The Egyptian National Hypertension Project (NHP)
Design and Rationale
Abstract Hypertension and its complications appear to be increasingly common in Egypt. The National Hypertension Project (NHP) is a collaborative Egyptian-American effort with the following objectives: (1) to determine the prevalence of hypertension and blood pressure–related preclinical and clinical complications in Egyptian adults, (2) to identify environmental factors associated with high blood pressure, and (3) to build an infrastructure for research and education in cardiovascular disease prevention in Egypt. The NHP surveys were conducted in six of Egypt’s 26 governorates, representing distinct geographic regions. In each of these six governorates, a probability sample of 600 households was surveyed. NHP was conducted in two phases. In phase I a team of specially trained physicians conducted household surveys in which all adult residents (age ≥25 years) were screened for hypertension. In phase II hypertensive adults identified in phase I (ie, those with systolic pressure ≥140 mm Hg and/or diastolic pressure ≥90 mm Hg and/or those receiving antihypertensive drug therapy) along with a randomly selected sample of normotensive control subjects were asked to participate in a more extensive evaluation. This included a detailed history and clinical examination, an electrocardiogram, an echocardiogram, laboratory studies, and skin color reflectance (on a subsample). As the first systematic, national survey of hypertension and its complications in an Arab country, NHP should provide data of great interest to the scientific, provider, and public health communities.
In many economically developing countries, such as Egypt, patterns of illness are changing dramatically. Specifically, communicable diseases are becoming less common, and the incidence and prevalence of noncommunicable chronic diseases, such as hypertension, are rising.1 Indeed, several lines of evidence suggest that hypertension and its complications are a major health problem in Egypt. First, the incidence of blood pressure (BP)–related clinical events such as myocardial infarction, stroke, and end-stage renal disease appears to be increasing dramatically. Vital statistics (such as cause-specific mortality and hospitalization discharge diagnoses) along with clinical observations support this notion.2 Second, preliminary surveys suggest that the prevalence of hypertension may be as high as 24%.3 4
In 1991 the National Hypertension Project (NHP) of Egypt was initiated. NHP is a joint Egyptian-American project supported by the US Agency for International Development and the Egyptian government. In this national survey Egyptian scientists from Cairo University and the Egyptian Ministry of Health collaborate with scientists from the National Heart, Lung, and Blood Institute and the Johns Hopkins University. The principal NHP objectives are (1) to determine the prevalence of hypertension and BP-related complications in Egyptian adults, (2) to identify environmental factors associated with high BP in Egypt, and (3) to establish an infrastructure for research and education in the prevention and treatment of hypertension as well as other cardiovascular disease risk factors.
For several reasons NHP should be of considerable interest and importance to the scientific, medical, and public health communities in Egypt and other countries. First, although hypertension surveys have been completed in certain economically developing countries and regions,5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 no such study of hypertension has been conducted in an Arab country. Whether the prevalence and determinants of hypertension and its complications in Egypt are similar to those identified in Western countries and in other economically developing countries is an important question. Second, Egypt’s population is extremely heterogenous in terms of its racial/ethnic composition, social class, and dietary habits. Furthermore, the types of communities (urban versus rural) and types of geographic environments (ie, coastal, frontier, delta, Upper Nile, and Lower Nile) are quite diverse. Such heterogeneity within the Egyptian population underscores the need for surveying representative samples of these groups and lends itself well to studies that examine the effect of environmental risk factors on the presence of hypertension and its complications. Third, the type and extent of the NHP data collection (eg, standardized BP measurements, echocardiography, electrocardiography, and lipid profiles on population-based samples of hypertensive and normotensive adults) present unique opportunities for descriptive and analytic studies. Finally, NHP has the potential to serve as a valuable model for future surveys of hypertension and other chronic diseases in economically developing countries.
NHP was conducted in a preparatory stage and two subsequent main phases (a household prevalence survey for hypertension [phase I] and a detailed clinic evaluation of all hypertensive subjects and a subset of normotensive individuals [phase II]). Before its implementation the study was approved by an Institutional Review Board convened in Cairo, Egypt.
