Maternal and Child Health Journal, Vol. 4, No. 2, 2000

Physicians’ Screening Practices for Female Partner Abuse During Prenatal Visits Linda Chamberlain1,2 and Katherine A. Perham-Hester1

Objective: Our purpose was to examine physicians’ screening practices for female partner abuse during prenatal visits and to identify barriers to screening. Methods: A self-administered questionnaire was developed to collect data on physicians’ screening practices and their beliefs about screening for female partner abuse. The survey was mailed to all primary care physicians practicing in Alaska. The response rate was 80% (305/383). These analyses were limited to physicians who indicated that they provided prenatal care (n ⫽ 157). Results: More than one-half of respondents providing prenatal care estimated that 10% or more of their female patients had experienced abuse. Less than one-half of respondents had recent training on partner abuse. Only 17% of respondents routinely screened at the first prenatal visit and 5% at follow-up visits. Respondents were more likely to screen at the first prenatal visit compared to follow-up visits. Multivariate analyses failed to support any associations between physicians’ characteristics and screening practices. Physicians’ perception that abuse was prevalent among their patients and physicians’ belief that they have a responsibility to deal with abuse were the only variables that were independently associated with screening at prenatal visits. Other barriers frequently cited in the literature were not predictive of screening. Conclusion: Most Alaskan physicians do not routinely screen for abuse during prenatal visits. Medical education should increase physicians’ index of suspicion for abuse, emphasize physicians’ responsibility to address partner abuse, and reinforce the importance of routine screening throughout the pregnancy. More research is needed to identify barriers to screening and strategies for integrating routine screening into prenatal care. KEY WORDS: Domestic violence; abuse; prenatal; screening; Alaska.

report that the abuse started or increased during their pregnancies (2, 3). Estimates of the prevalence of abuse during pregnancy ranged from 0.9% to 20.1% in a review of 13 studies on battering during pregnancy (4). Adolescents experience even higher rates of abuse during pregnancy than adult women (5). For teen mothers who had recently delivered a live birth in Alaska, the prevalence of having experienced physical abuse during the past 2 years ranged from 15.5% to 31.9% depending on where they resided in the state (6). Most estimates of abuse during pregnancy are limited to physical violence. When one study included verbal abuse in the abuse assessment of low-income pregnant women, 65% of participants disclosed a history of physical violence or verbal abuse during their pregnancies (7). Research on the relationship between violence

INTRODUCTION A recent nationwide survey revealed that nearly one in three women reported being kicked, hit, choked or otherwise physically abused by a spouse or boyfriend at some time in her life (1). While some victims of domestic violence experience a decrease in physical abuse during pregnancy, many other women 1

Alaska Family Violence Prevention Project, Section of Maternal, Child and Family Health, Division of Public Health, Department of Health and Social Services, State of Alaska, Anchorage, Alaska. 2 Correspondence should be addressed to Linda Chamberlain, PhD, MPH, Alaska Family Violence Prevention Project, State of Alaska, DPH, MCFH, 1231 Gambell St., Anchorage, Alaska 99501; e-mail: [email protected].

