PEDIATRICS Vol. 121 No. 3 March 2008, pp. e590-e596 (doi:10.1542/peds.2007-1571)
ARTICLE |
Motivations of Mothers to Enroll Their Newborn Infants in General Clinical Research on Well-Infant Care and Development
a Department of Neonatology, Edmond and Lili Safra Children's Hospital, Sheba Medical Center, and Sackler Faculty of Medicine, Tel Aviv University, Tel Aviv, Israel
b Psychology Department, Bar-Ilan University, Ramat-Gan, Israel
 |
ABSTRACT |
|---|
 TOP ABSTRACT METHODSRESULTSDISCUSSIONCONCLUSIONSREFERENCES |
|---|
OBJECTIVE. The purpose of this work was to identify possible motivations for mothers to consent to having their newborn infants participate in medical research on well-infant care and to compare between mothers of term infants and mothers of stable preterm infants.
METHODS. Two-hundred mothers answered a questionnaire regarding their consent to have their newborns participate in 5 simulated studies at different risk levels and their willingness to provide a telephone number for future contact. Demographic data, attitudes about medicine, medical research, and evaluation of research conditions served as predictors of the degree of consent.
RESULTS. Degree of consent was affected mainly by perceived risk, because the research did not offer a direct personal benefit; that is, 80% consented to a psychological study as opposed to 25% who consented to a vaccine study. The strength of the predicting variables differentiated according to the suggested study. No significant difference was found between the mothers of term infants (n = 127) and the mothers of preterm infants (n = 73), either in the degree of consent to the 5 suggested studies or in the predicting variables, except for the measure of actual behavior (ie, revealing a telephone number). Only 23% of the mothers of term infants in comparison with 48% of the mothers of preterm infants were willing to reveal their telephone numbers.
CONCLUSIONS. There is some willingness to consent when the infant is healthy and the research is not directed at solving a specific problem of the infant. The degree of consent decreases in accordance with the increase in risk. The altruistic motive is the main predictor for research that is perceived as very risky. The benefit of learning about their infant's development served as a motivating force for less risky studies. We deduce that pointing out personal benefits to balance the usual conveyed information on risks or burdens of the research can increase the willingness to consent.
Key Words: clinical research • altruism • ethics • parents • newborn • attitudes
Abbreviations: MANOVA—multivariate analysis of variance • df—degrees of freedom • Fwilks—Wilks lambda approximation F statistic
Clinical research is vital to promote medical knowledge and improve medical care. Because infant participation in research depends on parental consent, understanding what motivates parents is of utmost importance. Most of the studies1–6 dealing with this issue focused on sick children and/or severely preterm infants, where the proposed research projects were usually relevant to the child's medical problem. Those studies did not focus on what motivates parents of healthy infants to enroll their child in well-infant care or diagnosis. The current research attempts to fill this gap.
Recent summaries7,8 of studies on parents' reasons for consenting to or opposing enrolling their premature infants can be divided into 4 main categories. First, perceived risk, that is, risk ranging from discomfort to actual danger of harm, was the main impediment to consent. Among children who are severely ill, however, the benefits may outweigh the fear of risk. Second, personal benefit, that is, learning about the child's illness4 and hoping that entering a trial will provide their infant with better than standard treatment,2 serves as the main motivation for consent in all studies. Third, altruism, that is, consenting for the sake of the welfare of other infants and for improving medical knowledge is a motivation that finds expression in all studies.1–6,9,10 Last is medical research attitudes, in which some of the studies1,2,5 found that trusting doctors and valuing medical research goals have a positive affect on consent.
