Factors Affecting Life-Sustaining Treatment Decisions and Changes in Clinical Practice after Enforcement of the Life-Sustaining Treatment (LST) Decision Act: A Tertiary Hospital Experience in Korea
Article information
Abstract
Purpose
In Korea, the Act on Hospice and Palliative Care and Decisions on Life-Sustaining Treatment (LST) was implemented on February 4, 2018. We aimed to investigate relevant factors and clinical changes associated with LST decisions after law enforcement.
Materials and Methods
This single-center retrospective study included patients who completed LST documents using legal forms at Asan Medical Center from February 5, 2018, to June 30, 2020.
Results
5,896 patients completed LST documents, of which 2,704 (45.8%) signed the documents in person, while family members of 3,192 (54%) wrote the documents on behalf of the patients. Comparing first year and following year of implementation of the act, the self-documentation rate increased (43.9% to 47.2%, p=0.014). Moreover, the number of LST decisions made during or after intensive care unit admission decreased (37.8% vs. 35.2%, p=0.045), and the completion rate of LST documents during chemotherapy increased (6.6% vs. 8.9%, p=0.001). In multivariate analysis, age < 65 (odds ratio [OR], 1.724; 95% confidence interval [CI], 1.538 to 1.933; p < 0.001), unmarried status (OR, 1.309; 95% CI, 1.097 to 1.561; p=0.003), palliative care consultation (OR, 1.538; 95% CI, 1.340 to 1.765; p < 0.001), malignancy (OR, 1.864; 95% CI, 1.628 to 2.133; p < 0.001), and changes in timing on the first year versus following year (OR, 1.124; 95% CI, 1.003 to 1.260; p=0.045) were related to a higher self-documentation rate.
Conclusion
Age < 65 years, unmarried status, malignancy, and referral to a palliative care team were associated with patients making LST decisions themselves. Furthermore, the subject and timing of LST decisions have changed with the LST act.
Introduction
As efforts to improve the quality of life of patients with life-threatening diseases are spreading worldwide, discussions on end-of-life (EOL) care are actively underway [1]. Advance care planning (ACP) is a process of discussing and planning future medical treatment with the patient in preparation for situations in which the patient is unable to make decisions on their own [2,3]. It is advisable to initiate these discussions when life expectancy is predicted to be 6 months to 1 year [4,5]. Quality of life in EOL state is influenced by social, psychological, and functional factors [6]. In particular, appropriate ACP, including the do-not-resuscitate (DNR) code, in patients with life-threatening diseases is important for improving the quality of life of patients at the EOL [7,8]. Furthermore, numerous hospitals in many countries operate palliative care programs, including ACP, to offer appropriate palliative care [9-11].
In Korea, which has a family-oriented culture, most treatment decisions for EOL are made by family members, who are usually reluctant to discuss the patient’s death [12]. Most DNR documents used to clarify the EOL treatment plan are completed by family members, not by the patient, usually within 1 month before death [13,14]. As medical support has become more advanced and complex, the need for proper EOL care for terminally ill patients has gradually increased, and discussions on EOL care have taken place in various fields [15]. Eventually, on February 4, 2018, the Life-Sustaining Treatment (LST) Decision Act was passed in South Korea [16].
According to the LST Decision Act, patients can legally document their opinions on LST. Physicians also receive legal protection against not performing LST. Furthermore, multidisciplinary hospice care for patients at the EOL in Korea has been reimbursed nationally since 2009 as a pilot reimbursement program for inpatient hospices. This program was expanded to include home and consultation hospice care [17]. Therefore, many hospitals treating life-threatening diseases in Korea operate separate teams that offer palliative care consultation and help with LST decisions. According to a previous study on the LST Decision Act, the number of patients making their own LST decisions was higher than before, and the timing of the decision was earlier [18,19]. Some studies reported the changes before and after enactment of the law [19-21]. However, there are insufficient data on practical changes in the LST Decision Act and the role of the palliative care team in LST discussions.
Therefore, we aimed to investigate changes in the trends of LST decisions after implementation of the LST Act and the factors affecting LST decisions.
