Blood pressure control amongst patients living with hypertension presenting to an urban district hospital outpatient clinic in Kwazulu-Natal

Background The prevalence of hypertension in South Africa has been estimated to be 20% of the adult population with over six million people being affected. Poor adherence to treatment plans lead to inadequate blood pressure control and high morbidity. Many studies have looked at factors contributing to poor blood pressure control in South Africa but few studies actually focus on district hospitals in Kwazulu-Natal in particular, despite the fact that the province has the most heterogeneous population in South Africa. Method The study was a descriptive cross-sectional study conducted at the chronic outpatient clinic of an urban district hospital involving 370 participants aged 18–90 years. Result The study showed poorly controlled blood pressure in 58% of the participants. Only 35% knew their blood pressure results on the day of interview and 19.2% were aware of their target blood pressure. Good adherence was self-reported by 95% of the participants, whist 51.4% reported significant side-effects to medication. Conclusion The majority of patients had poor knowledge about blood pressure and little awareness of their blood pressure reading. These may be precursors to poor blood pressure control and this needs further investigation. A high level of self-reported adherence to medication did not translate into effective blood pressure control. A significant number reported medication side-effects which may have contributed to the poor blood pressure control. The high adherence rate may therefore have been over reported. An objective way to measure adherence will be necessary for future research.


Introduction
Hypertension is a major worldwide health challenge and a key risk factor responsible for adverse cardiovascular events. Although, largely preventable and modifiable, uncontrolled hypertension has been reported to negatively affect the global population's health, causing cerebrovascular accidents and cardiac diseases. 1 Hypertension 'accounts for greater than 5.8% of the total deaths, 11.9% of the years of life lost, and 1.4% of the disabilityadjusted life years worldwide'. 2 It has a major influence in the development of cardiac and vascular diseases. Despite being a highly treatable condition, studies have shown that hypertension remains poorly controlled. 3 In addition, the diagnosis, management, and control of hypertension pose a great challenge for physicians and researchers. 4 Hypertension affects approximately one billion individuals worldwide and fifty million individuals in the United States of America. 5 A number of clinical trials have shown that optimal blood pressure (BP) control can be achieved in most hypertensive patients using multiple antihypertensive medications. 6 A retrospective study conducted in communitybased settings in the USA indicated that 70% of the population had blood pressure levels that were optimally controlled. This was similar to the National Health and Nutrition Examination Survey (NHANNES) study done between 2007 and 2008, which indicated that 69% of patients in this cohort had controlled blood pressure. This study showed that optimally controlled patients living with hypertension were significantly more likely to be younger and have less co-morbidity. 7 A study which looked at uncontrolled hypertension in the United States concluded that the majority of cases were due to systolic hypertension. This report reaffirms the need to assess the extent to which patients are aware of their systolic blood pressure as systolic blood pressure awareness and education is an important component of the interventional programme that is used in the control of hypertension. 7,8 Hypertension control has been positively linked to adequate knowledge about blood pressure goals. A recent study done on 7649 participants across Europe showed that poor blood pressure control bears a strong correlation to lack of knowledge of target blood pressure. The study found that 50.4% of participants had controlled blood pressure and 49.4% had knowledge of their target blood pressure. 9 Numerous studies have concluded that poor medication adherence, lack of patient awareness about their target blood pressure levels, physician inertia as well as the presence of drug side-effects should be considered as possible causes of inadequately controlled blood pressure. 3 Adequate assessment and treatment of hypertension has always been a major challenge for public health care workers. 5 An article in Time magazine in 2004 stated that 'with sub-Saharan Africa's population of 650 million and increasing longevity and westernization, hypertension has now changed from a relatively rare condition to a major problem. This observation necessitated the African Union to regard hypertension as one of the continent's greatest challenges after AIDS'. 10 The prevalence of hypertension in rural studies undertaken in sub-Saharan Africa in the 1970s, 1980s and 1990s using the previous definition of blood pressure of 160/95 mmHg found hypertension to be as low as 4.1% in Ghana, 7.0% in Lesotho, 5.9% in Nigeria, and 7.37% in rural Zululand in South Africa. 11 Estimates based on the 1998 demographic survey put the prevalence of hypertension at 11% and 14% of South African males and females respectively. This implies that there were six million hypertensive persons in South Africa. 12 This number is likely to be even higher in 2012 because of population growth and growing urbanisation. 13 In the year 2000, 47 000 deaths in South Africa were attributed to high blood pressure, and elevated blood pressure was the second leading contributory cause of death. 14 A study amongst Durban adults reporting hypertension trends using the WHO criteria of 160/90 mmHg found hypertension to be more common in urban black South Africans of Zulu origin (25%), moderate in white South Africans (17%), and Indian populations (14%), and the lowest in rural black South Africans. 15 The 2011 South African Hypertension Guidelines recommend that the target blood pressure for patients on hypertensive therapy should be below 140/90. 16 In patients with major cardiovascular risks such as those with diabetes mellitus, associated clinical conditions or established target organ damage, the optimal blood pressure should be 130/80 mmHg. 16 In well conducted clinical trials, antihypertensive drugs either as mono-therapy or combination therapy have been shown to effectively control blood pressure and reduce cardiovascular morbidity and mortality. 17 However, outside of clinical trials, control of hypertension remains disappointing with an estimated 30% of patients on treatment being effectively controlled. 16,18 Despite the availability of multiple effective antihypertensive drugs, control of hypertension remains poor. 5,19 In both high and low income countries, less than 27% and 10% respectively of hypertensive patients have achieved their target blood pressure. 5,19 Many reasons have been ascribed for the poor control of hypertension worldwide. Patient knowledge of hypertension also plays an important role in the ability to achieve successful control. 17 In addition, poor treatment adherence, inappropriate prescriptions, therapeutic inertia and unhealthy lifestyle contribute to poor control. 3,20,21 The 2011 South African Hypertension Guidelines emphasised poor adherence to medication as one of the main causes of uncontrolled blood pressure and was thought to be due to treatment factors, patient factors and illness related factors. 16 Other factors identified as contributors to uncontrolled hypertension include sex, age, and race. 8 In South Africa, white and Asian individuals have been noted to have better blood pressure control compared to their black counterparts. 12 Black individuals also develop hypertension earlier in life, have more severe hypertension related complications and respond differently to antihypertensive medication. 22 In the first South African Demographic and Health Survey (1998), awareness, treatment status, and control varied substantially between men and women, as well as between different ethnic groups. 12 Women were substantially more likely to be aware of their hypertension than men, 51% and 26% respectively. The population survey also found that 36% of women and 21% of men had hypertension but were not on treatment. Amongst those on treatment, only 18% were controlled. With respect to ethnic differences, black South Africans had the lowest rates of awareness, treatment rates and control of hypertension. 12 There is little published data on hypertension control amongst patients attending district hospitals in the KwaZulu-Natal Province, South Africa. The aim of this research was to describe the pattern of blood pressure control amongst patients attending a busy outpatient clinic in an urban district hospital. The specific objectives were to describe the demographic and clinical characteristics, knowledge of hypertension targets, degree of adherence to medication and knowledge of side-effects to antihypertensive medications. Approval for the study was granted by the Bioethics Research Committee of the University of KwaZulu-Natal (UKZN), the postgraduate education committee, hospital management and the KwaZulu-Natal Department of Health. Written consent was obtained from all study participants.

