Journal of Hospital Infection (2007) 66, 346e351

www.elsevierhealth.com/journals/jhin

Surgical nurses and compliance with personal

protective equipment
M. Ganczak*, Z. Szych

Department of Hygiene, Epidemiology and Public Health, Pomeranian Medical University,

Szczecin, Poland

Received 7 July 2006; accepted 4 May 2007

Available online 27 July 2007

KEYWORDS Summary The study objectives were to evaluate self-reported compli-

Nurses; Personal ance with personal protective equipment (PPE) use among surgical nurses
protective equipment and factors associated with both compliance and non-compliance. A total
of 601 surgical nurses, from 18 randomly selected hospitals (seven urban
and 11 rural) in the Pomeranian region of Poland, were surveyed using a
confidential questionnaire. The survey indicated that compliance with
PPE varied considerably. Compliance was high for glove use (83%), but much
lower for protective eyewear (9%). Only 5% of respondents routinely used
gloves, masks, protective eyewear and gowns when in contact with
potentially infective material. Adherence to PPE use was highest in the
municipal hospitals and in the operating rooms. Nurses who had a high or
moderate level of fear of acquiring human immunodeficiency virus (HIV)
at work were more likely (P < 0.005 and P < 0.04, respectively) than staff
with no fear to be compliant. Significantly higher compliance was found
among nurses with previous training in infection control or experience of
caring for an HIV patient; the combined effect of training and experience
exceeded that for either alone. The most commonly stated reasons for
non-compliance were non-availability of PPE (37%), the conviction that the
source patient was not infected (33%) and staff concern that following locally
recommended practices actually interfered with providing good patient care
(32%). We recommend wider implementation, evaluation and improvement
of training in infection control, preferably combined with practical experi-
ence with HIV patients and easier access and improved comfort of PPE.
ª 2007 The Hospital Infection Society. Published by Elsevier Ltd. All rights
reserved.

* Corresponding author. Address: Department of Hygiene, Epidemiology and Public Health, Pomeranian Medical University,
_
Zo1nierska 48, 70-250 Szczecin, Poland. Tel.: þ48 91 4871392; fax: þ48 91 4800952.
E-mail address: mganczak@sci.pam.szczecin.pl

0195-6701/$ - see front matter ª 2007 The Hospital Infection Society. Published by Elsevier Ltd. All rights reserved.
doi:10.1016/j.jhin.2007.05.007
Use of protective barriers by nurses 347

