Skip to main content
NCBI home page
NIHPA Author Manuscripts logoLink to NIHPA Author Manuscripts
. Author manuscript; available in PMC: 2017 Sep 1.
Published in final edited form as: J Consult Clin Psychol. 2016 Sep;84(9):803–812. doi: 10.1037/ccp0000111

Processes of Change in Preventing Alcohol Exposed Pregnancy: A Mediation Analysis

Danielle E Parrish 1, Kirk von Sternberg 2, Yessenia Castro 3, Mary M Velasquez 4
PMCID: PMC5061601  NIHMSID: NIHMS806794  PMID: 27176661

Abstract

Objective

To examine mechanisms of the treatment effect for CHOICES, a motivational intervention to reduce risk of alcohol exposed pregnancy (AEP). Grounded in constructs from the Transtheoretical Model (TTM) and Motivational Interviewing (MI), the intervention targeted: risk drinking (>4 drinks/day or >7 drinks/week); ineffective contraception; and AEP risk (both behaviors). The experiential and behavioral processes of change (POC), posited to describe the mechanisms through which individual behavior change occurs, were examined. It was hypothesized that each of the targeted treatment outcomes at 9-month follow-up would be mediated by the experiential POC at 3-months, and that these would then be mediated by the behavioral POC at 9-months.

Method

830 women at-risk for AEP were randomized to CHOICES (Information Plus Counseling; IPC) condition (n=416) or Information Only (IO) condition (n=414). Primary outcomes and proposed mediators (POC) were assessed at 3- and 9-months. Path analyses using weighted least squares estimation with mean- and variance-adjusted chi-square statistic were conducted separately for each outcome.

Results

Model fit indices indicated good fit, and the indirect effect of treatment on outcome via POC was significant for hypothesized models predicting risky drinking and ineffective contraception. The indirect effect of treatment on AEP risk through POC for ineffective contraception was significant, but the indirect effect of POC for risky drinking was not.

Conclusions

These findings support the temporal relationship between experiential and behavioral POC consistent with the TTM. Opportunistic, motivation-based interventions may benefit from directly targeting experiential POC early in treatment and behavioral POC later in treatment.

Keywords: alcohol-exposed pregnancy, mediation, Transtheoretical model, processes of change, alcohol

Alcohol consumption during pregnancy is the leading preventable cause of developmental disabilities and birth defects in the United States (CDC, 2012; Wiliams & Smith, 2015). Alcohol-exposed pregnancies (AEP) can result in fetal alcohol spectrum disorders (FASD), which include Fetal Alcohol Syndrome (FAS) at the more severe end of the spectrum. The lifetime consequences of FASD can include all or some of the following: intellectual disability, behavioral and emotional challenges, learning disabilities, difficulty with executive functioning, abnormal facial features, speech and language delays, growth deficits, and other medical issues (Behnke & Smith, 2013; Sokol, Delaney-Black, & Nordstrom, 2003). FASD prevalence rates are estimated at 24–48 per 1000 children age 6–7 years or 2.4–4.8%, while more severe FAS rates have ranged between 6 to 9 cases per 1000 children (May et al, 2014). A recent study using a large, probability sample estimated that almost 2 million U.S. women were at risk of an AEP during a 1 month period (Cannon et al., 2014). While it is not uncommon for women to reduce alcohol consumption once they learn they are pregnant, over half of all pregnancies in the U.S. are unplanned (Finer & Zolna, 2014). Consequently, many women consume alcohol well into the first trimester before realizing they are pregnant. Risky drinking during this early and critical period of fetal development can result in FASD, suggesting a need to intervene during the preconception period to reduce risky drinking and/or improve effective contraception (Floyd, Decouflé, & Hungerford, 1999; Nykjaer et al., 2014).

Project CHOICES, a line of research funded by the Centers for Disease Control and Prevention (CDC), has developed and tested the efficacy of the CHOICES intervention to reduce the risk of AEP by targeting two risk behaviors – risky alcohol consumption and lack of effective contraception to prevent unplanned pregnancy (Floyd, et al., 2007; Project CHOICES Intervention Research Group, 2003; Project CHOICES Research Group, 2010; Velasquez, et al., 2010). In these studies, a woman was identified as at risk of AEP if she was 18–44 (childbearing age), and within the last 90 days, was sexually active with a man, fertile, did not consistently use effective contraception as defined by the American Congress of Obstetricians and Gynecologists (ACOG; 2013) and was drinking at risky levels identified by the National Institute on Alcohol Abuse and Alcoholism definition of risky drinking at the time of the study (i.e. more than 4 standard drinks in a day, or more than sevendrinks in a week). The CHOICES intervention, referred to in the study as information plus counseling (IPC), was efficacious in reducing the risk of AEP in a large, multisite randomized controlled trial with women of childbearing age during the preconception period (Floyd, et al., 2007). Sixty-nine percent of the women in the IPC treatment condition were at reduced risk of AEP at 9 months and were more than 2 times more likely to be at reduced risk of AEP at 3, 6, and 9 months compared to women in the information only (IO) control condition (Floyd, et al, 2007). Given the strong impetus within the social and health sciences to improve our understanding of the mechanisms of change (Apodaca & Longabaugh, 2009; DiClemente, 2007; Kazdin, 2007), this manuscript reports on the analysis of the aforementioned CHOICES efficacy study data to test whether theoretically-based psychosocial mechanisms mediate the treatment on the reduction of the risk of AEP.

The CHOICES IPC Intervention utilizes a Motivational Interviewing (MI) counseling approach and incorporates the Transtheoretical Model (TTM), which offers an integrative model or “template” for both developing intervention approaches/materials and for examining behavior change across disparate individuals, target behaviors and contexts (DiClemente, 2005, p. 9; Velasquez, et al, 2010; Connors, DiClemente, Velasquez, & Donovan, 2013). The TTM has played a critical role in the conceptual development of MI and other brief interventions that utilize motivational approaches (DiClemente & Velasquez, 2002). The TTM posits that individuals move through various stages of change when changing behavior, including precontemplation, contemplation, preparation, action and maintenance. Within each stage of change, there are corresponding processes of change or critical tasks associated with behavior modification (DiClemente, 2007). As described by DiClemente (2003), these processes “represent the internal and external experiences and activities that enable individuals to move from stage to stage” (p. 32). As these experiences and activities occur, individuals progress from being unaware of a problem or unwilling to change, to perhaps beginning to consider a change or considering a change (precontemplation/contemplation), to then making a change and maintaining it (action/maintenance) (DiClemente & Velasquez, 2002).

