Study design

This study was conducted in three surgical intensive care units in National Taiwan University Hospital. Patients with medical uncertainty or conflict regarding value-laden issues, and requests made by the attending physicians or nurses for HCEC from December 1, 2009 to April 30, 2012 were randomly assigned, to the usual care group (UC group) and the intervention group (HCEC group). The patients in the UC group did not receive HCEC, but still received usual care such as family meeting, consultation to social workers and so on, which were considered appropriate by the health care team. The patients in the HCEC group received HCEC conducted by an individual ethics consultant. If a case was assigned to the HCEC group but the attending physician did not want to receive HCEC, the preference of not receiving HCEC was honored. If a case was assigned to the UC group but the attending physician wanted to receive HCEC, the preference of receiving HCEC was honored.

Data collection

Data was concurrently and retrospectively collected from the review of medical records and HCEC records. We collected independent variables such as patient demographics and severity of illness as indicated by the total of Elixhauser comorbidity measures. Although the total of Elixhauser comorbidity measures is not an acute severity score, it is the only severity of illness which can be collected in all the three intensive care units. Three groups of outcome measurements were also collected: first, patient status at hospital discharge; second, consumption of medical resources as indicated by length of ICU stay, length of hospital stay, post-conflict length of ICU stay, and post-conflict length of hospital stay; and third, whether a consensus regarding patient care was achieved, as an indicator of whether the goal of HCEC was achieved. A consensus regarding the goal of medical care was achieved in the HCEC group if any of the morally acceptable options suggested by the individual ethics consultant was followed, and in the UC group if patients/family members and health care team members agreed on any options for the goal of medical care. Health care team members (i.e., the nurse in charge of the patient, head nurse, or the primary care resident) were contacted about whether there was a consensus on the goal of medical care after the onset of medical uncertainty or conflict regarding value-laden issues in the UC group, or after the HCEC was done in the HCEC group.

Intention-to-treat principle

Data analysis was based on the intention-to-treat principle, implying that the comparison of variables is based on the initial treatment assignment and not on the treatment eventually received. For example, if a patient with medical uncertainty or conflict regarding value-laden issues was randomly assigned to the HCEC group, but ultimately did not receive HCEC service, the patient was still retained in the HCEC group when the data were analyzed.

Statistical analysis

Our analysis examined univariate characteristics (central tendency, dispersion, and distribution) and bivariate relationships (correlations). These exploratory techniques were based on proportions (categorical variables) and medians (variables measured on an interval scale). The appropriate non-parametric tests, such as Mann–Whitney test and the Chi-squared test, were used depending on the scale of measurement. All statistical analyses were executed using STATA/MP 11.0 for Windows PC. This study was approved by the Research Ethics Committee in National Taiwan University Hospital.

Main findings

Examining the effectiveness of HCEC demonstrated that HCEC was associated with achieving a consensus regarding the goal of medical care, shorter length of entire ICU and hospital stay, and shorter length of ICU and hospital stay after patients encountering medical uncertainty or conflict regarding value-laden issues. In addition, patients in the HCEC group did not have a higher mortality rate than those in UC group at hospital discharge.

Strengths and limitations

This is the first study to evaluate the effectiveness of HCEC conducted in East Asian medical encounters, where the core value of medical decision-making may be distinct from North America/Europe [21]. Our individual ethics consultants were encouraged to follow the way of conducting HCEC suggested by the scholars from North America [20]. In addition, we proposed several novel outcome measurements to evaluate the effectiveness of HCEC, such as whether a consensus regarding the goal of medical care was achieved, the length of ICU stay and hospital stay after the occurrence of ethical conflicts. Whether a consensus was achieved is a better outcome measurement than most of the outcome measurements reported in the literature because it conforms to the goals of HCEC [1]. Lastly, our study used randomization and intention-to-treat principle to evaluate the effectiveness of HCEC [11, 12]. Although there are some concerns about incomplete double blindness [14], randomization and the principle of intention-to-treat are still considered the most rigorous study design to evaluate the effectiveness of HCEC.

