Nosocomial outbreaks linked to staff shortages
June 1, 1999 10 minutes read
Nosocomial outbreaks linked to staff shortages
Investigators find fewer staff may mean more staph
Health care staffing fluctuations, increasingly common in the managed care era, can spark nosocomial outbreaks if rising patient census or acuity outstrip nursing resources, clinical investigators are reporting.
As health care delivery changes continue, nosocomial infection rates in general may rise as patient severity of illness increases and all but the sickest patients are rapidly discharged.1-3 But clinicians — including several recently in San Francisco at the Society of Healthcare Epidemiology of America (SHEA) Conference — also are increasingly tracing nosocomial infections to staff problems caused by department mergers or addition of new services, spikes in census, cutbacks in nursing, or increased use of "pool" and agency nurses.
Though it can be difficult to establish a clear epidemiological link between such conditions and subsequent infections, the general consensus is that staffing problems — particularly if they occur in conjunction with an increase in patient acuity — may undermine aseptic technique, catheter care, and hand washing compliance by harried health care workers.4 (See Hospital Infection Control, June 1996, pp. 69-72.) HIC asked William Schaffner, MD, chairman of the department of preventive medicine at Vanderbilt University in Nashville, TN, to review some of the SHEA findings on the emerging trend.
"When we have difficulties in our surgical and trauma ICUs, it is when the personnel are stressed with more admissions and sicker patients," says Schaffner. "Obviously, there has to be a certain degree of flexibility in staffing, but when you reach the upper limits of that, [our nurses] are absolutely convinced — just by their observations — that health care personnel cut corners. They just have so much to do and they stop washing their hands, or they may be gloved but they go from patient to patient with the [same] gloves. Lots of little things like that begin to frazzle."
If so, current conditions generally ascribed to restructuring under managed care — higher inpatient acuity and tight-fisted budgets for nursing staff — would seem to be a virtual recipe for such problems to occur. But the link between staffing problems and subsequent infections is not a newly reported phenomena, as evidenced by studies reporting the problem in the early 1980s.5 In one such study, Schaffner and colleagues linked an outbreak of multiresistant Klebsiella pneumoniae infections in an intensive care nursery to inadequate nurse-to-patient staff ratios.6
"Clearly, the Klebsiella was being spread from person to person, and we think that it was on the hands of caregivers," he says. "Staffing ratios appeared to play a very important role."
While it is extremely difficult to tease out the variables in such studies to clearly ascribe the infections to staff changes as opposed to an increase in severity of patient illness, efforts to document the elusive connection may be used to sway administration from making staff cuts, he adds.
"It’s important to publish these because [ICPs] can take them to their administrators kind of prophylactically’ as reductions in force are anticipated," he says. "They can educate them that at some point, these reductions in force can have direct adverse consequences on the patients."
Latest findings presented at SHEA
In that regard, the latest wave of research on the issue was reported at the annual SHEA conference.
In one study, a staffing shortfall during a period of increased patient census was linked to an outbreak of Staphylococcus aureus in a neonatal intensive care unit at University Hospital in Cincinnati, reports Amy Beth Kressel, MD, director of infection control at the facility.7 Last year, the hospital became a referral center for high-risk pregnant women within its multihospital group. As a result, the census of the NICU increased and staffing fell below optimum levels. During the period of May 14-31, five infections with S. aureus occurred in the NICU. Four of the five infected neonates were in the same area of the NICU, and four of five infected babies were intubated.
"Usually, we have fewer than two infections per month, and they had five that were in close temporal association with each other," Kressel says. "When we looked at where the beds were [of] the infected neonates, four of them were in close proximity to each other — actually adjacent beds in one pod."
Given the close proximity of the cases, an initial theory was that a colonized health care worker was spreading the pathogen from patient to patient. Nurses were typically assigned to certain patients or pods of neonates on various shifts.
"But when we actually tried to find out who had been taking care of the babies, the nurses told us that they had a nursing shortage and in reality people were helping each other out, and even in rare instances they were going into other pods," she says. "So we really can’t say who specifically took care of these babies because there was a lot of cross-coverage. We also found that a disproportionate number of the infected neonates were intubated, so we [also investigated] respiratory therapy. We found that in both areas — respiratory therapy and nursing — there was a staff shortage."
