Q Manage Health Care Vol. 23, No. 2, pp. 86–93 C 2014 Wolters Kluwer Health | Lippincott Williams & Wilkins
Factors Influencing the Length of Hospital Stay in Patients With Acute Exacerbations of Chronic Obstructive Pulmonary Disease Admitted to Intensive Care Units Chok Limsuwat, MD; Charoen Mankongpaisarnrung, MD; Nat Dumrongmongcolgul, MD; Kenneth Nugent, MD Background: The cost of hospital admissions for acute exacerbations of COPD (AECOPD) accounts for 70% of total costs for the treatment of COPD patients. We wanted to identify clinical parameters associated with a longer length of stay (LOS) in these patients. Methods: We reviewed electronic medical records of patients with AECOPD admitted between January 1, 2006, and December 31, 2010. The inclusion criteria were age 45 years or older, the diagnosis of AECOPD by GOLD (Global Initiative for Chronic Obstructive Lung Disease) guideline criteria, and admission to an intensive care unit. We compared the quartile with the longest LOS group with the other 3 quartiles using routine clinical data. Results: 217 patients met inclusion criteria. The mean age was 67.4 ± 10.9 years, 47% were male, the mean FEV1 s (forced expiratory volume in 1 second) was 42.4% ± 17.4% of predicted, and the mean LOS was 9.0 ± 6.0 days. Univariate analysis demonstrated that nursing home status, low albumins, the presence of pleural effusions, intubation, and high APACHE II scores were associated with increased LOS (P < .05 for each factor). Multivariate logistic regression demonstrated that the need for intubation (P < .001) predicted an increased LOS. Conclusions: Our study demonstrates that intubation for mechanical ventilation increased the LOS in patients with AECOPD. More intensive interventions in these patients might decrease the LOS and improve outcomes. Key words: acute flare, length of stay, COPD, mortality, risk factors
C
hronic obstructive pulmonary disease (COPD) is a global problem that accounts for up to 500 000 hospitalizations per year in the United States for acute exacerbations (AECOPD).1,2 It is projected that COPD will be the third most common cause of death by 2020.3 Acute exacerbations of COPD have relatively high hospital costs, and 70% of the total cost for the treatment of COPD occurs during hospital admissions.4,5 Patients with COPD frequently have extrapulmonary disorders, such as cardiovascular disease, cancer, and depression, and these comorbidities may increase morbidity, mortality, and length of stay (LOS).6-8 Recent studies on AECOPD admissions have focused on reducing duration of admission by improving care processes and efficiency.9,10 We wanted to identify factors associated with an increased LOS in AECOPD patients who were admitted to intensive care units (ICUs). We hypothesized that underlying medical conditions, such as congestive heart failure, diabetes mellitus and cancer, nursing home residence, initial laboratory tests on admission, chest x-rays, and disease severity classification prediction scores, such as CURB 65 and APACHE II, might help physicians identify patients requiring longer LOS. Author Affiliations: Department of Internal Medicine, Texas Tech Health Sciences Center, Lubbock, Texas. Correspondence: Kenneth Nugent, MD, Department of Internal Medicine, Texas Tech Health Sciences Center, 3601 4th St, Lubbock, TX 79430 ([email protected]). The author declares no conflicts of interest. DOI: 10.1097/QMH.0000000000000024
