• Alcoholic gel;
  • educational programme;
  • hand-washing;
  • nosocomial infection;
  • risk factor


  1. Top of page
  2. Abstract
  3. Introduction
  4. Patients and Methods
  5. Results
  6. Discussion
  7. Transparency Declaration
  8. References

Clin Microbiol Infect 2011; 17: 894–900


The objectives of this study were to determine the risk factors associated with nosocomial infection (NI) in the intensive-care unit and to assess the influence of a hand-washing promotion programme on the NI rate. Over a 6-month study period (P1), a prospective study of NI cases was performed, and risk factors for NI were analysed. Data were compared with those corresponding to a second period (P2), during which a health workers’ hand-washing promotion programme was carried out; alcohol-based solution was also placed at the patients’ headboard. Eight hundred and six patients were included (395 patients in P1 and 411 in P2). The mean APACHE II score was 11.41; there were no statistically significant differences in epidemiological or clinical variables between P1 and P2, and there were no differences in risk factors for NI. The rate of infection in P1 was 26%, and that in P2 was 16% (p <0.05). The hand-washing rate was higher in P2 than in P1 (before patient care, 45% and 35%, respectively; after contact with the patient, 63% and 51%, respectively). In the multivariate analysis, only central venous catheterization (>5 days) and tracheostomy were statistically significant risk factors for NI; having been included in the study during P1 or P2 was not statistically associated with NI. In conclusion, there was a significant increase in hand-washing frequency in P2; the incidence of NI during P1 was significantly higher than during P2, but having been included in the study in P1 was not significantly associated with a higher rate of NI; only central venous catheterization (>5 days) and tracheostomy were significantly associated with NI.


  1. Top of page
  2. Abstract
  3. Introduction
  4. Patients and Methods
  5. Results
  6. Discussion
  7. Transparency Declaration
  8. References

Nosocomial infections (NIs) increase patient morbidity and mortality, hospital stay and economic cost, and are correlated with antibiotic resistance. According to a prevalence study carried out by the WHO in 55 hospitals of 14 countries, 8.7% of patients admitted to a hospital suffered an NI; this means that approximately 1.4 million people will develop a health-related infection [1]; for these patients, the mean hospital stay will be 14 days longer [2,3]. NIs are especially relevant in certain hospital areas (intensive-care units (ICUs)), where their incidence is two-fold to five-fold higher [4,5]. Identification of the factors related to NI is crucial for the development of preventive strategies [1,2,6], which require an integrated and continuous education programme to improve the care provided to patients by healthcare workers [6]; adequate hand hygiene is probably the most cost-effective measure [1,7–11].

Although there is enough evidence to show a relationship between improvements in hand hygiene practices and a decrease in NI incidence [6–13], there is a low level of adequate compliance with these recommendations in the routine of daily practice, rarely exceeding 40–50% [10,14–18].

Given the relevance of NI control in the ICU and the benefit attributed to the use of barrier measures (especially hand hygiene of healthcare staff) in the prevention of such infections, we performed this study with the following aims: (i) to determine the frequency and epidemiology of NIs in the ICU at a second-level hospital; (ii) to determine the main risk factors associated with them; (iii) to analyse the prognostic factors associated with mortality in patients admitted to this ICU; and (iv) to evaluate the influence of a hand-washing programme on compliance with hand-hygiene practices and on the development of NI.

Patients and Methods

  1. Top of page
  2. Abstract
  3. Introduction
  4. Patients and Methods
  5. Results
  6. Discussion
  7. Transparency Declaration
  8. References

Hospital characteristics

The study was performed in the 16-bed ICU of Santa María del Rosell Hospital in Cartagena (Murcia), a second-level healthcare centre.

Study periods

There were two different 6-month periods (P1 and P2), during which an observational prospective study was carried out by collecting epidemiological and clinical data of all patients admitted to the ICU. Throughout the second period (P2), a health workers’ hand-washing promotion programme was carried out, and alcohol-based solution was also placed at the patients’ headboards.


