Background: Tracheostomy is a procedure commonly required in the intensive care unit. In the last two decades, the use of the percutaneous method has increased in parallel with improvements in the technique. Objective: To describe our experience in employing the percutaneous method over the last 3.5 years. Methods: We created, retrospectively, a database of prospective tracheostomy data related to 78 patients evaluated between January 2000 and July 2003. We used the percutaneous tracheostomy techniques of either progressive dilatation (in 36 patients) or forceps dilatation (in 42 patients). Data are expressed as number of occurrences or median with interquartile ranges. Results: The mean age of the patients was 66 (range, 43-75), and the median APACHE II score was 16 (range, 12-21). The median time spent on mechanical ventilation prior to tracheostomy was 14 days (range, 10-17 days). Of the 78 patients studied, 18 (23%) died while in the intensive care unit. The most common cause of admission was acute central nervous system disturbance (in 45%). Most of the tracheostomies performed were indicated due to difficulty in weaning from mechanical ventilation (in 50%) or to Glasgow Coma scores consistently lower than 8 (in 49%). Bronchoscopy was used in all but 6 of the procedures. There were complications in 33% of the procedures. The most common complication was light bleeding, without need for transfusion. No patient died due to complications arising from the procedure. Conclusion: Percutaneous tracheostomy is reasonable and safe when performed in an intensive care unit.

Keywords: Tracheostomy/methods. Respiration, Artificial/methods. Respiratory insufficiency/therapy. Respiratory insufficiency/complications.

Poor sleep quality is a consistently reported by patients in the ICU. In such a potentially hostile environment, sleep is extremely fragmented and sleep architecture is unconventional, with a predominance of superficial sleep stages and a limited amount of time spent in the restorative stages. Among the causes of sleep disruption in the ICU are factors intrinsic to the patients and the acute nature of their condition, as well as factors related to the ICU environment and the treatments administered, such as mechanical ventilation and drug therapy. Although the consequences of poor sleep quality for the recovery of ICU patients remain unknown, it seems to influence the immune, metabolic, cardiovascular, respiratory, and neurological systems. There is evidence that multifaceted interventions focused on minimizing nocturnal sleep disruptions improve sleep quality in ICU patients. In this article, we review the literature regarding normal sleep and sleep in the ICU. We also analyze sleep assessment methods; the causes of poor sleep quality and its potential implications for the recovery process of critically ill patients; and strategies for sleep promotion.

Keywords: Sleep; Sleep deprivation; Intensive care units.

Introduction: The diagnosis of diffuse lung disease is still a challenge for the pediatric intensive care physician. Routine clinical examinations and laboratory tests are frequently negative. The objective of this study was to evaluate the diagnostic potential, the impact on therapy and the rate of complications of open lung biopsy in children with undiagnosed diffuse lung disease, respiratory failure and inappropriate response to initial therapy. Methods: From January 1987 to January 1997, 29 children with diffuse pulmonary disease of unknown etiology, respiratory failure (PaO2 /FiO2 < 300) and no response to previous treatments were considered for open lung biopsy. Newborns, children with known chronic pulmonary disease and children with untreatable shock or coagulopathy were excluded. All biopsies were performed by a thoracic surgeon by a microthoracotomy in the lung shown to be the most affected by X-ray examination. Tissue samples were analyzed in terms of cultures, light microscopy, electron microscopy and immunofluorescence microscopy, according to the pathologist's decision. Results: All biopsies (100%) resulted in at least one histological diagnosis and in 20 patients (68.9%) it was obtained a specific diagnosis. The most frequent histological patterns found were: non-specific interstitial pneumonitis with variable degrees of fibrosis in 18 cases; bronchiolitis in eight cases and pulmonary hypertension in three cases. Regarding the most frequent specific diagnosis, six children were found with cytomegalovirus infection, three with Pneumocystis carinii, three with adenovirus and three with respiratory syncytial virus infection. These data induced a change in therapy in 20 children (68.9%). The most frequent changes in therapy were the use of corticosteroids in 14 children and a review of the antibiotic regimen in six patients. Seven patients (24.1%) presented with complications that were easily resolved. There were 13 deaths, probably due to the critical conditions of these patients, all unrelated to the procedure. Conclusions: Open lung biopsy, though an invasive procedure, should be considered in the evaluation of selected children with undiagnosed diffuse lung disease, respiratory failure and with no satisfactory response to previous therapies.

