- Email: [email protected]
Stages of change and prenatal alcohol use
Journal of Substance Abuse Treatment 32 (2007) 105 – 109
Brief article
Stages of change and prenatal alcohol use Grace Chang, (M.D., M.P.H.)a,b,4, Tay McNamara, (Ph.D.)a, Louise Wilkins-Haug, (M.D., Ph.D.)c,d, E. John Orav, (Ph.D.)e,f a
Department of Psychiatry, Brigham and Women’s Hospital, 75 Francis Street, Boston, MA 02115, USA b Department of Psychiatry, Harvard Medical School, Boston, MA 02115, USA c Department of Obstetrics and Gynecology, Brigham and Women’s Hospital, 75 Francis Street, Boston, MA 02115, USA d Department of Obstetrics and Gynecology, Harvard Medical School, Boston, MA 02115, USA e Department of General Medicine and Primary Care, Brigham and Women’s Hospital, Boston, MA 02115, USA f Department of Medicine (Biostatistics), Harvard Medical School, Boston, MA 02115, USA Received 17 May 2006; accepted 4 July 2006
Abstract This study evaluated stage of change as a predictor of alcohol use in a sample of 301 pregnant women who were either in the precontemplation (62%) or in the action (38%) stage of change in their first trimester. Stage of change distinguished between different patterns of alcohol consumption before and after pregnancy. Those in the precontemplation stage drank more per episode and more often before pregnancy than those in the action stage. The precontemplation group also had a significantly greater quantity of alcohol after pregnancy. However, stage of change did not directly predict subsequent prenatal alcohol use. Previous alcohol use, age, and education were the most significant predictors of prenatal drinks per drinking day. Temptation to drink alcohol was the best predictor of prenatal drinking frequency after study enrollment. Women in both stages of change reduced the quantity and the frequency of their alcohol consumption while pregnant and achieved comparable rates of abstinence. D 2007 Elsevier Inc. All rights reserved. Keywords: Stages of change; Prenatal alcohol use
1. Introduction The transtheoretical model of intentional behavioral change has been an important and influential attempt to characterize disengagement from harmful behaviors such as smoking or drinking (Allsop, 2003; DiClemente, 2003; DiClemente, Bellino, & Neavins, 1999). In this model, stages of change depict the motivational and dynamic fluctuations of the change process over time. Because motivation is a key element in treatment and recovery, stage-based interventions that take into account the current This study was supported by grants (R01 AA12548 and K24 AA 00289, to G.C.) from the National Institute on Alcohol Abuse and Alcoholism. 4 Corresponding author. Brigham and Women’s Hospital, Department of Psychiatry, 75 Francis Street, Boston, MA 02115, USA. Tel.: +1 617 732 6775; fax: +1 617 264 6364. E-mail address: [email protected] (G. Chang). 0740-5472/07/$ – see front matter D 2007 Elsevier Inc. All rights reserved. doi:10.1016/j.jsat.2006.07.003
stage an individual has reached have been widely used, although empirical justification may be lacking (Riemsa et al., 2002). The enduring popularity of the transtheoretical model notwithstanding, the need for further evaluation has been increasingly appreciated due to apparent limitations (Sutton, 2001; West, 2005). For example, Kavanagh, Sitharthan, and Sayer (1996) examined the predictive utility of stage of change, self-efficacy, and alcohol dependence among a group of 166 adults enrolled in a correspondence treatment for alcohol abuse. Whereas self-efficacy predicted both treatment retention and later alcohol consumption, and previous alcohol consumption constituted the best predictor of subsequent intake, the role of stage of change was considerably weaker. Maisto et al. (2001) found that readiness to change was not related to changes in drinking by primary care study participants randomized to standard care or motivational enhancement interventions.
106
G. Chang et al. / Journal of Substance Abuse Treatment 32 (2007) 105–109
Possible limitations in the transtheoretical model may be explained by difficulties in measuring readiness to change. Five stages of change were originally conceptualized, but the model has since been modified to three stages (Sutton, 2001). Readiness to change begins with precontemplation, the stage at which there is no intention to change in the foreseeable future. Contemplation is the next stage, the stage when an individual is aware of the problem and is considering making a change in response to the problem but has not yet made a commitment to effect such a change. The action stage occurs when an individual modifies one’s behaviors (Prochaska, DiClemente, & Norcross, 1992). The Readiness to Change Questionnaire (RCQ) is the most popular measure of the stages of change and assigns drinkers to the precontemplation, contemplation, or action stage (Rollnick, Heather, Gold, & Hall, 1992). It was specifically developed for use in brief opportunistic interventions among excessive drinkers and has been studied in at least five countries, in as many different languages and cultures, but among a more limited range of subjects (Defuentes-Merillas, Dejong, & Schippers, 2002; Forsberg, Halldin, & Wennberg, 2003; Hannover et al., 2002; Rodriguez-Martos et al., 2000). Findings have been almost as diverse, with some since questioning the threestage model and instead advocating for either a continuous (Budd & Rollnick, 1996) or a two-stage alternative (precontemplation and contemplation/action; Willoughby & Edens, 1996). Furthermore, the transtheoretical model of change may not be relevant to all groups of drinkers, such as pregnant women. A few studies have examined the utility of the transtheoretical model among pregnant smokers, but not among pregnant drinkers. A comparison of 103 economically disadvantaged pregnant smokers and 103 matched community smokers found that the majority of pregnant smokers were not highly motivated to quit smoking simply because they were pregnant (Ruggiero, Tsoh, Everett, Fava, & Guise, 2000). A study of 54 low-income pregnant smokers randomized to either motivational intervention or usual care found that stage of change did not progress in the course of treatment, thus failing to support the investigators’ hypothesis that a motivational intervention would increase motivation for change (Stotts, DeLaune, Schmitz, & Grabowski, 2004). Although findings from a cross-sectional study of 637 pregnant women who were either current or previous cigarette smokers supported the overall relevance of the transtheoretical model of change, the investigators also concluded that a staged approach was less necessary for pregnant women (Slade, Laxton-Kane, & Spiby, 2006). Because no amount of alcohol is safe during pregnancy, a better understanding of why some women avoid alcohol and others do not is therefore highly desirable (Kost, Landry, & Darroch, 1998; Mukherjee, Hollins, & Abou-Salah, 2005). The overall prevalence of any prenatal alcohol consumption is 13%, with rates of binge drinking (defined as five or more drinks on any one occasion) and frequent drinking (defined
as seven or more drinks per week, or five or more drinks on any one occasion) during pregnancy estimated to be about 6% (Centers for Disease Control and Prevention, 2002). Thus, screening and brief interventions for prenatal alcohol use may help to reduce, if not eliminate, fetal risk and to maximize pregnancy outcome (Chang, 2004/2005). The purpose of this study is to evaluate stage of change as a predictor of alcohol use in a sample of alcohol-screenpositive pregnant women who participated in a randomized trial of brief interventions. Thus, if the transtheoretical model is applicable, then pregnant women in the action stage should demonstrate alcohol use less than that of those assigned to other stages.