The study of the patients was carried out according to a previously established data collection protocol, which included epidemiological and clinical characteristics; in cases of NI, variables related to the infection were also analysed. All patients admitted to the ICU for more than 24 h in either of the two periods of study were included.

Study of the infections

The diagnosis of the different types of NI was carried out with the diagnostic criteria suggested by the CDC [1] and adapted for the ICU by the Infectious Diseases Work Group of the Spanish Society of Intensive Medicine and Coronary Units (GTEI-SEMICYUC) [19].

Educational programme for hand-washing promotion

In order to assess compliance with the use of barrier measures in healthcare, all members of the ICU staff who cared for the patients were considered as potential subjects of study during two observational surveys carried out in P2. In the first 2 weeks and the last 2 weeks of the second period of study, healthcare provided by health workers in the ICU was assessed during 50 h of observation, distributed in sessions of 3 h in the morning shifts and 2 h in the evening shifts. All of the procedures related to patient care that required contact with the patient or with his or her environment were considered to require hand-washing. During P2, there was an attempt to improve health workers’ hand-washing by means of an educational programme, which included different measures, such as distribution of posters and educational leaflets, informative talks, and acquisition of a hydroalcoholic antiseptic gel and dispensers, which were distributed close to each of the ICU rooms as well as in other parts of the ward.

Data processing and statistical analysis

Data were analysed with SPSS 15.0 (SPSS Software, Chicago, IL, USA). A descriptive study was performed for the clinical and epidemiological characteristics of all included patients. The relationships or associations between pairs of qualitative variables were determined through analyses of contingency tables by means of Pearson’s chi-squared test, complemented by an analysis of residues, with the aim of determining the directional dependence. In the case of quantitative variables, means were compared with Student’s t-test. The difference was considered to be significant at p <0.05. Bivariate and multivariate analyses were performed in order to detect those factors related to the development of NI and to mortality. The multivariate analysis comprised a non-conditioned logistic regression, in which the development of the NI or mortality was taken as the dependent variable, and all those variables statistically associated with NI or mortality in the bivariate analysis were included as independent variables.


  1. Top of page
  2. Abstract
  3. Introduction
  4. Patients and Methods
  5. Results
  6. Discussion
  7. Transparency Declaration
  8. References

Epidemiological and clinical characteristics of the study population

A total population of 806 patients was included (P1, group A, 395 patients; and P2, group B, 411 patients), whose characteristics are shown in Table 1. The comparison between the patients of P1 and P2 showed that there were no statistically significant differences, except for the mean age (67 vs. 65 years), high blood pressure prevalence (58.22% vs. 49.9%), and chronic liver dysfunction (0.3% vs. 2.9%).

Table 1.   Patients in group A and group B
VariablesGroup A (n = 395)Group B (n = 411)p
  1. COPD, chronic obstructive pulmonary disease; HIV, human immunodeficiency virus; ICU, intensive-care unit; PDA, parenteral drug addiction.

  2. aStable/unstable patients, patients with CDC code 1–3/patients with CDC code 4–5.

