Comparison of Mothers' and Counselors' Perceptions of Predelivery Counseling for Extremely Premature Infants
http://www.100md.com
《小儿科》
Departments of Social Medicine Pediatrics, University of North Carolina, Chapel Hill, North Carolina
Compassionate Passages, Inc, Chapel Hill, North Carolina
ABSTRACT
Objective. To understand mothers' and counselors' perceptions of their roles in decision-making about resuscitation of extremely premature infants at delivery and to assess mothers' and counselors' satisfaction with the counseling and decision-making process.
Methods. Mothers who delivered an infant between 22 and 27 completed weeks of gestation and their self-identified counselor were interviewed using a structured interview format. Mothers' and counselors' perceptions of the content, tone, and directiveness of predelivery counseling and their satisfaction with the decision-making process were compared. Demographic data were collected for the mothers, infants, and counselors. Simple descriptive statistics described demographic characteristics of mothers, counselors, and infants. Pearson’s correlation coefficient was used to determine agreement within individual mother-counselor pairs about the content and directiveness of counseling.
Results. Thirty-three counselors and 15 mother-counselor pairs were interviewed. The majority (66.7%) of mothers stated that the counselor had made a treatment recommendation, and 60% stated that they had no choice in how their infant would be treated. Only 27.3% of counselors stated that they had made a recommendation, saying instead that they had described the treatment plan or offered options. Counselors believed that mothers were given a treatment choice in 57.6% of encounters. Specific mother-counselor pairs showed little correlation in their perceptions of whether a treatment recommendation had been made (R = 0.0) or a choice had been given about resuscitation (R = 0.07). Despite a lack of perceived choice, mothers generally believed that they were included in treatment decisions (66.7%) and were satisfied with the amount of influence that they had in the decision-making process (73.3%).
Conclusions. The decision-making process in this study conforms most closely to a model of informed assent. Mothers may have been satisfied with this type of counseling because they felt informed and included in the decision-making process. Physicians and nurses need to elicit mothers' preferences to incorporate them into the treatment plan, as counseling about the resuscitation of extremely premature infants at delivery is considered directive by mothers even when it is not intended to be directive.
Key Words: counseling decision-making ethics extremely preterm infants resuscitation
Counseling parents about whether to resuscitate their extremely premature infant at delivery is fraught with difficulty. The counseling is often done by medical professionals who are relative strangers to the parents, at a time when the infant's birth is imminent and parents are in crisis. In addition, the statistics are not encouraging: infants who are born at <26 weeks of gestation have a reported mortality rate of >50%, with half of surviving children normal at 30 months of age, 25% with mild to moderate disabilities, and 25% with severe disabilities.1 Few factors distinguish these groups in advance, so predicting the outcome of any specific premature infant is profoundly uncertain. Even these general statistics are a moving target, as advances in the care of very premature infants change potential outcomes. How much responsibility parents wish to take in making these extremely grave decisions and how much authority health care providers wish to relinquish are currently unclear.
In the United States, the past several decades have witnessed a shift in the norms of medical ethics from physician paternalism to patient autonomy.2 Accounts of neonates who were treated aggressively against their parents' wishes appear not infrequently in the lay press and on the Internet, reinforcing the notion that parents resent physician paternalism and wish to have complete autonomy in making resuscitation decisions.3
Empiric research describing delivery room decision-making for premature infants is sparse but demonstrates physicians' desire to maintain some decision-making authority for very premature infants, with most accepting a parental role in decision-making for infants who are born at the 22- to 24-week range of gestation but not for infants who are born at 26 weeks.4 In a Canadian study that contrasted physicians' and parents' views about who should have the greater role in making delivery room decisions for extremely low birth weight infants, Streiner et al5 found that 64% of the parents who had a premature infant survive into their teens believed that all extremely low birth weight infants should be resuscitated regardless of their prognosis, compared with only 6% of the neonatologists and neonatal intensive care nurses. Parents' responses did not differ according to whether their child had a handicapping condition or not. Because of the study design, parents of children who were not resuscitated at delivery or who died after resuscitation were excluded, so the sample was biased.
Most recently, in a qualitative study of parents of premature infants in Norway, Brinchmann et al6 found that parents wished to be informed, listened to, and consulted, but generally did not wish to bear the burden of making the final decision about withdrawal of support in the NICU. A similar study of NICU parents in the United States has not been done.
Our aims in this study were to gain a better understanding of mothers' and counselors' perception of their roles in the decision-making process about the resuscitation of very premature infants at delivery and to gain insight into mothers' and counselors' satisfaction with the decision-making process. We believe that this is necessary information for an informed debate about how controversial health care decisions for very premature infants should be made.
METHODS
Design
This descriptive study was conducted over a 2-year period at a Southeastern public teaching hospital with a level III NICU. The hospital treats pregnant women from the immediate area and also serves as a regional referral center for perinatal and neonatal care. The study included women who presented to the hospitals' obstetric service and delivered an infant between 22 and 27 completed weeks of gestation and the person whom each woman identified as having primarily counseled her about how her infant would be treated at birth. Obstetric and neonatal patients are cared for by attending physicians, fellows, residents, nurse practitioners, and nurses.
All mothers who delivered a singleton infant between 22 and 27 weeks of gestation and received at least 1 session of predelivery counseling were eligible to participate. Exclusions were induced abortions, stillbirths in which the fetus's death was known before labor, fetuses with known lethal anomalies, non–English-speaking mothers, mothers <18 years of age (as adolescents may represent a distinct subpopulation and had not been included in development of this questionnaire), and mothers with multiple gestations (because there may have been differential counseling for each fetus). During the study period, it was policy that infants who were born at <23 weeks of gestation were considered nonviable and were not resuscitated; infants who were born at >25 weeks of gestation were considered viable and were resuscitated; and infants who were born at 23, 24, or 25 weeks of gestation were considered potentially viable and resuscitated or not on the basis of a decision made after counseling the parents about the outcomes of extremely premature infants. The study population was chosen to encompass a spectrum of counseling styles (directive to nondirective) and outcomes.
Before their discharge from the hospital, eligible mothers were asked to identify who primarily counseled them about the treatment that their infant would receive at delivery. The counselor interviewed was the person whom the mother identified regardless of who else counseled her. Pictures of attending physicians, fellows, residents, and nurse practitioners were available to help mothers with identification. When a mother remembered being counseled but could not definitively identify her counselor, the medical team was asked to identify who that person would have been. Counselors then were contacted to ensure that the match was correct. Mothers were also notified before hospital discharge that someone would call them in 6 weeks to invite them to participate in a research study about preterm delivery.
Six weeks after delivery, mothers were contacted by a nurse, who asked whether they were willing to be called by a researcher to discuss participation in a study about preterm delivery. When a mother accepted, she then was contacted by an interviewer and informed consent was obtained by telephone. Informed consent was also obtained from all counselors. This study was approved by the medical school's Institutional Review Board.
