Abstract
Background
There is limited documentation of non-medical methods of labor induction and pain management during childbirth in the U.S. We estimated the prevalence of non-medical interventions for induction and pain management and examined the association between medical and non-medical care during labor.
Methods
We used a nationally-representative survey of U.S. women who gave birth in 2005 (N=1,382) to examine use of non-medical methods of labor induction and pain management. Using logistic regression, we calculated odds of non-medical and medical interventions to induce labor or mitigate pain, and the odds of medical induction and obstetric analgesia by whether non-medical methods were used.
Results
Nearly 30% of women used non-medical methods to start labor, and over 70% of women used non-medical pain management. Doula support was the strongest predictor of non-medical methods of labor induction (Adjusted Odds Ratio (AOR) = 3.0) and labor pain management (AOR = 5.7). Use of non-medical pain management was significantly associated with decreased odds of medical pain management (OR = 0.65); this relationship was attenuated with covariate adjustment.
Conclusions
Non-medical methods to induce labor and manage pain during childbirth are commonly used by U.S. women. Future research should examine effectiveness of these strategies and their influence on medical services use.
Keywords: complementary and alternative therapies, childbirth, obstetrics, labor induction, labor pain management
Introduction
Childbirth is a common reason for use of medical services by U.S. women (1). Both induction of labor and medical management of labor pain are increasingly common interventions that offer important potential benefits to childbearing women, but they also carry risks. Medical induction of labor has increased dramatically, from 9.5% in 1990 to 23.2% in 2011, and approximately 75% of all women receive neuraxial analgesia (epidural or intrathecal) during labor to manage pain during childbirth (2–4). Clinicians can induce labor by rupturing the amniotic sac, applying synthetic prostaglandin gel to the cervix, or giving synthetic oxytocin (Pitocin) intravenously (5). Labor induction is an important and sometimes necessary procedure, but when used without medical indication, especially before 39 weeks gestation, it can have adverse consequences. The rise in non-indicated labor induction is associated with decreasing mean birth weight (6); labor induction is also associated with increased odds of cesarean delivery (7), which, while often performed to avoid potential adverse events, also carries maternal and infant health risks (8, 9). Neuraxial analgesia is generally effective in managing labor pain, but may increase the chances of instrumental delivery, prolonged second stage of labor, need for labor augmentation, severe headache, maternal hypotension, maternal fever, and urinary retention (10–13).
Women of childbearing age are also frequent users of alternative health strategies and complementary and alternative medicines (CAM), the use of which has been steadily increasing in the United States, and currently comprises about 3% of national health expenditures for ambulatory care (14). CAM use is associated with being female, aged 30–65, higher levels of income and education, private (vs. public) insurance, and living in the West census region (15).
While the use of medical care during labor is well-documented, less is understood about the use of non-medical means of labor induction and pain management in the U.S. population. There is a lack of national data on how many women try to start their own labor, yet the results of small studies suggest that women frequently attempt “self-induction.” Even so, evidence on the efficacy of these non-medical techniques for starting labor is sparse, with small sample sizes and limited generalizability (16–18). Studies have had mixed or inconclusive results regarding the effectiveness of sexual intercourse for labor induction (17, 19–21). The evidence for nipple stimulation is tentatively positive (22). A systematic review of the use of castor oil found only one study that was of sufficient quality to include, and results showed it to be ineffective at starting labor (23); and a case report actually suggests potential harms of castor oil ingestion (24). In addition, early labor, iatrogenic birth, and delivery prior to 39 weeks gestation entails maternal and neonatal risks, whether labor is self-induced or medically induced without a clear clinical need (8, 9).
While non-medical pain management techniques during labor are commonly discussed in childbirth preparation books, manuals and courses (25), no nationally-representative studies provide information on the prevalence of their use. Effectiveness studies have generally evaluated individual methods separately despite the fact that multiple techniques are likely used together (26). Some evidence supports the effectiveness of immersion in water, acupuncture, acupressure, massage, and relaxation techniques in labor (10, 27, 28). The use of hypnosis did not affect the probability of using pharmacologic pain relief or having a spontaneous vaginal birth (29). A Cochrane review on this topic noted the difficulty in comparing studies due to variation in process and outcome measures, which can include outcomes such as length of labor, women’s reports of pain relief and satisfaction with pain relief, use of pharmacologic pain management, and spontaneous vaginal birth (10).
Having the support of a trained labor companion may also reduce a woman’s need for medical pain management in labor. A doula is a non-medical professional who “provides continuous physical, emotional and informational support to the mother before, during and just after birth; or who provides emotional and practical support during the postpartum period” (30). Randomized controlled trials have shown the clinical benefits of continuous labor support to include shorter labors, higher reported levels of patient satisfaction, higher rates of spontaneous vaginal birth, lower rates of cesarean delivery and instrument-assisted vaginal delivery, and lower rates of regional (i.e. epidural) analgesia (31).
In this study, we characterized the use of non-medical care to induce labor and manage pain during childbirth in the U.S. We further report the prevalence and types of non-medical methods of labor induction and pain management, maternal characteristics associated with non-medical methods, and the relationship between use of medical and non-medical methods.
Materials and Methods
Data
Data came from the Listening to Mothers II Survey, a survey of English-speaking women, ages 18–45, who gave birth to a singleton infant in a U.S. hospital during 2005 (N = 1,573). The survey was conducted by Harris Interactive via internet and telephone, using validated sampling methods (32, 33). Full information about the survey questionnaires, methodology, reports and related materials are publicly available (34). Data from the Listening to Mothers surveys have previously been used in public health and maternal health services research (4, 35–37), but have never before been used to examine the use of complementary and alternative medicines or non-medical forms of health care related to labor induction and pain management during childbirth.