141 1092-7875/00/0600-0141$18.00/0  2000 Plenum Publishing Corporation

142 and adverse pregnancy outcomes is sparse and the results are conflicting. While some studies have reported that women who are abused during pregnancy are at higher risk for complications including preterm labor, chorioamnionitis, low birth weight, miscarriages, and preterm delivery (8–13), a recent review of the literature concluded that no pregnancy outcome was consistently found to be associated with violence during pregnancy (14). Women experiencing abuse during their pregnancies are more likely to delay entry into prenatal care (15, 16). Several studies have substantiated the correlation between substance abuse and experiencing abuse during pregnancy (17–20). Other health risks that have been associated with abuse during pregnancy include smoking, less than optimal weight gain, and an unhealthy diet (10, 21, 22). Given the high likelihood that a woman will access health care services during her pregnancy, physicians providing prenatal care are in a strategic position to screen for partner abuse. Because victims are often isolated from friends, family, and other social support, prenatal visits may be one of the few opportunities to screen and assist women who are being abused during their pregnancies. In 1985, then Surgeon General C. Everett Koop issued public health recommendations on violence that included routine assessment of abuse for pregnant women (23). The American College of Obstetricians and Gynecologists (ACOG) revised instructional objectives for medical students, placing special emphasis on identification and intervention for battered women, and released technical bulletins on domestic violence in 1989 and 1995 (24, 25). The Diagnostic Guidelines and Treatment Guidelines on Domestic Violence released by the American Medical Association in 1992 recommend routine screening with all female patients (26). Notwithstanding recommendations and guidelines that advocate for routine screening, there is considerable evidence that screening for abuse is not routine in the prenatal care setting (27–31). Studies examining physicians’ screening practices for domestic violence during prenatal visits have limited their inquiries to the first prenatal visit or have not distinguished between initial or follow-up prenatal visits (29–31). Our previous work revealed that primary care physicians’ screened often or always when a female patient presents with an injury, but they rarely screened at initial or annual visits (32). Hypothesized barriers to screening for partner abuse have been described in the literature, but the relationships be-

Chamberlain and Perham-Hester tween these barriers and physicians’ screening practices have not been explored (33–36). Analyses of factors associated with physicians’ screening practices have been limited to a few physician characteristics (30, 31). The objectives of this study included (1) to investigate screening practices for partner abuse among primary care physicians providing prenatal care in Alaska, (2) to determine whether physicians’ screening practices varied between the first prenatal visit and follow-up prenatal visits, (3) to examine how physician characteristics may influence physicians’ prenatal screening practices, and (4) to explore the relationship between hypothesized barriers to screening for abuse and physicians’ prenatal screening practices.

METHODS Study Population All physicians in the specialities of family practice, general practice, obstetrics-gynecology, and internal medicine that were licensed to practice in the State of Alaska were surveyed. The study population was identified from the 1994 Alaska State Medical Association (ASMA) Physician Directory. The ASMA Directory is compiled from the State of Alaska medical licensing data and physician rosters from military bases, the Indian Health Service, and Native Health Corporations. Survey respondents were asked, ‘‘Are you currently engaged in clinical practice in Alaska?’’ and ‘‘Do you see FEMALE patients, age 16 or older in your current practice?’’ Any respondent who answered ‘‘no’’ to either of these filter questions was excluded from the study sample. A letter of introduction soliciting participation in the survey was sent out prior to the first survey mailing in June 1994. There were three mailings of the questionnaire plus a postcard reminder after the first mailing. A cover letter was sent with each mailing of the questionnaire that explained the purpose of the study and assured anonymity. A toll-free number was provided on the questionnaire for participants who had questions about the survey. The third and final mailing of the questionnaire was sent by certified mail. Of the 383 primary care physicians licensed to practice in Alaska, 305 (80%) responded to the survey. Eight respondents did not meet eligibility crite-

Screening Practices for Partner Abuse During Prenatal Visits ria, leaving a total of 297 respondents in the study sample. These analyses were limited to primary care physicians who indicated they provided prenatal care for both first and follow-up visits (n ⫽ 157). A comparison of all survey respondents to nonrespondents on selected demographics revealed that the percentage of male physicians was higher among physicians who did not respond (77.9%) compared with physicians who returned the survey (69.6%). However, this difference was not statistically significant (p ⫽ .133). Survey respondents differed significantly from nonrespondents by medical specialty. Obstetrician-gynecologists and general practice physicians were more likely to respond than internal medicine or family practice physicians (p ⫽ .022). The median number of years in practice was 15.0 for respondents compared with 18.5 for survey nonrespondents (p ⫽ .028).