Data Collection
Mothers were interviewed 6 weeks after delivery by telephone using a standardized interview form developed for this study. At least 5 attempts were made to contact each mother at different times of day. Charts of eligible mothers and their infants were reviewed for maternal age, marital status, parity, previous preterm delivery, education, gestational age and birth weight, treatments received at delivery, and outcome (including survival status and major diagnoses known by 6 weeks after delivery). All counselors were interviewed by telephone within 72 hours of delivery using a standardized interview format.
Questionnaire Design
The questionnaire for mothers was designed using expert opinion, including a neonatologist, an obstetrician, an epidemiologist, a mother of an infant who had been hospitalized in the NICU, and the NICU bereavement counselor; data from the literature about which factors in counseling are important to patients7,8; and information from a small focus group of mothers who had delivered premature infants. The questionnaire then was pretested on 4 mothers of preterm infants. Opinion was also solicited from these mothers about the appropriate timing for an interview after a preterm birth. Six weeks after delivery was chosen to minimize intrusiveness from the interview but to maximize recall of predelivery counseling. The final questionnaire addressed the following items: the mother's satisfaction with the counseling that she received about her infant's treatment at delivery (including the type of information given, how the information was given, and whether it was too much or too little information), how directive the counseling was, which factors she considered when thinking about resuscitation, who she believed made the final decision about delivery room treatment, whether she had a choice about resuscitation, whether she was satisfied with the amount of influence that she had in the decision, and whether she was satisfied with the care that her infant received at delivery.
Counselors were asked similar questions about the content, tone, and amount of information discussed; how directive they were in their counseling; whether they gave the mother a choice about resuscitation; who made the final decision; and whether they were satisfied with the outcome. Answers to open-ended questions were recorded on the questionnaire.
Analytic Approach
Simple descriptive statistics (frequencies and percentages) were used to describe the demographic characteristics of mothers, counselors, and infants and the salient features of counseling and outcome. Mothers who chose to be interviewed were compared with those who did not participate to ascertain whether there was any systematic bias in who was included in the study cohort. Nonparametric comparisons were made using the Mann-Whitney U test. Exact methods were used when the expected cell size was <5. Pearson’s correlation coefficient was used to determine whether individual mother-counselor pairs agreed about the counseling and decision-making process, including the type of information discussed, how directive the counseling was, and whether the mother had a choice about resuscitation. Interviews were reviewed qualitatively to delineate the main themes of mothers' and counselors' responses to open-ended questions.
RESULTS
Forty-four mothers were eligible to participate in the study. Nine of the 44 were discharged before they or their counselor could be enrolled. One mother with eclampsia could not remember being counseled. One mother was unable to identify her counselor. This left 33 (75.0%) available mothers to enroll. Thirty-three counselors were identified by these mothers, and all counselors were interviewed. Fifteen mothers agreed to be interviewed at 6 weeks, yielding 15 mother-counselor pairs for analysis. The most frequent reason for nonenrollment was the inability to contact mothers by telephone (disconnected numbers). Demographics by maternal interview status are presented in Table 1. Participating mothers were similar to the nonparticipating group, with no statistically significant differences at a P .05 level.
Counselors
Characteristics of the counselors and their impressions of the counseling sessions are shown in Table 2. Counselors were predominately female (75.8%), had usually counseled the mother only once (63.6%), and were usually obstetric residents (45.5%) or neonatal fellows (30.3%). Two mothers identified the labor and delivery nurse as her primary counselor. Counselors reported that they discussed the infant's chance of survival (90.9%) more often than the potential for handicap (69.7%). Fewer than half said that they discussed suffering. Most of the counselors (81.8%) believed that they knew what the mother wanted for her infant because the mother told them. Counselors said that 64% of the mothers wanted "everything" done, and 18% wanted "treatment if the infant looks viable."
A majority (57.6%) of the counselors believed that they gave the mother a choice about delivery room resuscitation, and most (69.7%) stated that they had not made a specific recommendation. Nine (27.3%) of the counselors believed that the physician made the final decision about resuscitation, 9 (27.3%) believed that the mother made the final decision, and 13 (39.4%) believed that the decision was made jointly.
Twenty-nine of the counselors knew which kind of treatment the infant actually received at delivery. Four counselors had gone off duty and did not know. Twenty-five (86%) of the infants whose delivery room treatment was known by counselors were resuscitated, and 4 (12.1%) received comfort measures. Twenty-eight (84.8%) of the counselors were somewhat or very satisfied with the outcome of the delivery room decision, and 1 counselor was neutral; however, 3 (9.1%) counselors believed that the infant's treatment had been excessive. None of the counselors believed that an infant had received too little care.
Mothers
Mothers' impressions of the counseling sessions are presented in Table 3. More than 90% of the mothers who were interviewed were very or somewhat satisfied with the counseling that they received before delivery and satisfied with the care that their infant received at birth. Two thirds of the mothers said that the counselor had made a treatment recommendation, and 60% said that they did not have a choice about how their infant would be treated. Nonetheless, most mothers agreed with the counselor's recommendation and believed that they had had a voice in the decision. Although a majority of the mothers said that they had no choice about the treatment that their infant would receive at delivery, 73% were very satisfied with the amount of influence that they had in making the resuscitation decision. Thirteen (86.7%) of the mothers said that everything that could be done for their infant was done. No mother said that everything was not done, although 2 said that they did not know.
Mother-Counselor Pairs
A comparison of the impressions of the 15 mother-counselor pairs about their counseling session(s) is presented in Table 4. Counselors were more likely than mothers to report that they had discussed the potential for handicap and less likely than mothers to report that they had discussed future suffering. Of note, twice as many mothers reported receiving a recommendation about delivery room resuscitation than counselors reported making a recommendation (66.7% vs 33.3%, respectively). When the responses of individual mother-counselor pairs were compared, there was almost no agreement between mothers and their counselors about either the content or the directiveness of the counseling session(s).
Responses to Open-Ended Questions
Mothers were asked open-ended questions about what they considered when deciding how their infant would be treated to delineate further their responses. Most mothers responded with personal values, beliefs, or experiences, rather than mentioning the medical information that was presented during their counseling. Many mothers said that they simply wanted everything done, for example, "Never a question not to do everything," "Wanted everything done for him." Others relied on their faith, as 1 woman said, "I put my faith in God and hoped He would help the infant and the doctors." Only 2 mothers specifically mentioned the infant's prognosis, whether their infant would suffer, or the infant's future quality of life. Some of the mothers relied on their previous experience with premature infants. Some wanted the doctors to make the decision.
When mothers were questioned about why they were satisfied or dissatisfied with their counseling, the predominant theme that emerged was a desire for information. They appreciated explanations and knowing what would happen in the delivery room. In general, mothers who believed that they had no choice in the resuscitation decision stated that they were satisfied with the amount of influence that they had in the decision-making process for 3 main reasons: (1) they were given information: "Explained step-by-step what they would do"; (2) they trusted the physicians' judgment: "The doctors knew what they were doing"; or (3) they felt included in the process: "They told us if they got to a place where decisions had to be made (about continuing support) they would tell us," "Asked my opinion before they did anything," "We all talked, and our feelings and opinions were considered."
