RSS Feeds
A Practice Guideline Revisited: Screening for Hypertension
- From the Washington University School of Medicine, St. Louis, Missouri. Requests for Reprints: Linda Johnson White, Director, Scientific Policy, American College of Physicians, Independence Mall West, 6th Street at Race, Philadelphia, PA 19106-1572. Product code 0115123. Current Author Address: Dr. Littenberg: Washington University School of Medicine, Campus Box 8005, 660 South Euclid Avenue, St. Louis, MO 63110.
Abstract
In 1993, the Clinical Efficacy Assessment Subcommittee began evaluation of new topics and reevaluation of previous guidelines on common screening tests, which were published in Annals of Internal Medicine between 1988 and 1990 and republished as a collection in 1991 (Eddy DM, ed. Common Screening Tests. Philadelphia: American College of Physicians; 1991). Of the 11 guidelines contained in Common Screening Tests, only 3 (for breast, colon, and cholesterol screening) will be published with new data supporting new recommendations. Dr. Littenberg's analysis of the evidence that has appeared since the publication of the original paper on screening for hypertension (Littenberg B, Garber AM, Sox HC. Screening for Hypertension. Ann Intern Med. 1990; 112:192-202) presents a new format for updating Clinical Efficacy Assessment Project (CEAP) guidelines that have not been altered by new evidence. This “updated guideline” reports on new published studies and its analysis affirms the approved American College of Physicians recommendations of 1990. The Clinical Efficacy Assessment Subcommittee carried out the internal and external review procedures that are used for all CEAP guidelines. The plan is to keep all American College of Physicians guidelines updated in this way on a regular basis. Direct any comments or suggestions to Director, Scientific Policy, American College of Physicians, 6th Street at Race, Philadelphia, PA 19106.
In 1990, Littenberg and colleagues [1] analyzed the practice of screening for hypertension using sphygmomanometry and found that the costs and risks of screening for mild hypertension were offset by the benefits gained. They recommended that every adult should be screened for hypertension and reported that screening is more cost-effective in older adults. In this report, I update that analysis by reviewing information that has become available since 1990. I did a MEDLINE search using the keywords hypertension and screening. I then reviewed 228 English-language papers published since 1989 for articles offering new information about the assumptions and estimates used by Littenberg and colleagues in 1990. Direct evidence would link sphygmomanometry to relevant, patient-centered, long-term outcomes such as stroke, cardiovascular death, and all-cause mortality. The most direct and convincing evidence of the value of screening would result from a randomized, controlled trial; no such trials have been reported for blood pressure screening. In the rest of this review, I focus on indirect evidence.
Is Hypertension Important?
In 1990, there was little doubt that moderate and severe hypertension were important risk factors for vascular disease. New data on adults from the Framingham study [2] suggest that even a borderline elevation of systolic pressure with normal diastolic pressure (isolated systolic hypertension) may be a risk factor. In 1995, hypertension is still an important condition with substantial implications for personal and public health.
Is Hypertension Detectable and Does Detection Lead to Treatment?
Littenberg and colleagues [1] concluded that sphygmomanometry “is useful in detecting asymptomatic hypertensive patients.” In other words, they considered only patients without manifestations of the sequelae of hypertension, such as stroke. Since then, reports on investigations of both screening (in persons who are not patients) and case-finding (among patients presenting for other reasons) have been published. Several reports show that screening persons who are not patients, such as persons at a shopping center or in the workplace, rarely leads to the treatment of previously untreated persons with hypertension [3-9]. The links between identification and control are often too weak to justify extensive community screening programs. Nonetheless, many patients with hypertension in the United States have become aware of their diagnosis and are receiving treatment [10]. I conclude that hypertension is detectable and that detection can sometimes lead to treatment. However, a good detection program must include a structured process for follow-up and treatment.
Is Treatment Effective?
In 1990, the efficacy of treatment of moderate and severe hypertension in middle-aged adults was unquestioned [1]. No new evidence has challenged the value of treating patients less than 70 years of age who have diastolic pressures consistently greater than 105 mm Hg. Several new trials have concentrated on older persons [11-13]: Each found that treatment significantly reduced the occurrence of strokes; one showed that treatment had a favorable effect on total mortality. Meta-analysis has shown that treatment of patients over 60 years of age produces a significant benefit [14].
What is the Effect on Health of Screening for Hypertension?
Previously published estimates of the efficacy of screening [1] have been qualitatively unchanged over the last several years. Screening an asymptomatic adult for hypertension yields a modest average benefit of less than 3 weeks additional expected life span; the benefit depends on the age and sex of the adult. As it would be in all screening programs, this average benefit is distributed unevenly throughout the screened population. Most persons will receive little or no benefit, but an important minority will reap tremendous advantages in morbidity and mortality averted.
Benefits and Harms
I know of no injuries or fatalities directly caused by sphygmomanometry. However, screening can lead to a cascade of indirect harms, including side effects of medication; labeling as sick an otherwise functional person; and initiation of a diagnostic evaluation that can be directly catastrophic (if it involves invasive tests) and indirectly dangerous (if it leads to inappropriate therapy). For many persons, the balance of harms and benefits will favor periodic screening. However, some adults will reasonably prefer to avoid the potential harms of screening. Individual decisions should be made on the basis of the values and preferences of individual persons.
Health Outcomes Versus Economic Outcomes
In 1990, Littenberg and colleagues [1] found the cost per quality-adjusted year of life gained by screening to vary according to age, sex, and various assumptions in the model. Since 1990, the cost of treating hypertension has increased substantially. The cost of the drug regimen used to treat hypertension is an important determinant of the cost-effectiveness of screening for high blood pressure. For instance, if medications, laboratory tests, and physician care for hypertension costs $1500 per year, the cost-effectiveness of screening 40-year-old women will be a substantial $119 000 per quality-adjusted year of life. If annual care costs only $500, the cost per quality-adjusted year of life decreases to $39 000 (Table 1). Because newer pharmaceuticals are much more costly than older agents, and because most of the trials that show treatment to be beneficial used older, less expensive medications, I fully endorse recommendations to use diuretics and β-blockers as first-line therapy for hypertension [10].
Limitations of this Analysis
I may have missed some information that would substantially alter these recommendations. However, given the high visibility of hypertension in the medical literature and the concordance of these recommendations with others [10, 15], a major omission appears unlikely. The analysis presented here depends critically on the previously published analysis by Littenberg and colleagues [1]. Any errors in that work will be propagated here.
Recommendation
The results of this review suggest that previous recommendations for screening [1] are still appropriate. Specifically, all adults should periodically be offered sphygmomanometry in a case-finding setting. This can be done during most visits to a physician, nurse, or clinic and as part of the periodic health examination. Treatment of hypertension should be done using the least expensive regimen that is consistent with efficacy and safety.
- Copyright ©2004 by the American College of Physicians
