Student Projects
Quantitative Research Proposal

Oral Evening Primrose Oil as a Cervical Ripening Agent in Low Risk Nulliparous Women

Paula Senner
Frontier School of Midwifery and Family Nursing
Philadelphia University, December 2003
Abstract
Evening primrose oil (EPO) is one of the most commonly prescribed non-pharmacologic cervical ripening agents used by midwives in the United States (McFarlin, Gibson, O'Rear, & Harman 1999). Cervical ripening is the softening, effacement, and dilation of the cervix that occur prior to active labor, and is an intervention that is used for certain indications, such as postdates pregnancy. Currently, multiple options exist for cervical ripening ranging from natural and herbal methods to more invasive medicines or procedures, such as prostaglandin preparations, membrane stripping, amniotomy, or mechanical means by Foley catheter balloon. Anecdotal data comprise much of the popular information available about herbs today. The literature review revealed that EPO has been studied for its efficacy and safety to treat a variety of conditions ranging from premenstrual syndrome, menopausal flushing, and preeclampsia to atopic dermatitis. EPO has been demonstrated to be safe, however studies show mixed results regarding its efficacy. The conceptual framework guiding this study is physiologic and biochemical in nature. Scientific studies of the efficacy and safety of herbs are less available than other drug trials because herbals are dietary supplements and are not regulated by the FDA (Cohen, Rousseau, & Robinson, 2000). Despite this lack of evidence, a growing number of childbearing women are looking for safe herbal alternatives for cervical ripening. A convenience sample of 250 women will be drawn from the pool of nulliparous clients at Loudoun Community Midwives. These women will then be randomly assigned to either the EPO group or the placebo control group, and both will take a prescribed regimen. The main outcome variable measured will be weekly Bishop scoring for cervical ripening scores until delivery. Additional outcome variables that will be investigated are gestational age at the time of delivery, incidence of postdates induction, amniotic fluid status, length of rupture, length of labor, type of delivery, incidence of abnormal labor patterns, infant weight, Apgar Scoring, and NICU admissions. This prospective, randomized, controlled clinical trial will examine the safety and efficacy of oral EPO as a cervical ripening agent in low risk nulliparous women with term pregnancies.

Chapter One
Introduction to the problem
Evening primrose is a biennial herb that grows wild in parts of North America and Europe (Salisbury Nursing School, 2003). Oil extracted from the seeds of the evening primrose contains linolenic acid, gamma linolenic acid, and vitamin E. Gamma linolenic acid is a known precursor of prostaglandin E and several other active substances, and is said to be the constituent of the oil responsible for its therapeutic effects (Kleijnen, 1994, Cohen, et al., 2000). Controlled clinical trials have been conducted for the use of evening primrose oil for a variety of disorders, including atopic dermatitis, rheumatoid arthritis, diabetic neuropathy, multiple sclerosis, many cancers, Raynaud’s phenomenon, ulcerative colitis, preeclampsia, premenstrual syndrome, menopausal flushing, breast cysts, mastalgia, Sjogren’s syndrome, schizophrenia, and hyperactivity (Kleijnen, 1994, Cohen et al. 2000). The results of these trials have been mixed. Many of the trials have been small.

Although evening primrose oil is one of the most common cervical ripening remedies recommended by midwives, it remains unclear whether this substance can ripen the cervix or induce labor (Tenore, 2003, McFarlin, et al., 1999). However, evening primrose oil appears to be safe, with reported mild side effects such as nausea, softening of the stools, and headaches (Kleijnen, 1994). One recent warning was issued with prolonged use greater than one year. These risks included inflammation, thrombosis, and immunosuppression (Kleijnen, 1994). In addition, optimal dosing and duration of treatment is unknown. In various trials, the daily dose has ranged from two to 16 500mg capsules, however, the most common dosing regimen appears to be three to six capsules twice daily (Kleijnen, 1994, Cohen, et al., 2000). According to Salisbury Nursing School (2003), pregnancy and lactation are listed as contraindications to its use, as well as a caution that the herb may reduce the seizure threshold in those taking phenothiazine drugs.

Multiple methods of cervical ripening have been proposed over the years, many of which have not been well researched. Current options in cervical ripening range from sexual intercourse, nipple stimulation, a variety of herbs, including evening primrose oil, and homeopathic solutions, castor oil, enemas, and acupuncture to more medical interventions, such as stripping of membranes, mechanical dilation, amniotomy, and pharmacologic preparations. Of the non-pharmacologic methods, only mechanical and surgical methods, such as foley catheter balloon, laminaria, membrane stripping, or amniotomy have demonstrated efficacy for cervical ripening or induction of labor, compared to waiting for natural ripening to occur (Tenore, 2003). Anecdotal data comprise much of the popular information available about herbs, and scientific studies of the efficacy and safety of herbs are less available than other drug trials (Cohen, Rousseau, & Robinson, 2000). Many herbal investigations have suffered from a lack of controlled clinical trials and lack of sufficient length of intervention to evaluate the effectiveness of the herbal treatments (Cohen, et al., 2000). This randomized clinical controlled trial will examine the safety and efficacy of oral evening primrose oil (EPO) as a cervical ripening agent in low-risk nulliparous women with term pregnancies.

Scope of the problem
Last year, there were 4,019,280 births in the United States (NCHS, 2003). More than a quarter of these births were by caesarean section (26.1%), which is the highest level ever reported in the United States (NCHS, 2003). Induction of labor has also been on a rising trend, with more than 13% of all births being induced in 1999 (Harman & Kim, 1999). Just two years later, more than one in five births were induced in the United States in 2001 (NCHS, 2003). According to Tenore (2003), the induction rate varies from 9.5 to 33.7 percent of all pregnancies annually. The most common reason for induction of labor is postdate pregnancy (Harman & Kim, 1999, Tenore, 2003). The incidence of postdates pregnancy remains steady at about 10%, which means that the induction rate is rising for other reasons (Harman & Kim, 1999). Postdates pregnancy is based on a 28-day cycle and occurs when pregnancy exceeds 42 weeks from the first day of the last menstrual cycle (Varney, 1997). Typically the drug of choice for labor induction is oxytocin when cervical examination reveals that the cervix is favorable (Harman & Kim, 1999, Tenore, 2003). Assessment for the need for cervical ripening prior to induction is accomplished through a Bishop score, which is calculated by assigning a score to cervical dilation, effacement, position, consistency, and fetal station. A Bishop score of less than six indicates the need for a ripening agent prior to induction, and a score that exceeds eight has been associated with successful induction leading to vaginal birth (Tenore, 2003).

When the cervix is found to be unfavorable, a variety of pharmacologic and non-pharmacologic cervical ripening agents can be used. Typically, research shows that the pharmacologic prostaglandin analogs enhance the success of labor induction (Harman & Kim, 1999). Cervical ripening, which is the softening, effacement, and dilation that occur prior to active labor, is mediated by multiple factors. Prostaglandin E2 and prostaglandin F2a appear to be the final mediators in this process (Harman & Kim, 1999). Therefore, although the exact mechanism is unknown, exogenously applied prostaglandins will induce cervical ripening (Harman & Kim, 1999). Prepidil and cervidil (PGE2/dinoprostone) are the most widely used agents and have been shown to be safe and efficacious in cervical ripening/labor induction (Harman & Kim, 1999). Misoprostol, a synthetic analog of PGE, has been associated with a significantly lower overall rate of cesarean section, a higher incidence of vaginal delivery within 24 hours of application, and a reduced need for oxytocin augmentation (Harman & Kim, 1999). The use of PGE2 is associated with uterine hyperstimulation, nausea, vomiting, diarrhea, and fever, and the use of misoprostol has been associated with tachysystole, hypertonus, and hyperstimulation, however maternal and neonatal outcomes were not statistically different between the misoprostil, dinoprostone, and oxytocin groups (Harman & Kim, 1999). Prepidil and cervidil cost between $100 and $175 per insert, whereas misoprostol costs $0.60 per tablet offering a significant cost savings (Harman & Kim, 1999).

Data remain inconclusive on castor oil, sexual intercourse, and breast stimulation as non-pharmacologic methods of cervical ripening (Tenore, 2003). However, these methods are inexpensive, can be used on an outpatient basis, and have shown no harm (Tenore, 2003). Mechanical methods share a similar mechanism of action of some form of local pressure that stimulates the release of prostaglandins. These options include laminaria cervical inserts and foley catheter balloon (Tenore, 2003). These methods require inpatient care and are more costly, however research indicates that they are safe, effective means of cervical ripening. Mechanical methods of cervical ripening have the potential risks of infection, bleeding, membrane rupture, and placental disruption (Tenore, 2003). Surgical methods include amniotomy and stripping of membranes, which come with the risks of infection, bleeding, accidental rupture of membranes, cord prolapse, and patient discomfort (Tenore, 2003). Current research does not support the use of either of these as effective methods for cervical ripening or induction of labor (Tenore, 2003).

Commonly prescribed herbal agents for cervical ripening include EPO, black haw, black and blue cohosh, and red raspberry leaves (Tenore, 2003). The risks and benefits of all of these agents remain unknown because the quality of evidence is based on a long tradition of use and anecdotal case reports (Tenore, 2003). The role of herbals in cervical ripening remains largely unstudied, however, if supported by controlled research, these products have the potential to offer an alternative to midwifery clientele. They are self administered, inexpensive, over the counter, and can be used on an outpatient basis. This inquiry will specifically look at the use of oral EPO as a safe and efficacious option for women in need of cervical ripening.

Significance to women’s health
Women constitute the largest proportion of health care consumers: they make 57% of the visits to doctors, they are prescribed 60% of all prescription medication, they undergo 59% of all hospital procedures, and they make 75% of the health care decisions in the family, while spending two out of three health care dollars in the US (Murphy, Kronenberg, & Wade, 1999). Two out of three of the leading surgeries in this country are performed on women: cesarean section and hysterectomy (Murphy, et al., 1999). Research shows that middle-aged, affluent women are becoming the largest consumers of complementary and alternative medicine in the US, with women of childbearing age not far behind (Murphy, et al., 1999). According to Belew (1999), from 1990 to 1997, the estimated use of at least one alternative modality increased from 38% to 42%, with the highest rates being attributed to women. Women perceive alternative and complementary therapies as more effective and having fewer side effects than pharmaceuticals (Belew, 1999).