Methods
This was a descriptive, cross-sectional study done at the chronic outpatient clinic at a busy urban public health district hospital. An average of 1375 hypertensive patients was seen monthly at the outpatient clinic in 2011. A sample size of 370 was selected as being appropriate for an observational descriptive study as confirmed with a biostatistician. Hypertensive patients visiting the outpatient clinic were systematically sampled by selecting every third patient that met the inclusion criteria over a three month period commencing October 2012. Sampling was stopped when the sample size was attained. The inclusion criteria were all patients living with hypertension older than 18 years attending the chronic clinic, and on treatment for hypertension for at least one year. Excluded from the study were pregnant women and cognitively impaired patients. Patients' clinical records were reviewed and an exit structured interview was conducted with each participant by the principal investigator after the participant had consulted the hospital medical officer. The study tool captured demographic data, anthropometric data and, clinical information from the clinical record. Patients were asked specific questions about their hypertension, comorbid conditions, and habits. Blood pressure readings from the last two clinic visits plus current blood pressure readings were captured. Poor control was defined as blood pressures greater than 140 mmHg systolic or 90 mmHg diastolic. Patients who had two of the three readings elevated were considered to be poorly controlled. Data was captured and analysed using SPSS Statistics for Window Version 21.0. (Armonk, NY: IBM Corp 2012). Results were summarised by frequencies and percentages (categorical variables) and mean, median, standard deviation and percentiles (numerical variables) as appropriate. Wilcoxon signed rank test was used in analysing the variations and trends between the categorical data.