Interventions incorporating tenets of the TTM are theoretically designed to activate or maintain these internal or environmental processes of change, providing fundamental units in the study of the mechanisms of behavior change (DiClemente, 2005; DiClemente, 2007). Grounded in the TTM, and as shown in Table 1, the CHOICES IPC intervention, depending on the participant’s stage of readiness to change, focused on reducing the risk of AEP by using materials and motivational counseling strategies to assist women in moving through the processes of change – which are identified as either experiential or behavioral – to reduce risky drinking and increase effective contraception (a change in one or both behaviors reduced overall risk of AEP).

Table 1.

Project CHOICES Intervention and TTM Processes of Change

Processes of Change (POC) CHOICES Treatment Components Target POC
Experiential Processes
 Consciousness Raising • Standardized feedback
• Introduction to fact sheets on AEP prevention (risky drinking and pregnancy; and birth control)
• Brochures on alcohol risks and contraception
• Daily journal to record risk behavior
 Dramatic Relief • Standardized feedback
• Introduction to fact sheets on AEP prevention
• Decisional balance exercise
 Environmental Reevaluation • Decisional balance exercise with counselor
 Self-Reevaluation • Decisional balance exercise with counselor
 Social Liberation • Decisional balance exercise with counselor
• Goal change statement; utilization of individual and environmental resources to reach goal
Behavioral Processes
 Reinforcement Management • Reinforcement of change talk, attendance of contraceptive visit, and behavior change on the part of the counselor.
 Helping Relationships • Trained counselor uses Motivational Interviewing techniques
 Counterconditioning • Referral for contraceptive visit to explore birth control options
• Self-evaluation (readiness) rulers
• Goal change statement
 Stimulus Control • Self-evaluation (readiness) rulers
• Revisit decisional balance exercises
• Goal change statement (planning to avoid or alter trigger situations)
 Self-Liberation • Goal change plan for alcohol and contraception
Open in a new tab

The processes of change were originally identified in a comparative analysis of 18 leading systems of psychotherapy, where 10 independent processes of change were identified–five cognitive/experiential and five environmental/behavioral (Prochaska, 1979). This development, as well as additional research exploring “self-changers” or those who improved without professional services, led to the recognition of the complexity inherent in an individual’s behavioral change process and the development of the TTM (DiClemente & Prochaska, 1982; Prochaska & DiClemente, 1983, 1984). It is posited in the TTM that the experiential processes are more salient in the earlier, pre-action stages of change, and have to do more with an individual’s thought processes, feelings, and unique situational experiences, while behavioral processes are much more action-oriented and strongly related to the later stages of change (DiClemente, 1993; 2005; Perz, DiClemente, & Carbonari, 1996; Prochaska & DiClemente, 1984; Velasquez, Gaddy-Maurer, Crouch, & DiClemente, 2001).

There are several studies that support the temporal relationship of the processes of change, with more experiential process use in pre-action and more behavioral process use associated with actual behavior change (Carbonari & DiClemente, 2000; Perz et.al., 1996; Project MATCH Research Group, 1997; Stotts, DeLaune, Schmitz, & Grabowski, 2004). However, this temporal relationship of the processes of change has not always been consistent across populations, behaviors, and contexts. Recent research on exercise and diet suggests that both experiential and behavioral processes are important early in behavioral change (Dishman, Vandenberg, Mott, & Nigg, 2010; Lewis, Williams, Martinson, Dunsigner, & Marcus, 2013; Plotnickoff, Hotz, Birkett, & Courneya, 2001; Rosen, 2000; Wilson & Schlam, 2004).

Considering the theoretical and empirical advancements, research is needed to better understand why and how interventions based on the TTM’s processes of change work (DiClemente, 2005; DiClemente & Velasquez, 2002; DiClemente, 2007; DiClemente, Carbonari, Zweben, Morrel, & Lee, 2001; Joseph, Breslin, & Skinner, 1999). Exploration of TTM variables gathered in RCTs, particularly when collected before the proposed outcome (Kazdin, 2007), can offer useful information to inform future intervention development that targets concrete behaviors and processes to which both clients and practitioners can relate (DiClemente & Velasquez, 2002). The CHOICES efficacy data set, which utilizes an intervention based on the TTM to target three behavioral outcomes, offers a novel opportunity to test the processes as mechanisms of change and the posited temporal relationship of the experiential processes to the behavioral processes in the reduction of these risks among women of childbearing age during the preconception period.

The goal of the current study was to test whether the CHOICES IPC treatment increased the use of experiential processes and subsequently the use of behavioral processes (see “Additional Measures”), which directly impacted the risk of AEP, a dichotomous variable indicating a combination of risky drinking and ineffective contraception. As such, our objectives were to test whether experiential and behavioral processes of change (in this order) mediate the relationship between treatment group and (a) risky drinking, (b) ineffective contraception, and (c) risk of AEP.

Method

Participants and Participant Characteristics

Participants in Project CHOICES (N = 830) were non-treatment seeking women of childbearing age (18–44). Inclusion criteria included the following: (1) 18–44 years old; (2) no condition causing infertility (tubal ligation, hysterectomy, menopause, or other reason); (3) not pregnant or planning to become pregnant in the next 9 months; (4) had vaginal intercourse during the previous 3 months (or 3 months before going to jail or residential treatment) with a fertile man (not surgically sterile) without using effective contraception (defined in “Outcome Measures”); (5) engaged in risky drinking (defined in Outcome Measures); and (6) available for the follow-up period.

Women at-risk of AEP were recruited from six settings from July 1, 2002 to January 30, 2004, with the follow up period ending on August 15, 2005. Details regarding these settings are provided elsewhere (Floyd et al, 2007), but included jails, residential alcohol and drug treatment facilities, OB/GYN clinics, primary care clinics and community recruitment. All of the women (N=830) were at risk of AEP at baseline (Floyd et al, 2007).

No significant differences were found on demographic or background clinical characteristics between the intervention and control groups at baseline (Floyd et al, 2007). While a detailed description of the study characteristics can be found in the original study (Floyd et al, 2007), participants were primarily African American (48%) and had annual incomes <$20,000 (55%).

Research Design

The Project CHOICES study utilized a two-group parallel RCT, with participants randomized into two groups: information only (IO; the control condition) and CHOICES information plus counseling (IPC; the intervention condition). Women assigned to IO received brochures on alcohol use and women’s health in general, and a referral guide to local resources. Women assigned to IPC received the CHOICES intervention over a 14 week period, with approximately 2 to 3 weeks between sessions. The counseling sessions and the contraception consultation visit were approximately 45 to 60 minutes. Participants were contacted at 3, 6, and 9 months for follow-up assessments. All participants were reimbursed for time and travel (jail inmates received deposits in their commissary accounts). Additional information concerning the implementation of the study protocol can be found in the original published study (Floyd et al, 2007).

