August 18th, 2013
What Patterns of Change in the Use of Nesiritide Reveal About Hospitals as Learning Organizations
Chohreh Partovian, MD, PhD
In 2005, safety concerns about nesiritide surfaced in the medical literature, in the media, and in official communiqués from the FDA. To determine how hospitals responded to those concerns and altered their patterns of nesiritide use, my colleagues and I retrospectively analyzed data from the Premier database, including 813,783 hospitalizations for heart failure at 403 hospitals from 2005 to 2010. We applied a growth mixture modeling approach to hospital-level, risk-standardized rates of nesiritide use, to classify hospital groups according to their changing patterns in use of the drug. Our findings, published in JACC: Heart Failure, reveal how hospitals varied in their responses to the new safety information.
Overall, the percentage of hospitalizations for heart failure that involved use of nesiritide declined from 15.4% in 2005 to 1.2% in 2010. An initial sharp drop occurred immediately after safety concerns were publicized, followed by a more gradual decline in subsequent years. That’s not surprising. However, when we analyzed the data at the hospital level, the speed and amplitude of change in nesiritide use varied widely among hospitals. Three distinct groups of hospitals emerged: low-users, fast de-adopters, and slow de-adopters.
The results suggest that common underlying factors within each group of hospitals may explain why they responded so differently. However, we found that these three hospital groups did not differ on traditional hospital characteristics such as size, urban setting, or teaching status. In multivariate regression analysis, none of those standard characteristics was significantly associated with a hospital’s likelihood of being in the slow de-adopter group compared with the other 2 groups.
Those “negative” results were frustrating but not really unexpected. Previous studies have shown that other organizational characteristics — such as team composition (number and type of specialists, inclusion of a pharmacist), internal culture (quality and frequency of communication and collaboration among team members), regulatory context (drug formularies), and availability and use of clinical-decision support systems —have greater impact on medical decision making, evidence-based practice, and hospital performance than do standard hospital categories. However, such organizational characteristics, often called “soft variables,” are not readily available in healthcare databases. Further research using qualitative and mixed methods are therefore required to determine what hospital characteristics actually foster or impede organizational responsiveness.
At this point, some might ask why we analyzed the data at the hospital level rather than physician level. First, hospitalized heart failure patients usually see multiple clinicians because they often have complex disease and multiple comorbidities. Therefore, it is not always possible to identify the prescribing physician. Second, we think the behavior should be attributed to the system structure. For example, it’s important to create an organizational environment where clinicians do not practice in isolated silos but where team building, collaboration, and communication are encouraged. The organization can also provide access to information technology and clinical-decision support systems that help clinicians access timely information, practice evidence-based medicine, and better cope with the uncertainty of accelerating change. This approach is more consistent with the emerging emphasis on systems of care, where the focus shifts from blaming individuals to redesigning how organizations function.
Here are some take-home messages from this study: Changing nature of medical evidence often requires a change in practice. This in turn requires not only access to high-quality, timely information but also a certain degree of “agility” in responding to new evidence. Measuring hospitals’ learning rates, or how quickly and efficiently they respond to new information, could provide important feedback to all stakeholders. It will be crucial to determine what transformations to underlying organizational structure will be associated with greater agility and an improved ability to learn.
As the Greek philosopher Heraclitus once said, “The only constant is change.” In our modern era, the accelerating changes in demographics, economic activity, and technology are challenging our institutions, practices, and beliefs. In my view, the extent to which we are capable of transforming and reinventing our institutions and the way we practice will be the most critical factor in our journey to a high-quality, high-performance healthcare delivery system.
JOIN THE DISCUSSION
Share your views on the study of nesiritide use patterns by Dr. Partovian and her colleagues. Is your institution agile enough to respond to change rapidly and effectively?
3 Responses to “What Patterns of Change in the Use of Nesiritide Reveal About Hospitals as Learning Organizations”
NEJM Journal Watch — Recent Cardiology Articles- Strategies and Criteria for the Diagnosis and Management of Myocarditis
- Sex-Based Disparity in CABG Outcomes: Are We Moving Toward Parity?
