Click for next page ( 103


The National Academies | 500 Fifth St. N.W. | Washington, D.C. 20001
Copyright © National Academy of Sciences. All rights reserved.
Terms of Use and Privacy Statement



Below are the first 10 and last 10 pages of uncorrected machine-read text (when available) of this chapter, followed by the top 30 algorithmically extracted key phrases from the chapter as a whole.
Intended to provide our own search engines and external engines with highly rich, chapter-representative searchable text on the opening pages of each chapter. Because it is UNCORRECTED material, please consider the following text as a useful but insufficient proxy for the authoritative book pages.

Do not use for reproduction, copying, pasting, or reading; exclusively for search engines.

OCR for page 102
APPENDIX D Research on Childbirth Settings: The Assessment of Psychological Variables Camille B. Wortman in consultation with Miriam C. F. Kelty At present there is a great deal of interest in conventional and nonconventional birth settings. The incidence of home delivery is believed to be increasing (Mehl et al., 1977), and alternative birth settings such as nonhospital childbearing centers, are becoming in- creasingly prevalent (Faison et al., 1979J Pragmatics, Inc., 1978). Although opponents of alternative birth settings stress the medical risks involved, advocates emphasize the psychological advantages of nonhospital environments and the freedom they provide from excessive medical intervention. Unfortunately, there is little objective evi- dence to support any of these claims (see Appendix A). A recent review of research on home and hospital birth settings emphasized that •this lack of data has been a major factor preventing effective and reasoned dialogue among health professionals and lay people, especially those holding widely divergent views• (Adamson and Gare, 1980). In coming years, more research should be directed toward studying both hospital and nonhospital birth settings. TO date, the limited research on this topic has focused on aedical outcomes such as fetal and neonatal death rates. This paper discusses the assessment of psychological variables in research on birth settings. The boundaries of this field have yet to be established, and the terrain remains virtually unmapped. The research findings suggest that aany oppor- tunities exist for productively using existing psychological concepts, constructs, and theories. Thus, explorations of the psychological aspects of childbirth settings might reward those who can overcame the substantive and methodological obstacles to conducting research in this field. After a review of the evidence regarding the importance of psycho- logical variables in the birth process, this paper discusses same methodological issues concerning the assessment of psychological vari- ables. These include: (1) the tt.ing of assessment, (2) the need to assess background and setting variables that may influence psychologi- cal variables, (3) the importance of longitudinal research (studying research participants over several points in time) and long-term follow- up (assessing the effects of a treatment or procedure at one or more points in time), (4) the importance of assessing psychological variables through multiple modes (e.g., objective observation and self-report), and (5) the need for multivariate approaches to psychological variables. 102

OCR for page 102
103 This discussion is followed by tables listing specific psychological variables likely to influence the birth process. For ease of pre- sentation they have been organized by the target of assessment (e.g., mother, father, infant, mother-father, mother-infant) and by the time of assessment (e.g., during pregnancy, during labor or delivery, just after birth). The tables also provide information regarding whether these variables have been assessed in prior research on the birth process, and if so, how this assessment was made. Instruments that are currently available for assessing these variables are described and evaluated, and areas where new scale development is necessary are discussed. RATIONALE FOR INCLUDING PSYCHOSOCIAL VARIABLES IN RESEARCH ON BIRTH SETTINGS In the past, most studies evaluating birth settings have focused on mortality and morbidity data (Adamson and Gare, 1980). Because ma- ternal mortality has become such a rare event, fetal and neonatal death rates are the common indicators used in such research. In some studies, investigators have taken into account the medical procedures used in a particular setting--such as the use of analgesia, oxytocin, low- or mid-forceps delivery, or episiotomies. Others have assessed the rate of such intrapartum and postpartum complications as meconium stain, hemorrhaging, or cesarean delivery (Barton et al., 19801 Goodlin, 19801 Mehl et al., 1977J Shy et al., 1980). Indicators of the infant's health status, such as birth weight and one- and five-minute Apgar scores, have also been recorded in a few studies (e.g., Chalmers et al., 1976a, 1976b, 1976C1 Faison et al., 1979J Mehl et al., 1977). However, the assessment of such psychological variables as parents• anxiety and emotional distress or parents' bonding to the infant has been notably absent in these studies. Advocates of home birth settings have emphasized the psychological advantages that they believe are conferred on the parents and the newborn infant. Some home birth advocates have argued that the woman's psychological well-being is jeopardized in hospital settings, where physicians are often perceived as authoritarian and impersonal (Arms, 1975). Others have maintained that the bonding between the baby and its parents is facilitated when labor takes place in a familiar, relaxed environment with supportive attendants (Stewart and Stewart, 1977). Still others have stressed the advantages for the other sib- lings and the positive effects on relationships in the family (e.g., Kitzinger and Davis, 1978). Many of these proponents believe that although risk factors can never be foreseen and eliminated in all cases, the positive aspects of home births outweigh the risks involved. For these reasons, research on home birth settings that does not include psychological variables may be dismissed as irrelevant by those sympa- thetic to the home birth movement. Although advocates of nonconventional birth settings have firm beliefs regarding the psychological superiority of these settings, virtually all of the evidence in support of their position is anecdotal

