Hypertension Control in the Elderly: Too Much of a Good Thing?

The past 15 years have witnessed truly dramatic improvements in hypertension control rates (defined as a systolic blood pressure [BP] <140 mm Hg and a diastolic BP <90 mm Hg). Using data from the National Health and Nutrition Examination Survey (NHANES), Egan and colleagues previously reported that from the period 1988–1994 to 2007–2008 national control rates among all hypertension patients (treated and untreated) increased from 27.3% to 50.1%. Among the subsample of patients taking treatment for hypertension, control rates increased from 50.6% to 69.1% during this same period.1 Within large integrated healthcare systems such as Veterans Affairs and Kaiser Permanente, hypertension control rates in the high 70% to >80% range are now being seen.2 Achieving this level of control has not been easy and requires broad‐based interventions. Foremost has been the alignment of performance measures such as the Healthcare Effectiveness Data and Information Set (HEDIS) measures with guideline‐recommended treatment goals for hypertension. Other strategies successfully employed by these healthcare systems include creation of disease registries, routine use of audit and feedback, shared care with nonphysician healthcare professionals, and increasingly pay‐for‐performance plans. Clinicians are being more aggressive in their treatment of uncontrolled hypertension and therapeutic inertia has been reduced.3 The net result of these efforts should be large declines in cardiovascular morbidity and mortality.

Despite this national success in hypertension control, it is reasonable to question whether similar improvements have been seen in specific subpopulations known to be at risk for poor control. The elderly, particularly patients older than 80 years, is one such group. Studies have consistently shown that older patients with hypertension are much less likely have controlled BP, that many physicians believe that hypertension treatment in the very old has more risks than benefits,4 and that many elderly hypertension patients, when given these risks and benefits, will refuse treatment intensification.5 The article by Bromfield and colleagues⁶ in this issue of The Journal of Clinical Hypertension provides some reassurance. 6Using NHANES data from 1988–2010 of people 80 years and older, they demonstrate similar large improvements in hypertension control rates. The control rate of 53.1% seen in 2005–2010 seems similar to the rate reported in the full NHANES sample. These impressive control rates in the elderly were partly the result of more intensive medication management, with the percentage of hypertension patients taking ≥3 medications quadrupling.

Clearly the medical community has responded to guideline‐derived performance measures that hold them accountable for achieving BPs <140/90 mm Hg. Yet the unintended consequences of quality measurement are well‐recognized and include a loss of professionalism, and potential patient harm, when clinicians focus on achieving the performance measure rather than what may be best for the patient.7 The use of specific all‐or‐nothing thresholds to define good quality of care is particularly problematic as it encourages overtreatment and overdiagnosis.8 Such overtreatment is evident in this sample of elderly hypertension patients.6 A total of 19% had a systolic BP <120 mm Hg and 49% had a diastolic BP <60 mm Hg. Moreover, one must wonder whether the reported near doubling (from 4.2 to 7.3 million) in the number of US adults aged 80 and older with hypertension reflects, in part, the initiation of therapy in people with prehypertension in order to avoid potential poor results on performance measures.

The hazards of overtreatment remain a concern and include adverse drug events, falls, and increased mortality. Randomized clinical trials in the elderly, albeit of limited power, persist in failing to demonstrate a benefit of treating systolic BP to ≤140 mm Hg when compared with a more conservative threshold of 150 mm Hg.9, 10 Observational studies of elderly hypertension patients have long shown increased mortality at achieved BPs <140/90 mm Hg.11 A secondary analysis from the International Verapamil SR‐Trandolapril Study (INVEST) of hypertension patients with coronary artery disease suggested that the optimal BP for patients aged 80 and older was a systolic BP of 140 mm Hg and a diastolic BP of 70 mm Hg.12 The hazard ratio for the composite primary outcome of death, myocardial infarction, and stroke was increased at both higher and lower BPs. In response to the uncertainty conveyed by these data, the recently published the Eighth Joint National Committee (JNC 8) guidelines recommend raising the target systolic BP goal in hypertension patients older than 60 years to 150 mm Hg.13

It is important to recognize that not all elderly patients are the same. Aging is characterized by a progressive, but highly variable, decline in the body's homeostatic mechanisms, often resulting in disability (dependence in activities necessary for independent living) and frailty. Frailty itself is a distinct biologic syndrome of decreased reserve characterized by weight loss, sarcopenia, weakness, poor endurance, and slowness in gait. The extent of disability and frailty might impact the desired BP threshold for elderly patients with hypertension. Using data on elderly people from NHANES, it was recently demonstrated that higher systolic BP was associated with increased mortality in patients with a fast gait speed, while in patients with a slow gait speed there was no association between mortality and BP. In patients unable to do the walk test, higher systolic BP was associated with a lower risk of death.14 Similarly, in the Leiden 85‐Plus Study, higher systolic BP was associated with a lower risk of stroke in those specific patients with physical and cognitive impairment.15

These results certainly suggest that in treating elderly patients with hypertension, one size may not fit all. Care will need to be taken in how JNC 8 recommendations are translated into the performance measures by which clinicians will be evaluated. It is often said that it is easier to develop performance measures that capture poor care than for excellent care. This will certainly be the case for a performance measure on optimal treatment for the elderly hypertension patient. Perhaps as more data on the elderly become available from clinical trials such as the Systolic Blood Pressure Intervention Trial (SPRINT), we will be able to better understand the nuances of hypertension care in frail elderly patients and what defines optimal hypertension care. We may also learn more on the risks of overtreatment and be able to advocate for a specific performance measure of hypertension overtreatment. Such a measure has been recently suggested for diabetes care in the elderly where recent clinical trials have called into question earlier HEDIS performance measures advocating a hemoglobin A_1c <7%.16 Until better data are available, clinicians should exercise their best judgment based on available evidence in tailoring a treatment target for their elderly patients with hypertension that minimizes the risks of both undertreatment and overtreatment.