Whatever the reasons that people are turning to alternatives to conventional medicine, they are frequently using remedies for which there is little or no scientific research on the product (Murphy, et al., 1999). As a result, complementary and alternative medicine is emerging as a significant area of scientific research. Women look to midwives for natural alternatives for cervical ripening should the need for induction of labor arise. In order to choose, women need information about the risks, benefits, and likely outcomes of the course that she has chosen. Women need to be able to choose from a variety of pharmacologic and non-pharmacologic options that are thoroughly researched and evidence-based.

Significance to midwifery/Philosophical Approach
Women are increasingly seeking out midwifery care in the United States for labor and childbirth. Certified nurse-midwives (CNMs) attended 297, 902 births in 2000, which accounted for 9.6% of the vaginal births that year (ACNM, 2002). Midwives have the expertise and ability to provide safe, effective care to the population while maintaining a low caesarean section rate of 11.6% in 2002, at a time when the national rate was 23.3% and an infant mortality rate less than half the national average (ACNM, 2002).

Women seek midwifery care because of the philosophy of care that provides for nonintervention in normal processes of pregnancy and birth (ACNM, 2003). In addition, individuals have the right to complete information in all aspects of their care. This means that midwives have an obligation to expand knowledge through research and to evaluate and revise care through quality assurance (ACNM, 2003). The ACNM Code of Ethics (2003) reiterates the responsibility of midwives to participate in developing and improving the care of women and families through research. According to Kennedy and Lowe (2001) “the central thread of our professional ethics, standards of practice, and core competencies is knowledge” that provides evidence for the practice of midwifery. Midwives need to weigh the lack of evidence relative to the potential benefits and risks of the vast array of herbs prior to advocating them in place of pharmaceuticals (Belew, 1999). Therefore, it is essential that further research be conducted on the use of oral EPO as a cervical ripening agent before it can take its place as a safe and efficacious option in the array of interventions for cervical ripening.

Key Concepts and Relationships
Evening Primrose Oil- Oil extracted from the seeds of the evening primrose contains linolenic acid, gamma linolenic acid, and vitamin E. Gamma linolenic acid is a known precursor of prostaglandin E and several other active substances (Kleijnen, 1994).

Term Pregnancy- Term pregnancy is identified as a pregnancy between 37 completed weeks gestation and 42 weeks completed weeks gestation from the first day of the last menstrual cycle (Cunningham et al. 2001).

Postdates Pregnancy- Postdates pregnancy is based on a 28-day cycle and occurs when pregnancy exceeds 42 weeks from the first day of the last menstrual cycle (Varney, 1997).

Cervical Ripening- Biochemically and physiologically, cervical ripening simply consists of the breakdown of cervical collagen bundles and the taking up of water by the cervix. Clinically, cervical ripening is manifested by the softening, effacement, anterior rotation, and dilatation of the cervix (Barrilleaux, et al., 2002) Cervical ripening is mediated by multiple factors. Prostaglandin E2 and prostaglandin F2a appear to be the final mediators in this process (Harman & Kim, 1999)

Bishop Score- The Bishop score is a cervical scoring system that was originally developed to assess multiparous women at the time of labor induction. The score is calculated based upon cervical dilation, effacement, station, position, and consistency. When the Bishop score is less than six, it is recommended that a cervical ripening agent be used before labor induction (Tenore, 2003). When the Bishop score exceeds eight, the likelihood of a successful vaginal delivery approaches that of a spontaneous labor (Harman & Kim, 1999). The duration of pregnancy before spontaneous labor is inversely correlated with the score, and further studies have demonstrated that a low Bishop score correlates with a prolonged labor or failed induction and the increased need for Cesarean delivery (Harman & Kim, 1999).

Relationship to Framework
Physiologically, research has shown that prostaglandins play a strong role in cervical ripening, EPO is, biochemically, a precursor of prostaglandin E that is in a readily bioavailable form to the human body. Theoretically, oral EPO administration should aid in cervical ripening and the improvement of Bishop scores. This inquiry will investigate oral EPO use in late gestation as a safe and efficacious cervical ripening agent that may aid in the prevention of postdates pregnancies.

Study Purpose/Type of Study/ Study Questions/Aims
This prospective, randomized, controlled clinical trial will examine the safety and efficacy of oral EPO as a cervical ripening agent in low risk nulliparous women with term pregnancies. This study will investigate some key questions related to oral evening primrose oil:

Does oral EPO have cervical ripening properties?
Does oral EPO have a statistically significant effect on Bishop Scores in term pregnant women?
Does the use of oral EPO decrease the incidence of postdates pregnancy?
Is oral EPO safe for use in pregnant women?

Limitations of the Study
The results of this study will not be able to be generalized to the pregnant population as a whole for several reasons. This study is limited to only low risk, nulliparous women who have self- selected to midwifery care. For these reasons alone, the results seen in this study will not necessarily be typical of the general population. An additional limitation is the fact that this study will take place in only one site. This will lengthen the amount of time that it will take for data collection, analysis and distribution. Additional sites would diversify the study population as well as speed the collection of data for analysis. However, this study will be a stepping stone that will help shape future research in the arena of using EPO for cervical ripening in pregnant populations.
Conceptual Framework and Review of Literature: Chapter Two
Conceptual Framework
According to Polit, Beck, and Hungler (2001) theories and conceptual frameworks are the primary mechanisms by which researchers organize their findings into a broader conceptual context. More specifically, theories are abstract generalizations that systematically explain the relationship between phenomena. Polit, et al. (2001) state that the “overall objective of theory is to make scientific findings meaningful, summarize existing knowledge into coherent systems, stimulate new research by providing direction and impetus, and explain the nature of relationships among variables.” The purpose of quantitative studies, such as in this inquiry, is to test hypotheses that are developed on the basis of an existing theory or conceptual model. In this inquiry, the proposed hypothesis is that EPO will promote cervical ripening in a safe and efficacious manner in term pregnant women. The underlying theory is that it is the prostaglandin precursor that is found in EPO that makes it an effective agent for cervical ripening, due to the presumed actions of prostaglandins in the human body and in natural cervical ripening. Although theories can never be proved, they represent our best efforts to describe and explain the phenomena at hand (Polit, et al., 2001). A conceptual model will broadly present an understanding of the phenomena of interest and reflect the assumptions of the model’s designer (Polit et al., 2001). This inquiry is based upon the use of physiologic theories that attribute to cervical ripening and the biochemistry of EPO and its potential bioactivity in the human body.
Physiology of Cervical Ripening
Biochemically and physiologically, cervical ripening simply consists of the breakdown of cervical collagen bundles and the taking up of water by the cervix. Clinically, cervical ripening is manifested by the softening, effacement, anterior rotation, and dilatation of the cervix (Barrilleaux, et al., 2002). According the Barilleaux, et al. (2002), the ideal cervical ripening agent would bring about these changes while provoking little if any uterine activity, however, no such agent has been discovered to date. It is believed that prostaglandins play a strong role in cervical ripening and the initiation of labor (Blackburn & Loper, 1992). Prostaglandins are organic compounds that act as chemical mediators or local hormones exerting their effect at or near the site of production (Blackburn & Loper, 1992). Prostaglandins are synthesized rapidly, are relatively unstable, and have a short half-life. They are formed by the enzymatic oxidation of arachidonic acid, a poly-unsaturated fatty acid precursor (Blackburn & Loper, 1992).

During pregnancy, the cervix increases in mass, water content, and vascularization. In order for cervical dilation and labor progress to occur, the cervix must ripen, or change from a relatively rigid to a soft distensible structure. The cervix is primarily composed of connective tissue, covered by a thin layer of smooth muscle (Blackburn & Loper, 1992). The connective tissue has a dense network of interlacing collagen and elastin fibers embedded in a gel-like ground substance matrix (Blackburn & Loper, 1992). Cervical ripening (the softening, effacement, and increased distensibility of the cervix) involves changes in collagen, proteoglycans, and smooth muscle with enzymatic degradation of collagen. Essentially, ripening involves changes in the solubility of collagen, which is under significant hormonal control that may involve changes in estradiol, progesterone, relaxin, prostacyclin, PGE2, and PGF2 alpha (Blackburn & Loper, 1992). More specifically, PGE2 and PGF2 alpha have a localized influence on cervical softening. Prostaglandin levels have been found to correlate with the degree of cervical dilation and to be higher in women with spontaneous onset of labor. PGE2’s action on the cervix is independent of uterine contractile activity and has been long used for its efficacy in improving cervical inducability (Blackburn & Loper, 1992).

Biochemistry of EPO
The literature review revealed several studies on the potential antioxidant properties of EPO, but none focused specifically on its properties as a prostaglandin precursor. Evening primrose is a biennial herb that grows wild in parts of North America and Europe (Salisbury Nursing School, 2003). Oil extracted from the seeds of the evening primrose contains linolenic acid, gamma linolenic acid, and vitamin E. Gamma linolenic acid is a known precursor of prostaglandin E and several other active substances, and is said to be the constituent of the oil responsible for its therapeutic effects (Kleijnen, 1994, Cohen, et al., 2000). One capsule of evening primrose oil contains about 320mg of linolenic acid, 40 mg of gamma linolenic acid, and 10 IU of vitamin E (Kleijnen, 1994). The Birch, Fenner, Watkins, and Boyd (2001) study sought to determine the antioxidant activity of the evening primrose seed. The study revealed that EPO is a highly unsaturated oil containing a significant source of gamma linolenic acid; a very active essential fatty acid and precursor of prostaglandin E1 and its derivatives. A second study by Wettsinghe, Shahidi, and Amarowicz (2002) also looked at the antioxidant properties of evening primrose seeds. This study revealed that the occurrence of gamma linolenic acid in the 1- and 3- positions of the triacylglycerol molecules of EPO makes them easily accessible to hydrolysis by pancreatic lipase in the small intestines. Because of this property, EPO is considered to be nutritionally more available than the other seed and vegetable oils containing gamma linolenic acid. This study also looked at the biochemical breakdown of EPO into antioxidant properties and found it safe and efficacious to exploit as a source of bioactive components for use in nutraceutical, as well as nonfood applications (Wettsinghe, et al., 2002).