Introduction Department. At each selected hospital, all wards

representing surgical subspecialities participated
Healthcare workers (HCWs) still fail to adhere to in the study. The head nurses at each hospital
standard precaution guidelines despite evidence were contacted to discuss the importance of the
that such a failure increases the risk of mucocuta- study and the study protocol.
neous blood and body fluid exposure resulting in An anonymous 32-item questionnaire was pre-
blood-borne infection (BBI).1e7 The major infec- pared by the authors using the guidelines from
tious occupational hazards in the healthcare indus- a previous American study.12 With a few excep-
try are hepatitis B and C viruses (HBV, HCV), and tions, all questions had fixed answer categories.
human immunodeficiency virus (HIV).8 Data pro- The questionnaire was pilot-tested and dealt
vided by the Central Register of Occupational Dis- with the following topics: demographic data; hep-
eases in Poland indicates that among 314 new atitis B vaccination status; fear of acquiring HIV in-
cases of occupational diseases in HCWs in 2005, fection at work; exposures to patients’ blood or
HBV and HCV represented 42.6% of all cases.9 De- body fluids during the previous year; compliance
spite the substantial reduction in HBV infection with the use of PPE, i.e. gloves, masks, protective
since vaccination was introduced in 1989, the inci- eyewear (spectacles included), and gowns when in
dence of HCV hepatitis in Poland is still on the contact with potentially infective material; poss-
increase in this occupational group.9,10 ible answers were ‘always’, ‘sometimes’, ‘never’;
Exposures to specific health hazards are likely to reasons for not complying with the barrier
affect certain high-risk groups of HCWs. For exam- precautions.
ple, the care of surgical patients with the in- Data collected from this questionnaire on the
creased chance of contact with blood means that incidence and risk factors of blood exposures have
surgical staff are more at risk of occupationally been described previously.13 This study examines
acquired infections.11 For this reason a better un- the data relating to the compliance with PPE
derstanding of surgical nurses’ adherence with use. All categoric data were analysed using the
PPE usage is important as it provides an assess- Chi-squared test with or without Yates’ correction,
ment of the efficacy of existing preventative strat- whilst the KruskaleWallis test or U-ManneWhitney
egies. This could then help to identify preventive test was used for numeric variables. Statistical
factors which are likely to improve the compliance significance for all analyses was presumed for
and reduce the risk of BBI. Finally, it is then poss- P  0.05.
ible to incorporate these preventative approaches For the purposes of comparing nurses with and
into institutional health strategies. without training in infection control and occupa-
There are many published studies which have tional experience with HIV-infected patients, we
focused on the use of PPE among HCWs as part of grouped all nurses who had had one or more
existing health preventive measures.3e6 However, courses together as having experienced the ‘in-
there have been no similar studies on the use of tervention’, and all nurses with experience of one
protective equipment by surgical nurses. This spe- or more known HIV patients as positive for that
cific study therefore intends to evaluate compli- ‘intervention’. Since occupational experiences had
ance with PPE use among surgical nurses in not been organized in a systematic manner for
selected Polish hospitals and to identify factors educational purposes, we placed the term inter-
associated with both compliance and non- vention in parentheses.
compliance. The results of this study are likely to
be of value in modifying existing barrier pro-
grammes in order to enhance the safety of staff Results
working on surgical wards.
None of the nurses present in the ward on the day
when the questionnaire was administered refused
Methods to participate.
The sample population surveyed consisted of
The study was conducted from January to March 601 nurses who were mostly young (aged 20e58
2003 in 18 Polish hospitals: seven of these were years, median 38), female (99.3%), and working
urban hospitals located in the city of Szczecin of full-time (565, 94%). Almost half of the respon-
which two were academic and five municipal. The dents (257, 42.8%) practised in municipal hospi-
remaining 11 hospitals were located in rural areas. tals, and more than one-third (229, 38.1%) in rural
All hospitals were selected randomly from a list hospitals. More than one-quarter of the respon-
obtained from the West Pomeranian County Health dents (162; 27%) had occupational contacts with
348 M. Ganczak, Z. Szych