  3. In bold: p < 0.05

Age (years)67.65 ± 13.90 (14–92)65.19 ± 15.65 (13–94)0.018
Age groups (years), no. (%)
 <4017 (4.30)26 (6.33)0.019
 40–5981 (20.51)121 (29.44)
 60–6985 (21.52)72 (17.52)
 70–7466 (16.71)58 (14.11)
 75–7973 (18.48)56 (13.62)
 ≥8073 (18.48)78 (18.98)
Men/women, no. (%)258 (65.32)/137 (34.68)286 (69.59)/125 (30.41)0.196
Underlying diseases, no. (%)
 Diabetes mellitus155 (39.24)136 (33.1)0.069
 Cardiopathy183 (46.33)167 (40.6)0.103
 High blood pressure230 (58.22)205 (49.9)0.017
 Dyslipidaemia148 (37.5)147 (35.8)0.616
 COPD90 (22.8)82 (20)0.326
 Chronic renal failure34 (8.6)34 (8.30.864
 Chronic liver dysfunction1 (0.3)12 (2.9)0.003
 Corticoid treatment6 (1.5)6 (1.5)0.945
 HIV infection1 (0.3)2 (0.5)0.586
 PDA3 (0.7)0.089
Reason for ICU admission, no. (%)
 Acute ischaemic coronary disease252 (63.8)272 (66.2)0.212
 Medical117 (29.6)101 (24.6)
 Surgical19 (4.8)31 (7.5)
 Polytrauma7 (1.8)7 (1.7)
CDC code, no. (%)
 CDC13 (0.76)10 (2.43)0.002
 CDC2185 (46.83)171 (41.60)
 CDC3108 (27.35)138 (33.58)
 CDC482 (20.76)58 (14.11)
 CDC517 (4.30)34 (8.27)
Stable/unstable patients, no. (%)a296 (74.9)/99 (25.1)319 (77.9)/92 (22.4)0.371
APACHE II11.72 ± 8.29 (0–47)11.10 ± 8.21 (0–49)0.287
APACHE II distribution, no. (%)
 0–592 (23.29)99 (24.09)0.734
 6–10128 (32.41)144 (35.04)
 11–1573 (18.48)77 (18.73)
  16–2042 (10.63)40 (9.73)
 21–2525 (6.33)23 (5.60)
 26–3022 (5.57)13 (3.16)
 >3013 (3.29)15 (3.65)
Days of stay before ICU1.63 ± 4.14 (0–31)1.27 ± 3.68 (0–38)0.19
Days of stay in ICU4.97 ± 7.73 (1–95)4.27 ± 5.58 (1–48)0.138

Exposure of the patients to the main invasive procedures considered to be risk factors for nosocomial infection was also analysed (Table 2). A statistically significant difference was found only for patient exposure to arterial catheterization (2.53% in P1 vs. 10.71% in P2); the use ratio of arterial catheters was also lower in group A than in group B (0.03 vs. 0.16). The distribution of the 166 episodes of NI in the ICU and the values for the accumulated incidence of each type of NI are given in Table 2.

Table 2.   Distribution of nosocomial infections in the intensive-care unit
Type of nosocomial infectionNo. of episodes of infection N (%)Accumulated incidence
Ventilator-associated pneumonia35 (21.1)4.3
Infection of the lower respiratory tract19 (11.5)2.4
Urinary infection associated with urinary catheter33 (19.9)4.1
Infection related to central venous catheter35 (21.1)4.3
Catheter-related bacteraemia6 (3.6)0.7
Primary bacteraemia13 (7.8)1.6
Secondary bacteraemia9 (5.4)1.1
Surgical wound infection16 (9.6)2
Total166 (100)20.6

Risk factors for the development of NI in the ICU

A bivariate study was carried out in order to analyse the characteristics and exposure to the different risk factors related to the developed of NI (Table 3). In the multivariate analysis, only central venous catheterization for more than 5 days (OR 18.815, 95% CI 10.6–33.399) and the presence of tracheostomy (OR 25.345, 95% CI 6.079–105.678) were statistically significant risk factors for the development of NI. The numbers of patients who developed some kind of NI were 13.7% in P1 and 8.3% in P2 (p <0.05), but having been admitted to the ICU during P1 or P2 was not significantly associated with NI.

Table 3.   Bivariate analysis of the characteristics and risk factors for the development of nosocomial infection
Characteristics and risk factorsNosocomial infection in ICU (n = 88)Without nosocomial infection in ICU (n = 718)p
  1. COPD, chronic obstructive pulmonary disease; CVC, central venous catheter; ICU, intensive-care unit.

