Recently, my wife and I had the pleasure of having dinner with Lewis Mehl-Madrona, MD, the author of "Coyote Medicine" and most recently "Narrative Medicine". During the course of the evening we had the opportunity of speaking with him about researched he conducted with patients on prenatal hypnosis and how it affected birth outcomes. I wanted to share the paper with you. Information is listed if you would like the study in its entirety.
Address communication and reprint requests to Dr. Mehl at the Center for Complementary Medicine, Shadyside Hospital, University of Pittsburgh Medical Center, 5230 Centre Ave., SON Bldg., Rm 216, Pittsburgh, PA 15232. Voice: 412-623-1365, fax: 412-623-1029. E-mail: madronalm@ssh.edu and/or mmadrona@aol.com.
ABSTRACT
Background: Prior research by the author showed that psychosocial factors distinguished abnormal from normal birth outcome. The purpose of this study was to determine if prenatal hypnosis could facilitate normal birth.
Methods:
Results: The use of hypnotherapy significantly inhibited negative emotional factors from being related to abnormal birth outcome.
Conclusions: Attention to reducing the impact of adverse psychosocial risk factors through the prenatal use of hypnosis did improve outcome.
INTRODUCTION
In these days of health care reform, an important opportunity exists to explore the utility of psychosocial interventions in improving birth outcome. These interventions are often less expensive than medical procedures and can be provided by non-physicians. One of these interventions is hypnosis.
Labor length and analgesic use have been reported to be decreased when hypnosis is done during pregnancy. [1] Anxiety about and during birth is decreased. [2] The incidence of postpartum depression is lessened.[3], [4] Increased self-confidence, increased calmness during labor, and easier transition into breast feeding has been described.[5]Decreased pain sensation has been reported.[6], [7], [8], [9], [10]Reductions have occurred in the number of complicated births.[11]Babies born have had higher Apgar scores.[12] Hypnosis subjects experienced greater belonging and security during labor, were less afraid of birth, and perceived birth more as a positive event.[13]Hypnosis helped women to be more relaxed during labor and birth, reduced hyperventilation, and increased feelings of participation and mastery.[14] A psychosocial intervention program that included hypnosis reduced the number of cesarean deliveries and oxytocin augmentations or inductions.[15]
Only one published report showed no improvement over Lamaze technique from the addition of hypnosis, though both LaMaze and hypnosis alone lessened pain equally during labor.[16]
Hypnosis has been reported helpful in the conversion of the breech presentation to vertex[17] and in the treatment of premature labor.[18]
The purpose of this research was to address the question of whether high stress and low social support contributed to birth complications and to determine if hypnosis would protect the high stress-low social support woman from developing birth complications.METHODS
Settings and sources of subjects: The author practiced holistic medicine in San Francisco, California, and, later, Tucson, Arizona. As part of this practice, he provided prenatal hypnosis. Two hundred sixty subjects in the first or second trimester of pregnancy were referred for prenatal hypnosis and included in this study. These patients paid for their visits, or their insurance was billed. Data obtained from the was initially obtained for clinical purposes. Referrals came from family phyicians, obstetricians, naturopathic physicians, and midwives practicing in these areas. Patients seen for their first appointment during the third trimester of pregnancy were not included in this study because the author feared that insufficient time would exist to conduct a thorough evaluation and to provide sufficient treatment to alter birth outcome. (Challenging this belief should be the topic of another study.)
During that same time, the author recruited subjects from these same groups of physicians and midwives for a study of psychosocial factors in the prediction of obstetrical risk. These subjects were told that information was needed from them to study how best to predict risk in childbirth. Subjects were paid for completion of questionnaires and for being interviewed. Informed consent was obtained and the study was approved by the author’s Institutional Review Board. Matched, comparison subjects were obtained from this cohort of subjects, who were not referred for hypnosis. Bias was expected to be against the hypnosis subjects, since 1) they came from the same groups of health care providers, and 2), their providers had not identified them as having special needs which could be addressed by hypnosis. There were approximately 2000 subjects from which to match for the comparison group.
Assessment: A complete medical and psychosocial history was obtained from all subjects, including: (1) Demographic information, (2) Complete family medical history, (3) Ob/gyn health history, (4) Psychosocial history, (5) Past medical history, and (6) Review of current symptoms.
