Published ahead of print on December 4, 2003, doi:10.1164/rccm.200305-645OC
© 2004 American Thoracic Society Junior versus Senior Physicians for Informing Families of Intensive Care Unit PatientsService de Réanimation Médicale, Hôpital Saint-Louis, Paris, France Correspondence and requests for reprints should be addressed to Élie Azoulay, M.D., Ph.D., Service de Réanimation Médicale, Hôpital Saint-Louis, 1 Avenue Claude Vellefaux, 75010 Paris, France. E-mail: elie.azoulay{at}sls.ap-hop-paris.fr
To compare the effectiveness of information delivered to family members of critically ill patients by junior and senior physicians, we performed a prospective randomized multicenter trial in 11 French intensive care units. Patients (n = 220) were allocated at random to having their family members receive information by only junior or only senior physicians throughout the intensive care unit stay; there were 92 and 93 evaluable cases in the junior and senior groups, respectively, with no significant differences in baseline characteristics. Between Days 3 and 5, one family representative per patient was evaluated for comprehension of the diagnosis, prognosis, and treatment in the patient; satisfaction with information and care; and presence of symptoms of anxiety and depression. No significant differences were found between the two groups for any of these three criteria. Family members informed by a junior physician were more likely to feel they had not been given enough information time (additional time wanted: 3 [0–6.5] vs. 0 [0–5] minutes, p = 0.01) and to have sought additional explanations from their usual doctor (48.9 vs. 34.4%, p = 0.004). Specialty residents, if given opportunities for acquiring experience, can become proficient in communicating with families and share this task with senior physicians.
Key Words: intensive care unit • comprehension • communication • residents • learning Meeting the informational needs of families is a major goal for intensive care physicians. Delivering honest, intelligible and effective information raises specific challenges in the stressful setting of the intensive care unit (ICU), yet remains not only an ethical necessity but also a practical one, if the families are to share in decision making and patient care (1). In recent years, several tools for evaluating the effectiveness of information delivered to ICU families have been developed and validated in prospective trials in Europe and North America (2–6). Studies have assessed three aspects of information: comprehension of the essential aspects of medical care, such as the diagnosis, disease severity, and main treatments (4), family satisfaction as evaluated by the Critical Care Family Needs Inventory (3, 7–9), and the prevalence of symptoms of anxiety and depression in family members (5). Another major goal for intensivists is teaching residents to interact with patients and family members (10–12). The ICU provides residents with unique opportunities for learning to communicate with families in situations where the diagnosis and prognosis are frequently uncertain, where bad news often needs to be imparted, and where treatment limitation decisions may need to be considered. In the time-constrained setting of the ICU, families need to receive information as quickly as possible, at admission of the patient and whenever the course of the disease changes abruptly (4). Thus, availability of a physician for providing information at any time during the day or night is a prerequisite to optimal handling of families' informational needs. However, emergencies are the sum and substance of ICU work, and senior physicians are often unable to leave their patients (2, 13). Families are more likely to receive information rapidly if informing is shared between senior and junior physicians. Furthermore, participation in family information may be a valuable learning opportunity for juniors, particularly if it occurs within a framework of debriefing by and support from seniors (14, 15). Whether juniors can carry out this task satisfactorily has not been evaluated. Poor information may have serious negative consequences, such as loss of the family's trust, a greater risk of conflict, and impairment of the family's ability to make appropriate decisions for the patient. Thus, the decision to entrust juniors with an information-giving role is far from trivial. Informing families requires emotional maturity, self-confidence, psychological sensitivity, and awareness of the factors that influence communication (16–19). Acquiring this amalgam of insight and knowledge probably takes time. A previous study showed that experienced intensivists benefited more from an Acute Physiology and Chronic Health Evaluation II training program than did residents (20). However, in a study of senior oncologists, time and experience failed to resolve communication problems (21). In addition, in an earlier study (9), we found that families were more likely to be satisfied when they received information from the junior physician in charge of their relative as well as from the senior physician. This finding prompted us to undertake a prospective, randomized, controlled, multicenter study of the effectiveness of information provided to family members by junior physicians (residents) and senior physicians (fellows and attending physicians). Our study hypothesis was that juniors would perform as well as seniors.