The Listening to Mothers II survey contains detailed data on medical and non-medical care used during pregnancy and childbirth as well as the reasons for use of this care and the self-reported effectiveness. This analysis uses the online component of the survey, in which the use of methods of labor induction and pain management was assessed (n=1,382). We excluded women with scheduled cesarean deliveries (n=224) and those who did not experience labor (n=31). The final analytic sample included 1,127 women who gave birth in a U.S. hospital during 2005 (weighted N=1,087).
Variable measurement
Labor induction
We used two measures of labor induction: self-induction and medical induction. Women were asked whether they themselves had done anything to start their labor (self-induction). Women who responded that they had tried to start their labor were asked whether they had used nipple stimulation, sexual intercourse, castor oil, herbal treatments, walking or exercise, or other methods. A second set of questions asked whether a medical provider had done something to start labor; if so, each was asked whether the provider had ruptured her membranes, swept her membranes, used Pitocin, cervical gel, an oral tablet or pill, or some other method. For both types of induction, respondents could indicate more than one method that was used. We categorized respondents as having used self-induction only, medical induction only, both, or neither.
Pain management during labor
The survey also asked about use of non-medical and medical techniques for pain management during labor. Respondents indicated whether they had used each of the following non-medical techniques: tub or pool, shower, position changes, birth ball, hot or cold objects, mental strategies, environmental changes such as music or aroma, hands-on techniques, breathing techniques, or other techniques. Women were categorized as having used a non-medical pain management technique if they had used at least one of the listed options. Respondents also indicated whether each of the following pain medications had been used: epidural, narcotics, nitrous oxide, local block, general anesthesia, or other. They were categorized as having used a medical pain management technique if one or more pain medications had been used. We also created a variable that categorized women as having used non-medical pain management only, medical pain management only, both medical and non-medical pain management, or neither type of pain management.
Covariates
We included several covariates in our analyses. Socio-demographic measures included: age category (18–24 years, 25–29 years, 30–34 years, 35 years or older), race/ethnicity (white, black, Hispanic, or other/multiple race), marital status (married or not), education level (high school or less, some college, Bachelor’s degree, or graduate education), census region (Northwest, Midwest, South, West), and nativity (foreign born or US born). Pregnancy-related covariates were doula support during the birth, parity (experienced mother or first-time mother), whether the pregnancy was unintended, and whether the woman agreed with the statement that “birth is a process that should not be interfered with unless medically necessary.” Other covariates included the type of health insurance that paid for maternity care (private vs. Medicaid or uninsured) and self-reported health status (excellent, very good or good, vs. fair or poor),
Analysis
First we examined descriptive statistics for the sample, characterizing the overall prevalence of labor induction and pain management techniques. We then used multivariate logistic regression to identify predictors of having used any non-medical method of labor induction and any non-medical method of pain management. Finally, we estimated unadjusted and adjusted odds of any medical method of labor induction and any medical method of labor pain management by use of non-medical interventions using logistic regression. All analyses were conducted in Stata12 and weighted to be nationally representative.
Results
Sample Characteristics
Characteristics of the study population are reported in Table 1. Approximately 41% of women in the sample had a high school education or less; about 29% had a Bachelor’s or graduate degree. Over 70% of women were white, about 9% were black, and nearly 15% were Hispanic. Women in the sample were fairly evenly distributed across the younger age categories (18–24, 25–29 and 30–34), with 25 to 30% in each category. A smaller number (18.5%) were age 35 or older. Over 70% of the sample was married and about 5% was foreign-born. Nearly two thirds of women were experienced mothers (parity>1), and over 40% reported that the pregnancy resulting in their recent birth was unintended. About 3% reported receiving support from a doula during labor. Nearly half (47.1%) agreed with the statement that childbirth is a process that should not be interfered with unless medically necessary. Forty-two percent of women reported public coverage or no insurance coverage for their maternity care, while 58% of women were privately insured. The vast majority (92.6%) of women in the sample reported being in good health.
Table 1.
Characteristics of sample.
| % (n) | |
|---|---|
| Socio-demographic characteristics | |
| Age category | |
| 18–24 | 27.9 (304) |
| 25–29 | 29.4 (320) |
| 30–34 | 24.2 (263) |
| 35+ | 18.5 (201) |
| Race | |
| White | 71.3 (775) |
| Black | 9.3 (101) |
| Hispanic | 14.7 (156) |
| Other/multiple race | 4.7 (52) |
| Married (vs. not married) | 71.4 (777) |
| Education | |
| H.S. or less | 40.9 (445) |
| Some college/Associate's degree | 29.9 (325) |
| Bachelor's degree | 21.2 (231) |
| Graduate education/degree | 8.0 (87) |
| Region | |
| Northeast | 18.3 (199) |
| Midwest | 23.0 (250) |
| South | 34.2 (372) |
| West | 24.5 (266) |
| Foreign born (vs. US born) | 5.4 (59) |
| Pregnancy Characteristics | |
| Doula support | 2.9 (31) |
| Experienced mother (vs. first time mother) | 62.9 (684) |
| Unintended pregnancy (vs. intended pregnancy) | 41.9 (455) |
| Belief that childbirth is a process that should only be interfered with if medically necessary | 47.1 (512) |
| Other Characteristics | |
| Public or no insurance (vs. private insurance) | 41.5 (636) |
| Good health (vs. poor health) | 92.6 (1007) |
Note: Table presents weighted Ns and percentages.