The objective of this research was to find out whether these 4 categories are applicable and to determine their predictive strength with respect to "healthy" subjects. The experimental design, which involved giving consent to 5 different scenarios manipulated by degree of risk, enables standardized measurement of the effect of the research content. Our sample of healthy newborns and stable preterms makes it possible to test whether severity has an effect on consent, taking into consideration that stable preterm infants are, in essence, healthy infants and do not suffer from illness severe enough to outweigh all of the other considerations.
|
METHODS |
|---|
|
TOPABSTRACTMETHODSRESULTSDISCUSSIONCONCLUSIONSREFERENCES |
|---|
Sample
The sample was composed of mothers of healthy term newborns (n = 127) and mothers of stable growing preterm infants with no active disease (n = 73) who had given birth at the Sheba Medical Center, a large tertiary hospital in Israel, during a 3-month period (January to March 2006). The time of year was optimal, because it was free of holidays that disrupt the routine, which enabled a more standardized distribution of questionnaires. Although in such a clinical study it is impossible to build a strict representative sample, we found that this sample's sociodemographic background was very similar to the sample in a previous study of 204 women in the same hospital wards 3 years earlier.11 Namely, in both samples the average age was 31 years and average education was 15 years; 98% of this sample was married versus 97% in the previous sample.
Procedure
The mothers were asked whether they were willing to participate in a survey on "parental approach to medicine, particularly pediatrics." They understood that answering the questionnaire was voluntary, not compulsory, and that anonymity was ensured. The mothers of term infants were approached during their very short period of postnatal hospitalization. The mothers of preterm infants, who were not hospitalized, answered on the hospital premises while they were with their infants. The many chores of the mothers during their time at the hospital explain why the dominant reason for rejection on approach was lack of spare time.
The clinical necessity to delay approaching mothers of preterm infants, because of both the trauma she just went through and waiting until the infant's clinical condition was stable, may have a slight affect on the differences of responses between the 2 study groups. However, it was more crucial that both groups answer under similar circumstances at the hospital.
Because of the anonymous design, it was not possible to compare those who rejected on approach to those who consented to answer. However, once the women accepted the questionnaire, the amount of "no answer" (missing values) to the questions on consent to research can give some sort of indication of possible bias of the rejecters. No significant correlations were found between the "missing value" score and the sociodemographic and health variables, including age (–.04), education (–.08), financial status (–.05), religion (–.08), mother's health (–.05), and infant's birth weight (0.06), or the source of the questionnaire, that is doctors or students (.05).
The Questionnaire
The research questionnaire was constructed specifically for this study, although it was based partially on previous research.2,3,5,10,11 The main contribution came from an instrument aimed at investigating parents' perceptions of research using newborns.12
Our questionnaire consisted of 5 parts in the following order: (1) attitudes toward medical research; (2) evaluation of conditions that may heighten consent to research; (3) degree of consent to 5 research scenarios; (4) demographic variables; and (5) willingness to reveal telephone number. This order was chosen to "ease" the way toward the questions of commitment.
Attitudes toward medical research (see Table 4) included 11 attitudes regarding the legitimacy of medical research in general and with infants in particular. The answers ranged on a Likert-type scale from 1, "completely disagree," to 5, "agree wholeheartedly."
|
Evaluation of research conditions (see Table 2) included 9 conditions that may have an effect on consent. The mothers were asked to indicate to what degree they think they would permit their infant to be enrolled in research under such a condition. The answers ranged on a Likert-type scale from 1, "I am sure I would not consent," to 2, "I think I would not consent," to 3, " I think I would consent," to 4, "I am sure I would consent."
|
For the degree of consent to the 5 scenarios, 5 possible studies that may be conducted in a hospital were presented. The mothers were asked to indicate to what degree, on a Likert-type scale, they think they would permit their infant to be enrolled in each of the 5 studies, according to the same 4 categories of consent mentioned above. The scenarios were described objectively, with a short description of aim and procedure, without labeling risk or benefit. The order of presentation was from the least penetrating to the most penetrating, which we hypothesized would be perceived as the most risky.
The first scenario (psychological) and the second scenario (physiologic) were defined as "the relationship between maturity at birth and psychological and physiologic development." The psychological study consisted of observing the infant during daily activity (sleeping, sitting, and eating). The physiologic study involved checking the infant's sight, hearing, weight, and reflexes. The third scenario (blood) required a single blood test, at the age of 3 months, for measuring hypercoagulability. It was explained that early diagnosis enables treatment. The fourth scenario (sleep) involved normal sleeping patterns. It required placing a "small watch" (Actigraph) on the infant's hand or leg (on top of the clothing) for 10 days to measure wake and sleep patterns. It was explained to the mothers that the instrument causes no harm and that it can be taken off during bath time, and so forth. The fifth scenario (vaccine) was explained as trying to advance the varicella vaccine to the young age of 1 month instead of the usual age of 1 year to overcome possible disease during the first year and that the efficacy of the vaccine and the possible adverse effects will be examined.