Survey Instrument Physicians’ beliefs about barriers to screening were identified through a review of the literature and semistructured, qualitative interviews with 30 health care providers (including 10 physicians) practicing in Alaska. The Alaska Network on Domestic Violence and Sexual Assault and regional domestic violence shelters and advocacy programs reviewed the survey questionnaire and made recommendations. Pilot testing of the questionnaire was conducted with 8 physicians followed by a focus group of physicians who discussed strategies to enhance participation. Female partner abuse was defined on the questionnaire as ‘‘the actual or threatened physical, sexual and/or emotional abuse of an individual, age 16 or older, by someone with whom they have or have had an intimate or romantic relationship.’’ The questionnaire defined screening for female partner abuse as ‘‘asking direct, specific questions about abuse’’ and provided two examples of screening questions. Screening practices were assessed for first prenatal visits and follow-up prenatal visits using a 4-point Likert scale (never, sometimes, often, or always). In addition to questions on physicians’ demographics (sex, medical specialty, practice setting, location of practice, number of female patients seen weekly, age, and years in practice), respondents were asked whether they had any training on partner abuse in the past 2 years. Physicians’ perceived prevalence of female partner abuse among their patients was assessed by asking them to estimate ‘‘what percent-

143

age of the female patients, age 16 or older, that you see, do you suspect have been abused at least once during the time you have been their physician?’’ Four perceived barriers to screening were identified and assessed using a 5-point Likert scale. Three barriers used a scale ranging from strongly disagree to strongly agree (belief that they can help a victim of abuse, level of comfort with screening, and a feeling of responsibility) and one used a scale ranging from not important to very important (time constraints).

Statistical Analyses All data analyses were done in SPSS v. 7.5.2. Fisher’s exact test was employed for 2 ⫻ 2 crosstabulations. The Pearson chi-square was used to determine the statistical significance of proportions for cross-tabulations with more than two rows or columns. When the cross-tabulation consisted of two ordinal variables, a test for linear by linear association was used to determine statistical significance. Age and the number of years in practice were the only continuous variables considered for analysis. The Kolmogorov–Smirnov (KS) goodness-of-fit test to a normal distribution was performed within each subcategory of the covariates. Depending on the outcome of the KS test, the appropriate parametric or nonparametric test for differences was performed. Statistical significance was defined as p ⬍ .05 for all statistical comparisons. Perceived barriers to screening were recoded into three levels of responses after examining the univariate distribution. Physicians’ screening practices were coded as never versus sometimes/often/ always for the bivariate and multivariate analyses. Perceived barriers measured by level of agreement were dichotomized (disagree/neutral versus agree) for bivariate and multivariate analyses. Time constraints as a barrier to screening was dichotomized as not important (levels 1, 2) versus important (levels 3, 4, 5). Forward stepwise logistic regression models were constructed to select predictor variables associated with physicians’ likelihood of screening for abuse during the first prenatal visit and follow-up prenatal visits. Since age and number of years in practice were highly correlated, we selected one variable for the regression model in order to avoid problems with multicollinearity and to limit the number of variables in the model. Years in practice was selected because it was felt to be a better measure of a physi-

144 cian’s clinical experience. A total of 23 records were excluded from the regression models due to missing values. No statistically significant differences were observed between the group of physicians included in the models versus those excluded. The likelihoodratio (LR) test was computed to select variables with an entry criterion of p ⫽ .01 and an exit criterion of p ⫽ .05. Assessment of interactions was limited to bivariate associations that were significant at the .001 level. Interaction was evaluated with the ‘‘chunk test’’ described by Kleinbaum (37). Goodness of fit for the final models was examined and validated with the Hosmer–Lemeshow statistic. Finally, we calculated 95% confidence levels for the odds ratio of each parameter in the final models.

RESULTS Respondents The characteristics of primary care physicians providing prenatal care are presented in Table I. The mean age was 42.9 years and the mean number of years in practice was 14.9. Two thirds of respondents were male. Approximately one half of physicians providing prenatal care were in private practice, and one third worked for Native Health Corporations or the Indian Health Service. Family practice was the specialty of approximately one half of respondents. An estimated 52% of physicians had not had any training on partner abuse in the past 2 years. Slightly more than one half (54%) of respondents estimated that 10% or more of their adult female patients had experienced female partner abuse while under their care. One half of physicians rated the statement, ‘‘Time constraints make it difficult to ask about abuse,’’ as not important in their decision to ask about abuse. Eighty percent of respondents agreed that physicians have a responsibility to address abuse, and more than 70% indicated that they were comfortable asking screening questions and believed that they could help a victim of abuse (Table II).

