Factors associated with mortality and high treatment expense of adult patients hospitalized with chronic kidney disease in Thailand

The study was approved followed an assessment by the Ethics Committee of the Faculty of Medicine, Khon Kaen University (certificate of approval No. HE541036), following the principles of the contemporary Declaration of Helsinki. The information to be analyzed was from the three main health scheme offices in Thailand: (a) the inpatient medical expense forms for fiscal year 2010 from the National Health Security Office; (b) the inpatient data from the CSMBS (Civil Servant Medical Benefit Scheme) from the Controller General’s Department; and (c) the Social Security Office. The data collected from inpatients hospitalized with CKD included: sex, age, comorbidities, complications, treatment, clinical outcomes, and hospital charge. The data were first checked for accuracy by examining for (a) overlapping information, (b) visit dates, (c) missing items, (d) incorrect coding, and (e) the correct fiscal year. We conducted a statistical analysis of age, sex, admission rates, death rates, comorbidity, complications, and the average hospital charges. Patients with CKD were identified in both primary and secondary diagnoses as code N18 of the International Statistical Classification of Diseases and Related Health Problems, 10th revision (ICD-10) [10]. Hemodialysis and peritoneal dialysis were identified as codes 39.95 and 54.98 according to the ICD-9-CM 2010 classification of procedures [11].

The outcome data of interest included (a) number, age, and sex of patients, (b) number of admissions, (c) length of stay, (d) hospital charges, and (e) mortality rates. The characteristics of patients with CKD and factors influencing their mortality and high treatment costs were analyzed.

The statistical analyses were conducted using SPSS Statistics for Windows, version 17 (SPSS Inc, Chicago, IL, USA). The respective continuous and categorical data were expressed as means ± SD or median (25^th -75^th percentiles), and percentage. Generalized estimating equation (GEE) and multiple logistic regression analysis (MLRA) were conducted to adjust the odds ratios for factors influencing a high cost accounting for multiple admissions within an individual and mortality rate, respectively.

CKD is defined as abnormalities in the kidney structure or function or a decreased glomerular filtration rate (GFR <60 mL/min/1.73 m²) for more than three months, or both [13]. Code N18 in the ICD-10 represents an older nomenclature for chronic renal failure—a decreased GFR comparable to patients with stage 3a–5 CKD (GFR <60 mL/min/1.73 m²). Our study found the major diseases associated with CKD were hypertension and DM. The less prevalent comorbidities, albeit causing long duration of admission, high treatment cost, and mortality were pneumonia, sepsis, respiratory failure, AKI in addition to CKD, stroke and ischemic heart disease. CKD with complications are also associated with higher mortality. The degree of CKD severity increased in-hospital mortality of patients with acute coronary syndrome [14-17], heart failure [18-20], cardiac surgery [21], and stroke [22-24]. The reasons for the poor outcome specifically among these patients are: (a) CKD is itself an independent factor, (b) CKD is usually associated with multiple comorbidities, (c) the associated diseases precipitate serious complications, and (d) standard care for comorbidities are underutilized [15, 17, 19, 20]. Moreover, in-hospital costs and length of hospitalization of CKD patients represent a substantial economic burden, which is associated with comorbidity— especially of cardiovascular diseases, DM, and infections [14, 19, 20, 25]. Optimum therapeutic interventions made early, and appropriate medication, might improve clinical outcomes and reduce the cost of hospitalization.

The present study revealed that patients with CKD aged between 70 and 80 years, and older had significantly increased mortality compared with younger patients. Prognosis of elderly CKD patients depended on several factors including functional, psychosocial, and cognitive status, occurrence of malnutrition, associated comorbidities and treatments. These patients should receive comprehensive geriatric assessment and multidimensional management for improving of prognosis [26].

An important finding was that severe infection, especially pneumonia, was a leading cause of associated disease resulting in the highest hospital charges and mortality. This supports a previous finding that pneumonia is a serious complication in older patients with CKD, who are more likely to have comorbidities—especially CVD, higher pneumonia severity index, and greater mortality than patients with pneumonia, but without CKD [27]. In addition, patients with CKD are susceptible to infection because of an impaired innate immune system including impaired neutrophil and macrophage function [28]. Further study is needed to identify risk factors and organism causing pneumonia in Thai patients with CKD to determine a strategy for preventing pneumonia and reducing mortality.

The prevalence of CVD—including coronary heart disease, congestive heart failure, left ventricular hypertrophy, and stroke—are increased in patients with CKD, and progress with the severity of CKD, leading to morbidity and mortality [14, 19, 22, 24, 29]. Occurrence of CKD in patients with CVD increased related hospital cost and mortality [15, 16, 18]. The interaction of cardiac and renal functions has been established and designated as cardio–renal syndrome [30]. The relationship between the heart and kidneys is bidirectional and mediated by several mechanisms: namely, hemodynamic change, hormonal effects, activation of the sympathetic nervous system and renin–angiotensin–aldosterone system, metabolic derangement, malnutrition, and inflammation [31, 32]. Moreover, common risk factors and complications of CKD (i.e., hypertension, diabetes mellitus, dyslipidemia, albuminuria, anemia, hyperkalemia, and volume overload) aggravate CVD. Therefore, treatment of these factors will improve CVD outcomes.

Data presented in the Tables presents the characteristics of patients with CKD with and without AKI, demonstrating that patients with CKD patients with AKI had significantly more comorbidity, CKD complications, dialysis treatment, length of hospital stay, hospital charges, and mortality. These findings confirm those found in other studies [33-35]. AKI and CKD have a bidirectional relationship; such that AKI is associated with progression to CKD and preexisting CKD increases the risk of AKI [36-38]. Therefore, more effort should be made to prevent AKI in patients with CKD.

Early and appropriate treatment of complications of CKD among outpatients may reduce the chance of admission. For example, receiving an adequate dose of erythropoiesis-stimulating agent and iron supplement decreases the need for blood transfusion and hospitalization [39, 40]. Patient education regarding limiting high potassium and salt intake and proper adherence to medication might reduce the occurrence of hyperkalemia and volume overload requiring hospital admission.

A strength of this study is that almost all Thai adult patients hospitalized with CKD were included. Therefore, the results should provide an overview of the situation. Notwithstanding, the lack of a registered nationwide laboratory system means that there is no defined staging of CKD in patients in the present study. However, because of the retrospective nature of the study design, some data might not be complete. The record of hospital charges for each group represents an average, which might not wholly characterize the severity of individual patients, nor include details of the procedures and medical instruments needed for each patient.

Pneumonia, sepsis, respiratory failure, AKI in addition to preexisting CKD, stroke, and ischemic heart disease were the major diseases associated with high hospital charges and mortality of CKD patients. Prevention and early treatment of these comorbidities and CKD complications should reduce the poor outcome of, and treatment costs for, patients hospitalized with CKD.