OCR for page 102
104 (see, e.g., Kitzinger and Davis, 1978). Because home birth is con- sidered unsafe and there are few rigorous empirical data supporting the purported psychological benefits, many physicians have not been recep- tive to the arguments. The lack of sound empirical data also makes it difficult for potential parents or consumers to make a reasoned judgment regarding the birth setting that would be best for thea. Well-designed, methodologically sophisticated studies that include measurement of relevant psychological variables could be effective in stimulating a dialogue among advocates of different settings. Evidence that psychological factors can influence birth outcomes provides a second reason for including these variables in research on birth settings. For example, the results of numerous studies have suggested that psychological events or conditions during pregnancy can affect the progress of labor and delivery. It also appears that the psychological climate during labor and delivery can influence the course of labor and fetal outcome. Same preliminary research even suggests that the psychological environment during delivery can influence mater- nal and infant behavior for years to ca.e. Evidence for these asser- tions is discussed below in more detail. Evidence Relating Psychological Factors to Labor Outcome In several prospective studies, investigators have noted a relationship between the woman's psychological state in pregnancy and outco.e in labor and delivery. Zuckerman et al. (1963) reported that anxiety, as assessed by an adjective checklist given during pregnancy, was directly related to the amount of analgesic required during labor and delivery. Davids et al. (1961) found that ca.pared to woaen who experienced a •normal• delivery, wcaen who experienced aa.plications in the delivery room or who gave birth to children with abnoraalities scored signifi- cantly higher on a scale measuring anxiety that had been adainistered during pregnancy. Although the evidence is not entirely consistent (Becket al., 1980J Burnstein et al., 1974), nu.erous studies have suggested that maternal anxiety in pregnancy can affect both aaternal and fetal outcomes (Crandon, 1979a, 1979bJ Erickson, 1976J Gorsuch and Key, 1974J Pilowsky, 1971). ror exa.ple, Brickson (1976) found that women who experienced uterine inertia, a prolonged first stage of labor, rotation of the infant's head, low forceps delivery, or whose infant's Apgar score was less than five, had previously scored significantly higher on a scale measuring •fear for self• than waaen who did not experience these ca.plications. WOmen with any one of the first four of these complications also scored higher on a •fears for bab7• inven- tory than women who did not experience complications. These ca.plica- tions were highly correlated with one another. lOr exaaple, prolonged first stage of labor was associated with an increased risk of uterine inertia. On the basis of these data, the investigator concluded that •psychological stresses during pregnancy may initiate a sequence of complications which directly affect both the .other and the infant• (Erikson, 1976).

OCR for page 102
105 Several significant associations were found in a prospective study conducted to determine the relationship between several psychological variables in pregnancy and progress in labor (Lederman at al., 1979). During the second stage of labor, both acceptance of pregnancy and identification with the motherhood role were associated with epineph- rine (a hormone that stimulates the sympathetic nervous system), Montevideo units, and length of labor. Each of these psychological variables was also related to the length of labor during stage three, and to the type of delivery (e.g., whether the delivery was forceps- assisted). In fact these investigators also found significant negative correlations between the infant's Apgar score at five minutes and two variables assessed during pregnancy--conflict in accepting pregnancy and fear of loss of self-esteem during labor (Lederman et al., 198la). In a larger prospective study conducted with 8,000 gravidas, Laukaran and van den Berg (1980) examined the relationship between maternal attitude and pregnancy outcome. The proportion of women with postpartum complications (infections or hemorrhage) was larger in the negative attitude group than in the group of women holding favorable, moderate, or ambivalent attitudes. Even more striking was the finding that the pregnancies of women with negative attitudes resulted in a prenatal death or a live-born infant with a severe congenital anomaly more often than the pregnancies of women with the other types of attitudes. Studies in Animals Numerous studies, including those using infrahuman species, have suggested that anxiety and disturbance during labor can result in protracted labor and poor fetal outcome (for reviews, see Myers and Myers, 1979r Newton, 1977). An advantage of such research is the ability to assign the animals randomly to different labor disturb- ance groups. In one study (Newton et al., 1966a) mice were gently cupped in the experimenter's hands for one minute at various times during labor. In these mice there was a 65 to 72 percent slowing of labor in comparison to the undisturbed controls. In another experiment on the effect of environment on labor, mice randomly assigned to an unfaailiar environment for the duration of their labor delivered their first pup significantly later. Also, they delivered approximately 54 percent more dead pups than did mice placed in a familiar environment or rotated between a familiar and an unfaailiar environment (Newton et al., 1966b). More recently, a number of investigators have examined the impact of maternal distress during labor on various physiological indices of the mother and the fetus. For exa.ple, causing stress in maternal monkeys ~ shining a bright light in their faces resulted in a decrease in fetal oxygenation and an increase in acidosis in the fetus (Mori- shima, 1978). Similarly, the presence of strangers standing in front of the cage of pregnant monkeys resulted in a drop in fetal heart rate, blood pressure, pH (acidity/alkalinity), and oxygen levels, and a rise in carbon dioxide levels. In some cases the investigators observed fetal asphyxia approaching fetal deaise (Myers and Myers, 1979). ....

OCR for page 102
106 Studies in Human Beings In a number of studies, investigators have found a relationship between maternal anxiety during labor and subse- quent outcome. In one study, maternal-state anxiety (a temporary epi- sode, rather than an underlying •trait• of anxiety) on admission to the labor room was predictive of labor length (Becket al., 1980). In another study, anxiety assessed during the beginning of second stage labor was related to type of delivery, e.g., forceps-assisted or not (Lederman et al., 1979). The physiological basis for such findings baa been explored in studies relating maternal anxiety and various physio- logical indices (Lederman et al., 1978, 198la). For example, anxiety reported by the patient at the onset of second stage labor was signifi- cantly associated with endogenous plasma epinephrine (Lederman et al., 1978). Higher epinephrine levels were found to be significantly cor- related with decreased uterine contractile activity and longer second stage labor. In a subsequent study (Lederman et al., 198lb) patients' self-reports of anxiety during labor were significantly correlated with plasma epinephrine levels. Both anxiety and high levels of epinephrine were associated with changes in the fetal heart rate in the third stage of labor. The fetal heart rate pattern was also correlated with Apgar scores at one and five minutes. The association between aaternal anxiety and plasma epinephrine is especially interesting in light of the evidence that catecholamine& (a group of compounds that affect the sympathetic nervous system) may decrease the blood supply to the placenta and prolong the first stage of labor (see Levinson and Shnider, 1979, for a review). There is also evidence that characteristics of the labor setting, which would presumably influence maternal anxiety, can affect the labor process. FOr example, in a study of 49 women in a childbirth education group, women whose husbands were unable to attend the sessions reported higher levels of pain during labor (Renneborn and COgan, 1975). In this study the direction of causality is difficult to ascertain: Husbands may have been leas likely to attend the birth if they expected their wives to experience a great deal of pain. Therefore, it is notable, as will be described below, that investigators who ~ave experimentally manipulated various aspects of the birth environments have produced similar findings. In one such study, healthy Guatemalan women were randomly assigned to one of two experimental conditions (Soaa et al., 1980). The women in the •experimental• group were accompanied during labor and delivery by a previously unencountered but supportive lay woman. They were compared to women in the second group who labored and delivered as usual, without a support companion. There was a highly significant difference in the number of subsequent perinatal problems in the two groups (e.g., meconium staining, stillbirths, cesarean sections, oxy- tocin augmentation, and forceps delivery). In fact it was necessary to admit 103 mothers to the control group, but only 33 mothers to the ex- perimental group, to obtain 20 in each group with uncomplicated deliv- eries. Only 12 women (37 percent) in the experimental group experi- enced complications as compared to 79 women (75 percent) in the control group. Even when mothers with complications were excluded, the length of time from admission to delivery was significantly shorter for mothers

