PrimeCuts: This Week in the Journals

July 6, 2009

beachIshmeal Bradley MD

Faculty Peer Reviewed

Let’s face it, the American population is getting older. With the graying of our populace (life expectancy 80 years for females, 74 for males), there has come an increased need for better quality of care and more cost-effective preventive care. The current healthcare fiscal crisis only highlights the need to more carefully spend our healthcare dollars. This week, the journals spent considerable space discussing the particular needs of the elderly and the preventive treatments that target them.

Take osteoporosis, a common condition of the elderly. Current guidelines recommend screening for osteoporosis at age 65 for post-menopausal women, or age 60 for post-menopausal women with certain risk factors (including low body weight and previous fragility fractures). Treatment for osteoporosis today rests heavily on the anti-resorptive agents, like the bisphosphonates. One question remains, should one monitor for response to treatment? In the British Medical Journal this week, Katy Bell et al.[1] examined whether it is necessary to routinely screen post-menopausal women undergoing bisphosphonate treatment with DEXA scans.

This team examined the original Fracture Intervention Trial (FIT) data from 1993 to see if there was much within-person variation in bone density over a 3-year follow up period. The original data set included 6,459 post-menopausal women with low bone mineral densities (≤ 0.68g/cm2). They used bone density measurements at enrollment and annually for three years. They found that patients treated with placebo experienced a net loss of 0.004 g/cm2/year of density (roughly 0.6% decrease from baseline) over the three years, while those treated with alendronate had a net increase in density of 0.0085 g/cm2/year (about 1.3% increase per year). After three years of treatment, the average increase in bone density in the alendronate group was 0.030 g/cm2 compared to an average decrease of 0.012 g/cm2 in the placebo group.

The authors also found that with routine monitoring, there were large measurement variations in bone density in individual patients over time. For example, the standard deviations for the alendronate group versus the placebo group were 0.012 g/cm2 vs. 0.014 g/cm2 respectively. This wide within-person variation may mask the true response to treatment.

The authors concluded that it was not necessary to screen treated women with DEXA scans. Because of the measured within-person variations that may not be clinically significant, the results of the DEXA scan may cause a physician to alter therapy when no change is actually needed. Furthermore, expending resources to monitor an intervention that has been proven to work is not cost-effective. In today’s age of quality control and comparative efficacy research, their work provides another way to cut down on unnecessary medical costs and allocate medical resources to a greater number of people. Unfortunately, what this paper fails to answer is what should the screening interval be, if not during the first three years of bisphosphonate treatment.

Many of those patients on bisphosphonates have probably experienced fractures resulting from falls. While on the inpatient wards, we have all called Physical Medicine and Rehab consults for gait training and walking aids for our elderly patients. The thought is that with these walkers and canes, our patients will be more mobile and less likely to fall at home. Unfortunately, that may not be the case. The New York Times [2] published an article last week showing that a whopping 47,000 Americans are brought to local emergency rooms each year for falls associated with the use of walking aids. Of these falls, 87% involved walkers and 12% canes. The majority of these falls occurred at home. With greater mobility comes the increased risk of falls, especially if the walkers and canes are not used appropriately. What this study shows us is that we have to do more than just offer these walking aids to our patients, but we have to ensure that they are using them appropriately. Remember, a renegade walker can be brutal in the hands of the untrained octogenarian.

Back in Primary Care Clinic, those of us with VA clinic are well aware of the clinical reminder for patients over age 65 that prompts us to order screening abdominal ultrasounds to assess for abdominal aortic aneurysms. In 2005 the U.S. Preventive Services Task Force came out with a class B recommendation to screen high-risk men, i.e. smokers, over the age of 65 with abdominal ultrasounds once. There is a class C recommendation for screening non-smokers. The British Medical Journal this week published the results of two studies looking at the cost-effectiveness of AAA screening. The first study by Ehlers [3] used data from all published studies and the Danish national health registries to statistically model a national screening program. They calculated a cost per quality of life year gained of roughly $71,000, making it not cost-effective to offer routine screening. A significant limitation of this study was that it did not take into account high-risk behaviors, like smoking, that greatly increase the incidence of AAA.

The second study by Thompson [4] was a randomized trial to assess if routine screening was cost-effective at 10 years and whether there was a mortality benefit. In their study, they found that the cost per quality life year gained was $67,000 at four years, $23,000 at seven years, and $12,500 at ten years. The authors noticed that the survival advantage increased over time, with the main costs of the screening program borne up front with initial screening and referrals to vascular surgeons. The costs of screening were outweighed by the costs of emergency surgeries. Mortality related to the aneurysm was halved between the control group and the screened group (OR 0.52). With this in mind, the study authors recommended routine one-time screening for AAA.

And, unfortunately, despite the widespread knowledge and use of CPR, survival rates after in-hospital CPR have not improved. Examining Medicare data from 1992-2005, Ehlenbach [5] and his colleagues found that even though the incidence of CPR did not change during the study period, neither did the overall survival-to-discharge rate. One explanation for this was that CPR was being offered to those who were unlikely to benefit from the intervention because of advanced primary disease. Not surprisingly, increasing age and illness severity were negatively associated with survival rates. On the other hand though, race was found to be a marker of survivability with whites more likely to survive to discharge than blacks and non-white others (19.2% vs. 14.3% vs. 15.9%). The authors discussed several reasons for this racial discrepancy: blacks tended to be hospitalized at institutions with lower overall survival rates; delayed defibrillation; and possible differences in the initial cardiac-arrest rhythm. Nevertheless, their work highlights the need for further research into the health disparities that still plague our healthcare system.

Dr. Bradley is a third year resident in internal medicine at NYU Medical Center.

Reviewed by Judith Brenner MD, Associate Editor, Clinical Correlations


1. Bell, Katy et al. “Value of Routine Monitoring of Bone Mineral Density After Starting Bisphosphonate Treatment: Secondary Analysis of Trial Data.” BMJ 2009;338,b2266.

2. Henry, Derrick. “Study Warns of Hazards for Elderly Using Walking Aids.” New York Times. 30 June 2009., accessed 2 July 2009.

3. Thompson, S. G. et al. “Screening Men for Abdominal Aortic Aneurysm: 10 year Mortality and Cost Effectiveness Results from the Randomised Multicentre Aneurysm Screening Study.” BMJ 2009;338,b2307.

4. Ehlers, Lars et al. “Analysis of Cost Effectiveness of Screening Danish Men Aged 65 for Abdominal Aortic Aneurysm.” BMJ 2009;338,b2243.

5. Ehlenbach, William J. et al. “Epidemiologic Study of In-Hospital Cardiopulmonary Resuscitation in the Elderly.” NEJM 2009;361:22-31.

Leave a Reply

Your email address will not be published. Required fields are marked *