When mothers were asked what could be done to help patients like them in the future, all who responded expressed that they wanted more information with less medical jargon: "When doctors would explain, the words kept getting bigger and bigger; it would be helpful to have someone to break it down into more simple explanations." Some mothers suggested pamphlets or booklets.
Counselors were asked what recommendations they gave about the infant's treatment at delivery, what treatment they thought the mother wanted, and why they were or were not satisfied with the resuscitation decision. Most counselors stated that they had not given a recommendation. Counselors were more likely to state that they gave a recommendation for infants of a younger gestational age, including, "Infant was previable and should not be resuscitated," and, "Recommended that if infant <24 weeks and <500 g not to resuscitate." One counselor stated that she gave an implied recommendation. "I explained what was usually done for 25-weekers. It was an implied recommendation that this would be done for her infant unless she wanted something different." Some counselors stated that they did not give a recommendation because the mother had no choice: "I felt that based on the infant's age that the prognosis was good and justified full measures. Did not offer the option of not doing anything." When counselors who gave no recommendation were asked how the mother wanted her infant treated at delivery, most stated that the mother "wanted everything done." Even for very premature infants (<24 weeks' gestational age), counselors reported, "Mom wanted everything done including cesarean section, intubation, and everything the NICU could do," "Mom wanted everything done." Sometimes, however, counselors reported that the mother had stated a goal and wanted the physician to determine the level of care on the basis of that goal: "Wanted the infant evaluated for its size and vigorousness and to resuscitate if the infant had a good chance to survive," "Wanted infant to survive if it had a reasonable chance to survive long term," "Was realistic," "Did not want the infant to survive with a great chance of suffering."
Most counselors were very satisfied or somewhat satisfied with the decisions that were made about resuscitation. Those who were very satisfied commented on "doing what was best for the infant," "gave the infant every chance," or doing "what the mom wanted." Those who were somewhat satisfied were concerned about being overly aggressive, although treatment was in alignment with the mothers' wishes: "(I was) sorry the infant had to be put through that, although it's what the mom wanted," "(It's) hard to separate my personal beliefs about the long-term prognosis."
DISCUSSION
The most striking finding of this study is the lack of concordance between mothers and counselors about what occurred during their counseling session(s). There was no concordance on clinical information such as survival and potential for the infant to have a handicap or about who made decisions and whether the mother had a choice in how her infant would be treated. The lack of concordance between counselor and mother on issues of clinical information in this study is similar to that found in a study by Zupancic et al.9 Zupancic et al had the parent(s) and the counseling neonatologist fill out a questionnaire about details of potential pregnancy-related and neonatal problems within 24 hours of the counseling session and before the infant's delivery. Concordance between the neonatologist and the parents on clinical detail was poor. They found that increased maternal anxiety decreased the concordance of maternal-physician responses. Other studies have shown that only a small portion of information is retained after parents are given traumatic news about their child.10
Despite this lack of concordance between counselors and mothers, satisfaction with counseling was high in our study. As is frequently the case with premature deliveries, in this study, there was often little time between many mothers' sole counseling session and their infant's birth. It is possible in this situation that the actual medical information was not as important to mothers as their core values and beliefs. Many mothers reported that they based their decision on their previous experiences with premature infants, their faith, and their confidence and trust in the medical team, although some said that they did consider information from the counseling sessions about the likelihood of their child's suffering. Many mothers reported that not "trying everything" would not have been an option. Therefore, maternal satisfaction with counseling could have been more dependent on the tone of the counseling rather than the actual content. Studies of adult patients' satisfaction with the medical encounter have found that the physician's warmth, ability to listen, and provision of information are more directly associated with patient satisfaction than the physician's perceived technical competence.11–13 In our study, 73% of the mothers rated their counselors as very caring (8–10 on a 10-point scale), and 60% said that they were provided with the right amount of information, which may have contributed to the overall high level of satisfaction in this study.
There was also little correlation between the mother and her counselor about how directive the counseling was. Overall, mothers in this study were much more likely than counselors to say that a recommendation about delivery room resuscitation had been made (67% vs 33%), and the concordance within individual pairs was poor (R = 0.0). Mothers were also more likely than counselors to say that they were given no choice about their infant's treatment at birth (60% vs 47%; R = 0.07), also with poor mother-counselor pair concordance. Despite this lack of concordance, most mothers were satisfied with the decision-making process. There are several potential explanations for the mothers' satisfaction.
We found that mothers of extremely premature infants were satisfied with the process of decision-making when they were informed and given explanations, even when they did not believe that they had a choice about how aggressive treatment would be at delivery. Mothers in this study believed that they participated in the decision about resuscitation and were satisfied with the amount of influence that they had on the decision 73.3% of the time, although they believed that they did not have a treatment choice 60% of the time. These results are in agreement with the study by Brinchmann et al,6 which found that Norwegian parents wanted their values and opinions recognized and included but did not wish to bear ultimate responsibility for end-of-life decisions in the NICU. These results are also in agreement with studies of adult patients in the United States, which found that patients' desire for autonomous decision-making decreased as medical decisions became more serious, although their desire to be informed did not decrease.14
Comments made by the mothers suggest that most of them viewed their counseling sessions as inherently interactive and participatory. In this context, mothers may have interpreted treatment plans put forward by the physicians or nurses as "recommendations" that they were engaging together. Because the mothers in this study viewed the treatment plans that were presented to them as recommendations and perceived themselves to be joint decision makers, they were satisfied even when the counseling was very directive.
Another possible explanation for the mothers' satisfaction in this study is that they shared the same view about treatment at delivery as their counselors. Indeed, 82% of the mothers said that they agreed with the recommendation that they received, and many said in their comments that it would not have been an option not to "try everything." We cannot say whether the lack of perceived choice would have led to greater dissatisfaction had there been more disagreement between mothers and their counselors, as might have been the case had the study included more infants who were born before 23 weeks of gestation, for whom resuscitation would not have been offered. However, we also cannot say whether different recommendations would have led to more disagreement. Doron et al15 found that infants who were born at 23 to 26 weeks of gestation were resuscitated at delivery when either the parent or the neonatologist wanted the infant to be treated. Infants were not resuscitated only when both the parent(s) and the neonatologist agreed that comfort care was preferable. Thus, whoever preferred the most aggressive treatment at delivery became the decision maker for the infant. This suggests that if mothers wanted their infant treated at delivery even after counseling about potential poor outcomes, then the treatment was likely to be delivered, and mothers were correct in their perception about the participatory nature of the decision-making process.
Most counselors reported that they told mothers what would happen at delivery and then used their lack of dissent as an indication of consent. This directive counseling practice conforms more closely to the model of obtaining informed assent, as opposed to informed consent, for medical treatment. The informed consent doctrine requires counselors to provide information about the pros and cons of alternative treatments, including the choice of no treatment, so that patients can recognize and weigh options themselves. By contrast, the process of assent limits patient choice by obscuring options. This form of counseling is somewhat paternalistic, but unlike pure paternalism, in which the physician may determine and carry out treatment without patient input, the model of informed assent requires that the physician inform and educate the patient about the treatment and then elicit his or her assent before proceeding. Our finding that most of the mothers in this study were satisfied with the decision-making process even when they perceived that they were given no explicit choices indicates that the process of informed assent was acceptable to them when they were being counseled about delivery room resuscitation.
