequency were related to nonadherence (Bangalore et al., 2007). Regarding patient-related factors, beliefs about medication were related to medication adherence (Gregoire et al., 2006; Lewis et al., 2010) and behavioral attitudes, perceived behavioral control, and subjective norms were positively related to intentions to self-administer medication in the elderly (Ben-Natan and Noselozich, 2011).

Purpose

The purpose of this study was to identify factors among demographic, socioeconomic, and health-status factors affecting nonadherence in patients with hypertension.

Methods

  1. Top of page
  2. Abstract
  3. Introduction
  4. Methods
  5. Results
  6. Discussion
  7. Conclusion
  8. Acknowledgment
  9. Contributions
  10. References

Data sources

This study used data from the Korean Medical Panel from 2008 and 2009. The Korean Medical Panel provides secondary data from that collected by the Korea Institute for Health and Social Affairs (KIHASA) and National Health Insurance (NHI). All researchers must submit research plans before given authority to use the data (Korea Institute for Health and Social Affairs & National Health Insurance, 2010). Authorization to access the data was granted by the KIHASA and NHI in 2011. The Korean Medical Panel surveyed all members of the households that were selected as representatives. The sample was randomly stratified by region from all households in Korea. The Korean Medical Panel surveyed 21 787 respondents of 7066 households in 2008 and 19 641 respondents of 6300 households in 2009 (Korea Institute for Health and Social Affairs & National Health Insurance, 2010). The survey in the first year was repeated in the second using the same questions.

The numbers of respondents who reported that they had hypertension were 2783 in 2008 and 2800 in 2009. First, we excluded patients who had not visited clinics for a diagnosis and who had missing data, such as that related to medication adherence and/or job and income status. Due to this process, 109 (3.9%) respondents in 2008 and 128 (4.6%) respondents in 2009 were excluded. Next, we excluded patients who answered “no” to the question, “Do you take medicine that lowers your blood pressure?” Fifteen in 2008 and seven in 2009 were additionally excluded. Finally, 2659 (95.5%) in 2008 and 2665 (95.2%) in 2009 were included in the final analysis. Among them, 2230 of the patients were surveyed for both years.

Variables

We defined nonadherence as to whether respondents in the analysis answered “no” to the question, “Do you take medicine according to an antihypertensive medication regimen?” The Korean Medical Panel asked patients who were diagnosed with hypertension by medical doctors and prescribed more than once.

Regarding predictors, we applied the demographic, socioeconomic, and health-related factors as surveyed by the Korean Medical Panel. The demographic variables were gender (male, female), age (< 45, 45–64, 65–74, ≥ 75), living alone (yes, no), and residence (capital area, outside the capital area). The socioeconomic variables included insurance (health insurance, medical benefit), job status (none, temporary, permanent) and average annual income (< 4000, 4000–7999, 8000–11 999, ≥ 12 000 US dollars). Average annual income was calculated by dividing the total household income by the number of family members. The type of insurance was considered to be a health-system-related factor (World Health Organization, 2003), while it also represents the socioeconomic status of the respondent. Therefore, we included insurance type as a socioeconomic factor. We included disabilities (presence, no presence) and the number of comorbidities except hypertension (0, 1–2, ≥ 3) as health-status variables. Age, income, and the number of comorbidities were originally measured as continuous variables, but ultimately we grouped them as a categorized variable.

One advantage of panel data is that it allows researchers to estimate the effects of changes of independent variables on dependent variables over time. However, we could not estimate these effects as the demographic, socioeconomic, and health status of patients typically does not change over a period of one year. For this reason, we included the survey year as a variable under the assumption that the predictors affect nonadherence independently despite the fact that many of the same people were surveyed in both 2008 and 2009.

Statistical analysis

First, we ran a chi-square test to find the differences between the adherence and nonadherence groups. Next, multivariate logistic regression models were used to identify significant factors influencing nonadherence. Adjusted odds ratios (OR) and 95% confidence intervals (CI) were estimated, and P values that were below 0.05 were considered to be statistically significant. Demographic (Model 1), socioeconomic (Model 2), and health status (Model 3) variables were entered in order to determine which characteristics predict noncompliance best. We assessed the fit of the multivariate models using c statistics at each step.

