- Email: [email protected]
Reducing percutaneous injuries at an academic health center: A 5-year review
Reducing percutaneous injuries at an academic health center: A 5-year review Marcia Trape´-Cardoso, MD, FACP, and Paula Schenck, MPH Farmington, Connecticut
Background: The University of Connecticut Health Center Employee Health Service collected and used National Surveillance System for Hospital Health Care Workers (NaSH) data to (1) improve surveillance of health care worker blood and body fluid exposures (BBFEs) and (2) target specific interventions for higher-risk groups (nursing staff, medical and dental students, and residents). Methods: All 870 BBFE incidents were abstracted from the NaSH database from the 1997 through 2002 academic years. Incidence rates per 100 full-time-equivalent workers were determined for each targeted occupation group with 95% confidence intervals. Results: The number of percutaneous injuries declined among medical/dental students and nursing staff, and to a lesser degree for residents. The incidence rates decreased from 7.9% in 2000 to 2001 to 2.6% in 2001 to 2002 for students and from 9.2% in 1997 to 1998 to 2.7% in 2001 to 2002 for nursing staff. Conclusions: Data from a surveillance database provided guidance for administrative, educational, and engineering control interventions. Active surveillance and periodic review of interventions are important aspects to reduce BBFEs in targeted high-risk occupational groups, especially when the workforce has a high turnover, as is typical in academic health centers. (Am J Infect Control 2004;32:301-5.)
The Employee Health Service (EHS) at the University of Connecticut Health Center (UCHC) participated in the National Surveillance System for Hospital Health Care Workers (NaSH) program in the academic years 1997 to 1998 and 1998 to 1999 and collected data on percutaneous injuries. Using the Centers for Disease Control and Prevention (CDC) software, the EHS integrated the NaSH information with the UCHC’s infection control program (1) to improve surveillance by raising awareness and encouraging reporting and (2) to target specific interventions to identified high-risk occupational groups to reduce percutaneous injuries and dangerous activities. From the Division of Occupational and Environmental Medicine, Department of Medicine, University of Connecticut School of Medicine. Reprint requests: Marcia Trape´-Cardoso, MD, FACP, Associate Professor of Clinical Medicine, 263 Farmington Avenue, Farmington, CT 06032-6210; E-mail: [email protected]. This work was partially supported by the Centers for Disease Control and Prevention, Grant H75/CCH117607-01. 0196-6553/$30.00 Copyright ª 2004 by the Association for Professionals in Infection Control and Epidemiology, Inc. doi:10.1016/j.ajic.2003.12.003
The CDC estimates that between 300,000 and 400,000 percutaneous occupational injuries occur per year in the United States, and some have estimated up to 500,000 per year, with approximately 1% from patients who tested positive for the human immunodeficiency virus (HIV).1 Thus, needlestick injuries and other percutaneous injuries are a major occupational risk to health care workers (HCWs). In the United States, 56 cases of occupational HIV transmission and 138 cases of possible conversion in HCW had been documented as of December 2000.2 Worldwide, there are 94 documented cases of HCWs who acquired HIV from an occupational exposure and another 170 possible occupational cases.3 In addition, hepatitis C and B viruses may be transmitted to HCWs after these percutaneous exposures. Delay in seroconversion to HIV after an occupational exposure with simultaneous infection with hepatitis C virus has been reported, and resistance to the usual antiretroviral employed for HIV postexposure prophylaxis (PEP) compound the severity of the problem.4,5 The UCHC has been tracking percutaneous injuries to its HCWs since 1993. The UCHC takes part in NaSH, a voluntary reporting system with a template to systematically collect information on occupational exposures and infections among workers. Specifically, 301
302
Trape´-Cardoso and Schenck
Vol. 32 No. 5
the reporting system, developed by the CDC, addresses immunization; tuberculin-skin-testing programs; and exposure events to (1) blood and body fluids, (2) vaccine-preventable diseases, and (3) tuberculosis. This system forms the core of the EHS’s surveillance program for blood and body fluid exposures (BBFEs). With the introduction of NaSH software in 1997, monitoring of BBFE rates improved, and occupational groups with a high risk for percutaneous injuries were more easily identified.6 Reports of percutaneous injuries with BBFE at the UCHC were 82 per 5220 HCWs (1.6%) in the year before NaSH. These rates increased to 155 per 5305 HCWs (2.9%) and 189 per 5422 HCWs (3.5%) in the 2 years after implementation of the NaSH program. The increased numbers and rates of injuries reported most likely reflect improved awareness of the importance of evaluation and treatment after an incident rather than increased problematic practices.
METHODS Implementation of targeted interventions and assessment of surveillance information were used to explore the effectiveness of the program to reduce BBFEs.