A period of 10 months (March 1991 through December 1991) was dedicated to preparation of the field survey. During this period coinvestigators were recruited, the organizational structure was established, the study design was finalized, the protocol and manual of operations were written, data collection staff were recruited and trained, and several pilot studies were conducted.
From Egypt’s most recent census conducted in 1986, we obtained a representative sample from which we could estimate the prevalence of hypertension and its complications in Egypt’s adult population (estimated to be more than 20 million in 1990).20 A multistage probability sample of clusters of households in geographically defined areas was prepared. In the first stage six of Egypt’s 26 governorates were selected, each representing a distinct region of Egypt (Fig 1⇓ and Table 1⇓). In brief, Cairo, the capital, was selected as a stratum by itself because a substantial fraction of Egypt’s adult population (13.6%) lives in this large urban area. Port Said was selected to represent the coastal cities, Sharkia to represent the mixed urban-rural regions of the Nile delta, Bani Sweif to represent northern Upper Egypt, Aswan to represent southern Upper Egypt and the border region with Sudan, and El Wadi El Gedid to represent the oases and frontier regions.
In the second stage of the sampling schema (Fig 2⇓) four areas (the capital of the governorate, a village associated with the capital, one rural center [a Markaze], and one village associated with this rural center) were surveyed in the combined urban-rural governorates of Sharkia, Bani Sweif, and Aswan. The capital was selected with certainty, and the other three areas were selected with probability proportionate to size. In Cairo and Port Said, three areas (ie, kisms) were selected based on socioeconomic conditions (ie, one with a high, one with a middle, and one with a low socioeconomic standard of living). In El Wadi El Gedid, three randomly selected areas were surveyed.
In the third stage five enumeration areas were randomly selected. Within each of these five areas, one fifth of the requisite number of households (30 in the areas of the urban-rural governorates and 40 in the areas of all other governorates) were identified for the survey with the use of a systematic sampling procedure with a random start. Thus, a total of 600 households were surveyed in each of the six governorates. In phase I of NHP all household members aged 25 years and older were asked to participate.
All hypertensive individuals (ie, those with an average systolic BP ≥140 mm Hg and/or diastolic BP ≥90 mm Hg and/or those receiving antihypertensive drug therapy) identified in phase I and a representative sample of nonhypertensive control subjects from phase I (matched for sex and enumeration area) were asked to participate in phase II of NHP. Because of the time-consuming battery of investigations, only one control subject was enrolled for every two hypertensive individuals.
The six governorates were studied sequentially over 18 months. In phase I social workers visited the identified households twice: once a week before and then 1 day before the physician survey team arrived. Survey teams of two physicians (one male and one female) screened four to five households per day. Normally, written informed consent was obtained before subjects participated in the survey. Verbal consent was deemed satisfactory in the setting of illiteracy. In addition to BP measurements, basic demographic and socioeconomic information was collected (see Table 2⇓ for a list of phase I data collection items). Two types of interviewer-administered questionnaires were given, one to the head of the household (enumerating all adults 25 years of age and older and collecting “common” data pertaining to all household members) and another to each adult in the household. An average of four BP readings was calculated. If the systolic BP was ≥140 mm Hg and/or the diastolic BP was greater than or equal to 90 mm Hg and/or the individual was currently on antihypertensive drug therapy, he or she was considered to be hypertensive and invited to participate in the phase II study.
All hypertensive subjects identified in phase I as well as a randomly selected sample of normotensive subjects were asked to participate in phase II, conducted in a local clinical center. In each area of each governorate phase II data collection activities began within a few days (generally <1 week) of the completion of phase I. In phase II, data collectors were blinded to participant status (normotensive or hypertensive). At the center, phase II participants were asked to provide more detailed information regarding their medical history. Weight, height, and waist and hip circumferences were measured; then the subject underwent a detailed physical examination. Laboratory samples included blood samples for fasting and postprandial blood sugar, serum lipids, blood urea nitrogen, serum creatinine, uric acid, and serum electrolytes (processed locally, frozen at −30°C, and then transported to Cairo for batch analyses). An aliquot from a 12-hour overnight urine collection was sent to the central laboratory for analysis of electrolytes. At the field center a midstream urine sample was analyzed for protein and sugar as well as examined microscopically for direct and indirect evidence of schistosomiasis (ova, hematuria, and/or casts). The participant then had a resting electrocardiogram (ECG) and echocardiogram (2-dimensional, M-mode, and Doppler). Ambulatory BP measurements, skin color reflectance, carotid wall thickness measurements, glycosylated hemoglobin, and fasting and postprodial insulin were obtained in selected samples of phase II participants.