2. Materials and methods 2.1. Setting This study took place at the Brigham and Women’s Hospital (Boston, MA). A screening survey with questions about health habits such as diet, exercise, sleep, and usual drinking, and the T-ACE questionnaire (a four-item alcohol screening instrument) were given to patients initiating prenatal care at one of three hospital obstetric practices (clinic, faculty, and private group affiliate), as well as to those responding to e-mail and other study announcements. 2.2. Subjects The subjects were 301 (99%) of 304 pregnant women who were enrolled in a randomized trial of brief interventions for prenatal alcohol use (Chang et al., 2005). The women were eligible to participate if they satisfied the entry criteria, which included a positive T-ACE alcohol screen (a score of 2 or more) and risk for prenatal alcohol use, as defined by any of the following: (1) alcohol use while pregnant in the 3 months before study enrollment; (2) consumed at least one drink per day in the 6 months before study enrollment; or (3) drank during a previous pregnancy. The T-ACE questionnaire is a four-item screening questionnaire based on the CAGE questionnaire that asks: How many drinks does it take to make you feel high (tolerance, T)? Have people ever annoyed you by criticizing your drinking (annoyed, A)? Have you ever felt you ought to cut down on your drinking (cut down, C)? Have you ever had a drink first thing in the morning to steady your nerves or to get rid of an hangover (eye opener, E)? The tolerance question is given two points if the respondent states that more than two drinks are needed to feel the effects, and affirmative replies to the A, C, and E questions
G. Chang et al. / Journal of Substance Abuse Treatment 32 (2007) 105–109
are each given one point. The T-ACE questionnaire has been well studied and validated in diverse clinical populations (Chang et al., 1998; Sokol, Martier, & Ager, 1989). Participants gave written informed consent for this study, which was reviewed and approved by the Institutional Review Board of the Brigham and Women’s Hospital. A Certificate of Confidentiality for the project was granted by the Department of Health and Human Services. 2.3. Interviews and intervention The participants completed two interviews. The first interview was conducted upon study enrollment, which occurred at a median of 11.5 weeks’ gestation. The pregnant participants completed: (1) the RCQ, a 12-item instrument that assigns nontreatment-seeking people to the precontemplation, contemplation, or action stage of change based on how they feel about their drinking at present (in the case of the participants, this would include the first trimester of pregnancy; Rollnick et al., 1992); (2) the Alcohol Timeline Follow Back (TLFB), a tool to obtain estimates of daily drinking for the 6 months before study enrollment, which would have included drinking before pregnancy and during early pregnancy (Sobell & Sobell, 1992); and (3) the Alcohol Abstinence Self-Efficacy (AASE) Scale, a tool to evaluate an individual’s perceived temptation to drink (a measure of cue strength) and efficacy (confidence) in abstaining from drinking in 20 common situations at present (in the first trimester; DiClemente, Carbonari, Montgomery, & Hughes, 1994). Scores on the temptation and confidence scales range from a minimum of 20 to a maximum of 100. The second interview occurred at an average of 78 days after delivery. At that time, 95% of the participants completed the Alcohol TLFB interview to provide information about alcohol consumption from the time of study enrollment until delivery. Randomly selected participants received a brief intervention after the first interview. Master’s-level nurses or the first author gave brief interventions to randomly selected participants and reviewed the impact of health habits such as smoking and drinking on pregnancy outcome in the course of 25 minutes. Two main results have been reported. First, brief interventions for prenatal alcohol reduced subsequent consumption most significantly for women with the highest consumption initially. Second, the effects of brief interventions were significantly enhanced when a partner participated. Additional details are available elsewhere (Chang et al., 2005). 2.4. Data analysis Data were analyzed using univariate and multivariate techniques with SAS 8.2 (SAS Institute, Cary, NC). Descriptive statistics are reported as percentages and means. The characteristics of pregnant women by stage of change were compared using the t-test or the chi-square test, as
107
appropriate. The frequency and the quantity of prenatal alcohol use after study enrollment were the dependent variables in ordinary least squares regression models. The models included demographic data (e.g., age, marital status, education, and race), clinical history (e.g., first pregnancy and history of obstetric problems), prior drinking history (as lagged dependent variable), and other study variables (e.g., brief intervention status and AASE temptation score) as predictors. Multiple imputations (five imputations) were used to manage missing data. All analyses were replicated with mean substitution to verify findings from multiple imputations (Rubin, 1987). The purpose of multivariate analyses was to focus on the impact of stage of change on the quantity and the frequency of alcohol use in this sample of pregnant women. Because the primary purpose of this study is limited to investigating these two relationships (stage of change vs. quantity and frequency), no formal correction for multiple testing was used.