Materials and Methods
1. Patients, study design, and data collection
This was a single-center retrospective study of patients who withheld or withdrew LST in accordance with the legal process at Asan Medical Center in Seoul, Republic of Korea, from February 4, 2018, to June 30, 2020. Patients aged < 18 years, those whose LST documents were not finalized, and those whose LST documents were withdrawn were excluded. We reviewed medical records and collected the following data: demographic characteristics, palliative care consultation, LST document type, decision date, decision maker, disease status at LST decision, terminal period versus EOL period, cancer information, death-related information, location, date, cardiopulmonary resuscitation (CPR) execution, and treatment information; chemotherapy, radiotherapy, operation, ventilator care, intensive care unit (ICU) treatment, hemodialysis, and parenteral nutritional support. We compared the data between the first year and subsequent year of implementation of the LST Act.
2. Definition [16]
The law is divided into terminal period and EOL period. Terminal period is a condition in which there is no possibility of recovery despite active treatment, symptoms gradually worsen, and death is expected within a few months. EOL period refers to a state in which death is imminent due to rapid deterioration without recovery despite treatment. LST means that treatments such as CPR, mechanical ventilation, renal replacement therapy, chemotherapy, transfusion, and the use of vasopressors extend the duration of the EOL period without improving prognosis. There are three steps to complete the LST decision according to the law. In step 1, two doctors (an attending physician and a specialist) determine that the patient has entered the EOL period (FORM 9). Step 2 is the process of confirming the patient’s or family’s wishes regarding LST. If the patient has previously completed an advance directive, it is verified that there have been any changes in their opinions (FORM 11). Patients who have not made an LST decision can complete the legal document for their ACP in collaboration with an attending physician (FORM 1). FORM 1 can be completed in advance during the terminal period, and in this case, there is no need to sign additional documents. If the patient is unable to express his or her preferences but has expressed in advance that he or she does not want LST, two family members can verify the opinion expressed by the patient in advance (FORM 11). If a patient whose opinion about LST is unknown is unable to express his or her preference, the decision to withhold or withdraw LST is made based on the consensus of all first-degree family members (FORM 12). Step 3 is the final process of implementing EOL care based on the legal documents precompleted by the patients or their families.
In this study, we defined self-determination only for those who completed FORM 1.
3. Statistical analysis
Categorical variables are reported as numbers (percentages) and continuous variables as medians (interquartile ranges). The chi-square test or Fisher’s exact test was used to compare categorical variables with a normal or non-normal distribution, while Student’s t test or Mann-Whitney U test was used to compare continuous variables with a normal or non-normal distribution. Univariate and multivariate logistic regression analyses were performed to adjust for covariates associated with self-determination. Covariates with a p-value of < 0.1 in the univariate analysis were included in the multivariate logistic regression analysis with the stepwise method using the backward selection approach. The odds ratios (OR) and 95% confidence intervals (CI) were calculated. Statistical significance was set at p < 0.05. All reported p-values were two-sided. All analyses were performed using SPSS ver. 26.0 (IBM Corp., Armonk, NY).
Results
1. Patients’ distribution by LST decision maker
This study included 5,952 patients who completed the LST documents. We analyzed data from a total of 5,896 patients, excluding 55 patients under 18 years and one patient with insufficient data (Fig. 1).
Of the 5,896 patients, 2,704 (46.5%), 1,080 (18.3%), and 2,112 (35.8%) completed the LST decision documents FORM 1, FORM 11, and FORM 12, respectively. A total of 2,357 patients completed the LST decision documents within one year of law enforcement (group 1), and 3,539 patients completed the LST decision documents one year after law enforcement (group 2). In group 1, there were 1,034 (43.9%), 419 (17.7%), and 904 (38.4%) patients with LST decision documents FORM 1, FORM 11, and FORM 12, respectively. In group 2, there were 1,670 (47.2%), 661 (18.7%), and 1,208 patients (34.1%) with LST decision documents FORM 1, FORM 11, and FORM 12, respectively.
