OBJECTIVES: To pilot-test a visual aid developed to help counsel pregnant women.
METHODS: After agreeing to participate, pregnant women at >28 weeks of gestation were assigned randomly to counseling with or without a visual aid. The visual aid contained pictures, graphics, and short messages about delivery room resuscitation, chances of survival, anticipated neonatal course, and long-term neurodevelopmental disabilities. A neonatal fellow performed counseling with a standardized script for an anticipated delivery at 23 weeks of gestation. In precounseling and postcounseling sessions, women were given a structured interview to assess their knowledge of chances of survival and disability and attitudes toward resuscitation.
RESULTS: Of the 89 women who participated, 76% were black and 59% read below a 9th-grade level. Compared with the no–visual aid group, women in the visual aid group recalled more disabilities and predicted longer neonatal stays (P = .01). For both groups, mothers' perceptions of the chances of survival were lower after counseling; the decrease was greater in the visual aid group (P = .03). The majority of women in each group opted for resuscitation, which was not affected by counseling. In multivariate analyses, use of the visual aid was a significant independent factor in explaining before/after differences in survival chances and recall of a long NICU stay and number of disabilities; higher literacy levels also were significant for recalling the number of disabilities.
CONCLUSIONS: Use of a visual aid improved mothers' knowledge and showed promise as a decision aid for counseling at the threshold of viability.
WHAT'S KNOWN ON THIS SUBJECT:
Several guidelines emphasize the need to counsel pregnant women with impending delivery of an extremely preterm infant to involve them in decisions regarding delivery room resuscitation. The most effective way to counsel women in such circumstances is not known.
WHAT THIS STUDY ADDS:
A visual decision aid improved parents' knowledge of chances of survival and disability without influencing their views regarding resuscitation.
Advances in neonatal intensive care have led to improved survival rates for extremely premature infants born between 22 and 24 weeks of gestation.1,2 Chances for survival increase from 6% for infants born at 22 weeks to 26% for those born at 23 weeks and 55% for those born at 24 weeks.2 However, the incidence of severe neurodevelopmental disability remains high among long-term survivors.3,–,9
Several guidelines and reports have emphasized the important role of perinatal counseling and the need to involve parents in making decisions about delivery room management.10,–,13 Parents' understanding of the information given and participation in decision-making are vital in helping the neonatologist develop a plan of care that parents find acceptable. The best practices for perinatal counseling are unknown. Both written information and provider counseling might be poorly understood by parents. Previous research found a discrepancy in what providers want parents to know and what parents actually understand.14
Limited health literacy might be a hidden problem in perinatal counseling. The Institute of Medicine reported that 90 million US adults have trouble understanding health information and that most health information is unnecessarily complex.15 Charts and tables that are easily understood by providers might be confusing to parents with limited literacy.16 Parents at all literacy levels might have difficulty understanding risk information, particularly percentages and probabilities.17,18
The Department of Health and Human Services National Action Plan to Improve Health Literacy states that people need information they can understand and act on to make informed health decisions. This action plan calls for novel approaches for improving techniques to provide health information.19 The objective of this study was to pilot-test a visual aid specifically developed to help neonatologists communicate more effectively with parents when delivery at the threshold of viability is imminent. We hypothesized that parents' information recall would be greater with the use of a visual aid.
Study Design and Sample
A randomized trial was conducted in the prenatal obstetrics clinic of a university hospital that serves primarily low-income individuals. All enrolled patients received a standardized hypothetical counseling scenario for delivering a premature infant at 23 weeks of gestation. One group received counseling with the use of a visual aid, and the other group received the same counseling but without the visual aid.
The Louisiana State University Health Sciences Center-Shreveport Human Research Protection Program required that the study be conducted in the prenatal clinic with women who were at least 28 weeks pregnant, rather than at 23 weeks of gestation. The Human Research Protection Program was concerned that the discussion at 23 weeks of gestation might generate unnecessary anxiety in the subjects. The study protocol was approved by the institutional Human Research Protection Program.
Randomization was performed by alternating the methods of counseling on different days of the week. Specifically, the visual aid was used on the first 3 days of 1 week and the last 2 days of the following week. This pattern was continued until the study was completed.
A sample size of 45 patients per group was calculated on the basis of the ability to detect a 25% increase in recall of information among mothers who received counseling with the visual aid; we used a 5% level of significance and a power of 80%. Pregnant women attending the prenatal clinic for regular checkups were recruited between October 2008 and November 2009. Patients were considered eligible if they were 17 years of age or older, were at ≥28 weeks of gestation, were not in active labor, were carrying a fetus without evidence of a congenital anomaly, and spoke English. Clinic nurses pulled the charts of eligible women waiting for their physician visits and gave them to the neonatal fellow.