3. Results Three hundred four participants were assigned to the precontemplation (61.5%), contemplation (1%), and action stages (37.5%) based on RCQ responses for drinking at present (in the first trimester). Results will focus on women
Table 1 Demographic data, clinical history, and drinking characteristics, by stage of change Variables
Precontemplation (n = 187)
Action (n =114)
Mean age (years) 30.5 30.6 Married (%) 77.6 86.5 African American (%) 6.4 6.3 Mean years 16.5 16.7 of education First pregnancy (%) 48.7 35.4 History of obstetric 20.1 31.8 problems (%) AASE score Confidence 93.8 93.7 Temptation 27.4 30.0 Drinking frequency (mean % drinking days) Before pregnancy 18.6 13.4 Early pregnancy, 2.9 2.1 before enrollment Pregnancy, 2.1 2.0 after enrollment After pregnancy 11.6 10.2 Drinking quantity (mean drinks per drinking day) Before pregnancy 2.1 1.6 Early pregnancy, 0.43 0.38 before enrollment Pregnancy, 0.34 0.37 after enrollment After pregnancy 1.45 1.02 Abstinent after 57 52 enrollment (%)
p (t-test or chi-square) .67 .08 .96 .33 .04 .05
.95 .04 .02 .31 .89 .49 b .01 .65 .68 b .01 .44
108
G. Chang et al. / Journal of Substance Abuse Treatment 32 (2007) 105–109
Table 2 Predictors of prenatal alcohol consumption after study enrollment
Predictors Intercept Action stage Brief intervention Interaction between action stage and brief intervention African American History of obstetric problems First pregnancy Married Age in years Education in years Previous alcohol consumption Temptation
Drinking frequency (% drinking days)
Drinking quantity (drinks per drinking day)
B
B
Pr
predictive of either the subsequent frequency or the quantity of prenatal alcohol use. The results of the analysis are summarized in Table 2.
Pr
0.31 0.11 0.096 0.95
.33 .27 .31 .53
8.0 1.03 0.20 1.4
.0008 .17 .76 .20
0.10 0.03
.50 .74
1.0 0.81
.37 .22
0.04 0.07 0.00004 0.0004 0.54
.60 .49 .09 .72 .17
0.13 0.18 0.0003 0.021 0.49
.83 .80 .036 .023 b .0001
0.008
.03
0.026
.33
in the precontemplation (n = 187) and action (n = 114) stages. When compared to those assigned to the action stage, precontemplation-stage women were significantly more likely to be pregnant for the first time (48.7% vs. 35.4%), with a less frequent history of obstetric problems (20.1% vs. 31.8%) and with less temptation to drink (27.4 vs. 30.0). However, there were no other systematic differences between the two groups in terms of age, marital status, racial background, years of education, or gestational age upon enrollment. With regards to drinking, precontemplation-stage women consumed significantly more alcohol before pregnancy, for both frequency (18.6% vs. 13.4%) and quantity (2.1 vs. 1.6 drinks per drinking), when compared to action-stage women. The precontemplation-stage women also had significantly more drinks per drinking day (1.5 vs. 1.0), but not more frequent drinking after pregnancy, than the action-stage women. Women from both stages reduced the quantity and the frequency of their alcohol consumption while pregnant and achieved comparable rates of abstinence. Table 1 summarizes the comparison between the women assigned to the precontemplation stage and the women assigned to the action stage. The stage of change was compared to demographic data (e.g., age and years of education), clinical history (e.g., past alcohol use and obstetric history), and AASE temptation score as predictors of the frequency and the quantity of prenatal alcohol use after study enrollment in regression models. The AASE temptation score was included because it differed by stage-of-change group, but the confidence score did not. Three predictors of drinks per drinking day were identified: age in years ( p = .04), education in years ( p = .02), and past drinking ( p b .0001). Temptation to drink was the only statistically significant predictor of drinking frequency ( p = .03). Stage of change was not
4. Discussion Stage of change distinguished between different patterns of alcohol consumption prior to pregnancy in this sample of 301 pregnant women. Those designated to be in the precontemplation stage drank more per episode and more often before pregnancy than those in the action stage. However, stage of change did not directly predict subsequent prenatal alcohol use. Previous alcohol use, age, and education were the most significant predictors of prenatal drinks per drinking day. Temptation to drink alcohol was the best predictor of prenatal drinking frequency after study enrollment. The precontemplation group also had a significantly greater quantity of alcohol after pregnancy. Explanations for the lack of direct association between stage of change and subsequent prenatal alcohol use are speculative. A possible explanation is that, although stage designation was associated with prepregnancy drinking, the study allowed a prospective evaluation with minimal loss to follow-up—two factors that have been cited as shortcomings in other investigations (DiClemente et al., 1999). Unexpectedly, the action group reported more temptation to drink, perhaps because they had made greater reductions in alcohol consumption. Temptation to drink has been found to predict responses to alcohol cues, even when the effects of typical drinking patterns were taken into account (Palfai, 2001). Another possible explanation may involve the very high levels of confidence expressed by both groups of women in managing their drinking. Half of the participants in the study sample became abstinent after study enrollment. Finally, it is possible that the difficulties in measuring stage of change are exacerbated when pregnant women are being assessed. Pregnant women have long been considered to be highly motivated to alter behaviors such as smoking (Boyd, 1985). Potential limitations to the generalizability of the study findings include sample characteristics. Higher education, non-Hispanic background, and employment have been shown to be associated with greater prenatal alcohol use (Centers for Disease Control and Prevention, 2002). Participants did not appear to be heavy drinkers, but they drank in excess of all recommendations for abstinence during pregnancy. Underreporting of alcohol use is possible. Reactivity to research protocols and regression to the mean are possibilities to consider as well (Clifford & Maisto, 2000; Fleming, Barry, Manwell, Johnson, & London, 1997). Finally, stage of change was not measured serially; thus, it was not possible to ascertain stage progression in this sample. Although stage-of-change designation did differentiate between prepregnancy levels of alcohol use, it was not
G. Chang et al. / Journal of Substance Abuse Treatment 32 (2007) 105–109