Labor Induction
Table 2 reports the prevalence of labor induction strategies among women in the study population. Nearly 30% of women had used some non-medical intervention to start their labor; over 50% had a provider attempt to induce their labor. Twenty-three percent of women in the sample used walking or exercise to try to start labor; this was the most common self-induction technique. A fifth of the sample had sexual intercourse to try to start labor. Nipple stimulation to induce labor was also fairly common, used by nearly 12% of women. The most common medical induction technique was use of Pitocin, used by 41% of all women in the study population. Rupturing the membranes or sweeping the membranes was also common, reported by 25% and 17% of women, respectively. About a third of women in the sample used medical interventions only to try to start labor; 10% used non-medical interventions only. Nearly a fifth of women used both medical and non-medical techniques to try to start labor, and about 39% did nothing to start labor.
Table 2.
Prevalence and type of non-medical and medical interventions for induction of labor and labor pain relief.
| % (n) | |
|---|---|
| Induction of labor | |
| Any non-medical interventions (%) | 28.6 (311) |
| Nipple stimulation | 11.9 (129) |
| Sexual intercourse | 20.3 (220) |
| Castor oil | 3.8 (41) |
| Herbal treatment | 4.1 (45) |
| Walk, exercise, movement | 23.4 (255) |
| Other | 2.1 (23) |
| Provider Induction (%) | 51.3 (557) |
| Rupture membranes/break water | 25.4 (276) |
| Sweep mebranes | 17.1 (186) |
| Pitocin | 41.1 (447) |
| Cervical gel | 12.2 (133) |
| Tablet/pill (oral) | 1.7 (19) |
| Other | 1.1 (12) |
| Medical interventions only | 32.7 (355) |
| Non-medical interventions only | 10.0 (109) |
| Both medical and non-medical interventions | 18.6 (202) |
| Neither medical nor non-medical interventions | 38.7 (421) |
| Labor pain relief | |
| Any non-medical interventions (%) | 70.5 (766) |
| Tub/pool | 6.2 (67) |
| Shower | 4.4 (48) |
| Position changes | 42.9 (467) |
| Birth ball | 6.9 (75) |
| Hot or cold objects | 6.5 (70) |
| Mental strategies | 25.4 (276) |
| Environmental changes (music, smells) | 3.7 (41) |
| Hands-on techniques | 21.0 (228) |
| Breathing techniques | 51.0 (555) |
| Other | 2.2 (23) |
| Any medical interventions (%) | 84.7 (921) |
| Epidural | 75.3 (819) |
| Narcotics | 25 (272) |
| Nitrous oxide | 1.5 (16) |
| Local block | 0.8 (9) |
| General anesthesia | 2.2 (24) |
| Other | 7.6 (82) |
| Medical interventions only | 26.1 (284) |
| Non-medical interventions only | 11.9 (129) |
| Both medical and non-medical interventions | 58.6 (637) |
| Neither medical nor non-medical interventions | 3.4 (37) |
| Used more than one non-medical intervention | 50.1 (545) |
Note: Table presents weighted Ns and percentages. Women who had a planned cesarean or did not experience labor are excluded. Categories are not mutually exclusive and may sum to more than 100%
Logistic regression results for use of non-medical interventions for labor induction are shown in Table 3. Women with doula support had substantially higher odds of using non-medical techniques for labor induction (Adjusted Odds Ratio (AOR) = 3.03, 95% CI [1.21–7.61], as did experienced mothers (vs. first-time mothers) (AOR = 1.91, 95% CI [1.28–2.86]) and women in good health (AOR = 2.34, 95% CI [1.07, 5.11]). Women over age 35 had lower odds of using non-medical labor induction techniques compared to women age 18–24 (AOR = 0.24, 95% CI [0.12, 0.50]). The unadjusted and adjusted odds of use of medical interventions for labor induction are also presented in Table 3. In the unadjusted model, use of non-medical labor induction methods was positively associated with use of medical labor induction methods (Odds Ratio (OR) = 2.20, 95% CI [1.51, 3.21]). After adjusting for covariates, the association between medical and non-medical methods of labor induction remained positive and statistically significant (AOR = 2.45, 95% CI [1.66, 3.62]). Women over age 35 (vs. 18–24) had higher odds of using medical labor induction methods (AOR = 1.81, 95% CI [0.99, 3.31]), although this finding was only marginally significant. Belief that birth is a process that should not be interfered with was negatively associated with medical labor induction (AOR = 0.67 [95% CI [0.48, 0.95]).
Table 3.