The fourth part of the questionnaire included sociodemographic data, such as age, number of children, level of education, income, and medical history, including fertility treatments, pregnancy, delivery, and infant's general condition. Finally, the mother was asked whether she would be willing to reveal her telephone number for future contact. This question served as a measure of actual behavior of consent, although, obviously does not comply her to consent to all of the hypothetical studies. The research ethics committee of Sheba Medical Center approved the study.
Statistical Analysis
On the 5 scenarios, A between X within mixed-model multivariate analysis of variance (MANOVA) was performed. The independent variables were the groups (preterm versus term) multiplied by research (5 scenarios) on the dependent variable, the degree of consent to participate. The categorical question of revealing telephone number by group was analyzed by 
2. MANOVA was used to analyze the differences between the 2 groups on each set of questions: attitudes toward medical research and conditions of research. Six linear regression equations were used to predict the degree of consent to each of the 5 experimental scenarios and to the willingness to reveal the telephone number. The predicting variables were the general questions and relevant demographic data. This analysis was conducted separately for the attitude questions and for the condition questions. All of the analyses were conducted by using SPSS 13.0 (SPSS Inc, Chicago, IL).
|
RESULTS |
|---|
|
TOPABSTRACTMETHODSRESULTSDISCUSSIONCONCLUSIONSREFERENCES |
|---|
Sociodemographic comparison between the 2 groups (term and preterm) showed that the groups did not differ on any of the sociodemographic variables, including average age (31.3 ± 4.7 vs 31.5 ± 6.1 years), average years of education (15.3 ± 2.4 vs 15.3 ± 2.5 years), percentage married (97.4% vs 98.6%), percentage with above-average income (67.0% vs 63.7%), and percentage of women who defined themselves as secular (57.1% vs 59.7%).
However, as to be expected, the groups differed on all of the health variables related to the problem of time of birth (P < .0001). Mothers of term infants compared with the mothers of preterm infants had >1 delivery (70.9% vs 43.8%), fewer fertility treatments (15.3% vs 39.1%), and fewer problems during pregnancy (24.5% vs 68.8%). The term birth weights were higher (3255 ± 505 vs 1752 ± 572 g), and the infants' gestational ages were greater (39.3 ± 1.25 vs 32.7 ± 3.12 weeks).
Consent to the 5 scenarios gradually dropped as risk increased (see Table 1). The mixed-model MANOVA yielded a significant difference between the scenarios (Fwilks = 52.83; degrees of freedom [df] = 4,165; P < .0001; 
2 = .56); no significant difference between the 2 groups (Fwilks = 0.66; df = 1,168; P > .417; 2 = .004); and no significant interaction (Fwilks = 0.85; df = 4,165; P > .497; 2 = .02). The statistically significant contrasts (P < .05) showed that consent to the psychological study was significantly higher than to all of the other studies. The lowest consent was for the vaccine study. The physiologic study showed greater consent than either the blood test or the sleep pattern study.
|
One third (32%) of all of the mothers revealed their telephone numbers, including 23.5% of mothers of term infants and 47.9% of preterm infants. The 2 groups differed significantly (
2 = 12.234; P < .0001; 
= .25). Predicting degree of consent from sociodemographic variables (age, education, income, medically oriented work, and religiosity) plus infant's term status (term or preterm) yielded different significant variables (regression coefficients) for each study and for revealing telephone number. Consent to the vaccine was not predicted by any demographic variable. The other studies were predicted as follows. As educational level rose, there was greater consent to the psychological (β = .22; P < .01), physiologic (β = .21; P < .02), and blood research (β = .26; P < .003) studies. As income rose, there was greater consent to the sleep research (β = .23; P < .01). Revealing the telephone number was predicted by the term of birth (β = .22; P < .010), as mentioned above. Age, type of work (medically oriented or not), and degree of religiosity did not predict consent. The following analyses were used to control for the influencing sociodemographic variables.