86 Copyright © 2014 Lippincott Williams & Wilkins. Unauthorized reproduction of this article is prohibited.
COPD Acute Flares and Hospital LOS
METHODS We conducted retrospective chart reviews of patients hospitalized in University Medical Center in Lubbock, Texas, between January 1, 2006, and December 31, 2010, who were diagnosed with a COPD exacerbation and needed ICU admission. We used International Classification of Diseases, Ninth Revision codes 491.21 (obstructive chronic bronchitis with [acute] exacerbation) to identify patients. The inclusion criteria were ages 45 years or older, a diagnosis of COPD exacerbation defined by at least 2 GOLD (Global Initiative for Chronic Obstructive Lung Disease) criteria (an increase in dyspnea, an increase in frequency and severity of cough, and an increase in sputum production in terms of volume and/or changed character), and the need for ICU admission. The exclusion criteria included a history of other respiratory disease, such as asthma, or a cardiac disorder resulting in decompensated congestive heart failure.4 Using electronic medical records and case management resources, we collected the patients’ ages, genders, body mass indices, baseline pulmonary function tests, comorbidities, complete blood cell count, complete metabolic profiles, initial arterial blood gases, sputum cultures, blood cultures, chest x-rays, APACHE II scores, CURB65 scores, final diagnoses, total duration of hospital stays, and in-hospital mortality. The primary outcome was the identification of the factors associated with increased LOS. To determine this outcome, we separated patients into 4 quartiles by LOS and analyzed variable factors associated with the longest LOS quartile. Descriptive statistics were used to calculate the medians, means, standard deviations, and frequencies for various study parameters. We then used logistic regression analysis to analyze which factors associated with increased LOS. To determine factors that predicted increased LOS, we selected significant factors with P values less than .05 from the univariate analysis to analyze in the multivariate logistic regression models. Statistical analysis was performed using SPSS version 16.0; P values less than .05 were considered statistically significant. The institutional
87
review board at Texas Tech University Health Sciences Center approved the study.
RESULTS We reviewed 325 patients admitted between January 1, 2006, and December 31, 2010, to ICUs with the International Classification of Diseases, Ninth Revision code for AECOPD. However, after chart reviews, 108 patients were excluded from the study (5 deaths, 103 survivors) because these patients either did not meet inclusion criteria or met exclusion criteria. The majority of excluded patients had a concurrent diagnosis of decompensated congestive heart failure or acute asthma. Therefore, this study included 217 cases with AECOPD who needed ICU admission; 26 died during hospitalization and 191 were discharged from the hospital. The patients were separated into 4 quartiles by LOS (Figure 1). The mean age of these patients was 67.4 ± 10.9 years, and 47% were men. The mean body mass index was 26.7 ± 8.4 kg/m2 . Nearly 40% (38.2%) of the patients had baseline pulmonary function tests performed within a year of admission, and 10.1%
Figure 1. Case selection process.
Copyright © 2014 Lippincott Williams & Wilkins. Unauthorized reproduction of this article is prohibited.
88
QUALITY MANAGEMENT
IN
HEALTH CARE/VOLUME 23, ISSUE 2, APRIL–JUNE 2014
had a FEV1 s (forced expiratory volume in 1 second) less than 30% of predicted. The majority of patients had comorbidity, and 59.5%, 28.6%, and 21.5% of study population had hypertension, diabetes, and a history of congestive heart failure, respectively (Table 1). The median LOS for this patient cohort was 7 days, and mean LOS was 9.01 ± 6.00 days. The mean LOS was 9.3 ± 6.0 days in men and 8.8 ± 5.1 days in women. Patients with very severe reductions in FEV1 had a mean LOS of 10.8 ± 6.3 days, patients with diabetes had a mean LOS of 9.9 ± 6.8 days, patients with a history of congestive heart failure had a mean LOS of 9.6 ± 6.1 days, and patients from a nursing home had a mean LOS of 11.9 ± 8.2 days (Table 2). Figure 2 shows the distribution of LOS. On the basis of LOS, we categorized patients into 4 quartiles: the first-quartile LOS was 1 to 4 days (N = 44), the second-quartile LOS was 5 to 6 days (N = 53), the third-quartile LOS was 6 to 10 days (N = 58), and the fourth-quartile LOS was 11 to 33 days (N = 62). Univariate logistic regression was used to compare the longest LOS group with the other 3 groups and demonstrated a statistically significant increase in LOS in patients from a nursing home (OR = 2.77; 95% CI: 1.27-6.04), with pleural effusions on chest x-ray (OR = 2.85; 95% CI: 1.16-70.1), requiring intubation (OR = 3.43; 95% CI: 1.86-6.33), and with a high APACHE II score (OR = 1.06; 95% CI: 1.004-1.11). Higher initial oxygen saturations (OR = 0.94; 95% CI: 0.89-0.99) and higher albumin levels (OR = 0.47; 95% CI: 0.27-0.80) were associated with a decreased likelihood of being in the fourth quartile (Table 3). Multivariate logistic regression demonstrated that only intubation (OR = 5.93; 95% CI: 2.7812.65) resulted in a statistically significant increase in hospital LOS based on membership in the fourth quartile (Table 4).