Odds of use of non-medical and medical interventions for labor induction.
| Any non-medical method of labor induction |
Bivariate: Any medical method of labor induction |
Multivariate: any medical method of labor induction |
||||
|---|---|---|---|---|---|---|
| AOR | 95% CI | OR | 95% CI | AOR | 95% CI | |
| Any non-medical labor induction method | 2.20 | (1.51, 3.21) | 2.45 | (1.66, 3.62) | ||
| Socio-Demographic Characteristics | ||||||
| Age category (Ref=18–24) | ||||||
| 25–29 | 0.78 | (0.47,1.29) | 1.42 | (0.88, 2.28) | ||
| 30–34 | 0.85 | (0.47,1.52) | 1.53 | (0.91, 2.56) | ||
| 35+ | 0.24 | (0.12,0.50) | 1.81 | (0.99, 3.31) | ||
| Race (Ref=white) | ||||||
| Black | 0.51 | (0.25,1.05) | 0.99 | (0.53, 1.83) | ||
| Hispanic | 0.86 | (0.44,1.67) | 0.73 | (0.40, 1.32) | ||
| Other/multiple race | 1.11 | (0.51,2.4) | 0.99 | (0.49, 2.00) | ||
| Married | 0.99 | (0.56,1.76) | 0.85 | (0.54, 1.36) | ||
| Education (Ref=H.S. or less) | ||||||
| Some college/Associate's degree | 0.95 | (0.61,1.49) | 0.78 | (0.52, 1.15) | ||
| Bachelor's degree | 1.13 | (0.64,1.97) | 0.67 | (0.39, 1.14) | ||
| Graduate education/degree | 1.45 | (0.74,2.83) | 0.86 | (0.47, 1.59) | ||
| Region (Ref=Northeast) | ||||||
| Midwest | 0.80 | (0.46,1.4) | 1.25 | (0.77, 2.22) | ||
| South | 1.07 | (0.62,1.84) | 1.25 | (0.74, 2.09) | ||
| West | 1.00 | (0.55,1.8) | 0.86 | (0.47, 1.59) | ||
| Foreign born | 0.76 | (0.28,2.08) | 0.67 | (0.31, 1.44) | ||
| Pregnancy Characteristics | ||||||
| Doula support | 3.03 | (1.21,7.61) | 0.60 | (0.22, 1.68) | ||
| Experienced mother (Ref=first time mother) | 1.91 | (1.28,2.86) | 0.73 | (0.50, 1.06) | ||
| Unintended pregnancy | 0.98 | (0.66,1.46) | 0.97 | (0.67, 1.42) | ||
| Belief that childbirth is a natural process | 0.80 | (0.55,1.17) | 0.67 | (0.48, 0.95) | ||
| Other Characteristics | ||||||
| Public or no insurance (Ref=private insurance) | 0.87 | (0.54,1.4) | 0.86 | (0.56, 1.31) | ||
| Good health (Ref=poor health) | 2.34 | (1.07,5.11) | 0.65 | (0.35, 1.21) | ||
Note: Models are weighted to be nationally representative. Women with planned cesareans and women who did not experience labor are excluded.
Pain management
The uses of both medical and non-medical strategies for managing pain during labor were very common (see Table 2). Over 70% of women used some non-medical pain management technique, and 85% used medical pain management during labor. Breathing techniques, position changes, and mental strategies were the most frequently-reported non-medical pain management techniques, with over half of women using breathing techniques. Epidural analgesia was by far the most common medical intervention for management of labor pain, used by three-quarters of women in the sample. Narcotics were used by 25% of women. Most women (58.6%) used both medical and non-medical interventions to manage their labor pain. About a quarter used medical interventions only, 12% used non-medical interventions only, and 3.4% reported no medical or non-medical pain management interventions. Half of women reported using multiple non-medical pain management strategies during labor.
Table 4 presents logistic regression results for use of non-medical pain management strategies. Similar to labor induction, doula support was the most important predictor of using non-medical pain management techniques (AOR=5.74, 95% CI [1.26–26.13]). Other predictors of non-medical pain management included census region (women in the West had twice the odds compared to women in the Northeast) and belief that childbirth is a process that should not be interfered with (AOR=1.83, 95% CI [1.27–2.64]). Table 4 also shows unadjusted and adjusted odds of use of medical pain management techniques. In the bivariate model, use of non-medical pain management methods was associated with decreased odds of use of medical pain management (OR = 0.65, 95% CI [0.42, 1.00]). After controlling for covariates, the negative association remained but was no longer statistically significant (AOR = 0.68, 95% CI [0.42, 1.11]). Being in the West census region (vs. the Northeast) was associated with higher odds of using medical labor pain management methods. Some college education (vs. high school or less), being married, being an experienced mother (vs. new mother), and believing that birth is a process not to be interfered with were all negatively associated with use of medical pain management.
Table 4.
Odds of use of non-medical and medical interventions for labor pain relief.
