When we look at published results of a clinical trial, particularly one that shows a dramatic improvement in outcomes with use of the drug or device being studied, we do so with a very skeptical eye. Study after study have used flawed methodologies to suggest to us their product is great. Or they highlight in the “conclusion” a subgroup analysis that had a positive trend when the primary outcomes measures were not met. See our February 16, 2010 Patient Safety Tip of the Week “Spin/Hype…Knowing It When You See It” for examples. If it looks too good to be true, it’s usually not.
So when we saw in bold headlines in all media that there was a 17% reduction in hospital-acquired conditions (HAC’s) in the past 3 years, saving 50,000 lives and $12 billion (AHRQ 2014), and largely attributed to government-sponsored programs, our “hype radar” went into high gear.
And this occurred at a time when public reporting showed substantial increases in serious adverse events in Massachusetts (Kowalczyk 2014, Biondolillo 2014) and Connecticut (Connecticut DPH 2014) and a report commissioned for the Betsy Lehman Center showed that many Massachusetts patients perceived medical errors to be common and not reported (Harvard School of Public Health 2014). And we’re not the only ones whose “hype radars” were raised. Healthcare bloggers Paul Levy and Susan Carr also discussed these mixed messages on patient safety.
Given the slow to nonexistent progress we’ve made in patient safety over the past 2 decades, is it plausible that such dramatic improvements could be made in 3 years or less? When we read a paper about a clinical trial we always look to see who sponsored or funded the trial. If it is funded by someone who stands to profit from results put in a positive light we are even more skeptical. For years we have lamented the fact that the percentage of studies funded and overseen by government organizations like NIH has dwindled and now most research is sponsored by those with vested interests. Notice we said we like government-sponsored research. That is because we always assume such research won’t be biased. But is that true for all government-sponsored studies? Government can also have self-interests that might lead to biases. Might it not want to justify large investments made in various projects?
But we did dig deeper into the evidence. First, much of the “increase” in serious events reported in Massachusetts and Connecticut appears to be due to changes in reporting requirements and may not reflect an actual increase in adverse patient events.
In the AHRQ study preliminary estimates for 2013 show a further 9 percent decline in the rate of hospital-acquired conditions (HACs) from 2012 to 2013, and a 17 percent decline, from 145 to 121 HACs per 1,000 discharges, from 2010 to 2013. About 40 percent of this reduction is from ADEs, about 20 percent is from pressure ulcers, and about 14 percent from catheter-associated urinary tract infections (CAUTIs)
The AHRQ report focused on the nine “core” events that were part of the improvement projects in the Partnership for Patients initiatives that worked with the HEN’s (Hospital Engagement Networks) funded through the Affordable Care Act (ACA). The hospital-acquired conditions that were the focus of the Partnership for Patients program are:
But those are really significant patient safety events that we all focus our attention on. The other events were lumped under the “all other HAC’s” category. And the AHRQ methodology was a sampling of all Medicare patients, not just those at the hospitals participating in the Partnership of Patients focused programs. So the methodology is actually pretty good for looking at trends over time.
Do we find other evidence that might corroborate the striking improvement in HAC’s in the AHRQ report? CDC has just reported its progress report on hospital-acquired infections (HAI’s) and it does show substantial improvement (CDC 2015). On the national level, the report found:
The significant reduction is CLABSI’s is not surprising as hospitals have adopted the practices developed by Peter Pronovost and promulgated by the success of the Michigan Keystone Project (see our March 2011 What’s New in the Patient Safety World column “Michigan ICU Collaborative Wins Big”).
And in our January 2015 What’s New in the Patient Safety World column “Beneficial Effect of EMR on Patient Safety” we highlighted a study using data from the Pennsylvania Patient Safety Authority (PPSA) and the HIMSS Analytics database that demonstrated a substantial decrease in patient safety events at hospitals with advanced EMR’s (Hydari 2014). The researchers found that advanced EMRs led to a 27 percent decline in patient safety events overall, driven by a 30 percent decline in events due to medication errors and 25 percent decline in events due to complications of tests, treatments and procedures. So as more and more hospitals have progressed to the “advanced EMR” status we likely are seeing a substantial reduction in events, particularly medication errors, which was one of the HAC’s with substantial improvement in the AHRQ report.