The Holmes-Rahe Life Stress Inventory was administered, along with the Taylor Manifest Anxiety Scale, the Dyadic Adjustment Scale, and the Beck Depression Inventory. The couple was interviewed whenever possible. The woman was assessed for her relationship to her body, awareness of body symptoms and patterns of body tension. The scores on the Taylor Manifest Anxiety Scale were normalized to a 0 to 1 scale, as the scores on the Dyadic Adjustment Scale. That scale was oriented so that higher scores meant greater marital satisfaction. The Holmes-Rahe Life Stress Inventory was assessed over the preceding 3 years.
Couples were interviewed whenever possible and during that interview, information was solicited from which an assessment of the couple's beliefs, experiences, expectations and affective states was made. Assessment was also made of the women's stressors, fears and social support. Interview formats and questions are present as Appendix A.
Matching: Subjects in the comparison group were matched with subjects in the hypnosis group so that their values lay within the same intervals defined below:
a) Age: 15-19, 20-24, 25-29, 30-34, 35-39, 40+
b) Socioeconomic status: No insurance, Medicaid, Commercial insurance or HMO, highest level private commercial insurance (ratings of insurance as average or high level was made by the author’s hospital’s billing department independent of this study).
c) Obstetrical risk (defined by the POPRAS system from Harbor General -- UCLA): Low, medium, high.
d) Marital status: Unmarried, married, separated, divorced (at the time of initial interview).
e) Education: Less than 12 years, high school graduate, junior college or college courses, college graduate, graduate courses, graduate degree.
f) Parity: 0, 1, 2, 3, 4+.
Evaluation: Initial interviews aimed to establish close rapport with each subject, so that feelings, fears and complaints could be freely expressed. Interviews usually lasted two hours. Information about the woman's past and current reactions toward herself, her family, partner, work, social, religious and physical experience was elicited. Her knowledge of the physiology of pregnancy and birth, her menstrual experience, family patterns of pregnancy and birth, changes in sexual relationships, attitudes toward body changes, the baby, nursing, and general experiences of pregnancy and birth were assessed.
Note was made of shifts in attitudes and reactions indicated by changes in tempo and intensity of verbalizations, slips of the tongue, innuendo, facial expression, vasomotor activity and tone of voice. Physiological monitoring was included when possible and changes in blood pressure, heart rate, skin conductance, skin temperature and muscle tension were noted when associated with specific subject matter.
The interest or capacity of each woman for participation varied. Some had but little experience thinking about themselves or that which had happened to them and expressed themselves poorly. Some subjects who verbalized poorly gave crucial, pertinent material once encouraged to talk about themselves. Others were defensive and produced scanty material.
These interviews represented different experiences to different patients. To most, they gave a much needed feeling of being an individual in whom others were interested. To a few, the interviews were simply an opportunity to be a part of a study, and to a couple of women, the interviews provided a serious threat. Most of the patients, however, soon came to regard the interview as a helpful experience--an opportunity to talk about anxieties and problems.
Observations of the prenatal care provider were elicited by telephone interview or a data form sent in the mail. Obstetric data and all physical examination findings during the course of pregnancy and childbirth were abstracted from prenatal care records, requested after delivery (the woman signed a records release form during the initial interview). Every effort was made to learn as much as possible about the patients in terms of their psychological functioning, cultural background and life experiences.
Interview records were examined in accordance with Glaser’s method of grounded theory.[19] In this method, qualitative data is examined with an eye toward data reduction. All possible categories which make sense clinically and theoretically are applied to the data. Categories are tabulated and reviewed. Categories are collapsed and combined when possible and logical to obtain a limited and manageable number of variables. A continued coding, sorting, and evaluating process eventually results in a data reduction scheme which makes sense and represents what is available in the data.
For example, the statement, "I am afraid of pain in childbirth", was coded as a fear response. Initially it was called “personal fear of birth, then fear of birth, and finally, fear, as categories were combined. By taking the verbal statements, reviewing any descriptions of associated affective expression, a statement could be made about the intensity of the fear. Statements made were rated on a +3 to -3 scale. "I am afraid of pain in childbirth" became a 'Fear' statement, with a numeral rating assigned to represent its intensity. Appendix B lists the final coding format that developed through the grounded theory process.