Increase in census, acuity
Both overall census and the respiratory acuity index had increased approximately two weeks prior to the outbreak, and the ratio of patients to nurses grew beyond recommended levels. As a result, evaluation of nursing care revealed deficiencies in hand washing technique and glove use, use of nonstandard soaps, and inconsistencies separating clean and dirty areas. Four of the five infected patients had staph with the same DNA fingerprint pattern. Efforts that successfully stopped the outbreak included presentations to administration on the importance of adequate staffing and education of nurses and respiratory therapists on infection control techniques, especially hand washing.
"We concluded that overall decreased staffing had probably played a role," she says. "There had been a spike in our census just before all of this happened, and that seemed to support this as well. Our staffing levels have improved, and we have not had recurrent problems with Staph aureus in the NICU."
In another MRSA outbreak reported at the SHEA conference, Robert Cunney, MB, a research fellow at McMaster University Medical Center in Hamilton, Ontario, Canada, examined the impact of nursing and environmental staffing levels on a busy general surgical ward over a six-month period during July to December 1998.8 There were 29 nosocomial acquisitions of MRSA by patients on the ward during the time period. A marked increase in the incidence of new cases of colonization was noted shortly after a cost-saving cutback reduced the number of environmental cleaning personnel on the ward. The cleaning duties affected included terminal room cleaning after patient discharge and routine cleaning on a day-to-day basis in the ward. There also was a prior increase in the use of temporary "agency" nursing staff on the ward.
"While there wasn’t a reduction in the overall nurse staffing levels, there was an increase in the use of temporary staff brought in from outside agencies," he tells HIC. "There was also some transfer of staff from other units in the hospital. So our impression was because it wasn’t the same stable nursing base in this particular ward, that this may have accounted for the increases in nosocomial MRSA acquisition."
On regression analysis, the use of agency staff was significantly associated with MRSA acquisition rates, though overall nursing and environmental staff levels were not. However, both the use of extra nursing staff (i.e., outside of those normally scheduled to work on that ward) and the number of hours of environmental cleaning carried out each week were found to be linked to MRSA acquisition rates on covariance analysis. Cunney concluded that ICPs should pay attention to nursing and environmental cleaning staffing levels on wards where MRSA is a potential problem. Environmental cleaning hours have been increased since the findings, and a similar effort will be made with administration regarding the nursing staff issues, he notes.
"It may well have been that these two changes were almost synergistic," he says. "That when you combined extra use of agency staff — who perhaps were not as familiar with the patients, who were generally working overtime and may not have the same skills base — with a reduction in environmental cleaning, they seem to translate into a marked increase in our MRSA rates."
Inadequate hand washing may have again been a factor, because patients who were acquiring MRSA on the ward appeared to be the ones who required the most handling, he added.
A similar finding regarding use of pool nurses was reported in another SHEA study on central venous catheter-related bloodstream infections (CVC-BSIs) presented by Robert Duncan, MD, epidemiologist at Lahey Clinic in Burlington, MA.9
Duncan and colleagues studied factors affecting CVC-BSIs after services were altered in a surgical intensive care unit (SICU). SICU infection rates had remained stable from 1993 until 1997, but then rates of BSIs and CVC-BSIs increased significantly. CVC-BSI rates rose most dramatically (by 3.1-fold). This coincided with facility changes that included incorporation in February 1997 of a Level II trauma center and establishment of a separate five-bed cardiothoracic surgery post-anesthesia care unit (PACU). Intensity of daily nursing care requirements remained steady, but average length of stay doubled (from three to six days) at the same time the hospital ran into a shortage of nurses.
"We had unanticipated change in nursing staffing," he told SHEA attendees. "We had expected with the new five-bed cardiothoracic PACU that we were going to have empty beds in the surgical ICU. Suddenly we were scrambling for nurses in a fairly tight, competitive market for nursing staffing."
More staff were drawn from overtime and nursing pools in an effort to keep the nurse-patient ratio steady, he noted. However, make-up staffing by nurses less familiar with the SICU adversely affected CVC care and infection rates. With additional hiring in 1998, BSIs dropped 22%, from 18.16 per 1,000 critical care days to 14.08, and CVC-BSIs dropped 41%. During the peak of infection problems, the typical 7% level of pool nursing staff had nearly doubled to 13%, but as infections subsided with the hiring of new staff, the level of pool workers dropped back to 8%. The CVC-BSI rate may be a sensitive indicator of costly adverse outcomes related to overburdened nursing staff, he noted.