4. Core Curriculum Unit on BBFEs at the UCHC Medical and Dental Schools: this curriculum includes a detailed lecture on blood-borne pathogen prevention and PEP. In addition, all medical and dental students have instructions on available personnel protective equipment (PPE), procedures to follow after an exposure, and hands-on practice with safety devices. 5. Broadly advertised open-to-public lectures (Ground Rounds) for faculty physicians, residents, medical students, and nursing staff on Percutaneous Injuries and PEP. 6. EHS Health and Wellness Monthly Cafeteria Exhibit: a tabletop and poster exhibit with handout fact sheets about hepatitis B and hepatitis C, safety devices, and risk assessment after a needlestick injury is set up directly outside the door where the majority of employees pass daily for lunch. Available safety devices are on display. 7. Annual Blood-borne Pathogen training as required by the Occupational Safety and Health Administration: computerized interactive instructional modules have been updated to include safety devices and use of gloves and other PPE. Each professional participant is automatically registered and rewarded with continuing education unit or continuing medical education credit.
Procedures/intervention program Intervention activities included increased education, changes in purchasing, and more administrative involvement. Beginning with the 1997 to 1998 year, the UCHC education program was expanded to include newly available safety devices and safe practice protocols for the targeted occupational groups. Specifically, the following training and educational activities were implemented: 1. Pocket cards with the CDC risk assessment for PEP given to all medical and dental students and to all incoming residents during orientation: these cards contain the EHS Clinic and Emergency Department telephone numbers to encourage reporting of injuries. 2. Extensive demonstration and instruction on proper utilization of the safety devices and on procedures to follow percutaneous injury at work: the EHS mails a self-learning module with a posttest survey to all residents and instructs incoming residents on the safety program. Staff members have the opportunity to try various shielded safety devices. 3. Lecture on Preventing Percutaneous Occupational Injuries through the VTEL (Video TELecommunication) interhospital system: the lecture is simulcast to all the area hospital sites served by UCHC trainees. The lectures use a case-study format.
Engineering controls implementation In 1994, the UCHC switched to safety piggyback (interlink) systems. In 1995, the institution introduced safety butterfly needles, followed by safety retractable lancets in 1998. Finally, after demonstration in 1999 (using the NaSH data) that most of the percutaneous exposures occurred from hypodermic needles attached to syringes, the institution implemented the most recent change, the provision of safety hollow-bore needle devices for injections. Beginning in the summer of 2000, the UCHC mandated that all needles attached to syringes had to be safety needles. This included the needles used for phlebotomies.
Data assessment The NaSH database was reviewed, and information was abstracted on the occupation of each injured HCW, the place where the incident happened, the HIV and hepatitis B and C status of the source patient, the presence of visible blood on the sharp, the manner in which the injury occurred, whether the injury occurred through gloves or other clothing, and the depth and body site of the injury. Incidence rates per 100 fulltime-equivalent workers (FTEs) were calculated to explore trends in reporting and to provide feedback on the intervention program. This study was reviewed
Trape´-Cardoso and Schenck
by the University Institutional Review Board and approved as exempt. Incidence rates were calculated by occupational groups and by year. Two-by-two contingency tables from absolute numbers were calculated using the Yates-corrected x2.7 To establish a reporting baseline, risk analyses for BBFEs (unfavorable outcomes) were compared each year with the initial year that NaSH was used. Significance was measured at a = 0.05. The incidence rate of BBFEs per 100 FTEs for each targeted occupation group was calculated with 95% confidence intervals (CIs).
RESULTS Using the NaSH data from the 1997 through 1999 academic years, 3 occupational groups at high risk for percutaneous exposures were identified. The groups are (1) residents; (2) medical and dental students; and (3) nursing staff, including nurse practitioners, physician assistants, registered nurses, medical assistants, clinical nurse specialists, licensed practical nurses, and nurses’ assistants. Other groups (dental assistants, dental hygienists, dentists, hospital operating room staff, and ambulatory procedures–surgical suite staff, including employees assigned to the endoscopic procedures rooms, same-day surgery rooms, and invasive cardiology and radiology suites) at the UCHC benefited from the overall program but were not targeted for specific interventions. From 1997 to 2002, residents, nursing staff, and medical and dental students accounted for over 80% of the BBFE reports. During the 60 months from May 1997 to April 2002, there were 870 blood and body fluid exposures at the