Other evidence comes from studies looking at the impact of CMS’s initiatives that withhold payments for certain HAC’s. Previous studies had suggested that CMS’s HAC (Hospital Acquired Condition) nonpayment initiative had not had significant impact on CLABSI’s or CAUTI’s (Lee 2012). But those results came from a limited data set. Now a new analysis (Waters 2015) of a much larger data set concludes that the CMS HAC nonpayment initiative was associated with significant reductions in CLABSI’s (11% decrease) and CAUTI’s (10% decrease) but no reduction in rates of injurious falls or decubiti (though the timeframe differs from that of the AHRQ study). The authors conclude that the success in reductng CLABSI’s and CAUTI’s reflected more robust evidence-based preventive interventions for those conditions than for the other two HAC’s. They also suggest that, since CLABSI’s and CAUTI’s are more frequent in specialized units like ICU’s, it is easier to get a small group of focused professionals addressing the issue. On the other hand, falls and decubiti occur throughout the hospital system and their reduction would require a much larger team-based approach.
The editorial accompanying the Waters study (Umscheid 2015) notes that the Lee study had 33 months of data on CLABSI’s and CAUTI’s before the CMS initiative began in 2008 whereas the Waters study only had 9 months of data. They argue that the more limited pre-implementation data may have been inadequate to define the true trend in CLABSI’s and CAUTI’s prior to the CMS action. Also, the CLABSI and CAUTI data in the Waters study apparently were limited to ICU events whereas CMS looks at data for the entire hospital.
That does raise an interesting point. In the AHRQ study the majority of deaths averted occurred as a result of reductions in the rates of pressure ulcers and ADEs, although declines in other HACs also contributed significantly to deaths averted. Yet the Waters study found no reduction in the rates of decubiti. But the timeframes differ. The Waters study looked at data from 2008 to 2010, the AHRQ study from 2010 to 2013.
So back to our original question – is it progress or is it propaganda? Looks like this is one instance where our “hype radar” was wrong. We’ll go with progress. It’s refreshing to see that a lot of hard work and plugging along are finally bearing fruit.
AHRQ. Interim Update on 2013 Annual Hospital-Acquired Condition Rate and Estimates of Cost Savings and Deaths Averted From 2010 to 2013. AHRQ Partnership for Patients 2014
Kowalczyk L. Mass. Hospitals’ Mistakes List Widens. Boston Globe August 14, 2014
Biondolillo M. Commonwealth of Massachusetts Department of Public Health. Serious Reportable Events 2011-2013. August 2014
Connecticut Department of Public Health. Legislative Report To The General Assembly. Adverse Event Reporting. Octobrer 2014
Harvard School of Public Health. The Public’s View on Medical Error in Massachusetts. Commissioned by Betsy Lehman Center for Patient Safety and Medical Error Reduction and Health Policy Commission. December 2014
Levy P. Falling behind on safety and quality in the Hub of the Universe. Not Running a Hospital Blog December 2, 2014
Carr S. Editor’s Note: Mixed Messages on Safety. PSHQ Blog December 9, 2014
CMS. Partnership for Patients.
