1. Background

Veterans hospitals are public medical institutions established to provide healthcare services for patriots and veterans who made sacrifices for our country and their families, and six veterans hospitals including Seoul VHS (Veterans Health Service) Medical Center are being operated in six regions of Korea. Veterans hospitals have an established systematic healthcare delivery system that allows patients to receive primary care at consigned medical institutions close to home, secondary care at veterans hospitals in 5 noncapital areas, and tertiary care at Seoul VHS Medical Center equipped with specialized healthcare facilities, and provides patient-centered integrated health and welfare services through cooperation among veterans hospitals [1]. The integrated health and welfare services aim to provide services related to the continuum of care consisting of prevention, treatment, rehabilitation, long-term care, home-based care, and end-of-life care by organically connecting them with community-based services through the interconnection of the medical and welfare services of the Korea Veterans Welfare and Medical Care Corporation [2].

In Korea, the population aged 65 and over accounts for 14.7% of the total population as of 2018, indicating that Korea has become an aged society [3]. The majority of the beneficiaries of benefits and services for patriots, veterans, and their families (henceforth MPVA (Ministry of Patriots and Veterans Affairs) beneficiaries) are wartime veterans that engaged in combat during a war. MPVA beneficiaries aged 65 or older are reported to be more than 0.5 million, and account for 73.9% of the total MPVA beneficiaries as of 2019 [4], and 93.4% of them are reported to be diagnosed with chronic diseases [5]. Elderly people use medical services including emergency rooms and hospitalization due to chronic diseases and deterioration in physical function [6], and long-term care due to these reasons results in an increase of hospital length of stay [7]. In elderly patients aged 65 or older among the users of health services covered by National Health Insurance, frequently diagnosed diseases of hospitalized patients are mostly acute diseases such as fractures [7]. However, in veterans hospitals, the most frequently diagnosed diseases of inpatients are chronic diseases [5]. The mean hospital length of stay of veterans hospitals is reported to be 35.3 days [5], which is longer than the mean hospital length of stay of 9.3 days in general hospitals [7]. The mean hospital length of stay of Seoul VHS Medical Center, a tertiary hospital, was found to be 17.6 days as of 2018 according to the results of an investigation by Seoul VHS Medical Center, and this value is also higher than the mean hospital length of stay of ordinary general hospitals [5].

Meanwhile, the management of length of stay in medical institutions can improve the financial conditions of hospitals by increasing the bed turnover rate, and can lead to the reduction of medical expenses and the improvement of quality of life through the use of outpatient treatment or home nursing care service after acute treatment [8]. The increase of the bed turnover rate makes it possible to rapidly provide inpatient care for patients waiting for hospital admissions and requiring acute treatment and to efficiently utilize limited medical resources. However, most users of veterans hospitals want to continue to stay in the hospital due to the benefits of the exemption or reduction of medical fees and good medical facilities and services [1]. For these reasons, veterans hospitals operate long-term care hospitals and nursing homes or operate hospitals by converting ordinary hospital beds into ones for long-term care to manage elderly patients with chronic diseases, and they are also establishing an integrated medical and welfare service system [2]. Koo et al. [2] reported that integrated medical and welfare services of veterans hospitals resulted in a lower hospital length of stay in the recipients of the services than nonrecipients, and the study attributed this result to the fact that recipients of the services become able to manage chronic diseases through the use of home nursing care service or visiting rehabilitation service or through the systematic use of outpatient services according to the prescription of physicians. Nevertheless, the majority of inpatients of veterans hospitals are patients with chronic diseases who need long-term care. According to a previous study of patients in Daegu VHS Medical Center, the mean hospital length of stay during the period of the past 6 months was 83.3 days in patients exempt from medical fees and 55.8 days in patients paying reduced medical fees due to government subsidy [9]. The study also reported that the influencing factor for hospital length of stay was the presence of out-of-pocket medical expenditure regardless of the degree of assistance in activities of daily living, and hospital wards with patients exempt from medical fees were practically being operated like long-term care wards [9]. In foreign healthcare facilities for veterans, home-based care services, such as visiting medical services [11] and telehealth which manages health status using traditional and digital communication devices [12], are being operated for the health management of elderly patients with chronic disease, and hospital use is mainly focused on acute treatment and medical expenses are managed through the case management of patients [13]. With respect to prolonged hospital length of stay in Korean veterans hospitals, patients' intention for discharge is important among patients exempt from medical fees unlike cases in foreign countries. Therefore, it is necessary to establish a linkage system for linking patients to community-based care services by identifying community-linked integrated medical and welfare services needed by patients after discharge.

