Antibiotic-resistant infections: millions in cost to hospitals, families, all of us

Folks, I mentioned that I'm way behind in working down a stack of great articles. Here's a very good one that I missed when it came out two weeks ago and is well worth your time.

A team from John H. Stroger Hospital (the new location of the iconic Cook County Hospital, public hospital for downtown Chicago) and from the Alliance for the Prudent Use of Antibiotics at Tufts University (headed by Dr. Stuart Levy, dean of antibiotic resistance scholarship in the US) has analyzed the direct and distributed costs of resistant infections, and their results are stunning. They took a random sample of patients seen at the hospital, sorted out a subgroup that suffered from resistant infections, and computed the costs that those infections imposed: in medical costs, increased length of stay, and excess deaths. Those sort of calculations have been done before at other institutions (cf. for instance the excellent work of Susan Cosgrove of Johns Hopkins), but what makes this Chicago study striking is an additional layer of analysis that computes the "social cost" to the families of those infected.

In the study's words:

In a sample of 1391 patients, 188 (13.5%) had [antibiotic-resistant infections]. The medical costs attributable to ARI ranged from $18,588 to $29,069 per patient in the sensitivity analysis. Excess duration of hospital stay was 6.4–12.7 days, and attributable mortality was 6.5%. The societal costs were $10.7–$15.0 million.

(Just to underline: These are almost certainly underestimates of the current problem and its current costs — because to get very solid data, the Stroger team went back in their database to patients who were treated in 2000. That's before the emergence and dominance of CA-MRSA USA300 nationwide, and its subsequent movement into hospitals. Since 2000, the MRSA epidemic has gotten worse.)

An accompanying editorial takes the next step in logic, stressing that if we're not going to work to reduce ARIs because it is good medicine to do so, we should do it because it is critically cost-saving:

...[T]he findings of Roberts et al [11] are significant, making a strong case for both the medical and financial benefits of reducing antimicrobial resistance. This is an important and timely question, considering the national focus on the prevention of health care–acquired infections, a significant proportion of which are caused by antimicrobial-resistant organisms, and the call for institutions to develop antimicrobial stewardship programs. These data should help inform decisions regarding the structure and implementation of health care initiatives designed to improve patient care while controlling unnecessary costs.

The cite for the study is: Rebecca R. Roberts, Bala Hota, Ibrar Ahmad et al. Hospital and Societal Costs of Antimicrobial‐Resistant Infections in a Chicago Teaching Hospital: Implications for Antibiotic Stewardship. Clinical Infectious Diseases 2009 49:8, 1175-1184.

Of course we would never have thought of that.

A new paper in the Annals of Internal Medicine suggests an astounding technique for figuring out whether patients experienced an adverse event while in the hospital:

Asking them.

No, really.

The study by Massachusetts researchers (from University of Massachusetts, Brown, Harvard, Massachusetts Department of Public Health and Massachusetts Hospital Association) looked back at the experience of more than 2,600 patients in 16 Massachusetts hospitals during 6 months in 2003. The researchers started from the assumption that the medical-records review done by many hospitals to spot adverse events was not capturing enough information — and that the interviews that some hospitals do with patients after discharge were asking the wrong questions because they focus only on satisfaction.

So the team did a 20-minute phone interview 6 to 12 months after discharge for 2,600 patients, asking about "negative effects, complications or injuries," and also reviewed the medical records of 1,000 patients who agreed to their charts' being released for review. For each arm of the study, two physician-reviewers checked results to be sure what was scored as an adverse event actually qualified as one.

And they found: That twice as many adverse events were uncovered when patients were asked about their experience. Among the interviewees, 23 percent reported an adverse event; when records were reviewed, only 11 percent of patients were judged to have experienced one.

Now, let's be clear: I'm very glad these researchers had the courage to do this study. Anything that supports better care, more transparency in care and more responsiveness to the patient's experience is a good thing and I support it.

But when I think of the dozens of hospital patients and family members who have told me about their experiences with poor infection control — lack of hand-washing, lack of housekeeping, bloody gauze on floors — and with being completely unable to get anyone in those hospitals to pay attention, it just makes me want to beat my head against a wall. Coming up with the idea of asking the patients about their experience... this is so hard?

As one of the co-authors, Saul Weingart of Dana-Farber Cancer Institute in Boston, said in an accompanying press release: "It's pretty clear that they can teach us important things about improving patient safety, if only we ask them."