predictive of subsequent prenatal consumption. Thus, the definitive role of stage of change as a predictor of prenatal drinking behavior requires further study. Nonetheless, because no amount of alcohol consumption can be considered safe during pregnancy, results from this study support the importance of identifying previous patterns of alcohol use, even among those who appear to enjoy the apparent advantages of age and education (Office of the Surgeon General, 2005). References Allsop, S. (2003). Addiction and change: How addictions develop and addicted people recover (Book review). Addiction, 98, 1782 – 1783. Boyd, M. D. (1985). Patient motivation during pregnancy. Maryland Medical Journal, 34, 977 – 981. Budd, R. J., & Rollnick, S. (1996). The structure of the Readiness to Change Questionnaire: A test of Prochaska and DiClemente’s transtheoretical model. British Journal of Health Psychology, 1, 365 – 376. Centers for Disease Control and Prevention. (2002, April). Alcohol use among women of childbearing age—United States, 1991–1999. Morbidity and Mortality Weekly Report, 51, 273 – 276. Chang, G. (2004/2005). Screening and brief intervention in prenatal care settings. Alcohol Research and Health, 28, 80 – 84. Chang, G., McNamara, T. K., Orav, E. J., Koby, D., Lavigne, A., Ludman, B., et al. (2005). Brief intervention for prenatal alcohol use: A randomized trial. Obstetrics and Gynecology, 105, 991 – 998. Chang, G., Wilkins-Haug, L., Berman, S., Goetz, M. A., Behr, H., & Hiley, A. (1998). Alcohol use and pregnancy: Improving identification. Obstetrics and Gynecology, 91, 892 – 898. Clifford, P. R., & Maisto, S. A. (2000). Subject reactivity effects and alcohol treatment outcome research. Journal of Studies on Alcohol, 61, 787 – 793. Defuentes-Merillas, L., Dejong, C., & Schippers, G. (2002). Reliability and validity of the Dutch version of Readiness to Change Questionnaire. Alcohol and Alcoholism, 37, 93 – 99. DiClemente, C. C. (2003). Addiction and change: How addictions develop and addicted people recover (pp. 3 – 61). New York, NY7 Guilford Press. DiClemente, C. C., Bellino, L. E., & Neavins, T. M. (1999). Motivation for change and alcoholism treatment. Alcohol Health and Research, 23, 86 – 92. DiClemente, C. C., Carbonari, J. P., Montgomery, R. P. G., & Hughes, S. O. (1994). The Alcohol Abstinence Self-Efficacy scale. Journal of Studies on Alcohol, 55, 141 – 148. Fleming, M. F., Barry, K. L., Manwell, L. B., Johnson, K., & London, R. (1997). Brief physician advice for problem alcohol drinkers. Journal of the American Medical Association, 277, 1039 – 1045. Forsberg, L., Halldin, J., & Wennberg, P. (2003). Psychometric properties and factor structure of the Readiness to Change Questionnaire. Alcohol and Alcoholism, 38, 279 – 280. Hannover, W., Thyrina, J. R., Hapke, U., Rumpf, H. J., Meyer, C., & John, U. (2002). The Readiness to Change Questionnaire in subjects with hazardous alcohol consumption, alcohol misuse and dependence in a general population survey. Alcohol and Alcoholism, 37, 362 – 369.
109
Kavanagh, D. J., Sitharthan, T., & Sayer, G. P. (1996). Prediction of results from correspondence treatment for controlled drinking. Addiction, 91, 1539 – 1545. Kost, K., Landry, D. J., & Darroch, J. E. (1998). Predicting maternal behaviors during pregnancy: Does intention status matter? Family Planning Perspectives, 30, 79 – 88. Maisto, S. A., Conigliaro, J., McNeil, M., Kraemer, K., Conigliaro, R. L., & Kelley, M. E. (2001). Effects of two types of brief interventions and readiness to change on alcohol use in hazardous drinkers. Journal of Studies on Alcohol, 62, 605 – 614. Mukherjee, R. A. S., Hollins, S., & Abou-Salah, M. T. (2005). Low level alcohol consumption and the fetus. British Medical Journal, 330, 375 – 376. Office of the Surgeon General (2005, February 21). The surgeon general releases advisory on alcohol use in pregnancy (Press release). Bethesda, MD7 Department of Health and Human Services (Available at www.hhs.gov/surgeongeneral/pressreleases/sg02222005.html). Palfai, T. P. (2001). Individual differences in temptation and responses to alcohol cues. Journal of Studies on Alcohol, 62, 657 – 666. Prochaska, J. O., DiClemente, C. C., & Norcross, J. C. (1992). In search of how people change: Applications to predictive behaviors. American Psychologist, 47, 1102 – 1114. Riemsa, R. P., Pattenden, J., Bridle, C., Sowden, A. J., Mather, L., & Walker, A. (2002). A systematic review of the effectiveness of interventions based on a stages of change approach to promote individual behaviour change. Health Technology Assessment, 6, 1 – 55. Rodriguez-Martos, A., Rubio, G., Auba, G., Santo Domingo, J., Rorralba, L., & Campilo, M. (2000). Readiness to Change Questionnaire: Reliability study of the Spanish version. Alcohol and Alcoholism, 35, 270 – 275. Rollnick, S., Heather, N., Gold, R., & Hall, W. (1992). Development of a short bReadiness to ChangeQ questionnaire for use in brief opportunistic brief interventions. British Journal on Addiction, 87, 745 – 754. Rubin, D. B. (1987). Multiple imputation for nonresponse in surveys. New York, NY7 Wiley and Sons. Ruggiero, L., Tsoh, J. U., Everett, K., Fava, J. L., & Guise, B. J. (2000). The transtheoretical model of smoking: Comparison of pregnant and nonpregnant smokers. Addictive Behaviors, 25, 239 – 251. Slade, P., Laxton-Kane, M., & Spiby, H. (2006). Smoking in pregnancy: The role of the transtheoretical model and the mother’s attachment to the fetus. Addictive Behaviors, 31, 743 – 757. Sobell, L. C., & Sobell, M. B. (1992). Timeline followback: A technique for assessing self-reported alcohol consumption. In J. P. Allen, & R. Z. Litten (Eds.), Measuring alcohol consumption: Psychosocial and biochemical methods (pp. 41 – 72). Totowa, NJ7 Humana Press. Sokol, R. J., Martier, S. S., & Ager, J. W. (1989). The T-ACE questions: Practical prenatal detection of risk-drinking. American Journal of Obstetrics and Gynecology, 160, 863 – 871. Stotts, A. L., DeLaune, K. A., Schmitz, J. M., & Grabowski, J. (2004). Impact of a motivational intervention on mechanisms of change in lowincome pregnant smokers. Addictive Behaviors, 29, 1649 – 1657. Sutton, S. (2001). Back to the drawing board? A review of applications of the transtheoretical model to substance use. Addiction, 96, 175 – 186. West, R. (2005). Time for a change: Putting the transtheoretical (stages of change) model to rest. Addiction, 100, 1036 – 1039. Willoughby, F. W., & Edens, J. F. (1996). Construct validity and predictive utility of the stages of change for alcoholics. Journal of Substance Abuse, 8, 275 – 291.