| Any non-medical method of labor pain management |
Bivariate: any medical method of labor pain management |
Multivariate: any medical method of labor pain management |
||||
|---|---|---|---|---|---|---|
| AOR | 95% CI | OR | 95% CI | AOR | 95% CI | |
| Any non-medical pain management method | 0.65 | (0.42, 1.00) | 0.68 | (0.42, 1.11) | ||
| Socio-Demographic Characteristics | ||||||
| Age category (Ref=18–24) | ||||||
| 25–29 | 1.02 | (0.62,1.67) | 0.76 | (0.40, 1.40) | ||
| 30–34 | 1.00 | (0.57,1.74) | 0.91 | (0.48, 1.72) | ||
| 35+ | 1.24 | (0.67,2.31) | 0.89 | (0.41, 1.93) | ||
| Race (Ref=white) | ||||||
| Black | 1.07 | (0.56,2.01) | 1.75 | (0.80, 3.81) | ||
| Hispanic | 1.14 | (0.6,2.14) | 0.93 | (0.43, 2.04) | ||
| Other/multiple race | 1.18 | (0.48,2.88) | 0.58 | (0.18, 1.87) | ||
| Married | 0.95 | (0.59,1.53) | 0.37 | (0.20, 0.69) | ||
| Education (Ref=H.S. or less) | ||||||
| Some college/Associate's degree | 1.38 | (0.92,2.09) | 0.47 | (0.27, 0.81) | ||
| Bachelor's degree | 1.05 | (0.60,1.86) | 0.53 | (0.26, 1.09) | ||
| Graduate education/degree | 1.47 | (0.72,2.99) | 0.61 | (0.26, 1.44) | ||
| Region (Ref=Northeast) | ||||||
| Midwest | 1.41 | (0.81,2.44) | 1.23 | (0.66, 2.26) | ||
| South | 0.97 | (0.57,1.66) | 1.79 | (0.97, 3.29) | ||
| West | 1.96 | (1.06,3.66) | 2.36 | (1.20, 4.65) | ||
| Foreign born | 0.47 | (0.19,1.14) | 0.78 | (0.30, 2.05) | ||
| Pregnancy Characteristics | ||||||
| Doula support | 5.74 | (1.26,26.13) | 0.69 | (0.27, 1.78) | ||
| Experienced mother (Ref=first time mother) | 0.85 | (0.57,1.28) | 0.30 | (0.18, 0.50) | ||
| Unintended pregnancy | 0.76 | (0.52,1.12) | 0.92 | (0.59, 1.44) | ||
| Belief in birth as a natural process | 1.83 | (1.27,2.64) | 0.37 | (0.23, 0.57) | ||
| Other Characteristics | ||||||
| Public or no insurance (Ref=private insurance) | 0.80 | (0.51,1.25) | 0.63 | (0.39, 1.01) | ||
| Good health (Ref=poor health) | 1.22 | (0.64,2.34) | 0.59 | (0.24, 1.48) | ||
Note: Models are weighted to be nationally representative. Women with planned cesareans and women who did not experience labor are excluded.
Discussion
Non-medical methods of labor induction and pain management are used nearly as frequently as medical methods among U.S. women who give birth. Almost a third of women used some non-medical intervention to start their labor, and half had a provider attempt to induce their labor. Women who used non-medical techniques for self-induction of labor also had higher odds of medical induction, perhaps owing to women’s discomfort or desire to hasten labor at the end of pregnancy or preference to avoid a medical induction by using techniques to encourage labor to begin spontaneously.
Over 70% of women used some non-medical pain management technique, and 85% used medical pain management during labor. Breathing techniques, position changes, and mental strategies were the most frequently-reported non-medical pain management techniques, with over half of women using breathing techniques. Many of these techniques are commonly recommended by clinicians and have shown effectiveness in mitigating labor pain (10, 38). Epidural analgesia was by far the most common medical intervention for management of labor pain, used by three-quarters of women in the sample, similar to the rates reported in birth certificates (13).
Given the frequency with which epidural analgesia is used for labor pain management, it is important to consider the potential impact of widespread use on the frequency of known risks associated with the procedure, including instrument-assisted delivery, prolonged second stage of labor, need for labor augmentation, severe headache, maternal hypotension, and maternal fever. Medical management of labor is also associated with added costs: the average professional fee paid by commercial health insurers to anesthesiologists for intrapartum care is $995 per birth (39). Our analysis showed that – in unadjusted comparisons across the study population - use of non-medical pain management methods was associated with decreased odds of use of medical pain management, of which epidurals comprise the vast majority. For women with a preference for non-medical methods of pain management, there may be potential cost savings and improved clinical outcomes when clinicians, health insurers or employers help enhance access to these methods. However, this relationship between medical and non-medical methods of pain relief may vary across subgroups of women or regions of the U.S.
Although doula support was a relatively low-prevalence characteristic in our study population, it was the strongest predictor of using non-medical methods of labor induction and labor pain management. Women with doula support had triple the odds of reporting non-medical techniques for labor induction, compared with women without such support. In addition, doula support was the most important predictor of use of non-medical pain management techniques, (AOR=5.74, 95% CI [1.26–26.13]). Our finding is consistent with results reported in a Cochrane review of continuous labor support, showing lower rates of epidural use for women with support from a trained childbirth professional (31). Use of doula services has increased in recent years, but is limited by financial and cultural barriers to access (40, 41). In addition to facilitating use of non-medical methods of labor induction and pain management, continuous labor support is associated with many other positive birth outcomes, so policies to increase access to doula care should be considered (31, 41). It may be useful for payers to consider benefits design strategies that allow for a range of choices for labor pain management.
Other significant predictors of non-medical pain management techniques included parity, U.S. Census region, and belief that childbirth is a process that does not require intervention. Regional patterns uncovered in our analysis are consistent with broad patterns of CAM use across the U.S. (15), and may reflect geographic differences in practice patterns and clinical norms as well as variation in the practice of midwifery, which is a care model that focuses on physiologic childbirth and may place greater emphasis on non-medical methods and CAM techniques (42). The role of beliefs about childbirth is also important in our analysis, with women who believe that childbirth does not require intervention having nearly double the odds of reporting non-medical means of pain management. This finding highlights the need to account for differences in preferences and attitudes toward pain relief during labor and delivery in order to better understand and address women’s needs (13, 43). Prior work has not explicitly addressed the relationship between women’s beliefs about childbirth and use of alternative health strategies during labor and delivery, and this analysis lays the groundwork for future research which could explore predictors of a belief in childbirth as a process that should not be interfered with and the relationship between these beliefs and quality of maternity care and health outcomes.