2. Baseline characteristics
Baseline demographics are summarized in Table 1. The median age was 65 years, and there were 3,600 males (61.1%) and 2,296 females (38.9%). Regarding marital status, 4,769 patients (80.9%) were married, followed by single (5.1%), divorced (2.3%), and bereaved (2.0%). As for educational background, 3,534 patients (60.0%) had a high school degree or less and 1,576 patients (26.7%) had a college degree or higher. A total of 3,900 patients (60.0%) lived in the metropolitan area and another 1,920 patients (32.5%) lived in the non-metropolitan area. To assess economic status based on health insurance status, 5,589 patients (94.7%) had health insurance and 302 patients (5.1%) received medical benefits. There were 3,977 patients (67.4%) with cancer and 1,919 patients (32.5%) with non-cancer disease. Among the cancer patients who made LST decisions, 1,452 (36.5%) were treated in consultation with the palliative care team. There were no statistically significant differences in any of the cases except for hematologic malignancy between groups 1 and 2. Hematologic malignancy was significantly higher in group 1 than in group 2 (8.6% vs. 6.9%, p=0.047).
3. Relevant factors of LST decision makers
Table 2 presents the factors related to LST decision makers. We considered FORM 1 to be indicative of self-determination. Patients with FORM 1 were younger than those with FORM 11 or FORM 12 (median, 62.0 [54.0-71.0] vs. 70.0 [60.0-78.0] or 67.0 [59.0-76.0]; p < 0.001). Women showed a higher completion rate of FORM 1 (46.8% vs. 45.3%) and FORM 11 (19.7% vs. 17.5%) than men but a lower completion rate of FORM 12 (33.4% vs. 37.2%) (p=0.006). The completion rate of FORM 1 was lowest among married patients (45.2%), followed by single (49.2%), bereaved (50.4%), and divorced patients (59.4%), while the completion rate of FORM 12 was similar in single and married patients (36.1% vs. 36.0%) (p=0.005). In addition, there were differences in LST decision makers depending on the type of health insurance. The completion rate of FORM 1 was higher in patients receiving medical benefits than patients having health insurance (53.2% vs. 45.5%) (p=0.022). Regarding cancer type, patients with head and neck cancer had the highest completion rate of FORM 1 (69.4%), while patients diagnosed with hematologic malignancies had the lowest completion rate (28.2%). The highest completion rate of FORM 12 was among patients with hematologic malignancy (53.4%), genitourinary (male) cancer (36.5%), lung cancer (32.6%), and hepatobiliary cancer (30.3%). On the other hand, female cancer showed lower FORM 12 completion in the order of gynecologic(female) (19.4%) and breast (19.9%) (p < 0.001). Only 3,977 patients diagnosed with cancer were eligible for palliative care consultation, and patients who received palliative care consultation (n=1,452, 36.5%) exhibited a higher completion rate of FORM 1 than patients who did not receive consultation (59.7% vs. 40.3%) (p < 0.001). Furthermore, when we compared the first year (group 1) and subsequent year (group 2), the self-documentation rate represented by FORM 1 showed a statistically significant increasing trend over time: 43.9% in group 1 versus 47.2% in group 2 (p=0.014) (Fig. 2). Even when only patients who did not receive palliative care team counseling were analyzed, group 2 (42.6%) had a statistically significantly higher self-determination rate compared to group 1 (39.6%) (p=0.049) (S1 Table).
To identify the factors affecting the self-determination rate, we conducted a logistic regression analysis. In univariate analysis, the following factors were significantly associated with a high self-determination rate: under 65 years of age (OR, 2.040; 95% CI, 1.839 to 2.264; p < 0.001); unmarried status (OR, 1.309; 95% CI, 1.097 to 1.561; p=0.003); medical benefits (OR, 1.346; 95% CI, 1.067 to 1.697; p=0.012); receiving palliative care consultation (OR, 2.086; 95% CI, 1.846 to 2.353; p < 0.001); presence of cancer (OR, 2.407; 95% CI, 2.146 to 2.669; p < 0.001); and LST decision timing of 1 year after enforcement of the LST Act, comparing within 1 year (OR, 1.140; 95% CI, 1.027 to 1.266; p=0.014) (Table 3). We performed multivariate analysis by including all the factors that were significant in the univariate analysis. The multivariate analysis results indicated that younger age (< 65 years) (OR, 1.724; 95% CI, 1.538 to 1.933; p < 0.001), unmarried status (OR, 1.204; 95% CI, 1.003 to 1.447; p < 0.001), receiving palliative care consultation (OR, 1.538; 95% CI, 1.340 to 1.765; p < 0.001), presence of cancer (OR, 1.864; 95% CI, 1.628 to 2.133; p < 0.001), and passage of time since the law enforcement (first year vs. following year) (OR, 1.124; 95% CI, 1.003 to 1.260; p=0.045) were significantly associated with a high self-determination rate.
