In the News

Defensive orders, arterial catheters in mechanically ventilated ICU patients, advanced practice providers, and more.


More than a quarter of orders by hospitalists partially defensive

About 28% of orders for services at 3 hospitals were judged by the hospitalists who ordered them to be at least partially defensive, a recent study found.

Researchers asked 42 hospitalists at 1 tertiary care and 2 community hospitals in Massachusetts to complete a survey about their attitudes toward defensive medicine—i.e., the overuse of medical tests and procedures due to fear of malpractice lawsuits. Hospitalists were shown their orders placed the previous day and were asked to rate, on a 5-point scale, the extent to which the orders were placed mainly due to concerns about malpractice. The researchers obtained itemized order costs, such as for room and board, from the hospitals' cost accounting systems. The outcomes were the percentage of orders and costs due to defensive medicine. Results were published as a research letter in the November JAMA Internal Medicine.

Thirty-six hospitalists filled out surveys and rated 4,215 orders for 769 patients. The median number of orders was 3 per patient and 97 per physician. Twenty-eight percent of the orders and 13% of the costs were at least partially defensive. Compared with physicians with fewer defensive orders, doctors with 10% or more defensive orders generated similar costs per patient and placed a similar number of overall orders. The mean cost per patient was $1,695, of which $226 was defensive. Completely defensive orders represented 2.9% of costs, mostly due to additional hospital days. Physician factors like sex, training, or litigation concerns weren't associated with defensive orders or costs. Compared with a 2008 study, the current study reported higher percentages of defensive medicine but lower percentages of completely defensive medicine, the study authors noted. “Our findings suggest that only a small portion of medical costs might be reduced by tort reform,” they concluded.

Mechanically ventilated ICU patients see no mortality benefit from arterial catheters

ICU patients on mechanical ventilation who had arterial catheters didn't have lower mortality rates than ventilated patients without arterial catheters, a recent study found.

Researchers performed a propensity-matched cohort analysis of data from 139 U.S. ICUs from 2001 to 2008 to examine the association between arterial catheter (AC) use and clinical outcomes. The main outcome was hospital mortality; other outcomes were days requiring vasopressor support, days of mechanical ventilation, ICU length of stay, and rate of packed red blood cell (PRBC) transfusions. The primary cohort comprised 60,975 medical patients who came to an ICU from anywhere but the operating room or postanesthesia unit and who needed mechanical ventilation at any point during their ICU stay. The researchers repeated the main analyses on 9 secondary cohorts to evaluate the generalizability of the findings. Results were published in the November JAMA Internal Medicine.

About 40% of primary cohort patients had an AC; with propensity matching, there was no association between AC use and hospital mortality. The same was true in 8 of the 9 secondary cohorts. In the ninth, all patients requiring vasopressors, the odds of death were 8% higher for those who received an AC than for those who didn't (odds ratio, 1.08; 95% CI, 1.02 to 1.14; P=0.008). In addition, days requiring vasopressors, duration of mechanical ventilation, and ICU length of stay were all greater for patients in the primary cohort who received ACs compared to those who didn't. There was no association between PRBC transfusions and AC use.

The results indicate that either there is no mortality benefit from using ACs or that patients who receive ACs are more likely to die but this tendency is curbed by use of ACs, the authors wrote. The latter explanation is unlikely since results were replicated across multiple analyses and cohorts, however. In the U.S. generally, 36% of patients receive an arterial catheter during an ICU stay, the authors noted, and ACs have been shown to be associated with measurable risks such as infections and limb ischemia.

Half of VHA hospitals employ advanced practice providers

About half of Veterans Health Administration (VHA) hospitals employ advanced practice providers (APPs), with nurse practitioners (NPs) and physician assistants (PAs) often doing similar tasks, a recent study found.

In an observational, cross-sectional cohort study, researchers analyzed survey results from 118 chiefs of medicine and 198 nurse managers from VHA hospitals to determine the typical scope of practice for NPs and PAs. A second goal was to identify outcomes that might be affected by having NPs and PAs on an inpatient medicine service, including patient satisfaction, nurse satisfaction, and coordination of care. Results were published in the October Journal of Hospital Medicine.

Of the 118 medicine services, 47.5% employed APPs, with 48.2% using only NPs, 26.8% using only PAs, and 25% using both. The daily caseload of NPs and PAs was similar at 4 to 10 patients (mean, 6.5 patients). The majority (58.9% of NPs and 65.4% of PAs) functioned somewhat or completely autonomously. There were few significant differences between tasks, with common tasks of both including writing orders (87.9%), coordinating discharge plans (86.7%), performing histories and physicals (82.5%), writing daily progress notes (80.7%), communicating with consultants (83.1%) and primary care clinicians (73.5%), and working with hospitalists (72.8%). Less common tasks were serving on committees (46.4%), championing quality improvement initiatives (40.6%), and conducting research (2.9%). Clinical outcomes, process-of-care measures, and cost-effectiveness were not determined.