UCHC. Of those 870, 718 (82.5%) were percutaneous, 145 (16.5%) were exposures of skin and/ or mucous membrane, and 7 exposures (1%) were human bites. The largest occupational group with BBFE was the residents, who constituted 50.8% (65 of 128) of the reported exposures in 1997 to 1998 and 46% (58 of 125) of the exposures in 2001-2002. Twenty-five percent of their reported exposures (218 of 870) occurred outside the UCHC campus. From these 218 injuries, 193 HCWs sustained injuries while working at correctional facilities throughout the state (HCWs from the university have been working under a contract with the Department of Corrections since 1998). Students and residents in training at other area hospitals make up the remaining exposures outside the UCHC buildings. When BBFEs were stratified by occupational groups, clinical staff (nurses, clinical technicians, nurse/medical assistants, residents, physicians, operating room and procedure room staff, dental staff, and phlebotomists) accounted for 92% (801 of 870) of the exposures. Moderate-risk occupational groups, including laboratory researchers, faculty, nonclinical technicians, lab-
August 2004
303
oratory/research assistants, housekeepers, and other maintenance staff, reported 6% (51) of all the injuries, and only 2 exposures were reported in the low-risk occupational groups, consisting of clerical and administrative workers. More women (59%) than men (41%) reported BBFEs. Of the 870 exposures, 77% (669) involved white non-Hispanics, 10% (85) Asian or Pacific Islanders, 5% (43) African American non-Hispanics, and 4% (38) Hispanics. Ethnicity data was missing in 4% of the BBFEs. The mean age of the population with BBFE was 36.7 years, with a range from 19 to 70 years, and 609 (70%) of the 870 exposures occurred in individuals born in the United States. Some individuals (110 of 870, 13%) had more than one injury. Of these, 90 (81.8%) had 2 percutaneous injuries, 13 (11.8%) experienced 3 percutaneous injuries, 3 had 4 (2.7%), and 4 had 5 injuries (3.6%). Only dental and medical students reported 3 or more BBFE. Most of these multiple exposures were of extremely low risk, suggesting the students’ awareness of the need to report promptly. The numbers of BBFE reports over 4 years from nursing staff (Table 1) and from medical and dental students (Table 2) are compared with the initial year (1997-1998) using NaSH data. Although not significant the first year, nursing staff showed a significant decrease in BBFE for the last 3 years when compared with the baseline year of data. Reporting by medical and dental students initially increased and then continued at the same level for 2 more years with minimal changes. The number of exposures significantly decreased in the fourth year in comparison with the baseline. When medical and dental students’ percutaneous BBFE reports for the fifth year (2001-2002) are compared with those from the second year (19981999, when educational interventions were implemented and some engineering controls added), injuries can be seen to have decreased significantly (x2 df (1) = 11.9, P = .001). Figs 1 and 2 present the reported percutaneous BBFE per 100 FTEs for the medical and dental students and nursing staff, respectively, over the 5 years studied (1997-1998 through 2001-2002). Incidence rates of percutaneous injuries among medical and dental students decreased significantly, from 7.9% (95% CI; range, 6%-10%) to 2.6% (95% CI; range, 1%-4%), between 2000-2001 and 2001-2002. There was also a significant decrease in percutaneous injuries among nursing staff, from 9.2% (95% CI; range, 6%-12%) in 1997 to 1998 to 2.7% (95% CI; range, 2%-4%) in 20012002. The residents had a minimal (not significant) decrease in the incidence rate of percutaneous BBFE. However, injuries among residents trended down after
304
Trape´-Cardoso and Schenck
Vol. 32 No. 5
Fig 2. Percutaneous exposures (with 95% confidence intervals) among nursing staff (registered nurses, licensed practical nurses, medical assistants, nurse practitioners, nurses’ assistants, physician assistants, and clinical nurse specialists).
Fig 1. Percutaneous exposures (with 95% confidence intervals) in medical and dental students.
Table 1. Nursing staff’s* percutaneous BBFEs per year compared with baseline year of tracking Year
BBFEs
FTEs
x2 (df 1)
38 61 41 50 25
415 830 841 834 952
Baseline year 1.000 7.931 3.760 26.572
1997-1998 1998-1999 1999-2000 2000-2001 2001-2002
P
.318 .005 .052 .000
BBFE, Blood and body fluid exposures; FTE, full-time-equivalent worker. *Nursing staff includes registered nurses, licensed practical nurses, nurses’ assistants, nurse practitioners, physician assistants, clinical nurse specialists, and medical assistants.
Table 2. Medical and dental students’ percutaneous BBFEs per year compared with first year of tracking Year
BBFEs
FTEs
x2(df 1)
P
21 39 32 38 12
498 505 494 480 468
Baseline year 4.874 2.079 5.267 1.528
.027 .149 .022 .001
1997-1998 1998-1999 1999-2000 2000-2001 2001-2002
BBFE, Blood and body fluid exposures; FTE, full-time-equivalent workers.
a peak 13.5% (95% CI; range, 11%-17%) rate of percutaneous injuries in 1999 to 2000, but then achieved lower rates in the next 2 years, with 7.6% (95% CI; range, 5%-10%) in 2000 to 2001 and 9.8% (95% CI; range, 7%-12%) in 2001 to 2002.