Childbearing women and their healthcare providers require more information about the benefits (and in some cases, potential risks) of non-medical methods of labor induction and pain management. In addition, more data are needed about the clinical and practical interactions between medical care and alternative health strategies used during labor and delivery. Whether clinicians are aware of women’s use of CAM techniques or non-medical strategies may impact the course of care during pregnancy and childbirth, and evidence is needed to guide this shared decision-making process (44). While birth certificates and hospital discharge records include information on medical procedures during childbirth, there is no routinely-collected information about non-medical methods of pain relief or about efforts women make to induce labor on their own (45).
Findings should be considered in light of potential limitations. First, our primary outcomes of interest are all based on self-report. When collecting data via surveys, there is potential for recall bias, social desirability bias, and other context effects that may influence the accuracy of self-reports (46). However, previous research comparing recall to medical record documentation suggests that women have relatively high accuracy of recall regarding events that take place during pregnancy and childbirth (47). Second, we classified all non-medical methods in one category for each outcome related to labor induction or labor pain management. These are heterogeneous groups of non-medical methods, each of which may have more or less effectiveness for the desired outcome. Grouping them together may have resulted in an attenuation of the relation between non-medical methods and decreased odds of medical management. Additionally, there is no detail about frequency, duration, or intensity of the non-medical interventions. To the extent that there is a dose-response relation, this lack of detail may obscure any findings. Third, our examination of the association between non-medical and medical methods is limited by the all-or-nothing nature of the measures of medical intervention (i.e., either had it or not). Even if non-medical methods did not prevent the use of medical interventions, it is possible that non-medical interventions might have delayed the use or decreased the amount of some of the medical methods used (e.g., narcotics, Pitocin). However, the available data do not include procedure codes, dosages, or timing of interventions and thus do not allow this more detailed analysis. Fourth, the survey is based on an online survey panel, which is a non-probability sample of those who have volunteered to participate in surveys and who have access to the internet. Thus, it may not be fully representative of all women giving birth in the US. However, evaluation of this online sample suggests that even though it may be a biased sample, it may produce more accurate estimates than other methods of national survey data collection (48). Finally, sample size was moderately small, especially with respect to some variables of interest (e.g., doula support). A larger overall sample, even with a relatively low prevalence exposure, like doula support, would allow more comprehensive analysis of this important facilitator of non-medical methods of labor induction and pain management.
Conclusions
Use of non-medical means of labor induction and labor pain management is very common, and it is important to understand the scope and effectiveness of use and to ensure that medical and non-medical means of care are well-integrated and mutually supportive. Clinicians should be aware of the needs and preferences of patients with respect to medical and non-medical methods of labor induction and pain management, and should encourage the use of trained labor support from a birth doula for those who prefer non-medical methods. Future research should examine effectiveness of alternative and non-medical strategies for induction and pain management and consider how they may influence the use of traditional medical care during labor for the benefit of mothers and infants.
Acknowledgments
The authors are grateful for helpful input provided by Judy Jou.
Funding Support
This research was supported by a grant from the Eunice Kennedy Shriver National Institutes of Child Health and Human Development (NICHD; grant number R03HD070868) and the Building Interdisciplinary Research Careers in Women’s Health Grant (grant number K12HD055887) from NICHD, the Office of Research on Women’s Health, and the National Institute on Aging, at the National Institutes of Health, administered by the University of Minnesota Deborah E. Powell Center for Women’s Health.
Contributor Information
Katy Backes Kozhimanil, Division of Health Policy and Management, University of Minnesota School of Public Health, Minneapolis, Minnesota.
Pamela Jo Johnson, Medica Research Institute and an Adjunct Assistant Professor in the Division of Epidemiology and Community Health, University of Minnesota School of Public Health, Minneapolis, Minnesota.
Laura B. Attanasio, Division of Health Policy and Management, University of Minnesota School of Public Health, Minneapolis, Minnesota.
Dwenda K. Gjerdingen, Department of Family Medicine and Community Health, University of Minnesota Medical School and University of Minnesota Physicians, Minneapolis, Minnesota.
Patricia M. McGovern, Division of Environmental Health Sciences, University of Minnesota School of Public Health, Minneapolis, Minnesota.
