Early readmission and its predictors among patients treated for acute exacerbations of chronic obstructive respiratory disease in Ethiopia: A prospective cohort study

Background Significant numbers of chronic obstructive respiratory disease patients are readmitted for Acute Exacerbation (AE) within 30 days of discharge. And these early readmissions have serious clinical and socioeconomic consequences. The objective of our study was to determine the rate of readmission within 30 days of discharge and it’s predictors among patients treated for acute exacerbations of asthma and chronic obstructive pulmonary disease (COPD). Methods A prospective cohort study involving 130 patients (asthma = 59, COPD = 71) was conducted from April-September, 2019, in Jimma Medical Center (JMC), South-West Ethiopia. Socio-demographic, clinical, laboratory, and drug-related data were recorded at admission and during hospital stay. Cox regression analysis was performed to identify risk factors for readmissions following an AE of asthma and COPD. Results During the study period, 130 (male, 78(60%)) patients were admitted with AE of asthma and COPD. The median age was 59(IQR, 50–70) years. Of 130 patients, 21(18.10%) had a new AE of asthma and COPD that required hospitalization in the 30 days after discharge. The overall median survival time to 30-day readmission was 20 days (IQR, 16–29). Multivariate analysis revealed prolonged use of oxygen therapy (AHR = 4.972, 95% CI [1.041–23.736] and frequent hospital admissions (AHR = 11.482 [1.308–100.793]) to be independent risk factors for early readmissions. Conclusion Early hospital readmission rates for AE of asthma and COPD were alarmingly high. Frequent hospital admission and long-term oxygen therapy during hospital stay were independent predictors of 30-day readmission.

determine the rate of readmission within 30 days of discharge and it's predictors among asthma and COPD patients presenting with acute exacerbations.

Study design and setting
This prospective cohort study was conducted at a University hospital in Ethiopia (Jimma Medical Center (JMC)) over a period of 6 months starting from April-September, 2019. JMC is located in Jimma town, 355 km from Addis Ababa, capital city. It is one of the oldest public hospitals found in the South -Western part of the country that runs under Jimma University. It is currently the only medical center in this part of the country. This medical center gives a service for a catchment's population of 15 million and has more than 800 beds, 1600 staff members. It has several specialty clinics. As one of the specialty clinics, chest clinic has outpatient department (OPD) and pulmonology unit, which was the major source of patients for this study.

Study population and variables
We included all adult (age>18 years) patients with acute exacerbations of asthma and COPD admitted to medical wards, ED, and ICU of JMC during the study period. Cardiac asthma, Asthma-COPD overlap, asthma or COPD misdiagnosed patients and patients unwilling to participate in the study were excluded.
Data were collected on patient-related factors (socio-demographic, body mass index (BMI)), social drug use (smoking status), medication adherence, living with pet animals), disease-related factors (laboratory findings, clinical presentation, severity ofasthma and COPD at admission, duration of AE of asthma and COPD, and co morbidities), and drug related factors (in-hospital medications, past medication history, medications at discharge). The lung function test (Forced expiratory volume (FEV 1 ), Forced vital capacity (FVC), Peak expiratory flow rate (PEFR), and FEV 1 /FVC) was performed by spirometryV1.34:CAR36ML3500SR4 (brand: carefusion-microlab3500 spirometry). Laboratory results of the first blood test in index admission (hemoglobin, white blood cell count (WBC), creatinine, blood urea nitrogen (BUN)) were collected. Complete blood counts (CBC) were measured using hematologic analyzers: XT-1800i (Sysmex, Japan), KX-21 N™ (Sysmex, Japan), and Cell-Dyn 18001(Abbot, USA). Blood tests including serum creatinine (Cr), blood urea nitrogen (BUN) were analyzed using chemistry analyzers: ABX Pentra 400(Horiba, USA), Dirui DR-7000D (DIRUI, Changchun, China) and HumaLyzer 3000 (HUMAN, Wiesbaden, Germany). The severity of asthma or COPD disease was evaluated using the global initiative for asthma (GINA 2019) and global obstructive lung disease (GOLD 2019) classification, respectively [23,24]. Adherence to asthma and COPD medications inhalation techniques were assessed by passive observation of patient's ability to demonstrate the10 steps of inhalation technique, the validated questionnaire widely used in chronic obstructive diseases.