The final seven categories to arise from the coding process included:
(1) Fear,
(2) Anxiety-stress,
(3) Maternal self-identity
(4) Beliefs,
(5) Psychosocial support from the partner,
(6) Psychosocial support from the mother's mother,
(7) Psychosocial support from friends.
A team of three clinicians assigned responses to the appropriate category and rated the response for intensity . The frequency of occurrence of a specific response and the magnitude or intensity of the responses were sufficient for clinicians to grade responses from -3 to +3 according to the strength of the statement. Comparative adverbs of 'very', 'mildly', etc., were included as indicators of the magnitude of the psychological state. The verbal responses of the women were differentially weighted in the specific content categories in proportion to the assumed intensity represented by statements made and interviewer notes. Values were assigned to all the verbal responses made. One type of direct verbal report of the subjective affective experience, such as, "I am anxious", would be classified in the 'anxiety-stress' category, and have a weighted value of a -2, while the same statement with a greater intensity, " I am very anxious", could be weighted -3. Each of the women's responses were assessed with the value weighted on each variable to develop a profile of her psychological attitude during pregnancy. The sign of the rating (+ or -) was in accordance with the hypothesized relationships of how this factor would affect the birthing process. The descriptors provided were those which worked for the raters to achieve over 85% agreement. They were developed through rating patients together prior to beginning the study. If the raters could not agree through consensus, the average of their ratings was taken and rounded. The inter-rater reliability was checked on every fifth patient and remained above 0.85.
Treatment: If hypnosis were successful it would be expected to prevent patients with many adverse factors from having abnormal birth complications. If hypnosis was not successful it would be expected to have no impact on outcome.
The general approach used for prenatal hypnosis was oriented toward problem-solving and was perceived as brief, not as insight-oriented nor psychoanalytically-based psychotherapy. Goals included increased relaxation, decreased anxiety, increased sense of trust of social support, realistic fear and a feeling of confidence that the woman could cope adequately with the pain of labor. Visualization was used to guide the woman through an imaginary experience of giving birth, thereby decreasing fear and anxiety. Careful notes were made of the interviews. Audio and videotapes of representative hypnosis sessions are available upon request.
The mean number of prenatal interviews was 5. The mode for number of psychosocial sessions was three. The minimum number was one and the maximum, 60. All hypnosis was provided by the author. Subjects in the comparison group were seen for two-three hours total, usually twice.
Outcome variables: A normal birth was defined as one without obstetrical intervention (no Cesarean, no uterine dysfunction, no fetal distress, no low Apgar scores, no infant resuscitation required, etc). Uterine dysfunction was diagnosed when treated by the doctor or midwife with oxytocin augmentation during labor or with induction. Fetal distress was noted when it was recorded on the labor and delivery record. Apgar scores were recorded on the labor and delivery record. Infant resuscitation efforts were described on the labor and delivery record. All of these were obtained from the hospital records, or, for home birth, from the midwife’s birth records.
An abnormal birth, therefore, required the use of obstetrical technology, including Cesareans, induction and augmentation of labor with oxytocin, fetal distress resulting in intervention or fetal scalp sampling, low Apgar scores, and postpartum hemorrhage. An obstetrician and two certified nurse-midwives reviewed each case to assess normal versus abnormal. They were blind to the existence of this study. They agreed on 95% of cases. For the remaining cases, their consensus was accepted. These were borderline cases in which, for example, blood loss was on the borderline of excessive or fetal distress was on the borderline of being excessive.
Data analysis: The T-test procedure and the discriminant analysis procedures from the Systat statistical package for the MacIntosh computer was used. Statistics provided are already corrected using the Bonferoni method for the number of comparisons made. Chi-square tests were used to test statistical signficance of differences between groups. Variables were compared within the total sample of 520 women by actual outcome group. Then comparisons were reconsidered with subjects grouped by the use of hypnosis or not.
RESULTS
Table 1 compares demographics between women having normal births and abnormal births in this sample. No significant differences were expected between women in the hypnosis and the comparison group, since matching procedures were used. Table 1 shows no signficant differences, either, in demographic variables, when women were compared between for actual outcomes. Age of the total sample of patients varied from 18 years to 39 years with an average of 27 years. Years of education ranged from 9 to 19, with an average of 13 years. Fifty-two percent of the women were primigravidous, 24% were secundigravidous, 11% were having their third child, and 13% were pregnant with their fourth or greater child. No significant differences in level of medical risk were found between subjects who had abnormal outcomes versus normal outcomes. No differences were found in the range of distance from place of birth for women in normal and abnormal outcomes.
