Prevalence, awareness and control of hypertension in Malaysia 1980 – 2017: A Systematic Review and Meta-Analysis

Background Hypertension is a common public health problem worldwide and is a well-known risk factor for increased risk of cardiovascular diseases, contributing to high morbidity and mortality. However, there is no systematic review and meta-analysis that has been done in a multi-ethnic population like Malaysia. This systematic review aims to determine the trend in prevalence, awareness and control of hypertension in Malaysia. Methods Systematic searches were conducted in PubMed, Scopus, Ovid, Cumulative Index to Nursing and Allied Health Literature, Malaysian Medical Repository and Malaysia Citation Index published between 1980 and 2017. All original articles in English were included. Studies included were those on adults aged 18 years and above. Studies of prevalence in children and adolescents and pregnancy related hypertension were excluded. Two authors independently reviewed the studies, carried out data extraction and performed quality assessment. Heterogeneity between studies and publication bias was assessed and effect size was pooled by the random effect model. Results Fifty-six studies with a total of 241,796 subjects were included. The prevalence of hypertension throughout Malaysia varied (I2 = 99.3%). The overall pooled prevalence of hypertension over the past 4 decades was 28.2% in adults aged 18 years and older (95% CI: 26.1 – 33.3) and the prevalence in those 30 years and older was 40.0% (95% CI: 35.3-44.8). For subgroup analysis, the prevalence of hypertension in male aged 18 and above was 31.4% (95% CI: 26.5 - 36.2) and 27.8% in female (95% CI: 20.7 – 34.9). The prevalence of hypertension among the ethnic groups aged 18 years and above were 37.3% in Malays (95% CI: 32.9 – 41.7); 36.4% in Chinese (95% CI 31.6 - 41.2) and 34.8% in Indians (95% CI: 31.2-38.4). The prevalence of hypertension was the lowest in the 1980s (16.2%, 95% CI: 13.4-19.0%), increases up to 36.8% in the 1990s (95% CI: 6.1-67.5), then came down to 28.7% (95% CI: 21.7-35.8) in the 2000s and 29.2% (95% CI: 24.0-34.4) in the 2010s. The prevalence of awareness was 38.7% (95% CI: 31.7 – 45.8) whereas the control of hypertension of those on treatment was 33.3% (95% CI: 28.4 – 38.2). Conclusion Three in 10 adults aged 18 years old and above have hypertension, whereas four in 10 adults aged 30 years old and above have hypertension. Four out of 10 are aware of their hypertension status and only one-third of them who were under treatment achieved control of their hypertension. Concerted efforts by policymakers and healthcare professionals to improve the awareness and control of hypertension should be of high priority.

Hypertension is a common public health problem over the past several decades [1][2][3]. It is one 3 of the major risk factors for cardiovascular diseases (CVD) like stroke, heart failure and 4 ischemic heart disease, [4]. Studies have been done repeatedly over the last 40 years on 5 prevalence, awareness and control of hypertension in Malaysia. The National Health and 6 Morbidity Surveys (NHMS), which are nationwide studies, were done every 10 years since 7 1986 in Malaysia. The NHMS has also shown the trend of hypertension in different genders 8 and ethnicities. Besides the NHMS, several other studies also examined the prevalence of 9 hypertension in specific groups and settings that were different from that of the NHMS [5,6]. 10 Prevalence in different settings may provide a slightly different picture and their accompanying 11 set of problems. Because of the different settings and their conceivably different prevalence, 12 there is a need for a systematic review and meta-analysis to gather and review all the 13 information provided by these studies. Furthermore there is no systematic review that has been 14 done to provide more information on the changes and trends of prevalence of hypertension in 15 the past 4 decades in Malaysia as well. Hence, this systematic review aimed to determine the 16 prevalence, control and awareness of hypertension in Malaysia over the past 4 decades.

19
Literature search strategy 20 Literature search was done between August 2017 and February 2018. Eligible articles were 21 identified independently by two authors (CZY, TCH) through performing an initial screen of 22 titles or abstracts, followed by full text review. Disagreements on study inclusion, data 23 extraction and quality assessment were resolved by discussion between the 2 authors or by 24 other senior authors. We comprehensively searched through 6 databases which are Pubmed, 5 1 Ovid, Scopus, Cumulative Index of Nursing and Allied Health Literature (Cinahl), MyCite [7] 2 and MyMedR [8]. The search strategy was based on terms related to: (prevalence) and 3 (awareness) and (control) and (hypertension or high blood pressure) and (Malaysia) and 4 combination of these using Boolean operators, adapted from previous reviews, and a 5 combination of expanded MeSH term and free-text searches were conducted. A combination 6 of expanded MeSH term and free-text searches were used as shown in Appendix 3. Cross 7 reference of all selected articles were scanned for additional studies. If more than one article 8 from a study was published, the article that provided the most updated data was selected. For 9 studies that used the same set of data, only the study with the primary data was chosen, in view 10 of possible duplication of data.  Exclusion criteria 6 1 We excluded intervention studies, case studies, pharmacogenetic studies, case series which 2 only included qualitative data, comments or letters, audits, narrative reviews, conference 3 proceeding, opinion pieces, methodological, editorials, animal studies or any other publications 4 lacking primary data and/or explicit method descriptions.