HIV patients during their professional carrier; Compliance was significantly influenced by
almost three-quarters (450; 74.9%) had partici- hospital location. It was lowest for nurses working
pated in a training course on infection control. As in academic hospitals (mean: 10.0  4.4), and
to hepatitis B vaccination status, 85.2% had been highest for nurses from municipal hospitals (mean:
immunized, 5.3% had been previously infected 12.1  4.7) (P < 0.0001).
with HBV, and 9.5% had never been either infected The survey showed that there was a significant
or immunized. association between the fear of acquiring HIV at
A high degree of fear of acquiring HIV at work work and PPE compliance. Nurses who had high
was reported by 63.9% of respondents (N ¼ 378/ or moderate levels of fear were more likely
592; 95% CI: 60e68%), moderate fear by 31.8% (P < 0.005 and P < 0.04, respectively) than staff
(N ¼ 188; 95% CI: 28e36%), no fear by 4.4% with no fear to be compliant (mean scores for
(N ¼ 26; 95% CI: 2e6%). the groups: 12.0  4.9; 11.1  4.3; 9.3  4.8, re-
Almost half of the respondents (276/601, 45.9%; spectively). Compliance was also related to the
95% CI: 42e49%) reported having an occupational past experience with HIV patients. It was signifi-
puncture injury during the past year and 134 (22.3%; cantly higher (P < 0.0001) among experienced
95% CI: 19.2e25.8) had sustained contact via their nurses (mean: 12.9  4.5) than non-experienced
mucous membranes. Over half the nurses (N ¼ 323, (mean: 11.1  4.8). Compliance was also signifi-
53.7%; 95% CI: 49.7e57.7) had worked at least once cantly higher among the trained nurses (mean:
with a recent abrasion or cut on their hands. 12.0  4.6) than non-trained (mean: 10.4  5.1)
In general, compliance with PPE varied consid- (P < 0.009). Compliance was found to be even
erably, and was highest for glove use (83%), and higher among nurses with both training and
lowest for protective eyewear use (9%) (Table I). experience using the KruskaleWallis test in
Compliance with all items (i.e. gloves, gowns, comparison with nurses who had neither, either
masks and protective eyewear) was 29/601 (4.8%; or both training and experience (P < 0.0001).
95% CI: 3.4e6.8). Comparison of age between The mean results included: no intervention
non-compliant and compliant nurses did not show 10.3  5.1; experience only, 11.0  5.1; training
significant differences for PPE use (P > 0.63). only 11.4  4.6; both training and experience
To examine nurses’ overall compliance to PPE 13.4  4.3 (Figure 1).
precautions, we calculated the overall score for Infection control training and experience
each respondent, giving 5 points for using each with HIV patients had no impact on the degree
single PPE ‘always’, 3 points for using it ‘some- of fear among nurses (P > 0.86 and P > 0.26,
times’, and 0 for ‘never’. Thus, the maximum respectively).
score for PPE compliance was 20, and the mini- Among nurses with perfect compliance for glove
mum was 0. Using this scoring scheme the mean for use, 217/501 (43.3%) experienced a sharps injury
the whole group of respondents was 11.6  4.8. during the past year which was significantly
There was no significant effect of age on the (P < 0.0007) less than the group with poor compli-
PPE compliance score (P > 0.32). However, a com- ance (62/100, 62%). Among the nurses who had
parison of surgical subspecialities showed signifi- used masks or protective eyewear regularly, the per-
cant differences for PPE compliance, with the centage of those who sustained splash contact via
highest compliance in the operating room (mean: their mucous membranes was 50% (20/40), which
15.4  3.5), and the lowest in the admitting area was not significantly (P > 0.7) higher than for the
(mean: 8.4  4.1) (P < 0.0001). group with poor compliance (179/386, 46.4%).
Only 396 of the respondents (65.9%) answered
the question on the reasons for non-compliance.
More than one reason per respondent was allowed
Table I Adherence to personal protective equip- and the most commonly stated reasons were non-
ment use among surgical nurses (N ¼ 601) availability of PPE (37%), the conviction that the
Personal No. of respondents (%)
source patient was not infected (33%), lack of time
protective (19.2%), staff concern that following locally rec-
Always Often Never ommended practices actually interfered with pro-
equipment
viding good patient care (32%) and a perception
Gloves 501 (83.4) 64 (10.6) 36 (6.0)
Gowns 242 (40.3) 222 (36.9) 137 (22.8)
that the equipment provided was ineffective, e.g
Protective 53 (8.8) 137 (22.8) 411 (68.4) poor-fitting gloves (9.8%). More nurses from the
eyewear academic and municipal hospitals believed that
Masks 202 (33.6) 245 (40.8) 154 (25.6) the source patient did not pose a risk compared to
the nurses from the rural hospitals (30.8% of 91,
Use of protective barriers by nurses 349

15 linked to a lower probability of having a sharps in-

14 13.4 jury, which could lead to HBV, HCV or HIV serocon-
13
Compliance (the overall score)

