Precursors of atherosclerotic and hypertensive diseases among adolescents in Addis Ababa, Ethiopia.

Precursors of atherosclerotic and hypertensive diseases among adolescents in Addis Ababa, Ethiopia D. Kebede1 & T. Ketsela2 The present study measured the prevalence of the precursors of atherosclerosis among 519 adoles- cents selected from high-school students in Addis Ababa. Also assessed was their knowledge about these precursors and about the hazards of smoking and of heavy alcohol consumption. A total of 13.8% of the adolescents smoked, 11.6% had a sedentary lifestyle, 9.2% consumed alcohol heavily, 14.1% were obese, 30.3% had one risk factor, and 4.4% had two risk factors for atherosclerosis. Altogether, 58% had inadequate knowledge about the precursors of atherosclerosis, and 62% and 51.4%, respec- tively, had inadequate knowledge about the hazards of smoking and of consuming alcohol. High-risk behaviours were positively associated with upper-income families and inadequate knowledge about the precursors of atherosclerosis. A substantial proportion of adolescents in Addis Ababa therefore exhibit the precursors of athero- sclerotic and hypertensive diseases. Further, similar studies should be carried out in other Ethiopian cities, and the Ministries of Health and Education should give due attention to the primary prevention of these diseases and formulate plans for appropriate actions. Introduction Recently a WHO Expert Committee concluded that cardiovascular diseases will emerge or be established as a substantial health problem in virtually every country in the world by the year 2000 (1). Although the major health problems in developing countries are malnutrition and communicable diseases, and the majority of their populations live in poverty, there have been marked changes in the living standards of the urban populations in these countries. It has been estimated that when life-expectancy reaches 50-60 years mortality from cardiovascular diseases accounts for 15-25% of all deaths (1). Among the urban communities of developing coun- tries life-expectancy at birth is currently over 50 years. In Addis Ababa, for example, life-expectancy is 63.3 years and about 12% of the Ethiopian popula- tion live in urban areas (2). As is the case in all developing countries, in Ethiopia national statistics grossly underestimate the effect of cardiovascular diseases on urban communities. 1 Assistant Professor, Department of Community Health, Faculty of Medicine, P.O. Box 1176, Addis Ababa University, Addis Ababa, Ethiopia. Requests for reprints should be sent to this author. 2 Senior Expert, Department of Epidemiology, Ministry of Health, Addis Ababa, Ethiopia. Reprint No. 5441 Extensive research conducted in developed countries on atherosclerotic diseases (coronary heart disease, atherosclerosis of the aorta and main arteries supplying the brain, the trunk, and lower extremities) and hypertensive diseases have shown that the major but potentially preventable precursors are smoking, obesity, a sedentary lifestyle, heavy alcohol con- sumption, diabetes, and hypertension (3). Trend analyses in these countries have revealed that the 1950-60 epidemic of cardiovascular mortality has been decreasing since 1970 (4) because of primary prevention activities. Secondary preventive strategies (begun after the onset of clinical disease) have not yet proved to be effective because of the short time between the first manifestation of these diseases and death, and the relatively high frequency of recur- rence, disability, and mortality among survivors (3). The precursors of atherosclerotic and hyperten- sive diseases start in childhood, and autopsies have shown early atherosclerotic changes in children and adolescents. Behavioural risk factors for these dis- eases are initiated in childhood and adolescence before they become established in adults. Preventive activities, if they are to be effective should, there- fore, begin among children and adolescents (5). If current projections of a major epidemic of cardiovascular diseases among the urban populations of developing countries by the year 2000 are correct, the health services of these countries, which are Bulletin of the World Health Organization, 71 (6): 787-794 (1993) © World Health Organization 1993 787 D. Kebede & T. Ketsela already overburdened, will be overwhelmed by demands to deal with diseases that require tertiary level diagnostic and management resources. A con- sensus is now emerging that national health policies should be examined to address this problem, in