Contribution of Medical and demographic variables to risk: Table 2 shows no differences in the two groups for members having previous live births, previous abortions and previous miscarriages. Women in the abnormal birth group showed significantly more previous (to the pregnancy) infections, injuries and hospitalizations. These events were not obstetrical or gynecological and did not increase their risk on the Popras Obstetrical Risk Screening Criteria. There were no differences in number of prior surgeries or diagnosed illnesses. Women in the abnormal birth group showed more frequent past drug use. Neither group was using drugs during the pregnancy. Women in the normal birth group were more physically active.
Emotional state variables. Table 3 shows the differences in the emotional state variables. All four variables were significantly different between groups. Women in the abnormal birth group showed more anxiety-stress and fear. Their beliefs were more negative toward birth. They showed less maternal identity.
Table 4 compares the emotional state variables between the normal and abnormal birth outcome groups, grouped for the use or non-use of hypnosis. The comparison group who did not receive hypnotherapy showed significant differences between anxiety and stress when normal and abnormal birth outcome groups were compared. These differences disappeared when comparisons were made in the presence of hypnosis. The presence of hypnotherapy seemed to inhibit cases of high anxiety-stress who received hypnosis from from having abnormal birth outcomes.
Fear variable. The control group who did not receive hypnotherapy showed significant differences between fear when normal and abnormal birth outcome groups were compared. These differences were still present but to a lesser degree when comparisons were made in the presence of hypnosis. The presence of hypnotherapy appeared to prevent women with high fear who received hypnosis from from having abnormal birth outcomes, but not to the same extent as for anxiety-stress. A significant difference was still present between women having abnormal births and women having normal births, both of whom received hypnotherapy.
Maternal self-identity variable. The control group who did not receive hypnotherapy showed significant differences between maternal identity when normal and abnormal birth outcome groups were compared. These differences disappeared when comparisons were made in the presence of hypnosis. The presence of hypnotherapy seemed to inhibit cases of low maternal identity who received hypnosis from from having abnormal birth outcomes.
Belief variable. The control group who did not receive hypnotherapy showed that negative beliefs about birth were significantly associated with abnormal birth outcome. These differences disappeared when comparisons were made in the presence of hypnosis. The presence of hypnotherapy seemed to inhibit cases of women with negative beliefs about birth who received hypnosis from having abnormal birth outcomes.
Depression (BDI). Among the comparison group, increased Beck Depression scores were significantly associated with abnormal outcomes. Within the hypnosis group, this difference was not signficant, indicating that hypnosis protected against the effects of depression on birth outcome. The greater levels of depression in the normals in the hypnosis group further suggests that hypnosis had a protective effect.
Taylor Manifest Anxiety Scale. Among the control group, increased Taylor Manifest Anxiety Scale scores was significantly associated with abnormal outcomes. Within the hypnosis group, this difference was not significant, indicating that hypnosis protected against the effects of manifest anxiety on birth outcome. The greater levels of manifest anxiety in the normals in the hypnosis group further suggests that hypnosis had a protective effect.
Life Stress Inventory. Among the control group, increased Life Stress Inventory scores were significantly associated with abnormal outcomes. Within the hypnosis group, this difference was not signficant, indicating that hypnosis protected against the effects of high levels of life stress on birth outcome. The greater levels of life stress in the normals in the hypnosis group when compared to normals in the non-hypnosis group further suggests that hypnosis had a protective effect.
Social Support Variables. Table 5 shows differences between normal birth outcome women and abnormal birth outcome women on measures of social support. Higher levels of perceived support from the woman's partner (husband, boyfriend, etc.) was signficantly associated with normal birth outcome. Higher levels of perceived support from the woman's own mother was, intriguingly, significantly associated with abnormal birth outcome. Higher levels of support from the woman's friends was statistically significantly associated with normal birth outcome. Higher levels of marital satisfaction as measured by the Dyadic Adjustment Scale was statistically significantly associated with normal birth outcome.
Support from mother's mother. No statistically significant differences or effects were observed.
Support from friends variable. Lower levels of support from friends were associated with abnormal birth in the presence of hypnotherapy than in its absence.