Our study has limitations. First, this is a single center study. The generalizability of the study results may be limited. Second, the differences observed in the outcome measurements could be overestimated because the involved parties in this study were not blinded regarding the HCEC service. If health care team members in the HCEC group reflected greater enthusiasm than those in the UC group because they knew that a HCEC service is being conducted for their patient, the differences observed in the outcome measurements might be partly associated with the greater enthusiasm. Third, although we evaluated the effectiveness of HCEC using one of the goals of HCEC as the outcome measurement, whether the other goals of HCEC were achieved was not examined. Fourth, some of the readers for this paper might be concerned that this is a self-reporting success because only the health care team members were inquired regarding “whether a consensus was reach”. However, until now, the health care team members still do not know that we evaluate the effectiveness of HCEC using “whether a consensus was reached” as an outcome measure.

Ethical issues for requesting health care ethics consultation

Several studies have reported the ethical issues which the patient or health care team members encountered for requesting HCEC. La Puma et al. reported that 49% of the cases requested HCEC for assistance with withdrawing or withholding life-supporting treatments, 37% for resuscitation issues, and 31% for legal issues [16]. Another study conducted by La Puma et al. showed that 74% of the cases requested HCEC for the decisions to forgo life-supporting treatments, 46% for resolving disagreement, and 30% for assessing patient competence for decision-making [15]. A recent study conducted by Johnson et al. reported that the requesters sought assistance with end-of life issues in 47% of the cases, in 41% of cases for shared decision-making, and in 14% of cases for professionalism [5].

Most of the HCEC requesters sought assistance with more than one issue. In addition, the majority of requests made for HCEC surrounded end-of-life issues, e.g. withdrawing or withholding life-supporting treatments, cardiopulmonary resuscitation/do-not-resuscitate, and disagreement. These studies reflected the educational needs for health care workers in resolving medical uncertainty or conflict regarding value-laden issues. They also demonstrated the issues that an ethics consultant should be familiar with, and capable of resolving.

Empowerment for requesting health care ethics consultation

According to Johnson et al., most of the requests for HCEC were placed by house officers (63%), nurses (12%), and attending physicians (11%) [5]. Our study showed that 37 (59.68%) of the 62 requests for HCEC were made by the primary care nurses or head nurses, and the remaining were made by the attending physicians. Both studies revealed that non-MDs participate in requesting HCEC service, particularly in our study, in which a higher percentage of HCEC requests were made by non-MDs than in the Johnson study. This may be associated with the fact that the nurses in the three ICU setting were encouraged to assist patients by requesting HCEC services.

We also identified that the physicians were more likely to request HCEC when encountering the issue of disagreement between health care team and family members than the issue of disagreement between health care team members. For the 37 disagreements between health care team and family members, 14 (37.84%) of them were requested by the attending physicians. For the 16 disagreements between health care team members, only three (18.75%) were requested by the attending physicians. These findings may imply that nurses are more likely than attending physicians to identify disagreements between health care team members as a problem, and that the attending physician may not see the disagreement as a problem or not be aware of the disagreement.

Appropriateness of outcome measurements

Numerous empirical studies have been initiated because of concerns related to accountability and quality assurance in HCEC. Many of these studies reported findings on physician satisfaction [11, 12, 15, 16], and physician’s perception of clarifying ethical issues, educating the health care team, making clinical decisions with confidence, and in patient management [18]. However, the satisfaction or perception of the parties involved may be influenced by factors not associated with the quality of the HCEC conducted [17, 22]. For example, a physician may be satisfied with the HCEC service provided because his/her suggestion is adopted by the ethics consultant and not because of the quality of the HCEC service.

Schneiderman et al. reported that, for those who did not survive to hospital discharge, HCEC was significantly associated with lower cost, shorter ICU stays, shorter hospital stays, and less use of life-supporting treatments. The studies also showed that HCEC was beneficial to patients who did not survive to hospital discharge [11, 12, 23].