Measurement and Quality-Control Procedures
Special procedures included (1) standardization of data collection for key variables; (2) training, certification, and recertification of data collectors; (3) direct observation of data collectors in the field; (4) assessments of intraobserver and interobserver variabilities; and (5) rigorous data entry and management techniques. Selected protocols for key NHP variables and procedures are described below.
BP was measured manually by trained and certified observers using a standard mercury sphygmomanometer and following a common protocol adapted from procedures recommended by the American Heart Association.21 In brief, an appropriately sized cuff was placed on the participant’s right arm. After the subject had rested 5 minutes in the seated position, a 30-second pulse was recorded. The cuff was then inflated and the pulse obliteration pressure recorded. For each BP measurement the cuff was inflated to 30 mm Hg above the pulse obliteration pressure. Two BP measurements separated by 30 seconds were then obtained. The final pair of BP measurements was recorded in a similar fashion after completion of the interviewer-administered questionnaire.
All BP observers in NHP were trained and certified in the standardized measurement of BP with criteria adopted from the Trials of Hypertension Prevention.22 The training process included a review of videotapes on BP measurement, supervised BP measurements with the use of a dual stethoscope, and audiotape tests. Recertification courses were held approximately every 6 months. In the field, intraobserver and interobserver repeatability was assessed on separate 5% samples of the phase I participants.
Weight, Height, and Body Circumference Measurements
Weight was recorded by trained staff using a certified double balance beam scale placed on a firm level surface. Height was measured from a Frankfort plane positioned at a 90° angle against a wall-mounted metal tape. All body circumference measurements were taken with an anthropometric centimeter measuring tape. Waist girth was measured from the horizontal plane across the minimum girth between the lowest lateral portion of the rib cage and iliac crest. Hip girth was measured at the level of maximal protrusion of the gluteal muscles.
Cardiologists specializing in clinical echocardiography followed a common protocol. In brief, echocardiography (2-D, M-mode, and Doppler) was performed with a portable machine (OTE Sim 500). Parasternal long-axis and short-axis views at the level of the papillary muscles were obtained. Left ventricular end-systolic and end-diastolic dimensions as well as septal and posterior wall thicknesses were assessed from M-mode measurements under 2-D guidance; measurements were based on an average of three sequential readings. Apical four-chamber and five-chamber views were used for assessment of regional wall motion abnormalities and for Doppler examination, which included recordings of both transmitral and transaortic flow patterns, detailed above. The initial interpretation was done in the field, and the second interpretation of the videotaped tracing was independently performed by another cardiologist at Cairo University Hospital. Replicate tracings were recorded in 10% of participants (5% by the same observer and another 5% by a different observer).
Selected, trained individuals obtained all ECG tracings. A 12-lead resting ECG was recorded on a three-channel recorder. At least three beats were obtained for each lead in addition to a double-speed strip for precise interval estimation. Subsequently, the ECGs were interpreted centrally by trained cardiologists following the Minnesota coding system.23
Data Management and Entry
In the field all forms were reviewed and edited. Subsequently, selected, trained staff entered all data at the central office. A number of techniques were used to minimize data entry errors, including range validation, cross-validation, duplicate entry of key variables (eg, BP), and complete duplicate entry of all data on a 10% sample of participants.
All laboratory personnel were skilled technicians with at least 5 years of experience. For all assays blinded replicate analyses for intratest and intertest variabilities were performed in 10% of assays.
Minimization of Nonresponse
A letter was delivered to each household in the NHP survey informing the residents about NHP and of the date of the scheduled visit. A week before the visit social workers went to the household to answer any questions and confirm the date. On the day of the visit social workers introduced the NHP teams to the household members. Separate days were planned for call-backs to reach absentees. Two call-backs were attempted.
