The study setting comprised the ICU of a second level hospital center (Santa María del Rosell Hospital in Cartagena, Murcia, Spain). The medical team consists of a Head of Unit, a Section Chief, 12 intensivists, 5 residents in training in Intensive Care, and a variable number of residents in other specialties undergoing rotation training in the ICU. The rest of the non-medical personnel include 63 workers (35 nurses and 28 nursing auxiliary personnel members with fixed assignment to the ICU) and orderlies not specifically ascribed to the ICU (i.e., pertaining to the general orderlies staff division), with distribution in shifts–each with 6 nurses, 4 nursing auxiliary personnel members and two orderlies. There were no modifications in the composition or distribution of the workers in the Unit during the study period.

The ICU facilities consist of a hospitalization zone with 16 individual rooms (13 of which are isolated), a room for reception and stabilization of the more seriously ill patients, and a treatment preparation room. There are two independent accesses to the hospitalization zone: one for the patients and the other for their families. In addition, there is a material storage area (two rooms adjacent to the hospitalization zone), an office area, meeting room and dressing area. The unit is equipped with the usual protocols for performing invasive techniques, and moreover has participated in the ENVIN-ICU program years before conduction of the present study.

The study was carried out between 1 February and 31 July 2006. In the first two weeks and in the last two weeks of the study period, we conducted 50h of observation of the in-ICU patient care activities of the healthcare personnel. The periods of observation were distributed into 3-h sessions in morning shifts and 2-h sessions in the afternoon shifts. Potential study subjects were taken to be all healthcare workers of the Unit attending patients admitted to the ICU during the observation sessions. The results of the first observation phase were not made known to the ICU personnel. Observation was carried out independently before and after contact with the patient, and the personnel were not aware of being under observation.

The indications for HH, the use of masks and gloves, or the adoption of aseptic techniques (mask, gloves, gown and drapes), were based on the usual recommendations10 and were referred to before and after the patient care activities.

Washing of the hands is considered hygienic when made with water and detergent soap, and is considered disinfecting or “surgical” when performed with substances approved for the purpose (iodine solution or water-soluble alcohol gel). The assessed techniques and times were adjusted to the recommendations referred to the effect.10 The following situations and activities were considered adequate for HH: before and after contact with the skin of the patient; before and after handling intravenous devices or bladder or nasogastric catheters; before and after contact with wounds; before and after contact with mucosal membranes; before and after contact with body fluids; following the removal of gloves or other barriers; and after cleaning, removal of waste, etc.

When performing procedures involving obligate sterility,10 and in addition to disinfection of the hands before and after the procedure, it was considered necessary to wear a cap, mask, sterile gown and sterile gloves, and to use sterile material and a sterile intervention field (“aseptic technique”).

Contact with the patient respirator or with the bladder catheter or urine bag was included within the category of contact with patient mucosal membranes. For increased simplification and to facilitate observation of the HH practices, simple contact with the bed of the patient or with inanimate objects in the surroundings was not taken into consideration, in the absence of contact with the skin of the patient.

Risk activity in turn was taken to constitute direct contact with mucosal membranes, wounds, blood or other biological material of the patient.

The following measures were adopted:

• (1)

  Acquisition of a water–alcohol antiseptic to add to the already existing products for hand hygiene and disinfection. The dispensers of the new product were distributed in the vicinity of each of the rooms of the ICU, together with a pair of dispensers in each of the treatment preparation rooms.

• (2)

  Informative sessions on the new water–alcohol antiseptic for HH, with indications and instructions for use.

• (3)

  Distribution of informative notices on the recommendations referred to the use of barrier measures and instructions on correct HH (with water and antiseptic soap or water–alcohol antiseptics) in strategic areas of the ICU: nursing controls, medication preparation rooms, offices and meeting rooms.

Patient severity upon admission to the ICU was established according to the criteria of the United States Centers for Disease Control (CDC),24 based on 5 codes:

• -

  Severity code 1 (CDC 1): postsurgical patients requiring routine postoperative monitoring but no intensive care.