order to institute preventive strategies among children and adolescents for the prevention of adult atherosclerot- ic and hypertensive diseases (6). Adolescents in Addis Ababa constitute about 25% of the population (2); the magnitude of the behavioural and other risk factors for atherosclerotic and hypertensive diseases among them is, however, not known. The present study was carried out to measure the proportion of adolescents in Addis Ababa who smoked, were obese, had a sedentary lifestyle, drank alcohol heavily, had diabetes, and were hypertensive, as well as to identify the social factors associated with these factors. The study also assessed the knowledge that adolescents had on the precursors of atherosclerosis, and on the hazards of smoking and of heavy alcohol consumption. Materials and methods Study sample The study was conducted among high-school stu- dents in Addis Ababa (current population: 1.8 mil- lion, of whom 27.6% are aged 10-19 years) (2). There are 37 high-schools in the city that are atten- ded by 116 704 students (Ministry of Education, unpublished statistics, 1988). The study population consisted of a random sample of students from grades 9, 10, and 11 who were attending govemment-owned high-schools in the city in the 1989-90 academic year. There are 20 such schools in Addis Ababa that are attended by 110 790 students (95% of the total high-school stu- dents in the city). Excluded from the study were four special (prison and vocational) schools (total popula- tion, 2479 students (2.1%)) and 13 private schools (3435 students (2.9%)). A cluster sampling method, proportionate to size, was employed to select the study population. The procedure was as follows. A list of schools and sections within the schools and their numbers of stu- dents was obtained from the Addis Ababa regional branch of the Ministry of Education. This was used as the first-level sampling frame. Since the sections differed in size, we calculated the cumulative popu- lation and a sampling interval (cumulative total population/number of clusters). After identifying one section using a table of random numbers, we sys- tematically selected 28 clusters (sections). These sections (classes) were located in 17 high schools. At the next level we used a systematic random sam- pling procedure to select 20 students from each of the 28 clusters. This procedure gives an unbiased sample estimate (7) and is widely used in health sur- veys.a Questionnaires A total of 519 (92.6%) out of the 560 students select- ed filled out the study questionnaires, which were in Amharic and self-administered. In order to increase the validity of the response, the questionnaires were administered anonymously (8). The questionnaires were first pretested among a group of high-school students in Addis Ababa (who did not participate in the study) before being used in the survey. Closed items were used in the questionnaires to collect the following information: data on demo- graphic, social, and medical characteristics of the students; their knowledge about the dietary and behavioural factors known to be precursors of ath- erosclerotic and hypertensive diseases; and their knowledge about the precursors of atherosclerosis, the hazards of smoking and of excessive alcohol consumption. Analysis of the results Students were classified as "smokers" if they report- ed currently smoking one or more cigarettes a day, and "hypertensive" or "diabetic" if they reported having been told by a doctor to have the condition. A physical activity index was used to classify students as "sedentary" by scoring their response to a 13-item physical activity questionnaire. Students were asked to state whether they engaged in one or more of the following activities and if so, for how many hours per week: leisure time physical activities (walking, tennis, gymnastics, running, and cycling) and house- hold activities (washing clothes or floor, chopping logs, grinding grain, digging, and climbing stairs). The physical activity items varied between those that required 5 kcal/min (moderate) to 14 kcal/min (strenuous) energy expenditure. The number of hours per week spent on the various activities were summed. Students were classified as sedentary if they did not regularly engage in at least 2 hours of moder- ate or 1 hour of strenuous physical activity per week, an approach that has been used in other studies (9, 10). Self-reported height and weight data were used in the study; the validity and reliability of such data have been established in several studies (11, 12). a Diarrhoea morbidity, mortality, and treatment procedures: household survey manual. Unpublished WHO document CDD/ SER/86.1 Rev. 1. 