Table 1
DISCUSSION
Abbreviations: ABG, arterial blood gas; APACHE, Acute Physiology and Chronic Health Evaluation; BiPAP, bilevel positive airway pressure; BUN, blood urea nitrogen; CXR, chest x-ray; CURB, confusion, urea, respiratory rate, blood pressure; FEV1 , forced expiratory volume in 1 second; GOLD, Global initiative for chronic Obstructive Lung Disease; PFT, pulmonary function test; WBC, white blood cell count. a Values indicate n (%) or mean ± SD.
Our patients had a median LOS of 7 days and a mean LOS of 9 days; other studies have reported median LOS of 5 to 6 days and mean LOS of 6 to 8 days.11-15 The longer LOS observed in our patients
DEMOGRAPHIC AND BASELINE CLINICAL CHARACTERISTICS OF AECOPD PATIENTSa Variable Age, y Sex Male Female Body mass index, kg/m2 PFT, tests (N) Mean FEV1 , % predicted Severity by GOLD criteria Mild FEV1 ≥ 80% Moderate 50% < FEV1 < 80% Severe 30% < FEV1 < 50% Very severe FEV1 < 30% Underlying disease Diabetes mellitus Hypertension Lung cancer Chronic kidney disease Cardiovascular accident Congestive heart failure From nursing home Mean arterial pressure, mm Hg Initial O2 saturation, % Initial blood sugar, mg/dL Hematocrit, % WBC, k/μL BUN, mg/dL Albumin, g/dL pH (ABG) Culture results: Positive blood culture Positive sputum culture CXR Present of pleural effusion Present of infiltration Intubation BiPAP APACHE II score CURB 65 score
Total 67.4 ± 10.9 102 (47%) 115 (53%) 26.7 ± 8.4 83 (38.2%) 42% ± 17.4% 2 (0.9%) 21 (9.7%) 36 (16.6%) 22 (10.1%) 62 (28.6%) 129 (59.5%) 7 (3.2%) 21(9.7%) 23 (10.6%) 53 (21.5%) 31 (14.3%) 82 ± 19.1 91.8 ± 5.3 156 ± 74.3 38.9 ± 7.1 12.8 ± 7.1 21.4 ± 16.4 3.6 ± 0.6 7.32 ± 0.11 14 (6.4%) 48 (22.1%) 22 (10.1%) 60 (27.6%) 92 (42.4%) 35 (16.1%) 13 ± 5.6 1.8 ± 1.2
Copyright © 2014 Lippincott Williams & Wilkins. Unauthorized reproduction of this article is prohibited.