• -

  Severity code 2 (CDC 2): stable medical patients requiring continuous prophylactic monitoring but no intensive care.

• -

  Severity code 3 (CDC 3): stable patients requiring intensive care.

• -

  Severity code 4 (CDC 4): unstable patients requiring intensive care.

• -

  Severity code 5 (CDC 5): unstable patients requiring intensive care, frequent re-evaluation and treatment adjustments.

There were no epidemics or outbreaks of multiresistant pathogens requiring a change in habits or procedures during the study periods.

The data collected with the different forms were entered on a Microsoft Excel 2000® spreadsheet for posterior analysis using the Statistical Package for Social Science (SPSS) version 15 for Microsoft Windows. A descriptive analysis was made of the observations referred to HH compliance. Qualitative variables were presented as percentages, while quantitative variables were expressed as the mean, standard deviation and range. The comparative analysis of the variables was carried out with the Student t-test in the case of quantitative variables, and applying the chi-squared test in the case of qualitative variables. Statistical significance was considered for p<0.05. Multivariate analysis in turn was performed using the stepwise binary logistic regression method to identify the factors independently correlated to compliance with the HH recommendations on the part of the healthcare workers in our ICU, before and after contact with the patient. The influencing or potentially influencing variables were entered in the multivariate analysis, with particular consideration of whether the fact that observation was made before or after the intervention program constituted an independent factor associated to HH compliance (both before and after contact with the patient).

We observed a total of 338 care activities in which HH compliance on the part of the intervening healthcare worker was assessed both before and after the care intervention. The categories of the observed workers, the patient severity according to the CDC code, and the different groups of activities observed during this period are shown in Table 1. Nurses were the most frequently observed professionals, while the patients most frequently implicated in the care activities corresponded to CDC severity code 4, and the most commonly observed care activity was simple contact with the skin of the patient.

Among the 338 observations prior to patient care or contact, HH was performed in 118 cases (34.9%); of these, 100 (84.7%) corresponded to hygienic washing and the rest (18 cases) to surgical or disinfectant washing with iodine solution. In the 338 observations after contact with the patient, HH was performed in 175 cases (51.8%), of which 170 (97%) corresponded to simple hygienic washing. In 44 (13%) of these 338 observed activities a completely aseptic technique was required, but was only performed in 35 cases (79.5%). Gloves were worn during contact with the patient in 94.1% of the cases, and a surgical mask in 5.9% (Table 2).

Hand hygiene, before and after patient care, and the wearing of a mask and gloves, proved more frequent during the second observation period (Table 2). Of the 355 observations prior to patient care or contact, HH was carried out in 161 cases (65.4%); of these activities of hygiene, 101 (62.7%) corresponded to hygienic washing, 40 (24.8%) involved use of the alcohol gel product, and 20 (12.4%) involved the use of iodine. In the 355 observations posterior to contact with the patient, HH was performed in 224 cases (63.1%), of which 198 (81.1%) represented simple hygienic washing and only 17 (7.6%) involved use of the alcohol gel product. A completely aseptic technique was required in 42 of the 355 observed activities (11.8%) but was only performed in 37 cases (88.1%) (Table 2).

The relationship between use of the barrier measures during patient care and the professional category of the intervening healthcare worker is reflected in Table 3. The association between HH compliance and patient severity and the type of activity carried out is in turn reflected in Table 4.

The multivariate analysis was performed to identify the factors independently that were related to HH compliance prior to contact with the patient, and which included 693 observations made before patient contact (338+355), the professional category “nurse” and the presence of a program for improvement in the use of barrier measures were associated to HH compliance before contact with the patient, while the wearing of gloves, the application of an aseptic technique and the professional category “orderly” were associated to failure to comply with hand hygiene before contact with the patient. In the multivariate analysis designed to identify the factors related to HH compliance after contact with the patient (and which likewise comprised 693 observations made after contact with the patient), only the presence of a program for improvement in the use of barrier measures was found to be significantly correlated to HH compliance (Table 5).