Brief article
Stages of change and prenatal alcohol use Grace Chang, (M.D., M.P.H.)a,b,4, Tay McNamara, (Ph.D.)a, Louise Wilkins-Haug, (M.D., Ph.D.)c,d, E. John Orav, (Ph.D.)e,f a
Department of Psychiatry, Brigham and Women’s Hospital, 75 Francis Street, Boston, MA 02115, USA b Department of Psychiatry, Harvard Medical School, Boston, MA 02115, USA c Department of Obstetrics and Gynecology, Brigham and Women’s Hospital, 75 Francis Street, Boston, MA 02115, USA d Department of Obstetrics and Gynecology, Harvard Medical School, Boston, MA 02115, USA e Department of General Medicine and Primary Care, Brigham and Women’s Hospital, Boston, MA 02115, USA f Department of Medicine (Biostatistics), Harvard Medical School, Boston, MA 02115, USA Received 17 May 2006; accepted 4 July 2006
Abstract This study evaluated stage of change as a predictor of alcohol use in a sample of 301 pregnant women who were either in the precontemplation (62%) or in the action (38%) stage of change in their first trimester. Stage of change distinguished between different patterns of alcohol consumption before and after pregnancy. Those in the precontemplation stage drank more per episode and more often before pregnancy than those in the action stage. The precontemplation group also had a significantly greater quantity of alcohol after pregnancy. However, stage of change did not directly predict subsequent prenatal alcohol use. Previous alcohol use, age, and education were the most significant predictors of prenatal drinks per drinking day. Temptation to drink alcohol was the best predictor of prenatal drinking frequency after study enrollment. Women in both stages of change reduced the quantity and the frequency of their alcohol consumption while pregnant and achieved comparable rates of abstinence. D 2007 Elsevier Inc. All rights reserved. Keywords: Stages of change; Prenatal alcohol use
1. Introduction The transtheoretical model of intentional behavioral change has been an important and influential attempt to characterize disengagement from harmful behaviors such as smoking or drinking (Allsop, 2003; DiClemente, 2003; DiClemente, Bellino, & Neavins, 1999). In this model, stages of change depict the motivational and dynamic fluctuations of the change process over time. Because motivation is a key element in treatment and recovery, stage-based interventions that take into account the current This study was supported by grants (R01 AA12548 and K24 AA 00289, to G.C.) from the National Institute on Alcohol Abuse and Alcoholism. 4 Corresponding author. Brigham and Women’s Hospital, Department of Psychiatry, 75 Francis Street, Boston, MA 02115, USA. Tel.: +1 617 732 6775; fax: +1 617 264 6364. E-mail address: [email protected] (G. Chang). 0740-5472/07/$ – see front matter D 2007 Elsevier Inc. All rights reserved. doi:10.1016/j.jsat.2006.07.003
stage an individual has reached have been widely used, although empirical justification may be lacking (Riemsa et al., 2002). The enduring popularity of the transtheoretical model notwithstanding, the need for further evaluation has been increasingly appreciated due to apparent limitations (Sutton, 2001; West, 2005). For example, Kavanagh, Sitharthan, and Sayer (1996) examined the predictive utility of stage of change, self-efficacy, and alcohol dependence among a group of 166 adults enrolled in a correspondence treatment for alcohol abuse. Whereas self-efficacy predicted both treatment retention and later alcohol consumption, and previous alcohol consumption constituted the best predictor of subsequent intake, the role of stage of change was considerably weaker. Maisto et al. (2001) found that readiness to change was not related to changes in drinking by primary care study participants randomized to standard care or motivational enhancement interventions.
106
G. Chang et al. / Journal of Substance Abuse Treatment 32 (2007) 105–109
Possible limitations in the transtheoretical model may be explained by difficulties in measuring readiness to change. Five stages of change were originally conceptualized, but the model has since been modified to three stages (Sutton, 2001). Readiness to change begins with precontemplation, the stage at which there is no intention to change in the foreseeable future. Contemplation is the next stage, the stage when an individual is aware of the problem and is considering making a change in response to the problem but has not yet made a commitment to effect such a change. The action stage occurs when an individual modifies one’s behaviors (Prochaska, DiClemente, & Norcross, 1992). The Readiness to Change Questionnaire (RCQ) is the most popular measure of the stages of change and assigns drinkers to the precontemplation, contemplation, or action stage (Rollnick, Heather, Gold, & Hall, 1992). It was specifically developed for use in brief opportunistic interventions among excessive drinkers and has been studied in at least five countries, in as many different languages and cultures, but among a more limited range of subjects (Defuentes-Merillas, Dejong, & Schippers, 2002; Forsberg, Halldin, & Wennberg, 2003; Hannover et al., 2002; Rodriguez-Martos et al., 2000). Findings have been almost as diverse, with some since questioning the threestage model and instead advocating for either a continuous (Budd & Rollnick, 1996) or a two-stage alternative (precontemplation and contemplation/action; Willoughby & Edens, 1996). Furthermore, the transtheoretical model of change may not be relevant to all groups of drinkers, such as pregnant women. A few studies have examined the utility of the transtheoretical model among pregnant smokers, but not among pregnant drinkers. A comparison of 103 economically disadvantaged pregnant smokers and 103 matched community smokers found that the majority of pregnant smokers were not highly motivated to quit smoking simply because they were pregnant (Ruggiero, Tsoh, Everett, Fava, & Guise, 2000). A study of 54 low-income pregnant smokers randomized to either motivational intervention or usual care found that stage of change did not progress in the course of treatment, thus failing to support the investigators’ hypothesis that a motivational intervention would increase motivation for change (Stotts, DeLaune, Schmitz, & Grabowski, 2004). Although findings from a cross-sectional study of 637 pregnant women who were either current or previous cigarette smokers supported the overall relevance of the transtheoretical model of change, the investigators also concluded that a staged approach was less necessary for pregnant women (Slade, Laxton-Kane, & Spiby, 2006). Because no amount of alcohol is safe during pregnancy, a better understanding of why some women avoid alcohol and others do not is therefore highly desirable (Kost, Landry, & Darroch, 1998; Mukherjee, Hollins, & Abou-Salah, 2005). The overall prevalence of any prenatal alcohol consumption is 13%, with rates of binge drinking (defined as five or more drinks on any one occasion) and frequent drinking (defined
as seven or more drinks per week, or five or more drinks on any one occasion) during pregnancy estimated to be about 6% (Centers for Disease Control and Prevention, 2002). Thus, screening and brief interventions for prenatal alcohol use may help to reduce, if not eliminate, fetal risk and to maximize pregnancy outcome (Chang, 2004/2005). The purpose of this study is to evaluate stage of change as a predictor of alcohol use in a sample of alcohol-screenpositive pregnant women who participated in a randomized trial of brief interventions. Thus, if the transtheoretical model is applicable, then pregnant women in the action stage should demonstrate alcohol use less than that of those assigned to other stages.