4. Changes in clinical practice of EOL care
Focusing on changes in clinical practice over time with enforcement of the law, the percentage of patients who received ICU care and CPR before completion of the LST documents was statistically lower in group 2 than in group 1 (35.2% vs. 37.8% for ICU care [p=0.045] and 2.9% vs. 4.3% for CPR [p=0.045]) (Table 4). Moreover, the percentage of patients who received parenteral nutrition before the LST decision was lower in group 2 than in group 1 (7.5% vs. 9.1%, p=0.022). In particular, the number of patients who wrote their LST documents before or during chemotherapy was significantly higher in group 2 (n=314, 8.9%) than in group 1 (n=155, 6.6%) (p=0.001). The rate of receiving radiotherapy or surgery after LST completion was not significantly different between the two groups (Table 4).
Discussion
Our study revealed that factors such as age under 65 years, unmarried status, receiving palliative care consultation, and presence of cancer were associated with self-LST decision-making. Furthermore, as the LST Decision Act was implemented and time passed, patients’ autonomy in making their LST decisions increased and their LST decision timing became earlier.
Previous national studies in Korea showed that being male, age range 30-70, low insurance premium level, no disability, no comorbidities, residence in the national capital region, history of hospitalization, and especially cancer diagnosis were associated with a high level of self-LST decisions [22,23]. The findings regarding age and cancer diagnosis are consistent with our results. However, most of the influencing factors, including unmarried status, palliative care consultation, and passage of time, were new findings in this study. Because our research focused on adult patients undergoing treatment in a tertiary hospital, the research subjects were quite different from those in the previous national studies, which included children, adolescents, and people who did not have comorbidities. Our study reflects critically ill adult patients’ decision-making better than the general population.
In our study, the rate of self-determination was higher in patients without a spouse, including divorced, widowed, or unmarried (52.1%), than in those with a spouse (45.2%). Considering these results, the high rate of self-LST decisions might be due to the lack of support of or dependence on family members. In terms of palliative care consultation, considering that the goal of the palliative care team is to achieve the physical and mental well-being of patients and caregivers by guiding the direction of treatment [24], the significant association between consultation and self-determination is an expected outcome. A previous pilot study conducted in Korea comparing before and after hospice counseling showed that there was an improvement in patient awareness, preparation for death, and feelings of anxiety [25]. Therefore, we believe that the intervention of the palliative care team would have influenced patients’ understanding of their disease prognosis and their ability to make decisions about LST on their own.
In Korea, both the patient’s family and doctors are culturally reluctant to inform patients of the disease prognosis [12]. This cultural background is similar to that in other Asian countries [26]. In a study conducted before enactment of the law, LST decisions were mostly made by patients’ families [14]. However, autonomy and quality of life at the EOL have been consistently emphasized, leading to enactment of the LST law not solely based on the opinions of healthcare professionals but rather in accordance with societal consensus in Korea [16]. In a previous study comparing before and after enforcement of the law, the self-determination rate increased from 1.5% to 63.5% after enforcement of the law, and the survival period after completion of the LST documents also increased from 14.4 days to 21.2 days [19]. These findings suggest that patients and physicians are improving their awareness of LST decisions with law enforcement, and a similar trend was observed in our study. Prior to law enforcement, researchers from the same institution reported that DNR decisions were made by family members in 100% of cases [27]; however, according to our data, the proportion of self-LST document completion during the 28 months following enforcement of the law was 45.8%. This result was numerically higher than the 32.1%-33.5% for the total population in previous national studies but was similar to 47.3% for cancer patients [22,23]. We believe this is because the proportion of cancer patients in our study was high. Comparing the overseas situation, a study that analyzed the rate of EOL decisions in six European countries (Belgium, Denmark, Italy, the Netherlands, Sweden, and Switzerland) reported that it was around 23% to 51% [28]. Furthermore, analyzing articles published in the United States from 2011 to 2016, the completion rate of living wills as advance directives was 29.3% [29]. According to previously reported Korean national data and this study, it seems that the self-documentation rate has risen to a level similar to that in Western countries after the enactment of the law.