Eighty percent of the participants completed the 3-month follow-up interview (n=665), 73% completed the 6-month follow-up interview (n=604), and 71% completed the 9-month follow-up interview (n=593), with approximately equal numbers in treatment and control at each of the phases. As reported in the CONSORT chart from the original manuscript (Floyd et al, 2007), the primary reason for follow-up attrition was the inability to locate women. Intent-to-treat analyses suggested that there was not a bias in results due to attrition at 9 months (Floyd et al, 2007).

Project CHOICES Preconception Intervention

The CHOICES IPC intervention is described in detail elsewhere (Velasquez et al., 2010). In brief, it consisted of four manual guided motivational counseling sessions with a mental health clinician and one contraceptive counseling session with a family planning clinician. The Transtheoretical Model (TTM; Prochaska & DiClemente, 1983; 1984; Prochaska, DiClemente, & Norcross, 1992) guided the development of the CHOICES IPC intervention (see Table 1), with an emphasis on the processes of change, and other TTM constructs such as decisional balance, temptation and confidence (Velasquez et al., 2010). While MI and the TTM were developed independently, they are highly compatible and referred to by Miller & Rollnick (2013) as “kissing cousins”. Sessions targeted both alcohol and contraception and were designed so that the counselor first addressed the behavior the woman was most ready to change, then moved to the other target behavior later in the session or in subsequent sessions. The intervention was designed to target the various processes of change as shown in Table 1.

More than 98% of the women in the CHOICES IPC intervention group received at least one session, and 63% received all four sessions. On average, women in the CHOICES IPC group attended 3.2 counseling sessions, and approximately 70% attended a contraception consultation visit. All of the women in the IO group received the intervention.

Outcome Measures

The primary outcomes included risky drinking, ineffective contraception and risk of AEP (the combination of both risk behaviors). All three variables were dichotomous risk variables. These were measured using the Timeline Followback (TLFB), which has extensive support for its reliability and validity in both clinical and non-clinical populations (Sobell, Maisto, Sobell, & Cooper, 1979; Sobell & Sobell, 1992; Sobell & Sobell, 1995). According to the NIAAA standards at the time, risky drinking was defined as more than 4 standard drinks in a day or more than 7 drinks in a week on average (Gunzerath, Faden, Zakhari, & Warren, 2004; Streissguth et al, 1994; Jacobson, Jacobson, Sokol, Ager, 1998; Konovalov, Kovetshy, Bobryshev, & Ashwell, 1998). Ineffective birth control use was defined as the participant’s self-reported deviations from the accepted, published guidelines for each kind of effective contraception (ACOG, 2013).

All women at baseline reported risky drinking and ineffective contraception use per study inclusion criteria. At follow-up, women were identified as at reduced risk of AEP if they reported no risky drinking, effective contraception, or both. Women were identified as at reduced risk (yes/no) rather than at no risk due to the potential of failure associated with each contraceptive method. The TLFB method provided a continuous record of daily drinking, vaginal intercourse, contraception use and effectiveness from 90 days prior to enrollment to 9 months post-enrollment.

At each point of data collection (baseline, 3 month, 6 month, and 9 month) participants provided detailed information about these behaviors for the prior 90 days. For analysis, these data were subdivided into 30 day increments over each 90 day period to calculate the following outcomes: if a woman consumed more than 4 drinks on any day, or on average more than 7drinks per week during the 30 day period, she was categorized as a risky drinker (yes/no); if she had the occurrence of vaginal intercourse without effective contraception, she was categorized as an ineffective contraceptive user for the 30 day period (yes/no). Women engaging in risky drinking and using ineffective contraception for the 30 day period were considered at risk of AEP for that 30 day period (yes/no). If a woman was at risk of AEP for any 30 day period within the 90 day period, she was considered at risk for AEP for that 90 day period.

Additional Measures (Including Study Mediators)

Women were assessed in person at baseline, 3 months and 9 months, and with an abbreviated assessment of main outcomes at 6 months by telephone. In addition to the aforementioned primary outcomes, other measures captured demographic information and psychosocial background variables, as reported in Floyd et al (2007). The Processes of Change for Alcohol (Prochaska, Velicer, DiClemente, & Fava, 1988; DiClemente, Carbonari, Addy, & Velasquez, 1996) 40-item scale was also administered at baseline, 3-, and 9-month time points, as was the 26 item Processes of Change for Contraception scale, which was adapted from the alcohol version. The adaptation included each of the 10 processes of change and involved expert consensus on final item pool and exploratory and confirmatory factor analyses (Floyd et al. 2007). The processes of change instruments are specific to each targeted health behavior (in this case alcohol use and contraceptive use), and measure, on a 5-point Likert scale, the covert and overt activities and experiences that individuals engage in when they are changing these risk behaviors. The items on the Processes of Change measures load onto five latent factors that represent the five experiential processes: consciousness raising, dramatic relief, environmental reevaluation, self-reevaluation, and social liberation and five latent factors that represent the five behavioral processes: reinforcement management, counter-conditioning, helping relationships, self-liberation, and stimulus control. The five experiential factors load onto the second order latent factor, experiential processes. The five behavioral factors load onto the second order latent factor, behavioral processes. An example of a self-reevaluation (experiential process) item from the processes of change for alcohol scale reads, “I consider that feeling good about myself includes changing my drinking behavior.” Likewise, a dramatic relief (experiential process) item from the processes of change for contraception scale reads, “I get upset when I think about getting pregnant when I don’t want to.” An example of stimulus control (behavioral process) item from the processes of change for alcohol scale reads, “I avoid situations that encourage me to drink”. Another example of a behavioral process item from the processes of change for contraception that loads onto the counter conditioning scale reads, “When I’m tempted to have sex without using birth control, I think about what could happen and do something else.” The alcohol processes of change measure has good internal consistency reliability with subscale alphas ranging from .82–.85 (von Sternberg, 2005), as does the contraception processes of change measure, which ranges from .86–.88. In confirmatory factor analyses for the adapted contraception processes of change, both the experiential (CFI=.958; RMSEA=.078) and the behavioral POC (CFI=.965; RMSEA=.073) had a moderate fit (unpublished). All data were collected in compliance with the IRBs at multiple sites described in the original outcome study (Floyd et al, 2007).