Results

  1. Top of page
  2. Abstract
  3. Introduction
  4. Methods
  5. Results
  6. Discussion
  7. Conclusion
  8. Acknowledgment
  9. Contributions
  10. References

Among the patients who were diagnosed with hypertension, 702 (13.2%) reported not taking their antihypertensive medication as prescribed. There were 364 (13.7%) in 2008 and 338 (12.7%) in 2009. Table 1 shows the characteristics of the study population. More than half the patients were female aged 65 years and over. The educational attainment of approximately 50% was below middle school, and full-time workers accounted for only 30%. The patients had a mean monthly income of US $1600. Approximately 10% had one or more disabilities, and more than 80% had other chronic diseases apart from hypertension.

Table 1. Characteristics of the study population (n = 5324)
CharacteristicsN%
Survey year
2008265949.9
2009266550.1
Gender
Male228342.9
Female304157.1
Age
< 452995.6
45–64224042.1
65–74189135.5
≥ 7589416.8
Residence
Capital area182234.2
Outside capital area350265.8
Family
Living alone66012.4
Not living alone466487.6
Insurance type
Health insurance494292.8
No health insurance3827.2
Education
Less than primary school88916.7
Primary school170131.9
Middle school87116.4
High school138526.0
Beyond high school4789.0
Job
Not economically active284653.5
Full-time workers143627.0
Temporary workers104219.6
Annual income (US dollar)
< 4 000103419.4
4 000–7 999155429.2
8 000–11 99983615.7
≥ 12 000190035.7
Number of disabilities
None468988.1
More than one63511.9
Number of chronic diseases
0103719.5
1–2244445.9
≥ 3184334.6

There were significant differences found for the three variables of age, educational attainment, and job status between the adherence and nonadherence groups (Table 2). More than 20% of those aged < 45 reported they did not take their medicine as prescribed. There was a trend toward patients those with less education reporting a higher incidence of nonadherence. The rate of nonadhering patients who were not economically active was lower than the rate of those who were.

Table 2. Percentage of self-reported nonadherence by demographic, socioeconomic, and health-status variables
CharacteristicsNNonadherence (%)P value
Survey year
200836413.70.278
200933812.7
Gender
Male28312.40.140
Female41913.8
Age
< 456722.4< 0.001
45–6430713.7
65–7421911.6
≥ 7510912.2
Residence
Capital area23212.70.482
Outside capital area47013.4
Family
Living alone9213.90.541
Not living alone61013.1
Insurance type
Health insurance64013.00.068
No health insurance6216.2
Education
Less than primary school14916.80.003
Primary school20712.2
Middle school11813.5
High school18113.1
Beyond high school479.8
Job
Not economically active34812.20.006
Full-time workers18613.0
Temporary workers16816.1
Annual income (US dollar)
< 4 00014313.80.204
4 000–7 99918411.8
8 000–11 99910612.7
≥ 12 00026914.2
Number of disabilities
None62113.20.733
More than one8112.8
Number of chronic diseases
014514.00.520
1–232613.3
≥ 323112.5

Age and gender were associated with nonadherence in the first model (Table 3). However, the statistical significance of gender disappeared after adjusting for the socioeconomic variables. Age increased the likelihood of medication adherence, but residence and whether or not a patient lived alone were not statistically significant. Furthermore, those with higher levels of educational attainment were more likely to be adherent than those with lower levels. However, there were no statistically significant differences related to insurance type, job status, or income. With regard to health status, there were no statistically significant variables in the analysis. The c statistics result for Model 1 was 0.557. Adding the socioeconomic variables produced a value for this model of 0.596, but health-status variables did not improve it any further.

Table 3. Odd ratios (OR) and 95% confidence intervals (CI) for factors associated with self-reported nonadherence
CharacteristicsModel 1Model 2Model 3
OR (95% CI)OR (95% CI)OR (95% CI)
  1. *P value <0.05.