It is possible that these mothers' satisfaction with their counseling and the process of assent may reflect acceptance of a power hierarchy in the counselor-mother relationship, such that mothers could not conceive of questioning or speaking against the proposed plan of care or taking a more assertive role in decision-making. This seems unlikely, however, as most (80%) mothers said that they were never afraid to ask questions, many linked their satisfaction to the perception that their opinions and views were included, and some did elect full intervention against the counselor's recommendation. Comments by the counselors support this perception. Many counselors said that following the mother's wishes was an important factor in their own satisfaction with the decision-making process.
An unexpected finding in this study was the high proportion of junior obstetrics residents identified by mothers as their primary counselor about the resuscitation decision. Most previous work on delivery room decision-making for extremely premature infants has studied the viewpoints of attending neonatologists or obstetricians.4,5 However, in the teaching hospital setting, this may not be the person whom the mother identifies as her counselor even though she has spoken with an attending physician. As 1 of the mothers' primary desires was for clear and accurate information, it is important that these residents who do counseling have knowledge about the outcomes of prematurity and familiarity with the NICU practices that support premature infants after birth. This finding has implications for obstetrics residency training and education.
This study has both weaknesses and strengths. The primary weakness is its small study size. Also, no mother who was interviewed had an infant who was born at <23 weeks of gestation, for whom resuscitation at delivery would have been refused. It is possible that mothers who were not satisfied chose not to participate. Because of the small sample size, there were inadequate numbers to reach theme saturation in analysis of open-ended responses. Therefore, typical responses were included to illustrate themes of mothers' perceptions. It is also possible that by 6 weeks after delivery, mothers had difficulty separating out the counseling that they received before delivery from subsequent counseling and other information that they learned about premature infants later. This period was chosen after consulting with mothers who had delivered premature infants about the appropriate time frame for an interview. However, the difference in time to interview between counselors and mothers does represent a differential recall bias. This bias would be difficult to overcome as this was a singular, important event for mothers but not for counselors; therefore, mothers are more likely to remember the counseling event. Furthermore, this study was done at a single, tertiary care hospital and may not be generalizable to other settings. Thus, the results from this study should be interpreted with caution.
This study also has strengths. Previous, related research has relied on interviews of the medical team only, questionnaires, or surveys of selected parents of premature infants many years after delivery. This study is unique in that we directly interviewed all consenting mothers about their recent experience and linked their responses to the responses of the person whom they identified as having counseled them about delivery room resuscitation. We did not assume that the mother's primary counselor was the neonatal or obstetric attending physician. Therefore, this study provides insight into whom the mothers perceived as their primary counselor about delivery room resuscitation, as well as into the mothers' own impressions of that counseling. Finally, because we interviewed medical professionals about real patients whom they had recently counseled, rather than about hypothetical patients and situations, this study may provide a more accurate depiction of counseling as it is actually perceived and practiced.
CONCLUSIONS
This study suggests that mothers of extremely premature infants perceived the counseling that they received about resuscitation before their child's birth as directive. In general, mothers were satisfied with this type of counseling and considered themselves to be joint decision makers even when they were given no explicit choice about their infant's treatment at birth. This kind of counseling is closer to a model of informed assent than informed consent. A related study found that one quarter of parents said that their preference was to have the physician make treatment decisions for their premature newborn; however, physicians were poor at judging which patients preferred this type of decision-making.9 It is possible that a directive form of counseling that gives information and recommendations but also elicits patient preferences allows mothers to choose their level of participation in the decision-making process. This may relieve some families of the burden of a choice that they do not wish to make alone, while allowing other families greater autonomy.
Mothers of extremely premature infants in this study wished to be well informed and wanted their values and opinions recognized and included in the decision-making process. Counselors should be careful to elicit mothers' preferences for treatment so that these can be incorporated into the medical plan, as mothers perceived counseling to be directive even when the counselor had not intended it to be so.
ACKNOWLEDGMENTS
This study was supported by University Research Council grant 5-43612 from the University of North Carolina (Chapel Hill).
We acknowledge the assistance of the late Michael J. McMahon, MD, MPH, who contributed to the design of this study.
FOOTNOTES
Accepted Oct 21, 2004.
No conflict of interest declared.
REFERENCES
Wood NS, Marlow N, Costeloe K, Gibson AT, Wilkinson AR. Neurologic and developmental disability after extremely preterm birth. N Engl J Med. 2000;343 :378 –384
Quill TE, Brody H. Physician recommendations and patient autonomy: finding a balance between physician power and patient choice. Ann Intern Med. 1996;125 :763 –769
Partridge JC, Freeman H, Weiss E, Martinez AM. Delivery room resuscitation decisions for extremely low birthweight infants in California. J Perinatol. 2001;21 :27 –33
Streiner DL, Saigal S, Burrows E, Stoskopf B, Rosenbaum P. Attitudes of parents and health care professionals toward active treatment of extremely premature infants. Pediatrics. 2001;108 :152 –157
Doron MW, Veness-Meehan KA, Margolis LH, Holoman EM, Stiles AD. Delivery room resuscitation decisions for extremely premature infants. Pediatrics. 1998;102 :574 –582
Brinchmann BS, Forde R, Nortvedt P. What matters to parents A qualitative study of parents' experiences with life-and-death decisions concerning their premature infants. Nurs Ethics. 2002;9 :388 –404
Kirschbaum MS. Life support decisions for children: what do parents value Adv Nurs Sci. 1996;19 :51 –71
Harrison H. The principles of family-centered neonatal care. Pediatrics. 1993;5 :643 –650
Ende L, Kazis L, Ashe A, Moskowitz MA. Measuring patients' desire for autonomy: decision making and information-seeking preferences among medical patients. J Gen Intern Med. 1989;4 :23 –30
Brody DS, Miller SM, Lerman CE, Smith DG, Lazaro CG, Blum MJ. The relationship between patients' satisfaction with their physicians and perceptions about interventions they desired and received. Med Care. 1989;27 :1027 –1035
Lerman CE, Brody DS, Caputo GC, Smith DG, Lazaro CG, Wolfson HG. Patients' perceived involvement in care scale: relationships to attitudes about illness and medical care. J Gen Intern Med. 1990;5 :29 –33
Buller JK, Buller DB. Physicians' communication style and patient satisfaction. J Health Soc Behav. 1987;28 :375 –388
Jedlicka-Kohler I, Gotz M, Eichler I. Parents' recollection of the initial communication of the diagnosis of cystic fibrosis. Pediatrics. 1996;97 :204 –209
Zupancic JAF, Kirpalani J, Barerett J, et al. Characterising doctor-patient communication in counseling for impending preterm delivery. Arch Dis Child Fetal Neonatal Ed. 2002;87 :F113 –F117(Heather T. Keenan, MDCM, )
Compassionate Passages, Inc, Chapel Hill, North Carolina
ABSTRACT
Objective. To understand mothers' and counselors' perceptions of their roles in decision-making about resuscitation of extremely premature infants at delivery and to assess mothers' and counselors' satisfaction with the counseling and decision-making process.