The joint Egyptian-American investigative team consisted of scientists from Cairo University; Johns Hopkins University; the National Heart, Lung, and Blood Institute; and the Egyptian Ministry of Health. A team of cardiologists, organized into various subcommittees, assisted the Principal Investigator in the design, implementation, and analysis of the surveys. Teams of physicians delegated from the Ministry of Health were the principal data collectors in the field and were the data entry personnel. Minimum requirements for the data collectors included a bachelor’s degree in medicine and surgery, willingness to travel to field sites, and a commitment to participate in the project for at least 2 years. Important aspects of the study organization included the high caliber and advanced training of all study personnel and frequent interaction between the Egyptian and American scientists. Key personnel are listed in “Acknowledgments.”
NHP is the first systematic, nationwide survey of hypertension and its complications in an Arab country. As such, it provides an impressive database for addressing a number of issues relevant to the scientific, provider, and public health communities in Egypt and other countries. First, using rigorous methodology, the study will provide information heretofore unavailable in Egypt, namely, accurate national estimates of the prevalence of hypertension in Egypt as well as similar information for specific geographic regions and populations in Egypt (eg, Nubians). Second, variation by region and population, if confirmed, will allow for more focused hypertension detection, treatment, and prevention initiatives in higher prevalence settings as well as scientific inquiries with the goal of explaining such variation. Third, the extent and high quality of NHP observations (eg, BP measurements, ECG, echocardiography, laboratory studies, and skin color reflectance) should help to elucidate the role of putative risk factors for hypertension in Egypt (eg, schistosomiasis, skin color) as well as the role of conventional risk factors such as obesity and dietary sodium. Fourth, the study provides what we believe is the first population-based survey using echocardiography. The infrequent use of antihypertensive medications increases the potential value of these data because the results are unlikely to be confounded by medication use. Finally, the design and organizational structure of NHP could serve as models for future surveys of hypertension and other chronic diseases conducted in developing countries.
An important strength of the NHP study is its organizational structure, in which Egyptian and American scientists have collaborated in the design, conduct, and analysis of the study. Such collaboration allows for rapid protocol development, use of state-of-the-art survey methodology, and efficient training of staff. Among the potential limitations of the study are its cross-sectional design, which limits inferences regarding causality when associations between exposures and diseases are identified. Follow-up of the NHP cohort should permit more satisfactory documentation of these relationships in longitudinal analyses. A second limitation relates to the lack of detailed dietary data. However, the collection of urinary electrolyte excretion concentrations, along with determinations of weight and reported alcohol intake, permit adjustment for the most important known risk factors for hypertension.24 An additional challenge pertains to classification of variables, such as socioeconomic status, that depend heavily on local culture and tradition. In the analyses we intend to correlate hypertension prevalence with common proxies of socioeconomic status, such as education and income, and relevant variables unique to the Egyptian population (eg, presence of electricity and air conditioning). In addition, we will develop and validate scales incorporating the responses to these variables.
Finally, in planning and implementing the protocol we encountered a number of practical challenges (eg, the recruitment of participants, retention of staff, and procurement of supplies). In each instance the problem was addressed in a satisfactory fashion without compromising the scientific integrity of the study. For instance, we relied on local social workers to enhance our response rates, thus ensuring a representative sample. Furthermore, standardization of measurement techniques and frequent training and certification courses allowed us to replace experienced staff with well-trained substitutes.
In summary, the NHP, a collaborative Egyptian-American project, is the first systematic, nationwide survey of hypertension and its complications to be conducted in an Arab country. The extent and quality of NHP data collection procedures should provide an impressive database on BP and other cardiovascular risk factors. Furthermore, it is anticipated that results from NHP will provide the scientific basis for launching national initiatives focusing on the prevention of BP-related clinical complications in Egyptian adults. Such efforts will include public and physician education programs and the development of national guidelines for hypertension prevention, screening, and treatment.
