St. Francis Medical Center
Pittsburgh, Pa 15201
Health care providers are trained to saved lives not to watch them helplessly slip away. Many physicians fear legal implications if a mutually acceptable decision affecting impending death, is not made with the patient or surrogates. When physician/family conflicts arise regarding the most appropriate treatment options, some physicians argue that the physician's decisions should prevail if no compromise can be made. Reasons cited in support of this viewpoint include:
Conversely, criticisms of physician autonomy in end of life decisions relate to the following issues:
The difficult decisions associated with the dying process, together with the advances made in medical technology, have led to concept of advanced directives, living wills by patients, the appointment of durable power of attorneys for health care, and surrogate decision-making by ethics committees, legislative rulings, and family members. Yet the efficacy of these concepts have been criticized. Families may simply be intellectually or emotionally incapable of dealing with the problem effectively. Questions have been raised as to whether some surrogates may execute advanced directives according to the patients instruction, or by imposing their own personal preferences (Lo 1995). There may be conflicts of interest, such as squabbles over inheritance between family members, or even possible threat of deportation for questionably legal aliens after the death of their citizen spouse.
Living wills have become increasingly fashionable but they can be a problem of authority and interpretation. They have limitations and legal provisions that vary by state. A physician has no way of knowing if the patient understood the document when it was formulated, meant it the way the physician interprets it, or currently feels the same about it as the day it was written. Durable power of attorneys are more comprehensive than living wills and often prove to be beneficial when conflict among family members is anticipated. But forms need to be witnessed or notarized, and must be consistent with decisions previously expressed by the patient (Burnell 1993, Lo 1995). The purpose of ethics committees is to facilitate the decision making process, by analyzing the ethical issues, improving communication between health care members and/or between health care members and patients or their families, while offering recommendations to resolve conflicts at hand. While the advantages of ethics committees have been documented, critics have challenged the issues of ethic consultants' expertise in making such decisions, the delay of consults, and the diffusion of responsibility among its members (Lo 1995).
Some families express feelings reflective of false hope; i.e., that if life support systems can maintain vital signs for a day or a week, "suspended animation" has the potential to sustain someone until a cure or treatment is found. These opinions are frequently supported by anecdotal articles in the popular press attesting to patients who have "awakened" after years of coma. The occasional inability of the above measures to formulate or negotiate a decision regarding an individual's life sustaining or life preserving intervention, often results in legal rulings (Miles 1992, Koch et al. 1992, Lo 1995). Landmark cases such as the Quinlan case, the Brophy case, and the Cruzan case, to name a few, have provided general guidelines for clinical practice decisions.
However, physicians, patients, and surrogates are not the only viewpoints that require consideration. Allied health care professionals, particularly nurses, are often asked to implement the decisions of physicians, without consideration of their individual viewpoints on the issues. Disagreements over, or discomfort with procedures nurses may feel are inappropriate or unethical often require delicate communication to supervisors, physicians and administration. However, such potential areas of friction may jeopardize a given nurse's position in the hospital's complex network (Jameton 1992, Lo 1995). Ideally, discussions between physicians, nurses and related staff should be carried out to provide physicians with an opportunity to explain their treatment, or its withdrawal when concerns arise. Although this appears to be an effective solution, it is not always the course of action chosen. Jameton (1992) notes that even though nurses have the right and responsibility to attempt to refute the logic of physicians, "the physician has no well-defined responsibility to listen to the nurse, and the physician's special expertise creates a presumption in the physician's favor" (p.104). Furthermore, the role of nurses as care-givers and patient advocates, has been cited as a complicating factor when emotional medical decisions need to be made (Jameton 1992).
The relationship nurses develop between patients and their families is often more complex than the one between physicians and their patients and/or families. This difference can be attributed to the amount of time that nurses spend with the patients in comparison to physicians. When disagreements arise over end of life decisions, it is not uncommon for this distinction to be cited as the root of the problem; i.e., the conflict between nurse professionalism and personal feelings about patients (Jameton 1992, and Lo 1995). The ability of all health care professionals to show compassion and empathetically relate to patients is essential to quality care, but more crucial is the ability to employ nonemotional and potentially unpopular decisions when required. The misconception that nurses are unable to distinguish the two, may be an issue of interprofessional relations that requires further attention.
