Article Text

other Versions


The attitudes of neonatal professionals towards end-of-life decision-making for dying infants in Taiwan
  1. Li-Chi Huang1,
  2. Chao-Huei Chen2,
  3. Hsin-Li Liu3,
  4. Ho-Yu Lee4,
  5. Niang-Huei Peng3,
  6. Teh-Ming Wang2,
  7. Yue-Cune Chang5
  1. 1School of Nursing, China Medical University, Taichung, Taiwan
  2. 2Diversion of Neonatology, Taichung Veterans General Hospital, Taichung, Taiwan
  3. 3Nursing College, Central Taiwan University of Science and Technology, Taichung, Taiwan
  4. 4Nursing Department, Chang Bing Show Chwan Memorial Hospital, Changhua, Taiwan
  5. 5Department of Mathematics, Tamkang University, Danshui, Taiwan
  1. Correspondence to Professor Niang-Huei Peng, Nursing College, Central Taiwan University of Science and Technology, No.666 Pu-tzu Road Betiun District, Taichung city 40601, Taiwan; lilyflower.nature{at}


The purposes of research were to describe the neonatal clinicians' personal views and attitudes on neonatal ethical decision-making, to identify factors that might affect these attitudes and to compare the attitudes between neonatal physicians and neonatal nurses in Taiwan. Research was a cross-sectional design and a questionnaire was used to reach different research purposes. A convenient sample was used to recruit 24 physicians and 80 neonatal nurses from four neonatal intensive care units in Taiwan. Most participants agreed with suggesting a do not resuscitate (DNR) order to parents for dying neonates (86.5%). However, the majority agreed with talking to patients about DNR orders is difficult (76.9%). Most participants agree that review by the clinical ethics committee is needed before the recommendation of ‘DNR’ to parents (94.23%) and nurses were significantly more likely than physicians to agree to this (p=0.043). During the end-of-life care, most clinicians accepted to continue current treatment without adding others (70%) and withholding of emergency treatments (75%); however, active euthanasia, the administration of drug to end-of-life, was not considered acceptable by both physicians and nurses in this research (96%). Based on our research results, providing continuing educational training and a formal consulting service in moral courage for neonatal clinicians are needed. In Taiwan, neonatal physicians and nurses hold similar values and attitudes towards end-of-life decisions for neonates. In order to improve the clinicians' communication skills with parents about DNR options and to change clinicians' attitudes for providing enough pain-relief medicine to dying neonates, providing continuing educational training and a formal consulting service in moral courage are needed.

Statistics from


Advances in neonatal medicine have dramatically improved neonatal survival in Taiwan.1 However, high technological advances in neonatal care have led to a corresponding improvement in long-term neurological outcomes,2 and to unprecedented dilemmas regarding end-of-life decisions for high-risk infants and extremely premature infants.3

When infants have no reasonable chance of survival due to various diseases, the end-of-life decision to withdraw or withhold neonatal intensive care may be made to relieve their suffering.4 Strictly defined, the do not resuscitate (DNR) order is the decision to forgo cardiopulmonary resuscitation and was formally introduced as an option for end-of-life care.5 The definition of withholding intervention was withholding potentially life-sustaining treatment which included withholding cardiopulmonary resuscitation and not providing additional intensive care interventions, for example, medicine. Withdrawing intervention was defined as withdrawing the mechanical ventilator and some medicines.4 Withdrawal of ventilation was the most common intervention associated with neonatal death in neonatal intensive care units (NICUs).4 6

A recent European study, European Project on Parents' Information and Ethical Decision Making in Neonatal Intensive Care Units: Staff Attitudes and Opinions (EURONIC) project,6 carried out on a large representative sample of NICUs in 10 European countries has found that the majority of neonatologists agreed with three strategies to limit care: (1) continuation of treatment without intensification; (2) withholding of emergency manoeuvres; and (3) withholding of intensive care. In addition, research also found that culturally-specific factors are more relevant than characteristics of individual neonatologists or NICUs in explaining the variation in non-treatment decisions.