788 WHO Bulletin OMS. Vol 71 1993 Precursors of atherosclerosis and hypertension in Addis Ababa There is no universal consensus on the definition of obesity in adolescents (13). In the survey, we there- fore used two definitions of obesity. Following the recommendations of a WHO meeting, we defined obesity using the body mass index (BMI) (kg/m2), with BMIs of 20, 25, and 28, respectively, as cut-off points for classifying < 14-year-olds, 15-year-olds, and . 16-year-olds as obese (5). On the other hand, a report on obesity by the Royal College of Physicians of London (14), recommended that BMI should not be used and that instead a reference weight-for- height value derived from studies performed on U.S. children in the 1920s (15) should be employed. Because only median weight-for-height values are reported, 95% cut-off points cannot be used. Using this approach we classified students as obese if they exceeded 100% of the median values. These are close to the 120% cut-off point recommended by the Royal College of Physicians and, we feel, are appro- priate for Ethiopia. Comparison of our results on obesity obtained using these two approaches indicated that the agree- ment was 76.7% (K = 0.25). The proportions of stu- dents classified as obese according to the method used are shown in Table 2. A semi-quantitative food-frequency question- naire (FFQ) was used to assess the students' dietary fat intake and alcohol consumption. The validity and usefulness of FFQs for classifying people according to their level of nutrient intake have been established in several studies (16, 17). Ethiopian food composi- tion tables were used to calculate the nutritional composition of each food item (18) and these were summed to obtain an individual's usual nutrient intake level. The dietary fat intake as a proportion of the total caloric intake was used to categorize adoles- cents into quartiles of fat intake. Proportions were used instead of absolute values in order to control for body size and physical activity differences between adolescents. The frequency of intake and amount of the vari- ous types of alcoholic beverages consumed were scored and summed using the FFQs. We defined heavy intake to be regular consumption of 20 ml of ethanol per day, which corresponds to 50% of the cut-off value used by the Centers for Disease Control for classifying adults as chronic heavy drinkers (19). To assess the students' knowledge about precur- sors of atherosclerosis, we used a 10-item question- naire on the role of smoking, hypertension, diabetes, obesity, physical exercise, high dietary intake of fat (including animal fat), salt, and sugar; and blood cholesterol levels. Students who correctly responded to .75% of the items were classified as having ade- quate knowledge. To assess the students' knowledge about the hazards of smoking cigarettes, we used a six-item questionnaire on the role of cigarette smok- ing on lung cancer, chronic lung disease, bladder cancer, cancer of the pharynx/larynx, low birth weight, and pregnancy wastage. To assess students' knowledge on the hazards of heavy alcohol intake, we used an eight-item questionnaire on the role of alcohol in liver disease, cancer of the liver, bladder cancer, diseases of the joints, pregnancy wastage, congenital malformation, mental retardation, and low birth weight. Students who responded correctly to >75% of each of the groups of items on smoking and alcohol were categorized as having adequate knowl- edge. Data were processed using SAS software on a microcomputer (20). The significance of associations was examined using X2 and Fisher's exact tests (21). Results The response rate for the study was 92.6% (519/560). Of the students who replied, 46.8% (n = 243), 37.4% (n = 194), and 15.8% (n = 82), re- spectively, were from the ninth, tenth, and eleventh grade. The sociodemographic characteristics of the study population are shown in Table 1. The age of Table 1: Sociodemographic characteristics of the study participants, Addis Ababa, 1990 Total Age (years)a 10-14 15-19 Ethnicity Oromo Amhara Gurage Tigrai Other Parental incomec <200 200-700 >700 School grade Nine Ten Eleven Total No. of No. of males females 144 (27.7)b 75 (26.1) 69 (29.7) 375 (72.3) 212 (73.9) 163 (70.3) 84 (16.2) 261 (50.3) 64 (12.3) 74 (14.3) 30 (5.8) 247 (47.6) 200 (38.5) 53 (10.2) 243 (46.8) 194 (37.4) 82 (15.8) 48 (16.7) 139 (48.4) 35 (12.2) 43 (14.9) 17 (5.9) 149 (51.9) 101 (35.2) 30 (10.4) 124 (43.2) 114 (39.7) 49 (17.1) 36 (15.5) 122 (52.6) 29 (12.5) 31 (13.4) 13 (5.6) 98 (42.7) 99 (42.7) 23 (9.9) 119 (51.3) 80 (34.5) 33 (14.2) 519 (100.0) 287 (100.0) 232 (100.0) a Only 7 students (5 of them female) were aged < 13 years. b Figures in parentheses are percentages. c Per month in birr (1 birr = US$ 0.48). WHO Bulletin OMS. Vol 71 1993 789 D. Kebede & T. Ketsela the students lay in the range 10-19 years, 72.3% (n = 375) being 15-19 years of age. A total of 55.3% (n = 287) were males, and 50.3%, 16.2%, 14.3%, and 12.3%, respectively, were from the Amhara, Oromo, Tigrai, and Gurage ethnic groups. The par- ents of 47.6% (n = 3247) of the students had a com- bined monthly income of less than 200 birr (1 birr = US$ 0.48), while only 10.2% (n = 53) had parents whose income exceeded 700 birr. A total of 13.8% (n = 72) of the students report- ed that they currently smoked one or more cigarettes a day, 11.6% (n = 60) were sedentary, and 9.2% (n = 48) consumed alcohol heavily (Table 2). In terms of their BMI, 25% (n = 130) were obese; how- ever, in terms of their reference weight-for-height values only 14.1% (n = 73) were obese. Altogether, 30.3% (n = 157) had one risk factor, while 4.4% (n = 23) had two risk factors of atherosclerosis. A higher proportion of males than females smoked and drank heavily (14.6% versus 12.9% and 9.7% versus 8.6%, resp.), but these differences were not statisti- cally significant. Females, on the other hand, were more sedentary (12.5% versus 10.8%) and obese (24.6% versus 5.5%) than males, and for obesity the difference was highly statistically significant (P <0.001). Furthermore, the prevalence of having one risk factor or two risk factors was higher for females (38.4% and 6.9%, resp.) than males (23.7% and 2.4%, resp.) Both these differences were statisti- cally significant (P <0.001 and P = 0.025, resp.). Hypertension and diabetes that had been diag- nosed by a doctor were reported by 1.7% (n = 9) and 1.0% (n = 5), respectively, of the students. Most of the students (64% (n = 334)), however, had never had their blood pressure measured. A total of 58.0% (n = 301), 62.0% (n = 322) and 51.4% (n = 267) of the students, respectively, had inadequate knowledge about the precursors of ath- erosclerosis, the hazards of smoking, and of heavy alcohol consumption. The differences between males and females in this respect were small and not satis- tically significant (P >0.05). As shown in Table 3, the proportion of students who smoked was highest (20.7%) among those from the wealthiest background and lowest (14.2%) among the poorest. This was also true for students Table 2: Precursors of atherosclerosis and hypertensive diseases, and knowledge about them among the study participants, Addis Ababa, 1990 Smoked Sedentary lifestyle Obese by: BMlb Weight-for-height Heavy alcohol consumption No. of risk factors Onec Twoc Hypertension Diabetes Inadequate knowledge on: Precursors of atherosclerosis Hazards of smoking Hazards of alcohol Total Total 72 (13.8)a 60 (11.6) 130 (25.0) 73 (14.1) 48 (9.2) 157 (30.3) 23 (4.4) 9 (1.73) 5 (1.0) 301 (58.0) 322 (62.0) 267 (51.4) 519 No. of males No. of females P-value 42 (14.6) 30 (12.9) 31 (10.8) 29 (12.5) 58 (20.2) 72 (31.0) 16 (5.5) 57 (24.6) 28 (9.7) 68 (23.7) 7 (2.4) 6 (2.1) 4 (1.4) <0.001 20 (8.6) 89 (38.4) 16 (6.9) 3 (1.3) 1 (0.4) <0.001 0.025 168 (58.5) 133 (57.3) 175 (60.9) 147 (63.3) 140 (48.7) 127 (54.7) 287 232 a Figures in parentheses are percentages. b BMI = body mass index. c Any one of the following behavioural precursors: smoking, sedentary lifestyle, obese, and heavy alcohol consumption. WHO Bulletin OMS. Vol 71 1993790 Precursors of atherosclerosis and hypertension in Addis Ababa Table 3: High-risk behaviours, by sociodemographic status, level of dietary fat intake, and knowledge of the precursors of atherosclerosis among the study participants, Addis Ababa, 199oa No. with: No. of No. No. No. who drank One risk Two risk smokers sedentary obese alcohol heavily factor factors Parental income b <200 35 (14.2),' 34 (13.7) 33 (13.3) 20 (8.1) 77 (31.2) 10 (4.0) 200-700 25 (12.5) 15 (7.5) 28 (14.0) 22 (11.0) 56 (28.0) 9 (4.5) >700 11 (20.7) 8 (15.1) 8 (15.1) 5 (9.4) 18 (33.9) 2 (3.7) Dietary fat intake (quartile) Upper 22 (15.9) 13 (9.4) 21 (15.2) 14 (10.1) 40 (28.9) 7 (5.1) Medium 36 (12.8) 38 (13.5) 39 (13.8) 28 (9.9) 91 (32.4) 14 (4.9) Lower 14 (14.0) 9 (9.0) 13 (13.0) 6 (6.0) 26 (26.0) 2 (2.0) Knowledge Inadequate 43 (14.3) 38 (12.6) 43 (14.3) 19 (6.3)d 91 (30.2) 9 (2.9) Adequate 29 (13.3) 22 (10.0) 