General Practitioners’ Approach to Hypertension in Urban Pakistan
Disturbing Trends in Practice
Background— Control of blood pressure (BP) remains suboptimal worldwide. The objective of this study was to determine whether (primary) general practitioners’ (GPs) approach to high BP is in accordance with international guidelines.
Methods and Results— We conducted a cross-sectional survey of 1000 randomly selected GPs from urban areas in Pakistan during 2002. A rigorously developed questionnaire on (1) type of practice and (2) detection, (3) evaluation, (4) treatment, and (5) source of information about high BP was administered by trained medical personnel. A total of 1051 GPs were approached, and 1000 (95%) consented to enroll; 766 were male and 655 had been in practice ≥10 years. The average number of patients (SD) seen per day was 48.2 (42.7). Overall, 30.6% (29.0% to 32.3%) and 79.7% (78.3% to 81.0%) of GPs used incorrect BP cutoffs to diagnose hypertension in patients <60 and ≥60 years, respectively. Appropriate therapy for hypertension in the elderly was initiated by only 34.7% (33.0% to 36.3%) of GPs. The use of sedatives either alone (23.8%) or in combination with antihypertensive agents as first-line medication for lowering BP was reported by 45.0% (43.2% to 46.7%). Thiazide diuretics were rarely prescribed (4.2%). Sublingual antihypertensive agents were prescribed by 68.7% (67.1% to 70.3%) of GPs for treating very high levels of BP. The practices of recent graduates from medical school were not better than those of older graduates.
Conclusions— GPs in Pakistan underdiagnose and undertreat high BP, especially in the elderly. Our findings underscore the need for urgent revision of teaching curricula in medical schools with regard to the risks, complications, and management of hypertension, as well as the initiation of widespread and intensive continuing medical education for all physicians involved in the management of patients with hypertension. Particular efforts are needed to encourage the use of low-cost thiazide diuretics as antihypertensive agents in developing countries.
Received September 16, 2004; revision received November 28, 2004; accepted December 14, 2004.
Levels of cardiovascular disease (CVD) are increasing rapidly in the developing world and are predicted to overtake infectious disease in many areas to become the leading cause of death in the next decade or so. This is particularly true for the Indian subcontinent, a population already at high risk of CVD. Hypertension is one of the most common risk factors for CVD and has one of the highest attributable risks for death worldwide.1–3 Despite convincing data about the benefits of blood pressure (BP) lowering and formulation of national and international guidelines on the detection and management of hypertension, data from the developed world at least indicate that control of hypertension remains remarkably poor.4 Only 24% of hypertensive patients in the United States have their BP controlled to the conventionally recommended target of <140/90 mm Hg.5 The reasons for this are not entirely clear but may relate to access to health care, knowledge and practice of healthcare providers, and patient compliance.
Data from the developing world, in particular the Indian subcontinent, in this regard are scanty, with details of the prevalence of the condition just beginning to emerge. The National Health Survey of Pakistan (NHSP) conducted during 1990 to 1994 highlighted the magnitude of the burden of hypertension in Pakistan. Hypertension was shown to affect 18% of adults >15 years and 33% of adults >45 years; however, <3% had their BP controlled to 140/90 mm Hg or below.6
The NHSP also showed that >70% of all hypertensive patients (85% in rural areas) in Pakistan are unaware of their disease, despite the fact that the average number of annual visits to the healthcare provider is 5.8 for women and 4.9 for men. This suggests that factors other than poor access to healthcare providers contribute to the lack of awareness and lack of control of hypertension in Pakistan.7 In addition, ≈80% of those with hypertension would be classified as being at high risk, because of either the elevated BP itself or comorbid conditions, and would therefore require immediate pharmacological intervention. The present study was designed to assess whether the knowledge and attitude of primary care practitioners in Pakistan are in accordance with the Joint National Committee VI (JNC-VI) and World Health Organization-International Society of Hypertension guidelines for management of hypertension5,8,9 and to identify areas for intervention.
We conducted a cross-sectional survey of general practitioners (GPs) in Pakistan to determine their approach toward the detection, evaluation, and treatment of high BP. We also sought to determine differences in the practices of those who had graduated relatively recently (≤10 years) versus older (>10 years) graduates of medical school.
The questionnaire initially was developed in English, then forward-translated to Urdu, and finally back-translated to English. It was then pretested in the pilot study and finalized after necessary amendments.