In Taiwan, the DNR document includes four options (to withhold intensive treatments, to withhold emergency treatment, to withdraw life-saving drugs and to withdraw mechanical ventilation) among which parents or families of dying infants may choose. A recent study investigated the conditions of infant death in the NICU in Taiwan through a review of 61 charts.7 In this research, 41 infants (67.2%) had a pre-existing DNR order at the time of death; moreover, the DNR order was obtained within a few hours of the infant's death in most cases (n=30, 28.7%). Of these 41 infants, 11 infants had withdrawal of mechanical ventilation. In another study, we investigated the conditions of decision-making for dying infants and the cultural effects on the process of infant death by reviewing the charts of 50 neonates who died in the NICU in Taiwan. In this research, 82% of the charts documented a DNR order and 16% of parents allowed discontinuation of ventilator support when they realised the futility of continued care. According to nursing notes in this research, various cultural issues affected both the families and the dying process of their infants including consideration of the time of death, use of alternative treatments and cultural traditions surrounding infant death. All these cultural issues could affect the process of obtaining DNR orders for dying infants.8 Another recent study has investigated the conditions of aggressive paediatric end-of-life cancer care in Taiwan.9 Research found that the majority of paediatric patients with cancer received chemotherapy and underwent mechanical ventilation in their last month of life.9 The above two studies revealed the phenomenon of the unavailability of paediatric hospice care in Taiwan.

When making critical decisions, parents of dying infants rely on physicians, rather than family or friends, to interpret the information and guide them during the decision-making process.10 Nurses usually play a key role in helping parents to understand information conveyed by physicians. When making a decision about withholding or withdrawing therapy, the considerations of neonatologists and nurses should also be taken into account. Studies have shown that clinicians differ considerably in their attitudes regarding the value of life and that clinicians' attitudes were reflected in their practices.5 10 11 Other investigators found that critical decision-making by parents for the care of newborns is influenced by the beliefs and attitudes of physicians and nurses.12–14 However, there exists little information on how neonatal clinicians in Taiwan actually confront this ethical issue and on underlying attitudes and views of neonatologists and nurses.

Research purposes

The purposes of this investigation are to describe the neonatal professionals' personal views and attitudes towards the neonatal ethical decision-making, to identify factors that might affect those attitudes and to compare the attitudes between neonatologists and neonatal nurses in Taiwan.

Research methods

This was a cross-sectional research, part of a larger research project which surveyed the attitudes and beliefs of neonatal staff towards applying palliative care in NICUs.

Sampling and settings

The research settings were the level III NICUs that met the inclusion criteria as routine care of high-risk infants, with at least 40 admissions per month as well as neonatal mortality rate about 4.05%, availability of long-term mechanical ventilation treatments as well as 24-h availability of a neonatologist. Convenience sampling was used to collect data. The research criteria were neonatologists and neonatal nurses with experience caring for dying infants and who worked in the above settings at four medical centres around the central region of Taiwan during the preceding 1 year.

Research instrument

A structured questionnaire was developed consisting of three components: demographic characteristics, attitudes regarding strategies of end-of-life decision-making for dying neonates and attitudes regarding DNR discussions. Seven questions for surveying the strategies of end-of-life decision-making were developed from the clinical practice in Taiwan and one previous research related to the EURONIC project.15 Six questions to investigate neonatal professionals' attitudes regarding DNR discussions were adapted from an earlier study and then modified based on the clinical practices in Taiwan.16

The attitudes towards end-of-life decision-making options were explored by asking respondents' agreement and coded accordingly: disagreement =1, agreement =2. Attitudes were quantified on a 5-point Likert type response scale (from ‘strongly agree’ to ‘strongly disagree’), with a list of six statements about the decision-making for a DNR order. The response scale was coded accordingly: strongly disagree =1; somewhat disagree =2; unsure =3; somewhat agree =4; and strongly agree =5.

Content validity was established through an expert review by an expert panel comprised of two neonatologists and three nursing scientists, all of whom were experienced in the care of terminally ill neonates. Each item in the questionnaire was appraised from ‘very inappropriate and not relevant’ =1 to ‘very appropriate and relevant’ =5. An overall content validity index was used to determine the validity of the structured questionnaire (content validity index =0.86), revealing high content validity. Part of the questions in questionnaire were originally prepared in English and subsequently translated into traditional Chinese characters. Attention was paid to the interpretation of the different end-of-life decisions. The translation accuracy was checked by a forward–backward translation procedure from English to Chinese. The research questionnaire's Cronbach α score (reliability Cronbach α score =0.8) acceptably demonstrated the reliability of the measured construct by previous studies.17 18

Data collecting procedure

The Committees for the Protection of Human Subjects of Research Hospital approved the conduct of this research. The researchers invited participants who met the inclusion criteria from all research NICUs. The respondents provided their consent to participate in the voluntary and anonymous survey when they completed the questionnaire. The survey was conducted from 2010 to 2011.