30 (13.8) 29 (13.3) 66 (30.2) 14 (6.4) Total 72 60 73 48 157 23 a Missing values are not shown; this accounts for the differences in totals for the various groups defined by risk variables. b Per month in birr (1 birr = US$ 0.48). c Figures in parentheses are percentages. d P<0.01. with a sedentary lifestyle (15.1% versus 13.7%), who consumed alcohol heavily (9.4% versus 8.1%), who were obese (15.1% versus 13.3%), and who had one risk factor (33.9% versus 31.2%). These differences were not, however, statistically significant. Although the differences between the highest and lowest income groups were consistent for all the high-risk subgroups, the values for the middle income group were not intermediate between those of the highest and lowest income groups. The proportion of the study population who smoked was higher among those in the upper quar- tile for fat intake than the lower quartile (Table 3). Similarly, the proportions of students who were sed- entary (9.4% versus 9.0%), obese (15.2% versus 13.0%), heavy alcohol consumers (10.0% versus 6.0%), had one risk factor (28.9% versus 26.0%), and had two risk factors (5.1% versus 2.0%) were greater among those whose dietary fat intake was in the upper quartile than the lower. The differences were not, however, statistically significant (P >0.05). Here also, although there was a consistency in the differences between the upper and lower quartiles of intake, the values for the medium intake group were not between those of the upper and lower income groups. The proportion of adolescents who smoked was higher among those whose knowledge of the precur- sors of atherosclerosis was inadequate rather than adequate (14.3% versus 13.3%). This was also the case for those who were sedentary and obese. On the other hand, the proportions who were heavy con- sumers of alcohol and had two risk factors were great- er among students whose knowledge was adequate rather than inadequate. For alcohol consumption the difference (13.3% versus 6.3%) was statistically sig- nificant (P <0.01). The proportion of smokers was higher among those whose knowledge about the hazards of smok- ing was adequate rather than inadequate (17.3% ver- sus 11.8%), but the difference was not statistically significant. Similarly, the proportion of students who consumed alcohol heavily was higher among those with adequate knowledge of the associated hazards than among those whose knowledge was inadequate (10.7% versus 7.8%). Discussion A substantial proportion of the adolescents in the study exhibited precursors of atherosclerotic and hypertensive diseases; these were associated with high socioeconomic status and inadequate knowl- edge about the precursors. The age-specific school-enrolment ratio (number of students: population of a specific age, expressed WHO Bulletin OMS. Vol 71 1993 791 D. Kebede & T. Ketsela as a percent) for adolescents in Addis Ababa is high. In 1984, it was 92.8%, 87.4%, 84.8%, 79.6%, 72.2%, and 65.5%, respectively, for those aged 13, 14, 15, 16, 17, 18, and 19 years (2). There were hardly any differences in this respect between males and females and our findings can, therefore, be general- ized to the adolescent population of the city. The results also compare favourably with reports of previous studies carried out in developing coun- tries. In 1978-80 the proportion of male adolescents who smoked on a daily basis was 38% in Gondar, Ethiopia (6), 71% in Senegal, and 17% in Nigeria (22). In industrialized countries the proportions of adolescents who smoke are higher; for example, in Finland in 1985, 30% of 18-year-olds smoked daily (23) and in the USA 26% of tenth-graders were reported to be smokers in 1987 (24). For females the proportion of smokers in our study (12.9%) is higher than the proportion in the above-mentioned Gondar (3%) and Nigerian (5%) studies, but less than that from Senegal (52%). Our findings for both males and females may be underes- timates because some adolescents may not have been willing to volunteer information about whether they smoked. If this is the case, the results indicate that females are now increasingly taking up smoking, a phenomenon that occurred in developed countries between 1965 and 1985 (25). Female teenagers in Addis Ababa could regard smoking to be a sign of equality with their male peers or with females in industrialized countries. Several studies on adults in both developing and developed countries have reported that the preva- lence of obesity among males ranges from 3.1% in Nicaragua to 12% in USA and among females from 