Necessity for this study
Εnd-of-Life decisions refer to decisions that Neonatologist may be called upon to take for neonates who have suffered severe brain injury associated with poor neurologic outcome, neonates, with serious congenital malformations or untreatable genetic conditions incompatible with life, marginally viable neonates at 22+ 0/6 -23+ 6 weeks gestation, and terminally ill neonates at a non-reversible state.4 22 Quality of provided antenatal/ prenatal care and improved prematurity survival rates, underline the necessity of empirical data collection for the implementation of specific policies in the NICUs.
Main findings and relevant studies
The present study showed that physicians are involved in various EoL practices. EoL practices differed in association to the attitude score, with a higher attitude score relevant to favoring quality of life to limitation of intensive care and a lower score relevant to sustaining life by continuation of treatment; additionally, religion importance and position on legal framework were the main determinants of the attitude score.
Daglas et al study on the attitude of Greek healthcare professionals in NICUs20 reported a mean attitude score of 3.1 indicative that Greek healthcare professionals tended to support the value of human life. Contrarily our study reports an average attitude score of 5.8 indicative of a shift towards the value for quality of life. This difference could be attributed to various factors: a) different sample composition; In Daglas et. al. study 251 healthcare professionals participated and only 71 were physicians, while our study is focused on physicians’ opinion (156 doctors’ respondents), a more than the double the size sample, b) different approach to calculate the attitude score: Daglas et al. study calculated the attitude score as the sum of each participant’s answers to the 7 questions found important; in our study 6 items of the questionnaire that were found important were weighted by their loadings before extracting the final attitude score. Finally c) our study data were collected 10 years after Daglas et al. study (from May 2009-May 2011). During these years Greece faced a severe economic recession with many burdens imposed on the population on social and cultural level. Furthermore the fact that during the time elapsed from the above mentioned period, physicians have experienced survivors with major disability that were born at the limit of viability or had faced major events at birth, may have changed their personal beliefs and their way of thinking.
Recent data in the literature shows increased frequency of withholding or withdrawing treatment.23 EURONIC multicentre ethical research study and subsequent studies showed that withholding or withdrawing treatment is a common practice in several European NICUs and worldwide.1 13 24–27 In our study, up to 2/3 of physicians were involved in EoL practices, the commonest being discontinuation of treatment. One out of five (1:5) physicians decided withholding of treatment, while one out of ten (1:10) decided withdrawing of treatment: these ratios are lower than those reported in other European countries and much similar to that in South Mediterranean countries. The absence of a clear distinction between treatment withdrawal and assisted dying is a key factor in differing practices with treatment withholding.28 Not all professionals accept that withholding and withdrawing treatment are morally equivalent. Moreover, there are arguments that withdrawal of treatment leads to assisted dying.9 The lack of a clear legal framework for NICUs, legal constraints, nonexistence of firm policies -even within NICUs- and not in the open ethical debate discussions, lack of psychological support29 lead to a more conservative EoL approach by physicians.
Decision-making is a multifactorial task dependent on knowledge, relationships, life experiences and subjective approach as attitude for life issues.16 The majority of studies on end-of-life care in severely ill newborns describe physicians’ attitudes and not implemented practices.26 Physicians’ EoL reported practices were associated with the attitude score; physicians with a higher attitude score showed a tendency for a positive answer, while those with lower attitude score were prone to non-participating. Attitude scores did not differ significantly only when physicians “continued ongoing treatment without adding other therapeutic interventions”. Moreover, the higher the physicians’ attitude score, the more probable action was to intervene in EoL practices, showing a clear preference to limit life without hope, while a lower score was oriented towards prolonging life. Differences concerning the attitude among physicians in other studies could be attributed to the different cultural and social background, in addition to physicians’ characteristics.11–14 28 30 Rebagliato et al study showed that after controlling for confounders, country remained a significant predictor of physicians’ attitude score and practices suggestive of social and cultural factors.1
Physicians’ characteristics with a statistically different attitude score were education, involvement in research, religion importance and position on changing the current legal framework; in the final analysis only religiousness and law change belief remained as the main predictors of the attitude score. Religious beliefs are highly influential factors when making life and death decisions for infants, both for the physicians and parents.31
The effect of physician’s characteristics and beliefs shows that clinical factors, legislation and social culture are not the only predictors for ethical decisions. Indeed, as ethical decisions derive from personal moral principles and values, parents have the right to be informed on physicians’ attitudes and personal beliefs.19 31
Additionally, male gender, younger age, less time of experience, participation in research projects and most importantly law change belief were found to be strong indicators for EoL practices towards limiting intensive care.
Strengths and limitations
Strong points of the present study are that it includes a national representative sample and its prospective design, detailed data collection and since objectives were met, it may provide a ground for generalization. There are several limitations to be underlined. First, the attitude score was derived through factorial analysis by selecting those statements that showed a high correlation and internal consistency, as well as content validity with attitudes towards life value and support. Second, the subjective nature of the study cannot exclude underreporting of EoLDs, despite protection of anonymity and confidentiality.
Implications
EoLDs vary considerably between countries in the decision- making process, the degree of parental involvement and the clinical indications for therapeutic limitations. As EoLDs are a key-factor for dying neonates, cross-cultural comparison of EoL practices is important when NICU outcomes between countries are compared. There are no universal standards for treating neonates at EoL, and literature reveals these variations, with few studies examining EoL protocols. The extent to which physicians’ personal values and attitudes are associated with their practices of EoLDs remains to be clarified.