The logistical difficulty, expense, and time constraints associated with assessing in real time, data regarding international decisions on life-sustaining or medically futile treatments have been largely unavailable. The unique forum of a medically oriented internet resource, allows such issues to be addressed and discussed among colleagues more efficiently than traditional survey methods. The advent of list servers permitted physicians and other health professionals to maintain "closed" interest and discussion groups serving a wide, yet consolidated population. Using a listserver, messages sent from one individual are simultaneously sent to only those others who have subscribed to the list. This allows interest groups to avoid unwanted "browsers." In August 1994, the first listserve group dedicated to the specialty of critical care medicine was founded. The purpose of the Critical Care List was to provide a forum to discuss and maintain a data bank for the holistic, high technological, often life-sustaining management of patients in the ICU setting.
The purposes of this pilot study were threefold: First, we sought to evaluate the usefulness of the ever-expanding internet as a tool to assess medical viewpoints, secondly, we hoped to be able to determine the international perceptions regarding the care of patients near the end of life, and finally we aimed to evaluate the reliability, face validity, and testing environment of the survey.
Methods:
The pilot survey conducted consisted of a case history involving futile care in the ICU followed by questions regarding attitudes and responses of health care professionals on related issues about life-sustaining treatments (See Appendix A and B). The information was used to collect international data on physician and allied health (primarily nurses) professionals to assess interprofessional, cultural, gender, and years of practicing experience differences, inherent in treatment decisions once a patient's condition is perceived to be futile. Two international internet discussion groups, CCM-L (critical care medicine) and NurseRes (nursing research) were selected for the postings. The moderators of both lists supported the surveying of its members.
The sampling pool consisted of physicians (both attending and residents), nurses, respiratory therapists and pharmacists. A preliminary posting was posted on each list informing members of the upcoming survey. Similar surveys on attitudes regarding medical futility for physicians and allied health care professionals were posted; making adjustments for the scope of practice for each profession.
Survey Construction:
The survey was constructed primarily by the authors, who do not claim to be experts in the field of medical ethics. The selection of items was based on controversial topics in the current literature of life sustaining and futile treatments. Respondents were invited to comment on the face validity of the survey, constructively criticize its contents, and to provide additional remarks on the topic.
Analysis:
Individual items were evaluated with frequencies and mean scores. Categorical and ordinal responses were cross tabulated by gender, location (USA or non-USA), affiliation (academic and non-academic), years of experience, and profession. Factor analysis was used to simply the data by identifying factors that represented the relationship between the set of items. Regression analysis was performed to determine if a significant relationship existed between the resulting factors and the descriptors noted above (location, affiliation, years of experiences, and profession). Standardized scores were used when combining various likert scale formats. The statistical analyses were performed using the personal computer version of the BMDP Statistical Software package.
Results:
Reported Number of Futility Cases
Physician Survey
Physician respondents report a mean number of 4.90 (SE+0.87) futility cases per year. Of the physician respondents, academically affiliated physicians (69.8% of respondents) reported a mean number of 5.30 (SE+1.78) futility cases, in contrast to non-academically affiliated physicians, combined affiliation or private practice, (30.2% of respondents) who reported a mean number of 4.00 futility cases (SE+ 0.94). Comparison of physician responses according to location, U.S.A. and non-U.S.A., revealed that U.S. physician respondents reported a mean number of 6.05 (SE+1.09) medical futility cases per year, while non-USA physicians reported a mean of 1.71 (SE+0.77) futility case per year. No significant difference in physician responses according to affiliation or location were found.
Allied Health Care Professionals Survey
The allied health survey was completed by 52 (91.2%) nurses, 2 (3.5%) respiratory therapists, and 3 (5.3%) pharmacists. Among the 52 nurses, 32 (61.5%) were intensive care nurses, 8 (15.4%) were advanced practice nurses, 8 (15.4%) classified themselves as "other nurses" -- but were primarily research nurses, 2 (3.9%) were staff nurses, 1 (1.9%) was an intermediate care nurse, 1 (1.9%) was a nursing administrator.
