Melissa A. Dalton-Richardson and Dr. Lynn Callister, Nursing
Within recent years, there has been a steady increase of births attended by midwives in the United States. The reemergence of midwives to the medical scene provides pregnant women with alternative methods of birth. In order to provide women with the most information possible about these two options of maternity care, there has been a good deal research done in the area of comparisons between obstetricians and certified nurse-midwives (CNM). Thus far, the research has done much in demarcating the differences in treatment and satisfaction of pregnant women. My research question took the next step in the research comparison. I attempted to discover if the differences in interventions of these two medical fields somehow affect a woman’s perceptions of self. My hypothesis was that those women in the CNM group would have higher perceptions of self-mastery and control than those in the obstetric group.
I ran a pilot study in which I hoped to sample 10 women in each care group: 10 from physician assisted births and 10 from nurse-midwife-assisted births). I obtained the name, city of residence and date of most recent birth for a sample of women in Utah County who had children sometime between three- to four-months previous to receiving the questionnaire. Before mailing out the survey, I phoned each respondent and asked them a series of questions that would enable me to control for such variables as yearly income, insurance availability, and level of risk in the childbirth. I mailed out 23 surveys (10 CNM, 13 OB/GYN). The data collected was measured according to two research tools: 1) the Utah Test of the Childbearing Year and 2) the Labor Agentry Scale (1,2).
For the Labor Agentry Scale, each subject had been given 29 items (14 positively- and 15 negatively-worded) listed in a random sequence. Each item was situated above a seven-step scale and given written instructions to mark the spot that best described her feelings during her childbirth experience. In order to obtain scores for each subject, I reversed the scores on the positively worded items (1=7 or 3=5, etc.). I then went through and selected from the questions 11 statements that measured a woman’s perceptions of her abilities and control. For the Utah Test for the Childbearing Year, each subject was given 64 items, both positively- and negatively worded, listed in a random sequence. Each item was situated next to a forced selection of agree or disagree and given written instructions to pick the option closest to how they feel. From this 64-item list, I selected 21 questions that I used to measure a woman’s perceptions of personal control. Finally, I included three open-ended questions in which I asked about their individual perceptions and control since this most recent birth.
In response to the survey, 82.6% of the respondents replied (7 of the certified nurse-midwife users, and 12 of the obstetrician users). Unfortunately, I had to cut my findings for analysis back to 6 in the CNM group and 10 in the OB/GYN group because of three respondents who didn’t fit into various areas that I was controlling for.
My main interest was not the differences in frequency of medical intervention imposed by both caregivers, but I did look at the variable of epidurals between the two groups. Only 1 (16.7%) of the certified nurse midwives in my sample had an epidural, whereas 7 (70%) of my obstetrician group had an epidural. This replicates the findings of previous research (about 20.8% vs. 63.3%) focusing on intervention.
I attribute the high response rate from the first mailing to a few things. First of all, the topic of childbirth is something very personal to women. Interest and therefore, participation is evoked when you give them an opportunity to speak about an experience that they were involved with and involves their children. Second, the fact that I had used multiple modes of conducting my survey (both telephone and mail) would have increased the response rate as well.
Upon analysis of the findings, it would seem that using measurements of significance and correlation provided few significant findings. The problem here is derived directly from my sample size. A measurement of significance is directly correlated with the size of the sample. Thus, if we just look primarily at the statistical points, it would seem that there is little significant difference between certified-nurse midwives care and obstetrician care.
But, when you look at the graphs comparing the percent of women from each group rather than the mathematical outputs of significance and correlation, there is a pattern. With a larger sample the significance would be ascertainable and the correlation between them very apparent. In this case, the findings point to a correlation between nurse-midwife care and a higher perception of control and self-mastery.
One thing to take into consideration in analysis of these findings is that this was not a longitudinal study. Even though a larger sample might produce more statistically significant results, I could still not claim causation until I had interviewed the women previous to birth. As stated in Lynn Callister’s work on women’s perceptions previous to the choice of a caregiver, there are definitely personality differences between those women who pick nurse-midwives and those who pick obstetricians (3). Until a longitudinal study is completed which analyzes the women’s perceptions previous to childbirth, I cannot make a statement of causation even with a larger sample size.