2. Materials and methods 2.1. Setting This study took place at the Brigham and Women’s Hospital (Boston, MA). A screening survey with questions about health habits such as diet, exercise, sleep, and usual drinking, and the T-ACE questionnaire (a four-item alcohol screening instrument) were given to patients initiating prenatal care at one of three hospital obstetric practices (clinic, faculty, and private group affiliate), as well as to those responding to e-mail and other study announcements. 2.2. Subjects The subjects were 301 (99%) of 304 pregnant women who were enrolled in a randomized trial of brief interventions for prenatal alcohol use (Chang et al., 2005). The women were eligible to participate if they satisfied the entry criteria, which included a positive T-ACE alcohol screen (a score of 2 or more) and risk for prenatal alcohol use, as defined by any of the following: (1) alcohol use while pregnant in the 3 months before study enrollment; (2) consumed at least one drink per day in the 6 months before study enrollment; or (3) drank during a previous pregnancy. The T-ACE questionnaire is a four-item screening questionnaire based on the CAGE questionnaire that asks: How many drinks does it take to make you feel high (tolerance, T)? Have people ever annoyed you by criticizing your drinking (annoyed, A)? Have you ever felt you ought to cut down on your drinking (cut down, C)? Have you ever had a drink first thing in the morning to steady your nerves or to get rid of an hangover (eye opener, E)? The tolerance question is given two points if the respondent states that more than two drinks are needed to feel the effects, and affirmative replies to the A, C, and E questions
G. Chang et al. / Journal of Substance Abuse Treatment 32 (2007) 105–109
are each given one point. The T-ACE questionnaire has been well studied and validated in diverse clinical populations (Chang et al., 1998; Sokol, Martier, & Ager, 1989). Participants gave written informed consent for this study, which was reviewed and approved by the Institutional Review Board of the Brigham and Women’s Hospital. A Certificate of Confidentiality for the project was granted by the Department of Health and Human Services. 2.3. Interviews and intervention The participants completed two interviews. The first interview was conducted upon study enrollment, which occurred at a median of 11.5 weeks’ gestation. The pregnant participants completed: (1) the RCQ, a 12-item instrument that assigns nontreatment-seeking people to the precontemplation, contemplation, or action stage of change based on how they feel about their drinking at present (in the case of the participants, this would include the first trimester of pregnancy; Rollnick et al., 1992); (2) the Alcohol Timeline Follow Back (TLFB), a tool to obtain estimates of daily drinking for the 6 months before study enrollment, which would have included drinking before pregnancy and during early pregnancy (Sobell & Sobell, 1992); and (3) the Alcohol Abstinence Self-Efficacy (AASE) Scale, a tool to evaluate an individual’s perceived temptation to drink (a measure of cue strength) and efficacy (confidence) in abstaining from drinking in 20 common situations at present (in the first trimester; DiClemente, Carbonari, Montgomery, & Hughes, 1994). Scores on the temptation and confidence scales range from a minimum of 20 to a maximum of 100. The second interview occurred at an average of 78 days after delivery. At that time, 95% of the participants completed the Alcohol TLFB interview to provide information about alcohol consumption from the time of study enrollment until delivery. Randomly selected participants received a brief intervention after the first interview. Master’s-level nurses or the first author gave brief interventions to randomly selected participants and reviewed the impact of health habits such as smoking and drinking on pregnancy outcome in the course of 25 minutes. Two main results have been reported. First, brief interventions for prenatal alcohol reduced subsequent consumption most significantly for women with the highest consumption initially. Second, the effects of brief interventions were significantly enhanced when a partner participated. Additional details are available elsewhere (Chang et al., 2005). 2.4. Data analysis Data were analyzed using univariate and multivariate techniques with SAS 8.2 (SAS Institute, Cary, NC). Descriptive statistics are reported as percentages and means. The characteristics of pregnant women by stage of change were compared using the t-test or the chi-square test, as
107
appropriate. The frequency and the quantity of prenatal alcohol use after study enrollment were the dependent variables in ordinary least squares regression models. The models included demographic data (e.g., age, marital status, education, and race), clinical history (e.g., first pregnancy and history of obstetric problems), prior drinking history (as lagged dependent variable), and other study variables (e.g., brief intervention status and AASE temptation score) as predictors. Multiple imputations (five imputations) were used to manage missing data. All analyses were replicated with mean substitution to verify findings from multiple imputations (Rubin, 1987). The purpose of multivariate analyses was to focus on the impact of stage of change on the quantity and the frequency of alcohol use in this sample of pregnant women. Because the primary purpose of this study is limited to investigating these two relationships (stage of change vs. quantity and frequency), no formal correction for multiple testing was used.