The study questionnaire was developed containing questions on 5 basic themes: (1) type of practice—location of practice, teaching versus nonteaching, and average number of patients seen; (2) detection of high BP—number of readings of BP, threshold levels of BP, and variation in these levels for the elderly or younger subjects for making a diagnosis of clinical hypertension; (3) evaluation of patients with high BP—whether laboratory investigations for coexisting conditions (diabetes, hyperlipidemia, renal failure) were obtained for hypertensive patients; (4) treatment of hypertension—patient education about deleterious effects of hypertension, advice on nonpharmacological measures to control hypertension, the level of target BP that is aimed for during treatment, and choice of antihypertensive medications; and (5) source of information about antihypertensive medication and ranking of medical journals.
Before the full-scale study was launched, a pilot study of 100 GPs was conducted to ensure feasibility of the study and to calculate the required sample size.
On the basis of the assumptions of results from our pilot study, which showed that at least 30% of GPs used inappropriate medications and did not follow the guidelines for treatment of hypertension, 350 subjects were needed, with error bounds of 2%. Furthermore, to detect a 1.5-fold difference with 80% power and a 2-sided α of 0.05 in practices between recent versus older medical school graduates with ratio of 1:2, a total sample size of 930 subjects would be required. In addition, to detect major differences in GP practices between Karachi (the most populous city, where the main survey [70%] was conducted) and 3 other major cities (Lahore, Quetta, and Peshawar), we also aimed to sample a subset of GPs from each of those cities. Detection of a 2-fold difference with a 2-sided α of 0.05 and 80% power would require ≈85 subjects from each city for comparison with 600 subjects from Karachi. We anticipated a nonresponse rate of ≈10%, so our overall targeted total sample size was 1050 GPs.
A cross-sectional study of the licensed GPs in Pakistan was performed mainly in the city of Karachi, the most metropolitan city of Pakistan, on 700 randomly selected GPs during 2002. In addition, a subset of GPs was selected from each of the 3 other major cities (100 GPs each from Lahore, Quetta, and Peshawar) of Pakistan. We obtained a comprehensive list of 6000 licensed GPs in Karachi registered by the Pakistan Medical and Dental Council, which is a requirement for all practicing GPs. A computer-generated random sample of 720 GPs was selected from this list. In addition, similar lists of GPs were obtained for the 3 other major cities (Lahore, Quetta, and Peshawar) in Pakistan. A total of 330 GPs (110 from each) was randomly selected from this sampling frame. Each GP was visited by a trained research officer who administered the study questionnaire after obtaining verbal consent. Each questionnaire was completed in ≈10 minutes. In case the selected GP was not present in the particular area, his/her replacement was chosen from the same location.
The following definitions were adopted for binary outcomes, which were based on the JNC-VI report on classification, detection, evaluation, and management of high BP5: (1) Correct threshold for the diagnosis of hypertension was defined as a systolic BP of ≥140 or a diastolic BP of ≥90 mm Hg in adults aged 18 years or older. (2) Correct threshold for initiating therapy (nonpharmacological or pharmacological) was defined as a systolic BP of ≥140 or a diastolic BP of ≥90 mm Hg. (3) Correct threshold for initiating therapy (nonpharmacological or pharmacological) was defined as a systolic BP of ≥130 or a diastolic BP of ≥85 mm Hg in patients with diabetes, CVD, or renal disease. (4) Correct target level of BP was defined as systolic/diastolic BPs <140/90 mm Hg, respectively. (5) Correct target level of BP was defined as systolic/diastolic BP of 130/85 mm Hg or lower in patients with diabetes, CVD, or renal disease.
Means and SDs were calculated for continuous variables, and proportions and 95% confidence intervals were calculated for discrete variables. The overall prevalence of outcomes was calculated after applying appropriate weights to samples that were in accordance with the distribution of GPs in each of the 4 major cities in Pakistan. Descriptive analyses of characteristics of GPs were also completed. GPs were grouped as recent (≤10 years) or older (>10 years) graduates of medical schools. We reasoned that this grouping would allow us to study the impact of sequential guidelines management of hypertension that has been emphasized over the past decade.5,10,11 The differences in characteristics between recent versus older graduates and GPs in Karachi versus other cities were assessed by χ2 or Student t test, as appropriate.