OCR for page 102
107 in the experimental group than for those in the control group (8.8 versus 19.3 hours). In the past, psychological variables have often been considered less important than indicators of physical well-being. The studies described above show that psychological variables play a central role in the birth process by exerting a strong influence on the progress of delivery and on the well-being of the mother and infant. Moreover, because these variables have been shown to influence such factors as the length of labor and likelihood of complications, they may also influence costs of health care. The effects of the woman's psychological state during labor may not be limited to the course of labor itself. Dysfunctional labor may have an adverse effect on the infant's subsequent development. Although the data are not entirely consistent (Broman et al., 1975), there is some indication that protracted labor and instrument-assisted delivery are associated with abnormalities in the child's speech, language, and hear- ing at three years of age and a lower IQ at four years (Friedman et al., 1977). Moreover, the events that occur during labor may influence the infant's development indirectly. For example, some psychological fac- tors such as anxiety or the presence of the woman's husband may influ- ence the use of anesthesia and other drugs that can be transferred to the fetus and thereby influence the newborn's behavior. Borgstedt and Rosen (1968) have shown that sedative or narcotic drugs administered to the mother during labor can cause at least transient central nervous system depression in the newborn. Also, parents may show less interest or different patterns of care for an infant who is depressed, limp, or unresponsive at birth (Klaus and Kennell, 1976). There is also some evidence that the psychological climate in which the birth takes place can directly influence subsequent parental behav- ior toward the offspring. Women randomly assigned to a group with a companion present during labor and delivery were more awake after de- livery, and they also stroked, smiled at, and talked to their babies more than the control mothers did (Sosa et al., 1980). Similarly, women who were randomly assigned to groups receiving 16 hours of extra contact with their infants shortly after birth behaved differently toward these infants at a follow-up visit one year later (Kennell et al., 1974). Extra-contact mothers were more preoccupied with their infants, more likely to soothe the child when it cried, and more likely to kiss the baby. In a follow-up of this group of mothers and infants after two years, significant differences were found in the speech patterns of mothers previously assigned to extra contact. While addressing their two-year-olds during informal play, those mothers given extra contact asked significantly more questions and used more adjectives and words per preposition, but fewer commands, than did control mothers (Ringler et al., 1975). These studies are very relevant to the study of birth settings because the variables found to be important (e.g., presence of a supportive companion and immediate postpartum contact with infant) are likely to differ as a function of birth setting. Given the profound effects shown to result from psychological fac- tors, it may be important to assess such variables not only in studies on birth location but also in research on other aspects of the process

OCR for page 102
108 as well. For exaaple, several randcaized studies have recently been conducted to exuine the impact of electronic fetal aonitoring. Most of these have shown an increased rate of cesarean section for the aon- itored group (Sosa et al., 1980). However, it is very likely that the control group of patients received aore ti•, more eJDOtional support, and aore physical contact from the nursing staff. T.bus, control pa- tients may have felt less anxious than the monitored group and there- fore aay have had a lower rate of cesarean section. In future studies of labor interventions, every effort should be made to ensure that the groups are equated on the relevant psychological variables. Including psychological variables in research on birth settings should also be helpful in uncovering the underlying biobehavioral pro- ceases that influence labor and delivery outccaea. For exaJIPle, what process can account for the superior outcomes among the waaen who had a supportive companion present? If Sosa et al. (1980) had noted the anxiety and plaa.. epinephrine levels in the waaen, we could begin to speculate about the underlying process involved. MB'l'BODOLOGIES IN 'l'BE ASSESSMBN'l' OF PSYCHOLOGICAL VARIABLES To assess psychological vari·ables in studies on alternative birth set- tings the investigator must consider nuaerous aethodological issues. For example, when should such asaessaents be made and what types of experimental designs should be used? Unless psychological variables are assessed with considerable aethodological sophistication, the results are unlikely to advance our knowledge about the birth process. Scae of the most coa.on concerns are explored below. T.be Timing of Assessment It is important to assess psychological variables as early as possible so that antecedents and consequences can be clearly distinguished. Some investigators have exuined such variables during labor or in the postpartua period. However, there are many advantages in assessing psychological variables at earlier stages of pregnancy. As suggested in the literature reviewed above, the woman's psychological reactions during pregnancy can have an independent influence on labor and delivery outccae. WOmen who have negative attitudes toward their pregnancy or who do not accept the mothering role have been shown to have aore com- plications than women with aore positive attitudes. Investigators who are interested in exaaining the effect of the birth setting on these same outccae variables will be able to conduct more sensitive analyses if the effects of earlier attitudes or anxiety patterns can be statis- tically partialled out or held constant. Assessing psychological variables before labor and delivery would be particularly important in nonrandcaized clinical trials comparing different birth settings. Wcaen who choose nonconventional birth set- tings are likely to differ with respect to important psychological vari- ables, and these alone may influence outccaes. For example, waaen who