Screening Practices Physicians were more likely to screen often or always at the first prenatal visit compared to followup prenatal visits (17% vs. 5%; p ⬍ .001). More than one fourth of respondents indicated that they never

Chamberlain and Perham-Hester Table I. Characteristics of Primary Care Physicians Providing Prenatal Care (N ⫽ 157) Characteristic Sex Male Female Medical specialty General practice Family practice Obstetrics/gynecology Internal medicine Practice setting Private practice NHC/IHSa Military/other Location of practice Urban Rural Number of female patients seen weekly 1–30 ⬎30 Partner abuse training in past 2 years None One or more Estimate of abuse among their female patients ⬍5% 5%–9% ⱖ10% Age, mean (SD) Years in practice, mean (SD)

n (%) 104 (66.2) 53 (33.8) 20 84 43 10

(12.7) (53.5) (27.4) (6.4)

78 (49.7) 55 (35.0) 24 (15.3) 85 (55.9) 67 (44.1)

58 (36.9) 99 (63.1) 82 (52.2) 75 (47.8)

33 (23.1) 33 (23.1) 77 (53.8) 42.9 (9.0) 14.9 (9.6)

NHC/IHS ⫽ Native Health Corporation/Indian Health Service.

a

screened for abuse at the first prenatal visit or followup prenatal visits (Table III). Physicians who had partner abuse training in the past 2 years were more likely to have screened for abuse at first prenatal visits compared to physicians who did report any recent training (p ⫽ .038). The proportion of physicians with recent training did not differ significantly for having screened at follow-up prenatal visits. Respondents with fewer years in practice were more likely to have screened at the first prenatal visit (p ⫽ .012), but this variable was not statistically significant for follow-up prenatal visits. Having screened at follow-up prenatal visits varied significantly by practice setting: 68% private practice physicians, 89% Native Health Corporation/Indian Health Service physicans, and 58% of military or other physicians (p ⫽ .004). This association was not observed for the first prenatal visit. Physicians who estimated a higher percentage of female partner abuse among their patients were more likely to report having screened at the first

Screening Practices for Partner Abuse During Prenatal Visits

145

Table II. Perceived Barriers to Screening for Female Partner Abuse n (%) Survey statement

Disagree

Neutral

I believe I can help a female patient who is being abused by her partner I feel comfortable asking female patients direct questions about abuse Physicians have as much responsibility to deal with partner abuse as they do to deal with other clinical problems

14 (9.0)

29 (18.6)

113 (72.4)

19 (12.2)

25 (16.0)

112 (71.8)

18 (11.5)

14 (9.0)

124 (79.5)

Not important

Important

76 (49.4)

33 (21.4)

Time constraints make it difficult to ask about abuse

prenatal visit (p ⬍ .001). This estimate was also statistically significant for having screened at follow-up prenatal visits (p ⬍ .001). Screening practices for the first prenatal visit and follow-up prenatal visits did not differ significantly by sex, specialty, location of practice, or the number of female patients seen weekly in the bivariate analyses. Respondents who disagreed with the statement that physicians have a responsibility to address female partner abuse were twice as likely to never screen at first prenatal visits compared to physicians who agreed with this statement (p ⫽ .004). The same association was observed for follow-up prenatal visits (p ⫽ .003). There were no significant differences between physicians having screened at first or followup prenatal visits and the following perceived barriers: time constraints, feeling comfortable about screening for abuse, and a physician’s belief that they can help an abused patient. We also evaluated the independent contribution of the variables in predicting whether physicians screen at the first prenatal visit and at follow-up prenatal visits. Physicians’ estimates of abuse and physicians’ belief that they have a responsibility to address abuse were the only variables that were independently associated with both screening at first and fol-

Agree

Very important 45 (29.2)

low-up prenatal visits (Table IV). No other evaluated variables were associated with these outcomes.