Allied health care professionals reported that they were aware of a mean number of 7.56 (SE+1.06) of patients' families that requested the medical staff to implement or continue care that they considered to be futile life-sustaining treatment. Of these 70.4% were observed by USA health care providers, while only 29.6% were noted by non-USA health care providers (p 0.05). Health care providers employed in an academically affiliated institution reported that approximately 57.4% of patient's families requested the medical staff to implement or continue care that they considered to be futile, in comparison to 42.6% observed by those health care providers employed in private, community or some combination of private/community/academic affiliated institution.
Allied health care providers reported that physicians requested their input in the decision to withdraw or withhold care for reasons of futility in 61.54% of the cases; academic physicians 44.83% of the time and non-academic physicians 59.09% of the time. The percentage of families that sought the input of allied health care professionals in their decision to discontinue life supportive measures varied little between academic and non-academic institutions, 50% and 48.39% respectively. Globally, these percentages indicate that 22.22% of USA and 100.00% of non-USA physicians requested the input of the health care staff in the decision to withdraw or withhold treatment in futile cases. Similarly, only 29.73% of families in USA hospitals, compared to 93.76% of families in non-USA hospitals sought the input of their family member's allied health care providers in their decision to discontinue life supportive measures. Accordingly, location (USA versus non-USA) was noted to be a significant factor (p < 0.05) in both the approximation of the number of families that requested the medical staff to implement futile life-sustaining treatment, and of those requesting the continuation of futile care, those who sought the opinion of the allied health care staff when making their decision.
Physician Response when Ambivalent about the Withdrawal of Life Supportive Measures:
Physicians were asked "when ambivalent about whether to withdraw life support" which action do you usually take: 1.) refer the decision to another physician on the case, 2.) follow the family's wishes, 3.) allow the institution's ethics committee to make a decision, or 3.) wait to make a decision until I feel more confident to do so. Table 1 presents the responses of physicians by affiliation.
Both academic and non-academic physicians expressed that most often when confronted with ambivalent feelings about whether to withdraw life support, they wait to make a decision until they feel more confident to do so (66.7% and 43.7% column totals respectively), with following the family's wishes next (27.8% and 37.5% column totals respectively). As noted in Table 1, caution must be made in the generalizability of these results since twice as many academic physicians responded to this question as non-academic physicians. Comparison of responses according to location (USA versus non-USA) resulted in similar findings. 56.4% of USA physicians and 69.2% of non-USA physicians expressed that they would wait to make a decision until they were more confident to do so when faced with ambivalent feeling regarding the withdrawal of life sustaining measures. However, the second most often reported attitude when ambivalent feelings were present for USA physicians was again to follow the family's wishes (38.5%) while non-USA physicians indicated that they would refer the decision to another physician on the case (23.1%). Once again, due to the greater response rate by USA physicians (75%) compared to non-USA physicians (25%) care must be taken in the interpretation of the results to this question.
Physicians' and Allied Health Care Providers' Views on Futile Life-Sustaining Measures:
Physician Survey
Respondents were asked 15 questions with the purpose of detecting the extent to which factors commonly reported in the literature, affected their decisions with regard to the care of patients near the end of life. All fifteen questions were placed in a likert scale format; 6 questions with a three scale format (never, sometimes, and always) and 9 questions with a five scale response format (strongly disagree, disagree, neural/indifferent, agree, strongly agree). Table 2 presents the overall distribution of responses in the three scale format, while Table 3 presents the overall distribution of responses in the five scale format. All fifteen questions were subject to a principle components factor analysis using direct oblimin rotation. Oblique rotation was chosen because of its appropriateness for scale evaluation, in that it allows the factors to be correlated with a greater likelihood that each item will be associated with a single factor. The factors were sorted accorded to the sum of squared loadings for each factor. The four distinct, yet interrelated factors contributing to the physician attitudes on life supporting measures were named: 1.) Availability of health resources - comprised of questions 3F, 5F, 5G, 5H; 2.) Personal responsibility - consisting of questions 5A, 5B, 5D, 5I; 3.) Professional multi-disciplinary approach to decisions on the appropriateness of clinical treatments - including questions 3A, 3B, 3D, 5E; and 4.) Reluctance to family participation in treatment decisions - consisting of questions 3C, 3E, 5C. No question had a loading greater than 0.50 on more than one factor. The factors and their corresponding loadings are provided in Table 4. Cronbach's alpha coefficient for the 15 items was 0.53.