6
Quality assessment and data extraction 7 Each article was undergone quality assessment by two authors using a modified critical 8 appraisal checklist (Appendix 4) [57]. Each article's quality was graded as 'high quality' if it 9 scores ≥7/11; or graded as 'low quality' if it scores score <7/11 [5,6]. The scoring result was 10 shown in Appendix 3. Study characteristics (first author, year of publication, place and year of 11 study, study setting, sampling method, sample size, blood pressure apparatus and classification 12 cut-offs), participant characteristics (age group, gender, ethnicity and geographical origin), and 13 prevalence (prevalence of hypertension, awareness and control) were extracted onto pre-coded 14 spreadsheets independently by two authors (CZY, TCH). Data were extracted at the lowest 15 possible disaggregate level referred to as subpopulation here. This review is presented 16 according to the PRISMA guideline [10].  OpenMetaAnalyst for data analyses [11]. Heterogeneity between studies for the pooled 24 estimates was examined using I 2 where an I 2 of ≥75% suggests considerable heterogeneity. If 25 the heterogeneity of the trials were ≥75%, a random effects model was used to pool the 7 1 prevalence, awareness and control of hypertension. Subgroup analysis was done to examine 2 the prevalence of hypertension by gender, age group, ethnicity, setting and geographical origin.

3
Characteristics of studies included 4 We identified a total of 1493 manuscripts in the initial search as shown in Figure 1. 1242 studies 5 were retrieved for further assessment after duplicated records were removed (n=251). 6 Evaluation of the inclusion/exclusion criteria showed that 52 studies fulfilled eligibility, 7 together with another 4 studies identified from cross-referencing making a total of 56 studies 8 to be included into our systematic review. Using a modified critical appraisal checklist, quality 9 assessment showed that 51 studies (50 community studies and 1 study done in healthcare 10 setting) were of high quality (Appendix 3). A total sample size of 241,796 respondents from 11 Malaysia were included into the analysis. Overall, the total subjects comprised of 51352     Awareness of hypertension 15 The overall awareness of hypertension in Malaysia as identified in 10 studies was 38.7% as 16 shown in Table 2  Malay has a control rate of blood pressure at 29.3%, while non-Malay had a control rate of 7 32.5%. The control of hypertension was higher in urban dwellers than in the rural area (36.5% 8 vs 34.1% respectively). 9 10 Sensitivity analyses 11 The main analysis for prevalence of hypertension was rerun by removing one subpopulation at 12 a time. The pooled estimates did not vary much from the original analysis during each removal. 13 Removal of five low quality studies or the smaller (size < 100) subpopulation did not affect the   A review showed that this prevalence was as high as developed countries despite being a 6 developing country [72]. In fact our prevalence of hypertension is even higher than that of 7 United States by 0.7% [73]. 8 9 The heterogeneity of the overall prevalence in this study was high. This could be due to our 10 wide variation of sample size in our review, which ranged from 56 to 106527. Geographical origin also played an important role in causing such high heterogeneity in our 3 study. In our study, we had a 9.9% difference in both geographical origin, in which urban 4 dwellers had a higher prevalence of hypertension compared to those living in rural areas. This 5 could be due to a more sedentary lifestyle and higher prevalence of obesity seen in urban 6 dwellers and these themselves are risk factors for developing hypertension. A rural population 7 is more likely to be in occupations or physical activities that require more energy expenditure 8 [75]. 9 10 It has been reported that digital blood pressure monitor show poor agreement with mercury 11 sphygmomanometer in measuring blood pressure, thus influencing the difference in prevalence prevalence of hypertension using mercury sphygmomanometer compared to digital blood 14 pressure device in our review, the difference between the two devices was not substantial.

15
Furthermore, in view of the safety issues of possible mercury leakage, the difference between 16 the prevalence of hypertension using these two devices should not be taken into consideration 17 during our daily clinical practice. As such, it is reasonable to use digital blood pressure device 18 as it is safer and more convenient than the "gold standard" mercury sphygmomanometer.  to developed country, we also found a similar doubling phenomenon in the prevalence of 4 1 hypertension, but only later in age, which is 63.1% in age group 60 years old and above from 2 33.2% among those from age group 40-59 years old [79]. This showed that the prevalence of 3 hypertension in Malaysia starts at an earlier age compared to the developed country which 4 makes young hypertension to be an anticipating issue to be concerned in the future in Malaysia.

5
This is expected as older age is closely related with hypertension because of the alteration in 6 the arterial structure and ongoing calcification that leads to increased arterial stiffness [78]. In 7 short, with increasing age it is expected that the prevalence of hypertension will rise. However, 8 focussing on the older population aged 60 years old and above, the prevalence of hypertension 9 in this age group in Malaysia is found to be the lowest among Asian countries such as Singapore a study done in Southern China which showed the similar result [9]. This may be due to limited  yet to be improved.