Prevalence of hypertension
This study provides important new evidence on the prevalence of hypertension in a representative sample of an elderly population in a geographically well defined Mediterranean area of the Tunisia center. Moreover, it has already established socio-demographic variables and drug consumption. A peculiar aspect of our population is the low level of education (89% for illiterates plus subjects with kuttabs education).
We observed that more than 50% of elderly in Monastir city suffer from hypertension. Since the use a BP value of 140/90 mmHg as cut-off, the prevalence of hypertension has increased. These percentage values have also been found in studies carried out in North African countries such as Morocco , and other Arab countries such as Egypt . Laouani reported a prevalence rate of 69.3%  but in 2000, Kammoun et al. reported in a representative sample of the Tunisian elderly population the prevalence of 32% . The increasing prevalence of hypertension among the elderly population of Monastir city may be attributed to change in lifestyle in the past decade. However, the prevalence of hypertension is still considerably lower than in other industrialised countries such as France (79.8%), England (62%), Greece (65.4%), Spain (62%), Italy (76.3%) and United States (84%) [17–22].
This study showed clear gender differences in the prevalence of hypertension, with women being more likely to be hypertensive than men. The gender differences in the prevalence of hypertension have been reported in many studies [21–23]. The present study found that hypertension prevalence declined after 80 years, the difference might partly be due to the shorter survival of people with AH and a low incident in the older age group. As demonstrated in this study, isolated systolic hypertension is the most common form of hypertension found in elderly . The predominance of systolic hypertension may reflect the consequences of a tendency for clinicians to treat it less aggressively than diastolic hypertension.
Awareness, treatment and control of hypertension
Only 19% of the respondents were newly diagnosed with hypertension. Although these proportions are quite similar to those reported from other developing countries . Population studies have shown different rate of hypertension awareness by hypertensive individuals. In European countries, these levels range from 52.7% in Germany to 70% in Sweden. In Canada and USA the awareness rate is 83% and 87.3% respectively [26, 27]. In these studies, the degree of awareness of hypertension was high (81%) than those reported by the India study . Wyatt et al. reported that awareness significantly increased with age, female sex, presence of major co-morbidity and receiving preventive care , a combination of these factors contributes to the high level of awareness in our study.
Among treated hypertensive individuals in this study, about 30% had blood pressure controlled. These data are also similar to those reported by Liau et al. (34.3%) , but lower than those reported by Cipullo et al. (52.4%). In Europe the blood pressure controls rate range from 22.9% in Spain to 37.7% in England . The favourable results in our study may be due to adequate public information about hypertension and availability of free government-supplied anti hypertensive drugs with relatively high access to health care and insurance coverage.
Correlates of hypertension
Regarding the educational level, there are an inverse but not significantly association between schooling and AH, the small number of educated population may influence this result. Although, this finding was consistent with previous reports [30, 31]. Cipullo et al. reported that the risk of hypertension was 2.8-times higher for those with lower schooling . There is an intricate relationship between physical health and psychological status. Depression is known to have diverse effects on body functions. A French study of the elderly in the community revealed that hypertension was associated with anxiety but not depression . Our study showed the lack of association between depression symptoms and incident hypertension.
The benefits of physical activity in the prevention and treatment of high blood pressure have been very well described . In the present study, physical activity was not at a reduced risk for hypertension, but self-reported bias and interviewer bias can result in no differential misclassification.
The higher, but not significant, prevalence of hypertension in non smokers should be interpreted with caution since it is possible that part of the existing difference may be due to the high percentage of current smokers for male compared to female. No association between AH and marital status, these results are in accordance to what has been reported in other study . In Martinique and China populations, the urban population exhibit a risk profile of hypertension [34, 35], but not urban/rural differences were observed in our study, it is important to note that in our study in Monastir City, only 14% are from rural sites, the small number may influence statistics. The higher prevalence rates in hypertension among the illiterate and in individuals living in urban area are often being attributed to differences in life style, modernization, a shift from an agricultural to non agricultural economy, physical activity and occupation. Socioeconomic difference play, also, an important role in health conditions influencing different factors such as access to the health system, degree of information and understanding of medical conditions [36, 37].
Similar to several population-based studies [26, 38, 39], it was observed that there is an important association between BMI and AH, as well as waist circumference. Overweight and obesity are actual risk factors for hypertension. The positive correlation between waist circumference and AH identifies a simple, low cost and easy-to-apply measure as an important marker for AH.
The strong association between diabetes mellitus and hypertension has been reported by numerous studies [28, 40]. It is widely appreciated that hypertension increase the risk of activities of daily living and instrumental activities of daily living decline and have an important impact on the dependency categories , our data support this observation.
Limitations of the study
Some limitations of this study should be mentioned: first the use of a single visit to ascertain hypertension status can result in an over-estimation of its prevalence. Other potential sources of bias include the self-reported hypertensive treatment by the participants. Treatment status was assessed by the question of whether or not the person was currently using antihypertensive drugs (within the last two weeks). No information on type and dosage of antihypertensive medication being used or medication compliance was collected from participants, nor were they asked about AH control measures other than pharmacotherapy. Therefore, "treatment" was restricted to the use of medication, without consideration of other, non pharmaceutical strategies such as dietary sodium restriction.
In this study, we did not assess the lipid profile and blood glucose measurement, a fact that did not allow us to evaluate the association between the metabolic changes and high BP.
We used BMI as a simple method to categorize people as "healthy weight," "overweight," or "obese". BMI formula may overestimates fatness in elderly because of loss of height resulting from vertebral compression (a problem commonly associated with aging). Knee-heel length and weight/knee height as alternative measurements in the elderly were not measured.
On the other hand, we described the prevalence of hypertension among the elderly in Monastir city, in order to provide estimates of the problem, this adding information about a group that is under researched. The use of a door-to door survey, the large sample size, adequate representation of women and the very old and a health professional for blood pressure measurement are strengths of our study.