"How to help nurses practice at the top of their game"
U.K. and U.S. media cover new research on deadly understaffing
August 5, 2012 -- Recent press items report that new research has revealed critical aspects of nurse understaffing in the United States and the United Kingdom. On July 31, the Philadelphia Inquirer ran a generally good piece by Don Sapatkin (with Meeri Kim) about a study by researchers at the University of Pennsylvania that used 2006 data from 161 Pennsylvania hospitals to analyze the link between worse staffing and higher rates of infection. The study focused on the burnout associated with poor staffing, which researchers found played a critical role in higher infection rates, taking lives and costing money. The Inquirer article quotes two of the nurses responsible for the study, as well as nurses from the American Nurses Association and elsewhere as outside experts. Although its account of the study findings is not totally clear, the report does give a sense of why having enough skilled, engaged nurses plays such an important role in patient outcomes, particular because of good input from a Pittsburgh union leader. And on July 31, the Telegraph (U.K.) published a good article by Laura Donnelly about a new study of staffing at 46 National Health Service (NHS) hospitals by nursing researchers at Kings College London. That study found that nurses had an average of eight patients during the day and 11 at night (in some places 15 patients at night). Not surprisingly, most of the nurses did not have enough time to do their work. The Telegraph piece is more about the government's responsibility for the poor staffing amid an ongoing public inquiry into the 2009 Stafford Hospital scandal, and the piece does not quote the researchers, though it does include key findings and quote two nursing leaders. The article stresses that understaffing is closely linked to the growing use of less-qualified support staff. The piece could have done more to explain what nurses do to save lives (like detecting infections) that other staff cannot; instead, we hear mostly about custodial care and "compassion." But both the Telegraph and the Inquirer convey the importance of nurse staffing and show that nurses can be academic and health policy leaders.
Cause and effect
The Inquirer piece "Penn study examines link between nurse burnout, care" first sets the stage by noting that nurses have long argued that hospital cost-cutting has resulted in understaffing that hurts patients. The article adds that California's "controversial" and "unique" response was to mandate minimum staffing ratios (of course, it isn't really "unique," since the Australian state of Victoria did it first). The piece then explains that in 2010, researchers found that applying the California ratios "would prevent 222 surgical deaths annually in New Jersey and 264 in Pennsylvania." Now, "members of that same University of Pennsylvania team" have done a follow-up study that suggests the problem is not just too few nurses but the "bad work environment" that entails, which leads to burnout.
To explain, the article relies in part on "lead author Jeannie P. Cimiotti," (right) who "led the research while at Penn and who is now executive director of the New Jersey Collaborating Center for Nursing at Rutgers University in Newark." Cimiotti says that when nurse understaffing combines with poor teamwork, inadequate support from superiors, and a lack of autonomy, "stress builds up and builds up and builds up until the giver of care just detaches," and "all of a sudden they are doing work, but they are not even cognizant of what they are doing, they are so stressed." One example, the piece notes, is that a nurse may forget to wash her hands. Much later, the piece returns to Cimiotti for more on how burnout works, and she notes that (in the piece's words) "when staffing is adequate, paying attention to issues such as physician-nurse relationships and having the right people to call at night could make a big difference at a hospital." Cimiotti observes that improving the organizational climate costs nothing. The piece also quotes Linda Aiken, another study author and "director of the Penn nursing school's Center for Health Outcomes and Policy Research," who says that it's "costing hospitals more money not to spend money on nursing."
These quotes are helpful, particularly in explaining how burnout can build up. But they could be more direct in bringing out the underlying conditions that have led to burnout, which relate very much to how nursing is undervalued and therefore underpowered. It would take a perceptive reader to translate the comments above into the idea that for this situation to change, nurses need more power and more respect, particularly from physicians and hospital managers. It also might have been good if the piece had cited the scholars' credentials (they have doctorates in nursing), although it is helpful that the article identifies the university-affiliated centers that they direct.