Objective: To identify factors predictive of mortality in patients admitted to the ICU with tuberculosis (TB)/HIV coinfection in the Manaus, Amazon Region. Methods: This was a retrospective cohort study of TB/HIV coinfected patients over 18 years of age who were admitted to an ICU in the city of Manaus, Brazil, between January of 2011 and December of 2014. Sociodemographic, clinical, and laboratory variables were assessed. To identify factors predictive of mortality, we employed a Cox proportional hazards model. Results: During the study period, 120 patients with TB/HIV coinfection were admitted to the ICU. The mean age was 37.0 ± 11.7 years. Of the 120 patients evaluated, 94 (78.3%) died and 62 (66.0%) of those deaths having occurred within the first week after admission. Data on invasive mechanical ventilation (IMV) and ARDS were available for 86 and 67 patients, respectively Of those 86, 75 (87.2%) underwent IMV, and, of those 67, 48 (71.6%) presented with ARDS. The factors found to be independently associated with mortality were IMV (p = 0.002), hypoalbuminemia (p = 0.013), and CD4 count < 200 cells/mm3 (p = 0.002). Conclusions: A high early mortality rate was observed among TB/HIV coinfected ICU patients. The factors predictive of mortality in this population were IMV, hypoalbuminemia, and severe immunosuppression.

Keywords: Mycobacterium tuberculosis; Critical care; Respiration, artificial; Acquired immunodeficiency syndrome.

Objective: To evaluate the impact that the implementation of therapeutic guidelines has on the empirical treatment of nosocomial pneumonia. Methods: A clinical trial, using historical controls and involving current ICU patients who had acquired nosocomial pneumonia, was carried out from June of 2002 to June of 2003. All were treated according to therapeutic guidelines developed by the Commission for Nosocomial Infection Control of the institution (group with intervention). As controls, the medical charts of the patients who acquired nosocomial pneumonia between June of 2000 and June of 2001 (group without intervention) were analyzed. Mortality and mean treatment period, as well as the length of hospital and ICU stays, were determined for the patients who acquired nosocomial pneumonia. Results: Mortality associated with pneumonia was lower in the group treated according to the therapeutic guidelines (26 vs. 53.6%; p = 0.00). As for overall mortality, there was no statistically significant difference between the two periods (51 vs. 57.9%; p = 0.37). There was also no difference in the type of microorganisms isolated, treatment period, length of hospital stay or length of ICU stay. Conclusion: The implementation of therapeutic guidelines for the treatment of nosocomial pneumonia acquired in the ICU can be efficacious in decreasing mortality rates.

Keywords: Pneumonia/treatment; Cross infection; Intensive care units; Mortality.

Objectives: To identify the degree of risk and the practice of prophylaxis to venous thromboembolism (VTE) in an intensive care unit (ICU). Methods: The subjects were patients admitted to the ICU between December 1997 and February 1998. Patients were excluded if they were in anticoagulant therapy or presented contra-indication to heparin. Determination of risk factors and classification of the degree of risk for VTE were done according to international consensus. The ICU medical staff did not know about the study. Results: The authors evaluated one hundred and eighty patients with a mean age of 58 years (±16.5). Risk factors more frequently found were: age ³ 40 years (85.0%), surgery (47.8%), thoracic or abdominal infection (22.8%). Two or more risk factors were present in 146 (81%) cases. In the evaluation of VTE risk, 142 (79%) patients were classified as moderate/high risk. Prophylactic measures were prescribed to 102 patients (57%), and heparin was used in 60% of moderate/high risk cases. There was a significant association between risk level and the number of risk factors with the use of prophylaxis (p < 0.05). Conclusion: VTE risk factors were frequent in the sample analyzed. However, 40% of patients considered moderate/high risk did not receive pharmacological prophylaxis for VTE.