Age (years)67 ± 13.8966.3 ± 150.616
Men/women, no. (%)64 (72.7)/24 (27.3)480 (66.9)/238 (33.1)0.267
Underlying condition, no. (%)
 Diabetes mellitus29 (33)262 (36.5)0.515
 Cardiopathy41 (46.6)309 (43)0.525
 High blood pressure49 (55.7)386 (53.8)0.733
 Dyslipidaemia33 (37.5)262 (36.5)0.853
 COPD24 (27.3)148 (20.6)0.150
 Chronic liver dysfunction1 (1.1)12 (1.7)0.707
 Chronic renal failure12 (13.6)56 (7.8)0.063
Reason for admission to ICU, no. (%)
 Acute coronary disease28 (31.8)496 (69.1)<0.001
 Medical46 (52.3)172 (24)
 Surgical12 (13.6)(5.3)
 Polytrauma2 (2.3)12 (1.7)
CDC Code, no. (%)
 CDC11 (1.1)12 (1.7)<0.001
 CDC213 (14.8)343 (47.8)
 CDC313 (14.8)233 (32.5)
 CDC435 (39.8)105 (14.6)
 CDC526 (29.5)25 (3.5)
Stable/unstable patients, no. (%)27 (30.7)/61 (69.3)589 (82)/130 (18)<0.001
APACHE II19 ± 9.310.5 ± 7.6<0.001
Days of stay before ICU2.8 ± 5.61.3 ± 3.60.017
Days of stay in ICU15 ± 14.83.3 ± 3.1<0.001
Days of mechanical ventilation9.6 ± 12.70.5 ± 1.9<0.001
Days of urinary catheter use14.3 ± 15.21.8 ± 3.3<0.001
Days of CVC use11.4 ± 15.20.9 ± 3.1<0.001
Days of CVC of peripheral access6.3 ± 11.51 ± 2.8<0.001
Days of arterial catheter use2.4 ± 5.80.2 ± 1<0.001
Days of Swan–Ganz catheter use0.07 ± 0.40.01 ± 0.10.240
Days of nasogastric tube use12.4 ± 15.71 ± 2.6<0.001
Days of enteral nutrition6.3 ± 9.80.4 ± 1.7<0.001
Days of parenteral nutrition6 ± 8.90.4 ± 1.7<0.001
Previous antibiotics (yes/no), no. (%)18 (20.5)/70 (79.5)54 (7.5)/664 (92.5)<0.001
Community infection (yes/no), no. (%)23 (26.1)/65 (73.9)141 (19.6)/577 (80.4)0.153
Nosocomial infection before ICU (yes/no), no. (%)6 (6.8)/82 (93.2)27 (3.8)/691 (96.2)0.172
Period of study (P1/P2), no. (%)53 (60.2)/35 (39.8)342 (47.6)/376 (52.4)0.026


During the two study periods, 121 patients died, giving a global mortality of 15%; 91 (11.3%) patients died during their stay in the ICU, and 30 (3.7%) in a different ward after ICU discharge (Table 4). In the multivariate analysis, variables statistically significant associated with mortality were: older age (>65 years), diabetes mellitus, administration of antibiotic treatment in a hospital ward before the patient’s admission to the ICU, APACHE II score  >15, CDC severity score >3, pulmonary artery catheterization by Swan–Ganz catheter, invasive mechanical ventilation for more than 5 days, and use of a urinary catheter for more than 3 days. The presence of community-acquired infection, the development of nosocomial infection in a hospital ward or during ICU stay, and having been admitted during P1 or P2 were not identified as independent risk factors for mortality (Table 5).

Table 4.   Differential characteristics of the patients who died or survived over the total duration of the study
Risk factorsSurvived (n = 685)Dead (n = 121)p
  1. CVC, central venous catheter; ICU, intensive-care unit; Peripheral central catheter, central venous catheter inserted peripherally.