Regression analysis to examine the effect of respondent location (USA versus non-USA), affiliation (academic versus non-academic), and years of experience, was performed with each of the four factors as dependent variables. No significant association was noted between factors one (availability of health resources) and four (reluctance to family participation in treatment decisions) and location, affiliation, or years of experience. A significant association (p 0.05) was noted between factor two (personal responsibility) and location , and between factor three (professional multi-disciplinary approach to decisions on the appropriateness of clinical treatments) and both location and affiliation. The final regression models from this data for both factors were:
Factor 2: Personal Responsibility = 8.0680 - 1.9469(Location)
Factor 3: Professional multi-disciplinary approach to decisions on the appropriateness of clinical treatments = 8.0066 - 1.6856(Location) + 1.0898(Affiliation)
Allied Health Survey:
Respondents were asked 14 questions which were similar in design and content to those asked of physicians, with the purpose of determining the extent to frequently identified factors influenced the attitudes, opinions, and behavior of allied health care persons caring for patients near the end of life. Eleven of these fourteen questions were placed in a likert scale format; 3 questions with a three scale format (never, sometimes, and always) and 8 questions with a five scale response format (strongly disagree, disagree, neural/indifferent, agree, strongly agree). In addition, three questions assumed the yes or no format. Table 6 shows the overall distribution of responses in the three scale format, while Table 7 shows the overall distribution of responses in the five scale format. A principle components factor analysis using direct oblimin rotation was performed on the standard scores of the eight questions in the five scale likert format. As noted in the physician survey analysis, oblique rotation was chosen because of its appropriateness for scale evaluation, in that it allows the factors to be correlated with a greater likelihood that each item will be associated with a single factor. One salient factor closely resembling an a-priori identified domain was interpreted as "the utilitarian approach to treatment decisions." Cronbach's alpha coefficient for the 8 items was 0.66.
Regression analysis to examine the effect of respondent location (USA versus non-USA), affiliation (academic versus non-academic), and years of experience, was performed with the one factor (utilitarian approach to treatment decisions) as the dependent variable. No significant association was noted between this factor and affiliation or years of experience. A significant difference (p 0.05 was noted between the factor (utilitarian approach to treatment decisions) and the location (USA versus non-USA) of the respondents. The final model from this data was: "Utilitarian approach to treatment decision" = 16.9757 + 3.7632(Location).
It can be noted from Table 5 that 83.9% of allied health care professionals report that they are adversely affected, at least sometimes, when physicians do not involve them in the termination of life supportive measures from their patients. A large percentage of the allied staff surveyed (83.9%) report that they would convey their feelings to the attending physician if he/she did not involve them in the discussions of their patient's end of life treatment decisions. Even more important, 86.6% of the allied staff who responded, stated that they would refuse to disconnect life support when/if ordered to do so by the patient's physician, if they felt the decision was against the family's wishes. These statistics support the concept of interprofessional communication of moral and ethical decisions that foster complementary health provider views on treatment decisions in their presentation to families.
Although the above findings note the substantial dissatisfaction experienced by allied health care providers when excluded from end of life treatment decisions concerning their patients, less than 15% of those surveyed asserted that they would disclose such cases to their superiors. The complexity of reasons associated with this apparent discrepancy among allied staff perceptions and actions may relate to their perceived subordinate role, poorly defined responsibility and authority in these matters, and fear of repercussions that may result from challenging a physician's decision. However, it is interesting that 8.8% of the allied staff report that they would discretely notify the media when hospital, family and courts all conflicted on how to best care for the needs of the patient. Whether or not these professionals would actually contact the media is unknown, however, the fact that 8.8% of them would consider doing so reveals their commitment to being the patient's advocate. More damaging, is the potential nightmare that such actions could cause the institution.