Keywords: thromboembolism, prevention & control

Objective: To establish the incidence of acute lung injury and acute respiratory distress syndrome, as well as related risk factors and mortality in an intensive care unit. To compare patients developing lung injury with at-risk patients not presenting acute lung injury or acute respiratory distress syndrome. Methods: The study was conducted in the intensive care unit of the Ribeirão Preto Hospital das Clínicas Emergency Room. All patients admitted between May 2001 and April 2002 were monitored prospectively. Clinical data, Acute Physiologic and Chronic Health Evaluation II score, complications, length of stay in the intensive care unit and lung injury data were recorded. Results: Of the 524 patients admitted, 175 (33.4%) presented risk factors for acute lung injury and acute respiratory distress syndrome, 33 (6.3%) developed acute respiratory distress syndrome, and 12 (2.3%) developed acute lung injury. The main risk factors were pneumonia (37.7%), shock (32.0%), multiple trauma (24.6%) and sepsis (21.1%). Patients developing acute lung injury had higher Acute Physiologic and Chronic Health Evaluation II scores (p < 0.05), more frequently presented sepsis (p = 0.001), developed more complications (p = 0.001) and presented greater mortality (p = 0.001). The main cause of death was multiple organ failure (38.5%). Conclusion: The incidence of acute lung injury and acute respiratory distress syndrome was 2.3% and 6.3%, respectively.

Keywords: Respiratory distress syndrome, adult/epidemiology; Respiratory distress syndrome, adult/mortality;

Objective: To translate the Richards-Campbell Sleep Questionnaire (RCSQ) and Sleep in the Intensive Care Unit Questionnaire (SICUQ) to Portuguese, making the appropriate cross-cultural adaptations for their use in Brazil, as well as to determine the interobserver reliability of the instruments. Methods: In this study, we evaluated medical and surgical patients admitted to the adult ICU of the Federal University of Paraná Hospital de Clínicas, in the city of Curitiba, Brazil, between June of 2017 and January of 2018. The translation and cross-cultural adaptation of the questionnaires involved the following steps: translation, synthesis, back-translation, revision by an expert panel, approval of the back-translation by the original authors, pretesting, and creation of the final versions. Two researchers applied the Portuguese-language versions in the evaluation of critically ill patients. Interobserver reliability was assessed by calculating the intraclass correlation coefficient (ICC) and 95% CI. Results: The sample comprised 50 patients, of whom 27 (54%) were women. The mean age was 47.7 ± 17.5 years. The main reason for ICU admission, in 10 patients (20%), was cancer. The interobserver reliability of the questionnaires ranged from good to excellent. For the RCSQ, the ICC was 0.84 (95% CI: 0.71-0.90). For SICUQ domains 1-5 (sleep quality and daytime sleepiness), the ICC was 0.75 (95% CI: 0.55-0.86), whereas it was 0.86 (95% CI: 0.76-0.92) for SICUQ domains 6 and 7 (causes of sleep disruption). Conclusions: The cross-culturally adapted, Portuguese-language versions of the RCSQ and SICUQ appear to have good interobserver reliability.

Keywords: Sleep; Intensive care units; Sleep deprivation; Surveys and questionnaires; Translations; Cross-cultural comparison.

Objective: To translate the Perme Intensive Care Unit Mobility Score and the ICU Mobility Scale (IMS) into Portuguese, creating versions that are cross-culturally adapted for use in Brazil, and to determine the interobserver agreement and reliability for both versions. Methods: The processes of translation and cross-cultural validation consisted in the following: preparation, translation, reconciliation, synthesis, back-translation, review, approval, and pre-test. The Portuguese-language versions of both instruments were then used by two researchers to evaluate critically ill ICU patients. Weighted kappa statistics and Bland-Altman plots were used in order to verify interobserver agreement for the two instruments. In each of the domains of the instruments, interobserver reliability was evaluated with Cronbach's alpha coefficient. The correlation between the instruments was assessed by Spearman's correlation test. Results: The study sample comprised 103 patients-56 (54%) of whom were male-with a mean age of 52 ± 18 years. The main reason for ICU admission (in 44%) was respiratory failure. Both instruments showed excellent interobserver agreement (κ > 0.90) and reliability (α > 0.90) in all domains. Interobserver bias was low for the IMS and the Perme Score (−0.048 ± 0.350 and −0.06 ± 0.73, respectively). The 95% CIs for the same instruments ranged from −0.73 to 0.64 and −1.50 to 1.36, respectively. There was also a strong positive correlation between the two instruments (r = 0.941; p < 0.001). Conclusions: In their versions adapted for use in Brazil, both instruments showed high interobserver agreement and reliability.

Keywords: Physical therapy modalities; Intensive care units; Translations; Validation studies.