Data Analysis

The overall aim of this analysis is to further understand how the CHOICES IPC Intervention worked, and whether it worked in the way theoretically intended. We hypothesized that the CHOICES IPC would impact both alcohol and contraception experiential and behavioral processes, in this order, to reduce risky drinking and increase effective contraception respectively (although a change in one or both behaviors reduced overall risk of AEP). These mediator variables were collected at baseline, 3 months and 9 months for each behavior (risky drinking and effective contraception) as a part of this RCT, supporting the use of path analysis. Specifically, a weighted least squares estimation with a mean- and variance adjusted chi-square statistic (WLSMV) was conducted using the Mplus software program (Muthen & Muthen, 2012). Model fit was assessed by examining the model chi-square as well as by examining the following model fit indices consistent with the recommendations of Kline (2005): comparative fit index (CFI) and Tucker-Lewis Index (TLI) > .90; root mean square error of approximation (RMSEA) < .08. A weighted root mean square residual (WRMR) < 0.9 (Schreiber, Nora, Stage, Barlow, & King, 2006) was also considered. Non-nested models were compared by re-estimating the models using the maximum likelihood estimator in order to obtain the Bayesian Information Criterion (BIC; Schwartz, 1978) for each model. There is no accepted value for the BIC; rather, the BIC of two non-nested models are compared, and differences greater than 10 are interpreted as “very strong” evidence that the model with the smaller BIC is the superior model (Kass & Raftery, 1995; Raftery, 1995). Indirect effects were estimated using the MODEL INDIRECT command in Mplus with bootstrapped 95% confidence intervals (CI) using 5,000 resamples.

As noted in the original study (Floyd et al, 2007), a parsimonious model was identified using a stepwise backward elimination logistic regression procedure that considered both the statistical significance and the standard error of the odds ratio associated with the sociodemographic and behavioral variables that were collected and the intervention effect. This analysis identified six baseline measures that were significant (<.01) and retained in the final model as confounders – number of male intercourse partners in the last 90 days, the total score from the Alcohol Use Disorder Identification Test (AUDIT), Readiness for Change for Contraception, Processes of Change for Alcohol, Decisional Balance for Alcohol, and Temptation for Alcohol. Each of the confounding variables used in the primary analyses from the original CHOICES Efficacy trial was tested to determine the relationship of each to the behavioral outcomes and to the posited mediators in the current study. Only those that were related to all dependent variables in the mediation models were retained (i.e. number of male intercourse partners in the last 90 days and the Alcohol Use Disorder Identification Test (AUDIT) at baseline). The AUDIT is a 10-item brief screening instrument designed to identify individuals with drinking problems at various levels of severity (Allen, Litten, Fertig, & Babor, 1997). Our mediation analyses included all study participants regardless of their dose of treatment (i.e. one to four sessions, with and without a contraception visit). Number of sessions and contraceptive visit attendance were not significantly related to the risk outcomes.

Primary Path Models

We tested the hypothesis that experiential processes are affected early in the change attempt and give way to behavioral processes, which directly affect the outcome. This hypothesis was tested separately for each outcome variable. In each case, treatment group assignment at baseline served as the independent variable. Experiential processes measured at the 3-month follow-up and behavioral processes measured at 9-month follow-up served as the mediating variables. Consistent with the hypotheses, the model for risky drinking consisted of the following two paths: 1) treatment group ➔ 3-month alcohol experiential processes ➔ 9-month alcohol behavioral processes ➔ risky drinking and; 2) treatment ➔ risky drinking. Similarly, the model for ineffective contraception consisted of the two paths: 1) treatment group ➔ 3-month birth control experiential processes ➔ 9-month birth control behavioral processes ➔ ineffective contraception, and; 2) treatment group ➔ ineffective contraception.

The model for AEP included the following three paths: 1) treatment group➔ 3-month alcohol experiential processes ➔ 9-month alcohol behavioral processes ➔ AEP; 2) treatment group ➔ 3-month birth control experiential processes ➔ 9-month birth control behavioral processes ➔ AEP risk, and; 3) treatment ➔ AEP risk. The AEP model additionally included the covariance path between the error terms of the two experiential process measures and the covariance path between the error terms of the two behavioral process measures. This was done to account for any correlation between the two measures not accounted for in the model, but particularly because these are the same measure only adapted for different health risk behaviors.

The Alternative Path Models

An alternative hypothesis is that experiential and behavioral processes have unique direct influences on the outcome (in comparison to combined, dual mediator influences). This hypothesis was tested with multiple mediation models using the experiential and behavioral process measures from the 3-month follow-up time points.

For risky drinking, the model included three paths: 1) treatment ➔ 3-month alcohol experiential processes ➔ 9-month risky drinking; 2) treatment ➔ 3-month behavioral alcohol processes ➔ 9-month risky drinking, and; 3) treatment ➔ 9-month risky drinking. For ineffective contraception, the three paths were: 1) treatment ➔ 3-month birth control experiential processes ➔ 9-month ineffective contraception; 2) treatment 3-month birth control behavioral processes ➔ 9-month ineffective contraception, and; 3) treatment ➔ 9-month ineffective contraception.

For AEP risk, five paths were modeled: 1) treatment ➔ 3-month alcohol experiential processes ➔ 9-month AEP risk; 2) treatment ➔ 3-month alcohol behavioral processes ➔ 9-month AEP risk, 3) treatment ➔ 3-month birth control experiential processes ➔ 9-month AEP risk; 4) treatment ➔ 3-month birth control behavioral processes ➔ 9-month AEP risk, and 5) treatment ➔ 9-month AEP risk. Just as with the previous AEP risk model, the covariance path between the error terms of the two experiential process measures and the covariance path between the error terms of the two behavioral process measures were modeled.

Results

Hypothesized Models

All model fit indices were indicative of good fit. For risky drinking, the fit indices were: χ2 (2) = 2.49, p = .29; RMSEA = .02; CFI =1.0; TLI = .99; WRMR = .29; and for ineffective contraception: χ2 (2) = 1.94, p = .38; RMSEA = .00; CFI =1.0; TLI = 1.0; WRMR = .23; for AEP: χ2 (6) = 5.84, p = .44; RMSEA = .00; CFI =1.0; TLI = 1.0; WRMR = .29. The final models with standardized path coefficients are shown in Figures 1a–c. In addition to good fit of the overall models, the indirect effect of treatment group on outcome via the process variables was significant for the models predicting risky drinking (Figure 1a, standardized estimate = −.02, 95% CI= −.034; −.002) and ineffective contraception (Figure 1b, standardized estimate = −.05, 95% CI = −.08; −.02). In both models, treatment group assignment at baseline was positively associated with experiential processes at the 3-month follow-up time point, which in turn was positively associated with behavioral processes at the 9-month follow-up time point, which was associated with reduced risk of the outcome at the 9-month follow-up time point. As depicted in Figure 1c, the indirect effect of treatment group on AEP through the birth control processes was significant (standardized estimate = −.03, 95% CI = −.05; −.01), but the indirect effect through the alcohol processes was not significant (standardized estimate = −.01, 95% CI = −.017; .002). The pattern of relations for the significant indirect effect on AEP was the same as those for risky drinking and ineffective contraception.