In Korea, although the community care pilot project is being implemented, elderly patients discharged from hospitals have difficulty in settling in the community due to the lack of various care services including visiting treatment, visiting rehabilitation, nutritional support, and residential environment improvement that can provide support needed for elderly patients discharged from hospitals to return to the community [14]. Although veterans hospitals are also making efforts to establish an integrated medical and welfare care service system, it has been suggested that there is a need for the control function for veterans hospitals, welfare institutions, and administrative welfare centers as well as linkage between them and specialized institutions. It has also been proposed that it is necessary to establish a community care model for veterans medical and welfare services to ensure the continuous operation of home-based veterans welfare services in the community social security systems [2]. Thus, this study aimed to investigate factors affecting hospital length of stay in elderly patients using veterans hospitals, identify community-linked integrated medical and welfare services needed by the participants after discharge in order to present basic data useful for the management of hospital length of stay and the future development and operation of intervention programs that utilize the accessibility of community resources.

2. Purpose

The specific objectives of this study are as follows:

• To investigate the general and disease-related characteristics of elderly long-term inpatients with chronic diseases of veterans hospitals, and examine differences in hospital length of stay according to these characteristics;

• to examine the levels of assistance of ADL, depression, anxiety, and hospital satisfaction as well as the relationship between each of these variables and hospital length of stay among elderly long-term inpatients with chronic diseases of veterans hospitals;

• to identify factors affecting hospital length of stay among elderly long-term inpatients with chronic diseases of veterans hospitals;

• to examine community-linked integrated medical and welfare services needed at discharge among elderly long-term inpatients with chronic diseases of veterans hospitals.

1. Study Design

This study is a descriptive correlational study to investigate the general and disease-related characteristics of elderly long-term inpatients with chronic diseases using veterans hospitals and their needs for community-linked integrated medical and welfare services needed after discharge, and identify factors related to prolonged hospital length of stay in the participants.

2. Participants

The participants were patients with chronic disease hospitalized in veterans hospitals in Korea. Based on the mean hospital length of stay of 9.26 days in the inpatients of general hospitals in Korea as of 2017 [7], the participants were selected among elderly patients aged 65 years or older and hospitalized in veterans hospitals for 10 days due to chronic disease, and they were selected according to the following criteria. The participants were limited to persons who were not staying in the hospital for rehabilitation treatment, antibiotic therapy, chemotherapy, or special procedures, and could be discharged since they were in a sufficiently stable health condition to allow the patients to receive their current medical services from community facilities linked to veterans hospitals. In addition, only persons who were provided with explanations about the purpose and contents of the study and risks involved, fully understood them, and signed the written informed consent form were included in this study. The sample size was calculated using the G*Power 3.1.9 software, and it was determined to be 184 persons with the significance level (α) of .05, power (1-β) of .95, the median effect size of .15, and 12 independent variables. Thus, a total of 250 questionnaires were distributed by distributing 100 copies at Seoul VHS Medical Center and 30 copies at each of the veterans hospitals in noncapital areas because a high dropout rate of the questionnaires was expected as a result of conducting a survey at 6 hospitals. A total of 232 questionnaires were collected, and they included of 93 copies from patients of Seoul VHS Medical Center and 139 copies from patients of veterans hospitals in noncapital areas. Among them, a total of 196 copies of questionnaires were used for the final analysis, excluding 6 copies of the participants aged under 65 and 30 copies of patients with the present hospital length of stay of less than 10 days.