This research was sponsored by the US Agency for International Development; the National Heart, Lung, and Blood Institute; and the Egyptian Ministry of Health and Education. The NHP Investigative Team includes the following members: Principal Investigator: M. Mohsen Ibrahim; Assistant Principal Investigator: Hussein H. Rizk; Clinical Investigators: Wael Abdel Aal, Wafaa El Aroussy, Zeinab Ashour, Hatem El Atroush, Saad Farrag, Soliman Gharieb, Sherif Helmy, Hossam Kandil, Hassan Khaled, Amal Khalifa, Karima Hassan Khalil, Salwa Morcos, Wael El Naggar, Mamdouh Rayan, Hisham Salah, Yasser Sharaf, Sameh Zaghloul, and Khaled Ziayada; Laboratory Investigators: Amal Rizk and Rasheed Bahgat; Egyptian Ministry of Health Field Coordinators: Hussein El Sayed Hussein and Magdy Aziz; National Heart, Lung, and Blood Institute Researchers: Ruth Hegyeli and Edward Roccella; Consultant Epidemiologists: Amal Samy, Paul K. Whelton, and Lawrence J. Appel; Blood Pressure Measurement Consultant: Jeanne Charleston; Data Entry: Amira Salah, Mohammed Aly el Sayed, and Mona Nour; Project Secretary: Lubna Abdel Fattah.
- Received February 13, 1995.
- Revision received March 23, 1995.
- Accepted July 11, 1995.
Dayem MKA, Ahmed A, Hamed R, El Sherbeeny M, Attia M. Ischemic and hypertensive heart disease in Egypt. Ain Shams Medical Journal. 1978;22:1-13.
Department of Cardiology, Cairo University Faculty of Medicine and Qualyub Training and Demonstration Center. Qualyub heart survey. Bull Eg Soc Cardiol. 1958;1:99-105.
Egypt, Arab Republic, Ministry of Health. Health profile of Egypt: narrative report. Ministry of Health, Arab Republic of Egypt; 1979.
Hughes K. Trends in mortality from hypertensive and cerebrovascular diseases in Singapore, 1959-1983. Int J Epidemiol. 1987;16:18-24.
Idahosa PE. Hypertension: an ongoing health hazard in Nigerian workers. Am J Epidemiol. 1987;125:85-91.
Seedat YK, Seedat MA, Hackland BT. Prevalence of hypertension in urban and rural Zulu. J Epidemiol Community Health. 1982;36:256-261.
Simmons D, Barbour G, Congleton J, Levy J, Meacher P, Saul H, Sowbery T. Blood pressure and salt intake in Malawi, an urban rural study. J Epidemiol Community Health. 1986;40:188-192.
M’Buyamba-Kabangu J-R, Fagard R, Lijnen P, Staesseb J, Ditu MS, Tshiana KA, Amery A. Epidemiological study of blood pressure and hypertension in a sample of urban Bantu of Zaire. J Hypertens. 1986;4:458-491.
Gupta SP, Siwachi SB. Epidemiology of hypertension in a north Indian population, based on rural urban community surveys. Jpn Heart J. 1985;37:380-385.
Poulter NR, Khaw KT, Hopwood BEC, Mugambi M, Peart WS, Rose G, Sever PS. The Kenyan Luo migration study: observations in the initiation of a rise in blood pressure. BMJ. 1990;300:967-972.
The Central Agency for Public Mobilization and Statistics (CAPMAS). The Annual Health Report of the Year 1990. Cairo, Egypt; CAPMAS, 1990.
Frohlich ED, Grim C, Labarthe DR, Maxwell MH, Perloff D, Weidman WH. Recommendations for human blood pressure determination by sphygmomanometers: report of a special task force appointed by the Steering Committee, American Heart Association. Hypertension. 1988;11:209-222.
Satterfield S, Cutler JA, Langford HG, Applegate WB, Borhani NO, Brittain E, Cohen JD, Kuller LH, Lasser NL, Oberman A, Rosner B, Taylor JO, Vogt TM, Walker WG, Whelton PK, for the Trials of Hypertension Prevention Collaborative Research Group. Trials of Hypertension Prevention: phase I design. Ann Epidemiol. 1991;1:455-471.
Prineas R, Crow R, Blackburn H. The Minnesota Code Manual: procedures for measurement and classification of electrocardiographic findings. In: Wright J, ed. Clinical Trials and Population Studies. Littleton, Mass: PSG; 1982:1-229.