Discussion:
Recent times have witnessed much discussion regarding the "life is sacred at any cost" philosophy (Ramos and Reagan 1989, Health Technology Case Study, 1984, and Campion 1981). Current technology is capable of indiscriminately maintaining some of the vital functions of the body, but the same technology does not necessarily allow us to heal underlying disease processes. Allotting scarce and expensive resources for potentially nonviable patients limits the amount which can be spent on potential survivors, and escalating medical costs make some form of rationing inevitable (Englehardt and Rie 1986). The problem has been how to determine which patients should receive the maximum benefit of intensive care technology and where to draw the line when a point of diminishing return is reached (Detsky 1981, Fedullo et al. 1983).
The above issues on life-sustaining treatments and medical futility are primarily Western viewpoints. Kilner (1992) examined the African perceptives on "Who shall be saved?" His results showed that the greater the level of education (from none for traditional healers to college educated for physicians) the more emphasis is placed on saving the lives of those who will be able to provide the greatest amount of societal benefit. In the African view, greater patient societal usefulness, generates more legitimate allocation of the scarce medical resources and personnel. Scarce resources limit the number of viable alternatives. Furthermore, it can seen that moral judgements are sensitive to cultural differences. Life-sustaining and medically futile treatments may not only be impossible due to the scarcity of the resources, but also conflict with those taught and encouraged within the country (Kilner 1992). Conversely in the Netherlands where medical resources are not as scarce as in developing countries, euthanasia (the compassionate termination of one's life by either active or passive measures taken by another person) is legally accepted on a case by case basis (Burnell 1993). Guidelines for euthanasia to be considered legal in the Netherlands do exist, but the practice is widely accepted. An important factor noted by Burnell (1993) is, "physicians are rarely, if ever, the subject of malpractice actions by patients . . . the doctor-patient relationship in the Netherlands is uniquely strong and is based on a deep mutual trust. The practice of defensive medicine is unknown, and the concept of liability is rarely mentioned in the context of medical practice" (p. 268).
Physician Survey:
This study is the first that we know of that aimed at identifying current global attitudes on life-sustaining measures as expressed by health care professionals, via an internet survey. Our data revealed that physicians in the United States report a greater number of futility cases per year than those located outside the United States, mean of 6.05 (SE 1.09) and 1.71 (SE 0.77), respectively. These numbers appear to support the premise that families of moribund patients in the United States seem to make decisions regarding continuation of life support based on the perception that these services are readily available to all, with little consideration given to the scarcity of health resources. Furthermore, Medicare law makes it illegal for families to be billed for any ICU care. The lack of financial responsibility that U.S. families must assume, provides little incentive to authorize withdrawal as long as the patient doesn't look uncomfortable and a glimmer of hope for recovery is held by the family. Anecdotal articles in the American popular press attesting to patients who have awakened after years of coma often confound the issue for families clinging to the notion that if life support systems can maintain vital signs for a day or week, "suspended animation" should be possible indefinitely.
Noting that futility cases appear to differ by location (USA versus non-USA) of physician, we sought to determine if they also differ according to practice affiliation. Our data showed that physicians in academic practice encountered a greater percentage of patients in futile situations than those in private practice, 69.8% compared to 30.2%, respectively. This discrepancy may be attributed to the reluctance of private hospitals to maintain patients on futile life sustaining support systems. Alternately, it may also be due to resource scarcity at the primary care level compared to tertiary referral centers. Other possibilities may include, that physicians in smaller community hospitals are more general practice oriented, may have developed a closer relationship to patients and families, may be in a more authoritative position to steer families away from emotionally derived decisions about futility, or simply may not see the number of critically ill patients seen by academic physicians. Similarly, private practice physicians on average, with all other factors held constant, are more likely to seek a multi-disciplinary approach to futility decisions, than are academic physicians. Again, the reasons for this may simply be rooted in a desire to diversify the authority for potentially unpopular and emotional decision making among close knitted physician-patient and/or physician-physician relationships. It may also be that University physicians perceive themselves as having more individual autonomy to make decisions within the Ivory Tower.
