Needlesticks more common than hospitals indicate

Study suggests means of lowering injury rates

Injuries from needlesticks occur more frequently than institutional reports suggest and do not happen at random, according to a multi-hospital study that also suggests solutions such as reducing the use of temporary nursing personnel, decreasing needle recapping, and implementing other organizational and behavioral changes.1

Report data were derived from a larger study of hospital AIDS care involving 40 inpatient units in 20 hospitals in 11 U.S. cities with a high incidence of AIDS. The larger study examined how the organization of inpatient hospital care relates to patient and nurse outcomes. The portion of the larger study included in the report pertains to the specific nurse outcome of occupational exposure to blood.

Researchers from the University of Pennsylvania (UP) in Philadelphia gathered and compared three kinds of data: prospective, retrospective, and institutional reports. They found that prospective and retrospective reports revealed similar injury rates, while institutional rates were significantly lower.

"It’s very clear that the large bulk of needlestick injuries continues to go unreported in hospitals," says researcher Douglas M. Sloane, PhD, associate professor in the UP’s sociology department and school of nursing and at the Catholic University of America in Washington, DC. "The expectation was that there would be a difference, but the magnitude of the difference was a surprise to all of us. [Institutional reports] provide only a small fraction of what actually occurs."

In addition, the authors found significantly lower needlestick rates among nurses in "magnet" hospitals than among those in nonmagnet hospitals. "Magnet hospitals, also known as professional nurse practice models, have distinct organizational attributes such as decentralized decision-making, policies promoting nurse autonomy and control, and work organization emphasizing continuity of care," the report explains.

Sloane suggests that because magnet hospitals place nurses in prominent administrative positions and experience less nurse turnover, nurses are more comfortable with the work environment, which results in lower needlestick injury rates.

Similarly, the study shows that staff nurses have lower needlestick rates than nonstaff or temporary nurses, "suggesting perhaps that familiarity with the setting in which the nursing practice is carried out has an impact," he says. Temporary and nonstaff nurses frequently are used in hospitals in times of restructuring when turnover takes place. In the study, prospective data showed that staff nurses had an injury rate of 0.84 per year, while nonstaff nurses’ injury rate was 1.38 per year.

The researchers continue to collect data related to occupational exposures in hospitals that are involved in restructuring, but the study’s findings indicate that "the recent downsizing and ‘deprofessionalizing’ of the hospital workforce is not without potential adverse consequences" in terms of injury rates. They point to a recent report on nurse staffing and safety from the Institute of Medicine in Washington, DC, which they say "calls for greater managerial attention to promising strategies for creating safer work environments for hospital nurses."2 (See related story in Hospital Employee Health, April 1996, pp. 43-47.)

To obtain their prospective data, the researchers documented percutaneous injuries for every shift during a 30-day period. Information on blood exposures was derived from coupons filled out by staff and nonstaff temporary nurses after each shift over two periods of one month each. Nurses indicated whether they had incurred a needlestick injury, defined as "a puncture with a needle or sharp instrument that is contaminated with blood." Participants worked a total of 14,379 shifts, and 12,349 coupons were returned, for an 86% response rate. Data were analyzed from 12,075 coupons (98%) — 11,039 completed by 920 staff nurses, and 1,036 completed by an unknown number of nonstaff temporary nurses.

Retrospective data were gathered by questionnaires that asked, "Have you ever been stuck with a needle or sharp object contaminated with blood?" Three additional questions followed for nurses who answered affirmatively:

• How many times has this occurred?

• How many of these incidents occurred in the past month?

• Did this incident go unreported to your institution’s office of employee health or comparable office?

A total of 762 of the 865 prospective study nurses who received the retrospective questionnaires returned them (88%). Researchers were able to use 732 (96%) of the questionnaires. Using common identification numbers, they linked retrospective injury reports to prospective injury reports for 732 (80%) of the 920 staff nurses in the prospective part of the study.

Institutional reports were provided by 15 of the 20 hospitals. Data included dates of reported injuries, unit on which the injury occurred, exposure circumstances, injury cause, action taken, and the nurse’s educational background. The six-month reporting period extended from the month before administration of the retrospective questionnaire to the end of prospective data collection.

Institutional reports reflect ‘only a fraction’

The researchers found no significant difference between prospective and retrospective data, but the monthly injury rate derived from the institutional data was significantly lower than either the prospective or retrospective rates. Forty-one injuries were reported on the study units of the 15 hospitals providing institutional injury data, a monthly injury rate of 0.46 per hospital. In the prospective data, 24 injuries were reported by nurses from those 15 hospitals over the course of one month, a monthly rate of 1.60 injuries per hospital. The nurses from those 15 hospitals who responded to the retrospective questionnaire reported 29 injuries over the preceding month, a monthly rate of 1.93 injuries per hospital.

"The institutional reports reflect only a fraction of the actual injuries that occur," the report states.

The researchers also examined whether certain behaviors or characteristics of nurses were related to the likelihood of an injury occurring. They found that the frequency with which nurses handled blood was positively related to whether they sustained injuries. Nurses who often or sometimes handled blood had odds of being injured that were more than 50% greater than those of nurses who rarely or never handled blood.

Recapping needles and taking precautions to avoid blood and body fluid contacts also were related to needlestick incidents. Nurses who sometimes or often recapped were more than 1.4 times more likely than those who never did to report a needlestick injury during the prospective part of the study, 2.2 times more likely to retrospectively report an injury in the previous month, and 1.8 times more likely to report ever having been stuck.

Nurses who said they did not always take precautions to avoid blood/body fluid contact were 1.4 times more likely to report a needlestick in the prospective part of the study than those who said they always take precautions, 0.9 times more likely to retrospectively report an injury in the previous month, and 0.5 times more likely to report ever having been stuck.

"You’d expect that somebody who says they take precautions — regardless of whether the patient is known to be HIV-positive or not — will be less likely to be stuck," says Sloane. "Exercising precautions does have a large impact, which is to say these things aren’t random acts entirely."

The study concludes that one of the most important interventions to prevent needlestick injuries is providing nurses with safer needle devices, "despite the higher cost of such devices and the seeming opposition of a sizable percentage of hospital managers to paying for them."

"There is very little question that hospitals are going to have to provide nurses with some of the newer technologies," Sloane says, "and now there are many needleless devices that some hospitals have been reluctant to use because of their cost. The average nurse can expect something akin to three or four needlesticks every five years, and it suggests that [hospital administrators] should probably take this quite seriously. Apart from trying to do more to make sure that nurses are working in environments in which they are very familiar and comfortable, they need to be investing in the kind of technology that would help prevent these accidents."


1. Aiken LH, Sloane DM, Klocinski JL. Hospital nurses’ occupational exposure to blood: Prospective, retrospective, and institutional reports. Am J Public Health 1997; 87:103-107.