The neonatal fellow asked eligible patients if they were willing to participate in a study about counseling of pregnant women who were facing threatened delivery of an infant at 23 weeks of gestation. He explained that this counseling did not pertain to their pregnancy or their infant. The patients were paid $10 for their time. A total of 101 women were approached, and 12 declined. Of those who declined to participate, 3 had other clinic appointments, 6 had transportation problems, and 3 were not interested in participating. A total of 89 patients were enrolled in the study and participated in the consent process with the fellow.
A standardized prenatal counseling scenario involving imminent delivery at 23 weeks was prepared by using survival data from our institution for the preceding 4 years and neurodevelopmental morbidity data reported in the current literature (see Table 1 for all topics covered). Counseling was nondirective and was prepared according to American Academy of Pediatrics guidelines and Neonatal Resuscitation Program recommendations.11,–,13 The plain-language counseling script, which was written at a 5th-grade reading level, was developed by the authors and delivered by a single neonatology fellow.
The visual aid (Appendix) contained graphics depicting delivery room resuscitation, a picture of an extremely premature infant on a ventilator, and ultrasound images of intraventricular hemorrhage and postsurgical necrotizing enterocolitis. Graphical representations were used to depict the critical gestational age, long-term developmental delay, and comfort care. Short, easy-to-read messages were placed next to each graphic.
The neonatology fellow administered a brief precounseling survey, which addressed perceptions about the chances of survival of the infant, chances of disability, and attitudes toward delivery room resuscitation at 23 weeks of gestation. A postcounseling survey was administered by a trained research assistant who was not blinded to the counseling method. Items in this survey addressed sociodemographic information, previous pregnancies, previous preterm delivery experiences (self and family/friends), previous NICU exposure, having a child with a disability, and how religious the subject considered herself to be (scale of 1–10). Questions assessing recall of information presented by the neonatal fellow, in the format of “how many out of 10 infants,” addressed chances of survival, chances of disability, and severity of disability. Questions also addressed the woman's choice between full resuscitation and comfort care in the delivery room, who should make decisions about infant care, and whether the woman would choose to talk to someone while making decisions.
After completing the survey, the research assistant administered the Rapid Estimate of Adult Literacy in Medicine (REALM), which is the most commonly used test of patient literacy in health care research. The REALM is a health word recognition test that measures an individual's ability to pronounce commonly used words in health materials. Raw REALM scores can be classified into 1 of 4 reading levels, namely, 3rd grade or less (score range: 0–18), 4th to 6th grade (score range: 19–44), 7th to 8th grade (score range: 45–60), or 9th grade or above (score range: 61–66). In this study, low literacy is defined as a score of <61, indicating a reading level of 8th grade or less.20
Comparisons between the 2 groups included demographic characteristics, numbers of previous preterm deliveries and exposure to prematurity, the level to which the woman considered herself religious, and the person she considered to be the decision-maker for the infant's care. The 2 groups also were compared with respect to variables thought to be potentially affected by counseling, including chances of survival, chances of disability and level of disability, attitudes toward delivery room resuscitation, number of possible short-term problems, anticipated interventions, and length of stay in the NICU. The Wilcoxon rank-sum test was used for comparisons of continuous variables. The χ2 or Fisher test was used for comparisons of categorical variables. The Wilcoxon signed rank test was used to determine significant overall differences in perceived survival scores before and after counseling. Multivariate linear and logistic regression analyses with backward elimination of variables were used to determine significant independent factors for outcomes for which the visual aid and no–visual aid groups differed significantly.
Participants ranged in age from 17 to 35 years (mean: 22 years), 69% had incomes of less than $10 000, 76% were black and 20% were non-Hispanic white, 43% had not completed high school, and 59% were reading below a 9th-grade level. Women rated themselves as moderately religious (6.7 on a 10-point scale). Compared with the visual aid group, the no–visual aid group contained larger proportions of black women, women reading below a 9th-grade level, and women who had previously delivered a premature infant. There were no other significant demographic differences between the 2 groups (Table 2).
Before counseling, women in both groups overestimated the chances of survival and underestimated the risks of disability. After counseling, the perceptions of chances of survival were reduced in both groups; a significantly greater decrease occurred in the visual aid group (P = .03). After counseling, most subjects were able to report accurately the 2 of 10 chance of survival (74% in visual aid group and 79% in no–visual aid group). The difference between the 2 groups was not significant. Although more women in the visual aid group underestimated the chance of survival (1 in 10), this difference also was not statistically significant. Compared with women in the no–visual aid group, women in the visual aid group recalled significantly more short-term problems and long-term disabilities and predicted longer NICU stays (P = .01). Attitudes toward resuscitation did not change after counseling in either group; the majority opted for full resuscitation (Table 3).