The report does have significant detail about the nature and importance of the study itself. On the latter point, the piece says that the study is "believed to be the first to examine why staffing matters" and "among the first to measure the potential harm," which it can do because the state of Pennsylvania requires the reporting of infection data. The piece notes that the study will appear in the American Journal of Infection Control, which, according to American Nurses Association (ANA) director of nursing practice and policy Cheryl Peterson, "takes it out of nurses' saying, yet again, 'Woe is me'" and makes it a "patient-safety issue for everyone. We appreciate that the Inquirer consulted the ANA, and we agree with Peterson's basic point, although the "Woe is me" part may suggest that nurses are normally a bunch of chronic whiners, when if anything nurses are often unwilling and/or unable to stand up for themselves as they should, which is a big reason they are understaffed--and patients are dying--in the first place.
As for the specific methods and findings, the piece explains that the researchers linked 2006 infection data from 161 acute-care Pennsylvania hospitals with staffing information and burnout data from questionnaires completed by more than 7,000 direct care nurses from those hospitals. They found that each nurse cared for an average of 5.7 patients and that the rate of catheter-related urinary tract infections was 8.6 per 1,000 patients. Adding one patient to that workload increased the number of urinary-tract infections by nearly one infection per 1,000 patients, which would apparently mean 1,351 additional infections statewide. The piece reports that surgical site infections, "which are half as common but which cost far more, went up at a similar rate." The researchers also measured the effect of burnout alone, finding that raising the number of "burnout-out nurses" by 10% increased the urinary tract infection rate by a similar amount, but raised the rate of surgical site infections from 4.2 to more than six per 1,000--more than 50 percent. But researchers were surprised when they combined the data and "discovered that burnout appeared to be responsible for nearly all the harm caused by greater workloads." This does not seem totally clear--is the piece saying that the researchers were able to determine that nearly all the understaffed nurses whose patients got infections were also burned out? Does that mean burnout is all we need to worry about, or that reducing burnout is possible and desirable even if we can't improve staffing, or simply that understaffing is the likely cause of the burnout? What is the statistical relation between understaffing and burnout? We worry that this data will cause hospitals to persuade nurses that they aren't really "burned out"--Happy Nurses Month!--and so, there's no staffing problem! But at least the study points out to thinking people the fact that if hospitals want to solve burnout, the first and most important place to start is to improve staffing ratios.
In any case, the piece says that the study found that reducing the number of burned-out nurses by 10 percent from the "typical" 30% would prevent 4,160 infections (of both types) annually and save $41 million statewide, noting that "insurers are increasingly unwilling to reimburse the expense of treating preventable infections." It's not clear from this account if the study proposed any ways to actually make such a reduction, apart from generally improving the organizational issues Cimiotti describes in her quotes, above. Is the idea that better staffing would still be the best way to improve care, in part because it would reduce burnout, but hospitals are looking for some other steps to take, so they don't have to spend any more money on nursing? And the piece mentions the "typical" 30% burnout rate for nurses but says nothing more about it. Isn't that rate, um, terrifying? Would we be OK if almost a third of some other group of professionals who hold lives in their hands were burned out?
The report also includes some context and reaction from those not involved in the study. It quotes Michael Leiter, an organizational psychologist at Acadia University in Nova Scotia, who says that (in the piece's words) "the study could not prove cause and effect"--perhaps meaning it could not prove that burnout causes more infections, or maybe that it could not prove what causes burnout, but it's not totally clear. In any case, Leiter notes that the findings make sense because burnout grows in "a cycle of exhaustion, cynicism, and lack of civility," and that nurses may also experience "'a value conflict with their employer' if they believe that finances are trumping patient care." Leiter adds that "basically, what makes burnout is bad management." So, if that is the case, does that mean at least 30% of Pennsylvania nurses are badly managed?
The reference to "exhaustion" suggests that understaffing is at least a factor in burnout, and the piece ends with a short discussion of recent controversies over staffing ratios. It notes that nurses have come under increasing pressure "as medical conditions have gotten more complex and lengths of patient stay have shortened." The report also says that nurses have recently backed proposed bills setting minimum staffing ratios in Pennsylvania, New Jersey, and many other states, but none has passed. The piece reports that the bill in Pennsylvania is based in part on a labor contract at Pittsburgh's Allegheny General Hospital that includes staffing ratios. The article consults Cathy Stoddart, "chapter president of SEIU Healthcare" (presumably at that hospital), who gives a great illustration of why staffing matters, noting that a slight change in a patient's condition--such as a small temperature change or decline in urine output--can signal a serious problem, but that a nurse with 10 patients is not going to notice that. Stoddart reports that staff turnover and infection rates both improved dramatically after the Allegheny contract created a "partnership" with the hospital; it's not clear what that meant apart from the staffing ratios.