Objective: To determine whether the use of a set of preoperative variables can predict the need for postoperative ICU admission. Methods: This was a prospective observational cohort study of 120 patients undergoing elective pulmonary resection between July of 2009 and April of 2012. Prediction of ICU admission was based on the presence of one or more of the following preoperative characteristics: predicted pneumonectomy; severe/very severe COPD; severe restrictive lung disease; FEV1 or DLCO predicted to be < 40% postoperatively; SpO2 on room air at rest < 90%; need for cardiac monitoring as a precautionary measure; or American Society of Anesthesiologists physical status ≥ 3. The gold standard for mandatory admission to the ICU was based on the presence of one or more of the following postoperative characteristics: maintenance of mechanical ventilation or reintubation; acute respiratory failure or need for noninvasive ventilation; hemodynamic instability or shock; intraoperative or immediate postoperative complications (clinical or surgical); or a recommendation by the anesthesiologist or surgeon to continue treatment in the ICU. Results: Among the 120 patients evaluated, 24 (20.0%) were predicted to require ICU admission, and ICU admission was considered mandatory in 16 (66.6%) of those 24. In contrast, among the 96 patients for whom ICU admission was not predicted, it was required in 14 (14.5%). The use of the criteria for predicting ICU admission showed good accuracy (81.6%), sensitivity of 53.3%, specificity of 91%, positive predictive value of 66.6%, and negative predictive value of 85.4%. Conclusions: The use of preoperative criteria for predicting the need for ICU admission after elective pulmonary resection is feasible and can reduce the number of patients staying in the ICU only for monitoring.

Keywords: Thoracic surgery; Risk factors; Intensive care units.

Objective: To determine the availability of noninvasive positive-pressure ventilation equipment, as well as the level of expertise and familiarity of physicians, nurses and physiotherapists with noninvasive positive-pressure ventilation in the intensive care units of public, private and teaching hospitals in the greater metropolitan area of São Paulo, Brazil. Methods: On-site administration of questionnaires. Results: Noninvasive positive-pressure ventilation equipment was widely available and was more commonly found in private hospitals than in teaching hospitals. Such equipment was least available in public hospitals, in which the predominant method was the use of mechanical ventilators designed for invasive ventilation and adapted to noninvasive positive-pressure ventilation. In private hospitals, continuous flow ventilators were more common, whereas, in teaching hospitals, ventilators specifically designed for noninvasive ventilation were typically employed. All physiotherapists felt themselves capable of initiating noninvasive positive pressure ventilation, compared with 72.6% of physicians and 33.3% of nurses. Physicians and physiotherapists presented high percentages of correct answers when asked about the indications and contraindications for the use of noninvasive positive-pressure ventilation. Over a one year period, more physiotherapists read articles about noninvasive positive-pressure ventilation and participated in related classes than did physicians, who in turn did so more than did nurses. Conclusion: Noninvasive positive-pressure ventilation equipment is widely available in the greater metropolitan area of São Paulo, although differences exist among public, private and teaching hospitals in terms of the type of equipment used. Physicians and physiotherapists exhibited considerable knowledge regarding the indications and contraindications for the use of noninvasive positive-pressure ventilation. More physiotherapists felt themselves able to initiate noninvasive positive-pressure ventilation, and their knowledge of the subject was more current than was that of physicians or nurses.

Keywords: Ventilators, mechanical; Positive-pressure respiration/methods; Respiratory insuficiency; Intensive care units; Hospitals

Objective: To assess the diagnostic performance of the Xpert MTB/RIF assay, a rapid molecular test for tuberculosis, comparing it with that of AFB staining and culture, in BAL fluid (BALF) samples from patients with clinically suspected pulmonary tuberculosis (PTB) who are sputum smear-negative or produce sputum samples of insufficient quantity. Methods: This was a retrospective study of 140 cases of suspected PTB in patients who were smear-negative or produced insufficient sputum samples and were evaluated at a tertiary teaching hospital in the city of Rio de Janeiro, Brazil. All of the patients underwent fiberoptic bronchoscopy with BAL. The BALF specimens were evaluated by AFB staining, mycobacterial culture, and the Xpert MTB/RIF assay. Results: Among the 140 patients, results for all three microbiological examinations were available for 73 (52.1%), of whom 22 tested positive on culture, 17 tested positive on AFB staining, and 20 tested positive on the Xpert MTB/RIF assay. The sensitivity, specificity, positive predictive value, negative predictive value, and overall accuracy for AFB staining were 68.1%, 96.1%, 88.2%, 87.5%, and 87.6%, respectively, compared with 81.8%, 96.1%, 90.0%, 92.4%, and 91.8%, respectively, for the Xpert MTB/RIF assay. The agreement between AFB staining and culture was 82.3% (kappa = 0.46; p < 0.0001), whereas that between the Xpert MTB/RIF assay and culture was 91.8% (kappa = 0.8; p < 0.0001). Conclusions: In BALF samples, the Xpert MTB/RIF assay performs better than do traditional methods, providing a reliable alternative to sputum analysis in suspected cases of PTB. However, the rate of discordant results merits careful consideration.