3. Results Three hundred four participants were assigned to the precontemplation (61.5%), contemplation (1%), and action stages (37.5%) based on RCQ responses for drinking at present (in the first trimester). Results will focus on women
Table 1 Demographic data, clinical history, and drinking characteristics, by stage of change Variables
Precontemplation (n = 187)
Action (n =114)
Mean age (years) 30.5 30.6 Married (%) 77.6 86.5 African American (%) 6.4 6.3 Mean years 16.5 16.7 of education First pregnancy (%) 48.7 35.4 History of obstetric 20.1 31.8 problems (%) AASE score Confidence 93.8 93.7 Temptation 27.4 30.0 Drinking frequency (mean % drinking days) Before pregnancy 18.6 13.4 Early pregnancy, 2.9 2.1 before enrollment Pregnancy, 2.1 2.0 after enrollment After pregnancy 11.6 10.2 Drinking quantity (mean drinks per drinking day) Before pregnancy 2.1 1.6 Early pregnancy, 0.43 0.38 before enrollment Pregnancy, 0.34 0.37 after enrollment After pregnancy 1.45 1.02 Abstinent after 57 52 enrollment (%)
p (t-test or chi-square) .67 .08 .96 .33 .04 .05
.95 .04 .02 .31 .89 .49 b .01 .65 .68 b .01 .44
108
G. Chang et al. / Journal of Substance Abuse Treatment 32 (2007) 105–109
Table 2 Predictors of prenatal alcohol consumption after study enrollment
Predictors Intercept Action stage Brief intervention Interaction between action stage and brief intervention African American History of obstetric problems First pregnancy Married Age in years Education in years Previous alcohol consumption Temptation
Drinking frequency (% drinking days)
Drinking quantity (drinks per drinking day)
B
B
Pr
predictive of either the subsequent frequency or the quantity of prenatal alcohol use. The results of the analysis are summarized in Table 2.
Pr
0.31 0.11 0.096 0.95
.33 .27 .31 .53
8.0 1.03 0.20 1.4
.0008 .17 .76 .20
0.10 0.03
.50 .74
1.0 0.81
.37 .22
0.04 0.07 0.00004 0.0004 0.54
.60 .49 .09 .72 .17
0.13 0.18 0.0003 0.021 0.49
.83 .80 .036 .023 b .0001
0.008
.03
0.026
.33
in the precontemplation (n = 187) and action (n = 114) stages. When compared to those assigned to the action stage, precontemplation-stage women were significantly more likely to be pregnant for the first time (48.7% vs. 35.4%), with a less frequent history of obstetric problems (20.1% vs. 31.8%) and with less temptation to drink (27.4 vs. 30.0). However, there were no other systematic differences between the two groups in terms of age, marital status, racial background, years of education, or gestational age upon enrollment. With regards to drinking, precontemplation-stage women consumed significantly more alcohol before pregnancy, for both frequency (18.6% vs. 13.4%) and quantity (2.1 vs. 1.6 drinks per drinking), when compared to action-stage women. The precontemplation-stage women also had significantly more drinks per drinking day (1.5 vs. 1.0), but not more frequent drinking after pregnancy, than the action-stage women. Women from both stages reduced the quantity and the frequency of their alcohol consumption while pregnant and achieved comparable rates of abstinence. Table 1 summarizes the comparison between the women assigned to the precontemplation stage and the women assigned to the action stage. The stage of change was compared to demographic data (e.g., age and years of education), clinical history (e.g., past alcohol use and obstetric history), and AASE temptation score as predictors of the frequency and the quantity of prenatal alcohol use after study enrollment in regression models. The AASE temptation score was included because it differed by stage-of-change group, but the confidence score did not. Three predictors of drinks per drinking day were identified: age in years ( p = .04), education in years ( p = .02), and past drinking ( p b .0001). Temptation to drink was the only statistically significant predictor of drinking frequency ( p = .03). Stage of change was not
4. Discussion Stage of change distinguished between different patterns of alcohol consumption prior to pregnancy in this sample of 301 pregnant women. Those designated to be in the precontemplation stage drank more per episode and more often before pregnancy than those in the action stage. However, stage of change did not directly predict subsequent prenatal alcohol use. Previous alcohol use, age, and education were the most significant predictors of prenatal drinks per drinking day. Temptation to drink alcohol was the best predictor of prenatal drinking frequency after study enrollment. The precontemplation group also had a significantly greater quantity of alcohol after pregnancy. Explanations for the lack of direct association between stage of change and subsequent prenatal alcohol use are speculative. A possible explanation is that, although stage designation was associated with prepregnancy drinking, the study allowed a prospective evaluation with minimal loss to follow-up—two factors that have been cited as shortcomings in other investigations (DiClemente et al., 1999). Unexpectedly, the action group reported more temptation to drink, perhaps because they had made greater reductions in alcohol consumption. Temptation to drink has been found to predict responses to alcohol cues, even when the effects of typical drinking patterns were taken into account (Palfai, 2001). Another possible explanation may involve the very high levels of confidence expressed by both groups of women in managing their drinking. Half of the participants in the study sample became abstinent after study enrollment. Finally, it is possible that the difficulties in measuring stage of change are exacerbated when pregnant women are being assessed. Pregnant women have long been considered to be highly motivated to alter behaviors such as smoking (Boyd, 1985). Potential limitations to the generalizability of the study findings include sample characteristics. Higher education, non-Hispanic background, and employment have been shown to be associated with greater prenatal alcohol use (Centers for Disease Control and Prevention, 2002). Participants did not appear to be heavy drinkers, but they drank in excess of all recommendations for abstinence during pregnancy. Underreporting of alcohol use is possible. Reactivity to research protocols and regression to the mean are possibilities to consider as well (Clifford & Maisto, 2000; Fleming, Barry, Manwell, Johnson, & London, 1997). Finally, stage of change was not measured serially; thus, it was not possible to ascertain stage progression in this sample. Although stage-of-change designation did differentiate between prepregnancy levels of alcohol use, it was not
G. Chang et al. / Journal of Substance Abuse Treatment 32 (2007) 105–109