3. Data Collection

Data was collected from July 18, 2019 to August 10, 2019, and data collection was conducted in six veterans hospitals located in Seoul, Busan, Kwangju, Daegu, Daejeon, and Incheon. Regarding the method of data collection, a questionnaire survey was conducted among the patients of 6 veterans hospitals in Korea to collect data on general characteristics, the degree of assistance of activities of daily living (ADL), hospital satisfaction, desired hospital length of stay, and needs of community-linked integrated medical and welfare service. In the case of disease-related data, the nurse in charge of data collection of each hospital collected disease-related data through the review of medical records of the participants. Data collection was carried out by sending questionnaires to the nurse in charge of data collection of each hospital after the researcher explained the purpose, methods and procedure of the study to the hospital director and the head of the department of each hospital and obtained their approval and promise to cooperate. The contents of the questionnaire included the purpose and methods of the study and contents about data collection. The nurse in charge of data collection was designated by each hospital, and the nurse provided explainations about the purpose and methods of the study and the guarantee of confidentiality of the data to the participants that met the participant inclusion criteria. During these explanations, the nurse showed the participant information sheet to the participants. She also explained to the participants that disease-related characteristics would be examined using medical records, the participants are allowed to withdraw from the survey and this study at any time if they want to, and the study results would be used only for the purpose of research. Then, after obtaining written informed consent from the participants, questionnaires were distributed to the participants to collect data. Although disease-related data was collected by the nurse in charge of data collection through the review of medical records of the patients, unclear contents of medical records were examined by asking attending physicians about them. In addition, in order to prevent errors in the interpretation of each item (general characteristics, disease-related characteristics, assistance of ADL, anxiety and depression, hospital satisfaction, desired hospital length of stay, and community-linked integrated medical and welfare service) and composition of the questionnaire, the nurse in charge of data collection of each hospital and the researcher met together before data collection to examine each item and discuss problems and questions about each item.

4. Measures

5. Data Analysis

The statistical analysis of the collected data was performed using SPSS/WIN 24.0 program. The general and disease-related characteristics and community-linked integrated medical and welfare services needed after discharge of the participants were analyzed using descriptive statistics, such as real numbers, percentages, mean and standard deviation. In addition, the t-test or ANOVA was used to analyze differences in hospital LOS according to general and disease-related characteristics, and relationships between variables was analyzed using Pearson's correlation coefficient. Also, factors related to hospital LOS were analyzed using stepwise multiple linear regression. For variables that were found to be significant by ANOVA, post-hoc analysis was performed using the Scheffé test.

6. Ethical Considerations

Data collection was conducted after receiving approval from the IRB of central verterans hospital (2019-03-033-004) about the study purpose, methods, the participant information sheet, and the informed consent form. To increase the reliability of the described contents of the questionnaire, the participants were instructed to respond to the questionnaire anonymously, and put each copy of the completed questionnaires in an individual document envelope and seal the envelope to prevent the survey data from being exposed to others. The participants were given the explanation that the survey data as well as data obtained from medical records would be used only for the purpose of research, the personal information in the questionnaires would be encrypted and stored in a separate file of a locked computer to keep personal information confidential, and the questionnaires would be kept in a locked document box and shredded and discarded 3 years after the completion of this study.

The time taken for the completion of the questionnaire was about 5 minutes for nurses in charge of data collection and about 15 minutes for patients. For the participants who were unable to read and understand Korean, the nurse in charge of data collection read the contents of each item of the questionnaire for them and checked responses to each question. After the questionnaire was completed, each participant was given a small gift as a token of appreciation.