Figure 1a. The Final Model for Risky Drinking.

Figure 1a

Open in a new tab

*p<.05; **p<.01;

POC=processes of change;

Alcohol Experiential POC were assessed at 3 months post-intake;

Alcohol Behavioral POC were assessed at 9 months post-intake;

Risky Drinking = >4 drinks on any day or >7drinks in a week on average in previous 90 days assessed at 9 months post-intake;

Analyses controlled for baseline scores on the Alcohol Use Disorders Identification Test and number of male partners in the past 90 days.

Figure 1c. The Final Model for Alcohol Exposed Pregnancy Risk.

Figure 1c

Open in a new tab

*p<.05; **p<.01;

POC=processes of change;

Alcohol Experiential POC were assessed at 3 months post-intake;

Alcohol Behavioral POC were assessed at 9 months post-intake;

Contraception Experiential POC were assessed at 3 months post-intake;

Contraception Behavioral POC were assessed at 9 months post-intake;

Alcohol-Exposed Pregnancy Risk = Any risky drinking (>4 drinks on any day or >7drinks in a week on average) and any vaginal sex without effective contraception in previous 90 days assessed at 9 months post-intake.

Note: 1) Analyses controlled for baseline scores on the Alcohol Use Disorders Identification Test and number of male partners in the past 90 days. For ease of viewing, these paths are not shown. 2) Standardized path coefficients are presented. 3) The full mediational path for alcohol POC (not shown): treatment ➔ 3-month alcohol experiential processes ➔ 9-month alcohol behavioral processes ➔ 9-month AEP risk was not significant (standardized estimate = −.01, 95% CI = −.017; .002).

Figure 1b. The Final Model for Ineffective Contraception.

Figure 1b

Open in a new tab

*p<.05; **p<.01;

POC=processes of change;

Contraception Experiential POC were assessed at 3 months post-intake;

Contraception Behavioral POC were assessed at 9 months post-intake;

Ineffective Contraception = any vaginal sex without effective contraception in previous 90 days assessed at 9 months post-intake;

Analyses controlled for baseline scores on the Alcohol Use Disorders Identification Test and number of male partners in the past 90 days.

Alternative Models

None of the three alternative models demonstrated good fit, suggesting the temporal order proposed in the aforementioned hypothesized model better explains the data. Fit indices for the alternative models were as follows: risky drinking: χ2(1) = 153.05, p < .001; RMSEA = .49; CFI = .54; TLI = .00; WRMR = 3.0; ineffective contraception: χ2(1) = 135.8, p < .001; RMSEA = .46; CFI = .33; TLI = .00; WRMR = 2.83; AEP: χ2(4) = 275.73, p < .001; RMSEA = .33; CFI =.50; TLI = .00; WRMR = 3.45. In addition, all BIC for the hypothesized models were substantially smaller than those of the alternative models (3795 vs. 3926 for risky drinking, 3955 vs. 4102 for ineffective contraception, and 6937 vs. 6997 for AEP). This indicates the hypothesized models were a superior fit to the data in all cases. In light of these results, the alternative models are not further discussed.

Discussion

The goal of this study was to test an aspect of the Transtheoretical Model (TTM) guiding the CHOICES IPC Preconception Intervention – whether experiential processes of change (POC) are affected early in the change process, subsequently impacting behavioral processes of change, which then reduce the risk of AEP. We were able to confirm this pattern in a large sample of non-treatment seeking women of childbearing age. Specifically, we found that the indirect effect of treatment group assignment (CHOICES IPC vs. IO Treatment as Usual), mediated by the experiential and then behavioral processes of change variables (in this order), was significant in predicting three different outcomes – risky drinking, contraception use and risk of AEP. Given the reliance of the CHOICES IPC intervention on the Transtheoretical Model of Behavior Change (TTM) and its emphasis on targeting POC, this finding supports this model’s tenet that experiential processes of change occur earlier during behavior change than behavioral processes (DiClemente, 2005; DiClemente & Velasquez, 2002). Likewise, it suggests the CHOICES preconception intervention is working as intended with non-treatment seeking women of childbearing age by assessing their readiness for change and then engaging them in activities and presenting materials relevant to completing stage tasks and moving forward in the processes of change. As such, given the efficacy of CHOICES (Floyd et al, 2007) and this study’s results, TTM guided motivation-based interventions may benefit from directly targeting experiential processes of change early in the intervention process with non-treatment seeking populations, while targeting behavioral POC in subsequent sessions (DiClemente & Velasquez, 2002). However, further research is needed to assess whether these findings hold true for other health behaviors among both men and women, and for both treatment and non-treatment seekers.

In the CHOICES IPC condition – consistent with recent research by Yin and colleagues (2013) suggesting that TTM interventions are more likely to result in paired action (the change of two targeted behaviors being more common than one) compared to control conditions- a larger proportion of women no longer at risk of AEP (47.3%) chose to reduce their risky drinking and engage in effective contraception use compared with 34.8% who did both in the control condition. Similar proportions of women in both groups chose to change only their contraceptive behavior (32.8% in the CHOICES IPC condition and 31.1% in the IO control group).

The indirect effect between group assignment via experiential and then behavioral processes of change on risky drinking was significant in the risky drinking model, but was not significant in the AEP risk model that incorporated both the alcohol POC and the contraception POC. More women changed both risk behaviors than changed either of the risk behaviors alone and were, therefore, engaging in the POC for two behaviors instead of just one. Either one of the behaviors (reducing risky drinking and using effective contraception) could lead to reduced risk of AEP. As such, the variance in the reduced risk of AEP associated with the use of the processes of change was shared between the alcohol POC and the contraception POC. This could explain why the alcohol POC mediational path was not a significant contributor to reduced risk of AEP.