DISCUSSION The decreases in annual incidence rates reported here are comparable to those in the current literature on HCWs.8 The Canadian Needle Stick Surveillance
Network9 reports that resident physicians had the highest incidence rate of BBFE events. The Canadian Network reported that from April 2000 through March 2001, the annual exposure rate per 100 FTEs was 4.88% for registered nurses, 6.61% for medical students, and 20.97% for resident physicians. The rate of exposures considering all exposures per 100 FTEs was 4.24%.9 In a study of HCWs in Italy, the highest rates of occupational BBFE occurred in nurses (7.8%) and physicians (1.9%) working in infectious disease units.10 Reddy and Emery11 reported a progressive improvement in the rates of BBFE before and after interventions. The overall incidence rate fell from 10.6% per 100 FTEs to 4.2% over a 5-year period. The same paper discusses some of the difficulties in comparing occupations and departments due to insufficient information about the individuals at risk who would constitute the denominator of the rate calculations. The decision to analyze reporting trends in the 3 high-risk occupational groups was in part driven by the fact that we have accurate denominators on those occupations or trainees. With guidance from analyzing NaSH data, the EHS has targeted interventions at selected groups within the UCHC. It is not surprising that the 5-year data for all 3 groups did not consistently trend downward in rate of injury reports. The rates probably reflect our success in encouraging reporting of incidents as well as the effectiveness of our interventions in reducing injuries. The trend in the rates of BBFE per 100 FTEs for students increased in the initial year, then generally decreased as the intervention program matured. The downward trend is stronger for nursing staff and may reflect the benefits of the well-established outreach program for
Trape´-Cardoso and Schenck
this group at the UCHC. As more experience with the student group accrues, it will be interesting to see if the downward trend in reports established over the last year strengthens. The variability in the incidence rate over time for residents raises concerns about unreliable reporting to the EHS from residents working at remote hospitals. It is possible that not all injuries have been reported, perhaps because residents may frequently be too busy to report and may find it difficult to come directly to our EHS from the hospital where they are assigned. Other confounding elements in analyzing the data include the variability in the UCHC workforce level over the 5 years. In particular, the addition of HCWs from the Department of Corrections in 1998 may have influenced the results. Significant layoffs occurred at the UCHC in 1999 and 2002, and that too may have an impact on the incidence and reporting of injuries. Changes in individuals’ workload over the 5 years also may have contributed to the results obtained. Other barriers to the success of our monitoring program are the variability of BBFE policy in the various institutions in which our personnel work, the diversity of safety devices available, and inconsistency of reporting procedures at the various hospitals in the area where residents and students train. The UCHC administration showed high levels of interest in participating and permitting their staff to attend training and educational activities planned. Moreover, there was an interest in cooperating with EHS staff after a percutaneous injury. Because the workers were more aware of the need for timely risk assessment and antiviral therapy, timely referrals for care increased. Complaints about insensitive supervisors and managers decreased to zero.
CONCLUSIONS To reduce percutaneous BBFE at the UCHC, we utilized data from the NaSH to demonstrate to the administration the importance of purchasing safety devices such as shielded hypodermic needles and needles attached to syringes. We also had to invest in educational and training activities targeted at the highrisk occupational groups identified through the NaSH database. After administrative interventions, engineering controls, and educational modules, there was
August 2004
305
a significant decrease in percutaneous injuries to medical and dental students and to nursing staff over the 5-year period. There was also a mild temporary decrease in these injuries among residents. Strict surveillance of BBFE at the UCHC is crucial to detect areas requiring interventions to reduce the rate of injuries among workers and trainees. The authors thank Dr Richard A. Garibaldi for guidance and support to the Blood and Body Fluid Exposure Control Program and for reviewing this paper, Dr Kenneth H. Dangman for his thoughtful review of the manuscript, and Andrew Warren, MPH, for data collection and initial data analysis.
References 1. Bell DM. Occupational risk of human immunodeficiency virus infection in healthcare workers: an overview. Am J Med 1997;102:9-15. 2. Centers for Disease Control and Prevention. Updated U.S. Public Health Service Guidelines for management of occupational exposure to HBV, HCV, and HIV and recommendations for post-exposure prophylaxis. Atlanta (GA): Centers for Disease Control and Prevention Guidelines; 2001:50 (RR-11). 3. Ippolito G, Puro V, Heptonstall J, Jagger J, Carli GD, Petrosillo N. Occupational human immunodeficiency virus infection in health care workers: Worldwide cases through September 1997. Clinical Infect Dis 1999;28:365-83. 4. Ridzon R, Gallagher K, Ciesielski C, Ginsberg MB, Robertson BJ, Luo CC, et al. Simultaneous transmission of human immunodeficiency virus and hepatitis C virus from a needlestick injury. N Engl J Med 1997;336:919-22. 5. Ippolito G, Puro V, Petrosillo N, De Carli G, Micheloni G, Magliano E, et al. Simultaneous infection with HIV and hepatitis C following occupational conjunctival blood exposure [letter]. JAMA 1998;280:28. 6. Centers for Disease Control and Prevention National Surveillance System for Hospital Health Care Workers. Available from: http: //www.cdc.gov/ncidod/hip/SURVEILL/nash.htm. Accessed July 2003. 7. Fleiss JL. Statistical methods for rates and proportions. 2nd edition. New York: John Wiley & Sons; 1981. 8. Wang SA, Panlilio AL, Doi PA, White AD, Stek M, Saah A, et al. Experience of healthcare workers taking postexposure prophylaxis after occupational HIV exposures: findings of the HIV postexposure prophylaxis registry. Infect Control Hosp Epidemiol 2000;21:780-5. 9. Health Canada. Update: surveillance of healthcare workers exposed to blood/body fluids and bloodborne pathogens, 1 April, 2000 to 31 March, 2001. Canada Communicable Disease Report (CCDR) December 2001;27. 10. Puro V, Carli GD, Petrosillo N, Ippolito G. Risk of exposure to bloodborne infection for Italian healthcare workers, by job category and work area. Infec Control Hosp Epidemiol 2001;22:206-10. 11. Reddy SG, Emery RJ. Assessing the effect of long-term availability of engineering controls on needlestick injuries among health care workers: 3 year preimplementation and postimplementation comparison. Am J Infect Control 2001;29:425-7.