Globally physicians do not appear anxious to withdraw life support from any patient. Table 1 reflects that most physicians, 46.2% of academic physicians, and 13.5% of non-academic physicians choose to postpone a decision on the withdrawal of life support when they are ambivalent about the most appropriate course of action, with "following the family's wishes" being the next most likely action. This supports the belief that most all physicians eventually reach a satisfactory comfort level in what they feel will most benefit the patient. It also provides evidence that the physician/patient and/or surrogate relationship is recognized as an important partnership in the handling of such sensitive decisions.
Tables 2 and 3 list the questions that physicians were asked to respond to based on their general practice patterns. The principle components factor analysis, using direct oblimin rotation, resulted in a four factor structure that best represented the dimensionality of the items. The four factors and their corresponding loadings are provided in Table 4. All questions met the widely accepted minimum factor loading of 0.30 (Cattell 1978, Nunnally 1978, Carmines and Zeller 1979, and Gorsuch 1983). The overall alpha reliability coefficient was 0.53. This is below the commonly accepted 0.70 minimum for new scales (Nunnally 1978), but was anticipated due to the small number of questions. However, we felt that the chosen forum and the time constraints of our target population prohibited a longer, and perhaps more comprehensive survey. A better approach may have been to limit the parameters being addressed, thereby allowing more questions related to each concept to be posed.
Regression analysis revealed that factor 2, named "personal responsibility", was significantly associated ( p < 0.05) with location. This means that location accounted for the predictive power of this factor. Thus, the predicted "personal responsibility" response for physicians located outside the USA is approximately 2 less than the predicted "personal responsibility" response for physicians located in the USA. The summated scale for this factor ranged from 3.90 to 16.99, with a mean response of 7.51 (SD+2.58). Because the "personal responsibility" factor was comprised solely of questions from the 5 category likert scale format, the lower the response number on the summated scale from 4 through 20, the more likely the physician disagreed with the questions comprising the factor. Accordingly, this indicates that non-USA physicians on average, are less likely to pacify families in an attempt to protect themselves from potential conflicts, either confrontational or legal, that may result from their actions.
We also noted from the regression analysis that factor 3, "professional multi-disciplinary approach to decisions on the appropriateness of clinical treatments" was significantly associated (p < 0.05) with location and affiliation. The final model for this factor was: Factor 3 = 8.0066 - 1.6856(Location) + 1.0898(Affiliation).
Factor 3 consisted of 3 questions (3A, 3B, and 3D) from the three category likert scale format, and one question (5E) from the five category likert scale format. Question 5E had a negative loading of -0.774, indicating that it is negatively correlated with the remaining 3 questions all of which had positive loadings. The summated scale for this factor ranged from 3.85 to 10.96, with a mean response of 7.86 (SD+1.93). Because this factor was comprised mostly of questions from the 3 category likert scale format, the lower the response number on the summated scale ranging from 1 through 11, the more likely the physician responded negatively to the questions comprising the factor. This indicates that non-USA physicians on average, with all other variables held constant, are less likely to seek a multi-disciplinary approach when a decision on life sustaining treatment is required. Furthermore, they are more likely to avoid outside assistance from ethics committees, other physicians, or consolidation to family's wishes, rather they feel that their objective, scientific decisions should prevail.
The reasons for this predisposition may lie in the supposition that non-USA physicians have more authority to control treatment choices. It may also be true that the resources available to maintain futile care outside the USA are so marginal that to use them in such a fashion visibly removes the possibility of meaningful care for large populations of patients with curable diseases. It is also possible that non-USA physicians are more likely to line up against requests for futile care because they know they are much less likely to be sued for defying familial wishes than in the USA. Finally, it is possible that non-USA physicians have not been tainted by pervasive technological care plans rampant in the USA, and have a better perspective of the point of diminishing returns.