This study, which used data from a large randomized controlled trial that identified an efficacious intervention, satisfies the following requirements for treatment mediators identified by Kazdin (2007): strong association, specificity, consistency, experimental manipulation, timeline, plausibility and coherence. In addition, it provides further research on an under-researched, yet potentially valuable aspect of the TTM – the role of the processes of change in targeted health behavior change (Armitage, 2009; DiClemente, 2005; Nigg et al. 2011; Connors, DiClemente, Velasquez & Donovan, 2013). Building upon studies that suggest processes of change are highly predictive of health behavior change (Aveyard et al., 2006; Carbonari & DiClemente, 2000; Project MATCH Research Group, 1997; Prochaska, Velicer, DiClemente, & Fava, 1988; Stotts, DeLaune, Schmitz, & Grabowski, 2004), this study also suggests that change is a process consisting of common and observable intermediate, short and long-term outcomes that are relevant and important at different times, and that practitioners can partner with clients to build upon and enhance these processes (DiClemente, 2005). Such information is important in guiding the development of interventions that focus on activating these processes of change to achieve health behavior change. While additional insight into the mechanisms of the CHOICES IPC intervention and the TTM guiding its development is provided by this study, additional research is needed to further understand the complex relationships between the processes of change and intentional health behavior change in disparate populations and contexts, particularly when more than one health behavior is targeted (DiClemente, 2005; Nigg et al., 2011; Yin et al., 2013). In addition, future research using statistical controls and experimental manipulation is needed to build the causal literature examining the POC and other TTM variables such as self-efficacy and decisional balance (Kazdin, 2007; Prochaska, Wright, & Velicer; 2008).

While this study offers valuable information to further understand the TTM, one potential limitation is the reliance on self-report as the primary outcome variable, despite its common use in clinical and behavioral research. This concern is at least partially offset by several major reviews (Conigrave, Davies, Haber, & Whitfield, 2003; Connors & Maisto, 2003; Del Boca, Babor, & McRee, 1994; Midanik, 1982; Poikolainen, Podkletnova, & Alho, 2002) that have identified retrospective self-reports of alcohol use as reliable and valid when collected in contexts that minimize bias (e.g., voluntary clients assured of confidentiality). Another potential limitation is the difference in contact time with a therapist in the CHOICES IPC intervention compared to IO control condition. The additional time spent with a therapist in the CHOICES IPC intervention could account for some portion of the treatment effect. While we recognize that other TTM or MI related mediators are important to examine in future research and perhaps this data set, this study provides valuable information in an important area with little research by examining the mediating role of the TTM processes of change (experiences/tasks that must be accomplished) in a TTM guided intervention designed to change health behavior (DiClemente, 2007; Norcross, Krebs, & Prochaska, 2011). Finally, it is important to note that the definition of risky drinking for women has changed since commencement of the original CHOICES efficacy study. At risk drinking for women is now defined as more than 3 drinks a day or more than 7 drinks per week. However, to ensure consistency with the original study, we have maintained the original definition. Like other RCTs, our study’s generalizability is limited to the settings targeted for this study.

Conclusion

This study examined theoretically based mediators – experiential and behavioral POC – of the CHOICES IPC preconception intervention treatment effect on the reduction of risky drinking, increased effective contraception use and reduced risk of AEP among women of childbearing age. Consistent with the Transtheoretical Model of Change, we hypothesized that experiential processes of change were affected early in the CHOICES IPC treatment process, subsequently impacting behavioral processes of change, which then directly reduced the risk of AEP. We were able to confirm this pattern in a large sample of non-treatment seeking women of childbearing age. These findings support the temporal relationship between experiential and behavioral POC consistent with the TTM and their relation to outcomes. As such, motivation-based interventions may benefit from directly targeting experiential POC early in treatment and behavioral POC later in treatment.

Public Health Significance.

This study explicates the mechanisms of behavior change in an efficacious intervention for preventing alcohol-exposed pregnancies. It provides support for the processes of change identified by the Transtheoretical Model as mediators of treatment, with reduced risk drinking and adoption of effective contraception as target behaviors. Understanding the role of the processes of change in both changing a negative behavior and in adopting a positive behavior may serve to inform the development of future behavioral interventions in order to increase their efficiency and effectiveness.

Acknowledgments

This work was supported by the Centers for Disease Control and Prevention (U84 CCU614576: PI: Velasquez) and the National Cancer Institute (K01CA157689, PI Castro).

Contributor Information

Danielle E. Parrish, Email: dparrish@uh.edu, Associate Professor, University of Houston, Graduate College of Social Work, Phone: (713) 743-8105, Fax: (713) 743-8149.

Kirk von Sternberg, Email: vonsternberg@mail.utexas.edu, Associate Professor, Associate Director, Health Behavior, Research and Training Institute, University of Texas at Austin, School of Social Work, Phone: (512) 232-0633, Fax: (512) 232-0638.

Yessenia Castro, Email: ycastro@austin.utexas.edu, Assistant Professor, University of Texas at Austin, School of Social Work, Phone: (512) 232-0778, Fax: (512) 232-0638.

Mary M. Velasquez, Email: velasquez@austin.utexas.edu, Centennial Professor in Leadership for Community, Professional and Corporate, Excellence; Director, Health Behavior, Research and Training Institute, University of Texas at Austin, School of Social Work, Phone: (512) 471-7019, Fax: (512) 232-0638.