A total of 1051 GPs were approached, and 1000 (95%) consented to enroll: 701 from Karachi, 99 from Lahore, and 100 each from Quetta and Peshawar. A total of 766 GPs were males, and 655 had been in practice for 10 years or more. There were no significant demographic differences in terms of gender, teaching versus nonteaching practice, and years since graduation between the GPs who did and did not consent to enroll.
Type of Practice
A total of 161 (15.6%) GPs were affiliated with teaching hospitals (Table 1). The average number of patients (SD) seen by the GP per day was 48.2 (42.7). The characteristics of the GPs and differences between recent versus older graduates are shown in Table 1.
A total of 661 (69.4%) of GPs used the correct BP threshold level for diagnosing hypertension in patients <60 years of age. Contrary to the JNC-VI guidelines, though, 79.7% of the GPs believed that this criterion was higher for those aged 60 years and above: 83.7% of recent graduates versus 77.8% of those in practice for 10 years or more (P≤0.001).
Laboratory tests for screening for diabetes, hyperlipidemia, and renal insufficiency were ordered by 89.8% of GPs. This proportion was higher in older (92.8%) versus recent (83.1%) medical school graduates (P<0.001).
The correct criterion was used for initiating therapy in patients <60 years of age by 64.9% of GPs. In patients >60 years of age, though, GPs appeared to be much more reluctant to follow standard criteria for the initiation of therapy (34.7%). This proportion was higher in older (36.2%) versus recent (31.3%) medical school graduates (P=0.005).
We tried to assess the prescribing practices of GPs by asking them about first-line treatment. Interestingly, the first-choice therapy for lowering BP was reported to be a sedative drug alone by 23.8%, whereas 21.1% prescribed a combination of an antihypertensive drug and a sedative as initial therapy. Only 55.1% appropriately prescribed an antihypertensive drug alone when initiating therapy for hypertension. This pattern of resorting to sedative drugs for the treatment of hypertension was greater in older (25.5%) versus recent medical school graduates (20.1%, P=0.001). Even more concerning was the practice of stopping antihypertensive therapy once BP control was achieved by ≈23% of GPs.
The most commonly prescribed medications were β-blockers (41.9%), angiotensin-converting enzyme (ACE) inhibitors (32.2%), and calcium channel blockers (7.1%). Thiazide diuretics were rarely prescribed (4.2%) by GPs. Approximately 68.7% of GPs reported prescribing sublingual antihypertensive medications on an ad hoc basis for rapidly lowering BP in patients whose BP was considered markedly elevated at presentation to the clinic. Furthermore, 49.4% administered intravenous medications to lower BP in their offices.
The correct levels of target BP were aimed for by 52.3% of GPs. The correct target BP for patients with diabetes, CVD, and kidney disease was aimed for by only 37.4%, 36.8%, and 35.6% of GPs, respectively. These targets did not differ significantly among recent versus older medical school graduates.
Sources of Information
The majority of GPs (63%) relied on representatives from pharmaceutical companies for updates on information about antihypertensive medications. Only 31.1% reported reading medical journals for this purpose. Approximately 95% of GPs expressed a willingness to participate in sessions on updates of hypertension.
Comparison Between Practices of GPs in Karachi Versus Other Cities
Table 2 shows a comparison of the approach of GPs toward management of hypertension in Karachi versus those in Lahore, Quetta, and Peshawar. Although intercity differences were observed in some practices, in general the practices were uniformly suboptimal.
The results of our GP survey, which is representative of medical practice in urban areas of Pakistan, show that 79.7% of GPs believe that the criterion for diagnosing hypertension differs according to age in adult subjects. A high proportion of GPs (45.0%) used sedatives for treating hypertension, and <5% prescribed thiazide diuretics as first-line antihypertensive agents. About 47.7% of GPs reported aiming for a target BP that was >140/90 mm Hg. Sublingual antihypertensive agents were being prescribed by 68.7% of GPs for treating very high levels of BP. The practices of recent graduates from medical school were not better than those of older graduates.