Age (years)65 ± 15.2 (13–94)70.8 ± 11.7 (29–91)<0.001
Men/women, no. (%)471 (68.8)/214 (31.2)73 (60.3)/48 (39.7)0.068
Reason for admission, no. (%)
 Acute coronary disease486 (70.9)38 (31.4)<0.001
 Medical143 (20.9)75 (62)
 Surgical42 (6.1)8 (6.6)
 Polytrauma14 (2)
APACHE II9.8 ± 7 (0–47)20.6 ± 9 (4–49)<0.001
CDC Code
 CDC113 (1.9)<0.001
 CDC2337 (49.2)19 (15.7)
 CDC3229 (33.4)17 (14)
 CDC491 (13.3)49 (40.5)
 CDC515 (2.2)36 (29.8)
Stable/unstable patients, no. (%)579 (84.5)/106 (15.5)36 (29.8)/85 (70.2)<0.001
Days of stay before ICU1.1 ± 3.1 (0–31)3.4 ± 6.7 (0–38)<0.001
Days of stay in ICU3.8 ± 4.5 (1–48)9.3 ± 12.8 (1–95)<0.001
Period 1/Period 2, no. (%)330 (48.2)/355 (51.8)65 (53.7)/56 (46.3)0.261
Mechanical ventilation, no. (%)92 (13.4)83 (68.6)<0.001
Urinary catheter, no. (%)275 (40.1)113 (93.4)<0.001
Central venous catheter, no. (%)105 (15.3)78 (64.5)<0.001
CVC peripheral insertion, no. (%)132 (19.3)55 (45.5)<0.001
Arterial catheter, no. (%)27 (3.9)27 (22.3)<0.001
Swan–Ganz catheter, no. (%)1 (0.1)6 (5)<0.001
Nasogastric tube, no. (%)139 (20.3)95 (78.5)<0.001
Enteral nutrition, no. (%)70 (10.2)53 (43.8)<0.001
Parenteral nutrition, no. (%)59 (8.6)56 (46.3)<0.001
Previous antibiotics, no. (%)39 (5.7)33 (27.3)<0.001
Infection in ICU, no. (%)54 (7.9)34 (28.1)<0.001
Infection in ward, no. (%)14 (2)19 (15.7)<0.001
Community infection, no. (%)118 (17.2)46 (38)<0.001
Days of exposure   
 Mechanical ventilation0.7 ± 3.1 (0–43)6.3 ± 10.5 (0–67)<0.001
 Urinary catheter2.1 ± 4.7 (0–48)8.9 ± 12.9 (0–95)<0.001
 CVC1.2 ± 4.9 (0–82)7.1 ± 11.6 (0–53)<0.001
 Peripheral central catheter1.2 ± 3.8 (0–53)3.9 ± 8.5 (0–65)0.001
 Arterial catheter0.2 ± 1.6 (0–33)1.7 ± 4.2 (0–19)<0.001
 Swan–Ganz catheter0.002 ± 0.04 (0–1)0.1 ± 0.5 (0–3)0.025
 Nasogastric tube1.2 ± 4.2 (0–48)7.8 ± 13 (0–33)<0.001
 Enteral nutrition0.6 ± 2.7 (0–32)3.7 ± 7.7 (0–40)<0.001
 Parenteral nutrition0.6 ± 2.7 (0–41)3.7 ± 6.6 (0–33)<0.001
Table 5.   Independent risk factors for mortality
VariablesOR (95% CI)
Age >65 years2.904 (1.664–5.065)
Diabetes mellitus1.910 (1.133–3.222)
Antibiotics before intensive-care unit3.274 (1.692–6.336)
CDC Score >34.348 (2.367–7.987)
APACHE II >154.042 (2.316–7.057)
Swan–Ganz catheter23.534 (2.091–264.871)
Mechanical ventilation >5 days7.273 (1.587–33.336)
Urinary catheter >3 days2.143 (1.051–4.369)
Non-urgent surgery0.218 (0.080–0.594)

Compliance with the recommendations on the use of barrier measures in P1 and P2

There were no statistically significant differences in professional categories, severity of disease or type of healthcare activities between the two observational surveys. Hand-washing in P1 (with water and soap or with alcoholic gel) before patient care was performed in only 35% of episodes vs. 45.4% in P2 (p <0.05); hand-washing after patient care was performed in 51.8% of episodes in P1 and in 63.1% in P2 (p <0.05), although alcohol gel was used in only 24.8% of the total of episodes in which some type of hand-washing before contact with the patient was observed (7.6% in cases of hand-washing after contact with the patient); in all other cases, hand-washing was performed with water and detergent.