version. Finally, apart from protecting the staff

12 11 11.4
11 10.3 from contagious diseases, gloves also protect
10 patients from micro-organisms during surgical pro-
9 cedures and the insertion of invasive devices as
8 well as from caregivers who may be carriers of
7
6 blood-borne viruses.
5 Gowns should be worn during procedures which
4 are likely to expose HCWs to spraying or splashing
3 with blood, body fluids, secretions and excre-
2
1
tions.15 However, only about one-third of our re-
0 spondents reported using them regularly. It is
Type of intervention also known that nurses can sustain blood contacts
(P < 0.0001)*
via mucous membranes, and one in five of our re-
‘Intervention’ spondents reported such contacts in the previous
Neither Training year. It has been reported in the literature that
Experience Training and experience some cases of exposure to HBV, HCV or HIV via mu-
cous membranes, leading to HCW seroconversion,
Figure 1 Compliance with personal protective equip- have occurred in this way.1,16e18 The regular use
ment among surgical nurses by training in infection con- of protective eyewear was reported by 9% of
trol and experience with HIV-infected patients, West nurses, and this probably reflected the use of cor-
Pomerania, Poland, 2003 (N ¼ 601). ) Kruskal-Wallis rective eyewear by older nurses rather than a de-
test for 4 groups.
liberate attempt to prevent exposure to blood.
As the facial protection used by nurses surveyed
and 25.8% of 236 vs 17.2% of 174; P < 0.02 and by us was not significantly linked to minimizing
P < 0.04, respectively). blood splashes, misunderstandings such as the be-
lief that spectacles can replace protective eye-
wear demonstrates a need for better education.
Discussion It would appear that the main reasons for non-
compliance with PPE use were similar to those
The response rate to the survey was excellent, reported by others2e7 and those identified in an
suggesting that this was an area of importance for earlier survey with surgeons.19 It is interesting to
surgical nurses. note that the main reason respondents in this study
It is recommended that HCWs should wear PPE gave for not using PPE was lack of availability. This
for any contact with blood or body fluids.1e2 How- problem would be relatively easy to address with
ever, this survey has identified that, despite the better institutional support. In Poland, recently
risk associated with blood contact, surgical nurses’ enacted regulation on occupational safety, up-
compliance with PPE was poor and placed them at dated in 2001, obliges employers to provide ade-
risk of contracting occupationally acquired BBI. quate protection for employees against harmful
In several published studies, including ours, substances including micro-organisms.20 Unfortu-
gloves have been shown to be the most frequently nately, a review of randomly selected state hospi-
used protective equipment, possibly reflecting the tals carried out by the Polish Ministry of Health and
long tradition of wearing them. Moreover, glove Occupational Safety Organization revealed that
use has been shown to be the largest contributor hospital directors were often failing to provide
to the efficacy of standard precautions.14 How- proper protective barriers to the staff, usually
ever, despite the fact that more than half of the due to financial constraints.21 Such a situation is
study respondents admitted that they had worked unacceptable. Whilst better protection would in-
while having a recent abrasion or cut on their volve increased costs for an adequate equipment
hands, and although almost half reported puncture supply, it is likely that this would be much less
injury in the last 12 months, regular glove use had than all the costs associated with treating HCWs
been neglected by 17% of them. This is even riskier who acquire BBI at the workplace.22
because glove wearing not only prevents the non- As reported by others, the respondents often
parenteral exposure to blood, but also can reduce blamed their non-compliance on a lack of percep-
the amount of virus inoculated.7,15 This study tion by the patient representing a health risk,
showed that glove wearing was also significantly despite the fact that patients with BBI can be
350 M. Ganczak, Z. Szych