COPD Acute Flares and Hospital LOS
Table 2 FACTORS AND LENGTH OF STAY BY PATIENT BASELINE CHARACTERISTICSa
Variable Age, y Overall (age 46-94) 70 (n = 83)
Median LOS, d
Mean LOS, d
7 12 7 7
9.0 ± 6.0 13.1 ± 4.8 8.9 ± 5.9 8.8 ± 6.2
Sex Male Female
7 7
9.3 ± 6.9 8.8 ± 5.1
Body mass index ≤30 (n = 148) 30 (n = 45)
7 7
8.8 ± 5.8 8.8 ± 5.3
19 5.5
19 ± 11.3 7 ± 4.1
7
8.9 ± 6.1
PFT, tests (N) Severity by GOLD criteria Mild FEV1 ≥80% (n = 2) Moderate 50% ≤ FEV1
Factors Influencing the Length of Hospital Stay in Patients With Acute Exacerbations of Chronic Obstructive Pulmonary Disease Admitted to Intensive Care Units Chok Limsuwat, MD; Charoen Mankongpaisarnrung, MD; Nat Dumrongmongcolgul, MD; Kenneth Nugent, MD Background: The cost of hospital admissions for acute exacerbations of COPD (AECOPD) accounts for 70% of total costs for the treatment of COPD patients. We wanted to identify clinical parameters associated with a longer length of stay (LOS) in these patients. Methods: We reviewed electronic medical records of patients with AECOPD admitted between January 1, 2006, and December 31, 2010. The inclusion criteria were age 45 years or older, the diagnosis of AECOPD by GOLD (Global Initiative for Chronic Obstructive Lung Disease) guideline criteria, and admission to an intensive care unit. We compared the quartile with the longest LOS group with the other 3 quartiles using routine clinical data. Results: 217 patients met inclusion criteria. The mean age was 67.4 ± 10.9 years, 47% were male, the mean FEV1 s (forced expiratory volume in 1 second) was 42.4% ± 17.4% of predicted, and the mean LOS was 9.0 ± 6.0 days. Univariate analysis demonstrated that nursing home status, low albumins, the presence of pleural effusions, intubation, and high APACHE II scores were associated with increased LOS (P < .05 for each factor). Multivariate logistic regression demonstrated that the need for intubation (P < .001) predicted an increased LOS. Conclusions: Our study demonstrates that intubation for mechanical ventilation increased the LOS in patients with AECOPD. More intensive interventions in these patients might decrease the LOS and improve outcomes. Key words: acute flare, length of stay, COPD, mortality, risk factors
C
hronic obstructive pulmonary disease (COPD) is a global problem that accounts for up to 500 000 hospitalizations per year in the United States for acute exacerbations (AECOPD).1,2 It is projected that COPD will be the third most common cause of death by 2020.3 Acute exacerbations of COPD have relatively high hospital costs, and 70% of the total cost for the treatment of COPD occurs during hospital admissions.4,5 Patients with COPD frequently have extrapulmonary disorders, such as cardiovascular disease, cancer, and depression, and these comorbidities may increase morbidity, mortality, and length of stay (LOS).6-8 Recent studies on AECOPD admissions have focused on reducing duration of admission by improving care processes and efficiency.9,10 We wanted to identify factors associated with an increased LOS in AECOPD patients who were admitted to intensive care units (ICUs). We hypothesized that underlying medical conditions, such as congestive heart failure, diabetes mellitus and cancer, nursing home residence, initial laboratory tests on admission, chest x-rays, and disease severity classification prediction scores, such as CURB 65 and APACHE II, might help physicians identify patients requiring longer LOS. Author Affiliations: Department of Internal Medicine, Texas Tech Health Sciences Center, Lubbock, Texas. Correspondence: Kenneth Nugent, MD, Department of Internal Medicine, Texas Tech Health Sciences Center, 3601 4th St, Lubbock, TX 79430 ([email protected]). The author declares no conflicts of interest. DOI: 10.1097/QMH.0000000000000024