predictive of subsequent prenatal consumption. Thus, the definitive role of stage of change as a predictor of prenatal drinking behavior requires further study. Nonetheless, because no amount of alcohol consumption can be considered safe during pregnancy, results from this study support the importance of identifying previous patterns of alcohol use, even among those who appear to enjoy the apparent advantages of age and education (Office of the Surgeon General, 2005). References Allsop, S. (2003). Addiction and change: How addictions develop and addicted people recover (Book review). Addiction, 98, 1782 – 1783. Boyd, M. D. (1985). Patient motivation during pregnancy. Maryland Medical Journal, 34, 977 – 981. Budd, R. J., & Rollnick, S. (1996). The structure of the Readiness to Change Questionnaire: A test of Prochaska and DiClemente’s transtheoretical model. British Journal of Health Psychology, 1, 365 – 376. Centers for Disease Control and Prevention. (2002, April). Alcohol use among women of childbearing age—United States, 1991–1999. Morbidity and Mortality Weekly Report, 51, 273 – 276. Chang, G. (2004/2005). Screening and brief intervention in prenatal care settings. Alcohol Research and Health, 28, 80 – 84. Chang, G., McNamara, T. K., Orav, E. J., Koby, D., Lavigne, A., Ludman, B., et al. (2005). Brief intervention for prenatal alcohol use: A randomized trial. Obstetrics and Gynecology, 105, 991 – 998. Chang, G., Wilkins-Haug, L., Berman, S., Goetz, M. A., Behr, H., & Hiley, A. (1998). Alcohol use and pregnancy: Improving identification. Obstetrics and Gynecology, 91, 892 – 898. Clifford, P. R., & Maisto, S. A. (2000). Subject reactivity effects and alcohol treatment outcome research. Journal of Studies on Alcohol, 61, 787 – 793. Defuentes-Merillas, L., Dejong, C., & Schippers, G. (2002). Reliability and validity of the Dutch version of Readiness to Change Questionnaire. Alcohol and Alcoholism, 37, 93 – 99. DiClemente, C. C. (2003). Addiction and change: How addictions develop and addicted people recover (pp. 3 – 61). New York, NY7 Guilford Press. DiClemente, C. C., Bellino, L. E., & Neavins, T. M. (1999). Motivation for change and alcoholism treatment. Alcohol Health and Research, 23, 86 – 92. DiClemente, C. C., Carbonari, J. P., Montgomery, R. P. G., & Hughes, S. O. (1994). The Alcohol Abstinence Self-Efficacy scale. Journal of Studies on Alcohol, 55, 141 – 148. Fleming, M. F., Barry, K. L., Manwell, L. B., Johnson, K., & London, R. (1997). Brief physician advice for problem alcohol drinkers. Journal of the American Medical Association, 277, 1039 – 1045. Forsberg, L., Halldin, J., & Wennberg, P. (2003). Psychometric properties and factor structure of the Readiness to Change Questionnaire. Alcohol and Alcoholism, 38, 279 – 280. Hannover, W., Thyrina, J. R., Hapke, U., Rumpf, H. J., Meyer, C., & John, U. (2002). The Readiness to Change Questionnaire in subjects with hazardous alcohol consumption, alcohol misuse and dependence in a general population survey. Alcohol and Alcoholism, 37, 362 – 369.
109
Kavanagh, D. J., Sitharthan, T., & Sayer, G. P. (1996). Prediction of results from correspondence treatment for controlled drinking. Addiction, 91, 1539 – 1545. Kost, K., Landry, D. J., & Darroch, J. E. (1998). Predicting maternal behaviors during pregnancy: Does intention status matter? Family Planning Perspectives, 30, 79 – 88. Maisto, S. A., Conigliaro, J., McNeil, M., Kraemer, K., Conigliaro, R. L., & Kelley, M. E. (2001). Effects of two types of brief interventions and readiness to change on alcohol use in hazardous drinkers. Journal of Studies on Alcohol, 62, 605 – 614. Mukherjee, R. A. S., Hollins, S., & Abou-Salah, M. T. (2005). Low level alcohol consumption and the fetus. British Medical Journal, 330, 375 – 376. Office of the Surgeon General (2005, February 21). The surgeon general releases advisory on alcohol use in pregnancy (Press release). Bethesda, MD7 Department of Health and Human Services (Available at www.hhs.gov/surgeongeneral/pressreleases/sg02222005.html). Palfai, T. P. (2001). Individual differences in temptation and responses to alcohol cues. Journal of Studies on Alcohol, 62, 657 – 666. Prochaska, J. O., DiClemente, C. C., & Norcross, J. C. (1992). In search of how people change: Applications to predictive behaviors. American Psychologist, 47, 1102 – 1114. Riemsa, R. P., Pattenden, J., Bridle, C., Sowden, A. J., Mather, L., & Walker, A. (2002). A systematic review of the effectiveness of interventions based on a stages of change approach to promote individual behaviour change. Health Technology Assessment, 6, 1 – 55. Rodriguez-Martos, A., Rubio, G., Auba, G., Santo Domingo, J., Rorralba, L., & Campilo, M. (2000). Readiness to Change Questionnaire: Reliability study of the Spanish version. Alcohol and Alcoholism, 35, 270 – 275. Rollnick, S., Heather, N., Gold, R., & Hall, W. (1992). Development of a short bReadiness to ChangeQ questionnaire for use in brief opportunistic brief interventions. British Journal on Addiction, 87, 745 – 754. Rubin, D. B. (1987). Multiple imputation for nonresponse in surveys. New York, NY7 Wiley and Sons. Ruggiero, L., Tsoh, J. U., Everett, K., Fava, J. L., & Guise, B. J. (2000). The transtheoretical model of smoking: Comparison of pregnant and nonpregnant smokers. Addictive Behaviors, 25, 239 – 251. Slade, P., Laxton-Kane, M., & Spiby, H. (2006). Smoking in pregnancy: The role of the transtheoretical model and the mother’s attachment to the fetus. Addictive Behaviors, 31, 743 – 757. Sobell, L. C., & Sobell, M. B. (1992). Timeline followback: A technique for assessing self-reported alcohol consumption. In J. P. Allen, & R. Z. Litten (Eds.), Measuring alcohol consumption: Psychosocial and biochemical methods (pp. 41 – 72). Totowa, NJ7 Humana Press. Sokol, R. J., Martier, S. S., & Ager, J. W. (1989). The T-ACE questions: Practical prenatal detection of risk-drinking. American Journal of Obstetrics and Gynecology, 160, 863 – 871. Stotts, A. L., DeLaune, K. A., Schmitz, J. M., & Grabowski, J. (2004). Impact of a motivational intervention on mechanisms of change in lowincome pregnant smokers. Addictive Behaviors, 29, 1649 – 1657. Sutton, S. (2001). Back to the drawing board? A review of applications of the transtheoretical model to substance use. Addiction, 96, 175 – 186. West, R. (2005). Time for a change: Putting the transtheoretical (stages of change) model to rest. Addiction, 100, 1036 – 1039. Willoughby, F. W., & Edens, J. F. (1996). Construct validity and predictive utility of the stages of change for alcoholics. Journal of Substance Abuse, 8, 275 – 291.