Trial data clearly demonstrate the effectiveness of antihypertensive therapy in reducing the risk of major outcomes, regardless of age, at least in the age ranges recruited to the studies. It is disappointing that such a large proportion of GPs (79.7%) think that thresholds for treatment should differ by age and that this belief was more prevalent among recent medical school graduates (83.7% versus 77.8%, P<0.001). The NHSP showed that despite a high rate of contact with healthcare providers, ≈70% of hypertensive patients in Pakistan are unaware of their condition. Our data suggest that poor knowledge of the criteria for the diagnosis of hypertension among healthcare providers may at least partly contribute to this paradox.12
The guidelines recommended for use by the Pakistan Hypertension League, an affiliate of the World Hypertension League, suggest the same BP thresholds for diagnosing and treating hypertension as JNC-VI. Nevertheless, in addition to failing to diagnose hypertension appropriately, Pakistani GPs appear to be significantly undertreating hypertension. Our results indicate that 47.7% of GPs in Pakistan aim for a target BP that is higher than 140/90 mm Hg, and the majority of GPs (62.6%) aim for targets greater than those recommended by the JNC-VI and -VII for patients with diabetes. This is of particular concern, because the prevalence of diabetes is high in South Asian populations.13,14
It has been established that medications with antianxiety properties only have no role in the treatment of clinical hypertension. It is alarming that 45.0% of GPs use sedatives either alone or in combination with other antihypertensive agents as first-line pharmacological therapy for hypertension. This approach is inappropriate, not only because it contributes to undertreatment of hypertension but also because it increases the risk of drug dependency in subjects in whom use of a sedative may not be indicated.15 The undertreatment of hypertension and inappropriate use of sedatives as antihypertensive agents probably contribute significantly to the poor rates of hypertension control in Pakistan.
The type of antihypertensive agent can have a significant impact on compliance with medications because of affordability.16 Of the 76% of GPs prescribing antihypertensive agents as first-line therapy, <5% prescribed diuretics. Previously, we and others have reported that ACE inhibitors should be the antihypertensive agent of choice in high-risk patients with compelling indications such as chronic kidney disease17; however, the superiority of ACE inhibitors compared with other antihypertensive drugs with regard to comorbid conditions has not been clearly established. Results of the ALLHAT study showed that for the majority of patients with hypertension, treatment with thiazide diuretics (the least expensive class of medication) confers the same benefit on adverse CV outcomes as calcium channel blockers and ACE inhibitors. In fact, diuretics were superior to α-blockers, calcium channel blockers, and ACE inhibitors in the prevention of one or more major forms of CVD, including stroke and heart failure.18,19 Furthermore, it is more important to lower BP to appropriate targets than to focus on the class of antihypertensive agent. For a low-income developing country like Pakistan, where the cost of medication is incurred by the patient, there is no doubt that inexpensive and effective antihypertensive agents such as thiazide diuretics should be a part of most antihypertensive regimens. Unfortunately, thiazide diuretics are not widely available as stand-alone drugs, owing to the fact that these drugs are inexpensive and lack profitability for the industry. This partly may explain the lack of their use. We believe that the federal drug regulation agency in Pakistan needs to mandate production of such effective medications before licensing other expensive brands of medications, which are more likely to result in low compliance because of financial constraints.
Guidelines for treatment of hypertension recommend that immediate therapy of very high BP is warranted only in patients with acute symptoms of end-organ damage. In such cases, intravenous administration of short-acting medications should be done under close monitoring so that the dose can be titrated to prevent excessive lowering of BP, which can cause hemorrhagic strokes in such patients.5 Furthermore, the use of sublingual antihypertensive medications should be avoided because of the risk of complications from unpredictable aggressive BP reduction.20,21 Our results showed that 68.7% of GPs in Pakistan prescribe sublingual antihypertensive agents for patients with very high BP. In addition, 49.4% of GPs administered intravenous antihypertensive medication in their outpatient offices. Although data on complications from such practices are lacking, the wide prevalence of such approaches suggest that complication rates would also be significant. Our findings emphasize the need for proper education of GPs about hazards from such practices.22
We found that the practices of recent graduates from medical school were not better than those of older graduates, and the former were more likely to underdiagnose and undertreat hypertension. It is alarming that contrary to studies from other parts of the world, where recent graduates are more likely to have updated knowledge, even recent graduates from medical schools are practicing inappropriate medicine.23 The situation is probably similar in neighboring developing countries.24
The vast majority of physicians (94.4%) were willing to participate in further education on updated management of hypertension. Our findings emphasize the need for continually updating the curricula of medical schools as well as structured continuing medical education in developing countries. These measures have been shown to be successful in other settings.23,25,26 These measures, coupled with federally mandated availability of low-cost, generic, antihypertensive drugs, are essential if Pakistan is to improve on the detection and treatment of hypertension.