CDC.Healthcare-associated Infections (HAI) Progress Report. January 2015
Hydari MZ, Telang R, Marella WM. Saving Patient Ryan - Can Advanced Electronic Medical Records Make Patient Care Safer? (September 30, 2014). Available at SSRN:
Lee GM, Kleinman K, Soumerai SB, et al. Effect of Nonpayment for Preventable Infections in U.S. Hospitals. N Engl J Med 2012; 367: 1428-1437
Waters TM, Daniels MJ, Bazzoli GJ, et al. Effect of Medicare’s Nonpayment for Hospital-Acquired Conditions. Lessons for Future Policy. JAMA Intern Med 2015; Published online January 05, 2015
Umscheid CA, Brennan PJ. Incentivizing “Structures” Over “Outcomes” to Bridge the Knowing-Doing Gap. JAMA Intern Med 2015; Published online January 05, 2015
You’ve heard us extol over and over the power of stories over statistics. One example we often note in our presentations and webinars was from Richard Shannon, MD back in 2007 at a conference on patient safety sponsored by the NY State Department of Health (Shannon 2007). He spoke about how for years data would be presented on rates of CLABSI’s and likely costs due to CLABSI’s and how this generated little interest in action. Then something caught his attention: over half his ICU patient who got a CLABSI died! That’s what spurred him and his colleagues to action.
A new study puts prevention of health care–associated infection (HAI’s) in perspective from both a human and financial standpoint (Dick 2015). The researchers looked at 5 years of Medicare data on HAI rates and used cost and quality of life estimates from the literature to compare patients with and without CLABSI or VAP and estimated the cost-effectiveness of multifaceted HAI prevention programs. For CLABSI they estimated the total life-years (LY’s) and quality-adjusted life-years (QALY’s) gained per ICU due to infection prevention programs were 15.55 and 9.61 respectively. For VAP the estimates were 10.84 LY and 6.55 QALY.
On the cost side, reductions in index admission ICU costs were $174,713.09 for CLABSI and $163,090.54 for VAP. The incremental cost-effectiveness ratios (ICER’s) were $14,250.74 per LY gained and $23,277.86 per QALY gained.
The authors conclude that the results underscore the importance of maintaining ongoing investments in HAI prevention.
Shannon R. Eliminating Hospital Acquired Infections: Is It Possible? Is It Sustainable? Is IT Worth It? Presentation at NYSDOH 2007 Patient Safety Conference
Dick AW, Perencevich EN, Pogorzelska-Maziarz M, et al. A decade of investment in infection prevention: A cost-effectiveness analysis. Am J Infect Control 2015; 43(1): 4-9
Since we’ve touched a lot this month on infection control, it is worth noting several new items pertinent to infection control in 2015.
First, CMS (Centers for Medicare & Medicaid Services) has adopted a new Hospital Infection Control Worksheet that will be used during hospital surveys. It has sections on the qualifications of the infection control officer, hospital leadership, infection control programs and indicators used, MDRO’s, antibiotic stewardship, hand hygiene, employee training, employee health, vaccinations, insulin pen safety, vial safety, sharps safety, PPE and precautions, lots of environmental services issues, sterile processing and reprocessing issues, a variety of indwelling catheter issues and respiratory therapy issues, isolation procedures, surgery, spinal injection procedures, point of care devices and others. It’s 49 pages long! Be prepared if you get surveyed!
Second, Joint Commission has announced a new Infection Prevention and HAI Portal. It basically provides you access to all materials pertinent to infection prevention and healthcare-associated infections in one place. It has outstanding tools and links to both Joint Commission resources (individual HAI’s, antibiotic stewardship, hand hygiene, etc.) and external resources like SHEA’s “A Compendium of Strategies to Prevent Healthcare-associated Infections in Acute care Hospitals: 2014 Updates” (SHEA 2014), the Surviving Sepsis Campaign, and many others.
And lastly the New York State Department of Health (NYSDOH 2014) list of locations for reporting CLABSI indicators has been expanded to include:
This is to become more consistent with CMS’s Hospital Inpatient Prospective Payment Systems (IPPS) expansion to medical, surgical, and medical-surgical wards effective January 2015.
CMS (Centers for Medicare & Medicaid Services). Hospital Infection Control Worksheet.
The Joint Commission. The Infection Prevention and HAI Portal.
SHEA (The Society for Healthcare Epidemiology of America). A
Compendium of Strategies to Prevent Healthcare-associated Infections in Acute
care Hospitals: 2014 Updates. Infection Control and Hospital Epidemiology 2014;
35(S2), September 2014
Society of Critical Care Medicine. Surviving Sepsis Campaign.