Setting Eleven ICUs in France participated in the study between June and October 2002. The study design is shown in Figure 1 . In each ICU, a single physician (the investigator) enrolled the participants and assigned them to their groups by opening sealed randomization envelopes. The investigator, who was not an informing physician, collected data on the ICU, patients, and families (see online supplement). In all the study ICUs, at least one junior and one senior physician were present at all times; thus information was always available to the family members in both groups. In each ICU, the physicians ensured that no families assigned to one category of physicians received information from physicians in the other category from the time of randomization. Between Days 3 and 5 of the ICU stay, the investigator interviewed the family representative about comprehension, satisfaction, and symptoms of anxiety and depression, as previously described (see online supplement). Although emotional support and facilitation in decision making are important components of communication with family members, they were not evaluated in the current study. The study was approved by the Ethics Committee of the French Society for Critical Care.
Junior physicians were defined as ICU residents and senior physicians as ICU fellows or attending physicians. Access to residency programs in France is obtained by sitting a national competitive examination after completing 6 years of medical school, with part-time hospital rotations throughout the last 3 years, when theoretical and clinical instruction is provided by residents and senior physicians. Residency programs consist of eight 6-month full-time hospital rotations. All the residents in this study were at their second year of specialty residency, but at their first ICU rotation. Formal training on interacting with patients and families is not part of the medical school curriculum or residency teaching programs in France. The trial was conducted at the midpoint of the 6-month ICU rotation, after the residents had had many opportunities to observe senior physicians informing families. The residents were not told beforehand of the trial, as this might have led them to direct more attention to information than they would have without the study. The senior physicians had finished their residency and had at least 2 years of ICU experience as senior physicians. Both junior and senior physicians spent 10 to 12 hours a day, 6 days a week, at the hospital and were on night call once a week. In all 11 participating centers, a junior and a senior physician were on duty at night (7.00 P.M.–9.00 A.M.).
Statistical Analysis
Sample Size
As planned, 220 patients were included in the study. Full sets of completed questionnaires were available for the representatives of 185 patients, 93 in the senior group and 92 in the junior group. Data from the remaining 35 patients could not be analyzed, for the following reasons: 5 patients died before the third day and, therefore, before their family members could be interviewed; 26 patients did not receive visits from family members after the third day; 3 family representatives declined to complete the Critical Care Family Needs Inventory and Hospital Anxiety and Depression Scale (HAD) forms; and the full set of data sheets was not evaluable for 1 patient. The patient flowchart is shown in Figure 1.
ICU Characteristics
Patient Characteristics The median age of the patients was 57 years (44–72 years), and 60% of the patients were male. Direct admissions, i.e., admissions from the emergency department or from an out-of-hospital site via a mobile emergency team, contributed 42.3% of the patients. The Simplified Acute Physiology Score II at admission was 39 (29–55), median length of ICU stay was 9 days (5–17), and ICU mortality was 33.6%. The characteristics of the study population, including comorbidities and reasons for admission, are detailed in Table 2 . None of the collected parameters differed significantly between the two randomized groups.
Characteristics of the Family Representatives The family representatives had a median age of 50 years (39–64) and most were spouses (43%) or children (27%) of the patients. Nearly 12% were healthcare professionals. Their other characteristics are reported in Table 3 .
Effectiveness of the Information Provided to Family Members As shown in Table 4 , no significant difference was found between the two groups for the main evaluation criterion: comprehension was poor in 35.5 and 30.1% of family representatives informed by juniors and seniors, respectively (p = 0.47).
There was no significant difference between the two groups for satisfaction as evaluated by the median Critical Care Family Needs Inventory score. The satisfaction data were encouraging: overall, nearly 93% of the family representatives were satisfied with the information received and 95% felt that their questions had always been answered. Two-thirds of the families knew the job title of the ICU caregiver who delivered information to them and 83% felt they had received no contradictory information, with no significant differences between the groups for either variable. Overall, only 4.8% of the family members said they had received too much information, whereas 61% said they had received too little information about the diagnosis, 63% about the prognosis, and 57% about the treatment; again, there were no significant between-group differences. Only two satisfaction parameters differed significantly between the two groups: the total time spent receiving information, as evaluated subjectively by the family representatives, was equal to the time wanted in the group informed by seniors but was shorter in the group informed by juniors, although no difference was found for clocked time, and family representatives informed by juniors were more likely to have sought explanations from their usual doctor. Symptoms of anxiety and depression were extremely common (Table 5) : overall, 74% of the family representatives had anxiety and/or depression. There was no significant difference between the juniors and seniors for symptoms of anxiety, depression, or anxiety and/or depression (Table 5).