Our study has limitations. It was mainly limited to the GPs residing in Karachi, an urban area, and therefore, may not represent the healthcare providers in rural regions as well as other cities of Pakistan. Still, the burden of hypertension is greater in urban areas of Pakistan, and attempts to address it should begin here.6 In addition, we included subsamples of GPs from 3 other major cities of Pakistan and found that the situation may be worse in many respects in other cities in Pakistan. Thus, the problems identified in our study are generalizable to other urban cities of Pakistan.
It must be borne in mind that our study was not designed to measure the appropriate detection of BP. Our study was designed to assess the knowledge of GPs about the detection and treatment of hypertension. Although 52.3% of GPs claim to aim for reasonable targets of BP, we cannot determine the proportion of GPs who set those targets while treating patients in practice and the proportion of subjects who achieved those targets.27 Judging by the dismal (<3%) rate of BP control in Pakistan, it seems likely that very few GPs are able to achieve these goals.
In conclusion, our study highlights the deficiencies in the knowledge and approach of GPs in Pakistan toward management of high BP. We identified serious limitations in current practice. On the other hand, there was also willingness to participate in further education. It is likely that this scenario is not unique to Pakistan. Our findings emphasize the need for continually updating the curricula of medical schools as well as continuing medical education for more senior physicians.
Furthermore, our study demonstrates an easy but valuable first step in determining what limitations we currently face in practice and how to address them. This type of approach could be adopted in other settings. Increasingly, researchers in the developing world should turn their attention to noncommunicable disease in terms of not simply establishing the burden of disease but also assessing the scope and quality of preventive and treatment practices. We have shown for hypertension that assessment of physician practice is readily doable in the developing world and provide important indicators for intervention and healthcare policy.
This study was supported by grants from the Fogarty International Center, NIH RO3 TWO5657-01A1 (Drs Levey, Jafar, and Chaturvedi) and NIDDK RO1 DK53869 (Dr Levey).
World Health Organization. Disease Statistics. World Health Report 1999: mortality by sex, cause and WHO regions, estimates for 1998. Geneva, Switzerland: World Health Organization; 1999.
Reddy KS, Yusuf S. Emerging epidemic of cardiovascular disease in developing countries. Circulation. 1998; 97: 596–601.
National Health Survey of Pakistan 1990–1994. Karachi, Pakistan: Pakistan Medical Research Council. 1998: 50.
Worrall G, Chaulk P, Freake D. The effects of clinical practice guidelines on patient outcomes in primary care: a systematic review. Can Med Assoc J. 1997; 156: 1705–1712.
Guidelines subcommittee of the WHO/ISH Mild Hypertension Liaison Committee. 1993 guidelines for the management of mild hypertension: memorandum from a World Health Organization/International Society of Hypertension meeting. Hypertension. 1993; 22: 392–403.
Allan K, Murphy P, Singleton S, Edwards R. Audit of diagnosis and management of hypertension in primary care: interpractice variation in prevalence of hypertension is due to inadequate detection. BMJ. 1997; 315: 314; author reply 315.
Seyffart G. Antihistamines and other sedatives. In: Haddad, LM, Winchester, JF eds. Clinical Management of Poisoning and Drug Overdose, 2nd ed. Philadelphia, Pa: WB Saunders; 1990: 839.
Jafar TH, Stark PC, Schmid CH, Landa M, Maschio G, de Jong PE, de Zeeuw D, Shahinfar S, Toto R, Levey AS. Progression of chronic kidney disease: the role of blood pressure control, proteinuria, and angiotensin-converting enzyme inhibition: a patient-level meta-analysis. Ann Intern Med. 2003; 139: 244–252.
Grossman E, Messerli FH, Grodzicki T, Kowey P. Should a moratorium be placed on sublingual nifedipine capsules given for hypertensive emergencies and pseudoemergencies? JAMA. 1996; 276: 1238–1231.
Evans CE, Haynes RB, Gilbert JR, Taylor DW, Sackett DL, Johnston M. Educational package on hypertension for primary care physicians. Can Med Assoc J. 1984; 130: 719–722.
Goldstein MK, Hoffman BB, Coleman RW, Musen MA, Tu SW, Advani A, Shankar R, O’Connor M. Implementing clinical practice guidelines while taking account of changing evidence: ATHENA DSS, an easily modifiable decision-support system for managing hypertension in primary care. Proc AMIA Symp. 2000: 300–304.