Survey year
2008ReferenceReferenceReference
20090.92 (0.79–1.08)0.91 (0.77–1.06)0.91 (0.77–1.07)
Gender
MaleReferenceReferenceReference
Female1.19 (1.00–1.40)*0.99 (0.82–1.21)0.99 (0.82–1.21)
Age
< 45ReferenceReferenceReference
45–640.53 (0.39–0.71)*0.46 (0.34–0.63)*0.46 (0.34–0.64)*
65–740.43 (0.31–0.58)*0.34 (0.24–0.48)*0.34 (0.24–0.49)*
≥ 750.44 (0.31–0.62)*0.33 (0.22–0.50)*0.33 (0.22–0.50)*
Residence
Capital areaReferenceReferenceReference
Outside capital area1.07 (0.91–1.27)1.00 (0.84–1.19)1.00 (0.84–1.19)
Family
Living aloneReferenceReferenceReference
Not living alone0.89 (0.70–1.14)0.96 (0.74–1.24)0.96 (0.74–1.23)
Insurance type
Health insurance ReferenceReference
Not health insurance 1.26 (0.93–1.72)1.27 (0.93–1.74)
Education
Less than primary school ReferenceReference
Primary school 0.64 (0.50–0.81)*0.64 (0.50–0.81)*
Middle school 0.64 (0.48–0.86)*0.64 (0.47–0.86)*
High school 0.55 (0.41–0.74)*0.55 (0.41–0.74)*
Beyond High school 0.36 (0.23–0.54)*0.36 (0.23–0.54)*
Job
Not economically active ReferenceReference
Full-time workers 1.03 (0.82–1.29)1.03 (0.82–1.29)
Temporary workers 1.23 (0.99–1.52)1.23 (0.99–1.52)
Annual income (US dollar)
< 4 000 ReferenceReference
4 000–7 999 0.86 (0.68–1.09)0.86 (0.68–1.09)
8 000–11 999 0.93 (0.70–1.23)0.93 (0.70–1.23)
≥ 12 000 1.06 (0.83–1.35)1.06 (0.83–1.35)
Disorder
Not present  Reference
Present  1.02 (0.80–1.32)
Number of chronic diseases
0  Reference
1–2  1.00 (0.80–1.25)
≥ 3  0.99 (0.77–1.27)
c-statistics0.5570.5960.596
Likelihood ratio chi-square (P value)32.245 (< 0.001)72.750 (< 0.001)72.782 (< 0.001)

Discussion

  1. Top of page
  2. Abstract
  3. Introduction
  4. Methods
  5. Results
  6. Discussion
  7. Conclusion
  8. Acknowledgment
  9. Contributions
  10. References

The self-reported nonadherence rate was 13.2%, and age and education attainment were associated with self-reported nonadherence after adjusting for other variables. Younger patients with relatively less education were not likely to take their medicine as prescribed compared to those who were older and who had more education attainment. Apart from age and education, the other variables, that is, demographic and socioeconomic variables, did not affect medication adherence. Factors pertaining to health status were not related to medication adherence, either.

Demographic variables

We noted that only age among the demographic variables was associated with nonadherence. Our findings thus support the results of previous studies. Many studies indicated that older people were more likely to be persistent or to adhere to taking their prescriptions (Jokisalo et al., 2002; Hyre et al., 2007). In fact, a one-year difference was associated with a rate of medication discontinuation that was 0.98 times greater (Degli Esposti et al., 2002). Recent research has also shown that those older than 55 years adhere to a refill schedule at higher rates than a younger group (Steiner et al., 2009). Patients who perceived hypertension as a risk tend to show greater adherence to an antihypertensive regimen (Gregoire et al., 2006). Older people tend to be more interested in health issues and perceive hypertension as a risk more than younger people, which make older people more adherent. We did not find statistically significant differences in adherence rates between groups by gender or residence. Our findings somewhat match those of previous studies in that some inconsistencies arose.