DISCUSSION We found that the majority of physicians do not screen routinely (often or always) at first prenatal or follow-up prenatal visits. These results are consistent with other surveys indicating that physicians do not routinely screen for abuse at prenatal visits (29–31). Our study provided definitions and examples of screening for female partner abuse to provide a more accurate assessment of physicians’ screening practices. Limiting the definition of screening to direct, specific questions may account for the lower frequencies of screening observed in our study. Horan et al. reported that 39% of respondents routinely screened at the first prenatal visit (30). In that study, screening for abuse was not defined and the study population was limited to ACOG fellows (obstetricians-gynecol-

Table IV. Final Logistic Models Using Forward Stepwise Regression to Select Predictors for Physicians’ Screening During Prenatal Visits, n ⫽ 134 OR (95% CI)

Table III. Percentage of Physicians who Screen for Female Partner Abuse during Prenatal Visits

First prenatal visit Follow-up prenatal visit

Never

Sometimes

Often/Always

n (%)

n (%)

n (%)

45 (28.7) 41 (26.1)

85 (54.1) 109 (69.4)

27 (17.2) 7 (4.5)

Variable in final model

First prenatal visits

Physicians’ estimates of abuse ⬍ 5% 1.0 5%–9% 3.4 (1.1, 10.1) ⱖ10% 25.3 (7.3, 87.5) Physicians’ responsibility to address abuse Disagree/neutral 1.0 Agree 7.4 (2.3, 23.8)

Follow-up prenatal visits 1.0 2.4 (0.8, 7.2) 12.2 (4.0, 37.6) 1.0 5.9 (1.8, 15.2)

146 ogists) who may have increased awareness of abuse due to ACOG’s educational initiatives on screening and recognition of abuse. Our estimate of physicians’ screening at the first prenatal visit is closer to that of Rodriquez et al., who provided definitions for domestic violence and screening for domestic violence similar to those found in our study, but did not provide examples of what is meant by ‘‘direct, specific questions about domestic violence’’ (31). We found that physicians were less likely to screen routinely at follow-up prenatal visits compared to the first prenatal visit. Horan et al. reported similar findings in their survey of ACOG fellows (30). The lower frequency of screening after the first prenatal visit may reflect physicians’ belief that they only need to screen a patient once or that assessment at the beginning of the pregnancy is sufficient. In the bivariate analyses, we found that training was significantly associated with screening at the first prenatal visit, but not at follow-up prenatal visits. One explanation may be that training has focused on screening as part of the initial obstetric history recorded at the first prenatal visit, but has not reinforced the importance of routine screening throughout the pregnancy. The decision not to screen for abuse at followup prenatal visits leads to missed opportunities for identifying abuse at the stages of pregnancy when women are at highest risk or are more likely to talk about the abuse. Studies have reported that the proportion of women disclosing abuse increased over the duration of the pregnancy (16, 38). A review of the literature on the prevalence of abuse during pregnancy found higher prevalence rates in studies that asked about violence more than once during detailed in-person interviews and in studies that asked about abuse during the third trimester (4). The potential for pregnant teens being abused by multiple perpetrators also highlights the importance of screening throughout the pregnancy (16). Our study investigated the relationship between hypothesized barriers to screening and physicians’ self-reported screening practices using multivariate analyses. In addition to examining physician characteristics and barriers identified in our presurvey interviews with health care providers, we included barriers first described by Sugg and Inui in their qualitative analysis of barriers to primary care physicians’ response to domestic violence (36). We found that factors described in the literature as potential barriers to screening (time constraints, level of comfort with screening, lack of training, and the belief that the