1. General Characteristics and Hospital Length of Stay (LOS) according to These Variables

The present hospital LOS and desired LOS of the participants were 37.78±32.66 days and 60.87±45.95 days, respectively. With respect to general characteristics, for gender, males accounted for 90.3% of the total participants. As for educational level, middle school graduates or lower accounted for 53.1%. In subjective economic status, 67.7% rated their subjective economic status as poor. In addition, 67.7% had no religious belief and 94.9% were married in marital status. Also, 90.3% had no occupations, and the main caregiver was the spouse in 67.3%. Regarding payment of medical fees, 62.9% were completely exempt from medical fees. As to the region of the hospital, 58.2% of the participants were inpatients of hospitals in non-capital areas (Table 1).

Table 1 General Characteristics and Differences in Hospital Length of Stay according to them (N=196) Click for larger image

Regarding differences in the length of hospital stay according to general characteristics, there was a significant increase in the desired length of hospital stay when the main caregiver was the spouse (t=2.48, p=.014), when patients were exempt form medical fees (t=2.40, p=.018), and when the hospital was located in a noncapial area (t=2.78, p=.006) (Table 1).

2. Disease-related Characteristics and Hospital Length of Stay (LOS) according to These Variables

With respect to disease-related characteristics, the number of chronic diseases was 3 or more in 30.1%, and the mean number of diseases was 2.9. Regarding present therapeutic diseases, the proportion of cardiovascular disease was highest at 59.7%, followed by endocrine and metabolic diseases including diabetes, respiratory disease, and genitourinary disease. As to therapeutic aspect at admission, treatment provided to the patient at admission was acute treatment in 35.7% of participants, symptom control requiring medical facilities in 30.1%, and symptom control feasible at home in 30.1%. Meanwhile, present therapeutic aspect was acute treatment in 35.2%, symptom control requiring medical facilities in 18.9%, and symptom control feasible at home in 45.9%, and patients that received discharge recommendation accounted for 31.6% (Table 2).

Table 2 Present Disease-related Characteristics and Differences for Hospital Length of Stay according to Them (N=196) Click for larger image

Table 2 shows disease-related characteristics affecting hospital LOS. Hospital LOS was higher in the group with two or chronic diseases than in the group with one chronic disease (F=2.51, p=.025). In addition, hospital LOS was higher when the primary disease is a genitourinary disease than when it was other kinds of disease (t=3.21, p=.002). It was also higher in the group with complications than in the group without complications (t=2.02, p=.048). As for desired hospital LOS, regarding therapeutic aspect at admission, desired hospital LOS was highest in the group with the feasibility of symptom control at home, followed by the group without the feasibility of symptom control at home and the acute treatment group (F=3.12, p=.047). In terms of the present therapeutic aspect, desired hospital LOS stay was highest in the group with the feasibility of symptom control at home, followed by the acute treatment group and the group without the feasibility of symptom control at home (F=3.34, p=.038). In addition, desired hospital LOS was significantly longer in the group with recommendation of discharge from medical staff than the group without recommendation of discharge (t=2.04, p=.043).

3. Assistance of ADL, Anxiety, Depression, Hospital Satisfaction, and Relationships between These Variables and Hospital Length of Stay (LOS)

In the participants of this study, the level of assistance of ADL was 5.38±7.28 out of 24 points, the levels of anxiety and depression were 1.00±0.68 out of 3 points and 1.34±0.81 out of 3 points, respectively. The level of hospital satisfaction was 8.18±1.76 out of 10 points. Hospital length of stay showed a significant but weak relationship with the degree of assistance of ADL and depression (Table 3).