OCR for page 102
109 decide on home births may, as a group, be leas anxious than those who decide on hospital deliveries. In a study ca.paring women who selected different birth alternatives, Cohen (1981) found that there were widely different attitudes toward desired involvement in the birth, toward pain, and toward hospital personnel. Even .,re importantly, wo.en who opted for nonboapital care were likely to be involved in supportive social relationships (Cohen, 1981). In contrast, approximately one- third of the hospital patients in this study could n... no genuinely supportive person. In several of the reaaining instances the woman opting for a hospital birth regarded her aate as ambivalent or unin- volved in the childbirth experience. If the a~M)Unt of social support available to the .,ther were not assessed prior to delivery, a number of aistaken inferences aight be drawn. Differences in social support, rather than the birth setting per ae, could result in improved outco.ea for the .,thers who have nonhoapital births. A second reason for early assessment of psychological variables is their possible interaction with birth setting variables to influence birth outcomes. Clearly, no one type of birth setting is ideal for everyone. SOme .,there want to be actively involved in the birth, whereas others want to be taken care of and are willing to •accept what may sometimes be impersonal, discontinuous, and routinized care while they relax and prepare themselves for the vicissitudes of the first weeks and .,nths at home• (Cohen, 1981, p. 11). However, •extra contact• may be a disaster for mothers with an unwanted pregnancy (de Chateau, 1977). In fact, some of these .,thers have refused extra contact with their infanta when it was offered. Investigators studying the effect of electronic fetal .,nitoring have also found divergent reactions dependent on the woman's personality characteristics and past experiences with pregnancy. Although a011e WOII8n judged the electronic fetal monitor to be reassuring, others found it upsetting (Starkaan, 1976). In abort, because women's reactions to a particular birth set- ting aay be dependent on psychological variables such as attitudes toward the pregnancy or personality disposition, it is important to assess such variables. Assessing Background and Setting Variables A wa.an•a attitudes toward pregnancy or feelings of anxiety during pregnancy may be influenced by such -.dical background factors as whether the pregnancy was planned or whether the mother had co-plica- tiona during a previous childbirth experience. Stailarly, a voaan•s feelings of anxiety during labor and delivery are likely to be affected by characteristics of the setting, e.g., the behavior of -.dical per- sonnel, the familiarity of the location, and the specific .edical proce- dures used. These studies have iaplications for the design of research to assess various birth settings. Hot only are there substantial differences in the attitudes of, and social support available to, women who select dif- ferent settings, but the settings theaaelves are likely to differ in aany ways (Cohen, 1981). The voaan•a position during labor, the aedical •

OCR for page 102
110 procedures performed, the supportiveness of attendants, the familiarity of the surroundings, and the amount of subsequent postnatal contact with the infant are just a few of the variables among birth settings. For this reason, investigators who compare the various settings and simply report differences in outcomes for mother and infant will shed little light on the birth process. Given the many differences among the settings, it will be difficult to ascertain which variables are responsible for any differences in outcome. By including careful assessments of background and setting variables in studies of alternative birth settings and by examining a large num- ber of such studies, it may be possible to make some preliminary judg- ments about the background and setting variables that are most impor- tant. Ideally, research in which various settings are compared should be paralleled by studies in which just one setting variable is manipu- lated while others are held constant. It is much easier to examine the effects of individual characteristics of the setting than the effects of the birth environment as a whole because discrete parameters of the setting (e.g., personnel, practices, clients, place) lend themselves more readily to randomized experimental designs. One might be skeptical about the use of randomized clinical trials in research on the birth process. However, there are a large number of studies that have effectively employed such sophisticated designs. In previous studies, randomized clinical trials have been conducted to examine the effect of such variables as the position of the mother dur- ing delivery (Humphrey et al., 1973)J the presence of a supportive lay person during delivery (Sosa et al., 1980)J whether electronic fetal monitoring was used (Kelso et al., 1978J Renou et al., 1976)J whether the Leboyer approach was used (Nelson et al., 1980)J whether the mother received extra contact with the infant (Kennell et al., 1974J Ringler et al., 1975) or was allowed •rooming in,• a situation in which the baby stays with the mother the entire time (Greenberg et al., 1973)J whether initial contact occurred immediately postpartum or was delayed 12 hours (Hales et al., 1977)J and whether the initial contact was made with a wrapped infant or skin-to-skin (Curry, 1979). The underlying processes are more likely to be elucidated by knowledge regarding the impact of specific variables on birth outcomes than by comparisons of birth settings. Moreover, such knowledge may be extremely useful in modifying birth environments to improve outcomes for the mother, child, and family. Prospective, Longitudinal Research Prospective, longitudinal studies are highly desirable in research on the birth process. If psychological variables are assessed only once and then found to be associated with outcome variables, the direction of causality is impossible to determine. For example, how should a positive association between maternal anxiety and labor difficulties be interpreted? Just as maternal anxiety may result in protracted labor, labor difficulties may enchance maternal anxiety. By assessing anxiety prior to labor, it is possible to draw inferences about the direction of casuality among the variables.

OCR for page 102
111 ~ny relationships among variables could be illuminated by using relatively short-term time lags between assessments (Walters and Walters, 1980). During labor, for example, women who do not receive support may be more likely to experience pain and express their discom- fort. Alternatively, those women who express pain and discomfort may receive different treatment from medical personnel than women who appear to be coping well. Assessments of women's emotional reactions and the supportive behaviors of health care providers at several points during labor should make it possible to determine the causal relationships among these variables. Long-Term Follow-Up Many advocates of home or other nonconventional birth settings have maintained that settings can influence such long-term outcomes as the child's emotional development or the relationship of the child to sib- lings. However, the evidence for these assertions consists almost exclusively of anecdotal evidence and case-study reports. As Cohen (1981) has noted, •There are indeed few, if any, long-range studies that support any claims at all. The time bas come for behavioral scientists [to explore the] childbirth experience as [it] relates to the development of the child.• Long-term assessment of psychological variables can determine whether outcomes initially appearing desirable prove to be detrimental in the long run. FOr example, mothers randomly assigned to a •rooming in• condition judged themselves as more competent in the infant's care and were also less likely to think they would need help with child care at home than mothers who were not provided a rooming-in option (Green- berg et al., 1973). Although the authors concluded that the impact of rooming in was positive, it would be interesting to know bow these mothers reacted to the full-time demands of child care once they re- turned home. Rooming-in mothers may have been less likely to arrange for help during the postpartum period and may subsequently have become more fatigued or experienced more strain in adopting the maternal role. Similarly, mothers who were given a few hours of extra contact with their child during the postpartum period were more likely to stand near their child during a physical examination or soothe the child if be or she cried (Kennell et al., 1974). These mothers also seemed much more preoccupied with their babies than were the mothers in the regular- contact condition. Extra-contact mothers were more likely to indicate that they thought constantly about the baby when they went out than mothers in the regular-contact condition. In fact, of those who had returned to school or work, five of the six extra-contact mothers (as compared to one of six control mothers) reported that they worried about or greatly missed their baby while away. Although the effects of extra contact appear to be beneficial in the short run, extended contact may intensify role conflict or distress over a longer sum period, especially among women who return to work. Given the considerable expense involved in long-term follow-up studies, one might ask whether the benefits of such research are likely

