The risks of treating high blood pressure in the elderly

When it comes to high blood pressure treatment in the elderly, the plot continues to thicken.

Last December, a minor controversy erupted when the JNC hypertension guidelines proposed a higher blood pressure (BP) treatment target (150/90) for adults aged 60+.

And now this month, a study in JAMA Internal Medicine reports that over 3 years, among a cohort of 4961 community-dwelling Medicare patients aged 70+ and diagnosed with hypertension, those on blood pressure medication had more serious falls.

Serious falls as in: emergency room visits or hospitalizations for fall-related fracture, brain injury, or dislocation of the hip, knee, shoulder, or jaw. In other words, we talking about real injuries and real patient suffering. (As well as real healthcare utilization, for those who care about such things.)

How many more serious falls are we talking? The study cohort was divided into three groups: no antihypertensive medication (14.1%), moderate intensity treatment (54.6%), and high-intensity treatment (31.3%).

Over the three year follow-up period, a serious fall injury happened to 7.5% of those in the no-antihypertensive group, 9.8% of the moderate-intensity group, and 8.2% of the high-intensity group. In a propensity-matched subcohort, serious falls happened to 7.1% of the no-treatment group, 8.6% of the moderate-intensity group, and 8.5% of the high-intensity group. (Propensity-matching is a technique meant to adjust for confounders — such as overall illness burden — between the three groups.)

The methodologists in the audience should certainly read the paper in detail and go find things to pick apart. For the rest of us, what are the practical take-aways?

The main one, in my mind, is that when it comes to people aged 70+, there are more risks to treating high blood pressure than are commonly recognized by clinicians and patients.  As the study authors note, real-world Medicare beneficiaries often have more chronic conditions than the older adults who are enrolled in randomized trials of blood pressure treatment.

Reducing the risk of cardiovascular events (the main purpose of treating high blood pressure) is laudable, but it’s been hard to prove a benefit to getting most people’s blood pressure below 150/90.

Given the findings of this study, we should be probably be more careful about starting — and continuing — treatment with blood pressure medications in elderly patients. And we should be especially careful when it comes to patients who seem prone to falls, or who are experiencing blood pressure levels well below the target of 150/90.

Because right now, when it comes to treating high blood pressure in older adults, we are often not careful. Meaning that many clinicians don’t:

  • Ask about falls or near-falls before starting or adjusting blood pressure meds.
  • Get more blood pressure data points before making an adjustment in therapy. The convention is to treat at a visit based on the blood pressure that the staff just obtained. It would be better to base treatment on multiple readings, preferably taken in the patients usual environment.
  • Check on blood pressure soon after making an adjustment in therapy. Often patients have their meds adjusted and nobody checks on things until the next face-to-face visit … which might be 6 months away.
  • Find out what the patient is actually taking before making adjustments. When looking at a given BP number, we should confirm that the patient is actually ingesting the meds we think they are, at the dose we think they are. Needless to say, this isn’t always the case! Also occasionally important to have figured out when medications were taken relative to when the BP was checked.
  • Act to reduce BP meds in vulnerable elders. If a frail older person on BP meds sits in front of me and registers SBP of less than 120, I generally look into things a little more. (I ask about falls, and I check orthostatics.) Why? Because now we seem to be fair ways below my usual target SBP of 140s. Is this person on more medication than they need? Are they dropping their BP into worrisome low range when they stand up?

Now, I’d love to see all primary care clinics for older adults implement the ideas above, but I’m not going to hold my breath. All of these ideas require a little more time, which is tough to find in today’s busy primary care environment.

And that extra time is something that patients and families have to contribute as well. Whether it’s time coming back to the office a little more often, or time tracking BP at home and connecting remotely with the clinical team: until we have the technology and systems to make monitoring and communication much easier, being more careful means patients and families will have to put in a little more effort.

Last but not least, we don’t know if outcomes would improve if the strategies above were routinely used in primary care. Specifically, we don’t know how changing our approach to blood pressure might reduce falls and other bad outcomes in older adults. (This JAMA study found that telemonitoring and pharmacist-managed medication adjustment improved BP control, but it’s a younger population and didn’t study potential harms of treatment.)

Still, I do recommend older adults get a good home blood pressure cuff, preferably one with the tech capabilities to make it easy to share data with a clinical team. If there have been any falls or near falls, taking a closer look at what is happening with blood pressure could very well help.

Less (medication) is often more (safety and well-being).

Leslie Kernisan is an internal medicine physician and geriatrician who blogs at GeriTechThis article originally appeared on The Health Care Blog.

Comments are moderated before they are published. Please read the comment policy.

  • Peter Elias

    I agree completely. In fact, though the guidelines (the opinion piece) have only recently acknowledged the issue, the data has been there for a while: we have good evidence that treating in the 150-160/90-100 range improves outcomes, but scant if any data for treating at pressures lower than that. (And, as a devotee of REAL evidence based medicine, I see evidence as one of three pieces of the individual decisions I make with patients. The other two are my clinical experience and the very important context and values/preferences of the informed patient. The evidence should never be allowed to make decisions in a vacuum. That’s just trading the paternalistic authoritarian physician for a paternalistic authoritarian algorithm.)

    Sadly, the approach you suggest and I support will meet with both active and passive resistance by institutions, EHRs, quality programs and P4P programs that think they can create quality by demanding that we worship specific numeric targets. (The EHR in my institution still tries to enforce a BP of <130/70 in diabetics, an A1c < 7.0 in all diabetics, LDL by numeric target rather than based on risk, and mammograms beginning at 40. And the P4P system in my institution does the same.)

    It takes time to know the patient and help the patient know their body and their illnesses. But it is more important to know the patient who has the disease, than the disease the patient has. (Osler)

    Peter Elias

    • Leslie Kernisan, MD

      Hi Peter, nice to hear from you here!

      I too used to get dinged in my primary care practice, because my elderly patients with diabetes had SBP >130. Very annoying.

      Separately, it’s interesting that EHRs and P4P programs are likely so slow to change that even if we all accept new evidence, it will take forever for those systems to reflect this!

      And of course, I agree w you that the “best evidence” is only a starting point, and that our real work is to adapt what we know in order to effectively partner with our patients, and help them reach their health goals.

  • PrimaryCareDoc

    Now, if only CMS will catch up with the evidence… PQRS still wants us to get people below 140/90.

    • Leslie Kernisan, MD

      Agree that this is a SERIOUS problem, the lag-time between publication of best-evidence and getting the quality measures to update.
      Also a big problem that the quality measures rely on office-based measurements, which tend to misclassify people

  • querywoman

    Hypertensive medicines are the most sacred of modern American medical cows! Questioning is overdue.

  • christythomas

    I’m 64, and several years ago, when my PCP saw that my office BP was 140/85, she wanted to start me on anti-hypertensives. I flatly refused, knowing first of all that I have classic white-coat hypertension, and second, that one single BP measurement meant nothing. Four weeks ago, I wore a 24 hour monitor (am enrolling in a study that is requiring certain baseline measurements) which showed an avg BP of 129/73, avg HR 62. I can’t even imagine what would have happened to me if I had agreed with my PCP. Probably would have fainted several times, and broken something serious.

    • Leslie Kernisan, MD

      Good for you!