"We had a significant increase in the number of polymicrobial catheter infections, which I take as an indicator of sloppy line care," Duncan told SHEA attendees.
Such links between infection rates and staffing would seem to suggest that increases in nursing workload in general — without staff reductions or skill-level fluctuations — would contribute to increased patient colonization with nosocomial pathogens. But another SHEA study that explored the same issue found that only prior administration of antibiotics — not nursing workload intensity — was a statistically significant risk factor for subsequent patient colonization with vancomycin-resistant enterococci. (See related story, p. 81.)
"It may well be that certain pathogens are more likely spread under these circumstances than are others," Schaffner says. "The literature has a principal interest in staphylococci, and as we get into other organisms — VRE, for example — it may be that the threshold for that as your indicator organism is indeed higher. It may also depend on whether you look at certain site-specific infections as opposed to all nosocomial infections."
Indeed, because it can be difficult to account for all such possibilities, it will be important to see the complete presentation of data and methods of the aforementioned SHEA studies if they are published in the peer-reviewed medical literature, he notes.
"The questions that these investigators are trying to address are very important because hospitals are seeing an increase in severity of illness, trying to shorten hospital stays, and they are looking very carefully at personnel," Schaffner says. "Lots of hospitals are undergoing staff reductions of various kinds. The possible relationship of those circumstances to nosocomial infections clearly needs to be looked at. Based on the literature and these [SHEA] observations, one suspects that at least under certain circumstances the two are related. The devil is in the details."
References
1. Calfee DP, Cage EG, Farr BM. Secular trends in nosocomial infections during the era of managed care. Abstract 59. Presented at the Conference of the Society for Healthcare Epidemiology of America (SHEA). San Francisco; April 18-20, 1999.
2. Taylor G. Nosocomial blood stream infections (NBSI): Ten year trends in one institution. Abstract K-60. Presented at the Interscience Conference on Antimicrobial Agents and Chemotherapy. San Diego; Sept. 24-27, 1998.
3. Taylor G, McKenzie M, Kirkland M, et al. Nosocomial blood stream infections (NBSI) in a restructured hospital. Abstract K-63. Presented at the Interscience Conference on Antimicrobial Agents and Chemotherapy. San Diego; Sept. 24-27, 1998.
4. Fridkin SK, Pear SM, Williamson TH, et al. The role of understaffing in central venous catheter-associated bloodstream infections. Infect Control Hosp Epidemiol 1996; 17:150-158.
5. Haley RP, Bregman DA. The role of understaffing and over-crowding in recurrent outbreaks of staphylococcal infection in a neonatal special-care unit. J Infect Dis 1982; 145:875-885.
6. McKee KT, Cotton RB, Stratton CW, et al. Nursery epidemic due to multiple-resistant Klebsiella pneumoniae: Epidemiologic setting and impact on perinatal health care delivery. Infect Control 1982; 3:150-156.
7. Kidd F, Heitkemper P, Kressel A. A neonatal intensive care unit outbreak of S. aureus associated with inadequate staffing. Abstract S74. Presented at the Conference of the Society for Healthcare Epidemiology of America. San Francisco; April 18-20, 1999.
8. Cunney RJ, Thornley D, Bialachowski A, et al. Environmental and nursing staff levels: Relationship to nosocomial acquisition of methicillin-resistant Staphylococcus aureus (MRSA). Abstract M29. Presented at the Conference of the Society for Healthcare Epidemiology of America. San Francisco; April 18-20, 1999.
9. Duncan RA, Levine A, Willey S, et al. Nursing staffing and central venous catheter-related bloodstream infections (CVC-BSIs) in a changing surgical intensive care unit (SICU). Revised Abstract. Presented at the Conference of the Society for Healthcare Epidemiology of America. San Francisco; April 18-20, 1999.
You have reached your article limit for the month. Subscribe now to access this article plus other member-only content.
- Award-winning Medical Content
- Latest Advances & Development in Medicine
- Unbiased Content