Conceptual Map
EPO + TP--(+)-->CR (IBS)
EPO + TP--(-)-->PP
EPO=Oral Evening Primrose Oil Capsules
TP=Term Pregnancy
CR=Cervical Ripening
IBS= Improvement in Bishop Score
PP=Decreased Incidence Postdates Pregnancy

Review of Literature
An initial review of literature was performed through a Medline search of medical literature. An additional search was performed in Proquest Nursing Database, which contains nursing, medical, and lay publications. A final search was performed within Journal of Nurse-Midwifery and Women’s Health, as it is the premier American journal of research currently available for nurse-midwives. Multiple studies were available for pharmacologic and mechanical methods of cervical ripening and labor induction. It was difficult to find studies that focused exclusively on cervical ripening alone without tying in induction of labor, as well. However, little research is available on natural methods of cervical ripening, such as the use of herbals, castor oil, intercourse, breast stimulation, amniotomy, and membrane sweeping. The intent of this literature review was to reveal research that focused on EPO and cervical ripening. When it became apparent that almost no research existed in this arena, this writer then expanded her search to include research that would demonstrate the safety and efficacy of EPO for other uses, as well as the safety and efficacy of natural, pharmaceutical, surgical, and mechanical cervical ripening methods in order to help shape this inquiry.

Cervical Ripening and EPO
A study by Dove and Johnson (1999) investigated the use of evening primrose oil on the length of pregnancy and selected intrapartum outcomes at an American free-standing birth center in low-risk nulliparous women. More specifically, the study examined the effect of oral evening primrose oil on length of pregnancy, length of labor, incidence of postdates induction, incidence of prolonged rupture of membranes, occurrence of abnormal labor patterns, and cesarean delivery. A two group retrospective quasiexperimental design was conducted on a sample drawn from the records of all nulliparous women at a free-standing birthing center over a seven year period from 1991 to 1998. All of the records were screened for accurate gestational age dating, cephalic presentation, low risk status and delivery between 38 and 42 weeks gestation. The study group consisted of 54 women who took oral evening primrose oil in their pregnancy (500 mg three times a day starting at 37 weeks gestation for the first week of treatment, followed by 500 mg once a day until labor ensued), and the control group was composed of 54 women who did not take anything. Antepartum and intrapartum records of all women were reviewed focusing on the above identified criteria. Differences between measured variables of maternal age, Apgar score, birth weight, length of pregnancy, and length of labor were tested using Student’s t test. Correlation between client age, and days of gestation and hours of labor were investigated because the researchers believed that age could be a significant covariate affecting these two dependent variables. The study used a two-stage data analysis approach using SPSS for Windows, Version 8. Descriptive and parametric statistics were generated. Results showed no significant differences between the evening primrose oil group and the control group on age, Apgar score, or days of gestation (P>.05). All variables were then subjected to Levine's test for equality of variance. This retrospective chart review showed no benefit from taking oral evening primrose oil for the purpose of reducing adverse labor outcomes or for reduction of length of labor. Those in the evening primrose oil group actually demonstrated a trend toward greater risk of protracted active phase, arrest of descent, and vacuum extractor usage.

This is the first study that attempts to demonstrate the benefits of evening primrose oil in facilitating cervical ripening. The most prominent limitation of this study is its retrospective design using client records to assess outcome criteria. There was no strict procedure for defining the onset of labor or latent phase, and there were many potentially confounding variables that were not controlled for, such as the incidence of sexual intercourse, breast stimulation, and use of homeopathic remedies or castor oil. Any of these variables may have influenced cervical ripening and labor onset. This was a small study population of low-risk nulliparous women, therefore these results cannot be generalized to the population at large. It is difficult to say whether the findings of adverse labor outcomes are incidental or related to the use of evening primrose oil. Further research is strongly encouraged. Replication of this study in a larger, more controlled population is indicated prior to recommending oral evening primrose oil for cervical ripening. This study helps to shape the direction of this research proposal by indicating that replication in a larger population is needed before the results can be generalized, as well as giving suggestions for tighter controls in future research.

One additional study discusses EPO as a cervical ripening agent (McFarlin, Gibson, O’Rear, & Harman, 1999). The purpose of this study was to document the use of herbal preparations for cervical ripening, induction, and augmentation of labor by certified nurse-midwives (CNMs) and nurse-midwifery education programs. Five hundred ACNM members and 48 educational programs were surveyed by questionnaire via a single anonymous mailing. The sample was obtained by requesting every 5th name on the ACNM membership list from each of the six regions in the United States. Data were collected and analyzed using the Epistat statistical package, and Student’s t test was used for univariate comparisons and Chi square for categorical data. Of the 500, 90 were returned from CNMs who use herbal preparations and 82 were returned from those who do not use herbals. Of the CNMs who use herbal preparations to stimulate labor, 93% use castor oil, 64% use blue cohosh, 63% use red raspberry leaf, and 60% use evening primrose oil, and 45% use black cohosh (McFarlin, et al., 1999). Interestingly, the most commonly cited reason for using herbals was that they are natural and safe, yet the most common reason for not using herbals was the lack of research or evidence of safety of these substances. CNMs reported that the use of EPO was the most efficacious for cervical ripening while the use of castor oil was most efficacious for labor stimulation. Interestingly, 69% of those using herbals learned about them from other CNMs, while only 4% reported learning from formal research, and none from their education programs (McFarlin, et al., 1999). Three quarters of those using herbals report choosing herbals before the use of pitocin. Twenty-one percent did report complications with use of herbals including precipitous labor, tetanic uterine activity, nausea and vomiting. However, there were no reported complications with the use of EPO and red raspberry leaf tea. Sixty four percent of educational programs reported instruction in the use of herbals to stimulate labor, with EPO being the most common herbal preparation discussed (McFarlin, et al., 1999). Education level and geographic region did not affect the use or nonuse of herbals. CNMs who used herbals were found to be slightly younger (43 vs. 45 years, P<.01) and more likely to deliver at home or in a birthing center (P<.0006) than CNMs who never use herbal preparations to stimulate labor (McFarlin, et al., 1999).

This writer concurs with McFarlin, et al. (1999) that this study clearly demonstrates an interest in the use of herbal therapies to stimulate cervical ripening, augmentation, and induction of labor by CNMs. Those who did not already use herbals responded that they would be interested in using them if there were scientific data available on the safety and use. This study re-emphasizes the fact that the use of herbals is not based on scientific research or formal education, but passed on by word of mouth. Strengths of this study include its stratified random sampling pattern, however, this writer believes that the results cannot necessarily be generalized to all midwives, due to the low response rate. The researchers could have done two follow-up mailings in order to increase their response rate. Of 500 ACNM members sampled, 172 responded, which is only a 34% response rate. Forty-eight nurse-midwifery education programs were also surveyed. Thirty- six responded, for a response rate of 75%. This study depicts interest in the use of herbals by some midwives in the US and shows a strong interest, even in those not currently using herbal therapies, and is evidence of the need for further research in this area. In fact, the authors urge CNMs who currently use herbals to document and report their experiences in prescribing data, efficacy, and complications.

EPO Use in Women’s Health
Research studies that demonstrate the use of EPO in pregnant women are very limited. D’Almeida, Carter, Anatol and Prost (1992) ran a placebo controlled, partially double-blinded clinical trial in a comparison of the use of EPO and fish oil to the use of magnesium oxide, or to a placebo (olive oil capsules) in preventing pre-eclampsia of pregnancy. The study was designed based upon the theory that prostaglandins and thromboxane are involved in the development and clinical expression of pre-eclampsia. The study groups consisted of 150 volunteers, less than four months into the pregnancy (age 14-40 years), selected from a population in Luanda, South Africa of similar ethnic background and socioeconomic status. They were then were randomized into three groups. A group of primiparous and multiparous women were given placebo, EPO and fish oil, or magnesium oxide as nutritional supplements for a time period of six months. Approximately one fifth of these women had a personal or family history of hypertension. Participants were asked not to alter their current diet or lifestyle during the time of the study. Compared to the placebo group (29%), those in the fish oil and EPO group had a significantly lower incidence of edema (13%, p=0.004) (D’Almeida, et al., 1992). The EPO group also reported a higher rate of vomiting (p=0.04625), while the magnesium group complained of higher rates of diarrhea during the pregnancy (p=0.0005) (D’Almeida, et al., 1992). Those in the magnesium oxide group had statistically significant fewer subjects who developed pregnancy-induced hypertension (two out of 50 subjects, or 4%), and those in the placebo group had the highest rates (five out of 50 subjects, or 10%). Proteinuria occurred evenly across all three treatment groups. Three cases of eclampsia were noted, all in the placebo group (D’Almeida, et al., 1992). Birth weight was higher in the EPO and magnesium groups than the placebo group.

According to the authors (1992) this study shows that the use of EPO and fish oil, either alone or in combination with elemental magnesium, may prevent pre-eclampsia. However, they do acknowledge that there is no difference in the incidence of pre-eclampsia in their subjects and other pregnant populations surveyed in the West and developing countries. The incidence of pre-eclampsia is commonly cited to be about 5%, although the incidence is influenced by parity, race and ethnicity, genetic predisposition, and environmental factors (Cunningham et al. 2001). Several factors were not controlled for in this study, including cigarette smoking and stress, which may both play a role in the development of pre-eclampsia. The authors also state that the study was performed during wartime, which may have had an effect on the stress levels of the participants. The results of this study are promising, but need to be researched further under different conditions and for a longer duration of time. This study is beneficial to this inquiry in that it demonstrates the potential efficacy and safety of EPO in an area of women’s health, while reinforcing that the side effects are minimal.