- Can Genetic and Social Risk Scores Improve Cardiac Risk Prediction?
- Extensive Ablation vs. Pulmonary Vein Isolation in Persistent Atrial Fibrillation
- Can a Novel Stent Design Improve Long-Term Outcomes of Percutaneous Coronary Intervention?
I used Nesitride a couple of times and stopped using it because of the limited benefice, secondary effects, better therapeutic options and high cost. Then, a few years later, I received confirmation of this in the 2005 FDA statement.
This paper analyzes how hospitals responded to those concerns and altered their patterns of nesiritide use.
I’ m more interested in why it took 4 years to find out these safety concerns.
The author identifies three distinct groups of hospitals: low-users, fast de-adopters, and slow de-adopters. But something catches my attention, the fact that in 2005, 15.4% of hospitalizations for heart failure involved the use of nesiritide. Taking in consideration the final recommendation of the FDA panel this number is to high or not?
Dr. Partorvian describes organizational characteristics, called “soft variables,” I believe this is the core of the problem and the solution, the human factor, the medical staff.
I would like to propose a different group classification based in the physician level. I identify three groups: The conservative physician, blind follower of published papers and the inquisitive published paper readers. This explains the “agility” in responding to new evidence in this case the low-users, fast de-adopters, and slow de-adopters.
We must acknowledge this new evidence works in two directions, adopting and stopping in this case the use of Nesitride. Then It is very possible that the transformations needed to obtain greater agility and an improved ability to learn in the organization is to convert the so called soft variables to hard variables.
Thank you for these great comments.
15.4% is indeed high: it’s almost double the proportion of heart failure hospitalizations using IV nitroglycerin. And interestingly, when use of nesiritide dropped, no “substitution effect” was observed: the utilization of other vasodilators or even inotropes did not increase, which strongly suggests a case of nesiritide overutilization in 2005.
The safety concerns in terms of higher mortality and renal toxicity were actually ruled out by the ASCEND-HF study published in 2011, which also established that the drug had no benefit on the outcomes.So for me and as suggested by others before 2005 the real question is: Should have nesiritide obtained FDA approval based only on small studies which did not assess the effect on important outcomes such as mortality and readmission?
I entirely agree with you about the importance of human factor as the core of the problem and the solution to higher quality and better performance.These soft variables should not be ignored because we currently do not have numbers for them. They should be assessed, become part of health care databases and be included in any model that tries to assess hospital or any delivery system performance.
Regarding your proposed new classification, I agree that physician’s ability to learn and think critically are very important however, we all as human being have biases and make judgmental errors. So it’s important to create structures and organizations in which even ordinary physicians can achieve extraordinary results.
Like I said evidence work in two direction ASCEND-HF observe no difference in the composite endpoint of death or CHF hospitalization at 30 days in patients hospitalized with ADHF, but a follow up of 30 days in ADHF to evaluate death is enough time? if you agree with this result, in their time the slow de-adopters were right because the initial safety concerns in terms of higher mortality and renal toxicity were actually ruled out by the ASCEND-HF, even if nesiritide is associated with mild improvements in dyspnea, the FDA initial safety concern were wrong.
OK I disagree 100% because like I said for me the main concern was limited benefice, secondary effects and cost, in resume a bad cost/benefit ratio. So the Nesitride was approved based only on small studies which did not assess the effect on important outcomes such as mortality and readmission and discarded with a study limited with only 30 days of follow up. Probably they are right but the method is questionable. The ASCEND-HF didn’t evaluate cost/benefice but I am sure is bad.
A few years back I have a discussion with the director of my teaching hospital. He was arguing that institution, this can be translated as organizations, make the physician. I disagree, I told him that people and in this case physician make the organization. I told him that a good student can by a great physician in any organization, school, hospital, etc. the principal and most important variable is the individual. A great deal of years has pass and his organization is full of politics, interest conflicts, tons of paper work, and burn physician. I decide to make my organization based around the physicians and I have fewer problems and I like to think with better results. So for me it’s important look for good physicians to create structures and organizations in which even a bad organization can achieve extraordinary results.