Chamberlain and Perham-Hester physician can help a victim of abuse) were not predictive of physicians having screened at prenatal visits. Multivariate analyses also failed to support any associations between physicians’ characteristics (sex, years in practice, specialty, practice setting, location of practice, number of female patients seen weekly) and having screened at prenatal visits. Chamberlain reported similar findings in a larger study of primary care physicians’ screening practices at initial visits, annual visits, and visits when a patient presents with an injury (32). While partner abuse has been identified as a public health priority and a leading women’s health issue, our results suggest that many physicians are not motivated to routinely screen for abuse. Physicians’ estimates of abuse among their female patients and physicians’ belief that they have a responsibility to address abuse were the only variables that were independently associated with having screened at the first prenatal visit or at follow-up prenatal visits. These findings have important implications for training and resources for physicians. Curricula on partner abuse should include a comprehensive discussion of clinical indicators that will help physicians to suspect and identify abuse even in the absence of injuries. Studies have shown that physicians are more likely to screen for abuse when a patient presents with an injury (31, 32). Training physicians to recognize the more subtle chronic health effects and behavioral indicators associated with a history of abuse will increase their index of suspicion for abuse and promote screening of all female patients. The correlation between mental abuse and physical abuse underscores the importance of increasing physicians’ awareness of the full spectrum of partner abuse (5, 7). Training and resources on partner abuse should define the role and responsibilities of physicians in addressing abuse and emphasize the importance of universal screening. Most women agree that health care providers should screen for abuse and most feel comfortable disclosing abuse to their doctors (39–41). Experts in the field of domestic violence have developed reliable screening tools and protocols that detect a higher percentage of cases of domestic violence in the prenatal setting when compared to standard patient interviews (42–44). The frequency of abuse assessment increased and was sustained for more than 1 year at prenatal clinics that implemented an abuse assessment protocol (44). Noting the high prevalence of undetected abuse among adult female patients, the potential value of this information in the care of the patient, and the low cost and low risk of harm from

Screening Practices for Partner Abuse During Prenatal Visits this type of screening, the U.S. Preventive Services Task Force recommends including a few direct questions about abuse as part of the routine history (45). Physicians need to know that screening and talking with patients about abuse can make a difference even when there are no services for victims available in their community. McFarlane et al. found that pregnant women who were screened and offered an intervention protocol reported a significant increase in the number of safety behaviors during and after pregnancy (46). One of the limitations of our study is that the temporal sequence of the events cannot be established to separate cause and effect in this cross-sectional survey. For example, it is possible that physicians’ experiences with screening female patients for abuse has led to their beliefs about partner abuse. Another significant limitation is potential misclassification of physicians’ screening practices due to social desirability. While physicians may have overreported their frequency of screening for abuse, the frequency of routine screening was very low in our study. Another limitation of our study is that we did not have data on physicians’ personal experiences with partner abuse or characteristics of patients that may influence physicians’ screening practices. While an 80% response rate is respectable, nonrespondents are always a potential source of bias in our results. The small size of our study population limited the precision of the confidence intervals in the multivariate analyses. Multivariate analyses of the larger dataset of all primary care physicians (n ⫽ 297) identified the same predictors of having screened for abuse with more precise estimates (32). Data from this study and the larger analysis of primary care physicians’ screening practices in Alaska provided baseline information to implement a training initiative—the Alaska Family Violence Prevention Project. We have developed data-driven curricula that are responsive to health care providers’ training needs and customized modules for medical specialties including prenatal care providers. Using a train-the-trainers model, we have been able to create sustainable training resources in rural and remote communities. These data were also pivotal in supporting legislation that has mandated domestic violence training for health care professionals in Alaska since 1996. While this report has focused on physicians who do not routinely screen for abuse during prenatal visits, it is important to note that over one half of Alaska physicians screened sometimes during first

147

and follow-up prenatal visits and that nearly one of five physicians screened often/always at the first prenatal visit. These data demonstrate the growing recognition of domestic violence as a reproductive health issue among physicians and the potential for universal screening in the prenatal setting. As screening policies and protocols for female partner abuse are adopted as a standard of care for pregnant women, these data will help us to understand how to promote screening and address barriers to assessment among prenatal care providers. In rural or remote communities where services for domestic violence victims are limited or nonexistent, screening for abuse in the prenatal care setting may be the only intervention that a pregnant woman being victimized will encounter. As prenatal care providers become more familiar with the risk of abuse around the time of pregnancy and the growing body of research on the consequences of victimization for pregnant women and their fetuses, it is likely that screening for female partner abuse will become an integral part of each prenatal visit. Future research needs to identify strategies to implement and sustain routine screening in different prenatal care settings, examine the value of screening as an intervention particularly in communities where services for victims are limited or nonexistent, and investigate the long-term impact of screening on the safety and quality of life for victims and their children.