Methods. Mothers who delivered an infant between 22 and 27 completed weeks of gestation and their self-identified counselor were interviewed using a structured interview format. Mothers' and counselors' perceptions of the content, tone, and directiveness of predelivery counseling and their satisfaction with the decision-making process were compared. Demographic data were collected for the mothers, infants, and counselors. Simple descriptive statistics described demographic characteristics of mothers, counselors, and infants. Pearson’s correlation coefficient was used to determine agreement within individual mother-counselor pairs about the content and directiveness of counseling.
Results. Thirty-three counselors and 15 mother-counselor pairs were interviewed. The majority (66.7%) of mothers stated that the counselor had made a treatment recommendation, and 60% stated that they had no choice in how their infant would be treated. Only 27.3% of counselors stated that they had made a recommendation, saying instead that they had described the treatment plan or offered options. Counselors believed that mothers were given a treatment choice in 57.6% of encounters. Specific mother-counselor pairs showed little correlation in their perceptions of whether a treatment recommendation had been made (R = 0.0) or a choice had been given about resuscitation (R = 0.07). Despite a lack of perceived choice, mothers generally believed that they were included in treatment decisions (66.7%) and were satisfied with the amount of influence that they had in the decision-making process (73.3%).
Conclusions. The decision-making process in this study conforms most closely to a model of informed assent. Mothers may have been satisfied with this type of counseling because they felt informed and included in the decision-making process. Physicians and nurses need to elicit mothers' preferences to incorporate them into the treatment plan, as counseling about the resuscitation of extremely premature infants at delivery is considered directive by mothers even when it is not intended to be directive.
Key Words: counseling decision-making ethics extremely preterm infants resuscitation
Counseling parents about whether to resuscitate their extremely premature infant at delivery is fraught with difficulty. The counseling is often done by medical professionals who are relative strangers to the parents, at a time when the infant's birth is imminent and parents are in crisis. In addition, the statistics are not encouraging: infants who are born at <26 weeks of gestation have a reported mortality rate of >50%, with half of surviving children normal at 30 months of age, 25% with mild to moderate disabilities, and 25% with severe disabilities.1 Few factors distinguish these groups in advance, so predicting the outcome of any specific premature infant is profoundly uncertain. Even these general statistics are a moving target, as advances in the care of very premature infants change potential outcomes. How much responsibility parents wish to take in making these extremely grave decisions and how much authority health care providers wish to relinquish are currently unclear.
In the United States, the past several decades have witnessed a shift in the norms of medical ethics from physician paternalism to patient autonomy.2 Accounts of neonates who were treated aggressively against their parents' wishes appear not infrequently in the lay press and on the Internet, reinforcing the notion that parents resent physician paternalism and wish to have complete autonomy in making resuscitation decisions.3
Empiric research describing delivery room decision-making for premature infants is sparse but demonstrates physicians' desire to maintain some decision-making authority for very premature infants, with most accepting a parental role in decision-making for infants who are born at the 22- to 24-week range of gestation but not for infants who are born at 26 weeks.4 In a Canadian study that contrasted physicians' and parents' views about who should have the greater role in making delivery room decisions for extremely low birth weight infants, Streiner et al5 found that 64% of the parents who had a premature infant survive into their teens believed that all extremely low birth weight infants should be resuscitated regardless of their prognosis, compared with only 6% of the neonatologists and neonatal intensive care nurses. Parents' responses did not differ according to whether their child had a handicapping condition or not. Because of the study design, parents of children who were not resuscitated at delivery or who died after resuscitation were excluded, so the sample was biased.
Most recently, in a qualitative study of parents of premature infants in Norway, Brinchmann et al6 found that parents wished to be informed, listened to, and consulted, but generally did not wish to bear the burden of making the final decision about withdrawal of support in the NICU. A similar study of NICU parents in the United States has not been done.
Our aims in this study were to gain a better understanding of mothers' and counselors' perception of their roles in the decision-making process about the resuscitation of very premature infants at delivery and to gain insight into mothers' and counselors' satisfaction with the decision-making process. We believe that this is necessary information for an informed debate about how controversial health care decisions for very premature infants should be made.
METHODS
Design
This descriptive study was conducted over a 2-year period at a Southeastern public teaching hospital with a level III NICU. The hospital treats pregnant women from the immediate area and also serves as a regional referral center for perinatal and neonatal care. The study included women who presented to the hospitals' obstetric service and delivered an infant between 22 and 27 completed weeks of gestation and the person whom each woman identified as having primarily counseled her about how her infant would be treated at birth. Obstetric and neonatal patients are cared for by attending physicians, fellows, residents, nurse practitioners, and nurses.
All mothers who delivered a singleton infant between 22 and 27 weeks of gestation and received at least 1 session of predelivery counseling were eligible to participate. Exclusions were induced abortions, stillbirths in which the fetus's death was known before labor, fetuses with known lethal anomalies, non–English-speaking mothers, mothers <18 years of age (as adolescents may represent a distinct subpopulation and had not been included in development of this questionnaire), and mothers with multiple gestations (because there may have been differential counseling for each fetus). During the study period, it was policy that infants who were born at <23 weeks of gestation were considered nonviable and were not resuscitated; infants who were born at >25 weeks of gestation were considered viable and were resuscitated; and infants who were born at 23, 24, or 25 weeks of gestation were considered potentially viable and resuscitated or not on the basis of a decision made after counseling the parents about the outcomes of extremely premature infants. The study population was chosen to encompass a spectrum of counseling styles (directive to nondirective) and outcomes.
Before their discharge from the hospital, eligible mothers were asked to identify who primarily counseled them about the treatment that their infant would receive at delivery. The counselor interviewed was the person whom the mother identified regardless of who else counseled her. Pictures of attending physicians, fellows, residents, and nurse practitioners were available to help mothers with identification. When a mother remembered being counseled but could not definitively identify her counselor, the medical team was asked to identify who that person would have been. Counselors then were contacted to ensure that the match was correct. Mothers were also notified before hospital discharge that someone would call them in 6 weeks to invite them to participate in a research study about preterm delivery.
Six weeks after delivery, mothers were contacted by a nurse, who asked whether they were willing to be called by a researcher to discuss participation in a study about preterm delivery. When a mother accepted, she then was contacted by an interviewer and informed consent was obtained by telephone. Informed consent was also obtained from all counselors. This study was approved by the medical school's Institutional Review Board.