86 Copyright © 2014 Lippincott Williams & Wilkins. Unauthorized reproduction of this article is prohibited.
COPD Acute Flares and Hospital LOS
METHODS We conducted retrospective chart reviews of patients hospitalized in University Medical Center in Lubbock, Texas, between January 1, 2006, and December 31, 2010, who were diagnosed with a COPD exacerbation and needed ICU admission. We used International Classification of Diseases, Ninth Revision codes 491.21 (obstructive chronic bronchitis with [acute] exacerbation) to identify patients. The inclusion criteria were ages 45 years or older, a diagnosis of COPD exacerbation defined by at least 2 GOLD (Global Initiative for Chronic Obstructive Lung Disease) criteria (an increase in dyspnea, an increase in frequency and severity of cough, and an increase in sputum production in terms of volume and/or changed character), and the need for ICU admission. The exclusion criteria included a history of other respiratory disease, such as asthma, or a cardiac disorder resulting in decompensated congestive heart failure.4 Using electronic medical records and case management resources, we collected the patients’ ages, genders, body mass indices, baseline pulmonary function tests, comorbidities, complete blood cell count, complete metabolic profiles, initial arterial blood gases, sputum cultures, blood cultures, chest x-rays, APACHE II scores, CURB65 scores, final diagnoses, total duration of hospital stays, and in-hospital mortality. The primary outcome was the identification of the factors associated with increased LOS. To determine this outcome, we separated patients into 4 quartiles by LOS and analyzed variable factors associated with the longest LOS quartile. Descriptive statistics were used to calculate the medians, means, standard deviations, and frequencies for various study parameters. We then used logistic regression analysis to analyze which factors associated with increased LOS. To determine factors that predicted increased LOS, we selected significant factors with P values less than .05 from the univariate analysis to analyze in the multivariate logistic regression models. Statistical analysis was performed using SPSS version 16.0; P values less than .05 were considered statistically significant. The institutional
87
review board at Texas Tech University Health Sciences Center approved the study.
RESULTS We reviewed 325 patients admitted between January 1, 2006, and December 31, 2010, to ICUs with the International Classification of Diseases, Ninth Revision code for AECOPD. However, after chart reviews, 108 patients were excluded from the study (5 deaths, 103 survivors) because these patients either did not meet inclusion criteria or met exclusion criteria. The majority of excluded patients had a concurrent diagnosis of decompensated congestive heart failure or acute asthma. Therefore, this study included 217 cases with AECOPD who needed ICU admission; 26 died during hospitalization and 191 were discharged from the hospital. The patients were separated into 4 quartiles by LOS (Figure 1). The mean age of these patients was 67.4 ± 10.9 years, and 47% were men. The mean body mass index was 26.7 ± 8.4 kg/m2 . Nearly 40% (38.2%) of the patients had baseline pulmonary function tests performed within a year of admission, and 10.1%
Figure 1. Case selection process.
Copyright © 2014 Lippincott Williams & Wilkins. Unauthorized reproduction of this article is prohibited.
88
QUALITY MANAGEMENT
IN
HEALTH CARE/VOLUME 23, ISSUE 2, APRIL–JUNE 2014
had a FEV1 s (forced expiratory volume in 1 second) less than 30% of predicted. The majority of patients had comorbidity, and 59.5%, 28.6%, and 21.5% of study population had hypertension, diabetes, and a history of congestive heart failure, respectively (Table 1). The median LOS for this patient cohort was 7 days, and mean LOS was 9.01 ± 6.00 days. The mean LOS was 9.3 ± 6.0 days in men and 8.8 ± 5.1 days in women. Patients with very severe reductions in FEV1 had a mean LOS of 10.8 ± 6.3 days, patients with diabetes had a mean LOS of 9.9 ± 6.8 days, patients with a history of congestive heart failure had a mean LOS of 9.6 ± 6.1 days, and patients from a nursing home had a mean LOS of 11.9 ± 8.2 days (Table 2). Figure 2 shows the distribution of LOS. On the basis of LOS, we categorized patients into 4 quartiles: the first-quartile LOS was 1 to 4 days (N = 44), the second-quartile LOS was 5 to 6 days (N = 53), the third-quartile LOS was 6 to 10 days (N = 58), and the fourth-quartile LOS was 11 to 33 days (N = 62). Univariate logistic regression was used to compare the longest LOS group with the other 3 groups and demonstrated a statistically significant increase in LOS in patients from a nursing home (OR = 2.77; 95% CI: 1.27-6.04), with pleural effusions on chest x-ray (OR = 2.85; 95% CI: 1.16-70.1), requiring intubation (OR = 3.43; 95% CI: 1.86-6.33), and with a high APACHE II score (OR = 1.06; 95% CI: 1.004-1.11). Higher initial oxygen saturations (OR = 0.94; 95% CI: 0.89-0.99) and higher albumin levels (OR = 0.47; 95% CI: 0.27-0.80) were associated with a decreased likelihood of being in the fourth quartile (Table 3). Multivariate logistic regression demonstrated that only intubation (OR = 5.93; 95% CI: 2.7812.65) resulted in a statistically significant increase in hospital LOS based on membership in the fourth quartile (Table 4).