1.5% in El Salvador to 32.0% in Trinidad (26) These studies used a value of BMI .30 as a cut-off point for obesity. In contrast, there have been few studies on the prevalence of obesity among adolescents; this is partly due to the lack of consensus on the defini- tion of obesity for this age group. A recent study in Taipei, China (Province of Taiwan), reported that 2.9% of males and 1.7% of females in the age group 15-29 years were obese (19), which was defined as those who were .120% of their ideal weight using the mid-value of the Metropolitan Life Insurance tables. A 1985 study in Finland reported a mean BMI in the range 19.9-21.7 for those aged 15-18 years (27). Our finding that 5.5% of the males were obese, which is considerably less than the value for females (24.6%), compares favourably with the results of other studies. A recent study in the USA reported that 29% of the adolescents surveyed were sedentary (24). In Finland 29% of 18-year-olds were reported to be sedentary (28), as were 85% of males and 82.1% of females aged 15-29 years in China (Province of Taiwan) (19). In our study only 11.6% of the adoles- cents had a sedentary lifestyle, although we included household activities in addition to leisure time physi- cal activity in our definition. Most of the studies in developed countries have included leisure time phys- ical activity only. In some countries up to 30% of adolescents reg- ularly consume alcohol (6). Although moderate alco- hol intake has not been shown to be associated with atherosclerotic disease, adolescents who drink alco- hol even moderately may become heavy drinkers in adulthood and heavy consumption does cause ath- erosclerotic and hypertensive diseases (3). Population surveys on hypertension carried out in developing countries have reported prevalences in the range 1-34%; for example, among Zulu men in 1976 the prevalence was 1%, among urban Bantu men it was 33% in 1983-84 (29), and among Ethio- pian males it was 3-11% and among females, 2-3% (30). The nine cases of hypertension among our study subjects are difficult to interpret because 64% of the students had never had their blood pressure measured. It is, however, unlikely that the study stu- dents erroneously reported having the condition since in Amharic the term is fairly specific and we enquired for a diagnosis made by a doctor. Individu- als who are diagnosed as hypertensive in adulthood have relatively higher blood pressure also as adoles- cents (6). Over 30% of the adolescents had one risk factor and 4.4% had two risk factors for atherosclerosis. These proportions are comparable to the findings of the above-mentioned study, in China (Province of Taiwan), in which 22% had one risk factor (19). The risk factors tended to be more prevalent among the higher income than the lower income groups. This is to be expected, since the adolescents whose parents were wealthy were usually able to afford a more affluent lifestyle. In Ethiopia obesity usually connotes a state of well-being, and adoles- cents from upper-income families do not usually per- form household chores. Also, economic reasons may discourage adolescents from lower-income families from starting to smoke or consuming large amounts of alcohol regularly. No firm conclusions can be drawn, however, since the findings for the middle income group did not lie between those for the upper and lower income groups. Our results also show a positive (albeit small) association between an inadequate knowledge about the precursors of atherosclerosis and smoking, and between a sedentary lifestyle and obesity. More importantly, over half the adolescents had inadequate knowledge about the precursors of atherosclerosis, or about the hazards of smoking and of drinking alco- WHO Bulletin OMS. Vol 71 1993792 Precursors of atherosclerosis and hypertension in Addis Ababa hol, on the one hand, and alcohol consumption and smoking, on the other. This could indicate that health education may be effective in the study area if institu- ted early enough. The negative association between inadequate knowledge about the precursors of athero- sclerosis, the hazards of smoking, and of drinking al- cohol, on the one hand, and alcohol consumption and smoking, on the other, could indicate that although those who smoked and drank knew about the hazards they were unable to stop practising these habits. Although the inherent limitations of the FFQ method preclude precise quantification of dietary fat intake, our findings indicate a positive association between a relatively high intake