Data Collection
Mothers were interviewed 6 weeks after delivery by telephone using a standardized interview form developed for this study. At least 5 attempts were made to contact each mother at different times of day. Charts of eligible mothers and their infants were reviewed for maternal age, marital status, parity, previous preterm delivery, education, gestational age and birth weight, treatments received at delivery, and outcome (including survival status and major diagnoses known by 6 weeks after delivery). All counselors were interviewed by telephone within 72 hours of delivery using a standardized interview format.
Questionnaire Design
The questionnaire for mothers was designed using expert opinion, including a neonatologist, an obstetrician, an epidemiologist, a mother of an infant who had been hospitalized in the NICU, and the NICU bereavement counselor; data from the literature about which factors in counseling are important to patients7,8; and information from a small focus group of mothers who had delivered premature infants. The questionnaire then was pretested on 4 mothers of preterm infants. Opinion was also solicited from these mothers about the appropriate timing for an interview after a preterm birth. Six weeks after delivery was chosen to minimize intrusiveness from the interview but to maximize recall of predelivery counseling. The final questionnaire addressed the following items: the mother's satisfaction with the counseling that she received about her infant's treatment at delivery (including the type of information given, how the information was given, and whether it was too much or too little information), how directive the counseling was, which factors she considered when thinking about resuscitation, who she believed made the final decision about delivery room treatment, whether she had a choice about resuscitation, whether she was satisfied with the amount of influence that she had in the decision, and whether she was satisfied with the care that her infant received at delivery.
Counselors were asked similar questions about the content, tone, and amount of information discussed; how directive they were in their counseling; whether they gave the mother a choice about resuscitation; who made the final decision; and whether they were satisfied with the outcome. Answers to open-ended questions were recorded on the questionnaire.
Analytic Approach
Simple descriptive statistics (frequencies and percentages) were used to describe the demographic characteristics of mothers, counselors, and infants and the salient features of counseling and outcome. Mothers who chose to be interviewed were compared with those who did not participate to ascertain whether there was any systematic bias in who was included in the study cohort. Nonparametric comparisons were made using the Mann-Whitney U test. Exact methods were used when the expected cell size was <5. Pearson’s correlation coefficient was used to determine whether individual mother-counselor pairs agreed about the counseling and decision-making process, including the type of information discussed, how directive the counseling was, and whether the mother had a choice about resuscitation. Interviews were reviewed qualitatively to delineate the main themes of mothers' and counselors' responses to open-ended questions.
RESULTS
Forty-four mothers were eligible to participate in the study. Nine of the 44 were discharged before they or their counselor could be enrolled. One mother with eclampsia could not remember being counseled. One mother was unable to identify her counselor. This left 33 (75.0%) available mothers to enroll. Thirty-three counselors were identified by these mothers, and all counselors were interviewed. Fifteen mothers agreed to be interviewed at 6 weeks, yielding 15 mother-counselor pairs for analysis. The most frequent reason for nonenrollment was the inability to contact mothers by telephone (disconnected numbers). Demographics by maternal interview status are presented in Table 1. Participating mothers were similar to the nonparticipating group, with no statistically significant differences at a P .05 level.
Counselors
Characteristics of the counselors and their impressions of the counseling sessions are shown in Table 2. Counselors were predominately female (75.8%), had usually counseled the mother only once (63.6%), and were usually obstetric residents (45.5%) or neonatal fellows (30.3%). Two mothers identified the labor and delivery nurse as her primary counselor. Counselors reported that they discussed the infant's chance of survival (90.9%) more often than the potential for handicap (69.7%). Fewer than half said that they discussed suffering. Most of the counselors (81.8%) believed that they knew what the mother wanted for her infant because the mother told them. Counselors said that 64% of the mothers wanted "everything" done, and 18% wanted "treatment if the infant looks viable."
A majority (57.6%) of the counselors believed that they gave the mother a choice about delivery room resuscitation, and most (69.7%) stated that they had not made a specific recommendation. Nine (27.3%) of the counselors believed that the physician made the final decision about resuscitation, 9 (27.3%) believed that the mother made the final decision, and 13 (39.4%) believed that the decision was made jointly.
Twenty-nine of the counselors knew which kind of treatment the infant actually received at delivery. Four counselors had gone off duty and did not know. Twenty-five (86%) of the infants whose delivery room treatment was known by counselors were resuscitated, and 4 (12.1%) received comfort measures. Twenty-eight (84.8%) of the counselors were somewhat or very satisfied with the outcome of the delivery room decision, and 1 counselor was neutral; however, 3 (9.1%) counselors believed that the infant's treatment had been excessive. None of the counselors believed that an infant had received too little care.
Mothers
Mothers' impressions of the counseling sessions are presented in Table 3. More than 90% of the mothers who were interviewed were very or somewhat satisfied with the counseling that they received before delivery and satisfied with the care that their infant received at birth. Two thirds of the mothers said that the counselor had made a treatment recommendation, and 60% said that they did not have a choice about how their infant would be treated. Nonetheless, most mothers agreed with the counselor's recommendation and believed that they had had a voice in the decision. Although a majority of the mothers said that they had no choice about the treatment that their infant would receive at delivery, 73% were very satisfied with the amount of influence that they had in making the resuscitation decision. Thirteen (86.7%) of the mothers said that everything that could be done for their infant was done. No mother said that everything was not done, although 2 said that they did not know.
Mother-Counselor Pairs
A comparison of the impressions of the 15 mother-counselor pairs about their counseling session(s) is presented in Table 4. Counselors were more likely than mothers to report that they had discussed the potential for handicap and less likely than mothers to report that they had discussed future suffering. Of note, twice as many mothers reported receiving a recommendation about delivery room resuscitation than counselors reported making a recommendation (66.7% vs 33.3%, respectively). When the responses of individual mother-counselor pairs were compared, there was almost no agreement between mothers and their counselors about either the content or the directiveness of the counseling session(s).
Responses to Open-Ended Questions
Mothers were asked open-ended questions about what they considered when deciding how their infant would be treated to delineate further their responses. Most mothers responded with personal values, beliefs, or experiences, rather than mentioning the medical information that was presented during their counseling. Many mothers said that they simply wanted everything done, for example, "Never a question not to do everything," "Wanted everything done for him." Others relied on their faith, as 1 woman said, "I put my faith in God and hoped He would help the infant and the doctors." Only 2 mothers specifically mentioned the infant's prognosis, whether their infant would suffer, or the infant's future quality of life. Some of the mothers relied on their previous experience with premature infants. Some wanted the doctors to make the decision.
When mothers were questioned about why they were satisfied or dissatisfied with their counseling, the predominant theme that emerged was a desire for information. They appreciated explanations and knowing what would happen in the delivery room. In general, mothers who believed that they had no choice in the resuscitation decision stated that they were satisfied with the amount of influence that they had in the decision-making process for 3 main reasons: (1) they were given information: "Explained step-by-step what they would do"; (2) they trusted the physicians' judgment: "The doctors knew what they were doing"; or (3) they felt included in the process: "They told us if they got to a place where decisions had to be made (about continuing support) they would tell us," "Asked my opinion before they did anything," "We all talked, and our feelings and opinions were considered."