References

  1. Allen JP, Litten RZ, Fertig JB, Babor T. A review of research on the alcohol use disorders identification test (AUDIT) Alcoholism: Clinical and Experimental Research. 2006;21:613–619. [PubMed] [Google Scholar]
  2. American College of Obstetricians and Gynecologists (ACOG) Understanding and using the U.S. selected practice recommendations for contraceptive use, 2013. 2013 (Committee Opinion No. 577). Retrieved from: http://www.acog.org/Resources-And-Publications/Committee-Opinions/Committee-on-Gynecologic-Practice/Understanding-and-Using-the-US-Selected-Practice-Recommendations-for-Contraceptive-Use-2013.
  3. Apodaca TR, Longabaugh R. Mechanisms of change in motivational interviewing: a review and preliminary evaluation of the evidence. Addiction. 2009;104(5):705–715. doi: 10.1111/j.1360-0443.2009.02527.x. [DOI] [PMC free article] [PubMed] [Google Scholar]
  4. Armitage CJ. Is there utility in the transtheoretical model? British Journal of Health Psychology. 2009;14:195–210. doi: 10.1348/135910708X368991. [DOI] [PubMed] [Google Scholar]
  5. Aveyard P, Lawrence T, Cheng KK, Griffin C, Croghan E, Johnson C. A randomized controlled trial of smoking cessation for pregnant women to test the effect of a transtheoretical model-based intervention on movement in stage and interaction with baseline stage. British Journal of Health Psychology. 2006;11:263–278. doi: 10.1348/135910705X52534. [DOI] [PubMed] [Google Scholar]
  6. Behnkey M, Smith VC. Prenatal substance abuse: short- and long-term effects on the exposed fetus. Pediatrics. 2013;131(3):e1009–e1024. doi: 10.1542/peds.2012-3931. [DOI] [PMC free article] [PubMed] [Google Scholar]
  7. Carbonari JP, DiClemente CC. Using transtheoretical profiles to differentiate levels of alcohol abstinence success. Journal of Consulting and Clinical Psychology. 2000;68:810–817. [PubMed] [Google Scholar]
  8. Center for Disease Control and Prevention (CDC) Alcohol use and binge drinking among women of childbearing age – United States 2006–2010. Atlanta, GA: Department of Health and Human Services; 2012. [PubMed] [Google Scholar]
  9. Chacko MR, von Sternberg K, Velasquez MM. Gonorrhea and chlamydia screening by young women: The processes of change. Journal of Adolescent Health. 2004;34:424–427. doi: 10.1016/j.jadohealth.2003.07.022. [DOI] [PubMed] [Google Scholar]
  10. Conigrave KM, Davies P, Haber P, Whitfield JB. Traditional markers of excessive alcohol use. Addiction. 2003;98(2):31–43. doi: 10.1046/j.1359-6357.2003.00581.x. [DOI] [PubMed] [Google Scholar]
  11. Connors GJ, DiClemente CC, Velasquez MM, Donovan DM. Substance abuse treatment and the stages of change: Selecting and planning interventions. New York: The Guilford Press; 2013. [Google Scholar]
  12. Connors GJ, Maisto SA. Drinking reports from collateral individuals. Addiction. 2003;98(2):21–29. doi: 10.1046/j.1359-6357.2003.00585.x. [DOI] [PubMed] [Google Scholar]
  13. Del Boca FK, Babor TF, McRee B. Reliability enhancement and estimation in multisite clinical trials. Journal of Studies on Alcohol. 1994;12:130–136. doi: 10.15288/jsas.1994.s12.130. [DOI] [PubMed] [Google Scholar]
  14. DiClemente CC. Changing addictive behaiviors: A process perspective. Current Directions in Psychlogical Science. 1993;2(4):101–106. [Google Scholar]
  15. DiClemente CC. Addiction and Change: How Addictions Develop and Addicted People Recover. The Guilford Press; New York: 2003. [Google Scholar]
  16. DiClemente CC. Conceptual Models and Applied Research: The Ongoing Contribution of the Transtheoretical Model. Journal of Addictions Nursing. 2005;16(1–2):5–12. [Google Scholar]
  17. DiClemente CC, Velasquez M. Motivational interviewing and the stages of change. In: Miller WR, Rollnick S, editors. Motivational Interviewing. 2nd. New York: The Guilford Press; 2002. pp. 201–216. [Google Scholar]
  18. DiClemente CC. Mechanisms, Determinants and Processes of Change in the Modification of Drinking Behavior. Alcoholism: Clinical and Experimental Research. 2007;31:13s–20s. doi: 10.1111/j.1530-0277.2007.00489.x. [DOI] [PubMed] [Google Scholar]
  19. DiClemente CC, Carbonari JP, Addy R, Velasquez MM. Alternate short forms of a processes of change scale for alcoholism treatment. Poster presented at the Fourth Internatinoal Congress on Behavioral Medicine; Washington DC. 1996. [Google Scholar]
  20. DiClemente CC, Carbonari JP, Zweben A, Morrel T, Lee RE. In: Motivation hypothesis causal chain analysis. Longabaugh R, Wirtz PW, editors. Vol. 8. Bethesda, MD: 2001. (Project MATCH Hypotheses: Results and Causal Chain Analysis. National Institute on Alcohol Abuse and Alcoholism Project MATCH Monograph Series). [Google Scholar]
  21. DiClemente CC, Prochaska JO. Self-change and therapy change of smoking behavior: A comparison of processes of change in cessation and maintenance. Addictive Behaviors. 1982;7(2):133–142. doi: 10.1016/0306-4603(82)90038-7. [DOI] [PubMed] [Google Scholar]
  22. Dishman RK, Vandenberg RJ, Mott RW, Nigg CR. Using constructs of the transtheoretical model to predict classes of change in regular physical activity: A multi-ethnic logitudinal cohort study. Annals of Behavioral Medicine. 2010;40:150–163. doi: 10.1007/s12160-010-9196-2. [DOI] [PMC free article] [PubMed] [Google Scholar]
  23. Finer LB, Zolna MR. Shifts in intended and unintended pregnacies in the United States, 2001–2008. American Journal of Public Health. 2014;104(S1):S43–S48. doi: 10.2105/AJPH.2013.301416. [DOI] [PMC free article] [PubMed] [Google Scholar]
  24. Floyd RL, Decouflé P, Hungerford DW. Alcohol use prior to pregnancy recognition. American Journal of Preventive Medicine. 1999;17(2):101–107. doi: 10.1016/s0749-3797(99)00059-8. [DOI] [PubMed] [Google Scholar]
  25. Floyd RL, Sobell M, Velasquez MM, Ingersoll K, Nettleman M, Sobell L, et al. Preventing Alcohol-Exposed Pregnancies: A Randomized Controlled Trial. American Journal of Preventive Medicine. 2007;32(1):1–10. doi: 10.1016/j.amepre.2006.08.028. [DOI] [PMC free article] [PubMed] [Google Scholar]
  26. Gunzerath L, Faden V, Zakhari S, Warren K. National institute on alcohol abuse and alcoholism report on moderate drinking. Alcoholism: Clinical and Experimental Research. 2004;28:829–847. doi: 10.1097/01.alc.0000128382.79375.b6. [DOI] [PubMed] [Google Scholar]
  27. Jacobson JL, Jacobson SW, Sokol RJ, Ager JW. Relation of maternal age and pattern of drinking to functionally significant cognitive deficit in infancy. Alcoholism: Clinical and Experimental Research. 1998;22:345–351. doi: 10.1111/j.1530-0277.1998.tb03659.x. [DOI] [PubMed] [Google Scholar]