Hypothetical conditions of the nonspecified research that may affect consent revealed expected results (see Table 2). Almost all of the mothers were willing to consent to participate if the research has a very relevant benefit, for example, solving a medical problem. Very few were willing to consent if participation required much effort, as in visits to the hospital. From the intermediate items, we learned that if the risk was perceived as minimal and if there was a sense of contributing to medical knowledge, >50% of the women would consent. On the other hand, if the promised benefits were not directly relevant to the research, most mothers would not consent.
The MANOVA results showed no significant difference between the 2 groups on the degree of consent to the various conditions as a unit (Fwilks = 1.46; df = 9,160; P < .15). However, 1 condition did differ between the 2 groups (F = 4.50; df = 1,168; P <. 035; 2 = .03). Mothers of preterm infants were more willing to repeat visits to the hospital than were the mothers of term infants (mean: 1.95 > 1.69; percent: 33% > 12%).
Predicting consent from the declarations of preferred conditions revealed differentials between the scenarios and the revealing of a telephone number. Table 3 includes those variables that remained significant in the final stage of the 6 linear regression analyses. The only condition predicting consent to the riskiest scenario, vaccination, was altruistic. Lack of suffering was the primary condition predicting consent to other scenarios. The psychology observation study was also predicted by overcoming a feeling of harm. The willingness to make the effort of visits to the hospital predicted the physiologic and sleep studies, as well as willingness to reveal a telephone number. An interesting finding arose with the condition "if the study serves to solve a medical problem that the infant suffers from." The negative β values indicate that the more participants' declared their agreement with the necessity of such a condition, the less they were willing to volunteer for the developmental studies (psychology and physiology).
|
Attitudes toward medical research (Table 4) indicate that most of the mothers agreed that they should be the ones to decide on consent after receiving information from doctors. Although most did not agree that research is simply for the purpose of advancing doctors or to make drug companies rich, a little less than half fully trusted that the doctors would not endanger their infants. Although many agreed that medical research is vital for the good of all, only a third agreed that it is not forbidden to do research on healthy infants, and even fewer agreed that it is their own moral obligation to participate.
MANOVA results showed no significant difference between the attitudes of the 2 groups (Fwilks = 0.38; df = 11,172; P > 1). These attitudes have predictive power with respect to consent (Table 5). The moral motive was predictive of the physiologic, blood, and vaccination studies and of whether to reveal a telephone number. The more participants felt that such research was not forbidden, the greater their consent was to the psychological study and their willingness to reveal a telephone number. Perception of benefit to the infant led to consent to the physiologic and blood studies. With respect to attitudes toward the doctor, more trust led to greater consent to psychology studies. The lower the feeling that the research was for doctors' advancement, the greater the consent was to the sleep study. Consent to the risky vaccine study was increased by the attitude that doctors should make the decision.
|
Analyzing within the preterm group showed a relationship between severity and consent. The lower the weight at birth, the greater the consent to the psychological study (r = –.29; P < .02) and to the sleep pattern study (r = –0.24; P < .05), as well as willingness to reveal a telephone number (r = –0.45; P < .001). Two attitudes were related to severity. The lower the weight at birth, the greater the mothers' willingness was to consent on condition that "the infant receive special attention from the medical staff" (r = –0.27; P < .03) and the more they did not think that "it is forbidden to do medical research on healthy children" (r = –0.42; P < .001).
|
DISCUSSION |
|---|
|
TOPABSTRACTMETHODSRESULTSDISCUSSIONCONCLUSIONSREFERENCES |
|---|
This pioneering study examines parental consent to have their healthy infant participate in a clinical research project that is not related to sickness. Among all of the studies cited on motivation to consent, only 113 sampled parents of healthy infants, as well as preterm infants. No study dealt with well-infant care and development research projects.