Khoo, Munro and Battistutta (1990) published a prospective, randomized, double-blinded, placebo (liquid paraffin) controlled trial of the use of EPO for the relief of 10 specific symptoms of premenstrual syndrome (PMS) over a period of six cycles. The trial was then crossed over after three cycles. The sample population was recruited from the female population of a hospital in Australia. Criteria consisted of age 20-40, cyclic menstruation, ability to complete six full cycles of treatment, and presence of PMS symptoms in the following categories (one symptom from the first and second category, and two from the third category): fluid retention, breast symptoms, and mood changes. Those already taking NSAIDS, systemic steroids, or oral contraceptives were excluded from the study. Either active or placebo treatment commenced on day one of the first cycle and continued until the end of the third cycle. At the crossover, treatment commenced on day one of the fourth cycle and continued for the next three cycles. Daily treatment consisted of four capsules in the morning and four capsules at night. Subjects were asked to complete a four point scoring assessment of 10 PMS symptoms prior to the initiation of the study, and again during each cycle of the study. The results of the study showed an improvement in the symptoms of PMS during the trial, but no significant differences in the scoring between the active and placebo groups were found over six cycles, with the mean difference being –0.026 with a standard error of 1.144 (test statistic from paired t test=0.02; 37df: P=0.982) (Khoo, Munro, & Battistutta, 1990). The beneficial effect on all symptoms (especially psychological, fluid retention, and breast) was rapid with scores decreasing in the first cycle, but increasing slightly at the crossover period after the third cycle, regardless of whether the active or placebo medication was the next given. The authors concluded that any improvement in symptoms in these women was merely a placebo effect, making it difficult to apply the data in favor of EPO as a viable treatment for PMS.

Like many others of its kind in the arena of EPO research, this trial was very small, with only 38 women participating, which again make it difficult to generalize the results to the population at large. In addition, crossover trials have been criticized for their design, especially since those taking EPO in the first period may experience extended effects during the second period, possibly affecting the scoring associated with placebo treatment. However, the group taking the placebo in the first round and EPO in the second round could serve as its own control for statistical analysis. A strength of this study design is the fact that subjects were randomized to treatment groups, which allows them to serve as their own control group (Polit, et al., 1999). An ethical strength of this study is that it exposes everyone to the potential benefit of treatment. This trial is beneficial to this inquiry in that it demonstrates safety of EPO, no adverse reactions were noted with its use, and in that it examines it’s potential prostaglandin effects. It is theorized that women with PMS have an abnormal sensitivity to normal circulating prolactin levels. This trial originated based on the theory of the role of prostaglandins and essential fatty acids in the disorder of PMS by attenuating the biological effects of prolactin. Additional research needs to follow, which replicate this trial for longer periods of time, with the elimination of the crossover component.

Chenoy, Hussain, Tayob, and Moss (1994) did a randomized, double blinded, placebo controlled pilot study that evaluated the efficacy of gamolenic acid (in the form of EPO) in treating hot flushes and sweating associated with menopause. This study was designed based upon anecdotal evidence from the lay public that EPO is effective in the control of vasomotor symptoms. The underlying theory in the use of EPO for the treatment of menopausal symptoms lies in the changes in the peripheral and central mechanisms dependent upon prostaglandins that stimulate the release of gonadotrophins through the hypothalamus. Unfortunately, the results of this study found no benefit by using EPO. Fifty-six volunteers were recruited in England to participate in this study, none of which had received estrogen replacement therapy or other essential fatty acid supplements in the previous two months. Also controlled for were concurrent treatment with systemic steroids, nonsteroidal anti-inflammatory agents, anti-convulsants, clonidine, and phenothiazides. A threshold change in the number of daytime flushes of 13% was desired. Women did not receive any treatment during the first month in order to establish a baseline. This was followed by random allocation to six months of treatment with active (500mg EPO) or placebo (500mg liquid paraffin) capsules, with four being taken twice a day. Patients were assessed at baseline, one, four, and seven months for the number and severity of flushing and sweating episodes per day, as well as any adverse effects. The change in the number of episodes of flushing and sweating between the initial control cycle and the last control cycle was calculated for each subject. The significance of improvement was assessed by the Wilcoxon signed ranks test, and differences between groups were assessed by the Mann-Whitney U test (Chenoy, et al., 1994). All women given the placebo showed a significant positive difference between the control and the last cycle (P<0.05) (Chenoy, et al., 1994). The women given EPO, did not show a significant improvement, except in the maximum number of night flushes (P=0.014) (Chenoy, et al., 1994). All patients tended to show improvement, but a change in the mean number of daytime flushes of less than 13% could not be detected (P=0.02) (Chenoy, et al., 1994). It is entirely possible that this is a developmental effect, in that the sweats would naturally decline over time regardless of treatment.

One major flaw in this study is that all data were included in the analysis irrespective of any lack of compliance or protocol violations. Only a total of 35 women completed the trial. The 21 women that did not complete the trial withdrew only because of a poor clinical response to treatment. This study was not able to demonstrate the benefits of EPO over placebo in menopausal flushing. One possible explanation for this may be because the metabolites of EPO have high concentrations of prostaglandins, which decrease the affinity of ligands, such as estrogens and other hormones for their receptors. In addition, prostaglandins may stimulate the release of high concentrations of pituitary gonadotropins that have long been implicated in the production of menopausal vasomotor disorders (Chenoy, et al., 1994). This study again demonstrated that side effects to the use of EPO were minimal, with only slight nausea, which was relieved with administration after meals (Chenoy, et al., 1994). This study helps shape this inquiry by demonstrating the safety and minimal side effects of EPO, as well as demonstrating its efficacy, but not to the degree desired by this study. This study also delineates possible safe dosing regimens to be employed in future research. Future replication of this study in larger samples is desirable in an attempt to finding acceptable natural alternatives to hormonal therapy in menopausal women.

Nonpharmacologic Methods of Cervical Ripening
Review of literature turned up little to no formal research on natural cervical ripening methods, such as intercourse, nipple stimulation, membrane sweeping, or other herbals, such as black cohosh, and castor oil.

Magann, McNamara, Whitworth, Chauhan, Thorpe, and Morrison (1998) designed and implemented a study looking at whether the risk for postdatism could be reduced by serial membrane sweeping in women with an unfavorable cervix and a negative fetal fibronectin test result at 39 weeks gestation. It has been proposed that serial membrane sweeping creates a natural prostaglandin release that aids in cervical ripening. A total of 65 volunteers were recruited in three different geographic locations if they were candidates for vaginal delivery at 39 weeks gestation, had an unfavorable cervix (Bishop score </= 4), and a negative fetal fibronectin test. Patients were randomly assigned to a group (n=33) for membrane sweeping every three days or to a control group (n=32) who received gentle examinations every three days. The two groups were similar in demographics of maternal age, race, gravidity, parity, gestational age at the onset of the study, and Bishop score at the beginning of the investigation. Estimation of gestational age and estimated date of confinement was tightly controlled. Outcome variables included gestational age and Bishop score at 39 weeks and on admission to labor and delivery, reason for delivery and delivery admission, route of delivery, and neonatal outcome. The statistical analysis was by Student t test, Chi square, and Fisher exact test. Relative risks and 95% confidence intervals were calculated. A P value of < .05 was considered significant. The gestational age on admission for delivery was earlier in the membrane sweeping group (39.9 +/- 0.3) versus the control group (41.5 +/- 0.6, P< .0001) (Magann, et al., 1998). The Bishop score on admission to labor and delivery was greater (8.8 +/- 2.1) in the membrane sweeping group than in the control group (6.2 +/- 2.7, P< .0001) (Magann, et al., 1998). The number of women admitted for labor inductions at 42 weeks gestation was 18 of 32 in the control group versus none in the membrane sweeping group (P< .0001) (Magann, et al., 1998). The authors concluded that serial membrane sweeping significantly reduces the risk of postdatism and induction of labor.

This is a well designed study that demonstrates a positive effect of serial membrane sweeping on the reduction of postdates pregnancies, thereby decreasing the risks associated with postdates pregnancies, such as oligohydramnios, Cesarean deliveries for macrosomia and fetal distress, meconium stained amniotic fluid, fetal postmaturity syndrome, and an increased risk of perinatal morbidity and mortality (Magann, et al., 1998). This research aids in the development of this inquiry by demonstrating methodology that will be easy to replicate and apply to this inquiry, such as the use of serial vaginal examinations and Bishop scoring as an evaluation tool. This study also shows that women with unfavorable cervices at 39 weeks are likely to remain undelivered at 41 weeks, thus necessitating the use of additional antenatal testing and possible labor induction, which may increase the cost of care during pregnancy. This research again follows the underlying premise that prostaglandins help ripen the cervix and decrease the incidence of postdates pregnancies, which is what this inquiry intends to investigate. The authors conclude that methods of cervical ripening are indicated to decrease the incidence of postdatism.

Although no articles could be obtained for the use of castor oil as a cervical ripening agent, Davis (1984) performed a retrospective study at an American out of hospital birthing center, via chart review over a five year period from 1976 to 1981, on the use of castor oil to stimulate labor in term 37 to 42 week, uncomplicated pregnant patients with premature rupture of membranes. Of 196 patient records studied, 107 received castor oil and 89 did not. One hundred seven patients presenting with PROM and at least a four-hour latency period received two ounces of castor oil orally. The main outcome variable measured was length of time to labor onset. Other variables that may have influenced the effectiveness of the castor oil were also evaluated, including parity, Bishop score, gestational age of patients, and birth weight. In addition, outcomes of labors were also analyzed for method of delivery, need for oxytocin intervention, Apgar scores, and the presence of meconium. Data were analyzed by t test to individually evaluate multiple factors on a single dependent variable. The two groups did not differ in respect to patient age, parity, gestational age and Bishop score.