Socioeconomic variables

We demonstrated that a lower education attainment level was associated with higher rates of nonadherence. Earlier research performed in the USA also showed that those with more education showed greater rates of medication adherence (Lowry et al., 2005). A study carried out in Finland explained the relationships between education and medication adherence as a result of an information seeking attitude, holding that higher education is associated with receiving more advice and information from a physician (Jokisalo et al., 2002). Further, a recent study found that the effects of education on medication adherence varied by sex, showing that lower educational attainment was associated with higher adherence in men but that lower educational attainment was related to lower adherence in women (Braverman & Dedier, 2009). However, the effects of education on medication adherence did not differ by gender in our additional analyses.

According to our findings, other socioeconomic variables excluding education did not relate with nonadherence. On the other hand, insurance type and income were related to medication adherence in Portugal (Santa-Helena et al., 2010) and Canada (Friedman et al., 2010). Medication adherence increases with the utility and continuity of health care (Fincham, 2007). Since most treatments and prescriptions pertaining to hypertension are covered by health insurance or medical benefits in Korea, nonadherence appears to be unaffected by insurance type or income level in our study.

Health-status variables

There were no statistically significant variables related to nonadherence among the health-status variables. A lack of mobility or communication can increase nonadherence by decreasing accessibility to health care or the availability of medical support (Fincham, 2007). For example, a study in Korea reported that mobility or communication impaired persons had lower adherence (Park et al., 2008a). Since there were few patients who had such disabilities in our study, we were unable to create separate categories for them. However, the results may differ if we analyze additional panel data further. The number of chronic diseases as another factor of the health-status variables was not associated with nonadherence in our study. Comorbidity can influence increased adherence by raising the awareness level about a disease, whereas it can also be a negative factor related to adherence because people with more comorbidities typically take more pills as well (World Health Organization, 2003). As we did not assign weights to particular diseases due to the limitations associated with self-reported questionnaires, it is difficult to conclude an effect of comorbidity on adherence at this point.

Implications for nurses

The results of our study showed that nurses should focus on the younger and less educated patients when they plan an antihypertension program in a clinical or community setting. These patients are not likely to adhere to a medication regimen due to a lack of knowledge about their disease (Uzun et al., 2009). Although there is little evidence about the positive effects of nursing care on increasing adherence levels (Haynes et al., 2008), it is still important to enhance the patient's belief or cognition through motivation, education, or information (World Health Organization, 2003; Uzun et al., 2009). One recent study confirmed that the program promoted the knowledge, self-efficacy, and health-related quality of life of hypertensive patients by educating, consulting, and discussing with them (Ham & Kim, 2011).

Strengths and limitations of the study

The strength of our study is that national survey data were analyzed for the first time in Korea and thus the results can be generalized to the Korean population. Our study is also meaningful since we explored multiple variables associated with self-reported nonadherence for the first time in Korea.

Although we found factors related to nonadherence in hypertensive patients, our study has some limitations. First, it is likely to overestimate nonadherence since the Korean Medical Panel measured nonadherence using a self-reported dichotomous question. Despite the evidence that self-reports provide estimates of overall adherence similar to those of other measures (Hansen et al., 2009), nonadherence rate was likely to be overestimated. Besides, dichotomous questions (e.g. “Do you take medicine according to an antihypertensive medication regimen?”) restrict the information provided, such as the prescribed length of time on medication. Various ways of measuring adherence, such as linking self-reporting and health-insurance data, would be helpful to understand the scale of nonadherent patients accurately. Second, our model of estimating nonadherence did not include therapy-related or patient-related factors. It is necessary to also consider these factors in order to find closely related variables in a further study. finally, we did not explore the mechanisms of our variables for nonadherence. It may be possible that demographic, socioeconomic, and/or health-status factors affect medication adherence indirectly via beliefs. It is therefore necessary to construct a model to explain the relationships between variables in an effort to find the most effective interventions.

Conclusion

  1. Top of page
  2. Abstract
  3. Introduction
  4. Methods
  5. Results
  6. Discussion
  7. Conclusion
  8. Acknowledgment
  9. Contributions
  10. References

This study analyzed the demographic, socioeconomic, and health-status variables that affect medication nonadherence in patients with hypertension. Based on a national survey data and multivariate logistic regression models, the results showed that the variables of age and education-attainment level were statistically significant in explaining low adherence rates. Therefore, nurses should pay more attention to hypertension patients who are young and who have low education-attainment levels when educating them about their medications. Further study can measure adherence more accurately by including therapy-related factors and by constructing a model that can offer a better understanding of the mechanisms that affect medication adherence.