When mothers were asked what could be done to help patients like them in the future, all who responded expressed that they wanted more information with less medical jargon: "When doctors would explain, the words kept getting bigger and bigger; it would be helpful to have someone to break it down into more simple explanations." Some mothers suggested pamphlets or booklets.
Counselors were asked what recommendations they gave about the infant's treatment at delivery, what treatment they thought the mother wanted, and why they were or were not satisfied with the resuscitation decision. Most counselors stated that they had not given a recommendation. Counselors were more likely to state that they gave a recommendation for infants of a younger gestational age, including, "Infant was previable and should not be resuscitated," and, "Recommended that if infant <24 weeks and <500 g not to resuscitate." One counselor stated that she gave an implied recommendation. "I explained what was usually done for 25-weekers. It was an implied recommendation that this would be done for her infant unless she wanted something different." Some counselors stated that they did not give a recommendation because the mother had no choice: "I felt that based on the infant's age that the prognosis was good and justified full measures. Did not offer the option of not doing anything." When counselors who gave no recommendation were asked how the mother wanted her infant treated at delivery, most stated that the mother "wanted everything done." Even for very premature infants (<24 weeks' gestational age), counselors reported, "Mom wanted everything done including cesarean section, intubation, and everything the NICU could do," "Mom wanted everything done." Sometimes, however, counselors reported that the mother had stated a goal and wanted the physician to determine the level of care on the basis of that goal: "Wanted the infant evaluated for its size and vigorousness and to resuscitate if the infant had a good chance to survive," "Wanted infant to survive if it had a reasonable chance to survive long term," "Was realistic," "Did not want the infant to survive with a great chance of suffering."
Most counselors were very satisfied or somewhat satisfied with the decisions that were made about resuscitation. Those who were very satisfied commented on "doing what was best for the infant," "gave the infant every chance," or doing "what the mom wanted." Those who were somewhat satisfied were concerned about being overly aggressive, although treatment was in alignment with the mothers' wishes: "(I was) sorry the infant had to be put through that, although it's what the mom wanted," "(It's) hard to separate my personal beliefs about the long-term prognosis."
DISCUSSION
The most striking finding of this study is the lack of concordance between mothers and counselors about what occurred during their counseling session(s). There was no concordance on clinical information such as survival and potential for the infant to have a handicap or about who made decisions and whether the mother had a choice in how her infant would be treated. The lack of concordance between counselor and mother on issues of clinical information in this study is similar to that found in a study by Zupancic et al.9 Zupancic et al had the parent(s) and the counseling neonatologist fill out a questionnaire about details of potential pregnancy-related and neonatal problems within 24 hours of the counseling session and before the infant's delivery. Concordance between the neonatologist and the parents on clinical detail was poor. They found that increased maternal anxiety decreased the concordance of maternal-physician responses. Other studies have shown that only a small portion of information is retained after parents are given traumatic news about their child.10
Despite this lack of concordance between counselors and mothers, satisfaction with counseling was high in our study. As is frequently the case with premature deliveries, in this study, there was often little time between many mothers' sole counseling session and their infant's birth. It is possible in this situation that the actual medical information was not as important to mothers as their core values and beliefs. Many mothers reported that they based their decision on their previous experiences with premature infants, their faith, and their confidence and trust in the medical team, although some said that they did consider information from the counseling sessions about the likelihood of their child's suffering. Many mothers reported that not "trying everything" would not have been an option. Therefore, maternal satisfaction with counseling could have been more dependent on the tone of the counseling rather than the actual content. Studies of adult patients' satisfaction with the medical encounter have found that the physician's warmth, ability to listen, and provision of information are more directly associated with patient satisfaction than the physician's perceived technical competence.11–13 In our study, 73% of the mothers rated their counselors as very caring (8–10 on a 10-point scale), and 60% said that they were provided with the right amount of information, which may have contributed to the overall high level of satisfaction in this study.
There was also little correlation between the mother and her counselor about how directive the counseling was. Overall, mothers in this study were much more likely than counselors to say that a recommendation about delivery room resuscitation had been made (67% vs 33%), and the concordance within individual pairs was poor (R = 0.0). Mothers were also more likely than counselors to say that they were given no choice about their infant's treatment at birth (60% vs 47%; R = 0.07), also with poor mother-counselor pair concordance. Despite this lack of concordance, most mothers were satisfied with the decision-making process. There are several potential explanations for the mothers' satisfaction.
We found that mothers of extremely premature infants were satisfied with the process of decision-making when they were informed and given explanations, even when they did not believe that they had a choice about how aggressive treatment would be at delivery. Mothers in this study believed that they participated in the decision about resuscitation and were satisfied with the amount of influence that they had on the decision 73.3% of the time, although they believed that they did not have a treatment choice 60% of the time. These results are in agreement with the study by Brinchmann et al,6 which found that Norwegian parents wanted their values and opinions recognized and included but did not wish to bear ultimate responsibility for end-of-life decisions in the NICU. These results are also in agreement with studies of adult patients in the United States, which found that patients' desire for autonomous decision-making decreased as medical decisions became more serious, although their desire to be informed did not decrease.14
Comments made by the mothers suggest that most of them viewed their counseling sessions as inherently interactive and participatory. In this context, mothers may have interpreted treatment plans put forward by the physicians or nurses as "recommendations" that they were engaging together. Because the mothers in this study viewed the treatment plans that were presented to them as recommendations and perceived themselves to be joint decision makers, they were satisfied even when the counseling was very directive.
Another possible explanation for the mothers' satisfaction in this study is that they shared the same view about treatment at delivery as their counselors. Indeed, 82% of the mothers said that they agreed with the recommendation that they received, and many said in their comments that it would not have been an option not to "try everything." We cannot say whether the lack of perceived choice would have led to greater dissatisfaction had there been more disagreement between mothers and their counselors, as might have been the case had the study included more infants who were born before 23 weeks of gestation, for whom resuscitation would not have been offered. However, we also cannot say whether different recommendations would have led to more disagreement. Doron et al15 found that infants who were born at 23 to 26 weeks of gestation were resuscitated at delivery when either the parent or the neonatologist wanted the infant to be treated. Infants were not resuscitated only when both the parent(s) and the neonatologist agreed that comfort care was preferable. Thus, whoever preferred the most aggressive treatment at delivery became the decision maker for the infant. This suggests that if mothers wanted their infant treated at delivery even after counseling about potential poor outcomes, then the treatment was likely to be delivered, and mothers were correct in their perception about the participatory nature of the decision-making process.
Most counselors reported that they told mothers what would happen at delivery and then used their lack of dissent as an indication of consent. This directive counseling practice conforms more closely to the model of obtaining informed assent, as opposed to informed consent, for medical treatment. The informed consent doctrine requires counselors to provide information about the pros and cons of alternative treatments, including the choice of no treatment, so that patients can recognize and weigh options themselves. By contrast, the process of assent limits patient choice by obscuring options. This form of counseling is somewhat paternalistic, but unlike pure paternalism, in which the physician may determine and carry out treatment without patient input, the model of informed assent requires that the physician inform and educate the patient about the treatment and then elicit his or her assent before proceeding. Our finding that most of the mothers in this study were satisfied with the decision-making process even when they perceived that they were given no explicit choices indicates that the process of informed assent was acceptable to them when they were being counseled about delivery room resuscitation.