Of the 107 that received castor oil, 80 (75%) has spontaneous onset of labor compared to 52 (58%) of 89 patients in the control group (P<0.05) (Davis, 1984). The effects of castor oil were most pronounced among primiparas with 75% entering spontaneous labor in the experimental group and only 54% in the control group (P<0.01) (Davis, 1984). Castor oil was also found to be significantly associated with labor onset at 41-42 weeks (P<0.01) (Davis, 1984). There were noted to be three times as many Cesarean sections in the control group (15.7%) as the castor oil group (5.6%) (P<0.01) (Davis, 1984). Davis (1984) proposes that this decreased rate may be due to the fact that castor oil may function as a ripening agent, thereby decreasing the rate of failed labors leading to Cesarean section. The castor oil group did not require oxytocin as often, but not at a significant level. The incidence of Apgar scores less than seven did not differ at one minute and five minutes between the two groups. The presence of meconium at the time of rupture of membranes was very low. No one developed meconium staining after castor oil administration. There was no significant maternal morbidity or mortality in either group, and there were no maternal deaths. The expected side effect of diarrhea did occur in almost all in the experimental group, however maternal dehydration and electrolyte imbalance were not problematic (Davis, 1984).

This study demonstrates that there are safe, convenient, effective, inexpensive, nonpharmacologic methods of labor induction available for those seeking alternative therapies. Although there has been recent anecdotal speculation about the presence of meconium stained amniotic fluid associated with the use of castor oil for labor stimulation, no formal evidence is available to back this claim. One weakness of this study is its retrospective nature that lacks randomization, which may create a bias. This study only looks at a population with PROM, and it is possible that they were more ready for labor than the average population at term. The author cautioned that the results that occurred with the unique population of women at a birthing center may not be able to be generalized to all groups of women. In vivo studies of oral administration of castor oil in pregnant rats have demonstrated a significant increase in portal venous prostaglandin E concentration, therefore it has been hypothesized that there is a relationship between intestinal prostaglandin synthesis and its effect on the uterus (Davis, 1984). This theory is helpful to the inquiry because it is the belief that the prostaglandin in EPO that will aid in cervical ripening.

Combined Natural and Pharmacologic Methods of Cervical Ripening
Damania, Natu, Mhatre, Mataliya, Mehta, and Daftary (1992) performed a study looking at two methods of cervical ripening in Nowrosjee Wadia Maternity Hospital in Mumbai, India: breast stimulation and oxytocin infusion. Breast stimulation was performed for one hour three times daily, each breast alternating every 10 minutes. Oxytocin infusion started at 2.5 milliunits per minute for three hours per day for three days. The authors are motivated by the fact that prostaglandins are not available in their country, therefore they need to find alternative means of cervical ripening. A total of 40 primigravida women were randomly selected to either breast stimulation or oxytocin administration, based upon an obstetric indication for induction of labor if they had a Bishop score of 5 or 6 and at least 37 completed weeks of gestation. An additional 20 patients were randomly selected to be a control group in which no method was employed. Outcome variables were Bishop score, which was assessed at 24 hours and at 72 hours, and neonatal outcomes.

The Bishop score improved in 41.2% of cases where breast stimulation was used compared to 75% where an oxytocin infusion was given (Damania, et al., 1992). The study was brought to a premature halt after three cases of fetal death were noted in the breast stimulation group. There was also one fetal death noted in the oxytocin group. The authors concluded that cervical ripening with an oxytocin infusion is the preferable method since the dosage can be precisely controlled, making the technique more precise, predictable and reliable. The authors then suggest that since breast stimulation is effective in ripening the cervix, it may be used only in cases of intrauterine fetal death as it may otherwise adversely affect fetal outcome. Inclusion criteria for this study was that all patients had to have a reactive nonstress test, so it is clear that fetal monitoring was available. However, the study fails to mention whether additional fetal monitoring was done during the course of treatment in various groups. The researchers do suggest in the discussion that perhaps contraction stress testing should be employed during breast stimulation. Another drawback to this study is that it does not delineate any demographic data on the study participants or indicate whether this was a charity hospital or more reasonably well off women.

This study demonstrates the importance of establishing the safety as well as efficacy of the different methods of cervical ripening. It assists this inquiry in methodology in the designation of simple outcome variables, such as Bishop scoring and neonatal outcome. Prior breast stimulation studies have demonstrated that most Bishop scores improved by a three to four point increase with breast stimulation in low risk groups who had no real indication for cervical ripening, and 36% of patients were in labor within 3 days (Damania, 1992). Breast stimulation has been criticized because it is difficult to control and quantify. It is difficult to say what variables contributed to the intrauterine fetal deaths in the breast stimulation group in this study. Postmortem examination of all three neonates revealed no anomalies. The higher risk patients in this study included postterm pregnancy, hypertension, and intrauterine growth retardation. It is more likely that the authors should have concluded that breast stimulation be limited to use in lower risk patients. In addition, this study was performed in a country with higher rates of maternal and fetal morbidity and mortality. The maternal mortality ratio in 1995 in the United States was 12, whereas in India, it was 440 for the same year (World Health Organization, 1995). Unfortunately, India did not report infant/neonatal mortality ratios to the World Health Organization for comparison with United States statistics. Therefore, it is difficult to generalize the findings of this study to all populations.

Pharmacologic and Mechanical Methods of Cervical Ripening
Sciscione, McCullough, Manley, Shlossman, Pollock, and Colmorgen (1999) compared intracervical prostaglandin E2 gel with the insertion of a Foley bulb for efficacy in preinduction cervical ripening. Women with a Bishop score of less than or equal to five, who were admitted for induction of labor were randomly assigned to treatment with either prostaglandin E2 gel or a Foley bulb. Inclusion criteria included a single gestation in the vertex position with a gestational age of > 28 weeks. Every effort was made for the same examiner to perform the pre and post ripening Bishop scores. The prostaglandin E2 gel dose was repeated every six hours as long as the Bishop score remained </= five, to a maximum of three doses. The Foley group had a number 14 Foley inserted, inflated, and placed on traction. Either immediately after Foley bulb extrusion or six hours after the last prostaglandin E2 gel placement, patients were placed on an oxytocin drip if they were not in labor. The primary outcome measure was change in Bishop score. Secondary outcome measures included time for preinduction cervical ripening, total preinduction and induction times, delivery route, uterine hyperstimulation, side effects, patient comfort, fetal heart rate disturbances, and patient charges. Students t test, Fisher exact test, and the Chi square test were used to compare the two groups. P< .05 was considered statistically significant (Sciscione, et al., 1999).

Sciscione, et al. (1999) concluded that the use of the Foley catheter resulted in higher pre- induction Bishop score, a greater change in Bishop score, a shorter induction time, and lower patient charges than did the prostaglandin E2 group. A total of 77 women were in the Foley group and 72 were in the prostaglandin group. The groups were similar with respect to maternal age, gestational age at inclusion, gravity, parity, previous Cesarean section delivery history, insurance providers, and Bishop scores at entry. Both the Bishop score after preinduction ripening (6.5 vs 5.1, P< .001) and the change in Bishop score (3.5 vs. 2.7, P= .015) were significantly higher in the Foley group (Sciscione, et al., 1999). The researchers found no difference between the groups in mode of delivery, infant weight, rate of hyperstimulation, shoulder dystocia, patient discomfort, epidural use, oxytocin use, or nonreassuring fetal heart patterns. The preinduction time (9.9 vs. 17.2 hours, P< .001) and the total induction time (22.4 vs. 30.4 hours, P< .001) were significantly shorter in the Foley group (Sciscione, et al., 1999). The Foley group also accrued 31% less in patient charges (P< .001) (Sciscione, et al., 1999). There were few maternal side effects in either group. The Foley group complained of discomfort upon insertion, and four patients in the prostaglandin group complained of nausea.

This study supports the use of an alternative, inexpensive nonpharmacologic cervical ripening agent. It highlights the safety of the Foley catheter for preinduction cervical ripening. A strength of this study is its larger sample size, which accounts for greater statistical significance of the findings. Foley catheters appear to effect cervical ripening not only through direct mechanical dilation, but also through the release of prostaglandins (Sciscione, et al., 1999), which is consistent with the proposed mechanisms of this inquiry. This study also assist in the formation of the methodology of this inquiry by demonstrating that the Bishop score is an effective outcome measure that can be used to assess the degree of cervical ripening as a result of the intervention being used.

Barrilleaux, Bofill, Terrone, Magann, May, and Morrison (2002) conducted a study that compared the efficacy of misoprostol, dinoprostone gel and a Foley catheter. Misoprostol is an inexpensive prostaglandin E1 analog that brings about cervical ripening, as well as significant uterine activity. Three hundred thirty nine women who needed cervical ripening and labor induction were randomized into one of three groups: (1) supracervical Foley catheter and intravaginal dinoprostone gel every four hours, (2) supracervical Foley catheter and 100mcg oral doses of misoprostol every four hours, or (3) serial 100mcg oral doses of misoprostol every four hours. Each patient was to receive up to four doses of medication for ripening. Intravenous pitocin was only administered via a set high dose protocol if there was a protraction disorder of labor. Criteria for inclusion in the study included Bishop score less than seven, an obstetric indication for induction of labor, an unscarred uterus, a singleton pregnancy, a cephalic presentation, and intact membranes. The main outcome variable was the interval of time from the first intervention to the time of delivery. Additional variables included the presence or absence of uterine tachsystole and the use of intrapartum terbutaline to control the tachysystole if there were moderate or severe fetal heart rate decelerations, repetitive late decelerations, or prolonged fetal bradycardia. Continuous data were compared with the use of t tests with the Bonferroni correction or Kruskal-Wallis analysis of variance on ranks. Dunn’s method for multiple pairwise comparisons was also used. Categoric data were compared using the Chi square test. And significance was to be at a probability value of <.05 (Barrilleaux, et al., 2002).