Acknowledgment

  1. Top of page
  2. Abstract
  3. Introduction
  4. Methods
  5. Results
  6. Discussion
  7. Conclusion
  8. Acknowledgment
  9. Contributions
  10. References

We would like to thank the Korea Institute for Health and Social Affairs and the National Health Insurance (NHI) for their help with database.

Contributions

  1. Top of page
  2. Abstract
  3. Introduction
  4. Methods
  5. Results
  6. Discussion
  7. Conclusion
  8. Acknowledgment
  9. Contributions
  10. References

Study design: SC, JK.

Data analysis: SC, JK.

Manuscript writing: SC, JK.

References

  1. Top of page
  2. Abstract
  3. Introduction
  4. Methods
  5. Results
  6. Discussion
  7. Conclusion
  8. Acknowledgment
  9. Contributions
  10. References
  • Bangalore S, Shahane A, Parkar S, Messerli FH. Compliance and fixed-dose combination therapy. Curr. Hypertens. Rep. 2007; 9: 184189.
  • Baune BT, Aljeesh Y, Bender R. Factors of non-compliance with the therapeutic regimen among hypertensive men and women: a case-control study to investigate risk factors of stroke. Eur. J. Epidemiol. 2005; 20: 411419.
  • Ben-Natan M, Noselozich I. Factors affecting older persons' adherence to prescription drugs in Israel. Nurs. Health Sci. 2011; 13: 164169.
  • Braverman J, Dedier J. Predictors of medication adherence for African American patients diagnosed with hypertension. Ethn. Dis. 2009; 19: 396400.
  • Burnier M. Medication adherence and persistence as the cornerstone of effective antihypertensive therapy. Am. J. Hypertens. 2006; 19: 11901196.
  • Cramer JA, Roy A, Burrell A et al. Medication compliance and persistence: terminology and definition. Value Health 2008; 11: 4447.
  • Degli Esposti E, Sturani A, Di Martino M et al. Long-term persistence with antihypertensive drugs in new patients. J. Hum. Hypertens. 2002; 16: 439444.
  • Fincham JE. Patient Compliance with Medications. Binghampton, NY: Haworth Press, 2007.
  • Fitz-Simon N, Bennett K, Feely J. A review of studies of adherence with antihypertensive drugs using prescription databases. Ther. Clin. Risk Manag. 2005; 1: 93106.
  • Friedman O, McAlister FA, Yun L, Campbell NR, Tu K. Antihypertensive drug persistence and compliance among newly treated elderly hypertensives in Ontario. Am. J. Med. 2010; 123: 173181.
  • Gregoire J, Moisan J, Guibert R, Ciampi A, Milot A. Predictors of self-reported noncompliance with antihypertensive drug treatment: a prospective cohort study. Can. J. Cardiol. 2006; 22: 323329.
  • Ham KO, Kim BG. Evaluation of a cardiovascular health promotion programme offered to low-income women in Korea. J. Clin. Nurs. 2011; 20: 12451254.
  • Hansen RA, Kim MM, Song L, Tu W, Wu J, Murray MD. Comparison of methods to assess medication adherence and classify nonadherence. Ann. Pharmacother. 2009; 43: 413422.
  • Hashmi SK, Afridi MB, Abbas K et al. Factors associated with adherence to anti-hypertensive treatment in Pakistan. PLoS ONE 2007; 2: e280.
  • Haynes RB, Ackloo E, Sahota N, McDonald HP, Yao X. Interventions for enhancing medication adherence. Cochrane Database Syst Rev 2008; (2): CD000011.
  • Hyre AD, Krousel-Wood MA, Muntner P, Kawasaki L, DeSalvo KB. Prevalence and predictors of poor antihypertensive medication adherence in an urban health clinic setting. J. Clin. Hypertens. 2007; 9: 179186.
  • Jang SM, Song HJ, Shin SY, Sung YN, Kim SO (eds). Development of Evaluation Indexes of Hypertensive Medicine. Seoul: Health Insurance Review and Assesment Service, 2008.