New York State Department of Health (NYSDOH). Proposed change to 2015 New York State hospital-acquired infection reporting requirements. September 5, 2014
Ever since we began discussing the ability of early warning systems like MEWS to detect clinical deterioration early we have commented on the need to add clinical impression (by a nurse or physician) to the prediction score.
In our March 2012 What’s New in the Patient Safety World column “Value of an Expanded Early Warning System Score” we noted a study (Smith 2012) from the Netherlands that showed a positive impact of a modification of the MEWS that added some new parameters, including a more subjective parameter: the nurse’s level of concern about the patient’s condition.
In our July 15, 2014 Patient Safety Tip of the Week “Barriers to Success of Early Warning Systems” we discussed an excellent study in the nursing literature (Watson 2014) that provided great insight into the barriers that impact implementation of an early warning system. One of those barriers was that there was a general perception by RN’s that the EWS was no better at predicting deterioration than their own clinical impression. They recommended adding RN or family concerns to the EWS score.
Now a new study has addressed the role of physicians’ clinical judgment in detecting early clinical deterioration (Patel 2015). Patel and colleagues utilized the Patient Acuity Rating (PAR) (Edelson 2011) as a clinical tool to predict clinical deterioration. The PAR has an interesting history. It was originally proposed as a way to summarize a patient’s risk of deterioration by a score that could simply be added to the signout/handoff to a covering physician. It simply consists of the response on a 7-point Lickert scale to the question “How likely is this patient to suffer a cardiac arrest or require emergent transfer to the ICU in the next 24 hours?” (a score of 7 being extremely likely and a score of 1 being very unlikely). When Edelson and colleagues tested the PAR on medical attendings, interns, residents and physician extenders on non-ICU medical patients they found it had reasonable inter-rater reliability and good ability to predict which patients would likely have a cardiac arrest or require urgent transfer to an ICU within 24 hours. A PAR of 4 or higher corresponded to a sensitivity of 82% and a specificity of 68% for predicting cardiac arrest or ICU transfer in the next 24 hours.
In the new study Patel and colleagues (Patel 2015) assessed PAR scores and MEWS (Modified Early Warning Scores) scores on over 3000 medical inpatients. Outcome measures were cardiac arrest, ICU transfer, RRT (rapid response team) activation, or a composite of the three. They found poor correlation between the MEWS and PAR scores and there was a median 84 minute gap between the PAR and MEWS scores. However, the combined PAR plus MEWS score was more accurate for the composite outcome than either the MEWS or PAR scores individually.
But be careful – there is a possibility that the PAR is a self-fulfilling prophecy. That is, the clinician doing the PAR might also be the one who may make the decision to transfer the patient to the ICU or might influence the receiving physician to do so. So using transfer to the ICU as an outcome variable may be somewhat biased.
As we refine some of the hi-tech collection and background analysis of physiological variables as noted in our November 11, 2014 Patient Safety Tip of the Week “Early Detection of Clinical Deterioration” it will be of interest to see how scores for predicting clinical deterioration might make better use of the clinician’s clinical judgment.
Some of our other columns on MEWS or recognition of clinical deterioration:
Smith T, Den Hartog D, Moerman T, et al. Accuracy of an expanded early warning score for patients in general and trauma surgery wards. British Journal of Surgery 2012; 99: 192-197
Watson A, Skipper C, Steury R, et al. Inpatient Nursing Care and Early Warning Scores: A Workflow Mismatch. J Nurs Care Qual 2014; 29(3): 215-222
Patel AR, Zadravecz FJ, Young RS, et al. The Value of Clinical Judgment in the Detection of Clinical Deterioration. JAMA Intern Med 2015; Published online January 05, 2015
Edelson DP, Retzer E,Weidman EK, et al. Patient acuity rating: quantifying
clinical judgment regarding inpatient stability. J Hosp Med 2011; 6(8): 475-479
Print “PDF version”