Table 1
DISCUSSION
Abbreviations: ABG, arterial blood gas; APACHE, Acute Physiology and Chronic Health Evaluation; BiPAP, bilevel positive airway pressure; BUN, blood urea nitrogen; CXR, chest x-ray; CURB, confusion, urea, respiratory rate, blood pressure; FEV1 , forced expiratory volume in 1 second; GOLD, Global initiative for chronic Obstructive Lung Disease; PFT, pulmonary function test; WBC, white blood cell count. a Values indicate n (%) or mean ± SD.
Our patients had a median LOS of 7 days and a mean LOS of 9 days; other studies have reported median LOS of 5 to 6 days and mean LOS of 6 to 8 days.11-15 The longer LOS observed in our patients
DEMOGRAPHIC AND BASELINE CLINICAL CHARACTERISTICS OF AECOPD PATIENTSa Variable Age, y Sex Male Female Body mass index, kg/m2 PFT, tests (N) Mean FEV1 , % predicted Severity by GOLD criteria Mild FEV1 ≥ 80% Moderate 50% < FEV1 < 80% Severe 30% < FEV1 < 50% Very severe FEV1 < 30% Underlying disease Diabetes mellitus Hypertension Lung cancer Chronic kidney disease Cardiovascular accident Congestive heart failure From nursing home Mean arterial pressure, mm Hg Initial O2 saturation, % Initial blood sugar, mg/dL Hematocrit, % WBC, k/μL BUN, mg/dL Albumin, g/dL pH (ABG) Culture results: Positive blood culture Positive sputum culture CXR Present of pleural effusion Present of infiltration Intubation BiPAP APACHE II score CURB 65 score
Total 67.4 ± 10.9 102 (47%) 115 (53%) 26.7 ± 8.4 83 (38.2%) 42% ± 17.4% 2 (0.9%) 21 (9.7%) 36 (16.6%) 22 (10.1%) 62 (28.6%) 129 (59.5%) 7 (3.2%) 21(9.7%) 23 (10.6%) 53 (21.5%) 31 (14.3%) 82 ± 19.1 91.8 ± 5.3 156 ± 74.3 38.9 ± 7.1 12.8 ± 7.1 21.4 ± 16.4 3.6 ± 0.6 7.32 ± 0.11 14 (6.4%) 48 (22.1%) 22 (10.1%) 60 (27.6%) 92 (42.4%) 35 (16.1%) 13 ± 5.6 1.8 ± 1.2
Copyright © 2014 Lippincott Williams & Wilkins. Unauthorized reproduction of this article is prohibited.
COPD Acute Flares and Hospital LOS
Table 2 FACTORS AND LENGTH OF STAY BY PATIENT BASELINE CHARACTERISTICSa
Variable Age, y Overall (age 46-94) 70 (n = 83)
Median LOS, d
Mean LOS, d
7 12 7 7
9.0 ± 6.0 13.1 ± 4.8 8.9 ± 5.9 8.8 ± 6.2
Sex Male Female
7 7
9.3 ± 6.9 8.8 ± 5.1
Body mass index ≤30 (n = 148) 30 (n = 45)
7 7
8.8 ± 5.8 8.8 ± 5.3
19 5.5
19 ± 11.3 7 ± 4.1
7
8.9 ± 6.1
PFT, tests (N) Severity by GOLD criteria Mild FEV1 ≥80% (n = 2) Moderate 50% ≤ FEV1