Marital satistaction. For the control group who did not receive hypnotherapy, higher levels of marital satisfaction were associated with normal births, and low levels with abnormal birth. Statistically significantly lower levels of marital satisfaction were still associated with normal birth in the presence of hypnotherapy.
Birth data. In comparing the mean scores between the normal and abnormal birth outcome groups on birth data variables, first stage labor length, Apgar score at 1 minute, and Apgar score at 5 minutes, significant differences were found as would be expected. First stage labor length was shorter, and the Apgar scores at 1 and 5 minutes were better for the normal birth outcome group than for the abnormal birth outcome group (by definition). No significant differences between the mean scores for the normal and abnormal birth outcome groups were found for gestational age, second stage labor length and birthweight, indicating that premature labor did not figure a role in these differences. These results are summarized in Table 6.
Discriminant function analysis was used to correctly classify 91% of the cases correctly into the normal birth outcome group (group N) or the abnormal birth outcome group (group A). Women having normal deliveries were classified correctly with 90.2% accuracy, compared to 92.1% for women having normal births. The canonical correlation was 0.7808 meaning that about eight times out of ten, correct classification occurred. The most significant psychosocial factors were fear and support from the baby's father and drug use. Hypnosis was signficantly associated with normal outcome.
No one particular practice group was significantly associated with abnormal outcomes, suggesting that practitioner practice patterns were sufficiently similar not to contribute to outcome differences. Practitioners who were willing to refer patients for hypnosis or for a study on psychosocial factors affecting risk tended to be more supportive of natural childbirth and had lower intervention rates than their colleagues who would not participate in making such referrals.
DISCUSSION
Examination of the group of women who were evaluated and not offered hypnotherapy revealed that psychosocial variables were related to abnormal birth outcome. Seven major psychosocial variables showed importance to birth problems. These included maternal stress and anxiety, fear, negative beliefs about birth, negative maternal self-identity. Psychosocial support variables of significance included support from the baby's father, support from friends, and marital satisfaction/dissatisfaction as measured with the Dyadic Adjustment Scale. Depression as measured by the Beck Depression Inventory was asociated with abnormal birth as was manifest anxiety (Taylor Manifest Anxiety Scale), and Life Stress over the past three years (Holmes-Rahe Life Stress Inventory).
Providing hypnosis to women at psychosocial risk (as judged by the above criteria) did seem to help them have a normal birth.
Of surprise was the finding that high levels of support from the woman's mother was associated with abnormal birth outcome. This may indicate that the manner in which we rated this variable keyed more into dependency and passivity, or that women need a different kind of support from their mothers than we imagined in conceptualizing our rating system.
This study shows that a psychosocial intervention program can have a positive effect on birth outcome among women who are having full-term labors. Prenatal providers might do well to incorporate a consideration of psychosocial risk factors during prenatal care and to utilize methods such as hypnosis (or others that may prove effective) to help reduce psychosocial risk during routine prenatal care. This can be done cost-effectively, through utilization of a trained mental health professional. This practitioner can interact with nurses and childbirth educators to improve co-ordination of patient care and provide attention to psychosocial risk. Specifically, psychosocial risk reduction involves:
(1) Identification, acceptance and resolution of fears.
(2) Identification of states of high anxiety-tension, with helping the client learn more effective coping styles.
(3) Identification of negative beliefs about birth and parenting, with provisions for reframing and emotional relearning.
(4) Identification of low maternal identity with anticipatory guidance for the mothering role and hypnotherapy to improve self-esteem.
(5) Consideration of strengths and stresses of the woman's support system. Needed interventions included:
(a) Couple's therapy to decrease stress, increase husband's emotional availability, improve lifestyles, etc.
(b) Individual counseling for the husband to address his unique concerns (when indicated).
(c) Network therapy (environmental intervention with the mother's friends (which can be done in childbirth classes),
(d) Therapy with the mother and her mother to facilitate transition and change of that relationship, and/or,
(e) Helping the woman with no psychosocial support to establish needed relationships and resources.
On a broader level, these findings may presage a time when all medicine is practiced more holistically, with mind-body interventions seen as important. When the way medicine is practiced changes to reflect our new understanding of the interactive nature of all aspects of the patient's life on health and disease, we will have come far.
0 comments:
Post a Comment