Regardless of the underlying issues, our findings are in agreement with those previously reported. In 1986, the American Medical Association conducted a survey of 1,000 physicians, of whom 78% favored the withdrawal or withholding of life sustaining measures from terminally ill patients, only if the decision was supported by the family. The general finding from this survey was that American physicians followed the consensus of the American public. A subsequent survey by the Physician's Management magazine (1991) revealed that 58.5% of the 2,000 physicians surveyed, admitted to removing life-sustaining therapies in certain circumstances. The largest obstacle reported by these physicians in easing a terminally ill patient's dying process, was the fear of legal prosecution. Although public opinion polls reveal an increasing trend toward the approval of euthanasia, up to 90% of those surveyed in some places, the large number of futility discrepancy cases noted by physicians, appears to indicate that people may diverge from their beliefs when the time comes to actually make the decision to terminate a loved one's life support (Kanoti and Orlowski 1991, Burnell 1993).
Allied Health Survey:
The largest group of allied health respondents were nurses, 91.2%. Of these, 61.5% were hospital-based intensive care (ICU) nurses. According to our nurse respondents, families utilizing academic centers were more likely to request futile care than families with loved ones in private facilities, 57.4% compared to 42.6%, respectively. This implies a public perception that financial resources are more readily available in academic centers. However, on a more professional note, non-academically affiliated physicians were more likely to request the input of allied health providers than academic ones, 59.09% and 44.83%, respectively. Plausible explanations for this finding include the need for a closer working relationship between private practice physicians and allied staff, as the academic physicians are more likely to spend greater periods of time in the ICU due to the proximity of their practice in comparison to some physicians in private practice. The roles and relationships of medical students, residents and fellows in relation to the attending physician may also contributed to the lack of input these physicians seek from the allied staff.
This concept became even more striking when comparing USA physicians to non-USA physicians. In so doing, we noted that relatively few of the USA physicians (22.22%) requested the advice of allied care workers concerning futility decisions, in comparison to the non-American physicians. Non-USA allied health care professionals reported that their physicians requested their advice on issues concerning futility 100% of the time. This appears to indicate that non-USA physicians and allied care providers function more as a team than in the USA, where physicians usually writes the orders, and the allied care provider carry them out.
Most allied health care professionals (83.9%) report that they are adversely affected, at least sometimes, when physicians do not involve them in the termination of life supportive measures from their patients. The percentage of those suffering psychological stress is increased in American ICU's, where allied health care providers are placed in a more subservient position than those in foreign hospitals. 83.9% stated that they would convey their feelings to the attending physician if not involved in the discussions of their patient's end of life treatment decisions. Even more important, 86.6% stated they would refuse to disconnect life support when/if ordered to do so by the patient's physician, if they felt the decision was against the family's wishes. However, less than 15% of those surveyed reported that they would disclose such cases to their superiors, indicating their perceived liability for job loss or sever censure. Whether or not these professionals would actually redress grievances by contacting the media remains unknown. These statistics support the concept of inter-professional communication of moral and ethical decisions that foster complementary health care provider views on treatment decisions in their presentation to families.
Unfortunately, the nurse's perception of the public's opinion on his/her role as a professional involved in the care planning of patients, also seems to be more respected by the people outside the USA. Nurse's employed in hospitals outside the USA report that 93.76% of families sought their input in withdrawal of life support issues, whereas only 29.73% of USA nurses reported that families sought their input on such issues. Why nurses appear to viewed in a lesser light by American physicians and public may be rooted in the tradition of nurses as "comforters" rather than primary healers. The notion that nurses simply implement the orders given to them by physicians, is seemingly not as antiquated as some professional nurses would like. The public image of nurses as more passive in the nurse-physician unit is a tradition that continues to be nurtured in the popular press, on television, and in motion pictures.