Statistical analysis

Statistical analyses were performed on a personal computer using the statistical package SPSS (V.18, SPSS). Fisher's exact tests were used for the comparisons of categorical data, for example, gender and religious beliefs. The independent t tests or one-way analysis of variance were used for two-group (or more than two-group) comparisons. The corresponding non-parametric methods, Mann–Whitney U or Kruskal–Wallis tests, were used whenever the normality assumption was not satisfied. Statistical significance was defined as p≤0.05.

Research results

Response rate and demographic characteristics

A total of 25 neonatologists and 84 neonatal nurses were invited and recruited from four participating NICUs, and 104 completed and returned the questionnaire (24 physicians and 80 nurses). The response rate was 95.4% because of a lot of missing data in five returned questionnaires. Of the responding nurses, 100% were female subjects and most of responding physicians were male subjects (66.7%). Most of responding nurses were younger than the responding physicians. Table 1 summarises the demographics and neonatal professional experience of the research participants.

Table 1

Demographic data of 104 survey respondents (80 neonatal nurses, 24 neonatologists)

Personal views regarding end-of-life decision-making for neonates

Table 2 shows the proportion of neonatal professionals' agreement or disagreement with the seven strategies for neonatal end-of-life decisions. The two strategies with which the majority of respondents agreed were: continuation of current treatment without addition of others (n=73, 70.19%) and withholding of emergency treatments (n=78, 75%; table 2). The administration of drugs with the purpose of ending life and the withdrawal of mechanical ventilation were not endorsed by most of the neonatologists or nurses. Neonatologists were significantly more likely than nurses to agree with the practices of withholding intensive care (t=2.197, p=0.03), withholding emergency treatment (t=2.802, p=0.007) and continuing current treatment without adding others (t=2.575, p=0.013).

Table 2

Personal views regarding end-of-life decision-making for neonates

Attitudes towards the DNR order for neonates

Table 3 shows the distribution of the responses regarding the DNR order for neonates. There was broad agreement among respondents that talking to parents about DNR orders is difficult (n=80, 76.9%, table 3). In addition, a large majority of participants agreed that a clinical ethics committee for discussing medical procedures is needed before the recommendation of DNR to parents of dying infants (n=98, 94.23%), with nurses agreeing significantly more than neonatologists (Z=−2.026, p=0.043). More nurses significantly agreed with the statement ‘DNR consent form is clear for me’ than did neonatologists (Z=−2.131, p=0.033). However, neonatologists showed more confidence in discussing consent for medical procedures than did nurses (Z=−2.776, p=0.005).

Table 3

Attitudes towards the DNR order for neonates (80 nurses, 24 neonatologists)

Factors impacting clinicians' attitudes towards end-of-life decisions

In this research, the distribution of age, the length of professional experience and religion of NICU clinicians were not significantly associated with participants' attitudes (p>0.05). Participants in this research were 88 female and 16 male subjects. The independent t test was used to examine the differences in scale scores according to the different genders of participants. There were several significantly different responses between different genders on the different options of DNR order for dying neonates and different attitudes towards the decision-making process for a DNR order. However, research nurses were female subjects and most of the neonatologists were male subjects.


Personal views on the end-of-life decision-making for neonates

The majority of neonatal clinicians felt that to continue the current treatment without adding any other interventions and to withhold emergency treatments may be ethically justifiable in case of fatal or terminal ill neonates in Taiwan. In contrast, the administration of drugs to end life and withdrawal of mechanical ventilation were not acceptable to most participants. Perhaps this is because neonatal euthanasia remains illegal in Taiwan and most participants believe that ethical decisions should not violate the law.

Most participants, especially neonatal nurses, disagreed or strongly disagreed with prescribed analgesics because of a high risk of decreasing respiratory function. This may be one reason that a previous study found that dying neonates received insufficient analgesic medicines in NICUs in Taiwan.4 To compare results with other studies conducted in Spain,19 USA,20 Italy, France, The Netherlands and Sweden,6 research participants were more likely to express concern regarding the use of analgesics in NICUs. This research result showed that there is an urgent need to change and improve the neonatal professionals' attitudes regarding pain control care for these terminally ill neonates in Taiwan.

Compared with the results of studies in several European countries15 17 21 and the USA,20 research participants were more likely to disagree with the withdrawal of mechanical ventilation and withdrawal of life-saving drugs from dying neonates.

There are several potential reasons for the differences found in this study. In Taiwan, neonatologists are required to inform parents and complete a DNR consent form before applying end-of-life interventions for terminally ill neonates. It is illegal to withdraw any emergency medicines and ventilator without the parents' consent. Furthermore, National Health Insurance in Taiwan will cover all medical treatments and emergency interventions for these terminally ill neonates. Thus, concern about cost may not provide an urgent concern to make a critical decision for dying infants.