Keywords: Tuberculosis; Molecular diagnostic techniques; Bronchoscopy; Bronchoalveolar lavage fluid; Mycobacterium tuberculosis.

Objective: To investigate the clinical application of CT and CT-guided percutaneous transthoracic needle biopsy (CT-PTNB) in patients with indeterminate pulmonary nodules (IPNs). Methods: We retrospectively studied 113 patients with PNs undergoing CT and CT-PTNB. Variables such as gender, age at diagnosis, smoking status, CT findings, and CT-PTNB techniques were analyzed. Data analysis was performed with the Student's t-test for independent samples the chi-square test, and normal approximation test for comparison of two proportions. Results: Of the 113 patients studied, 68 (60.2%) were male and 78 (69%) were smokers. The diameter of malignant lesions ranged from 2.6 cm to 10.0 cm. Most of the IPNs (85%) were located in the peripheral region. The biopsied IPNs were found to be malignant in 88 patients (77.8%) and benign in 25 (22.2%). Adenocarcinoma was the most common malignant tumor, affecting older patients. The IPN diameter was significantly greater in patients with malignant PNs than in those with benign IPNs (p < 0.001). Having regular contour correlated significantly with an IPN being benign (p = 0.022), whereas spiculated IPNs and bosselated IPNs were more often malignant (in 50.7% and 28.7%, respectively). Homogeneous attenuation and necrosis were more common in patients with malignant lesions (51.9% and 26.9%, respectively) Conclusions: In our sample, CT and CT-PTNB were useful in distinguishing between malignant and benign IPNs. Advanced age and smoking were significantly associated with malignancy. Certain CT findings related to IPNs (larger diameter, spiculated borders, homogeneous attenuation, and necrosis) were associated with malignancy.

Keywords: Solitary pulmonary nodule; Tomography; Image-guided biopsy.

Objective: To evaluate the influence of the surgical team (general surgery or thoracic surgery) and the surgical technique (with or without isthmectomy) on the incidence of postintubation injuries in the airways of tracheostomized patients. Methods: Between January 1st and August 31st, 2007, 164 patients admitted to the adult intensive care unit and tracheally intubated for more than 24 h were studied prospectively at the Sumaré State Hospital, located in the city of Sumaré, Brazil. When tracheostomy was necessary, these patients were randomly assigned to thoracic or general surgery teams. All of the patients were submitted to fiberoptic tracheoscopy for decannulation or late evaluation of the airway. Results: Of the 164 patients in the study, 90 (54.88%) died (due to causes unrelated to the procedure), 67 (40.85%) completed follow-up, and 7 (4.27%) were lost to follow-up. Of the 67 patients who completed follow-up, 32 had undergone tracheostomy (21 by the general surgery team and 11 by the thoracic surgery team), and 22 had been submitted to isthmectomy (11 by the general surgery team and 11 by the thoracic surgery team). There was no difference between the surgical teams in terms of the incidence of stoma complications. However, there was a significant difference when the surgical techniques (with or without isthmectomy) were compared. Conclusions: Not performing isthmectomy in parallel with tracheostomy leads the surgeon to open the tracheal stoma more distally than expected. In such cases, there were more stoma complications.

Keywords: Tracheostomy; Intensive care units; Tracheal stenosis.