The design of our study measured motivation and consent at the same time (the same questionnaire), in contrast to most previous studies1–6,9 that differentiated between immediate consent and later subjective explanation of the decision. Often there was a large time gap, in 1 study6 up to 18 months, between the consent and the questions about reasons for consent. However, even when the time gap was in hours,1 the estimated motivation was based on parents' memory, which may include rationalizations "after the fact." Our design of asking about attitudes unrelated to request for consent and seeking the statistical relation between motivation and consent overcame some of those problems.
The objective of this research was to determine whether the findings of previous studies can apply to "healthy" research. The comparisons were tested along the same 4 categories of motivation found previously.
Perceived risk was found to be a very strong negative motivation, similar to the findings in most studies.1–3,9,13 The less perceived risk in the scenario, the greater the consent. Perceived risk seems to be determined by the degree of invasion required. Therefore, the physiologic study was "riskier" than the psychological study, and the blood test was riskier than the sleep study. The individual differences in the perception of risk (suffering) were the main predictors in all of the scenarios except for the very risky vaccine.
Direct benefit of solving the infant's medical problem was the strongest motivation for consent in all of the studies.1,3–5,7–10 Although our mothers agreed hypothetically, such benefit did not apply in this case. Minor benefit of learning about the infant did predict consent for all of the mothers and was more pronounced within the low-weight preterm group. The indirect benefit of monetary payment was not considered as important by the mothers and did not have predictive powers. The benefit of receiving special attention from the medical staff, which was considered by some of the studies3 to be a very important motivator, was not considered so by our sample as a whole, nor did it have any significant predictive power. However, within the preterm infants group, this motivation gained in importance as birth weight decreased.
Altruism, as in most studies,1–5,9,10,13 was declared by mothers to be an incentive to consent. However, agreement with the truism that "medical research is vital" did not lead to agreeing that it is their personal moral obligation to participate in such research. Within the preterm group, this duality was even more pronounced. One mother of twins expressed this quite vividly. "On the one hand, since my infants are premature I am interested in follow-up and in the knowledge that the researchers may provide; on the other hand, since they've suffered from so much invasive treatment, I am not interested in having them participate in any research that involves invasive procedures, such as pricking and injections."
The predictive power of the altruistic motive of the whole sample was the strongest for the vaccination scenario. The few women who regarded research as their moral obligation were also more willing to participate in the physiologic development and blood test scenarios, as well as to reveal their telephone numbers.
Attitudes toward medicine can be broken down as described here. First is informed consent. As in most studies,2 the vast majority wants to be informed and to make their own decisions. However, in the case of the risky vaccine study, the doctor's decision may increase consent. Second is the perceived motive of the researcher: only a small minority thought that all research is done for ulterior motives (similar to findings in other countries13); most held an intermediate attitude. What is more troublesome is that a fifth of the mothers do not trust doctors not to "endanger infants just for the research."
Several limitations of this present study must be acknowledged. First, handing out the questionnaire in the hospital did not enable us to question the fathers. Second, the anonymous design did not make it possible to understand the underlying reasons for rejection, although most claimed it was a time problem. However, there was no sociodemographic or health selection among those who responded partially or fully. Third, the scenarios were hypothetical and not based on "real life." We tried to overcome this to some extent by explaining that these are proposed research studies. The telephone number request was real.
Before outlining the specific conclusions of this research, we would like to point out the communality of the findings between our study and previous ones. Despite differences in designs, analysis methods, and samples (size, gender, and country), the basic motivations are present for all. The differences lie in the magnitude of the motivations, depending on the perceived risk or benefit of each study. This perception itself is influenced by the condition of the infant.
|
CONCLUSIONS |
|---|
|
TOPABSTRACTMETHODSRESULTSDISCUSSIONCONCLUSIONSREFERENCES |
|---|
When healthy infants are studied, parental consent cannot be taken for granted. Altruistic components of the research should be emphasized to increase consent. This emphasis should include making sure that the doctors or the researchers convey that the study is not just for self-promotion, financial motivations, or other nonaltruistic reasons, and that the doctors can be trusted not to harm the child. If the risk is not too great, personal benefit can be made salient. Most parents are interested in knowing about their child's development. Ensuring individual information at the end of the study will greatly increase cooperation.
|
FOOTNOTES |
|---|
Accepted Jul 31, 2007.