Results
Forty-nine percent of the participants were black South Africans, 25.4% were Indians, 17.8% were Mixed race and 7.6% were white South Africans with their age groups being presented in Table 1. Approximately 62.3% of the participants had secondary and tertiary education, 59.1% were pensioners or patients collecting a social grant. Sixty percent of the participants were either overweight or obese.
The study showed uncontrolled blood pressure (BP) in 58% of the participants during the preceding three months. Other determinants of blood pressure control are shown below in Table 2. Eighty percent had other co-morbid diseases such as diabetes mellitus, asthma or arthritis and 60% were either obese or overweight. Approximately 81% did not know what their target blood pressure should be, and 65% could not recall what their blood pressure for the current visit was. Most participants claimed that they were not told what their blood pressure was by the health care providers managing their hypertension. Eighty-eight percent of participants claimed a 95% adherence to treatment. This information was not verified. Fifty-one percent of the participants reported significant side-effects ranging from cough, dizziness, skin rashes and urinary frequency, but almost 60% of these did not know if the drug treatment was causing these sideeffects. More than 70% of the study participants indicated that they practiced at least one form of exercise to decrease their weight. Only 48% of the hypertensive patients could describe what a DASH diet was and these participants claimed to adhere to it. Sixty-six percent of the participants claimed to be psychologically stressed.
The associations between the various factors contributing towards blood pressure control are shown in Table 3 A and B below. There was no statistically significant difference between blood pressure control, and adherence (p = 0.514). There was no statistically significantly association between blood pressure control and increase in number of antihypertensive medications used (p = 0.527).There was a statistically significant association between side-effects to medication and blood pressure control (p = 0.004). There was a statistically significant association between BP control and ethnic group (p = 0.004).

Discussion
The study showed a blood pressure control rate of 41.9% amongst the patients attending a district hospital in Kwazulu-Natal. The control rate was slightly higher than the average rate of 40% that was previously reported in South Africa but consistent with a global control rate of 19% -54%. 23 However, this degree of control is too low considering the complications of hypertension. Poor blood pressure control translates into increased cardiovascular morbidity and mortality such as ischemic heart diseases which affects the burden placed on health care services in South Africa.       The demographics of the study group presented in Table 1 showed that the predominant racial group in this study was black South African people and the predominant gender was female. The cross tabulation for BP control versus race showed a significant relationship (p = 0.004.) These further affirm findings of earlier studies which showed highest reported cases to be amongst urban black South Africans and females when viewed along ethnic and gender lines. 12,15 It has been argued that greater medication adherence and better blood pressure control improves with patients' knowledge of hypertension. 24 This study showed that only 19.2% of the participants had knowledge of their target blood pressure and the cross tabulation for the blood pressure control vs. knowledge of target blood pressure did not show a statistically significant relationship (p = 0.585).
The self-reported adherence rate of 88.4% does not correlate with the degree of control noted (42%). A high adherence should translate into a well-controlled blood pressure. The cross tabulation between BP control and side-effects indicated that there is a statistically significant association between side-effects reported and BP control. This would suggest that patients who experienced side-effects to drugs may have over reported their adherence rates. This over reporting of adherence is consistent with findings from other studies. 25,26 This study showed that 51% of the participants reported side-effects to antihypertensive medications but 60% of these participants could not identify which drugs caused the sideeffects. It has been reported that 35% of hypertensive patients will stop taking their drugs in the first six months if they experience a side-effect. Another reported reason for stopping medication was patient dissatisfaction with treatment. 27 The significantly higher proportion of side-effects noted in this study was associated with multiple antihypertensive medications and this was statistically significant (p = 0.042).
This study showed that 70% of the participants were using more than three medications, which is of concern as they constituted the largest percentage of uncontrolled hypertensive patients. These participants would be classified as having resistant hypertension and in South Africa should have been managed at a regional level hospital. The study also showed that 80.3% of the participants had other co-morbid diseases necessitating poly-pharmacy, which increased the possibility of adverse drug-drug interactions.

Implications of the research
The findings of this study support previous studies which demonstrated that only about 40% of hypertensive patients attain blood pressure control in South Africa. A plan to improve the control of blood pressure from the current 42% to greater than 80% is urgently required. The poor knowledge of blood pressure control and hypertensive targets amongst this cohort of participants should be addressed by clinicians through improved patient education. The high number of reported side-effects to anti-hypertensive medication should alert physicians that prescription practices in hypertensive patients may need reviewing. In addition, quality assurance teams at district level should be encouraged to do regular clinical audits of the management of hypertensive patients and monitor outcomes. Regular monitoring and evaluation of hypertension should be part of routine clinical practice and poorly controlled patients with resistant hypertension must be referred to higher levels of care. This study supports the development and implementation of educational programmes that improve the knowledge of hypertension in patients.

Recommendations
Educational intervention aimed at counselling patients about the cardiovascular implications should be instituted.
Pharmacists, nurses and doctors managing hypertensive patients should aim to address the gaps in the management of hypertension identified by this study. There is need for future research on patients' adherence rates using a more objective approach than self-reporting.

Limitations
It is worth mentioning that a degree of bias may exist in this study. Adherence rates may have been over reported as the investigator is a doctor known to patients at the clinic. The predominant age group are elderly with multiple comorbidities, which might have made hypertension control more difficult. The generalisation of these results should be limited to urban district hospitals and communities with heterogeneous ethnic populations.