There are many obstacles and taboos in Korea and other Asian nations regarding discussions on death. Proxy decision-making for end-of-life (EOL) is overwhelming, and the EOL discussion takes place approximately 2 to 8 days before death [1,2]. The use of advanced directives can promote patient participation in EOL discussions [3]. In Korea, the Act on Hospice and Palliative Care and Decisions on Life-Sustaining-Treatment for Patients at the EOL was enacted in 2016 and implemented in 2018 to enhance patients’ involvement in making decisions about EOL [4]. This law allows terminally ill patients with no chance of rehabilitation to withdraw or withhold life-sustaining treatment (LST) with their own consent or that of their family members. The patient’s intention for LST on the Act is a decision on four items, including cardiac resuscitation, mechanical ventilation, hemodialysis, and anti-cancer drugs. This law covers 43 pages, including the act, enforcement decree, enforcement rules, a table, and seven forms. The seven forms required by law include the following: form 1 (LST plan form), form 6 (advanced directive form), form 9 (determination of whether the patient is at the EOL process), form 10 (confirmation of the patient’s intention by advanced directive), form 11 (confirmation by consistent statements of two or more of the patient’s family members), form 12 (confirmation by unanimous consensus of the patient’s family), and form 13 (implementation of LST). Form 6 is written in advance by a person aged ≥ 19 years with a direct submission of his or her decision on whether to use a hospice, and should be written directly by the registration authority designated by the Minister of Health and Welfare; this form does not involve patient decisions regarding LST at the EOL. With the exception of form 6, the other six forms must be written by a doctor in a hospital, and occasionally also a specialist. With the commencement of the Act, more than one of the six forms should be prepared in the hospital and used to confirm the patient’s intention to withdraw or withhold LST, determine whether the patient is at the EOL stage, and implement the patient’s decision regarding LST.

Here we evaluated the components of the six forms that should be written in the hospital and are required by law to make plans for EOL treatment. We also analyzed the preparation of the forms and the implementation of LST decisions from the database of the National Agency for Management of LST in the year following the enforcement of the Act.

The database of the National Agency for Management of LST includes seven forms: forms 1, 6, and 9–13. Form 6 can be prepared regardless of the disease and is excluded from the documentation required by the hospital. We collected the terminal status and EOL information required by each of the forms under the Act on Decisions LST for Patients in Hospice and Palliative Care or at the EOL as follows (S1 Fig. is Korean version of forms 1, 9–13): form 1, LST plan form; form 9, determination of whether the patient is at the EOL process; form 10, confirmation of the patient’s intention by advanced directive; form 11, confirmation by consistent statements of two or more of the patient’s family members; form 12, confirmation by unanimous consensus of the patient’s family; and form 13, regarding the implementation of LST. A doctor with a patient who is in terminal status or at the EOL process writes form 1 to leave his/her intention of LST. If the patient is unable to leave his/her intention to the doctor, form 11 or 12 is completed by a doctor or specialist with his/her family.

We analyzed the preparation of the form’s from the database of the National Agency for Management of LST between February 4, 2018 and January 31, 2019. Form 10 was excluded from this analysis because it was indirectly prepared in form 13 and details of form 10 were not available from dataset the National Agency for Management of LST. We defined patients with form 1 as “self-determination,” and those with forms 11 or 12 as “family decisions”. We collected the following data on forms other than form 10 from the database of the National Agency for Management of LST: Form 1, patient information (sex, age, status, address), institution (type and address), intention to use hospice services, and four decisions relating to LST, including the date on which the patient’s intention was identified and the date of creation; form 9, patient information (sex, age, diagnosis), date of the doctor’s decision, date of the specialist’s decision, and date of creation; form 11, patient information (sex and age), information relating to family members (total number, number of making a statement, relationships), and date of creation; form 12, patient information (sex and age), number of family members, and date of creation; and form 13, patient information (sex and age), institution (type and address), four decisions for LST, including verification of the patient’s intention, the date on which the patient’s intention was identified, and the date of creation. We compared the four items of LST on form 13, and the order of documentation and time interval between forms 1, 11, or 12 and form 13 according to the decision subject. Formally, after decision were made on LST in forms 1, 11, or 12, LST in form 13 were implemented. It is generally the order in which LST is implemented after a decision has been made. The time interval dates from the creation of forms 1, 11, or 12 to form 13 reflect the time from the decision to the implementation of LST. We also analyzed several dates that required more than one date within forms 1, 9, and 13.