ACKNOWLEDGMENTS This work was funded by the Johns Hopkins School of Hygiene and Public Health, the federal Maternal and Child Health Bureau, and the State of Alaska Department of Health and Social Services. The authors wish to acknowledge physicians practicing in the state of Alaska who made this research possible.

REFERENCES 1. Collins KS, Schoen C, Joseph S, Duchon L, Simantov E, Yellowitz M. The Commonwealth Fund 1998 Survey of Women’s Health. Baltimore, MD: Johns Hopkins University Press, 1999. 2. Perham-Hester KA, Chamberlain L. Domestic violence in Alaska. Family Health Dataline 1999;5:1–4. 3. Adams-Hillard PJ. Physical abuse in pregnancy. Obstet Gynecol 1985;66(2):185–90. 4. Gazmararian JA, Lazorick S, Spitz AM, Ballard TJ, Saltzman LE, Marks JS. Prevalence of violence against pregnant women. JAMA 1996;275(24):1915–20.

148 5. Parker B, McFarlane J, Soeken K, Torres S, Campbell D. Physical and emotional abuse in pregnancy: A comparison of adult and teenage women. Nurs Res 1993;42(3):173–7. 6. Perham-Hester KA, Gessner B. Teen live births in Alaska during 1991–94. Family Health Dataline 1997;3(1). 7. O’Campo P, Gielen AC, Faden RR, Kass N. Verbal abuse and physical violence among a cohort of low-income pregnant women. Women’s Health Issues 1994;4(1):29–37. 8. Berenson AB, Weimann CM, Wilkinson GS, Jones WA, Anderson GD. Perinatal morbidity associated with violence experienced by pregnant women. Am J Obstet Gynecol 1994; 170(6):1760–9. 9. Bullock LF, McFarlane J. The birth-weight/battering connection. Am J Nurs 1989:1153–5. 10. Parker B, McFarlane J, Soeken K. Abuse during pregnancy: Effects on maternal complications and birth weight in adult and teenage women. Obstet Gynecol 1994;84(3):323–8. 11. Schei B, Samuelsen SO, Bakketeig LS. Does spousal physical abuse affect the outcome of pregnancy? Scand J Soc Med 1991;19(1):26–31. 12. Dye TD, Tolliver NJ, Lee RV, Kenney CJ. Violence, pregnancy and birth outcome in Appalachia. Paediatr Perinat Epidemiol 1995;9:35–47. 13. Mooney J. The method used and a selection of the findings from the North London domestic violence survey. Occasional Paper. London: Middlesex University, Centre for Criminology, 1993. 14. Peterson R, Gazmararian JA, Spitz A, Rowley DL, Goodwin MM, Saltzman LE, Marks JS. Violence and adverse pregnancy outcomes: A review of the literature and directions for future research. Am J Prev Med 1997;13(5):366–73. 15. Dietz PM, Gazmararian JA, Goodwin MM, Bruce FC, Johnson CH, Rochat RW. Delayed entry into prenatal care: Effect of physical violence. Obstet Gynecol 1997;90(2):221–4. 16. McFarlane J, Parker B, Soeken K, Bullock L. Assessing for abuse during pregnancy: Severity and frequency of injuries and associated entry into prenatal care. JAMA 1992;267(23): 3176–8. 17. Berenson AB, Stiglich NJ, Wilkinson GS, Anderson GD. Drug abuse and other risk factors for physical abuse in pregnancy among White non-Hispanic, Black, and Hispanic women. Am J Obstet Gynecol 1991;164:1491–9. 18. Amaro H, Fried LE, Cabral H, Zuckerman B. Violence during pregnancy and substance use. Am J Public Health 1990;80(5): 575–9. 19. Campbell J, Poland M, Waller J, Ager J. Correlates of battering during pregnancy. Res Nurs Health 1992;15:219–26. 20. Perham-Hester KA, Gessner BD. Correlates of drinking during the third trimester of pregnancy in Alaska. Matern Child Health J 1997;1(3):165–72. 21. Campbell J. Addressing battering during pregnancy: Reducing low birth weight and ongoing abuse. Semin Perinatol 1995;19(4):301–6. 22. Stewart DE, Cecutti A. Physical abuse in pregnancy. Can Med Assoc J 1993;149(9):1257–63. 23. U.S. Public Health Service. Surgeon General’s Workshop on Violence and Public Health Report. Washington, DC: Author, 1985. 24. American College of Obstetricians and Gynecologists. The battered woman. Washington, DC: Author, 1989. 25. American College of Obstetricians and Gynecologists. Domestic violence. Washington, DC: Author, 1995.