This study found no significant difference in the time from the first intervention to delivery in all three groups (P= .546) (Barrilleaux, et al., 2002). A similar percentage of women required oxytocin administration in all three groups (P= .103), however the need for oxytocin was least in the balloon/misoprostol group (P= .033) (Barrilleaux, et al., 2002). The rates of Cesarean section were similar among all three groups (P= .722) (Barrilleaux, et al., 2002). Rates of tachysystole were high in all three groups (P= .209), the use of intrapartum terbutaline more common in the misoprostol group (P= .007), and there were no significant differences of Apgar scores or umbilical artery pH (Barrilleaux, et al., 2002). The authors concluded that 100mcg of oral misoprostol with and without the use of Foley catheter was equivalent in efficacy to the use of a Foley catheter and dinoprostone gel for cervical ripening and induction of labor.

This study assists this inquiry in demonstrating the effectiveness of a medication, misoprostol, being used for an off label purpose of cervical ripening. However, the authors did note significant rates of uterine tachysystole and a significant number of these episodes that needed to be treated with intrapartum terbutaline. Tachysystole remains the most common complication of misoprostol use for cervical ripening and labor induction. The appropriate dose of misoprostol for cervical ripening is yet to be determined. Perhaps 100mcg is too high a dose, and this study might yield different outcomes with regard to complications in lower dosages. Its strengths lie in the fact that it is a randomized prospective trial with a large sample size. A drawback to this study is that it combines methods of cervical ripening within two of the three groups, which makes it difficult to determine exactly what the effective agent was. In addition, the most common indication in the study for induction was pregnancy induced hypertension with oligohydramnios. These factors may have skewed the resulting tachysystole and fetal heart rate abnormalities, as well as the rate of Cesarean section. This study also did not control for parity, which can certainly skew the main outcome variable being measured.

Based upon the literature review, we know that EPO is a popular herb that has been studied for a variety of conditions in women’s health, with mixed results. Many of the trials are small in sample size and have been crossover trials, which may create a bias in the results. This makes the results difficult to generalize to the population. EPO has been studied only once as a cervical ripening agent with disappointing results, and additional prospective research with better controls is indicated. There is a multitude of research on pharmacologic and mechanical methods of cervical ripening, but little on alternative, more natural methods of cervical ripening. These pharmacologic studies will help design and guide the direction of this inquiry.


Chapter Three Quantitative Research Proposal
Research Design
The review of scientific literature revealed a lack of evidence supporting the use of oral evening primrose oil for a variety of medical conditions, as well as cervical ripening. Only one research article was found examining its use as a cervical ripening agent. This research study was retrospective in nature and had several weaknesses. It failed to show EPO as an efficacious cervical ripening agent and even questioned its association with an increased incidence of prolonged rupture of membranes, oxytocin augmentation, arrest of descent, and vacuum extraction. The authors of this study concluded that it was essential that prospective, randomized clinical trials be performed in larger settings before any generalizations about its safety and efficacy as a cervical ripening agent could be made (Dove & Johnson, 1999). This study will develop knowledge in this area by building on prior research.

Experiments are the most powerful designs for testing hypotheses of cause-and-effect relationships (Polit, Beck, & Hungler, 2001). In an experimental study, the researcher is primarily interested in the relationship between the independent variables and dependent variables (Polit, et al., 2001). Testable research hypotheses state the expected relationship between the independent variable (EPO) and the dependent variables (cervical ripening/Bishop score and a decreased incidence of postdates pregnancies) within a population. Research designs need only possess three properties in order to qualify as an experiment. These properties are manipulation of a variable(s), the use of controls, including a control group, and randomization (Polit, et al., 2001).

This experimental research proposal is for a prospective double-blinded randomized controlled clinical research trial, using a between-subjects design that will examine the use of oral EPO versus placebo as a cervical ripening agent. The subjects will be randomly assigned into two groups: experimental (EPO) and control (placebo) groups, and outcome data (weekly Bishop scoring and other selected outcomes) will be collected. This will be a longitudinal study using a before-after design that aims to determine a correlation between the use of oral EPO and an increase in cervical ripening measured via the Bishop score, as well as a decrease in the incidence of postdates pregnancies in low risk nulliparous women.

The research design notation for this study is as follows:

R1 O X O1 (group 1 is randomly assigned to oral evening primrose oil administration)
R2 O Y O2 (group 2 is randomly assigned to oral placebo administration)

X-oral evening primrose oil administration
Y-oral placebo administration
R-randomization
O-baseline observation/Bishop scoring
O1/O2-weekly observation/Bishop scoring until birth

Threats to Internal and External Validity
Internal validity is the extent to which it is possible to make an inference that the independent variable is truly influencing the dependent variable (Polit, et al., 2001). Selection can be a potential internal threat to a research study. An additional threat is the homogeneity of the women who self select themselves into midwifery care. These women tend to be low risk and seek low/noninterventive forms of care. Randomization is essential, as it gives every participant an equal chance of being included in any group. This eliminates selection/systematic bias (Polit, et al., 2001). This means that randomly assigned groups are expected to be comparable with respect to an infinite number of biologic, psychological, and social traits at the outset of the study. Any group differences emerging after randomization can then be attributed to the treatment (Polit, et al., 2001). An additional threat to internal validity is history. This is the potential occurrence of events concurrent with the independent variable that can affect the dependent variable (Polit, et al., 2001). The history threat will be decreased substantially by having an experimental research design. Maturation is a potentially strong threat to internal validity because it refers to processes occurring within subjects as a result of time (Polit, et al., 2001). For example, the exact physiology of cervical ripening is not completely understood. Therefore, it is entirely possible that the subjects in this study will show increasing Bishop scores over time as they near term regardless of the administration or oral EPO. Finally, mortality is a threat that arises from attrition. Some subjects will withdraw due to loss of interest or motivation. Northern Virginia is known to have a very transient population, therefore it can be expected that a small percentage will move away during the course of the study.

External validity refers to the generalizability of the research findings to other settings or samples (Polit, et al., 2001). A threat to external validity in this particular study is the choice to include only nulliparous women. Approximately a quarter to a third of a practice can be expected to be nulliparous. An additional threat is the fact that all participants will be a low risk midwifery clientele, therefore the findings may not necessarily be able to be generalized to the pregnant population as a whole, but will be intended to be used as a springboard for future study designs of EPO as a cervical ripening agent in broader populations of pregnant women. This researcher feels that it is more important to maintain tighter controls to increase internal validity, while sacrificing external validity to a lesser degree.

Measures to Control
Research control is mainly concerned with eliminating possible extraneous influences on the dependent variable so that the true relationship between the independent and dependent variables can be understood (Polit, et al., 2001). This researcher has decided to only include low risk nulliparous women in the study, as multiparous women tend to have more spontaneous ripening near term. High-risk pregnancies will automatically be excluded from inclusion to the study due to the potential for confounding variables affecting the course of care. There will be multiple confounding variables that will need to be controlled. These include other natural activities that may have an impact on cervical ripening during the time under study. Women in both groups will be asked to limit such activities as nipple stimulation, intercourse, and to refrain from taking other herbal supplements or castor oil that may have an impact on cervical ripening. These activities are not recommended in terms of this study, but some women may choose to do so anyway. If women do participate in these activities, they are asked to be respond honestly by recording the activities in their daily journal (See Appendix F).

Participants in the study will all follow a set prescribed dosing regimen, whether they are in the experimental or control groups. Review of literature reveals a maximum of two to 16 500mg capsules to be a safe daily dosage of oral EPO (Kleijnen, 1994). This researcher has followed the lead of the only research study for cervical ripening and has chosen a dosing regimen of 500mg three times per day starting at 37 weeks gestation to be continued through delivery for the experimental group, and a paraffin placebo to be taken three times per day in the control group for the same duration (Dove & Johnson, 1999). Subjects will be asked to keep a record of the dosing regimen. If a subject fails to take greater than five doses, her results will be omitted from the study results.

Sampling Methods
Site/Population of Interest
A convenience sample will be drawn from the population of prenatal care clients attending Loudoun Community Midwives, in Leesburg, Virginia. All nulliparous women that meet inclusion criteria in this practice will be offered information on this clinical trial between 35 and 36 weeks gestation. Because this is a solely midwife owned and operated practice, those women attending care in the practice have already been screened as low risk. Those choosing to be included in the study will be randomly assigned to either the experimental or control group and will begin administration of the prescribed regimen of oral EPO or placebo at 37 weeks gestation.

In quantitative studies, larger sample sizes are preferred because they reduce the sampling error (Polit, et al., 2001). The larger the sample, the more statistically significant are the findings (Polit, et al., 2001). Loudoun Community Midwives averages 25 deliveries per month, with approximately half being nulliparous women. This researcher estimates a pool of 12 nulliparous candidates per month, with approximately 90% (or 10 women) choosing to enroll in the study per month. This research midwife will use the “n versus v” rule of thumb to estimate a sample size, where 10-30 subjects are needed for each variable measured in the study. To achieve a total sample of 250 women, the study will need to be carried out over an extensive period of time, slightly greater than two years.

Entry to Site/Access to Population
Entry to the site and access to the population under study will not be an issue. This researcher is currently one of three certified nurse-midwives in the practice and already has access to the intended study population. The other two midwives in the practice already recommend the use of EPO as a cervical ripening agent and are eager to see further research on the topic. In addition, application to Loudoun Hospital Center Ethics Committee will be made for approval of this research project.