Taiwanese parents tend to propose that physicians should make resuscitation decisions without parental involvement. In paternalistic cultures, such as Taiwan, parents hope for their child's survival and agree to the physician's authority.22 23 Nevertheless, neonatal professionals often feel a sense of failure at the death of an infant.8 24 Therefore, high-risk infants often receive several critical interventions before they die without a pre-existing DNR order in Taiwan.7

In order to provide quality care for dying infants and their families, further studies are needed to deal with these ethical dilemmas for neonatal professionals. Continuing education regarding pain management in terminally ill neonates and moral courage for neonatal professionals will be needed to overcome fear and to stand up for the core values surrounding compassionate end-of-life decision-making.25 26

Attitudes towards the DNR order for neonates

In this research, most participants agreed to suggest a DNR order to parents of dying neonates and most participants agreed that obtaining consent of DNR orders is beneficial for terminally ill neonates. However, some participants thought that the definition of the DNR informed consent form was unclear and also felt that it is a difficult subject to discuss with parents (table 2). Compared with a previous study conducted in USA,16 research participants in this research were more likely to report that discussing the DNR order with parents is difficult. Our research also found that most neonatal professionals acknowledged the importance of reviewing by the clinical ethics committee before suggesting the DNR options to parents. Some previous studies have also found that most paediatricians favoured formal review of medical decisions by colleagues together with ethical or legal experts.27 28

Impact factors in clinicians' attitudes towards end-of-life decisions

Neonatologists were more likely to agree ‘to withhold intensive care’, ‘to continue current treatment without adding others’ and ‘to withhold emergency treatment' than were neonatal nurses (table 2). Compared with neonatologists, nurse participants were less confident discussing consent for medical procedures (p=0.005, table 3). Research outcomes implied that nurses likely face more moral distress than neonatologists during the end-of-life decision-making process. As the main caregivers, neonatal nurses are in an important position to contribute to the decision-making process regarding the care of the terminally ill neonates.12 16 29 30 At least NICU nurses are often asked to implement any decisions made with regard to providing or withdrawing certain treatments.31–33 In order for nurses to contribute to the decision-making process in the most effective manner, continuing education for neonatal professionals in theory and moral encouragement are needed.

According to literature, the religious background19 34 and the length of professional experience of clinicians may be important impact factors;9 14 however, in our research, no differences were found in attitude score among different religious beliefs and professional backgrounds. The reason for the differences between this research and previous literature may be that demographic data in these research participants presented a centralised tendency, such as a predominance of Buddhism or Taoism (table 1). There were several differences responds between female participants and male participants. However, most research neonatologists were male subjects and all neonatal nurses were male subjects. Further study may be needed to investigate the impact of these personal characteristics towards end-of-life decisions for terminally ill neonates.


Although the research questionnaire was designed to evaluate the attitudes of professionals towards the limitation of intensive care and neonatal ethical decision-making, all responses were by self-report and participants may have inflated or minimised their responses. To reduce this response bias, future research should enlarge the number of research participants by expanding to different areas of Taiwan. Also different research methods may be used to investigate the neonatal clinicians' ethical dilemmas and conflicts related to the status of DNR.


To our knowledge, this is the first study exploring the neonatal professionals' personal views and attitudes towards the neonatal end-of-life decisions in Taiwan. In our study, neonatologists and nurses held similar views and attitudes towards end-of-life decisions for neonates. Based on research findings, continuing educational services and a formal counselling service may be needed to relieve neonatal professionals' anxiety regarding communicating with parents about the DNR options for dying infants and to change clinicians' attitudes towards providing them sufficient pain-relief medication. Additionally, a clinical ethics committee is also essential to resolve ethical dilemmas during the process of end-of-life critical decision-making for dying infants.


The authors thank all research neonatal professionals for their attending and many contributions.


View Abstract


  • Competing interests None.

  • Patient consent All research participants (neonatal professionals) had signed the consent forms before the research.

  • Ethics approval The Committees for the Protection of Human Subjects of Research Hospital approved the conduct of this research.

  • Provenance and peer review Not commissioned; externally peer reviewed.

Request permissions

If you wish to reuse any or all of this article please use the link below which will take you to the Copyright Clearance Center’s RightsLink service. You will be able to get a quick price and instant permission to reuse the content in many different ways.

Linked Articles