The cite is: Weissman, JS et al. Comparing Patient-Reported Hospital Adverse Events with Medical Record Review: Do Patients Know Something That Hospitals Do Not? Ann Intern Med 2008; 100-108.

Isolation: Doesn't work if healthcare workers contaminate themselves afterward

In the new Emerging Infectious Diseases, there is a small but very smart study that ought to get wider play. It was done by a PhD candidate at University of North Carolina, Chapel Hill named Lisa Casanova, with the help of faculty and the local health department.

Background: In certain highly infectious environments — including in-hospital isolation — healthcare workers wear what is usually known as "personal protective equipment" or PPE. PPE generally includes gloves, gown and an eye shield, goggles or face-splash guard (also called "barrier precautions") as well as a mask or a respirator ("respiratory protection"). PPE protects the healthcare worker while he or she is in the patient's presence, but it poses a problem when the worker leaves that environment, because the PPE is likely to be carrying the disease organism on its surface. If the worker doesn't doff the PPE very carefully, he or she might contaminate himself/herself and become infected or colonized, or spread the organism further in the healthcare environment.

This accidental contamination was a significant problem in the 2003 SARS epidemic — so after SARS was over, the Centers for Disease Control and Prevention came up with a recommended procedure for taking off PPE (on this page, half-way down). Casanova decided to test how well the protocol actually works.

Answer: Not so much. She had 10 volunteers (men and women, left- and right-handed) dress in PPE, contaminated the equipment in certain spots ("front shoulder of gown, back shoulder of gown, right side of N95 respirator, upper right front of goggles, and palm of dominant hand") with a benign virus, had the volunteers take off their PPE, and then tested them for the virus's presence. Results:

Transfer of virus to both hands, the initially uncontaminated glove on the nondominant hand, and the scrub shirt and pants worn underneath the PPE was observed in most volunteers.

Casanova recommends changes: additional PPE; different PPE and doffing protocols, such as are used in surgical suites; or PPE impregnated with antimicrobials. (#1 and #3 of course would be more costly; #2 would require procedural change but not necessarily additional garments).

She also raises a vital ongoing issue for MRSA infection control: that healthcare workers may not be punctilious about hand hygiene because they believe that gloves are adequate protection. Only, as this study demonstrates, they are not:

This study also indicates the need for continued emphasis on hand hygiene. A barrier to improving hand hygiene compliance rates is the belief that gloves make hand hygiene unnecessary (14). This is contradicted by our study and others showing that organisms can spread from gloves to hands after glove removal (15). Even if double gloving is incorporated into protocols for PPE use, it is not a substitute for proper hand hygiene.

The cite is: Casanova L, Alfano-Sobsey E, Rutala WA, Weber DJ, Sobsey M. Virus transfer from personal protective equipment to healthcare employees’ skin and clothing. Emerg Infect Dis. 2008 Aug; [Epub ahead of print]

Antibiotics and the tragedy of the commons

I've been turfing through the pile of research materials I've gathered over the past two years and finding gems I had forgotten about. Here's one, from a post at Pioneering Ideas, a blog dealing with projects funded by the Robert Wood Johnson Foundation. One of RWJF's projects is "Extending the Cure," which seeks to reframe thinking about antibiotic use and stewardship by viewing them as economic and social-science problems rather than as medical problems.

In a guest post last year, Ramanan Laxminarayan, PhD, a senior fellow at Resources for the Future, explains the central problem in beautifully clear language:

[A]ntibiotic resistance is not always going to show up as a ‘crisis’ type of problem. ...

Sure - a set of circumstances such as a new strain of influenza that facilitates secondary infections such as Staphylococcus aureus that cannot be treated easily would result in a crisis. But the more likely scenario is one where, ten years from now, we are all using antibiotics that cost $1,000 or more a treatment course and none of the cheaper antibiotics work. This has a ripple effect on productivity of other medical technologies such as transplants and surgeries and on overall health care costs and insurance premiums. Since no one insurer is affected differently than others, there is no incentive for anyone to act.

Just as there is not going to be a single ‘day that the oil runs out’ and we have to think of solutions to our energy needs twenty-to- fifty years from now, we also have to start thinking of our antibiotics portfolio twenty-to-fifty years from now.

The full post is here.