Background: The University of Connecticut Health Center Employee Health Service collected and used National Surveillance System for Hospital Health Care Workers (NaSH) data to (1) improve surveillance of health care worker blood and body fluid exposures (BBFEs) and (2) target specific interventions for higher-risk groups (nursing staff, medical and dental students, and residents). Methods: All 870 BBFE incidents were abstracted from the NaSH database from the 1997 through 2002 academic years. Incidence rates per 100 full-time-equivalent workers were determined for each targeted occupation group with 95% confidence intervals. Results: The number of percutaneous injuries declined among medical/dental students and nursing staff, and to a lesser degree for residents. The incidence rates decreased from 7.9% in 2000 to 2001 to 2.6% in 2001 to 2002 for students and from 9.2% in 1997 to 1998 to 2.7% in 2001 to 2002 for nursing staff. Conclusions: Data from a surveillance database provided guidance for administrative, educational, and engineering control interventions. Active surveillance and periodic review of interventions are important aspects to reduce BBFEs in targeted high-risk occupational groups, especially when the workforce has a high turnover, as is typical in academic health centers. (Am J Infect Control 2004;32:301-5.)
The Employee Health Service (EHS) at the University of Connecticut Health Center (UCHC) participated in the National Surveillance System for Hospital Health Care Workers (NaSH) program in the academic years 1997 to 1998 and 1998 to 1999 and collected data on percutaneous injuries. Using the Centers for Disease Control and Prevention (CDC) software, the EHS integrated the NaSH information with the UCHC’s infection control program (1) to improve surveillance by raising awareness and encouraging reporting and (2) to target specific interventions to identified high-risk occupational groups to reduce percutaneous injuries and dangerous activities. From the Division of Occupational and Environmental Medicine, Department of Medicine, University of Connecticut School of Medicine. Reprint requests: Marcia Trape´-Cardoso, MD, FACP, Associate Professor of Clinical Medicine, 263 Farmington Avenue, Farmington, CT 06032-6210; E-mail: [email protected]. This work was partially supported by the Centers for Disease Control and Prevention, Grant H75/CCH117607-01. 0196-6553/$30.00 Copyright ª 2004 by the Association for Professionals in Infection Control and Epidemiology, Inc. doi:10.1016/j.ajic.2003.12.003
The CDC estimates that between 300,000 and 400,000 percutaneous occupational injuries occur per year in the United States, and some have estimated up to 500,000 per year, with approximately 1% from patients who tested positive for the human immunodeficiency virus (HIV).1 Thus, needlestick injuries and other percutaneous injuries are a major occupational risk to health care workers (HCWs). In the United States, 56 cases of occupational HIV transmission and 138 cases of possible conversion in HCW had been documented as of December 2000.2 Worldwide, there are 94 documented cases of HCWs who acquired HIV from an occupational exposure and another 170 possible occupational cases.3 In addition, hepatitis C and B viruses may be transmitted to HCWs after these percutaneous exposures. Delay in seroconversion to HIV after an occupational exposure with simultaneous infection with hepatitis C virus has been reported, and resistance to the usual antiretroviral employed for HIV postexposure prophylaxis (PEP) compound the severity of the problem.4,5 The UCHC has been tracking percutaneous injuries to its HCWs since 1993. The UCHC takes part in NaSH, a voluntary reporting system with a template to systematically collect information on occupational exposures and infections among workers. Specifically, 301
302
Trape´-Cardoso and Schenck
Vol. 32 No. 5
the reporting system, developed by the CDC, addresses immunization; tuberculin-skin-testing programs; and exposure events to (1) blood and body fluids, (2) vaccine-preventable diseases, and (3) tuberculosis. This system forms the core of the EHS’s surveillance program for blood and body fluid exposures (BBFEs). With the introduction of NaSH software in 1997, monitoring of BBFE rates improved, and occupational groups with a high risk for percutaneous injuries were more easily identified.6 Reports of percutaneous injuries with BBFE at the UCHC were 82 per 5220 HCWs (1.6%) in the year before NaSH. These rates increased to 155 per 5305 HCWs (2.9%) and 189 per 5422 HCWs (3.5%) in the 2 years after implementation of the NaSH program. The increased numbers and rates of injuries reported most likely reflect improved awareness of the importance of evaluation and treatment after an incident rather than increased problematic practices.
METHODS Implementation of targeted interventions and assessment of surveillance information were used to explore the effectiveness of the program to reduce BBFEs.
4. Core Curriculum Unit on BBFEs at the UCHC Medical and Dental Schools: this curriculum includes a detailed lecture on blood-borne pathogen prevention and PEP. In addition, all medical and dental students have instructions on available personnel protective equipment (PPE), procedures to follow after an exposure, and hands-on practice with safety devices. 5. Broadly advertised open-to-public lectures (Ground Rounds) for faculty physicians, residents, medical students, and nursing staff on Percutaneous Injuries and PEP. 6. EHS Health and Wellness Monthly Cafeteria Exhibit: a tabletop and poster exhibit with handout fact sheets about hepatitis B and hepatitis C, safety devices, and risk assessment after a needlestick injury is set up directly outside the door where the majority of employees pass daily for lunch. Available safety devices are on display. 7. Annual Blood-borne Pathogen training as required by the Occupational Safety and Health Administration: computerized interactive instructional modules have been updated to include safety devices and use of gloves and other PPE. Each professional participant is automatically registered and rewarded with continuing education unit or continuing medical education credit.