More services reported that PAs performed procedures compared to NPs (50% vs. 22%; P=0.02); the same was true for PAs' teaching of nonphysician students (50% vs. 24.4%; P=0.04). PAs were also more likely to work weekends and federal holidays. NP and PA presence wasn't associated with patient or nurse manager satisfaction. Chiefs of medicine ratings of overall inpatient coordination were nonsignificantly higher in facilities with only NPs than in those without either NPs or PAs (P=0.08). Nurse manager ratings of overall discharge coordination were nonsignificantly lower for facilities using only PAs (P=0.06). An editorialist noted that chiefs of medicine and nurse managers aren't necessarily members of an inpatient care team, however, so the validity of the care coordination measure is questionable.

At least in the VHA, NPs and PAs who work in hospitals “are more widely used and have a broader scope of practice on inpatient medicine than previously known or appreciated,” the authors wrote.

AHA statement advises on DVT

A recent scientific statement from the American Heart Association on the postthrombotic syndrome offers evidence-based prevention, diagnosis, and treatment strategies for inpatient and outpatient care.

The statement, which appeared in the Oct. 28 Circulation, begins with an overview of the postthrombotic syndrome (PTS) (including epidemiology, diagnosis, and risk factors) followed by practical recommendations for clinicians caring for patients with deep venous thrombosis (DVT).

Recommendations for primary and secondary prevention of DVT to prevent PTS include using thromboprophylaxis in patients at significant risk for DVT (class I recommendation; C evidence) and providing anticoagulation of appropriate intensity and duration to treat the initial DVT (class I recommendation; B evidence). To optimize anticoagulation, the statement recommends frequent, regular internal normalized ratio monitoring in patients treated with a vitamin K antagonist, especially for the first few months (class I recommendation; B evidence). The effectiveness of new oral anticoagulants compared with vitamin K antagonists is unknown, the statement says (class IIb recommendation; C evidence). Also unknown is the effectiveness of low molecular weight heparin (LMWH) alone compared to LMWH followed by a vitamin K antagonist (class IIb recommendation; B evidence).

Effectiveness of elastic compression stockings for preventing PTS is also uncertain, but the stockings are reasonable to use for reducing symptomatic swelling in patients with proximal DVT (class IIb recommendation; A evidence). The statement also includes recommendations on thrombolysis and endovascular treatment approaches for acute DVT to prevent PTS, including catheter-directed thrombolysis, pharmacomechanical catheter-directed thrombolysis, and balloon angioplasty.

The statement's recommendations on treatment of PTS address graduated and intermittent compression; pharmacotherapy with rutoside, hidrosmin, and defibrotide; exercise training; and endovascular and surgical treatment. For venous ulcer management, the statement says compression should be used (class I recommendation; A evidence), multi-component compression systems are more effective than single-component ones (class I recommendation; B evidence), pentoxifylline can be useful alone or with compression (class IIa recommendation; A evidence), and neovalve reconstruction may be considered in refractory cases (class IIb recommendation; C evidence).

Most inpatient folate deficiency testing is unnecessary

Fewer than 1% of Canadian inpatients tested for folate deficiency were actually deficient, a recent study found.

Researchers retrospectively reviewed inpatient records from 3 academic hospitals in Toronto for all red blood cell folate and vitamin B12 levels taken in 2010. Emergency department patients and outpatients treated at the hospital were excluded from the analysis. The researchers also did chart reviews of patients found to have low red blood cell folate (<254 nmol/L) and decided by consensus the likely indications for testing and etiology of the deficiency. Results were published online Sept. 5 by the American Journal of Medicine.

In 2010, 2,563 red blood cell folate and 3,154 vitamin B12 level tests were performed at the 3 hospitals studied. The mean red blood cell folate level was 1,436±20 nmol/L. Four red blood cell folate levels (0.16%) fell below the normal range, 1 of which appeared to be a testing error. The chart reviews indicated the other 3 folate deficiencies were related to alcohol abuse, a malabsorption syndrome, and decreased oral intake secondary to schizophrenia. All patients with low folate levels were anemic, but just 2 had elevated mean corpuscular volume. The mean vitamin B12 level was 467±10 pmol/L, with 98 levels (3.1%) falling below the lower limit of normal (<138 pmol/L) and 426 (13.5%) in the intermediate range (138 to 221 pmol/L). None of the patients with low folate levels had low or intermediate B12 levels.

The results were similar to those of a 2013 U.S. study that found 0.1% of serum folate levels were deficient in inpatients in 2011 at a large academic medical center. The rarity of folate deficiency is likely the result of grain fortification, the authors wrote, and indicate folate testing should be restricted to “only the most unique scenarios.” Also, folate deficiency as an etiology of macrocytic anemia should be de-emphasized in medical education, and more emphasis should be placed on vitamin B12 deficiency as an etiology, in light of the 1-in-6 low or intermediate levels of vitamin B12 found, the authors wrote.