The appropriateness of randomized controlled trials in evaluating HCEC has been a concern because these trials are not double-blinded [14]. Moreover, researchers have argued that monetary saving should not be included as an outcome measurement to evaluate the effectiveness of HCEC because lowering costs is not one of the goals of HCEC [13]. Rigorous scientific research results supporting the effectiveness and quality of HCEC appear to lag far behind the rapid growth of HCEC and concerns about its accountability and quality assurance.

The effectiveness of HCEC is evaluated based on measurable outcomes that are consistent with the intended goals of HCEC [17]. If the intended goals of HCEC are followed, the quality of HCEC is satisfactory. Therefore, to determine the outcome measurements required to evaluate the effectiveness and quality of HCEC, we can refer to the established goals of HCEC [9].

The Society for Health and Human Values-Society for Bioethics Consultation Task Force on Standards for Bioethics Consultation organized a consensus panel with professionals and experts [1]. The goals of HCEC were proposed by the task force. As pointed out by the task force, we noted that whether a consensus regarding the goal of medical care was achieved is a critical outcome measurement for evaluating the effectiveness of HCEC. Therefore, whether a consensus regarding the goal of medical care was achieved was examined and compared between the HCEC group and UC group in this study. We identified that HCEC services facilitated achieving a consensus regarding the goal of medical care effectively, thus conforming to the goals of HCEC proposed by the American Societies for Bioethics and Humanities.

The results of our study agree with those of several previous studies showing, for example, that HCEC is associated with short lengths of entire ICU stay and entire hospital stay [11, 12]. However, the entire ICU stay and hospital stay in our study were considerably longer than those reported by Schneiderman et al. Therefore, we examined the outcomes of HCEC by using the length of ICU stay and hospital stay after the occurrence of medical uncertainty or conflict regarding value-laden issues, which more directly measured the influence of HCEC than the length of entire ICU stay and entire hospital stay.

Cultural differences in conducting health care ethics consultation

Our ethics consultants were encouraged to conduct HCEC following the ethics facilitation approach as proposed by Aulisio et al. Part of the rationale to support this approach to conducting HCEC in the U.S., according to Aulisio et al. [20], are that the U.S. is a pluralistic society, and the main societal value is individual autonomy. To honor each moral stakeholder from different racial/ethnic backgrounds, and also to uphold the societal value of respecting individual autonomy, the voice of each moral stakeholder surrounding the ethical conflict should be heard, and his/her preferences should be respected. Therefore, ethics facilitation approach for conducting HCEC is highly suggested in the U.S.

However, the ethics facilitation approach to conduct HCEC in the medical encounters in Taiwan might be of concern because individual autonomy may not be the main societal value. For several thousand years, Confucian philosophy has deeply influenced societal values, and ethical considerations in East Asian countries such as Taiwan [24]. One phenomenon rooted in Confucian philosophy highlighting the difference between East Asian countries and North America/Europe is the locus of authority in decision-making: North America/Europe demands and promotes the value of individual autonomy; East Asian countries typically honor and uphold the value of family autonomy [25]. Although the ongoing westernization of East Asian biomedical ethics in Taiwan is convincing, family autonomy seems to remain as the main societal value [21]. As such, the appropriateness of applying the ethics facilitation approach to conducting HCEC in Taiwan’s medical encounters should be further deliberated.

Before this study was conducted, HCEC services were not formally announced to National Taiwan University Hospital. There were only few formal and informal HCEC services conducted by two individual ethics consultants who had several years of clinical ethics training as well as medical training. Currently, given that HCEC services have been formally announced to National Taiwan University Hospital and the institutional supports in place for HCEC services, a group of individual ethics consultants (composed of physicians, nurses and social workers) are conducting daily HCEC services, and, as a result, healthcare professionals’ requests for HCEC are dramatically increasing.