No differences were observed between family representatives of patients with or without decisions to forgo life-sustaining treatment or between family representatives of patients who survived or died (data not shown). No conflicts between family members and ICU caregivers occurred during the study period.
This is the first multicenter prospective randomized study comparing the ability of senior and junior physicians to deliver effective information to families of ICU patients. Using previously validated tools for assessing the effectiveness of information, we found that junior physicians performed as well as seniors in this area. Conflicting results have been obtained in earlier studies of the ability of residents to communicate with families or patients, particularly when bad news must be imparted or end-of-life issues discussed. Most studies of residents stress the need for improving interviewing techniques (23), communication skills (24, 25), and knowledge of ethics (11), implying that residents are not sufficiently skilled to be entrusted with interviewing and informing patients and families. Moreover, in a study by Orlander and coworkers, 73% of residents said that their first experience with delivering bad news occurred when they were medical students or interns and that only 5% had been assisted in this task by a senior physician (26). Thus, it is reasonable to hypothesize that, even without formal training, junior physicians strive to develop communication skills early on to meet the challenges brought to them by their clinical work. In a previous multicenter trial conducted in France to evaluate how well ICU caregivers satisfied the needs of ICU families, we found that families were very satisfied with their interactions with residents and that information provided by a junior physician was among the six independent predictors of family satisfaction in the multivariate analysis (9). This suggested to us that we might be underestimating the ability of residents to deliver information and prompted us to conduct the present study. When we compared the components of family satisfaction in the two groups, we found that only two variables were significantly different: family members who were informed by junior physicians felt that they had not spent enough time receiving information, and they were more likely to have sought explanations from their usual physician. Interestingly, the time spent receiving information as clocked by the physicians was the same in the two groups. The perception by the families in the junior group that information time was inadequate is consistent with a minor degree of dissatisfaction, possibly related to differences in style of communication. Further work is needed to look for such differences. For instance, it has been reported that meetings should never be ended without giving the families the opportunity to ask questions they might hesitate to formulate (16). In some cases, it may be desirable to provide answers to questions that have not been asked. Thus, one possibility is that the juniors in our study ended the interviews too abruptly or failed to provide information that the families wanted but did not specifically request, for instance about the prognosis, for which a trend toward poorer comprehension was noted in the group of family members informed by junior physicians (16). The style of communication may be at least as important to families as the content of the information imparted. In a study of families informed of the death of patients with trauma, 72 and 70% of families felt that the attitude of the physician and the clarity of the message were the most important factors, respectively, as compared with only 57% for the ability to answer questions (27). In our earlier study (9), families who received explanations from their own doctor had higher satisfaction scores (9). A family's usual doctor can provide clarification about the patient's condition in a way that is appropriate for that family's unique structure and dynamics, thus contributing to improve comprehension and to alleviate distress. In the present study, family members informed by junior physicians were more likely to have sought additional explanations from their usual doctor. The similar level of satisfaction in our two study groups may be ascribable to this difference, as the family doctors may have compensated for any shortcomings related to the junior physicians' inexperience (28). Similarly, interpretation of the levels of satisfaction and of the prevalence of symptoms of anxiety and depression in this study should take into account the fact that families in both groups received information from a single physician, so that contradictions were eliminated (5, 9). Our results can probably be extrapolated to other hospital departments. In the ICU, families are likely to experience more suffering than anywhere else in the hospital, yet the time available to meet their needs is often shorter. If junior physicians can perform well in this challenging environment, as shown by our study, they can probably perform at least equally well in other departments. On the other hand, ICU rotations give residents innumerable opportunities for observing senior physicians as they interact with families, break bad news, and lead end-of-life discussions. Thus, the juniors in our study, who had already spent 3 months in the ICU, were perhaps more skilled in this area than residents with no ICU experience (9). In this study, both junior and senior physicians were on-site around the clock, and senior physicians were probably more available than in other settings. This may have offered the junior physicians greater opportunities to observe and model information delivery techniques. In addition, families who see that senior physicians are present at all times during the day and night may have a greater sense of satisfaction with information delivered by junior physicians than families who see no