  1. Top of page
  2. Abstract
  3. Introduction
  4. Patients and Methods
  5. Results
  6. Discussion
  7. Transparency Declaration
  8. References

The main aims of our study were to analyse risk factors for NI in an ICU and the influence of a hygiene promotion programme on infection control. The data concerning the development of NI in our cohort were obtained from the patients admitted to the ICU during the two observation periods; despite the fact that P1 was distributed in different months from those of P2, the analysis of the population characteristics in both periods showed that they were clinically comparable. The statistically significant difference between the average ages of the populations, which was probably of little relevance from the clinical point of view, was not translated into a clinical difference in past or present pathology between the two groups, and only chronic liver dysfunction was more frequent in group B than in group A. Although patient severity of disease as estimated by the APACHE II index was similar, the classification according to the CDC score showed inequalities. Nevertheless, on grouping of patients into two categories including, on the one hand, CDC codes 1–3 (corresponding to clinically stable situations), and, on the other, CDC codes 4 and 5 (unstable clinical situations), significant differences were not found between the groups. With regard to patient exposure to most of the invasive procedures considered to be risk factors for the development of NI, the results were also similar, with the exceptions of arterial catheterization (less frequent in group A) and the administration of antibiotics before ICU admission (more frequent in group A). The fact that exposure to antibiotic treatment before ICU admission was more frequent in group A could be explained by the higher frequency of community infections in this group of patients (23.29% vs. 17.52%; p <0.05).

Among the NIs diagnosed in our ICU, ventilator-associated pneumonia (VAP) was the most frequent (21.1% of all episodes of NI), as reported by previous studies [4], but it was closely followed by urinary infection associated with urinary catheter use (19.9%). The rate of infections related to central catheter use (without bacteraemia) (21.1%) was equal to that of VAP, but most of the publications do not refer to this type of infection, and they only assess bacteraemia associated with venous catheter use.

Identification of the factors that increase the risk of developing NI is paramount, in order to enable the elaboration of strategies for the prevention of such complications. In our cohort, in the multivariate analysis, we mainly found extrinsic risk factors, in agreement with previously published data [5,14,20–22]. These extrinsic variables were central venous catheter inserted for more than 5 days (OR 18.8) and the presence of tracheostomy (OR 25.3). Both are directly related to the types of NI most frequently diagnosed in our ICU: VAP and central venous catheter infection (CVC). However, urinary catheter use was not an independent risk factor for the development of NI, which is somewhat striking; perhaps the higher incidence of the other two types of NI is the cause of this result.

Fifteen per cent of the patients in our cohort died; the factors that were statistically associated with a fatal outcome were older age (>65 years) (OR 2.9), diabetes mellitus (OR 1.3), administration of antibiotic treatment during hospital stay before ICU admission (OR 3.3), level 3 according to the CDC severity score (OR 4.3), APACHE II score >15 (OR 4), pulmonary artery catheterization (OR 23.5), invasive mechanical ventilation for more than 5 days (OR 7.3), and presence of a urinary catheter for more than 3 days (OR 2.1). However, having undergone non-urgent surgery was a protective factor, probably because these patients developed fewer complications than the patients who were admitted for ‘urgent’ reasons. The identification of prolonged mechanical ventilation and central venous catheterization as risk factors associated with mortality coincides with the findings of previous studies [22]. Other risk factors mentioned in the literature were included in our analyses, such as renal failure, parenteral nutrition, and the presence of tracheostomy, but statistical significance was not detected. Unlike in other studies, in our cohort the development of NI was not a risk factor for death [23,24].