Table 7 lists the questions that allied health care providers were asked to respond to based on their opinions and attitudes toward life sustaining measures. The principle components factor analysis, using direct oblimin rotation, resulted in one salient factor. This factor was named, "utilitarian approach to treatment decisions." The corresponding loadings are presented in Table 8. Six of the questions met the widely accepted minimum factor loading of 0.30 (Cattell 1978, Nunnally 1978, Carmines and Zeller 1979, and Gorsuch 1983). Two questions did not load significantly on this factor, 3A -- role as patient advocate (-0.084), and 3H -- contribution of possible legal repercussions (-0.025). Since one's responsibility as patient advocate may be too general a question, and the potential for legal repercussions may be not perceived as too much of a threat, a rewording of these questions may help to generate more reliable information on the underlying concepts we attempted to address. The overall alpha reliability coefficient was 0.66. This is below the commonly accepted 0.70 minimum for new scales (Nunnally 1978), but as with the physician survey , it was anticipated due to the small number of questions. However, because we attempted to design the two surveys as closely as possible, so comparisons between the two could be made, a longer and perhaps more subject specific survey was prohibited.
Regression analysis revealed a significant association (p < 0.05) with the factor, "utilitarian approach to treatment decisions," and location. The final model being: Factor 1 = 20.7368 + 3.7632(Location).
This means that location accounted for the predictive power of this factor. Accordingly, the predicted "utilitarian approach to treatment decisions" response for allied health care providers working outside the United States is approximately 4 more than the predicted similar response for allied health care workers employed in U.S. hospitals. The summated scale for this factor ranged from 12.00 to 33.00, with a mean response of 21.85 (SD+5.61). Because this factor consisted solely of questions from the 5 category likert scale format, the lower the response number on the summated scale from 5 through 40, the more likely the allied health care providers surveyed, responded negatively to the questions comprising the factor. Therefore, this indicates that allied health care employees outside the United States on average, are more likely to agree that physicians should be able to assume a greater amount of autonomy due to their experience and objective medical knowledge than their American colleagues. This finding again displays the discrepancy between practitioners in the United States, and elsewhere. Whether this apparent difference is due solely to cultural, or economic disparities, is unknown. However, we have previously demonstrated that the nurse-physician working relationship is acknowledged and more respected by physicians outside the United States than within. We have also discussed the possibility that non-USA physicians may have a better awareness of the allocation of scarce health care resource. Together, these concepts may contribute to the perceived notions we found by allied health care workers.
Summary and Conclusions:
Critical Care is a medical and surgical specialty that seeks to reverse organ system failure by the application of high technology monitoring and treatment systems. Since we deal with patient care on the cutting edge of medical technology, it is extremely important for us to have rapid access to technologically related data. Prior to the advent of rapid and reliable internet access, options for accessing data useful to us was difficult, ponderous and frequently expensive. The ability of like-minded individuals to communicate in real time by electronic mail, set the stage for an exponential increase in working physicians ability to access and use meaningful medical data and expertise related to patient care. The formation of "closed" medical interest group lists, allows a large multi-national cross section of physicians, nurses, researchers, pharmacists, and other interest groups to provide each other with advice, share new findings, and debate treatment care plans. This enables specific patient problems to be quickly and efficiently addressed within the cyber-global physician community.
We accessed a large contingent of multi-national physicians and allied providers to assess their perceptions on a current health care issue, namely futility. The answers we received suggest that physician views and thought about the ethical and clinical problem of futile care are relatively similar. Their actions in dealing with them, however, are markedly different and in many ways, are shaped by their environment. The fact that we were able to access an international population of health care providers in a relatively short period of time is remarkable, and will surely shape the future practice of medicine in ways not currently known. Nonetheless, the small scale nature of this pilot survey, together with the biases inherent in the population sampled, prevent making any generalizations to the larger medical and allied health care populations. However, it was felt that the two chosen internet lists would elicit the target population of health care professionals involved in futility decisions; critical care physicians and allied health care professionals, particularly nurses. Because the complexity of pediatric cases involves even more dilemmas associated with end of life decisions, only adult health care providers were surveyed.
References