asymptomatic or unaware that they are in- strategies is required, including continuous edu-
fected.4,5 Therefore, HCWs should regard the cation in infection control, easy accessibility to
blood of all patients as potentially infectious and PPE, and improvement in the comfort and con-
protect themselves routinely when exposure to venience of barrier precautions.
blood is expected.1,23
Another reason respondents gave for not
wearing PPE related to poor dexterity, which is Acknowledgements
an issue that has often been raised in other
surveys.2,6 This suggests that protective equip- Thanks to M. Milona, RN MSc, for help in perform-
ment used in Polish healthcare facilities needs ing the survey.
to be better designed and more comfortable.
Further studies to identify improvements to PPE
References
would be valuable, followed by an evaluation
of changes in compliance. 1. Updated US Public Health Service Guidelines for the Man-
In this study, compliance with PPE use was agement of Occupational Exposures to HBV, HCV, HIV and
significantly related to the fear of acquiring HIV Recommendations for PEP. MMWR Morb Mortal Wkly Rep
(and probably other BBIs) at work, with a dosee 2001;50(RR-11):1e67.
response effect evident. This is similar to previous 2. Bennett G, Mansell I. Universal precautions: a survey of
community nurses’ experience and practice. J Clin Nurs
studies among surgical staff in which acquiring HIV 2004;13:413e421.
from patients was ranked as a strong influence on 3. Ferguson KJ, Waitzkin H, Beekmann SE, Doebbeling BN.
compliance with PPE use.4 A fear of BBI tends also Critical incidents of nonadherence with standard precau-
to be the prime motivator for hospital personnel to tions guidelines among community hospital-based HCWs.
change their behaviour.22,23 J Gen Intern Med 2004;19:726e731.
4. Hoffman-Terry M, Rhodes 3rd LV, Reed 3rd JF. Impact on HIV
As both training programmes and practical on medical and surgical residents. Arch Intern Med 1992;
experience in working with HIV-positive patients 152:1788e1796.
positively influenced nurses’ compliance to PPE 5. Michalsen A, Delclos GL, Felknor SA, et al. Compliance with
use, and the combined effect was greater than universal precautions among physicians. J Occup Environ
either alone, we recommend that all surgical Med 1997;39:130e137.
6. Nelsing S, Nielsen TL, Nielsen JO. Noncompliance with uni-
nurses should receive effective training in infec- versal precautions and the associated risk of mucocutane-
tion control methods, preferably with known HIV ous blood exposure among Danish physicians. Infect
patients. Control Hosp Epidemiol 1997;18:692e698.
The study has a number of limitations. First, 7. Rabaud C, Zanea A, Mur JM, et al. Occupational exposure to
‘compliance’ is difficult to quantify reliably. Com- blood: search for a relation between personality and behav-
iour. Infect Control Hosp Epidemiol 2000;21:564e574.
parisons of observed and self-reported adherence 8. Hersey JC, Martin L. Use of infection control guidelines by
to barrier precautions among HCWs found signifi- workers in healthcare facilities to prevent occupational
cant differences in the respective rates for more transmission of HBV and HIV: results from a national survey.
protective barriers.8 Second, because the study Infect Control Hosp Epidemiol 1994;15:243e252.
sampled only surgical nurses, the results may not 9. Wilczyńska U, Szeszenia-Da ˛browska N, Szymczak W. Occu-
pational diseases in Poland, 2005. Med Pr 2006;57:225e234.
be generalizable to all hospital-based nurses. Third, 10. Wac1awik J, Ga ˛siorowski J, Inglot M. Epidemiology of occu-
recall bias is possible. Finally, a cross-sectional pationally acquired infectious diseases among health care
study design can be used only to show associations workers in Wroclaw region. Med Pr 2003;54:535e541.
and cannot confirm a cause-and-effect relation- 11. O’Neal TM, Abbott AV, Radecki SE. Risk of needlesticks and
ship.5 The 100% response rate to the questionnaire occupational exposures among residents and medical stu-
dents. Arch Intern Med 1992;152:1451e1456.
and the variety of hospitals selected randomly 12. Lowenfels AB, Wormster GD, Jain R. Frequency of puncture
means that the results of this study are likely to injuries in surgeons and estimated risk of HIV infection.
reflect accurately the situation regarding the use Arch Surg 1989;124:1284e1286.
of PPEs in the surgical wards in Polish hospitals. 13. Ganczak M, Milona M, Szych Z. Nurses and occupational
Our findings indicate that despite common exposures to blood-borne viruses. Infect Control Hosp
Epidemiol 2006;27:175e180.
contacts with blood, compliance with PPE use 14. Wong ES, Stotka JL, Chinchilli VM, Williams DS, Stuart CG,
among Polish surgical nurses is unacceptably low, Markowitz SM. Are universal precautions effective in reduc-
which shows that the existing strategies to ing the number of occupational exposures among HCWs?
control infections to HCWs have not been ade- J Am Med Assoc 1991;265:1123e1128.
quate. There is a need to consider factors that 15. Damani NN. Manual of Infection Control Procedures.
2nd edn. London/San Francisco: GMM 2003; 191e206.
enable people to change their behaviour, and 16. Do AN, Ciesielski CA, Metler RP, Hammett TA, Li J,
also the availability, cost and convenience of the Fleming PL. Occupationally acquired HIV infection: national
preventive barriers.24 Thus, a combination of case surveillance data during 20 years of the HIV epidemic
Use of protective barriers by nurses 351

in the United States. Infect Control Hosp Epidemiol 2003; 20. Infectious diseases e Act from 24th September 2001,
24:86e96. Dziennik Ustaw. 01.126. 3184.
17. Ippolito G, Puro V, Heptonstall J, Jagger J, De Carli G, 21. G1adysz A, Inglot M, Knysz B. Hepatitis as a nosocomial
Petrosillo N. Occupational HIV infection in HCWs: worldwide infection. Mag Med 2001;15:24e29.
cases through September 1997. Clin Infect Dis 1999;28:365e 22. Fahey BJ, Koziol DE, Banks SM, Henderson DK. Frequency of
383. nonparental occupational exposure to blood and body fluids
18. Ippolito G, Puro V, Petrosillo G, De Carli G, Micheloni G, before and after universal precautions training. Am J Med
Magliano E. Simultaneous infection with HIV and hepatitis 1991;90:145e153.
C virus following occupational conjunctival blood exposure. 23. Goldmann DA. Blood-borne pathogens and nosocomial infec-
J Am Med Assoc 1998;280:28. tions. J Allergy Clin Immunol 2002;110(Suppl. 2):S21eS26.
19. Ganczak M, Boroń-Kaczmarska A, Bialecki P, Szych Z. Use of 24. Wright JG, Young NL, Stephens D. Reported use of strategies
preventive strategies by surgeons and the risk of HIV expo- by surgeons to prevent transmission of blood-borne dis-
sure. Wiad Lek 2004;57:221e228. eases. Can Med Assoc J 1995;152:1089e1095.