In the current study, we found that the self-determination rate of LST was 31% and the mean time interval from self-determination to implementation of LST was approximately 8.6 days, which is higher and longer than those from previous studies [1,2,5]. Indeed, the self-determination rates from recent single center and national retrospective studies were 29% and 33.5%, respectively [6,7]. There was little difference in the self-determination rate as a result of differences in the hospital settings, enrolled subjects, or research period. The time interval dates from the creation of forms 1, 11, or 12 to form 13 reflect the time from the decision to the implementation of LST. In the case of family decision, the mean time interval was about 1 day. It remains to be filled out when a patient is close to dying. Our results showed that approximately 56% of patients in the self-determination group and approximately 80% in the family decision group decided their LST and implemented it on the same day. In other word, these patients implemented LST on the day of the decision of LST. Family decisions are much more likely to be made and implemented on the verge of death. Therefore, it is necessary to encourage patients to participate in the discussion of LST at an earlier stage of illness.

Advanced care planning (APC), including EOL discussion, is important to help patients meet a peaceful and dignified death. Both the advanced directive and POLST are APC forms, although there are some differences between the two in terms of the population, who completes the form, and the time frame [8]. An advanced directive is a legal document that can be written by anyone regardless of his/her illness and includes a future medical care plan. The POLST form is a medical document that includes mainly EOL discussions. In 1991, physicians in Oregon developed the POLST program [9], which converts patients’ wishes for treatment into medical orders. In Korea, the legislation of the POLST program is the Act on the decision of LST for patients at the EOL. This Act established an approach to EOL planning that is based on conversations between patients, family members, and doctors to determine and honor the wishes of seriously ill patients. The POLST forms need to be consistent and easy to write to allow patients’ preferences regarding the use of LST to be honored. Incomplete and contradictory POLST forms may cause confusion among healthcare providers and may result in patients receiving treatment contrary to their wishes [10,11].

The implementation of the Act includes both withholding and withdrawing LST. Generally, there is no ethical or legal distinction between withdrawing and withholding LST. However, approximately 70% of Koreans think there should be ethical and legal differences between withdrawing and withholding LST [12], withholding being acceptable, but withdrawing socially unacceptable. As one law attempted to control acceptable withholding and difficult-to-accept withdrawing LST, complex forms and penalties were included in accordance with the latter. Under the Act, a person who violates the this LST law may be sentenced to up to 3 years in prison or fined up to 30,000,000 won. However, legislating and penalizing issues with insufficient social consensus do not change the social awareness of death. To this end, a complex, multifaceted, and longitudinal intervention such as continuous social efforts, institutional publicity, and education should be accompanied [13].

On the forms required for the Act in Korea, information about a patient and his/her doctor is repeatedly written. Information about hospitals is also repeated in that it is included in the information kept at the workplace of doctors and specialists or requires administrative information. With the exception of forms 11 and 12, the other forms require more than one date to be input, such as the date of creation, the date of identification, and the date of decision. In more than 95% patients who have completed forms 1 and 9, several dates within the forms were same. Conversely, it is relatively simple to write the items relating to the LST decision. The LST decision item is written to express the patient’s intention by marking the item and not expressing his/her wishes as I will or will not. Marking could cause confusion regarding whether a patient wishes to either receive or postpone LST.