Objective: To determine the feasibility of performing tracheostomy in the intensive care unit (ICU) environment and to assess procedurerelated complications and mortality. Methods: The medical records of the 73 patients submitted to tracheostomy in the ICU of the Federal University of São Paulo Hospital São Paulo between January and November of 2003 were evaluated retrospectively. All operations were performed by surgical residents, under the supervision of a thoracic surgeon, using the open technique standardized at the facility. Results: The mean age of the patients was 55.2 years. Of the 73 patients evaluated, 47 (64.4%) were male and 26 (35.6%) were female. The most common indication was prolonged orotracheal intubation (76.7%). There was no procedure-related mortality, and, in all patients, the procedure was successfully performed in the ICU. Early complications occurred in 2 patients (2.7%), who presented increased local bleeding, which was controlled using compression. The late complication was infection at the incision site, which occurred in 2 patients (2.7%) and was treated by applying local dressings, without further clinical repercussions. Conclusions: Based on the results of our analysis, which are comparable to those found in the literature regarding tracheostomy performed in the operating room, we concluded that tracheostomy in the ICU is feasible and presents a low rate of complications, even when performed in critically ill patients and by surgeons in training. Therefore, in our view, it is possible to state that performing tracheostomy in the ICU is worthwhile.

Keywords: Tracheostomy; Intensive care units; Mortality.

Objective: To evaluate and compare subjective sleep quality in medical students across the various phases of the medical course. Methods: This was a cross-sectional study involving medical undergraduates at one medical school in the city of Botucatu, Brazil. All first- to sixth-year students were invited to complete the Pittsburgh Sleep Quality Index, which has been validated for use in Brazil. Participants were divided into three groups according to the phase of the medical course: group A (first- and second-years); group B (third- and fourth-years); and group C (fifth- and sixth-years). The results obtained for the instrument components were analyzed for the total sample and for the groups. Results: Of the 540 students invited to participate, 372 completed the instrument fully. Of those, 147 (39.5%) reported their sleep quality to be either very or fairly bad; 110 (29.5%) reported taking more than 30 min to fall asleep; 253 (68.0%) reported sleeping 6-7 h per night; 327 (87.9%) reported adequate sleep efficiency; 315 (84.6%) reported no sleep disturbances; 32 (8.6%) reported using sleeping medication; and 137 (36.9%) reported difficulty staying awake during the day at least once a week. Group comparison revealed that students in group A had worse subjective sleep quality and greater daytime dysfunction than did those in groups B and C. Conclusions: Medical students seem to be more exposed to sleep disturbance than other university students, and first- and second-years are more affected than those in other class years because they have worse subjective sleep quality. Active interventions should be implemented to improve sleep hygiene in medical students.

Keywords: Pulmonary disease, chronic obstructive; Risk groups; Exercise.

Objective: To assess the direct and indirect costs of diagnosing and treating community-acquired pneumonia (CAP), correlating those costs with CAP severity at diagnosis and identifying the major cost drivers. Methods: This was a prospective cost analysis study using bottom-up costing. Clinical severity and mortality risk were assessed with the pneumonia severity index (PSI) and the mental Confusion-Urea-Respiratory rate-Blood pressure-age ≥ 65 years (CURB-65) scale, respectively. The sample comprised 95 inpatients hospitalized for newly diagnosed CAP. The analysis was run from a societal perspective with a time horizon of one year. Results: Expressed as mean  standard deviation, in Euros, the direct and indirect medical costs per CAP patient were 696  531 and 410  283, respectively, the total per-patient cost therefore being 1,106  657. The combined budget impact of our patient cohort, in Euros, was 105,087 (66,109 and 38,979 in direct and indirect costs, respectively). The major cost drivers, in descending order, were the opportunity cost (lost productivity); diagnosis and treatment of comorbidities; and administration of medications, oxygen, and blood derivatives. The CURB-65 and PSI scores both correlated with the indirect costs of CAP treatment. The PSI score correlated positively with the overall frequency of use of health care services. Neither score showed any clear relationship with the direct costs of CAP treatment. Conclusions: Clinical severity at admission appears to be unrelated to the costs of CAP treatment. This is mostly attributable to unwarranted hospital admission (or unnecessarily long hospital stays) in cases of mild pneumonia, as well as to over-prescription of antibiotics. Authorities should strive to improve adherence to guidelines and promote cost-effective prescribing practices among physicians in southeastern Europe.

Keywords: Pneumonia; Cost of illness; Costs and cost analysis; Health care costs; Hospitalization; Severity of illness index.