Address correspondence to Ayala Maayan-Metzger, MD, Department of Neonatology, Sheba Medical Center 52621, Tel Aviv, Israel. E-mail: maayan{at}post.tau.ac.il
The authors have indicated they have no financial relationships relevant to this article to disclose.
Drs Maayan-Metzger and Kedem-Friedrich contributed equally to the article as first authors from different disciplines.
This trial has been registered at www.clinicaltrials.gov (identifier NCT 00332150).
|
REFERENCES |
|---|
|
TOPABSTRACTMETHODSRESULTSDISCUSSIONCONCLUSIONSREFERENCES |
|---|
- Hoehn KS, Wernovsky G, Rychik J, et al. What factors are important to parents making decisions about neonatal research?
Arch Dis Child Fetal Neonatal Ed.2005; 90
(3):F267
–F269
[Abstract/Free Full Text] - Mason SA, Allmark PJ. Obtaining informed consent to neonatal randomised controlled trials: interviews with parents and clinicians in the Euricon study. Lancet.2000; 356 (9247):2045 –2051[CrossRef][Web of Science][Medline]
- Morley CJ, Lau R, Davis PG, Morse C. What do parents think about enrolling their premature babies in several research studies?
Arch Dis Child Fetal Neonatal Ed.2005; 90
(3):F225
–F228
[Abstract/Free Full Text] - Rothmier JD, Lasley MV, Shapiro GG. Factors influencing parental consent in pediatric clinical research.
Pediatrics.2003; 111
(5 pt 1):1037
–1041
[Abstract/Free Full Text] - Zupacnic JA, Gillie P, Streiner DL, Watts JL, Schmidt B. Determinants of parental authorization for involvement of newborn infants in clinical trials. Pediatrics.1997; 99 (1). Available at: www.pediatrics.org/cgi/content/full/99/1/e6
- Stenson BJ, Becher JC, McIntosh N. Neonatal research: the parental perspective.
Arch Dis Child Fetal Neonatal Ed.2004; 89
(4):F321
–F323
[Abstract/Free Full Text] - McKechnie L, Gill AB. Consent for neonatal research.
Arch Dis Child Fetal Neonatal Ed.2006; 91
(5):F374
–F376
[Abstract/Free Full Text] - Platt MW. Participation in multiple neonatal research studies.
Arch Dis Child Fetal Neonatal Ed.2005; 90
(3):F191
[Free Full Text] - Hayman RM, Taylor BJ, Peart NS, Galland BC, Sayers RM. Participation in research: informed consent, motivation and influence. J Paediatr Child Health.2001; 37 (1):51 –54[CrossRef][Web of Science][Medline]
- Madsen S, Holm S, Riis P. Ethical aspects of clinical trials: the attitudes of the public and outpatients. J Intern Med.1999; 245 (6):571 –579[CrossRef][Medline]
- Maayan-Metzger A, Kedem-Friedrich P, Kuint J. To vaccinate or not to vaccinate: that is the question—why are some mothers opposed to giving their infants hepatitis B vaccine? Vaccine.2005; 23 (16):1941 –1948[CrossRef][Medline]
- Oberle K, Singhal N, Huber J, Burgess E. Development of an instrument to investigate parents' perceptions of research with newborn babies.
Nurs Ethics.2000; 7
(4):327
–338
[Abstract/Free Full Text] - Singhal N, Oberle K, Burgess E, Huber J. Parents' perceptions of research with newborns. J Perinatol.2002; 22 (1):57 –63[CrossRef][Medline]
PEDIATRICS (ISSN 1098-4275). ©2008 by the American Academy of Pediatrics
CiteULike 
Connotea 
Del.icio.us 
Digg 
Facebook 
Reddit 
Technorati 
Twitter What's this?
| ||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||