OCR for page 102
112 to justify the costs. Extended follow-up studies are rare. sa.e have provided data indicating that significant effects fro. birth-associated variables were still apparent years after the birth (Kennell et al., 1974). Several investigators have found differences in aaternal behav- ior toward their child two years later as a result of extra contact at birth (Ringler et al., 1975). Similarly, protracted labor, which aay be more likely to occur in some birth settings tban in others, is asso- ciated with differences in speech, language, and I.Q. as long as four years after the birth (Friedman et al., 1977). One study showed a sig- nificant relationship between a mother's attitude toward her baby at one month (a variable that could presumably be influenced by the type of birth experience) and the child's behavior more than four years later (Broussard and Hartner, 1971). In this study babies judged by their mothers as worse than average at one month of age were significantly more likely to require •therapeutic intervention• as deterained by an independent clinical assessment at age 4.5. Although it will be diffi- cult and expensive to conduct saae of these long-tera studies, their inforaation will be critical to uncovering psychosocial differences due to aspects of different birth settings. Multiple Methods of Assess..nt In assessing psychological variables such as the mother's emotional state during childbirth or the social support available to her, it is extremely important to use multiple aethods of assessment. As other investigators have noted, any one means of assessing a construct is necessarily i~rfect (campbell and Piske, 1959). Por ex..ple, if nurses or doctors are asked to assess a wo.an•s .-otional state during labor, their role as providers aay aake it objectively difficult for them (Standley and Nicholson, 1980). Similarly, data taken fraa aedical records uy be incaaplete and inaccurate. By . .aauring a given con- struct in several different ways, however, an investigator can increase the likelihood of demonstrating its validity. In assessing reactions to pain, for ex.-ple, the investigator aay get a more caaplete picture by exaaining a caabination of self-report measures (e.g., subjective distress or anxiety), attendants• observations of the client's behavior, (e.g., observers• judgaents of the wo.an•s distress), and physiological indicators (e.g., frontalis auscle tension and breathing irregular- ities). Multivariate Data Analysis When assessing a large nuaber of variables, a aultivariate approach should be considered. In a recent study on the effects of extended contact, 35 different mother-infant interactions were recorded 36 hours after birth (de Chateau, 1976). At a 6-.onth follow-up, 61 behaviors were scored during a mother-infant play session. Three statistically significant effects were found at 36 hours and 4 at the 6-.onth assess- ment. Given the large nuaber of analyses conducted, it is iiiPOrtant

OCR for page 102
TABLE 2 Continued Otbac POMible RefacaDCa ~at .._ of Infoc..tion on AaM-nt DiecuaMa 'ftlia A8Ma-nt ~ of leUabiUty and COnceptual Variable Parloda Vadabla 'I'Ool 'u a~) Inatc-nt Cbaractariatic.VDaecciption Validity llhetbac 110tbar baa Dud~ labor/ opportunity for caapariaon daUva~, with otbar 110tbara and aubaaquant babiaa, and vbatbar aucb poatpart• cc.parlaona are !Mda Batiafaction with infant'• Bubaaquent Doari~ and •aeaction to Motbara vaca aaka4 what the Reliability not appearance poatpartua Bntwiale, lt75 tba baby• baby looked like and r~ reported aponMa 1••9•• •u9ly, with hair aticki~ up all over•) -vera acored by intarvievera. ('l'ba quaation uaa4 in the pilot study, •how did you fHl about tba baby• tan4ed .... w to produce only aocially 01) deairable ceponMa) Avaranaaa of diatiDCtiva Bubaaquant feature• of the newborn postpartum Plaaaura or enj~nt Subsequent of contact witb tba infant postpartum latiafaction witb MX Subsequent of infant postpartum Reaction to infant'• Subsequent c~i~ postpartwa Willi~naaa to let otbara Subsequent care for tba infant postpartua 8UUIIQUDI'l' TO !'U POI'!PAJmJIC Plla.IOD Maternal babavioc ltannall at al., Obaarvar 'ftle nu.bar of behaviors 70' of interobMrvac lilt rati~a, uai~ rated waa not indicated. rati~a vera 9reatar a cbackliat, Such babaviora aa raactiona than .85r 91' vaca of 110tbar'a to infant'• behavior vaca 9caatac than .eo location an4 racocda4. babavioc avery 15 MOOn4a durin9 pbyaical axaa of infant 1 year aftac birth an4 durin9 fra~play period follov- lft9 the exa•

OCR for page 102
Maternal behavior Klaua at al., 1172 Salt-reported as apeoific aotivltiaa vera Rapor~ rallabllity c- Curry, 1111, reaction to in- racorcled, auob as care- ooafficianta ara or Kontoe, 1178, fant's cryi119. taking skills, fondling, bi9bar than .ao for aiailu 8alf-r~rted ancl cudcllin9 obaervation raaponaa to . . .auras) 90i119 out ainoa tha infant's birth, Obser- vation of aothar durin9 axaaination of infant1 obsarftr rati119 ...Sa on 3-point acala. Observer rat1119• of tiae- lapsad Ulaa of aothara feeding tbeir infants (15 ainutaa wara Ulaad, and aaob f r - of tha first 600 wara rated) Maternal acceptance Cbaabarlain, 1976 'ftla Dubee- Q-sort s• 'ftla o-sort consists of Mona raportacl Miohaal Child atat...nta about child's Behavior Q-Sort1 babavior. Motbar sorts tba• Darbeaand into 11 pilaa aooorcli119 to Micbaal, 1170 tha way sba psrcaivea bar .... w cbilcl to behave, ancl bow \C aha voulcl ideally lika bar child to babava, A corre- lation batwean tha t - ia tben oalculatacl Motbar-to-intant apeecb Just after Ri119lar at al., Obsarven 'ftla saquanoa of uttaranoaa Mona raportacl birth 1975 analysad taped obtained in aach apaacb conversations aaapla vas claaaifiacl of aothar and aocorclin9 to a nuabar of child duri119 a atanclarcl linguistic criteria •fraa play• C•·9·• rata1 la119th1 variety aituation. of uttaranoaa1 vr....tical Tranaoriptiona structural for• claaa1 and vera divided type of aantanca, such •• into 3-ainuta question or ~nd) intervals which wra further divided into sequential units of aothar and cbilcl apaec:b c.