and the other high- risk behaviours. Dietary fat intake is an important determinant of atherosclerotic changes in blood ves- sel (1). Thus, our results indicate the importance of an early dietary intervention programme. A possible limitation of our study is its use of self-reported data. For logistical reasons, population- based studies usually rely on such data. Methodolog- ical studies conducted to evaluate the validity of self- reported information on smoking (5), weight and height (11, 12) and diet (16, 17) have shown, never- theless, that valid data can be obtained using such an approach. In our study the use of pre-tested and anonymous questionnaires makes it unlikely that the students volunteered inaccurate information. Also, non-response was minimal (7.4%). Although stu- dents with high-risk behaviours may have tended not to respond, this would imply that the true preva- lences of such behaviours were greater than those we have reported here, underscoring the gravity of the situation. With increasing economic development, the importance of commercial sectors in smaller towns and of wealthy farmers in rural areas will increase. It is thus reasonable also to search for the emergence of the precursors of atherosclerotic and hypertensive diseases in smaller towns and rural areas in Ethiopia. A substantial proportion of adolescents in Addis Ababa therefore exhibited the precursors of athero- sclerotic and hypertensive diseases and possessed inadequate knowledge about the precursors of ath- erosclerosis. Our findings indicate that unless health interventions are undertaken among this adolescent population to control the emergence and establish- ment of these precursors, the predicted epidemic of atherosclerotic and hypertensive diseases will be realized by the end of the century (1). We, therefore, recommend that similar, larger studies be carried out among adolescents in other urban areas and semiur- ban areas of Ethiopia and that the Ministry of Health and the Ministry of Education give due attention to the primary prevention of these diseases and formu- late plans for appropriate actions. Acknowledgements We gratefully acknowledge the support received from Addis Ababa University and the Ministry of Health, the Addis Ababa Branch of the Ministry of Education, and the principals and teachers of the high schools involved in the study. Mr S. Zewdie is thanked for his assistance with the data processing, Ms Ketsela for entering the data, and Ms M. Kebede for typing the manuscript. Resume Precurseurs de I'ath6roscilrose et des maladies hypertensives chez les adolescents a Addis-Abeba, Ethiopie Un comite OMS d'experts a estime que, d'ici I'an 2000, les pays en developpement connaitraient des 6pidemies de maladies cardio-vasculaires. Comme les precurseurs de l'ath6rosclerose et des maladies hypertensives apparaissent des l'enfance et l'adolescence, les pays en d6veloppement doi- vent agir des maintenant pour empecher la surve- nue de telles 6pid6mies. La pr6sente 6tude a 6t6 consacr6e a la mesure des precurseurs de l'ath6rosclerose chez les adolescents a Addis-Abeba, et a l'6valuation des connaissances de ces adolescents sur ces pr6curseurs ainsi que sur les risques du tabagis- me et de la consommation de quantites impor- tantes d'alcool. Les donnees ont et6 recueillies au moyen d'autoquestionnaires pretest6s chez 519 jeunes selectionnes parmi 1'ensemble des eleves du secondaire d'Addis-Abeba au moyen d'une m6tho- de d'6chantillonnage proportionnelle a la taille des grappes. Au total, 13,8% des adolescents 6taient actuellement fumeurs, 11,6% avaient un mode de vie sedentaire, 9,2% consommaient des quantites importantes d'alcool, 14,1% 6taient obeses, 30,3% avaient un facteur de risque pour l'atheroscl6rose et 4,4% avaient deux facteurs de risque. Dans 1'ensemble, 58% d'entre eux avaient des connais- sances insuffisantes sur les pr6curseurs de l'ath6- rosclerose, et 62% et 51,4% avaient respective- ment des connaissances insuffisantes sur les risques associ6s au tabac et a l'alcool. Nos r6sultats indiquent qu'une proportion notable des adolescents d'Addis-Abeba pr6sen- tent des precurseurs de l'ath6rosclerose et des maladies hypertensives. Des 6tudes analogues devraient etre effectuees dans d'autres villes d'Ethiopie et les Ministeres de la Sante et de l'Education devraient etre attentifs a la prevention primaire de ces maladies et 6laborer des plans d'action appropries. WHO Bulletin OMS. Vol 71 1993 793 D. Kebede & T. Ketsela References 1. WHO Technical Report Series No. 797, 1990 (Diet, nutrition, and the prevention of chronic diseases: report of a WHO Study Group). 