References
- 1.Cylus J, Hartman M, Washington B, Andrews K, Catlin A. Pronounced gender and age differences are evident in personal health care spending per person. Health Aff. 2011;30(1):153–160. doi: 10.1377/hlthaff.2010.0216. [DOI] [PubMed] [Google Scholar]
- 2.Berns SD, Kott A. Toward Improving the Outcome of Pregnancy III: Enhancing Perinatal Health Through Quality, Safety and Performance Initiatives. White Plains, NY: March of Dimes National Foundation; 2010. [Google Scholar]
- 3.Martin JA, Hamilton BE, Ventura S, et al. National Vital Statistics Reports. 1. Vol. 62. Hyattsville, MD: National Center for Health Statistics; 2013. Births: Final data for 2011. [PubMed] [Google Scholar]
- 4.Declercq ER, Sakala C, Corry MP, Applebaum S. Listening to Mothers II: Report of the Second National Survey of Women’s Childbearing Experiences. New York: Childbirth Connection; 2006. [Google Scholar]
- 5.ACOG Committee on Practice Bulletins - Obstetrics. ACOG Practice Bulletin No. 107: Induction of Labor. Obstet Gynecol. 2009;114(107):386–397. doi: 10.1097/AOG.0b013e3181b48ef5. [DOI] [PubMed] [Google Scholar]
- 6.Zhang X, Joseph KS, Kramer MS. Decreased term and posttermbirthweight in the United States: impact of labor induction. Am J Obstet Gynecol. 2010;203(2):124.e1–124.e7. doi: 10.1016/j.ajog.2010.03.044. [DOI] [PubMed] [Google Scholar]
- 7.Ehrenthal DB, Jiang X, Strobino DM. Labor induction and the risk of a cesarean delivery among nulliparous women at term. Obstet Gynecol. 2010;116(1):35–42. doi: 10.1097/AOG.0b013e3181e10c5c. [DOI] [PubMed] [Google Scholar]
- 8.Ecker J, Frigoletto F. Cesarean Delivery and the Risk-Benefit Calculus. N Engl J Med. 2007;356(9):885–888. doi: 10.1056/NEJMp068290. [DOI] [PubMed] [Google Scholar]
- 9.Zelop C, Heffner LJ. The downside of cesarean delivery: short- and long-term complications. Clin Obstet Gynecol. 2004;47(2):386–393. doi: 10.1097/00003081-200406000-00012. [DOI] [PubMed] [Google Scholar]
- 10.Jones L, Othman M, Dowswell T, et al. Pain management for women in labour: an overview of systematic reviews. Cochrane Database Syst Rev. 2012;(Issue 3) doi: 10.1002/14651858.CD009234.pub2. Art. No.: CD009234. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 11.Simkin P. Moving beyond the debate: a holistic approach to understanding and treating effects of neuraxial analgesia. Birth. 2012;39(4):327–332. doi: 10.1111/birt.12011. [DOI] [PubMed] [Google Scholar]
- 12.Anim-Somuah M, Smyth R, Jones L. Epidural versus non-epidural or no analgesia in labour. Cochrane Database Syst Rev. 2011;(Issue 12) doi: 10.1002/14651858.CD000331.pub3. Art. No.: CD000331. [DOI] [PubMed] [Google Scholar]
- 13.Osterman MJK, Martin JA. Epidural and Spinal Anesthesia Use During Labor: 27-state Reporting Area, 2008. Natl Vital Stat Rep. 2011;59(5):1–14. [PubMed] [Google Scholar]
- 14.Davis Ma, Martin BI, Coulter ID, Weeks WB. US Spending On Complementary And Alternative Medicine During 2002–08 Plateaued, Suggesting Role In Reformed Health System. Health Aff. 2013;32(1):45–52. doi: 10.1377/hlthaff.2011.0321. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 15.Barnes PM, Bloom B, Nahin RL. Complementary and alternative medicine use among adults and children: United States, 2007. Natl Health Stat Report. 2008;(12):1–23. [PubMed] [Google Scholar]
- 16.Chaudhry Z, Fischer J, Schaffir J. Women’s Use of Nonprescribed Methods to Induce Labor: A Brief Report. Birth. 2011;38(2):168–171. doi: 10.1111/j.1523-536X.2010.00465.x. [DOI] [PubMed] [Google Scholar]
- 17.Schaffir J. Sexual intercourse at term and onset of labor. Obstet Gynecol. 2006;107(6):1310–1314. doi: 10.1097/01.AOG.0000220507.91904.4a. [DOI] [PubMed] [Google Scholar]
- 18.Bayles BP. Herbal and other complementary medicine use by Texas midwives. J Midwifery Womens Health. 2007;52(5):473–478. doi: 10.1016/j.jmwh.2007.03.023. [DOI] [PubMed] [Google Scholar]
- 19.Kavanagh J, Kelly AJ, Thomas J. Sexual intercourse for cervical ripening and induction of labour. Cochrane Database Syst Rev. 2008;(Issue 4) doi: 10.1002/14651858.CD003093. Art. No.: CD003093. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 20.Tan PC, Andi A, Azmi N, Noraihan MN. Effect of coitus at term on length of gestation, induction of labor, and mode of delivery. Obstet Gynecol. 2006;108(1):134–140. doi: 10.1097/01.AOG.0000223229.83920.af. [DOI] [PubMed] [Google Scholar]
- 21.Tan PC, Yow CM, Omar SZ. Effect of Coital Activity on Onset of Labor in Women Scheduled for Labor Induction: A Randomized Controlled Trial. Obstet Gynecol. 2007;110(4):820–826. doi: 10.1097/01.AOG.0000267201.70965.ec. [DOI] [PubMed] [Google Scholar]
- 22.Kavanagh J, Kelly A, Thomas J. Breast stimulation for cervical ripening and induction of labour. Cochrane Database Syst Rev. 2010;(Issue 1) Art. No.: CD003392. [Google Scholar]
- 23.Kelly A, Kavanagh J, Thomas J. Castor oil, bath and/or enema for cervical priming and induction of labour. Cochrane Database Syst Rev. 2012;(Issue 5) doi: 10.1002/14651858.CD003099. Art. No.: CD003099. [DOI] [PubMed] [Google Scholar]