OCR for page 102
TABLE 2 Continued Otberr PoAible ---·-lit . .ferreDae '!bat Diacu...a ftia ..__lit .._of 'l'ype of Infomation on leliability ancl Coaoeptual Vuiable Perrioda Varriable 'l'aol (if any) Inatrr-nt Cbarr.cterriatic.tDaacrription ValicJity c-it.ent to the infant Juat aftarr birrtb ConficJence in ability to Juat aftarr Leicleman et al., Self-rreporrt IIOtbarr aakecl to COIIiperre !lone .otberr tbe infant birrth 1173 herrMlf with 5 otherr carre- takerra (e.g,, fatherr, grrancJ- -therr, expedencecl ~herr, pecliatrrio IIIIJrM, ancJ clootOl') on e.ch of 6 carretaking taaka (e.g., calaill9 orr feecling tbe baby), Perroent- -v• of inata-• in llbich . ... 0 abe Uata herrMlf aa - • t able carretakerr ia notacJ Carretaking akill J.aat afterr bil'tb llbetberr the infant 1a Juat afterr prrowiclecJ witb ati•lation bil'tb Prreocoupation with infant Juat afterr (e.g,, extant to which birrth tboughta ancJ -.rrna arre clcainatacl by infantr willing- neaa to go out an laave tbe infant) j l I

OCR for page 102
141 ~ ... ... .sJ11 !....e :I ... .... ,.... .. ... .::• . ::r .. Jl c • . .. ."l . !• . ! I I .s !. i a ... J .., :: .. i '8 • ... . . 1:: 0 - J I i i . . ! 0 -;: J .., 1: c: !'. 0 )5 .g • ... • ..! I = 1"4 • ~ u • ! .8 ~ j.. • : I

OCR for page 102
TABLE 4 Psychological Research on the Father Otbar Po••ible Reference 'l'bat • - of Infor-tion on A8H8-nt Diec:u•H• 'l'hi• AaH•-nt 'l'ypa of Reliability and Conceptual Variable Period• Variable 'I'Ool ( if any) In•tr-nt Characteri•tic•/Daec:ription Validity PJIIOR 'lO PRBGIWICY !i: N Background and panonality variable• (it. .i•ed in 'fable 1) DURING PRIIGIIAIICY Pather'• attitude•, Ju•t after Wapner, lt76 Self-report 'l'he it. .a were divided No interit. . relia- fHling•, and behavior• birthr inventory with into Hparats ec:ale•• bility coefficients •ubaequent 63 it.... . . _ reported for the poatpartu• it. .a (e.g., • Pear• and attitude• Hparate ec:ale• tho•• on about fatherhood husband'• • PHling8 and attituda8 physical about tba pregnancy IIJIII)tc.s and • PHUng• and attitudes participation about the . .rital rela- in childbirth tionshipC•• influenced ClaSH8) by the preg~) • PHlings and attitudes about saxual/phy•ical aapacta of the Mrital relationship C•• in- fluenced by the prer nancy) • Behavioral inwl-nt in the pregnancy, • Incidence of husband' • physical ey.ptc.a ra- latacl to pravnancy

OCR for page 102
143 ..... .!.. .... H ... ... ?I' ... ... .a .... •t >o ....I!:'I! ... .... ... • ... .. • . 8 .. 1 ·J!-.. j:1 .... .... ... . ... . ... .... ~s 11 .. ~1 .. t 1,:: ...... . c . co :"' ·-• r::: ... >• .: .... ~ 0 •I • ... ec ! ....... j .. ;~ao: 0 ... """' ....... !- ... . . ! ... .a·~ . . . Cl. • c ....... .. , .. ' !i -r 1'"' ... 1 J'i! 11§::: • .I ......... 'J·ej::J II ..: ... ·~ • . ,t. II ... .. ,a .:• 5"' ~- .... ... ...... ~: o.o 0 ... • .. 1 :i. , .:: •.:.e•-o-8 • .: ::,., a ... "rc•o.o ~.::~ ... :: ... ..a. ...r .....r ·t• ....e. ~ . =...... ~ . ....... ~ ;.~:=:n:Eal : .. """ ...... .... ...l::i.S.a ~i! 2 i .I .I • ~- ~~ r ... 0 f!i-; ..• .:. ... II: .. ~ ..·~ g!. . ..... • I . . .... .... .... , !~j i! ..... 1!" ..... ••a ; .... icc 8 ... 8. .... J. ::.:.; ....:: .c ...... .. t-1::: .. e .... J :::51 tB:: .a .. I 1 . ! ... .... ...... o ... J • .. ~!1!· .. Ba•-e. II ~J J:: f! ~" ...... il .. ! . •g ,... Q ::o.aa... - ... ..... ,.,. ... ! . . !r ..; .• ... ... • .I .... 1.. ... ... - ... a Is! •• : ji~ ~~] !a.- .. ; ... • ..• ..• •! ...=: o:l.a • 0 :t - ";;-'"5 ... "::' ... 5 ... ::I - 11! 1 J ~.:: a B J ~.:: a B .. ......... 8:.• ~ :.. I.... .. .. .. .at .. ·=.af !=! ~-=·· i!~! i ....:: ..:.a . •. ... : ow .a ... ... ~8·3 • .a .. •1! ... .a .... 8~ .s •c• .. -" D 0 D i c"'i" ... 0 ... .~~i ... " 1:J: .. g I .. :1 ";;- ~ ~g 1 .. u-e .. . 41 • ... I::!!'..... • • ... 0 . J ..... :::::5:: ...... •I~J .. 0 ....... i •• 8 ... ... ' ... ... ! ~s:: j' t ..... J !:if.t:~ 5 ~'i!" ~~i • . a .. •• 8 ... ... . •a ... 1 ..... "• :a•-:: 1 • .... 2 ~~ . • s ... R~r • 0 il 8~ J=l~ ' "'!' ... 0 ..g.• .. . 1: .. H"' ~~.. I , ...... •1 ::.!:.:~uJ1 ••se .... iJ .... ... o .. "0!: ;:: ••• II "8 .......... I f ...... o,, .... ~ .. -~ ... .II . • eUI . :: ... ...... ... . •" .,.CI !I. Q •• . "'::f! >of u j i.B i