Procedures/intervention program Intervention activities included increased education, changes in purchasing, and more administrative involvement. Beginning with the 1997 to 1998 year, the UCHC education program was expanded to include newly available safety devices and safe practice protocols for the targeted occupational groups. Specifically, the following training and educational activities were implemented: 1. Pocket cards with the CDC risk assessment for PEP given to all medical and dental students and to all incoming residents during orientation: these cards contain the EHS Clinic and Emergency Department telephone numbers to encourage reporting of injuries. 2. Extensive demonstration and instruction on proper utilization of the safety devices and on procedures to follow percutaneous injury at work: the EHS mails a self-learning module with a posttest survey to all residents and instructs incoming residents on the safety program. Staff members have the opportunity to try various shielded safety devices. 3. Lecture on Preventing Percutaneous Occupational Injuries through the VTEL (Video TELecommunication) interhospital system: the lecture is simulcast to all the area hospital sites served by UCHC trainees. The lectures use a case-study format.
Engineering controls implementation In 1994, the UCHC switched to safety piggyback (interlink) systems. In 1995, the institution introduced safety butterfly needles, followed by safety retractable lancets in 1998. Finally, after demonstration in 1999 (using the NaSH data) that most of the percutaneous exposures occurred from hypodermic needles attached to syringes, the institution implemented the most recent change, the provision of safety hollow-bore needle devices for injections. Beginning in the summer of 2000, the UCHC mandated that all needles attached to syringes had to be safety needles. This included the needles used for phlebotomies.
Data assessment The NaSH database was reviewed, and information was abstracted on the occupation of each injured HCW, the place where the incident happened, the HIV and hepatitis B and C status of the source patient, the presence of visible blood on the sharp, the manner in which the injury occurred, whether the injury occurred through gloves or other clothing, and the depth and body site of the injury. Incidence rates per 100 fulltime-equivalent workers (FTEs) were calculated to explore trends in reporting and to provide feedback on the intervention program. This study was reviewed
Trape´-Cardoso and Schenck
by the University Institutional Review Board and approved as exempt. Incidence rates were calculated by occupational groups and by year. Two-by-two contingency tables from absolute numbers were calculated using the Yates-corrected x2.7 To establish a reporting baseline, risk analyses for BBFEs (unfavorable outcomes) were compared each year with the initial year that NaSH was used. Significance was measured at a = 0.05. The incidence rate of BBFEs per 100 FTEs for each targeted occupation group was calculated with 95% confidence intervals (CIs).
RESULTS Using the NaSH data from the 1997 through 1999 academic years, 3 occupational groups at high risk for percutaneous exposures were identified. The groups are (1) residents; (2) medical and dental students; and (3) nursing staff, including nurse practitioners, physician assistants, registered nurses, medical assistants, clinical nurse specialists, licensed practical nurses, and nurses’ assistants. Other groups (dental assistants, dental hygienists, dentists, hospital operating room staff, and ambulatory procedures–surgical suite staff, including employees assigned to the endoscopic procedures rooms, same-day surgery rooms, and invasive cardiology and radiology suites) at the UCHC benefited from the overall program but were not targeted for specific interventions. From 1997 to 2002, residents, nursing staff, and medical and dental students accounted for over 80% of the BBFE reports. During the 60 months from May 1997 to April 2002, there were 870 blood and body fluid exposures at the UCHC. Of those 870, 718 (82.5%) were percutaneous, 145 (16.5%) were exposures of skin and/ or mucous membrane, and 7 exposures (1%) were human bites. The largest occupational group with BBFE was the residents, who constituted 50.8% (65 of 128) of the reported exposures in 1997 to 1998 and 46% (58 of 125) of the exposures in 2001-2002. Twenty-five percent of their reported exposures (218 of 870) occurred outside the UCHC campus. From these 218 injuries, 193 HCWs sustained injuries while working at correctional facilities throughout the state (HCWs from the university have been working under a contract with the Department of Corrections since 1998). Students and residents in training at other area hospitals make up the remaining exposures outside the UCHC buildings. When BBFEs were stratified by occupational groups, clinical staff (nurses, clinical technicians, nurse/medical assistants, residents, physicians, operating room and procedure room staff, dental staff, and phlebotomists) accounted for 92% (801 of 870) of the exposures. Moderate-risk occupational groups, including laboratory researchers, faculty, nonclinical technicians, lab-