Moreover, we observed an increasing trend in the self-determination rate during the first year and subsequent year following enactment of the law. This trend was maintained even in the subgroup of patients who did not receive palliative care team consultation. Over time, there were not only changes in the decision makers for LST but also in the timing of decisions and healthcare utilization patterns. The rate of transfer to the ICU, receiving CPR, and administration of total parenteral nutrition before the LST decision decreased, whereas the number of patients who received chemotherapy after the LST decision increased. The increasing number of patients receiving chemotherapy after making LST decisions is interpreted as earlier decision-making during palliative chemotherapy, whereas in the past, EOL decisions were mainly made after discontinuing chemotherapy. Furthermore, the rate of transfer to the ICU and receiving CPR before the LST decision decreased, which also supports earlier decision-making. A previous multicenter study that compared healthcare utilization before and after enforcement of the law for patients who had LST documents and died in the hospital reported a significant reduction in CPR performance (3.6% vs. 0.2%) and the use of anticancer drugs (20.7% vs. 8.6%) [21]. The time points of the comparison groups in the previous study and our study were different. Additionally, the previous study focused on patients who died in tertiary hospitals and excluded patients transferred to a hospice; therefore, the results regarding receiving chemotherapy after LST completion may differ from those of our study.
This study had several limitations. First, since the study included single tertiary hospital data, the results can be interpreted within the context of critically ill patients and may not reflect outcomes in long-term care hospitals or community care. Second, non-self-determination was divided into cases in which patients’ opinions were previously confirmed (FORM 11) and cases in which opinions were unknown (FORM 12); however, both were considered non-self-determination in this analysis. Third, because our study included only patients who completed their LST documents, it may not fully reflect all patients who received EOL care. To overcome this limitation, a multicenter study that includes all deceased patients regardless of the completion of LST documents is needed. Despite these limitations, this is the first study to elucidate the role of the palliative care consultation team in LST decisions and analyze chronological changes after enactment of the LST Decision Act in Korea.
In conclusion, the LST Decision Act was established to respect patients’ preferences for EOL care and it is demonstrating its application in actual practice. Younger patients and those diagnosed with cancer showed a tendency to decide on LST on their own. Among cancer patients, those referred to a palliative care consultation team showed a higher self-determination rate than non-referred patients. As the law has been implemented and time has passed, there has been an increased proportion of patients making LST decisions earlier, especially among cancer patients, even during anticancer drug treatment. Future follow-up studies are needed to confirm our findings through multi-institutional research.
Electronic Supplementary Material
Supplementary materials are available at Cancer Research and Treatment website (https://www.e-crt.org).
Notes
Ethical Statement
The Institutional Review Board of Asan Medical Center approved the study protocol (2020-1463) and waived the requirement for informed consent because this was a retrospective analysis. This study was conducted in accordance with the ethical standards of institutional research and the Declaration of Helsinki.
Author Contributions
Conceived and designed the analysis: Jang YJ, Seo S.
Collected the data: Jang YJ, Yang YJ, Koo HJ, Yoon HW, Uhm S, Kim SY, Kim JE, Huh JW, Kim TW, Seo S.
Contributed data or analytical tools: Jang YJ, Yang YJ, Koo HJ, Yoon HW, Uhm S, Kim SY, Kim JE, Huh JW, Kim TW, Seo S.
Performed the analysis: Jang YJ, Yang YJ, Koo HJ, Yoon HW, Uhm S, Kim SY, Kim JE, Huh JW, Kim TW, Seo S.
Wrote the paper: Jang YJ, Seo S.
Interpreted, reviewed, and commented: Jang YJ, Seo S.
Conflict of Interest
Conflicts of interest relevant to this article was not reported.
Acknowledgements
We express our sincere gratitude to the Korea Institute of Radiological & Medical Sciences for providing English proofreading services in support of the Medical Scientists Incubation Program.