OCR for page 102
TABLE 5 Psychological Research on the Father and Infant Otbe~ Pouible lef•~•- 2bat .... of Inforution on Aaaae-nt Diecua"• ftie Aaaae-nt type of leUabiUty and Co~tual Vuiabla ••doda Vadable \'ool (if any) Inet~-nt Cba~ecta~ietice/Daec~iption Validity JUn AftBil 81111'11 Patba~ 0 8 babaYiO~ towa~d 8ubeaquant llaDonald, 1978 Di~act obaar- 211• followinv baba•ion we~• Obae~. .n obtained infant poatpart- ntion of eoo~ad• bo'ladnv, p~oJ.onvad high agree..nt in Yidaotapae of 9aaing, •ieual contact, identifying theae be- ~ fatba~'• ba- baYio~ towa~d pointinv, fece-t~fece oon- tact, finga~tip oontact, and haviors (198), and moderately high agre e- • • infant. aa..n pawinv oontact ment in scoring each patunal ba- paternal behavior baYio~• we~• during each 3-aecond obae~Yad dudnv interval (9 of 21 tb~•• 3-.J.nute lendal w •aluee intanale in tbe ~P9ad f~011 .51 to Unt t poat- .84J t otba~• ~angad pa~t- •inutae f~OII • 71 to 1.0) Contact witb infant 8ubeaquant poetpa~t- Pla..ura in o~ enjo,.ant 8ubeaquant of CXIIItact witb infant poatparta 8atiefection with ... of 8ubeaquant infant poetpa~t­ 8atiefection witb in- SubMquant fant'• appea~­ poetpa~t­ a.a~anaee of dietincti.. SubMquant featu~•• of newbo~n poatpart- C•·9•, appauanca, cey) Patba~'• •nv~oe-nt witb Juetaft•~ aleba~, lt7t 'l'wo Mlf-~apo~t Pa thera were asked wbat wae tba infant bi~tb itau the "niceat thi nv• about tbe daya their wivee we~• in tba hoapital, and t he •nioeet thing about the fi~et few daya at hoae.• ftey we~• alao aaked why tbei~ ..~~i­ age ia happy. ttl• n.-e~ wbo epon~ely Mntioaad tbe baby in eMb ceae we~• noted

OCR for page 102
Patber'• inwol....nt in llallion, 1977 ...bi' • • 'l'ypical 8elf-report Contain• queationa on tba Autbor noted tbat faaa infant caretaklnv Day• fatber'a participatica in and oontant validity oaratakinv actlvltiaa durinv bad been eatabliabad an arbitrarily cboaan tt.a in a pilot atudy, but (e.g,, bow aany tt.aa in no dataila are given tba pravioua - k ba bad batbed, diapered, rocked, and fad tba babi' I Patber'a at~nt to Peter- at al., Obeener ratinva ObMrvera nota extant to Intarobaarver agrae- infant 1979 baaed on ob- wbicb father inter.ctad and Mnta of .85 ware ob- aervatlon of cared for infant, fatbar'a tained fatber'a be- oonfidence in oadnv for baYlor and baby, father'• fnUnva of reaponaea to cloaanaaa to tba babi', and interview fatbar'a tendency to inter- queatiODa .ct witb babi' in a . .y tbat ia plaaaurabla for botb(a,g., oauainv botb to laugb) . .... Ul

OCR for page 102
TABLE 6 P.ychological Reaearch on the Infant Otbell: Pouible . .f • r - ftat .... of Info~Uon on aaa.-nt Di-aftb Aaae-nt of 'l'ype . .liability and Conceptual Vadable Pedoda vuiable 'l'Ool ( U any) Inatr-nt Cbaracter1atic.toeacr1pt1on Validity JVft AftU llllft lleonatal bebaYiOII: lra•elton et al., lleonatal .... "-ata niUII:~ . .ba,ioul it_, t110 global All UHaii:Cbell:a and 19741 lra..lton, baYioul Ar logic adequacy 418ena1ona (attractinneaa cliniciana wbo plan to 1!176, 19781 - - n t aoalea vitb 20 uflex and need for atgglatlon) u.. tbe acale are a-roff, 1978 ara..lton, •aauree an4 ue 4edYe4 fr- an4 are urged to 1rielt one of 1973 bebaYioral rated on 4-point KAl... thll:.. training centere re~H• to 'ftle ind1Y14ual' • interaction for a 2-day reli- lnYir-ntal repertoire 1• tben aaHaaed ability Haalon. atggU. 'l'akea on 27 apeciflc beba1rioral lra•elton (1978) 20 ainutea to it. .• tbet are belieYed to report• tbat tbia pufom, 10 reflect 4 bebaYioral ayatea produce • ainutea to df.8ena1onaa (l) interact1Ye acceptable and blgb - u reliabi- capacltiea (ability to reliability, vbicb lity. Deaigne4 attend to and proce•• en- can be Mintained for infanta in •111:-ntal ennta) 1 (2) - for 2 year• vitb- .... tbe Urat aontb aftu bill:tb1 inappropriate toric capacitiea (e.g., ability to control -tor bebaYior, aucb aa bdnging out ~:.....1nat1on Although interrater reliability is high , • 0\ for pr-t.are banda to -tb) 1 (3) 011:9an1- teet-retest reli- i11fanta less •ational cepacitiea vitb ability scores are tban 37-ka reapect to atate control moderate to low (se e 9eatat1on. (bow vall infant -illtai11a Saaerotf, 1978, for loOII:II tbe in- a cala, alut state 4eap1te further discussion) . fant•a •beat• incraaaed atgglation)l an4 Also, the scale perforMncl 011 141 011:9ani•ational capaci- appean to lack tbe df.8enaion tiea--pbyaiological reaponae pre41ct1YI Yalidity in queation to atr••• le•9•• bow aucb vitb reapect to infant ia able to 111b1b1t denlos-ental out- atartlea) - ( H I s-roff, 1978, for further diacu. .lon) lleonatal bebaYior llor-it•, et al,, lleonatal .... '1'0 enDCIIIIPil8• fte other four new acalea ..llabllitiea 011 19781 8ull1Yan, baYioul Ae- - o f tbe area all ecalea are 1977 H - n t Scale vitb biiNI obeenationa aany testers . Quality of infant'• bi9bu tban .90 llo41Ucationa (DU-It) exaa, but which generally go . make durinq the alert re~•1•1ty &xaalner'• peraietence 1•·9•• bow bard ea- unrecorded, ulnar baa to -rk to Sullivan de- elicit alertneaa froa veloped S new ~~eales. One of . illfant) General Irritability these, Orienta- t1011 tO lnani- . of i11fant . .l11for~nt •alue of Mtl Vbual infant (t.v., bOw aYir- and Aullitory ai" or rewarcUnv it ia lti.aU, ia dt- to interact vitb infant) \