Outcome measurement and method of validations
Both Spirometry measurements and inhalation techniques were assessed after patients became clinically stable without respiratory distress. Before spirometry measurements, patients were asked to omit short-acting inhaled bronchodilators for 4-6 hours, and long-acting oral and inhaled agents for 12 hours. Spirometry was done initially then Salbutamol 400mcg inhaler administered in four separate doses and spirometry was repeated after 15minutes [24].
Patients were diagnosed as AECOPD, if the post bronchodilator (400mcg of SABA in our setting) peak flow measurement after 15 minutes was increased by less than 12% or 200ml from baseline of PEF or FEV 1 (pre-bronchodilator) or evaluated FEV 1 was less than or equal to 80% predicted from age, height, and sex, and FEV 1 /FVC ratio was less than 0.7, plus clinical feature of dyspnea, chronic cough and sputum productions [24].
Patients were diagnosed as AE of asthma if the post bronchodilator (400mcg of SABA) peak flow measurement after 15 minutes was increased by 12% or 200ml from baseline of PEF or FEV 1 (pre-bronchodilator) and FEV 1 /FVC ratio was less than 0.7 and oxygen saturation�90, plus clinical feature of cough, wheezing shortness of breath and presence of risk factors [25]. Finally, 30-day readmission information was obtained when readmitted to the hospital or through telephone from the family or the patient itself.

Sample size determination and sampling technique
Since the number of patients encountered during the study period was very limited, we included all patients who fulfilled the eligibility criteria and there is no special technique employed.

Data collection tool and procedure
The data collection tool was developed based on previous similar studies [11,14,16,[26][27][28][29][30] and active patient follow-up chart. Data were collected both from active patient follow-up chart and patient interview.

Data quality assurance
A carefully designed data collection tool was used to collect important data required to meet the stated objectives. Prior to the actual data collection process, pre-test was conducted on 10 randomly selected eligible patients and the data collection tool was modified accordingly. Four data collectors (two pharmacists with Bachelor of Pharmacy degree and two clinical nurses with Bachelor of Science degree) and four supervisors (general practitioners) were hired and two-day training on the data collection tool and general procedures was provided by the principal investigator. The supervisors coordinate data collectors and facilitate the daily activities. All filled checklists were reviewed for completeness and consistency on a daily basis by supervisors and principal investigator.

Statistical analysis
Data were entered into Epi-Data 4.02.01 for cleaning and exported to STATA 14 for analysis. Descriptive analysis was performed and results were presented by text, tables and figures. Survival estimates for 30-day readmission was checked by Kaplan-Meier (log-rank test). Multicolinearity test was performed to check for co-linearity between independent variables. X 2 test was performed to check adequacy of cells before performing Cox regression. Cox regression model assumption of proportional hazards was checked by testing an interaction of covariates with time. Bivariate Cox regression was performed to identify candidate variables for multivariable Cox regressions. Variables with p-value < 0.25 in bivariate regression were considered as candidates for multivariable Cox regression. Multivariable Cox regression was performed to identify independent predictors of 30-day readmission. Hazard ratio was used as a measure of strength of associationand p-value < 0.05 was considered todeclare statistical significance.

Ethical approval and consent to participate
The study was approved by the Institutional Review Board (IRB) of Jimma University. It has designated with an IRB number of IHRPGD/549/109. After informing the overall concern of the study and confidentiality of their personal information, participants were requested for verbal consent; for those patients unable to speak, consent was requested from their relatives. For those who could write and read, written consent was taken.