The Oregon POLST registry is a 1- or 2-page format [14]. In Korea, there are six forms that have been created in the hospital, among which, at least two should be written. Six forms required an Act involving various basic information on patients, doctors, specialist doctors, family members, and institutions. Information about patients and doctors as subjects is required in all six forms. Information about hospitals is repeated in that it is included in the workplace information of doctors and specialists or requires administrative information. It is necessary to reduce both the number and items of forms to reduce the repetition of information between forms and to reduce the effort required to prepare the forms.

As four items of LST are based on a social consensus that took place in 2009 [15,16], they need to be organized according to LST intensity or patient status when a patient need LST and then modified to the current perception or medical judgment. In the revised version of March 26, 2019, extracorporeal membrane oxygenation, blood transfusion, inotropics, and other LST were added LST which terminally ill patient with no chance of rehabilitation could decide to withhold or withdraw. The Act not only relates to cancer patients but also terminally ill patients with chronic disease. Moreover, given that anti-cancer drugs with fewer toxicities, such as targeted agents, have been developed, the exclusion of anti-cancer drugs among LST items should be reconsidered. Considering that Oregon is using version 11 of the POLST in 2017 [13], constant renewal are needed to make it easier to verify the patient’s intention of LST. Furthermore, the education of people who help patients understand the medical situation and write POLST properly is needed. According to a previous study, approximately 90% of patients who completed POLST and wished to receive chemotherapy in the first step changed their intention to not receive chemotherapy in the second step after the doctors had thoroughly explained the time of EOL [17].

To improve EOL care by reflecting the patient’s value, designating an agent in case the patient cannot make a decision is another way. Establishing patients’ rights to forgo and the authority of surrogate decision makers were achievements in the first phase of improving EOL care [18]. There is no concept of a surrogate in the act of LST; therefore, without designating an agent, the decision is made based on indirect decisions by family unanimity or after receiving two or more statements from family members regarding the patient’s wishes. However, the patient’s family may not know the value of the patient’s wishes for EOL care and may sometimes show an aggressive attitude toward EOL care [19,20]. In addition, there might be conflicts among family members or an issue of no family members who are available to make decisions. For this limited case, it might be important to establish surrogate decision makers about EOL process on behalf of incompetent patients. Actually, Korea has a family oriented Confucian culture, and many patients prefer to consider the best interests of their family members as opposed to their own. Indeed, some patients may suspend their LST to lessen the economic burden on the family [17]. Non-family surrogate decisions can cause another conflicts if there are no authority on the law. Therefore, earlier discussions on EOL care can not only help patients have time for self-determination, but also mediate different perspectives among patients and family members regarding life prolongation. This requires continuous social effort.

Our study had several limitations. First, since we examined data within a year of law enforcement, the results also included data related to trial and error in the settlement process. Approximately 17% of patients with at least one form had not completed form 13, 5% of patients with form 13 had not completed any other forms, and 0.5% of patients had completed forms 1, 11, and 12. There were also about 1–2% of patients whose decisions were made after LST was implemented. Second, we could not fully investigate all of the problems associated with form preparation, such as the completeness of the form and the difference in LST between forms, because of limited accessibility to data. Third, we excluded several patients with form 10 as advanced directive because it was not available. This can explain the lower self-determination ratio compared to the previous national survey. Although these results involve small numbers of patients, they are all self-determinators and will also increase in the future as more people complete form 6. Finally, there is an ambiguity about the fact that the date of creation of form 13 was indirectly regarded as the date of death. The dates of the completion of form 13 are expected to be similar to the date of death when the LST decision is implemented, but in practice, it implies uncertainty that the patient can survive and the condition can improve.

In conclusion, we found that the self-determination rate of LST was 31%, and the mean time interval from self-determination to implementation of LST was approximately 8.6 days after the enforcement of the Act. However, the creation of these forms still takes place when the patient is near to death. Moreover, in the early stages of implementation, there are many types of forms, and some information on patients and doctors, and the date of creation need to be written repeatedly. Therefore, continuous revisions and updates of forms are needed. Moreover, social efforts and communication are important to change the perception of death and move forward with the discussion of death earlier.