August 2004
303
oratory/research assistants, housekeepers, and other maintenance staff, reported 6% (51) of all the injuries, and only 2 exposures were reported in the low-risk occupational groups, consisting of clerical and administrative workers. More women (59%) than men (41%) reported BBFEs. Of the 870 exposures, 77% (669) involved white non-Hispanics, 10% (85) Asian or Pacific Islanders, 5% (43) African American non-Hispanics, and 4% (38) Hispanics. Ethnicity data was missing in 4% of the BBFEs. The mean age of the population with BBFE was 36.7 years, with a range from 19 to 70 years, and 609 (70%) of the 870 exposures occurred in individuals born in the United States. Some individuals (110 of 870, 13%) had more than one injury. Of these, 90 (81.8%) had 2 percutaneous injuries, 13 (11.8%) experienced 3 percutaneous injuries, 3 had 4 (2.7%), and 4 had 5 injuries (3.6%). Only dental and medical students reported 3 or more BBFE. Most of these multiple exposures were of extremely low risk, suggesting the students’ awareness of the need to report promptly. The numbers of BBFE reports over 4 years from nursing staff (Table 1) and from medical and dental students (Table 2) are compared with the initial year (1997-1998) using NaSH data. Although not significant the first year, nursing staff showed a significant decrease in BBFE for the last 3 years when compared with the baseline year of data. Reporting by medical and dental students initially increased and then continued at the same level for 2 more years with minimal changes. The number of exposures significantly decreased in the fourth year in comparison with the baseline. When medical and dental students’ percutaneous BBFE reports for the fifth year (2001-2002) are compared with those from the second year (19981999, when educational interventions were implemented and some engineering controls added), injuries can be seen to have decreased significantly (x2 df (1) = 11.9, P = .001). Figs 1 and 2 present the reported percutaneous BBFE per 100 FTEs for the medical and dental students and nursing staff, respectively, over the 5 years studied (1997-1998 through 2001-2002). Incidence rates of percutaneous injuries among medical and dental students decreased significantly, from 7.9% (95% CI; range, 6%-10%) to 2.6% (95% CI; range, 1%-4%), between 2000-2001 and 2001-2002. There was also a significant decrease in percutaneous injuries among nursing staff, from 9.2% (95% CI; range, 6%-12%) in 1997 to 1998 to 2.7% (95% CI; range, 2%-4%) in 20012002. The residents had a minimal (not significant) decrease in the incidence rate of percutaneous BBFE. However, injuries among residents trended down after
304
Trape´-Cardoso and Schenck
Vol. 32 No. 5
Fig 2. Percutaneous exposures (with 95% confidence intervals) among nursing staff (registered nurses, licensed practical nurses, medical assistants, nurse practitioners, nurses’ assistants, physician assistants, and clinical nurse specialists).
Fig 1. Percutaneous exposures (with 95% confidence intervals) in medical and dental students.
Table 1. Nursing staff’s* percutaneous BBFEs per year compared with baseline year of tracking Year
BBFEs
FTEs
x2 (df 1)
38 61 41 50 25
415 830 841 834 952
Baseline year 1.000 7.931 3.760 26.572
1997-1998 1998-1999 1999-2000 2000-2001 2001-2002
P
.318 .005 .052 .000
BBFE, Blood and body fluid exposures; FTE, full-time-equivalent worker. *Nursing staff includes registered nurses, licensed practical nurses, nurses’ assistants, nurse practitioners, physician assistants, clinical nurse specialists, and medical assistants.
Table 2. Medical and dental students’ percutaneous BBFEs per year compared with first year of tracking Year
BBFEs
FTEs
x2(df 1)
P
21 39 32 38 12
498 505 494 480 468
Baseline year 4.874 2.079 5.267 1.528
.027 .149 .022 .001
1997-1998 1998-1999 1999-2000 2000-2001 2001-2002
BBFE, Blood and body fluid exposures; FTE, full-time-equivalent workers.
a peak 13.5% (95% CI; range, 11%-17%) rate of percutaneous injuries in 1999 to 2000, but then achieved lower rates in the next 2 years, with 7.6% (95% CI; range, 5%-10%) in 2000 to 2001 and 9.8% (95% CI; range, 7%-12%) in 2001 to 2002.