It is possible that these mothers' satisfaction with their counseling and the process of assent may reflect acceptance of a power hierarchy in the counselor-mother relationship, such that mothers could not conceive of questioning or speaking against the proposed plan of care or taking a more assertive role in decision-making. This seems unlikely, however, as most (80%) mothers said that they were never afraid to ask questions, many linked their satisfaction to the perception that their opinions and views were included, and some did elect full intervention against the counselor's recommendation. Comments by the counselors support this perception. Many counselors said that following the mother's wishes was an important factor in their own satisfaction with the decision-making process.
An unexpected finding in this study was the high proportion of junior obstetrics residents identified by mothers as their primary counselor about the resuscitation decision. Most previous work on delivery room decision-making for extremely premature infants has studied the viewpoints of attending neonatologists or obstetricians.4,5 However, in the teaching hospital setting, this may not be the person whom the mother identifies as her counselor even though she has spoken with an attending physician. As 1 of the mothers' primary desires was for clear and accurate information, it is important that these residents who do counseling have knowledge about the outcomes of prematurity and familiarity with the NICU practices that support premature infants after birth. This finding has implications for obstetrics residency training and education.
This study has both weaknesses and strengths. The primary weakness is its small study size. Also, no mother who was interviewed had an infant who was born at <23 weeks of gestation, for whom resuscitation at delivery would have been refused. It is possible that mothers who were not satisfied chose not to participate. Because of the small sample size, there were inadequate numbers to reach theme saturation in analysis of open-ended responses. Therefore, typical responses were included to illustrate themes of mothers' perceptions. It is also possible that by 6 weeks after delivery, mothers had difficulty separating out the counseling that they received before delivery from subsequent counseling and other information that they learned about premature infants later. This period was chosen after consulting with mothers who had delivered premature infants about the appropriate time frame for an interview. However, the difference in time to interview between counselors and mothers does represent a differential recall bias. This bias would be difficult to overcome as this was a singular, important event for mothers but not for counselors; therefore, mothers are more likely to remember the counseling event. Furthermore, this study was done at a single, tertiary care hospital and may not be generalizable to other settings. Thus, the results from this study should be interpreted with caution.
This study also has strengths. Previous, related research has relied on interviews of the medical team only, questionnaires, or surveys of selected parents of premature infants many years after delivery. This study is unique in that we directly interviewed all consenting mothers about their recent experience and linked their responses to the responses of the person whom they identified as having counseled them about delivery room resuscitation. We did not assume that the mother's primary counselor was the neonatal or obstetric attending physician. Therefore, this study provides insight into whom the mothers perceived as their primary counselor about delivery room resuscitation, as well as into the mothers' own impressions of that counseling. Finally, because we interviewed medical professionals about real patients whom they had recently counseled, rather than about hypothetical patients and situations, this study may provide a more accurate depiction of counseling as it is actually perceived and practiced.
CONCLUSIONS
This study suggests that mothers of extremely premature infants perceived the counseling that they received about resuscitation before their child's birth as directive. In general, mothers were satisfied with this type of counseling and considered themselves to be joint decision makers even when they were given no explicit choice about their infant's treatment at birth. This kind of counseling is closer to a model of informed assent than informed consent. A related study found that one quarter of parents said that their preference was to have the physician make treatment decisions for their premature newborn; however, physicians were poor at judging which patients preferred this type of decision-making.9 It is possible that a directive form of counseling that gives information and recommendations but also elicits patient preferences allows mothers to choose their level of participation in the decision-making process. This may relieve some families of the burden of a choice that they do not wish to make alone, while allowing other families greater autonomy.
Mothers of extremely premature infants in this study wished to be well informed and wanted their values and opinions recognized and included in the decision-making process. Counselors should be careful to elicit mothers' preferences for treatment so that these can be incorporated into the medical plan, as mothers perceived counseling to be directive even when the counselor had not intended it to be so.
ACKNOWLEDGMENTS
This study was supported by University Research Council grant 5-43612 from the University of North Carolina (Chapel Hill).
We acknowledge the assistance of the late Michael J. McMahon, MD, MPH, who contributed to the design of this study.
FOOTNOTES
Accepted Oct 21, 2004.
No conflict of interest declared.
REFERENCES
Wood NS, Marlow N, Costeloe K, Gibson AT, Wilkinson AR. Neurologic and developmental disability after extremely preterm birth. N Engl J Med. 2000;343 :378 –384
Quill TE, Brody H. Physician recommendations and patient autonomy: finding a balance between physician power and patient choice. Ann Intern Med. 1996;125 :763 –769
Partridge JC, Freeman H, Weiss E, Martinez AM. Delivery room resuscitation decisions for extremely low birthweight infants in California. J Perinatol. 2001;21 :27 –33
Streiner DL, Saigal S, Burrows E, Stoskopf B, Rosenbaum P. Attitudes of parents and health care professionals toward active treatment of extremely premature infants. Pediatrics. 2001;108 :152 –157
Doron MW, Veness-Meehan KA, Margolis LH, Holoman EM, Stiles AD. Delivery room resuscitation decisions for extremely premature infants. Pediatrics. 1998;102 :574 –582
Brinchmann BS, Forde R, Nortvedt P. What matters to parents A qualitative study of parents' experiences with life-and-death decisions concerning their premature infants. Nurs Ethics. 2002;9 :388 –404
Kirschbaum MS. Life support decisions for children: what do parents value Adv Nurs Sci. 1996;19 :51 –71
Harrison H. The principles of family-centered neonatal care. Pediatrics. 1993;5 :643 –650
Ende L, Kazis L, Ashe A, Moskowitz MA. Measuring patients' desire for autonomy: decision making and information-seeking preferences among medical patients. J Gen Intern Med. 1989;4 :23 –30
Brody DS, Miller SM, Lerman CE, Smith DG, Lazaro CG, Blum MJ. The relationship between patients' satisfaction with their physicians and perceptions about interventions they desired and received. Med Care. 1989;27 :1027 –1035
Lerman CE, Brody DS, Caputo GC, Smith DG, Lazaro CG, Wolfson HG. Patients' perceived involvement in care scale: relationships to attitudes about illness and medical care. J Gen Intern Med. 1990;5 :29 –33
Buller JK, Buller DB. Physicians' communication style and patient satisfaction. J Health Soc Behav. 1987;28 :375 –388
Jedlicka-Kohler I, Gotz M, Eichler I. Parents' recollection of the initial communication of the diagnosis of cystic fibrosis. Pediatrics. 1996;97 :204 –209
Zupancic JAF, Kirpalani J, Barerett J, et al. Characterising doctor-patient communication in counseling for impending preterm delivery. Arch Dis Child Fetal Neonatal Ed. 2002;87 :F113 –F117(Heather T. Keenan, MDCM, )