Table 3 Assistance of ADL, Anxiety, Depression, Hospital Satisfaction, and Their Correlations with Hospital Length of Stay (N=196) Click for larger image

4. Factors affecting hospital length of stay (LOS) in elderly Long-term Inpatients with Chronic Diseases

To identify specific predictor variables affecting hospital LOS in elderly long-term inpatients with chronic diseases, variables related to hospital LOS among general and disease-related characteristics were set as independent variables. More specifically, general characteristics such as main caregiver, payment of medical fees, and area of the hospital and disease-related characteristics such as the number of chronic diseases, therapeutic aspect at admission, present therapeutic aspect, and presence of genitourinary illness were used as independent variables. In addition, the degree of assistance of ADL, anxiety, depression, and hospital satisfaction, which were presumed to be related to hospital LOS, were also set as independent variables, and these variables were entered and analyzed in a hierarchical manner. Variables measured on a nonimal scale (main caregiver, area of hospital, payment of medical fees, presence of chronic genitourinary disease, and therapeutic aspect) were analyzed by treating them as dummy variables, and reference values were set as 'the non-spouse caregiver for the main caregiver', 'Seoul for the region of hospital', 'reduction of medical fees for payment of medical fees', 'non-genitourinary diseases for the main disease', and 'acute treatment for therapeutic aspects.' Tests for checking the assumptions of regression analysis were performed before conducting stepwise regression analysis, and the results showed that the assumptions were all satisfied. The Durbin-Watson test was used to test autocorrelation for the dependent variable, and the Durbin-Watson value was determined as 1.64, indicating that there was no autocorrelation. To assess multicollinearity between independent variables, the variance inflation factor (VIF) was calculated, and the VIF value ranged from 1.00 to 1.19, values less than 10, indicating that there is no multicollinearity between independent variables. As a result, it was confirmed that the data of this study was suitable for regression analysis. In addition, residual analysis was conducted to confirm the suitability of the model, and the results showed that the assumptions of normality and equal variance were satisfied. The factors related to desired hospital LOS in elderly long-term inpatients of veterans hospitals were found to be urologic disease, assistance of ADL, area of hospital, payment of medical fees, and hospital satisfaction. The explanatory power of each factor was as follows: urologic disease (β=.30, p<.001), assistance of ADL (β=.34, p<.001), region of hospital (β=.31, p<.001), type of payment of medical fees (β=.15, p=.026), and hospital satisfaction (β=-.14, p=.036). The adjusted explanatory power of these variables was 26.4% (F=13.30, p<.001) (Table 4).

Table 4 Factors related to Estimated Hospital Length of Stay (N=196) Click for larger image

5. Needs of Discharge-related Community-linked Integrated Medical and Welfare Services

Table 5 shows the needs of community-linked integrated medical and welfare services in relation to discharge among elderly patients with chronic diseases. As for opinions of medical staff on the directions of the management of chronic diseases needed for discharge of the participants, alleviation of symptoms (56.2%) accounted for the largest proportion of the responses, followed by simple health management (16.5%). 52.0% of respondents answered that their discharge would be possible if they were able to receive medical services linked to community integrated medical and welfare services. Among 102 respondents that expressed the possibility of discharge, regarding the type of care needed after discharge, 55.9% wanted to receive facility-based care, and the most needed type of facility was long-term care hospitals, followed by rehabilitation institutions, and nursing homes. 41.0% wanted to receive visiting care, and in the case of visiting care, the largest proportion of patients needed home nursing care. In response to the question about the necessity for community-linked integrated healthcare after discharge, 90.8% of respondents answered positively. Regarding the needs of community-linked integrated medical and welfare services, 67.4% of patients wanted home nursing care. As to the contents of home nursing service, the proportion of observations of the disease state was highest at 83.3%, and other services needed by at least 10% of patients were infusion or injection therapy (29.2%), counseling (15.9%), rehabilitation treatment (14.2%), and wound care (13.7%). Some participants needed transportation services for dialysis, blood transfusion or abdominal paracentesis. Regarding self-management for chronic disease management, 55.7% of the total participants did not have the experience of self-management education, and 67.5% did not have any intention to receive self-management education, constituting a larger proportion than the group with an intention to receive self-management education (Table 5).

Table 5 Discharge related Opinion and Requiring Community Linked Medical-Welfare Service after Discharge (N=196) Click for larger image