DISCUSSION The decreases in annual incidence rates reported here are comparable to those in the current literature on HCWs.8 The Canadian Needle Stick Surveillance
Network9 reports that resident physicians had the highest incidence rate of BBFE events. The Canadian Network reported that from April 2000 through March 2001, the annual exposure rate per 100 FTEs was 4.88% for registered nurses, 6.61% for medical students, and 20.97% for resident physicians. The rate of exposures considering all exposures per 100 FTEs was 4.24%.9 In a study of HCWs in Italy, the highest rates of occupational BBFE occurred in nurses (7.8%) and physicians (1.9%) working in infectious disease units.10 Reddy and Emery11 reported a progressive improvement in the rates of BBFE before and after interventions. The overall incidence rate fell from 10.6% per 100 FTEs to 4.2% over a 5-year period. The same paper discusses some of the difficulties in comparing occupations and departments due to insufficient information about the individuals at risk who would constitute the denominator of the rate calculations. The decision to analyze reporting trends in the 3 high-risk occupational groups was in part driven by the fact that we have accurate denominators on those occupations or trainees. With guidance from analyzing NaSH data, the EHS has targeted interventions at selected groups within the UCHC. It is not surprising that the 5-year data for all 3 groups did not consistently trend downward in rate of injury reports. The rates probably reflect our success in encouraging reporting of incidents as well as the effectiveness of our interventions in reducing injuries. The trend in the rates of BBFE per 100 FTEs for students increased in the initial year, then generally decreased as the intervention program matured. The downward trend is stronger for nursing staff and may reflect the benefits of the well-established outreach program for
Trape´-Cardoso and Schenck
this group at the UCHC. As more experience with the student group accrues, it will be interesting to see if the downward trend in reports established over the last year strengthens. The variability in the incidence rate over time for residents raises concerns about unreliable reporting to the EHS from residents working at remote hospitals. It is possible that not all injuries have been reported, perhaps because residents may frequently be too busy to report and may find it difficult to come directly to our EHS from the hospital where they are assigned. Other confounding elements in analyzing the data include the variability in the UCHC workforce level over the 5 years. In particular, the addition of HCWs from the Department of Corrections in 1998 may have influenced the results. Significant layoffs occurred at the UCHC in 1999 and 2002, and that too may have an impact on the incidence and reporting of injuries. Changes in individuals’ workload over the 5 years also may have contributed to the results obtained. Other barriers to the success of our monitoring program are the variability of BBFE policy in the various institutions in which our personnel work, the diversity of safety devices available, and inconsistency of reporting procedures at the various hospitals in the area where residents and students train. The UCHC administration showed high levels of interest in participating and permitting their staff to attend training and educational activities planned. Moreover, there was an interest in cooperating with EHS staff after a percutaneous injury. Because the workers were more aware of the need for timely risk assessment and antiviral therapy, timely referrals for care increased. Complaints about insensitive supervisors and managers decreased to zero.
CONCLUSIONS To reduce percutaneous BBFE at the UCHC, we utilized data from the NaSH to demonstrate to the administration the importance of purchasing safety devices such as shielded hypodermic needles and needles attached to syringes. We also had to invest in educational and training activities targeted at the highrisk occupational groups identified through the NaSH database. After administrative interventions, engineering controls, and educational modules, there was
August 2004
305
a significant decrease in percutaneous injuries to medical and dental students and to nursing staff over the 5-year period. There was also a mild temporary decrease in these injuries among residents. Strict surveillance of BBFE at the UCHC is crucial to detect areas requiring interventions to reduce the rate of injuries among workers and trainees. The authors thank Dr Richard A. Garibaldi for guidance and support to the Blood and Body Fluid Exposure Control Program and for reviewing this paper, Dr Kenneth H. Dangman for his thoughtful review of the manuscript, and Andrew Warren, MPH, for data collection and initial data analysis.
References 1. Bell DM. Occupational risk of human immunodeficiency virus infection in healthcare workers: an overview. Am J Med 1997;102:9-15. 2. Centers for Disease Control and Prevention. Updated U.S. Public Health Service Guidelines for management of occupational exposure to HBV, HCV, and HIV and recommendations for post-exposure prophylaxis. Atlanta (GA): Centers for Disease Control and Prevention Guidelines; 2001:50 (RR-11). 3. Ippolito G, Puro V, Heptonstall J, Jagger J, Carli GD, Petrosillo N. Occupational human immunodeficiency virus infection in health care workers: Worldwide cases through September 1997. Clinical Infect Dis 1999;28:365-83. 4. Ridzon R, Gallagher K, Ciesielski C, Ginsberg MB, Robertson BJ, Luo CC, et al. Simultaneous transmission of human immunodeficiency virus and hepatitis C virus from a needlestick injury. N Engl J Med 1997;336:919-22. 5. Ippolito G, Puro V, Petrosillo N, De Carli G, Micheloni G, Magliano E, et al. Simultaneous infection with HIV and hepatitis C following occupational conjunctival blood exposure [letter]. JAMA 1998;280:28. 6. Centers for Disease Control and Prevention National Surveillance System for Hospital Health Care Workers. Available from: http: //www.cdc.gov/ncidod/hip/SURVEILL/nash.htm. Accessed July 2003. 7. Fleiss JL. Statistical methods for rates and proportions. 2nd edition. New York: John Wiley & Sons; 1981. 8. Wang SA, Panlilio AL, Doi PA, White AD, Stek M, Saah A, et al. Experience of healthcare workers taking postexposure prophylaxis after occupational HIV exposures: findings of the HIV postexposure prophylaxis registry. Infect Control Hosp Epidemiol 2000;21:780-5. 9. Health Canada. Update: surveillance of healthcare workers exposed to blood/body fluids and bloodborne pathogens, 1 April, 2000 to 31 March, 2001. Canada Communicable Disease Report (CCDR) December 2001;27. 10. Puro V, Carli GD, Petrosillo N, Ippolito G. Risk of exposure to bloodborne infection for Italian healthcare workers, by job category and work area. Infec Control Hosp Epidemiol 2001;22:206-10. 11. Reddy SG, Emery RJ. Assessing the effect of long-term availability of engineering controls on needlestick injuries among health care workers: 3 year preimplementation and postimplementation comparison. Am J Infect Control 2001;29:425-7.