  • Jokisalo E, Kumpusalo E, Enlund H, Halonen P, Takala J. Factors related to non-compliance with antihypertensive drug therapy. J. Hum. Hypertens. 2002; 16: 577583.
  • Korea Institute for Health and Social Affairs, National Health Insurance Service. The Second Korean Medical Panel Sourcebook. Seoul: Korea Institute for Health and Social Affairs and National Health Insurance Service, 2010.
  • Krousel-Wood M, Hyre A, Muntner P, Morisky D. Methods to improve medication adherence in patients with hypertension: current status and future directions. Curr. Opin. Cardiol. 2005; 20: 296300.
  • Krousel-Wood M, Joyce C, Holt E, Muntner P, Webber LS, Morisky DE, Frohlich ED, Re RN. Predictors of decline in medication adherence: results from the cohort study of medication adherence among older adults. Hypertension 2011; 58: 804810.
  • Lagi A, Rossi A, Passaleva MT, Cartei A, Cencetti S. Compliance with therapy in hypertensive patients. Intern. Emerg. Med. 2006; 1: 204208.
  • Lewis LM, Askie P, Randleman S, Shelton-Dunston B. Medication adherence beliefs of community-dwelling hypertensive African Americans. J. Cardiovasc. Nurs. 2010; 25: 199206.
  • Lowry KP, Dudley TK, Oddone EZ, Bosworth HB. Intentional and unintentional nonadherence to antihypertensive medication. Ann. Pharmacother. 2005; 39: 11981203.
  • Morris AB, Li J, Kroenke K, Bruner-England TE, Young JM, Murray MD. Factors associated with drug adherence and blood pressure control in patients with hypertension. Pharmacotherapy 2006; 26: 483492.
  • Park CM, Jang SM, Jang SH, Lee HJ. Medical Expediture and Health Outcome According to Medication Adherence. Seoul: Health Insurance Review and Assessment Service, 2010.
  • Park JH, Lee SY, Kim SY, Shin Y. Disparities in antihypertensive medication adherence in persons with disabilities and without disabilities: results of a Korean population-based study. Arch. Phys. Med. Rehabil. 2008a; 89: 14601467.
  • Park JH, Shin Y, Lee SY, Lee SI. Antihypertensive drug medication adherence and its affecting factors in South Korea. Int. J. Cardiol. 2008b; 128: 392398.
  • Santa-Helena ET, Nemes MI, Eluf Neto J. Risk factors associated with non-adherence to anti-hypertensive medication among patients treated in family health care facilities. Cad. Saude Publica 2010; 26: 23892398.
  • Shaya FT, Du D, Gbarayor CM, Frech-Tamas F, Lau H, Weir MR. Predictors of compliance with antihypertensive therapy in a high-risk medicaid population. J. Natl Med. Assoc. 2009; 101: 3439.
  • Statistics Korea. The Statistics of Death Cause. Daejon: Statistics Korea, 2011.
  • Steiner JF, Ho PM, Beaty BL et al. Sociodemographic and clinical characteristics are not clinically useful predictors of refill adherence in patients with hypertension. Circ. Cardiovasc. Qual. Outcomes 2009; 2: 451457.
  • Uzun S, Kara B, Yokusoglu M, Arslan F, Yilmaz MB, Karaeren H. The assessment of adherence of hypertensive individuals to treatment and lifestyle change recommendations. Anadolu Kardiyol. Derg. 2009; 9: 102109.
  • Vawter L, Tong X, Gemilyan M, Yoon PW. Barriers to antihypertensive medication adherence among adults – United States, 2005. J. Clin. Hypertens. 2008; 10: 922929.
  • World Health Organization. Adherence to Long-Term Therapies. Geneva: World Health Organization, 2003.
  • Yang SJ, Jung D, Choi AS. Prediction model of blood pressure control in community-dwelling hypertensive adults in Korea. Nurs. Health Sci. 2010; 12: 105112.