Anthrax checklist, email security

Summaries from ACP Hospitalist Weekly.

CDC checklist can help quickly distinguish anthrax from nonanthrax illnesses

A checklist developed by the CDC can rapidly distinguish most anthrax illnesses from those not related to anthrax, a recent study found.

Researchers from the CDC compared clinical presentation of adults with anthrax in case reports published from 1880 through 2013 with that of patients who presented for evaluation at two epicenters of the U.S. anthrax attacks in 2001. They then developed a screening checklist to rapidly assign patients to IV treatment, anthrax-specific diagnostic testing, or oral treatment and postexposure prophylaxis after a wide-area aerosol release of Bacillus anthracis spores. They measured diagnostic test characteristics that included positive and negative likelihood ratios (LRs) and patient triage.

Image by Getty Images
Image by Getty Images

The checklist, which involved both paramedical and medical evaluation, included heart rate, respiratory rate, temperature, mental status, diaphoresis, dyspnea, severe headache, characteristic skin lesions, abnormal lung sounds, and increased girth. Patients were adjudicated via the checklist based first on vital and other signs, next on symptoms, and finally on findings such as abnormal lung sounds and characteristics of skin lesions. The results were published by Annals of Internal Medicine on March 19 and appeared in the April 16 issue.

The study included 408 case-patients who had inhalation, ingestion, and cutaneous anthrax and primary anthrax meningitis, along with 657 controls. Triage according to the checklist, without diagnostic testing, correctly classified 95% of 353 adult anthrax case-patients and 76% of 647 control patients (positive LR, 3.96 [95% CI, 3.45 to 4.55]; negative LR, 0.07 [95% CI, 0.04 to 0.11]; false-negative rate, 5%; false-positive rate, 24%). Up to 5% of case-patients and 15% of controls needed diagnostic testing for triage (positive LR, 8.90 [95% CI, 7.05 to 11.24]; negative LR, 0.06 [95% CI, 0.04 to 0.09]; false-negative rate, 5%; false-positive rate, 11%). Sensitivity of the checklist was higher when only cases of inhalation anthrax were evaluated (97%; 95% CI, 94% to 100%) and when only higher-quality case reports were used (98%; 95% CI, 96% to 100%).

The authors acknowledged that the data on case-patients were limited and incomplete, that reporting bias favored more severe anthrax cases, and that most exposures were accidental, among other limitations. However, they concluded that the checklist in this study can help distinguish anthrax from nonanthrax cases according to signs and symptoms and thereby minimize medical and diagnostic resources in the event of wide-area aerosol release of Bacillus anthracis spores. “Its use in such a setting should improve response logistics and patient outcomes by limiting the need for medically trained staff and diagnostic testing and by directing potentially limited resources, such as beds and antimicrobials, to those with the greatest need,” the authors wrote.

Hospital employees frequently fell for simulated phishing emails, study finds

Hospital employees frequently clicked on emails that simulated phishing attacks, a recent study found.

The retrospective study included six U.S. health care institutions that ran phishing simulations in the period from Aug. 1, 2011, through April 10, 2018. The analysis included 95 simulated phishing campaigns with a total of 2,971,945 emails. Results were published by JAMA Network Open on March 8.

Almost one in seven of the simulated emails (422,062 [14.2%]) were clicked on by employees. The median institutional click rates for campaigns ranged from 7.4% (interquartile range [IQR], 5.8% to 9.6%) to 30.7% (IQR, 25.2% to 34.4%). Across all campaigns and institutions, the overall median click rate was 16.7% (IQR, 8.3% to 24.2%). A regression model showed that repeated phishing campaigns were associated with decreased odds of employees clicking on a subsequent email (adjusted odds ratios, 0.511 for six to 10 campaigns [95% CI, 0.382 to 0.685] vs. 0.335 for more than 10 campaigns [95% CI, 0.282 to 0.398]).

The finding of how many employees click on the emails is consistent with data in other industries, the study authors said. However, they noted that health care systems are uniquely vulnerable to phishing attacks, due to factors such as high turnover (including many brand-new employees), many devices on their networks, highly interdependent information systems, and difficulty locking down systems. They noted that the studied hospitals had information security programs robust enough to be running phishing simulations, so the results may actually reflect a conservative estimate of phishing click rates in U.S. hospitals.

The decrease in clicks with repeated campaigns suggests that simulated campaigns may reduce this risk, the authors said. Other potential strategies include email filters, multifactor authentication, special access controls for specific systems, and other awareness and training efforts, such as antiphishing laptop decals and posters. “It is necessary for all members of the health care community to understand this risk, particularly as safe and effective health care delivery becomes increasingly dependent on information systems,” the authors wrote.

Hospital-level antibiotic use associated with rates of hospital-onset C. diff

Hospitals that used more antibiotics had higher rates of hospital-onset Clostridioides difficile infection, and those that decreased antibiotic use had corresponding drops in C. diff, a recent study found.