One of the goals of this work was the implementation of an educational programme to improve hand-washing practices and the placement of alcohol gel at the patients’ headboards [11–13,25]. As previously mentioned [1,10–18], hand-washing is usually considered to be the most important measure for the prevention of NI and dissemination of microorganisms. Even so, compliance with this measure does not exceed 50% at most, and although the relevance of hand-washing is generally admitted, some authors find that, under ideal circumstances, hand hygiene may influence only 40% of all NIs in an ICU [26]. Nevertheless, the impact of hand-washing on NI prevalence seems to have been clearly proved [7,12,13], and, even taking into account what has previously been stated, and the fact that, at worst, there is a beneficial effect on 40% of the NIs in the ICU, where the incidence is usually higher than in the rest of the hospital, hand-washing has already been shown to be important.

The compliance with hand-washing (considering both hygienic washing with water and soap, and disinfection with iodine or alcohol gel) obtained in the first observation period of healthcare activities was similar to that reported by other authors [18,26,27], usually under 50%. The prevalence of compliance before patient care was lower than that obtained after care [18,27], which can be explained by workers’ perception of their own health risk. In P2, the frequency of hand-washing, both before and after healthcare, increased significantly, as has been found in previous studies [17,18,27]. However, it should be pointed out that, even though the frequency of hand-washing increased, it continued to be performed mainly with water and soap (hygienic washing), although hand disinfection with hydroalcoholic antiseptics is more advisable [7,10,25]. Considering all the products for hand hygiene available in the second observation period, disinfection with alcohol gel was performed in only 24.8% of hand-washing episodes before patient care and in 7.6% after care. Therefore, it is obvious that, with the implementation of the programme, we did not succeed in raising enough awareness of the importance of this recommendation, which is currently considered to be the reference standard for hand hygiene [27]. Perhaps this explains why, despite the fact that the number of patients who developed NIs in the ICU showed a significant decrease in the second period of study (13.7% in P1 vs. 8.3% in P2; p <0.05), having been admitted during P1 was not a statistically significant risk factor for the development of NI. Multiple factors are involved in the development of NI, and establishing the association between improvement in hand hygiene and the reduction in NIs is difficult. In a meta-analysis of clinical studies [26], only nine publications registered a positive impact of compliance with hand hygiene measures on the prevention of NI, and only four reported important reductions in NIs.