Operational definition of terms
Active smokers: refers to patients smoking a cigarette in the past 3 months before the current hospital admission [31].
Adherence (inhalation techniques): is the degree to which known asthma and COPD patients' medication-taking behavior (inhalation techniques steps) corresponds with agreed recommendations from a healthcare provider. Good adherence was defined as an average adherence to study medications of >80% over the whole period the subject was in the study except critical steps. Poor adherence (not adhere to inhalation techniques) was defined as average adherence to inhaled medication �80% including critical steps or missed at least one critical steps [32].
Critical step: procedure if it is missed which leads to poor asthma control. These are place mouthpiece between teeth and lips, simultaneously press canister and breathing in slowly and inhaler out of mouth and hold breathe for 5-10 sec [33].
Efficient: Refers when patients can perform all critical steps regardless of non-critical steps [34].
Inefficient: Refers when patients miss any one or more critical steps regardless of performance of non-criticalsteps [34].
Index admission: The first occurrence of the patient's hospital encounter for asthma or COPD within the study period [35] and inpatient stay discharged alive with no missing length of stay (LOS) [36].
Readmission is an inpatient stay that begins within 30 days of the discharge date of an index admission.
The readmission rate is the percentage of index admissions that are readmitted within 30 days [36].
30-day readmission timing: any unplanned consecutive admission within 30 days (1-30) after being discharged alive from a hospital stay where asthma or COPD was a principal diagnosis during the study period [35].
Spirometer: the apparatus commonly used to measure the volume of air exchanged during breathing and the respiratory rate.
Forced vital capacity (FVC): the maximum volume of air in the lung after a forceful exhalation following forceful inhalation.
Forced expiratory volume (FEV 1 )-the volume of air expired forcefully in one second. Peak expiratory flow rate (PEFR)-a maximum expiratory air flow beyond which the flow cannot be increased any more even with greatly increased additional force. FEV 1 /FVC (FEV 1 %): the percentage of the FVC expired in the first second of maximal forced expiration following full inspiration. Predicted values greater than 80% is usually considering as normal.
Chronic obstructive respiratory diseases: are obstructive lung diseases that include either asthma or COPD (ICD-10, 2019).
Long-term oxygen therapy: use of oxygen for greater or equal to 16hrs per days and for multiple days (GOLD-2019).

Characteristics of study cohort
In total, 162 asthma/COPD adult patients were admitted to hospital for acute exacerbations. Twelve patients were excluded from enrolment because of not meeting inclusion criteria and not willing to participate. During the follow-up period, 20 patients were excluded due to loss to follow-up, 4 patients, 14 patients not obstructive, and two patients lung function test unknown. One hundred thirty patients met the eligibility criteria (asthma = 59, COPD = 71) and included into the final analysis. Eligible participants were followed for a median period of 5 days (IQR, 3-14 days). (Fig 1).

Clinical characteristics
Of total patients,111(85.38%) had history of previous chronic obstructive respiratory disease with the median duration of 3 years (IQR, 2-6 years) for COPD, and 10 years (IQR, 4.5-19 years) for asthma. The overall median length of hospital stay was 5 days (IQR, 3-14 days). The median length of hospital stay for acute exacerbation of asthma was 3 days (IQR, 3-6 days); while for acute exacerbation of COPD was 8 days (IQR, 4-19 days). Seventy-nine (60.77%) patients had history of hospital admission and nearly half, 38(48.10%) of them had two admissions per annum. Eight (6.15%) patients had history of ICU admission (asthma = 4, COPD = 4).
Spirometer results for pre and post bronchodilator. On admission, lung function test was performed for all candidate patients. The FEV 1 /FVC was <0.7 in 130 patients showing obstructive pattern. The relationship between pre-bronchodilator values and bronchodilator response was estimated and declared as acute exacerbation of asthma or COPD (severity).

Measurements of inhalation techniques adherence
For all eligible patients, 10 step-wise inhalation techniques were provided to check the adherence of effective inhalations. Fourteen (10.78%) patients totally could not perform inhalation techniques correctly including the three critical steps and declared as "inefficient". The remaining 116(89.23%) patients missed at least one steps, except the three critical inhalation techniques and declared as "efficient". As depicted in Fig 4, 114 patients performed step one correctly (blue color), 16 patients cannot perform correctly (red color), and so on. As steps goes up, the ability to remember and perform effectively was decreased i.e. 91(70%) could not perform correctly the last steps (exhale and wait for 30-60 seconds before the other inhaler) (red color). (Fig 4).