Multivariate analysis indicated that use of the visual aid was the single significant independent factor responsible for the observed differences between the 2 groups. A higher literacy level also was found to be a significant independent factor with respect to recall of more disabilities (Table 4).
Prenatal counseling at the threshold of viability is a regular practice, but optimal methods of counseling are not known. Training providers improves their knowledge.21 However, previous attempts to improve counseling and care at the threshold of viability focused on achieving consensus among providers with respect to both content and approach.21,–,24 This is the first study of which we are aware that has assessed parents' knowledge and attitudes before and after counseling.
Counseling is challenging because of the complexity of the information, the uncertainty of outcomes for individual infants, the lack of institutional standardization, and the need for parents to be involved in making critical decisions at a very stressful time. To help parents make informed decisions, communication needs to be clear and accurate, and attention should be paid to parents' understanding, attitudes, and beliefs.
Communicating risk is often as difficult for clinicians as understanding risk is for patients.18 The Joint Commission encourages providers to use decision aids in the shared decision-making process, to inform patient decisions more effectively. In the absence of shared decision-making, interventions are provided to people who would not choose them and are withheld from those who would.25 Decision aids have been found to be useful in various clinical situations in which diagnostic and treatment dilemmas exist, such as cancer treatment, prenatal counseling, and genetic testing. Decision aids have been found to reduce difficulties in decision-making by improving knowledge and risk perception in such situations.26,27
In this study, a visual aid containing graphics and short messages was developed and used as a decision aid to enhance parents' knowledge about delivery room resuscitation, neonatal courses, and morbidity and mortality rates for infants born at the threshold of viability. Counseling was standardized, so that both groups were given the same information. Women in both groups anticipated severe disabilities among survivors after counseling, and the visual aid group recalled more short-term problems, more long-term disabilities, and a longer NICU stay. Mothers' expectations of chances of survival were decreased after counseling in both groups, and the reduction in expectations was greater among women counseled with the visual aid. These findings indicate that the visual aid improved provider communication regarding the problems and disabilities seen among infants born at the threshold of viability.
Previous surveys showed that the majority of parents preferred full delivery room resuscitation.28,29 Our findings were similar. In our study, however, women's attitudes toward resuscitation did not change with the use of the visual aid.
Fifty-five percent of the mothers in our study reported that they should be the primary decision-makers regarding delivery room resuscitation. The values were similar in the 2 groups and were in line with previous studies, which found that most parents wanted to decide on delivery room management.29,30 Providers and parents might have differing views about delivery room resuscitation of extremely premature infants.31 Blanco et al21 found that both obstetric and neonatal care providers underestimate chances for survival and overestimate long-term disability rates. In a survey of parents and physicians in Canada, 64% of parents but only 6% of physicians wanted to intervene to save an extremely premature infant, regardless of the weight or condition of the infant.29 In our study, before they were counseled, mothers overestimated the chance of survival and underestimated the risk of disability. These types of discrepancies have led to attempts to improve provider knowledge and to achieve consensus within a single center24 and in multicenter settings.22
Standardized counseling and the use of decision aids might help clarify the perceptions of both care providers and parents regarding shared decision-making, thereby improving parents' understanding. In 1 study, 60% of mothers reported that they were not given the choice of treatment for their infants, even when neonatal care providers documented involvement of the mothers in the decisions regarding resuscitation. Most of those women indicated that they wanted to participate in decision-making.31
Differences in understanding also can arise as a result of the way in which providers frame the message,28 the stress levels of parents,14 medications the mother is receiving, language differences between providers and families, and parents' ability to comprehend risk conveyed in quantitative terms.18 In previous studies, parents found printed gestational age-based survival outcomes32 and printed consensus of institutional policy24 to be helpful. However, parents' understanding of the information was not measured in those studies. Parents of premature infants in an Australian NICU who received audiotapes of conversations with the neonatologists reported that they had better knowledge of diagnoses, treatments, and outcomes.33
There is no consensus regarding whether or under what circumstances a directive versus nondirective approach to counseling should be used.12 We chose to provide a nondirective method when explaining the advantages and disadvantages of full resuscitation versus comfort care. Our study results differed significantly from those of a study that used directive counseling with printed, institution-based, consensus, management guidelines for infants born at 22 and 27 weeks.24 The latter approach resulted in a majority of parents endorsing the physician-preferred guidelines; 100%, 61%, and 38% of parents opted for comfort care at 22, 23, and 24 weeks of gestation, respectively.24 Although the parents were satisfied with the counseling and were comfortable in making such decisions, the knowledge base of the parents was not assessed. When discussing long-term disabilities, we focused on profound impairments that are more likely to be viewed by parents as “worse than death.”34
Our study had several limitations. The 2 groups differed with respect to numbers of previous preterm births, racial distribution, and literacy skills; we could not identify a cause for the observed differences other than chance. This was a pilot study conducted in the artificial setting of a prenatal clinic, and patients generally were of low income and resided in 1 area of the country. The ability of mothers to recall information might be different in the stressful environment of the delivery room. However, we think that information presented in an easy-to-understand format with a visual aid might be even more helpful in the stressful situation of imminent delivery. Similarly, unlike in “real-life” situations, the decisions made did not have any direct impact on pregnancy outcomes. However, mothers' preferences for delivery room resuscitation in this study are in line with findings reported for other studies.18 We thought that, while being pregnant, women would better understand the real-life situation of having an extremely premature infant. Mothers in the group not using the visual aid had experienced more previous preterm deliveries. However, studies showed that previously having a premature infant does not change attitudes toward resuscitation.30
In real-life situations, many other factors might need to be included in the counseling session, such as fetal anomalies, chorioamnionitis, multiple births, and maternal and socioeconomic conditions. These additions could be included in the framework of standardized information that is provided with the visual aid. We did not record the conversations to verify the content. However, a standardized script was used every time, thus minimizing the potential for discrepancies.