  1. Top of page
  2. Abstract
  3. Introduction
  4. Patients and Methods
  5. Results
  6. Discussion
  7. Transparency Declaration
  8. References
  • 1
    Pittet D, Donaldson L. Clean care is safer care in worldwide priority. Lancet 2005; 366: 12461247.
  • 2
    Burke JP. Infection control—a problem for patient safety. N Engl J Med 2003; 348: 651655.
  • 3
    Stone PW, Larson E, Kawar LN. A systematic audit for economic evidence linking nosocomial infections and infection control interventions: 1990–2000. Am J Infect Control 2002; 30: 145152.
  • 4
    Vincet JL. Nosocomial infections in adult intensive-care units. Lancet 2003; 361: 20682077.
  • 5
    Álvarez Lerma F, Palomar Martínez M, Olaechea Astigarraga P, Insausti Ordeñana J, Bermejo Fraile B, Cerdá Cerdá E. Grupo de estudio de vigilancia de infección nosocomial en UCI. Estudio nacional de vigilancia de infección nosocomial en Unidades de Cuidados Intensivos. Informe del año 2002. Med Intensiva 2005; 29: 112.
  • 6
    World Health Organization. Prevention of nosocomial infections. Clinical guidelines, 2nd edn. Geneva, WHO: 2005.
  • 7
    Ferrer C, Almirante B. Higiene de manos: una prioridad para la seguridad de los pacientes hospitalizados. Enferm Infecc Microbiol Clin 2007; 25: 365368.
  • 8
    Teare EL, Cookson B, French G, Gould D, Jenner E, McCulloch J. Handwashing—a modest measure with big effects. BMJ 1999; 318: 686.
  • 9
    Jarvis WR. Handwashing—the Semmelweis lesson forgotten? Lancet 1994; 344: 13111312.
  • 10
    World Alliance for Patient Safety. WHO Guidelines on hand hygiene in health care. 2009. Available at: (last accessed 16 January 2009).
  • 11
    Allegranzi B, Storr J, Dziekan G, Leotsakos A, Donaldson L, Pittet D. The first global safety challenge ‘Clean care is safer care’: from launch to current progress and achievements. J Hosp Infect 2007; 65: 115123.
  • 12
    Hugonnet S, Pittet D. Hand hygiene—beliefs or science?. Clin Microbiol Infect 2000; 6: 348354.
  • 13
    Pittet D, Allegranzi B, Sax H et al. Evidence-based model for hand transmission during patient care and the role of improved practices. Lancet Infect Dis 2006; 6: 641652.
  • 14
    Nystrom B. Impact of hand washing on mortality in intensive care: examination of the evidence. Infect Control Hosp Epidemiol 1994; 15: 435436.
  • 15
    Lund S, Jackson J, Leggett J, Hales L, Dworkin R, Gilbert D. Reality of glove use and handwashing in a community hospital. Am J Infect Control 1994; 22: 352357.
  • 16
    Pittet D, Mouroaga P, Perneger TV. Compliance with hand washing in a teaching hospital. Ann Intern Med 1999; 130: 126130.
  • 17
    Pittet D, Hugonnet S, Harbarth S et al. Effectiveness of a hospital-wide programme to improve compliance with hand hygiene. Infection Control Programme. Lancet 2000; 356: 13071312.
  • 18
    Sánchez Payá J, Galicia García MD, Gracia Rodríguez RM et al. Grado de cumplimiento y determinantes de las recomendaciones sobre la higiene de manos. Enferm Infecc Microbiol Clin 2007; 25: 369375.
  • 19
    Sociedad Española de Medicina Intensiva y Unidades Coronarias. Grupo de Trabajo de Enfermedades Infecciosas. Manual del Programa para la Vigilancia de Infección Nosocomial en Servicios de Medicina Intensiva y del Estudio Nacional de Vigilancia de Infección Nosocomial en UCI (ENVIN-UCI). Grupo de Trabajo de Enfermedades Infecciosas de la Sociedad Española de Medicina Intensiva y Unidades Coronarias (SEMICYUC); 19941995; 191.
  • 20
    Van der Kooi TI, de Boer AS, Manniën J et al. Incidence and risk factors of device-associated infections and associated mortality at the intensive care in the Dutch surveillance system. Intensive Care Med 2007; 33: 271278.
  • 21
    Ibrahim EH, Tracy L, Hill C, Fraser VJ, Kollef MH. The occurrence of ventilator-associated pneumonia in a community hospital. Chest 2001; 120: 555561.
  • 22
    McCusker ME, Périssé A, Roghamann MC. Severity of illness markers as predictors of nosocomial infection in adult intensive care unit patients. Am J Infect Control 2002; 30: 139144.
  • 23
    Fabbro-Peray P, Sotto A, Defez C et al. Mortality attributable to nosocomial infection: a cohort of patients with and without nosocomial infection in a French university hospital. Infect Control Hosp Epidemiol 2007; 28: 265272.
  • 24
    Ylipalosaari P, Ala-Kokko TI, Laurila J, Ohtonen P, Syrjälä H. Intensive care acquired infection is an independent risk factor for hospital mortality: a prospective cohort study. Crit Care 2006. Available at: (last accessed 7 January 2009).
  • 25
    Pittet D, Boyce JM. Revolutionising hand hygiene in health-care setting: guidelines revisited. Lancet Infect Dis 2003; 3: 269270.
  • 26
    Silvestri L, Petros AJ, Sarginson RE, de la Cal MA, Murray AE, Van Saene HKF. Handwashing in the intensive care unit: a big measure with modest effects. J Hosp Infect 2005; 59: 172179.
  • 27
    Wendt C, Knantz D, Von Baum H. Differences in hand hygiene behavior related to the contamination risk of healthcare activities in different groups of healthcare workers. Infect Control Hosp Epidemiol 2004; 25: 203220.