The study examined cross-sectional and temporal associations in 2006 through 2012 between hospital-level antibiotic use and hospital-onset C. diff infections at 549 acute care hospitals. Hospital-onset C. diff was defined by presence of the ICD-9 discharge code and treatment with metronidazole or oral vancomycin three or more days after admission. Results were published by Clinical Infectious Diseases on March 1.

Overall, the rate of hospital-onset C. diff was 7.3 per 10,000 patient-days (95% CI, 7.1 to 7.5) and antibiotics were used on 811 days per 1,000 patient-days (95% CI, 803 to 820). Cross-sectional analysis found that for every 50-day per 1,000 patient-days increase in antibiotic use, there was a 4.4% increase in C. diff. For every 10-day per 1,000 patient-day increase in the use of third- and fourth-generation cephalosporins or carbapenems, hospital-onset C. diff infections increased by 2.1% and 2.9%, respectively. The time-series analysis found six hospitals that decreased their antibiotic use by at least 30% over the study, and this was associated with a 33% decrease in hospital-onset C. diff infections (rate ratio, 0.67; 95% CI, 0.47 to 0.96). Hospitals that decreased their use of fluoroquinolones or third- and fourth-generation cephalosporins by more than 20% had corresponding drops in C. diff of 8% and 13%, respectively.

This ecologic analysis, which the authors called the largest on the subject to date, supports the idea that effective antibiotic stewardship programs can have a major impact on hospital-onset C. diff infections. “Although achieving a 30% decrease in total [antibiotic use] within a hospital may represent a challenge, such reduction may be feasible in many hospitals based on findings of a recent study suggesting antibiotic use can be improved in 37.2% of the most common prescription scenarios,” they said. An effective alternative strategy might be to focus on high-risk antibiotic classes, particularly fluoroquinolones and third- and fourth-generation cephalosporins, the authors suggested. They noted that fluoroquinolone use in U.S. hospitals did drop between 2006 and 2012, without any corresponding drop in C. diff infections, so it might be most effective to concentrate particularly on replacing cephalosporins with penicillin-based drugs.

Limitations of the study include its reliance on administrative data and lack of data on changes in hospital infection control and diagnostic practices, which might have confounded the observed associations.

Chlorhexidine bathing wasn't effective for preventing infections on the wards

Having non-ICU patients bathe with chlorhexidine and use nasal mupirocin if they were known to carry methicillin-resistant Staphylococcus aureus (MRSA) did not reduce multidrug-resistant organisms, a study found.

Image by Getty Images
Image by Getty Images

The cluster-randomized trial included 53 hospitals, which measured 194 of their medical or surgical units' MRSA or vancomycin-resistant enterococcus (VRE) cultures during a baseline period from March 1, 2013, to Feb 28, 2014, a phase-in period from April 1, 2014, to May 31, 2014, and an intervention period from June 1, 2014, to Feb 29, 2016. During the intervention, the hospitals were randomized to either usual care or daily chlorhexidine bathing for all patients plus mupirocin for known MRSA carriers.

There were 189,081 patients in the baseline period and 339,902 patients in the intervention period (156,889 receiving routine care and 183,013 undergoing decolonization). On the primary outcome of unit-attributable MRSA-positive or VRE-positive clinical cultures, the control and intervention groups both saw decreases between the baseline and intervention periods and the difference between groups was not significant (hazard ratios for intervention period vs. baseline, 0.79 [95% CI, 0.73 to 0.87] with decolonization vs. 0.87 [95% CI, 0.79 to 0.95] with routine care. There were 25 adverse events, all involving chlorhexidine. Results were published by The Lancet on March 5.

“These results were neither significant nor clinically meaningful for a broad-based intervention strategy,” the study authors said. A post hoc analysis did find significant benefits to the intervention among a high-risk subgroup of patients with medical devices, they noted. Similar decolonization strategies have been successful at reducing infection rates in ICUs and might be more effective in a population with a higher prevalence or risk of multidrug-resistant infection, the authors said.

Another recent study found that declines in hospital-onset MRSA infections in the U.S. have slowed in recent years. According to a report in the March 5 Morbidity and Mortality Weekly Report, hospital-onset MRSA bloodstream infections decreased by 17.1% annually in 2005 through 2012, but the decline slowed during 2013 through 2016. “To reduce the incidence of these infections further, health care facilities should take steps to fully implement CDC recommendations for prevention of device- and procedure-associated infections and for interruption of transmission,” the authors said.

Another article in the same issue reported that at Veterans Affairs Medical Centers, hospital-onset MRSA decreased by 66% from 2005 through 2017, with infections declining more among patients who had negative admission surveillance MRSA screening tests (annual 9.7% decline) than those with positive admission MRSA screening tests (4.2%).