OCR for page 102
ai9ned to par- allel tbe IIMI acalt on orien- tation to ani- •t•otbtr tbe att.I!Ur -1•• are oo~ pletaly - · Sullivan ..Sa - ainor aod i- fioationa to tbe ori9inal -1•• and in •tbod • for acol'in9 tbta aa . .11 Baby'• beba9lor atata . .leon at al., Obtel'ftr . .unga of bebaviou ranging Mean interobHrver during tbe firat bour of life 1110 ratinga froa deep al. .p tbrQU9b a9r-nt wu .18 quiet alert activity to irritable cryift9. Available tbrQU9b RAPS (Docuatnt 103588), P.O. Boll 3513, Grand Central Station, . . . York, •Y 10017 Baby'• aucking behavior Kl'on at al., IOttle waa attached to in- 1166 atr-nta tbat ....ure tbt rate of feedift9, preaaure of aucking, and - t • Qualitiea of infant'• lub8equent leakind and Deacl'ibea bow 0D tbe Hlf-report inatl'u- Pearaon oouela- cry po11tputua Leatar, 1171 erie• can be oate90r bed atnt, cdea can bt rated eooording to bow 9rating, tiona aaong ratift98 of tbe ei9bt different .... ~ eooording to objective "aick," ur9ent, diatreaaift9, piercift9, diacoaforting, cry qualitiea ranted froa .73. to .91 .... j~nta of avenive, and arouain9 they pitch, latency, are. A factor anelyaia of etc., a• ..11 tbeae rat1n9• r.,.aled two at factor• •jor dt.tnaionaa a tbat elicit tbe "diacoafort" factor and a cry and per- MOOne! factor oonveyin9 tbe cantata of tt.t "aick" nature of tbe cry apent cryin9. Alae deacdbea a Hlf-report inatr-nt for uHaaing aub- jactive re- action• to tbe cry lleelnatal bebavioral Grabaa, et al., !'be GrabuV Obtel'ftr !'be ecale oontaina Hveral Moat intaraoore reli- deficit 1956, IIGHnblitb IIGHnblitb ~ ratingaof aubecalea, aucb aa-auacle ability cotfficitnta 1961J IIGHnblitb bavioral Bxui- infant'• be- tenaion, viaion, and are bi9btr than .ao et al., in pr••• nation for wew- havior •turation borna

OCR for page 102
TABLE 6 Continued Otber Po .. ible Reference That 11a. . of Infor.. tion on AaHa-nt Dillcu..ea 'l'llia AaHa-nt of '1'ype Reliability and Conceptual Variable Perioda Variable Tool (if any) Inatru.ent Characteriatica/Deacription Validity SUBSIIQUBIIT TO TIIB POS'l'PAR'l'UM PERIOD Infant t.-per...nt Juat after carey, 1970 Self-report in- Iteas caapriae 9 acalea, in- Several aothera were birth ventory de- cluding activity, rhytha- interviewed and coa- aigned for icity, adaptability, pleted queation- babiaa 4-8 approach, threahold, inten- nairea, the author 110ntha old can aity, IIOod, diatractibility report• the reaulta be cc.pleted by and peraiatence. Queationa were in agreeaent. the 110ther in focua on apecific behavior• Author reported approociaately of infant (e.g., when high teat-reteat 20 ainutea and already full, hov doea reliability (ape- acored in leaa infant reapond to feeding cific coefficient tban 10. The atteapta not provided) Infant developaant Bayley, 1969 Bayley Scalea inventory haa 70 atat-nta, each with 3 choice a Deaigned for The ecale contain• 3 partaa (1) Split-half reli- . ... CD of Infant 2-30 aonth (1) a Mental scale, ability coefficient• Developaant childrenr deaigned to aaHaa Hnaory- ranged froa .81 to tha teat ia perceptual acuitiea and .93 Teat-reteat reli- adainiatered diecriainationa, aeaory, ability waa 76.4, by an e:uainer learning, and problea- interobaerver agrea- and takea aolving abilityr vocaliza- waa 89.4. approxi..tely tionar early evidence of (2) Split-half reli- 45 ainutea to ability to fora generali- ability coefficient• to cc.plete aationar (2) A Motor Scale, ranged froa .68 to daaigned to aeaaure degree .92. Teat-reteat of control of the body and reliability 75.3, coordinationr and (3) An inter-obHrver agree- Infant Behavioral Reco~ .. nt waa 93.4 focuaing on the child'• (3) Not reported aocial and objective orientation• toward hia/her environaent aa expreaaed in intereata, .-otiona, energy activity, and tendenciea to approach or withdraw froa atiaulation Failure to thrive