The piece rightly includes the perspective of hospital management, noting that "the industry generally has fought staffing mandates, arguing that hospitals need flexibility to respond to unique challenges." But Nancy Foster, the American Hospital Association's "vice president of quality and patient safety policy," says the new Penn study "raises an interesting question," namely "exactly how can we look at nursing burnout and its impact on patient safety." She reportedly adds that "hospital leaders she speaks with are all looking at how to help nurses practice at the top of their game." We assume many hospitals would be attracted to measures they could take to make nurses happy and improve care that would not cost much money. And that sounds great to the extent those things include improving the organizational climate (as Cimiotti puts it), perhaps by empowering nurses and increasing understanding of nursing among physicians and others, as opposed to the kind of angel-oriented pats on the head often seen during Nurses Week and in certain television and radio commercials. It would be worrisome, however, if those organizational measures were seen as a substitute for adequate nurse staffing. All the "respect" in the world will not enable overworked nurses to "practice at the top of their game."
Their condition is deteriorating
The Telegraph piece's headline gets at the bottom line: "Nurses look after 15 patients at a time." That should scare anyone who knows anything about what nurses actually do, although of course that category is much smaller than we would like. The piece explains that the "major study" of 46 NHS hospitals was led by the National Nursing Research Unit at Kings College London as part of "international research ordered by the European Commission." The study showed a "sharp increase" in nurses' workloads and a surge in the use of "unqualified healthcare assistants," who now outnumber nurses "at one in six [hospital] trusts." The piece delivers other major findings in bullet-point form:
* On average, nurses were allocated 11 patients at night and eight by day;
* At some trusts, nurses were asked to care for 15 patients at night;
* Across the country, the proportion of qualified nurses to unqualified staff has fallen
below safe thresholds outlined by England's most senior nurse last year;
* Five years ago, on average there were two nurses for each assistant; now the ratio
is closer to 1:1;
* Of 3,000 nurses who took part in the research, two-thirds said they ran out of time
to offer "comfort" to patients and relatives;
* More than one-third said they ran out of time to safely supervise their patients.
The Telegraph article notes that five years earlier, nurses had "an average of seven patients on day shifts and nine at night, previous research shows." And the report includes a few other findings in the text, including that the average trust now has 56 per cent "qualified nurses" to 44 per cent "healthcare assistants," and in some trusts the percentage of nurses is as low as 43 per cent. More than three quarters of the nurses polled (76%) said there were "not enough staff to get the work done." Nurses said that older patients got "the least care and attention."
The piece also provides some of the context surrounding the new study, which it says "follows growing public concern that hospital patients, especially the elderly, are being denied basic dignity and left thirsty and hungry." In particular, the piece refers to the pending inquiry into "the  Stafford Hospital scandal - where hundreds died amid 'appalling' failings in care," including the replacement of nurses with cheaper staff; the inquiry will apparently issue a report in the fall about the hospital's problems and how to improve care. That inquiry and "a review of nursing ordered by the Prime Minister" are reportedly considering whether there should be new rules about nurse staffing, which could establish "a maximum number of patients per nurse, or a ceiling for the proportion of assistants to qualified nurses on any shift." The article notes that NHS has no minimum nurse staffing levels now, although "in parts of the US and Australia, mandatory levels have been imposed -- typically at around six patients per nurse." Actually, the levels in California vary by unit, ranging from 2 for ICU to 5 for the medical-surgical units. Even five patients is an unsafe number of patients for a nurse in most units, according to a 2002 Penn research finding that for every additional patient over four that a nurse had to care for, mortality rose by 7% (though of course five is much better than eight, 11, or 15).