Inclusion/Exclusion Criteria
Inclusion Criteria
Nulliparous women between 35 and 36 weeks gestation, regardless of age or race
All study candidates must be willing to participate in the study by taking the prescribed regimen of EPO or placebo
Low risk status
Strong gestational dating by certain LNMP or 1st trimester sono
No size dates discrepancy
Candidate for vaginal delivery
Cephalic presentation
Exclusion Criteria
Multiparous women
Development of conditions that change risk status to high risk, such as PIH or other pregnancy related conditions
Any size dates discrepancy
Non-candidate for vaginal delivery
Non-cephalic presentation
Women who have had cervical surgery that may have created scar tissue and inhibit the normal ripening process

Sampling Methods and Strengths/Weaknesses
The primary strength of this sampling method is randomization of the subjects to control and EPO groups. This will eliminate selection bias, as discussed earlier in this proposal. Convenience sampling entails the use of the most conveniently available people as study participants. This method was chosen simply to keep cost down by gaining easy access to a local study population, thereby eliminating the increased time and monetary expenditure that would be necessary if the researchers had to travel to other locations. The primary weakness of convenience sampling is the possibility that the available subjects may be atypical of the population being targeted for study (Polit, et al., 2001). Because this study is only in low risk midwifery clients, it will be difficult to generalize the findings to other pregnant women without further research trials in broader populations.

Ethical Considerations
This proposal will be submitted to the Philadelphia University Institutional Review Board for approval. Following approval from Philadelphia, this proposal will be submitted to Loudoun Community Midwives lead midwife for formal approval. Finally, this proposal will go before the Loudoun Hospital Center Ethics Committee for final approval prior to the commencement of data collection. Individual informed consent (see Appendix A) will be obtained from subjects as they enroll in the study. Three primary ethical principles will be followed during the course of research: beneficence, respect for human dignity, and justice (Polit, et al., 2001). U.S. government requirements governing research with pregnant women will be respected and followed. This means that the research must meet the health needs of pregnant women and the risks to her and the fetus will be minimized (Polit, et al., 2001).

Each participant in the study will be fully educated on the nature of the study, the purpose of the study, the risks and benefits of using cervical ripening agents, and will be assured of the confidentiality of the study. Each participant will be assigned a number, and a record of names and numbers will be kept separate from all study research materials and will be kept in a locked file cabinet. This identifying list will be destroyed upon completion of this study.

One potential ethical concern with this study is that the potential participants will be known to the research midwives due to the fact that they are patients in that practice. Patients of the practice may feel obligated to participate in order to please their midwife even though they may not really want to participate at all. Secondly, the research midwife will need to remain completely impartial and not express disappointment, nor treat the patients any differently in their care should they choose not to participate in the study. The research midwife also has an ethical obligation to accurately report the findings, even if they do not support this long held practice. An additional ethical consideration is given to the fact that herbal use in pregnancy is relatively unstudied. Herbal preparations are not regulated by the FDA, therefore the exact strength and actions of herbals are not quantified, and the clients will be informed of this upon consenting to participate in the study.

Data Collection Procedures
Timeframe
Immediately following approval by Philadelphia IRB, Loudoun Hospital Ethics Committee and Loudoun Community Midwives, the clients that meet inclusion criteria will be able to give informed consent and enroll in the study, and data collection will immediately follow. In order to accomplish a valid research study with results that can be applied to the future of midwifery care for clients seeking natural alternatives for cervical ripening, the timeframe for this study will be in excess of two years for data collection, followed by six months for interpretation and statistical analyses. An additional six months will be allotted for the final results to be condensed into a manuscript and application made for publication. Please see Appendix D for Timeline.

Procedure
Qualifying nulliparous clients, consenting to being included in the study, will be randomized via a computer software package to either the experimental or control group. The client will then receive a packet containing instructions for the TID dosing regimen, a demographic information/data collection sheet to be filled out at each prenatal visit and at delivery (Appendix E), a journal for documenting each dose taken, and an unmarked bottle of pills that will contain enough for the client to maintain TID dosing through 42 weeks gestation. Neither the client, nor the midwife will know to which group the client has been randomized. The client instructions will ask that the client take one pill three times per day starting at 37 weeks gestation and continuing through delivery. Each time she takes a dose, she should record the date and time in the journal provided, and the journal should be brought with her to each prenatal visit for the midwife to verify continuing eligibility for the study. (If the client misses more than 5 doses or develops any of the excluding criteria listed above, she will be excluded from the final study results. However, data will continue to be collected unless a client withdraws themselves from the study entirely.) The journal will be relinquished to the midwife at the time of admission for delivery.

Description of Each Instrument and its Reliability and Validity
All three midwives will be trained on the study objectives and outcome criteria. The main outcome variable measured will be (a biophysiologic, in vivo measure) weekly Bishop scoring for cervical ripening scores until delivery. The Bishop score is a cervical scoring system that was originally developed to assess multiparous women at the time of labor induction. The score is calculated based upon cervical dilation, effacement, station, position, and consistency. When the Bishop score is less than six, it is recommended that a cervical ripening agent be used before labor induction (Tenore, 2003). When the Bishop score exceeds eight, the likelihood of a successful vaginal delivery approaches that of a spontaneous labor (Harman & Kim, 1999). The duration of pregnancy before spontaneous labor is inversely correlated with the score, and further studies have demonstrated that a low Bishop score correlates with a prolonged labor or failed induction and the increased need for Cesarean delivery (Harman & Kim, 1999).

All research midwives are very familiar with Bishop scoring as they use it routinely in their practice. All three will be involved in collecting Bishop score data on the study participants at their regular scheduled weekly office visit. Unfortunately, it is not as ideal to have three different researchers doing the Bishop scoring on varying clients. A better design would entail the same examiner assigning a Bishop score to the same client at each visit. However, the midwives feel strongly that the patients need to continue to rotate through the providers for their last few weeks of care prior to delivery.

Reliability is the degree of consistency or accuracy with which an instrument measures the attribute it was designed to measure (Polit, et al., 2001). Validity is the degree to which an instrument measures what it is intended to measure (Polit, et al., 2001). The reliability and validity of this study will depend on the honesty and diligence of the study participants using the journal instrument to record medication dosing, as well as other demographic information or confounding variables. Bishop scoring will vary from midwife to midwife, based upon their individual training and experience, and this may slightly impact the validity and reliability of the results, however the Bishop score is the best known method of assessing cervical ripening and its relationship to labor. Interrater reliability could be improved by having two or more researchers making simultaneous Bishop scores, however this technique creates increased discomfort for the woman, as well as bringing up additional ethical considerations. If the woman consents, a second midwife will be available to assign a Bishop score in order to increase interrater reliability. Biophysiologic measures are considered highly objective means of providing valid measures of the targeted variables. A concern to be highlighted when using biophysiologic measures as instruments is that the measuring tool may affect the variables it is intending to measure (Polit, et al., 2001). This means that weekly cervical manipulation may be a confounding variable in and of itself, which may contribute to cervical ripening.

A secondary outcome variable of interest is the incidence of induction at 42 weeks gestation for postdates pregnancy. This research midwife proposes that taking EPO capsules from 37 weeks through delivery will decrease the incidence of induction for postdates by ripening the cervix and assisting the woman to go into spontaneous labor. Additional outcome variables will be investigated once the client has delivered. These are gestational age at delivery, amniotic fluid status, length of rupture, length of labor, type of delivery, incidence of abnormal labor patterns, infant birth weight, Apgar Scoring at one and five minutes, and NICU admissions. The delivering midwife will be responsible for recording this data at the time of delivery.

Data Analysis Plan
Data Management and Analysis Software/ Statistics
The main research question this study is asking is: What is the relationship between oral EPO and cervical ripening? Secondarily: What is the incidence of postdates induction in women who use oral EPO versus a placebo? A statistical test is a procedure for testing hypotheses and evaluating the believability of the findings (Polit, et al., 1999). Analysis of covariance (ANCOVA) will be used to control extraneous variables statistically. Although randomization is used to the experimental or control groups, the sample population being used is one of convenience. Therefore, ANCOVA offers the possibility of post hoc statistical control (Polit, et al., 1999). This will aid in controlling statistical differences between the experimental and control groups, such as maternal age, which may potentially have an impact on the outcome variables.

Students t test for dependent groups is one of the most appropriate parametric tests that will be applied to the data collected in assessing the two groups of people on a dependent variable, ie. EPO and control and cervical ripening assessed through Bishop score. This test measures the difference between the two dependent group means. According to Polit et al. (1999) experimental repeated measure designs typically call for a dependent groups t test or a repeated measures ANOVA because the same people are being measured more than once after being randomly assigned to different treatments.

Additional analysis will include multiple regression analysis on all outcome variables of interest to identify any correlations between the primary independent variable (oral EPO) and other outcome characteristics, such as gestational age at the time of delivery, incidence of postdates induction, amniotic fluid status, length of rupture, length of labor, type of delivery, incidence of abnormal labor patterns, infant weight, Apgar Scoring, and NICU admissions. The correlation coefficient statistic can show the strength of the relationship between variables of interest (Polit et al., 1999). The closer the number is to zero, the weaker the relationship, the closer it is to one, the stronger the relationship, however correlation should not be confused with causality (Polit, et al., 1999). All data will be entered in the SSPS (Statistical Software for Social Sciences) program for Windows for data computation.