2. Population and housing census of Ethiopia, 1984: analytical report for results on Addis Ababa. Addis Ababa, Office of the Population and Housing Cen- sus Commission, 1987. 3. Inter-society Commission for Heart Disease Resources. Atherosclerosis Study Group and Epidemiology Study Group. Optimal resources for primary prevention of atherosclerotic disease. Circu- lation, 70: 153A-205A (1984). 4. Uemura, K. & Pisa, Z. Trends in cardiovascular dis- ease mortality in industrialized countries since 1950. World health statistics quarterly, 41: 155-178 (1988). 5. Tell, G.S. et al. Studies of atherosclerosis determi- nants and precursors during childhood and adoles- cence. Bulletin of the World Health Organization, 64: 595-605 (1986). 6. WHO Technical Report Series, No. 792, 1990. (Pre- vention in childhood and youth of adult cardiovascu- lar diseases: time for action. Report of a WHO Expert Committee). 7. Levy, P.S. & Lemeshow, S. Sampling for health professionals. Belmont, Wadsworth Inc. 1980, p. 250. 8. Vaughan, J.P. & Morrow, R.H. Manual of epidemi- ology for district health management. Geneva, World Health Organization, 1989. 9. Briazgonov, I.P. The role of physical activity in the prevention of and treatment of non-communicable diseases. World health statistics quarterly, 41: 242-250 (1988). 10. Lawrence, M. & Whitehead, R.G. Physical activity and total energy expenditure of child-bearing Gam- bian village women. European journal of clinical nutrition, 42: 145-160 (1988). 11. Stunkard, A.J. & Albain, J.M. The accuracy of self- reported weight. American journal of clinical nutri- tion, 34: 1593-1599 (1981). 12. Palta, M. et al. Comparison of self-reported and measured height and weight. American journal of epidemiology, 115: 223-230 (1982). 13. WHO Working Group. Use and interpretation of anthropometric indicators of nutritional status. Bul- letin of the World Health Organization, 64: 926- 941 (1986). 14. Obesity - a report of the Royal College of Physi- cians. Journal of the Royal College of Physicians of London, 17: 5-65 (1983). 15. WHO Technical Report Series, No. 724, 1985 (Energy and protein requirement report of a joint FAO/WHO Ad Hoc Expert Consultation). 16. Willett, W.C. et al. Reproducibility and validity of a semi-quantitative food frequency questionnaire. American journal of epidemiology, 122: 51-65 (1985). 17. Sampson, L. Food frequency questionnaires as a research instrument. Clinical nutrition, 4: 171-178 (1985). 18. Agren, G. et al. Food composition table for use in Ethiopia, vol. 1. Addis Ababa, Ethiopian Nutrition Institute, 1964-1975. 19. Kloong, S.L. et al. A prevalence survey of behav- ioral risk factors in Taipei city, Taiwan. International journal of epidemiology, 19: 154-159 (1990). 20. SAS/STAT guide for personal computers, version 6. Cary, NC, SAS Institute Inc., 1987. 21. Snedecor, G.W. & Cochran W.G. Statistical methods, 7th ed. Ames, IA, Iowa State University Press, 1980. 22. WHO. Technical Report Series, No. 731, 1986 (Young people's health - a challenge for society. Report of a WHO Study Group on Young People and "Health for All by the Year 2000"). 23. Byckling, T. & Sauri, T. Atherosclerosis precursors in Finnish children and adolescents XII. Smoking behavior and its determinants in 12-18-year-old subjects. Acta paediatrica Scandinavica (suppl.), 318: 195-203 (1985). 24. Results from the National Adolescent Student Health Survey. Journal of the American Medical As- sociation, 261: 2025-2031 (1989). 25. Fielding, J.E. Smoking and women: tragedy of the majority. New England journal of medicine, 317: 1343-1345 (1987). 26. Gurney, M. & Gorstein, J. The global prevalence of obesity - an initial overview of available data. World health statistics quarterly, 41: 251-254 (1988). 27. Dahlstrom, S. et al. Atherosclerosis precursors in Finnish children and adolescents 11. Height, weight and body mass index and skin-folds, and their cor- relation to metabolic variables. Acta paediatrica Scandinavica (suppl.), 318: 65-78 (1985). 28. Telama, R. et al. Atherosclerosis precursors in Fin- nish children and adolescents X. Leisure time physi- cal activity. Acta paediatrica Scandinavica (suppl.), 318: 169-180 (1985). 29. Nissinen, A. et al. Hypertension in developing countries. World health statistics quarterly, 41: 141-154 (1988). 30. Zein, A.Z. & Makonnen, A. Blood pressure levels and hypertension in rural Ethiopian communities. Ethiopian medical journal, 24: 169-178 (1986). 794 WHO Bulletin OMS. Vol 71 1993