The piece also includes reaction from experts not involved in the study, including two nursing leaders. The report says patient advocacy groups argue that the sharp rise in nurses' workload has "compromised safety and left little time for compassion," despite "the lessons of the Stafford Hospital scandal." The piece quotes Patients Association chief executive Katherine Murphy (right), who describes the situation as "absolutely disgraceful": "One nurse cannot care safely for 15 patients, and identify when their condition is deteriorating, let alone offer comfort or compassion." Similarly, Royal College of Nursing director of nursing Janet Davies says: "When there aren't enough nurses, it has a huge impact on the quality of care for patients; they are left unsupervised for too long, and it is harder to provide compassionate care." The piece notes that the RCN favors limiting healthcare assistants, who are "only supposed to carry out non-medical tasks such as washing and feeding," to 35% of the nursing staff on each ward. Of course, many feel that nurses should also be doing the washing and feeding, since that provides potentially life-saving opportunities to observe, educate and counsel patients. The piece also includes comment from Dame Christine Beasley, England's recently retired Chief Nursing Officer, who told the Stafford inquiry that the assistant number should be no higher than 40%, and that the apparent ratio in the Stafford setting of the reverse--with six assistants for every four nurses--was "unacceptable."
And the piece briefly includes reaction from government officials. Health Secretary Andrew Lansley (right) reportedly conceded that the number of nurses had fallen by 4,500 in the past two years, despite apparent promises not to cut "frontline jobs" during the current focus on budget cuts. Health Minister Anne Milton said the NHS had 10,000 more nurses than it did five years ago, and that the government's review of nursing had been asked to (in the piece's words) "find ways to free up staff so they could spend as much time as possible with patients."
Overall this article is a helpful look at nurse understaffing that threatens patients' lives, and commendably it does include comment from two nursing leaders. The piece effectively brings out that this understaffing matters a great deal, referring repeatedly to the well known Stafford scandal, which apparently included serious breakdowns in basic custodial care, and emphasizing that a nurse with too many patients lacks time for compassion or to get his work done in general. And the patients association leader Murphy rightly notes that a nurse with 15 patients cannot "identify when [a patient's] condition is deteriorating." But that is the only specific indication in the piece of how understaffing prevents nurses from using their vital physiological assessment and treatment skills, as opposed to simply being there to avoid basic problems or to provide the "comfort" and "compassion" and "time with patients" that nurses are so well-known for, but which has not been enough to save the profession from severe underfunding. Of course, terms like those also mask the real psychosocial expertise nurses also need to care for patients, and anyway, we doubt much of the public would regard that expertise as comparable to the technical skill it associates with physicians. Even the "deteriorating" formulation here does not really convey how much skill is required to detect subtle changes in conditions that can have huge effects, like the deadly infections measured in the Penn study. And the public needs to know that hospital nurses also save lives by using their skills to initiate treatments to help patients recover.
In any case, the Telegraph piece is very effective in highlighting the problem posed by the growth in the use of cheaper, unlicensed assistive staff rather than nurses, which is also a major concern in the United States. This problem has developed because public and private sector decision-makers in an era of rising health costs and shrinking national budgets often do not understand the nature or value of what nurses do, and therefore they have seen fit to cut nurse staffing, assuming that assistants can do much of it. Press items like these, and the public advocacy of the nurses and others they quote, are vital if that situation is to be reversed.
We thank those responsible for both of these important reports on nurse understaffing.
See the article "Nurses look after 15 patients at a time" by Laura Donnelly, published on the Telegraph (London) website on August 5, 2012. Also see the article "Penn study examines link between nurse burnout, care," by Don Sapatkin, posted July 31, 2012 on the Philadelphia Inquirer website.
See the University of Pennsylvania study:
Jeannie P. Cimiotti, Linda H. Aiken, Douglas M. Sloane, Evan S. Wu; “Nurse staffing, burnout, and health care–associated infection,” American Journal of Infection Control, 40, (6) 486-490, August 2012, available at http://tinyurl.com/prg2vzd
See the U.K. study:
Jane E. Ball, Trevor Murrells, Anne Marie Rafferty, Elizabeth Morrow, Peter Griffiths “‘Care left undone’ during nursing shifts: associations with workload and perceived quality of care,” British Medical Journal Quality & Safety, July 2013, available http://qualitysafety.bmj.com/content/early/2013/07/08/bmjqs-2012-001767.full.pdf