Although American Academy of Pediatrics guidelines emphasize the importance of perinatal counseling with the impending delivery of infants at the threshold of viability, the best method of counseling is not currently known. The results of our pilot study show that the addition of the visual decision aid improved mothers' recall of information provided in counseling and, thus, might be a useful tool for helping parents make more-informed decisions. A visual decision aid and plain-language counseling script might be even more important for patients with low literacy levels and those experiencing a great deal of emotional stress. Larger trials using a visual aid in real-life situations are needed to clarify its usefulness.
This project was funded solely by the Department of Pediatrics, Louisiana State University Health Sciences Center (Shreveport, LA).
- Accepted August 16, 2011.
- Address correspondence to Venkatakrishna Kakkilaya, MD, Division of Neonatal-Perinatal Medicine, University of Texas Southwestern Medical Center at Dallas, 5323 Harry Hines Blvd, Dallas, TX 75390-9063. E-mail:
FINANCIAL DISCLOSURE: The authors have indicated they have no financial relationships relevant to this article to disclose.
- REALM —
- Rapid Estimate of Adult Literacy in Medicine
- Stoll BJ,
- Hansen NI,
- Bell EF,
- et al
- Hintz SR,
- Kendrick DE,
- Vohr BR,
- Poole WK,
- Higgins RD
- Hintz SR,
- Kendrick DE,
- Wilson-Costello DE,
- et al
- MacDonald H
American Academy of Pediatrics; American Heart Association. Textbook of Neonatal Resuscitation. 5th ed. Elk Grove Village, IL: American Academy of Pediatrics; 2006:1–16
- Batton DG
- Kattwinkel J,
- Perlman JM,
- Aziz K,
- et al
- Zupancic JA,
- Kirpalani H,
- Barrett J,
- et al
- Nielsen-Bohlman L,
- Panzer A,
- Kindig DA
- Doak C,
- Doak L,
- Root J
- Lipkus I,
- Samsa G,
- Rimer B
Joint Commission. “What Did the Doctor Say?” Improving Health Literacy to Protect Patient Safety. Oakbrook Terrace, IL: Joint Commission; 2007
US Department of Health and Human Services, Office of Disease Prevention and Health Promotion. National Action Plan to Improve Health Literacy. Washington, DC: US Department of Health and Human Services; 2010
- Schwartzberg JG,
- Van Geest JB,
- Wang CC
- Davis T,
- Kenneth E,
- Gazmararian J
- Blanco F,
- Suresh G,
- Howard D,
- Soll RF
- Ohlinger J,
- Kantak A,
- Lavin JP Jr.,
- et al
- Zabari M,
- Suresh G,
- Tomlinson M,
- et al
- Kaempf JW,
- Tomlinson MW,
- Campbell B,
- Ferguson L,
- Stewart VT
- Sepucha KR,
- Fowler FJ Jr.,
- Mulley AG Jr.
- O'Connor AM,
- Bennett CL,
- Stacey D,
- et al
- Haward MF,
- Murphy RO,
- Lorenz JM
- Streiner DL,
- Saigal S,
- Burrows E,
- Stoskopf B,
- Rosenbaum P
- Lee SK,
- Penner PL,
- Cox M
- Keenan HT,
- Doron MW,
- Seyda BA
- Koh TH,
- Butow PN,
- Coory M,
- et al
- Copyright © 2011 by the American Academy of Pediatrics