- 24.Sicuranza GB, Figueroa R. Uterine rupture associated with castor oil ingestion. J Matern Fetal Neonatal Med. 2003;13(2):133–134. doi: 10.1080/jmf.13.2.133.134. [DOI] [PubMed] [Google Scholar]
- 25.Kennedy HP, Nardini K, McLeod-Waldo R, Ennis L. Top-selling childbirth advice books: A discourse analysis. Birth. 2009;36(4):318–324. doi: 10.1111/j.1523-536X.2009.00359.x. [DOI] [PubMed] [Google Scholar]
- 26.Rooks JP. Labor pain management other than neuraxial: what do we know and where do we go next? Birth. 2012;39(4):318–322. doi: 10.1111/birt.12009. [DOI] [PubMed] [Google Scholar]
- 27.Anderson FWJ, Johnson CT. Complementary and alternative medicine in obstetrics. Int J Gynaecol Obstet. 2005;91(2):116–124. doi: 10.1016/j.ijgo.2005.07.009. [DOI] [PubMed] [Google Scholar]
- 28.Lee S-L, Liu C-Y, Lu Y-Y, Gau M-L. Efficacy of warm showers on labor pain and birth experiences during the first labor stage. J Obstet Gynecol Neonatal Nurs. 2012;42(1):19–28. doi: 10.1111/j.1552-6909.2012.01424.x. [DOI] [PubMed] [Google Scholar]
- 29.Madden K, Middleton P, Am C, Matthewson M, Jones L. Hypnosis for pain management during labour and childbirth. Cochrane Database Syst Rev. 2012;(Issue 11) doi: 10.1002/14651858.CD009356.pub2. Art. No.: CD009356. [DOI] [PubMed] [Google Scholar]
- 30.Anon [Accessed April 1, 2013];DONA. Available at: http://www.dona.org/
- 31.Hodnett ED, Gates S, Hofmeyr GJ, Sakala C, Weston J. Continuous support for women during childbirth. Cochrane Database Syst Rev. 2011;(Issue 2) doi: 10.1002/14651858.CD003766.pub3. Art. No.: CD003766. [DOI] [PubMed] [Google Scholar]
- 32.Taylor H, Brenner J, Overmeyer G, Siegel JW, Terhanian G. Touchdown! Online polling scores big in November 2000. Public Perspective. 2001;12:38–39. [Google Scholar]
- 33.Terhanian G, Bremer J, Smith R, Thomas R. Correcting data from online surveys for the effects of nonrandom selection and nonrandom assignment. Harris Interactive White Paper. 2000:1–13. [Google Scholar]
- 34.Childbirth Connection. [Accessed February 20, 2013];Listening to Mothers Surveys and Reports. Available at: www.childbirthconnection.org/listeningtomothers/.
- 35.Declercq E, Cunningham DK, Johnson C, Sakala C. Mothers’ reports of postpartum pain associated with vaginal and cesarean deliveries: results of a national survey. Birth. 2008;35(1):16–24. doi: 10.1111/j.1523-536X.2007.00207.x. [DOI] [PubMed] [Google Scholar]
- 36.Declercq E, Labbok MH, Sakala C, O’Hara MA. Hospital practices and women’s likelihood of fulfilling their intention to exclusively breastfeed. Am J Public Health. 2009;99:929. doi: 10.2105/AJPH.2008.135236. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 37.Young RL, Declercq E. Implications of subdividing marital status: Are unmarried mothers with partners different from unmarried mothers without partners? An exploratory analysis. Matern Child Health J. 2010;14(2):209–214. doi: 10.1007/s10995-009-0450-9. [DOI] [PubMed] [Google Scholar]
- 38.Simkin P, Bolding A. Update on nonpharmacologic approaches to relieve labor pain and prevent suffering. J Midwifery Womens Health. 2004;49(6):489–504. doi: 10.1016/j.jmwh.2004.07.007. [DOI] [PubMed] [Google Scholar]
- 39.Truven Health Analytics. The Cost of Having a Baby in the United States. 2013 Available at: http://transform.childbirthconnection.org/wp-content/uploads/2013/01/Cost-of-Having-a-Baby1.pdf.
- 40.Lantz PM, Low LK, Varkey S, Watson RL. Doulas as childbirth paraprofessionals: results from a national survey. Womens Health Issues. 2005;15(3):109–116. doi: 10.1016/j.whi.2005.01.002. [DOI] [PubMed] [Google Scholar]
- 41.Kozhimannil KB, Hardeman RR, Attanasio LB, Blauer-Peterson C, O’Brien M. Doula care, birth outcomes, and costs among Medicaid beneficiaries. Am J Public Health. 2013:1–9. doi: 10.2105/AJPH.2012.301201. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 42.Rooks JP. The midwifery model of care. J Nurse Midwifery. 1999;44(4):370–374. [PubMed] [Google Scholar]
- 43.Simkin P. Moving beyond the debate: a holistic approach to understanding and treating effects of neuraxial analgesia. Birth. 2012;39(4):327–332. doi: 10.1111/birt.12011. [DOI] [PubMed] [Google Scholar]
- 44.Gee RE, Corry MP. Patient engagement and shared decision making in maternity care. Obstet Gynecol. 2012;120(5):995–997. doi: 10.1097/aog.0b013e31827046ac. [DOI] [PubMed] [Google Scholar]
- 45.Lain SJ, Hadfield RM, Raynes-Greenow CH, et al. Quality of data in perinatal population health databases: a systematic review. Med Care. 2012;50(4):e7–e20. doi: 10.1097/MLR.0b013e31821d2b1d. [DOI] [PubMed] [Google Scholar]
- 46.Groves RM. Survey errors and survey costs. Hoboken, NJ: John Wiley & Sons; 1989. [Google Scholar]
- 47.Githens PB, Glass CA, Sloan FA, Entman SS. Maternal recall and medical records: an examination of events during pregnancy, childbirth, and early infancy. Birth. 1993;20(3):136–141. doi: 10.1111/j.1523-536x.1993.tb00438.x. [DOI] [PubMed] [Google Scholar]
- 48.Chang L, Krosnick JA. National surveys via RDD telephone interviewing versus the internet: Comparing sample representativeness and response quality. Public Opin Q. 2009;73(4):641–678. [Google Scholar]