Predictors of 30-day readmission
Several variables such as history of hospital admission, use of high dose salbutamol inhaler, chest clinic physician visit, occupational status, residence, and other four variables (age, BMI, smoking status, and educational status) were omitted from analysis due to violation of cell adequacy test and two variables (comorbidity and living status) removed due to co-linearity effect. Therefore, nine variables were candidates for bivariate Cox-regression analysis (p<0.25) depicted in Table 5. Further multivariate Cox-regression identified prolonged use of oxygen therapy and frequent hospitalization per yearto be independent predictors of 30-day readmission.
Accordingly, patients who were on oxygen therapy for �16hrs per 24hrs had 5 times increased hazard of 30-
In this study, one hundred sixteen (89.23%) patients were adherent (efficient) to inhalation technique. This adherence rate is relatively better than the study conducted by Kebede et al, 87 (62.14%) [28]. In a study conducted in Turkey, nearly 23% of the patients had suboptimal adherence to inhalation technique [43]. Difference in the methodology used to measure adherence level might contribute to the variation. Furthermore, the direct involvement of clinical pharmacists in the patient care process and regular provision of counseling on medication use in our set up may contributed for better adherence. Although we have evaluated a wide range of risk factors for hospital readmission, only two clinical factors were found to be independent predictors of 30-days readmission in the current study. In our study, patients who were on oxygen therapy for �16hrs per day had 5 times increased hazard of 30-day readmission (AHR = 4.972.95% CI [1.041-23.736]. A number of previous similar studies [39,41], have also reported long term oxygen therapy as a risk factor associated with an increased risk of readmission among patients with chronic obstructive airway diseases. Individuals who stay longer on oxygen therapy likely have severe and advanced disease and are at higher risk for hospital readmission. Furthermore, patients who admitted to the hospital three times or greater in the previous year had 11 times increased hazard of 30-day readmission (AHR = 9.5, 95% CI [1.12-80.16]. This finding is in agreement with several other studies [40,42,44,45]. According to Guerrero M et al., [45], the rate of 30 days readmission was significantly higher in COPD patients with a history of >2 exacerbations in the previous year. Frequent hospital admission is an indicator of disease severity and patients with advanced disease are very prone for readmission.
No association was found between FEV 1 % predicted and the risk of asthma and COPD readmission. This finding supports the result of a previous study by Bahadori et al., [29] that found no differences between FEV 1 % predicted and the risk of COPD readmission. However, other studies [27,30,46] found FEV 1 <50% predicted to be predictive of a higher risk of 30-days readmission. The possible justification for the lack of association in the current study might be related to the small sample size, which reduces statistical power to find true association.
The prevalence of COPD/asthma in Ethiopia is comparable with neighboring Sub-Saharan African countries. A prevalence rate of 17.6% was reported for COPD [47] and this is in-line with studies done in Uganda [48] and Tanzania [49]. The prevalence of asthma was reported to be 29.6% [50]. This prevalence was relatively higher compared to the study conducted in Egypt [51], and Uganda [52]. Differences in climate conditions, growing population size, and radical urbanization might contribute to the increment in the prevalence of asthma.

Limitations
The findings of the present study should be interpreted in the context of several possible limitations. First, inclusion of a small number of patients might under power the study. Second, follow-up period of 6 months may be relatively short, but we focused on early 30-days readmission. Third, merging two populations of obstructive diseases may affect the validity of the outcome. Moreover, salbutamol alone was used to test the lung functions in the study due to unavailability of ipratropium, as most of bronchodilator reversibility studies were conducted using the combinations of short-acting beta agonists and anti-muscarinic. Finally, the present study was conducted in a single institution, and so the extrapolation of these findings to other settings must be done with care.

Conclusion
In conclusion, hospital readmission rates for AE of asthma and COPD within 30 days of discharge were alarmingly high. Frequent hospital admission and long-term oxygen therapy during hospital stay were independent predictors of 30-day readmission.