evidence of senior physician involvement. However, this study could not have been performed in another setting. Indeed, the only way to compare the effectiveness of information by junior and senior physicians was to perform the study in ICUs where physicians in both categories were available at all times to avoid depriving family members from important information. This point is particularly important for those patients and families arriving in the ICU (4). ICUs should be viewed as a rich source of experience in the field of communication and ethics. Further work is needed to determine how this source can be used optimally to hone the communication skills of residents, especially with family members of patients requiring end-of-life decisions. In addition, potential stress reactions induced by having to inform family members should be carefully monitored in the residents. In an earlier study (4) of comprehension evaluated using the same instrument as in the present study, we found poor comprehension in 54% of family members. Poor comprehension was noted in 33% of family representatives overall in the present study. This improvement may be ascribable in part to the use by 10 of the 11 study centers of a family information leaflet. In a previous randomized study (22), we found that a family information leaflet significantly reduced the proportion of family members with poor comprehension. The results of the current study also highlight the need for improving comprehension of the prognosis, most notably when junior physicians provide information. Also, the observed discrepancies between the high level of family satisfaction and the high rate of poor comprehension as measured and reported by family members should be confirmed using other tools (6, 14, 15, 17). Comprehension and satisfaction may assess different domains of communication. Another possibility is that physicians put more effort into empathy, listening skills, and emotional support than into the content of the information provided. Further studies should identify means of helping families achieve greater levels of understanding. Our study has several limitations. First, we did not control for information provided by other caregivers (e.g., nurses), which might introduce bias. However, a reasonable assumption is that information by other caregivers had similar effects in the two groups because these were defined by randomization. Second, we compared the effectiveness of information in the two groups determined by randomization, but we did not assess how emotional support for decision making was provided. Third, although formal training on interacting with patients and families is not part of the medical school curriculum or residency teaching programs in France, the ICUs were part of the French FAMIREA Group and may therefore have provided residents with more training about interacting with families and patients than is usually provided in ICUs. In addition, residents with an interest in families may tend to choose ICUs that are FAMIREA members. Should ICUs without a special focus on families decide to extend information duties to junior physicians, evaluation of the impact of this decision on the effectiveness of information would be of considerable interest. Furthermore, in our study, we did not debrief the junior physicians after family meetings and consequently cannot assess the possible burden imposed on the junior physicians by our intervention. Fourth, we did not record the type of information requested from family physicians; this would have suggested specific means of improving communication skills in junior physicians because families in the junior physician group were more likely to ask questions of their family physician. In summary, if given enough opportunity and experience, junior physicians can hone their communication skills and share with senior physicians the time-consuming yet rewarding task of interacting with the families of critically ill patients. This strategy would be expected not only to ensure that all families receive adequate information even when no senior physicians are available but also to provide residents with valuable learning opportunities. Senior physicians can support residents in their quest for better communication skills by sharing knowledge, providing feedback to the residents about their performance, and remaining alert to the emotional responses of residents. Further studies should assess how well junior physicians perform in providing emotional support and decision-making assistance, two important components of communication skills in the ICU.
Supported by a grant (AOR1004) from the Assistance Publique—Hôpitaux de Paris and the Direction Régionale de la Recherche Clinique, two nonprofit government-funded organizations in France. This article was written on behalf of the FAMIREA Study Group: Intensive Care Unit of the Saint-Louis Teaching Hospital and University of Paris 7, Assistance Publique—Hôpitaux de Paris, Paris, France. The study was done in the 11 intensive care units listed in the online supplement. This article has an online supplement, which is accessible from this issue's table of contents online at www.atsjournals.org Conflict of Interest Statement: D.M. has no declared conflict of interest; D.G-T. has no declared conflict of interest; C.A. has no declared conflict of interest; M.J. has no declared conflict of interest; C.A. has no declared conflict of interest; D.A. has no declared conflict of interest; M.G-O. has no declared conflict of interest; J-Y.L. has no declared conflict of interest; L.P. has no declared conflict of interest; P.Q. has no declared conflict of interest; F.P. has no declared conflict of interest; E.A. has no declared conflict of interest. Received in original form May 13, 2003; accepted in final form November 25, 2003
This article has been cited by other articles:
|
|||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||
TOP
ABSTRACT
METHODS