Chamberlain and Perham-Hester 26. American Medical Association. Diagnostic and treatment guidelines on domestic violence. Chicago, IL: Author, 1992. 27. Helton AS, McFarlane J, Anderson ET. Battered and pregnant: A prevalence study. Am J Public Health 1987;77(10): 1337–9. 28. McGrath ME, Hogan JW, Peipert JF. A prevalence survey of abuse and screening for abuse in urgent care patients. Obstet Gynecol 1998;91(4):511–4. 29. Parsons LH, Zaccaro D, Wells B, Stovall TG. Methods of and attitudes towards screening obstetrics and gynecology patients for domestic violence. Am J Obstet Gynecol 1995; 173(2):381–7. 30. Horan DL, Chapin J, Klein L, Schmidt LA, Schulkin J. Domestic violence screening practices of obstetrician-gynecologists. Obstet Gynecol 1998;92(5):785–9. 31. Rodriquez MA, Bauer HM, McLoughlin E, Grumbach K. Screening and intervention for intimate partner abuse. JAMA 1999;282(5):468–4. 32. Chamberlain L. Primary care physicians’ screening practices for female partner abuse in Alaska. Dissertation. Baltimore, MD: Johns Hopkins School of Hygiene and Public Health, 1996. 33. Brown JB, Lent B, Sas G. Identifying and treating wife abuse. J Fam Pract 1993;36(2):185–1. 34. Ferris LE. Canadian family physicians’ and general practitioners’ perceptions of their effectiveness in identifying and treating wife abuse. Med Care 1994;32(12):1163–72. 35. Ferris LE, Tudiver F. Family physicians’ approach to wife abuse: A study of Ontario, Canada, practices. Fam Med 1992;24:276–82. 36. Sugg NK, Inui T. Primary care physicians’ response to domestic violence. JAMA 1992;267(23):3157–60. 37. Kleinbaum DG. Logistic regression. New York: Springer-Verlag, 1994. 38. Webster J, Sweett S, Stoltz TA. Domestic violence in pregnancy: A prevalence study. Med J Aust 1994;161:466–70. 39. Caralis PV, Musialowski R. Women’s experiences with domestic violence and their attitudes and expectations regarding medical care of abuse victims. South Med J 1997;90(11): 1075–80. 40. McNutt LA, Carlson BE, Gagen D, Winterbauer N. Reproductive violence screening in primary care: Perspectives and experiences of patients and battered women. JAMWA 1999;54(2):85–90. 41. Friedman LS, Samet JH, Roberts MS, Hudlin M, Hans P. Inquiry about victimization experiences. A survey of patient preferences and physician practices. Arch Intern Med 1992;152:1186–90. 42. Covington DL, Dalton VK, Diehl SJ, Wright BD, Piner MH. Improving detection of violence among pregnant adolescents. J Adolesc Health 1997;21:18–24. 43. Norton LB, Peipert JF, Zierler S, Lima B, Hume L. Battering in pregnancy: An assessment of two screening methods. Obstet Gynecol 1995;85(3):321–5. 44. Wiist WH, McFarlane J. The effectiveness of an abuse assessment protocol in public health prenatal clinics. Am J Public Health 1999;89(8):1217–21. 45. U.S. Preventive Services Task Force. Guide to clinical preventive services; Report of the U.S. Preventive Services Task Force, 2nd ed. Baltimore, MD: Williams and Wilkins, 1996:555–65. 46. McFarlane J, Parker B, Soeken K, Silva C, Reel S. Safety behaviors of abused women after an intervention during pregnancy. J Obstet Gynecol Neonatal Nurs 1998;27(1):64–9.