Plan for Disseminating Findings
Evening primrose oil remains one of the most commonly midwife recommended non-pharmaceutical agent for cervical ripening in the United States. This author proposes that results of this research will have an impact on making a common midwifery practice evidence based for the future. This researcher plans to submit her research findings for publication in the Journal of Midwifery and Women’s Health because it is the premier research journal available to midwives in the United States, and the information can be widely disseminated in an efficient manner. In addition, this researcher proposes to present her findings at an annual conference of the American College of Nurse-Midwives, as this is another efficient way to share the findings with a large audience of midwives at one time.
References
American College of Nurse-Midwives (ACNM) (2002, December 12). Basic Facts about certified nurse-midwives. Retrieved July 23, 2003 from the World Wide Web: http://www.midwife.org/prof/display.cfm?id=6
American College of Nurse-Midwives (ACNM) (2003). ACNM philosophy. Retrieved July 23, 2003 from the World Wide Web: http://www.midwife.org/prof/print.cfm?id=132
American College of Nurse-Midwives (ACNM) (2003). Code of Ethics. Retrieved July 23, 2003 from the World Wide Web: http://www.midwife.org/prof/print.cfm?id=136
Barrilleaux, P, Bofill, J., Terrone, D., Magann, E., May, L., & Morrison J. (2002). Cervical ripening and induction of labor with misoprostol, dinoprostone gel, and a Foley catheter: a randomized trial of 3 techniques. American Journal of Obstetrics and Gynecology, 186(6), 1124-1129.
Belew, C. (1999). Herbs for the childbearing woman: guidelines for midwives. Journal of Nurse-Midwifery, 44(3), 231-252.
Birch, A., Fenner, G., Watkins, R., & Boyd, L. (2001). Antioxidant properties of evening primrose seed extracts. Journal of Agriculture and Food Chemistry, 49(9), 4502-4507.
Blackburn, S. & Loper, D. (1992). Maternal, fetal, and neonatal physiology. Philadelphia: W.B. Saunders Company.
Chenoy, R., Hussain, S., Tayob, Y., & Moss, M. (1994). Effect of oral gamolenic acid from evening primrose oil on menopausal flushing. British Medical Journal , 308(6927), 501-506.
Cohen, S., Rousseau, M., & Robinson, E. (2000). Therapeutic use of selected herbs. Holistic Nursing, 14(3), 59-68.
Cunningham, F., Gant, N., Leveno, K., Gilstrap, L., Hauth, J., & Wenstrom, K. (2001). Williams obstetrics (21st ed.). New York: McGraw Hill.
D'Almeida, A., Carter, J., Anatol, A., & Prost, C. (1992). Effects of combination of evening primrose oil (gamma linolenic acid) and fish oil (eicosapentaenoic (plus) docahexaenoic acid) versus magnesium, and versus placebo in preventing pre-eclampsia. Women and Health, 19(2-3), 117.
Damania, K., Natu, U., Mhatre, P., Mataliya, M., Mehta, A., & Daftary, S. (1992). Evaluation of two methods employed for cervical ripening. Journal of Postgraduate Medicine, 38(2), 58-59.
Davis, L. (1984). The use of castor oil to stimulate labor in patients with premature rupture of membranes. Journal of Nurse-Midwifery, 29(6), 366-370.
Dove, D & Johnson, P (1999). Oral evening primrose oil: its effect on length of pregnancy and selected intrapartum outcomes in low-risk nulliparous women. Journal of Nurse-Midwifery, 44(3), 320-324.
Harman, J. & Kim, A. (1999). Current trends in cervical ripening and labor induction. American Family Physician, 60(2), 47-56.
Kennedy, H., & Lowe, N. (2001). Science and midwifery: paradigms and paradoxes. Journal of Midwifery & Women's Health, 46(2), 91-97.
Khoo, S., Munro, C., & Battistutta, D. (1990). Evening primrose oil and treatment of premenstrual syndrome. Medical Journal Australia, 153, 189-192.
Kleijnen, J. (1994). Evening primrose oil. British Medical Journal, 309(6958), 824-828.
Kramer, R. (1997). A randomized trial of misoprostol and oxytocin for induction of labor: safety and efficacy. Obstetrics and Gynecology, 89, 387-391.
Magann, E., McNamara, M., Whitworth, N., Chauhan, S., Thorpe, R. & Morrison, J. (1998). Can we decrease postdatism in women with an unfavorable cervix and a negative fetal fibronectin test result at term by serial membrane sweeping? American Journal of Obstetrics and Gynecology, 179(4), 890-894.
McFarlin, B., Gibson, M., O'Rear, J., & Harman, P. (1999). A national survey of herbal preparation use by nurse-midwives for labor stimulation. Journal of Nurse-Midwifery, 44(3), 205-216.
Murphy, P., Kronenberg, F., & Wade, C. (1999). Complementary and alternative medicine in women's health: developing a research agenda. Journal of Midwifery and Women's Health, 44(3), 325.
National Center for Health Statistics (NCHS) (2003, June 25). US birth rate reaches record low. Retrieved July 23, 2003 from the World Wide Web: http://www.cdc.gov/nchs/releases/03news/lowbirth.htm.
Polit, D., Beck, C., & Hungler, B. (2001). Essentials of nursing research: methods, appraisal, and utilization (5th ed., Rev.). Philadelphia: Lippincott, Williams, & Wilkins.
Salisbury Nursing School (2003). Evening primrose oil. Retrieved June 30, 2003 from the World Wide Web: http://www.salisbury.edu/Schools/Henson/NursingDept/Herbalremedies/evening_primrose.html
Sciscione, A., McCullough, H., Manley, J., Shlossman, P., Pollock, M., & Colmorgen, G. (1999). A prospective, randomized comparison of Foley catheter insertion versus intracervical prostaglandin E2 gel for preinduction cervical ripening. American Journal of Obstetrics and Gynecology, 180(1).
Tenore, J. (2003). Methods for cervical ripening and induction of labor. American Family Physician, 67, 2123-2128.
Varney, H. (1997). Varney's Midwifery (3rd ed.). Sudbury, MA: Jones and Bartlett Publishers.
Wettasinghe, M., Shahidi, F., & Amarowicz, R. (2002). Identification and quantification of low molecular phenolic antioxidants in seeds of evening primrose (Oenothera biennis L.). Journal of Agriculture and Food Chemistry, 50(5), 1267-1271.
World Health Organization (1995). Revised 1995 estimates of maternal mortality. Retrieved October 19, 2003 from the World Wide Web: http://www3.who.int/whosis/mm/mm_country_1995.cfm?path=whosis,mm,mm_country...

Appendix A Consent Form
I, __________________________, consent to participate in research entitled, “Evening Primrose Oil as a Cervical Ripening Agent in Low Risk Nulliparous Women” conducted by Paula Senner CNM, MS.

I understand the purpose of this study is to identify a relationship between the use of oral evening primrose oil capsules and increased cervical ripening, as well as a decrease in the incidence of postdates pregnancies. I understand that I will be randomly assigned between 35 and 36 weeks gestation to either the experimental group that will receive oral evening primrose oil capsules in a set prescribed dose or to the control group that will receive an oral placebo in the same dosing regimen as the experimental group. Both groups will begin the dosing regimen at 37 weeks gestation to be continued through delivery. I understand that I will be asked to keep a record of the administration of the herb/placebo, and my data will continue to be collected, but will be excluded from the final study results if I miss taking more than 5 doses during the course of the study. I also understand that although midwives have been using oral evening primrose oil as a cervical ripening agent for a long time, there are no conclusive scientific studies at this time that have identified the efficacy of this method of cervical ripening. However, multiple research trials have demonstrated its safety. The most commonly reported side effects to EPO and placebo administration are stomach upset, softening of the stools, or headache.

By participating in this study, I am contributing to important research into the use of herbals in pregnancy. I understand that herbals are not regulated by the FDA, therefore it is difficult to know their exact strength and potential actions. I understand that data collected in this study will be prepared for publication in midwifery journals and will also be presented at a national midwifery conference, however all identifying information will be excluded. All identifying information will be held confidential under lock and key.

I acknowledge that I have had the opportunity to obtain additional information about this study, and that my questions have been answered to my full satisfaction. I understand that I may withdraw from this study at any point in time without penalty.

I fully understand that by voluntarily signing this form, I am consenting to participation in this study and giving permission to the researchers to use my personal and demographic information as it relates to the study objectives. I also understand that this research proposal has been approved by the IRB at Philadelphia University, as well as the Ethics committee of Loudoun Hospital Center. Thank you for participating in this study.

Subject Signature:____________________ Subject Number:________ Date:____________
Witness Signature:____________________ Date:___________

Appendix B Feasibility
This study appears to be logistically very feasible, as a site will not need to be recruited and a population of low risk pregnant nulliparous women will be readily available as a research pool to this research midwife. However, the study does carry substantial costs in time, as the study will need to be conducted over a period greater than two years in order to obtain a sample large enough to produce valid and reliable results. Funding will be sought from various women’s health research organizations and the Federal government in order to help defray budget costs (See Appendix C Budget).

Appendix C Budget
Forms, copying, copy supplies, etc.: $200
Research Assistant: $10/hr x 10 hours/week x 130 weeks= $13,000
Research Midwife: $30/hr x 8 hours/week x 130 weeks=$31,200
Statistical Software SPSS+= $300
EPO 500mg capsules (125 women x $10/bottle of 120)= $1250
Paraffin Capsules (125 women x $5/bottle of 120)= $625
Total Budget: $46,575

Appendix D Timeline
Philadelphia University IRB approval: 1 month
Loudoun Hospital Center Ethics Committee approval: 1 month
Site Recruitment: 0 month
Data Collection: 2+ years
Data Entry, Cleaning, & Analysis: 6 months
Dissemination of Findings: 6 months
Total Timeline: 3.5 years

Appendix E Demographic Information
Subject Number:________________
Date:_________________________
Age:__________________________
Total Years of Education:_________
Ethnicity:______________________
Marital Status:__________________
Employment Status:______________
Average Household Income:________
Gravida:________________________
Parity: 0
Hx Miscarriage:__________________
Hx Abortion:_____________________
Data Collection
If still pregnant:
Current Gestational Age:____________
Bishop Score:_____________________
Labor Record:
Gestational Age at Delivery:__________
Rupture of Membranes:
Spontaneous
Artificial
Length of Rupture:__________________
Color of Fluid:_____________________
Postdates Induction
Yes
No
Use of oxytocin:
Induction
Augmentation
Use of other Labor augmentation tools:
Castor Oil
Nipple Stimulation
Total Length of Labor:__________________
Incidence of Abnormal Labor Patterns:
Prolonged latent phase
Prolonged active phase
Secondary arrest of dilation
Arrest of Descent
Delivery Type:
Vaginal
Operative:
Vacuum
Forceps
Caesarean section
Apgar Scores:
One minute_________
Five minutes________
NICU Admission:
Yes
No

Appendix F Sample Diary
DateTime Dose
#1
Time Dose
#2
Time Dose
#3
Side Effects/
Unusual Symptoms
Please indicate confounding activities, such as intercourse, nipple
stimulation, use of other herbals, or castor oil.
. . . . . . .