  28. Joseph J, Breslin C, Skinner H. Critical perspectives on the transtheoretical model and stages of change. In: Tucker JA, Donovan DM, Marlatt GA, editors. Changing addictive behavior: Bridging clinical and public health strategies. New York: Guilford Press; 1999. pp. 160–190. [Google Scholar]
  29. Kass RE, Raftery AE. Bayes factors. Journal of the American Statistical Association. 1995;90:773–795. [Google Scholar]
  30. Kazdin AE. Research design in clinical psychology. 3rd. Needham Heights, MA US: Allyn & Bacon; 1998. [Google Scholar]
  31. Kazdin AE. Mediators and mechanisms of change in psychotherapy research. Annual Review of Clinical Psychology. 2007;3(1):1–27. doi: 10.1146/annurev.clinpsy.3.022806.091432. [DOI] [PubMed] [Google Scholar]
  32. Kline RB. Principles and Practice of Structural Equation Modeling. 2nd. New York: The Guilford Press; 2005. [Google Scholar]
  33. Konovalav HV, Kovetsky NS, Bobryshev YV, Ashwell KWS. Disorders of brian development in the progeny of mothers who used alcohol during pregnancy. Early Human Development. 1997;48:153–166. doi: 10.1016/s0378-3782(96)01848-8. [DOI] [PubMed] [Google Scholar]
  34. Lewis BA, Williams DM, Martinson BC, Dunsiger S, Marcus BH. Healthy for life: A randomized trial examining physical activity outcomes and psychosocial mediators. Annals of Behavioral Medicine. 2012;45:203–212. doi: 10.1007/s12160-012-9439-5. [DOI] [PMC free article] [PubMed] [Google Scholar]
  35. May PA, Baete A, Russo J, Elliott A, Blankenship J, Kalberg WO, Buckley D, Hoyme HE. Prevalence and characteristics of fetal alcohol spectrum disorders. Pediatrics. 2014;134(5):855–866. doi: 10.1542/peds.2013-3319. [DOI] [PMC free article] [PubMed] [Google Scholar]
  36. Midanik L. The Validity of Self-Reported Alcohol Consumption and Alcohol Problems: A Literature Review. British Journal of Addiction. 1982;77(4):357–382. doi: 10.1111/j.1360-0443.1982.tb02469.x. [DOI] [PubMed] [Google Scholar]
  37. Miller WR, Rollnick S. Motivational interviewing: Helping people change. 3rd. New York: The Gilford Press; 2012. [Google Scholar]
  38. Muthen LK, Muthen BO. Mplus Statisical Analysis with Latent Variables User’s Guide. Los Angeles, CA: Muthen & Muthen; 2012. [Google Scholar]
  39. Nigg CR, Geller KS, Motl RW, Horwath CC, Wertin KK, Dishman RK. A research agenda to examine the efficacy and relevance of the Transtheoretical Model for physical activity behavior. Psychology of Sport and Exercise. 2011;12:7–12. doi: 10.1016/j.psychsport.2010.04.004. [DOI] [PMC free article] [PubMed] [Google Scholar]
  40. Norcross JC, Krebs PM, Prochaska JO. Stages of change. Journal of Clinical Psychology: In Session. 2011;37(2):143–154. doi: 10.1002/jclp.20758. [DOI] [PubMed] [Google Scholar]
  41. Nykjaer C, Alwan NA, Greenwood DC, Simpson NAB, Hay AWM, White KLM, Cade JE. Maternal alcohol intake prior to and during pregnancy and risk of adverse birth outcomes: evidence form a British cohort. Journal of Epidemiology & Community Health. 2014 doi: 10.1136/jech-2013-202934. Advance online publication. [DOI] [PMC free article] [PubMed] [Google Scholar]
  42. Perz CA, DiClemente CC, Carbonari JP. Doing the right thing at the right time? The interaction of stages and processes of change in successful smoking cessation. Health Psychology. 1996;15(6):462–468. doi: 10.1037//0278-6133.15.6.462. [DOI] [PubMed] [Google Scholar]
  43. Plotnikoff RC, Hotz SB, Birkett NJ, Courneya KS. Exercise and the transtheoretical model: A longitudinal test of a population sample. Preventive Medicine. 2001;33:441–452. doi: 10.1006/pmed.2001.0914. [DOI] [PubMed] [Google Scholar]
  44. Poikolainen K, Podkletnova I, Alho H. Accuracy of quantity-frequency and graduated frequency questionnaires in measuring alcohol intake: Comparison with daily diary and commonly used laboratory markers. Alcohol and Alcoholism. 2002;37(6):573–576. doi: 10.1093/alcalc/37.6.573. [DOI] [PubMed] [Google Scholar]
  45. Prochaska JO. Systems of psychotherapy: A transtheoretical analysis. Homewood, IL: Dorsey Press; 1979. [Google Scholar]
  46. Prochaska JO, DiClemente CC. Stages and processes of self-change of smoking: Toward an integrative model of change. Journal of Consulting & Clinical Psychology. 1983;51(3):390–395. doi: 10.1037//0022-006x.51.3.390. [DOI] [PubMed] [Google Scholar]
  47. Prochaska JO, DiClemente CC. The transtheoretical approach: crossing the traditional boundaries of therapy. Homewood, IL: Dorsey/Dow Jones-Irwin; 1984. [Google Scholar]
  48. Prochaska JO, DiClemente CC, Norcross JC. In search of how people change: Applications to addictive behaviors. American Psychologist. 1992;47:1102–1114. doi: 10.1037//0003-066x.47.9.1102. [DOI] [PubMed] [Google Scholar]
  49. Prochaska JO, Velicer WF, DiClemente CC, Fava JL. Measuring the process of change: Applications to the cessation of smoking. Journal of Consulting & Clinical Psychology. 1988;56:520–528. doi: 10.1037//0022-006x.56.4.520. [DOI] [PubMed] [Google Scholar]
  50. Prochaska JO, Wright JA, Velicer WF. Evaluating theories of health behavior change: A hierarchy of criteria applied to the transtheoretical model. Applied Psychology. 2008;57:561–588. [Google Scholar]
  51. Project CHOICES Intervention Research Group. Reducing the risk of alcohol-exposed pregnancies: a study of a motivational intervention in community settings. Pediatrics. 2003;111(5):1131–1135. [PubMed] [Google Scholar]
  52. Project CHOICES Research Group. Alcohol-exposed pregnancy: characteristics associated with risk. American Journal of Preventive Medicine. 2002;23(3):166–173. doi: 10.1016/s0749-3797(02)00495-6. [DOI] [PubMed] [Google Scholar]
  53. Project MATCH Research Group. Matching alcoholism treatments to client heterogeneity: Project MATCH post-treatment drinking outcomes. Journal of Studies on Alcohol. 1997;58:7–29. [PubMed] [Google Scholar]
  54. Raftery AE. Baysian model selection in social research. Sociological Methodology. 1995;25:111–163. [Google Scholar]
  55. Rosen C. Is the sequence of change processes by stage consistent across health problems? A meta-analysis. Health Psychology. 2000;19:593–604. [PubMed] [Google Scholar]
  56. Schreiber JB, Nora A, Stage FK, Barlow EA, King J. Reporting structural equation modeling and confirmatory factor analysis results: A review. Journal of Educational Research. 2006;99(6):323–337. [Google Scholar]
  57. Schwartz G. Estimating the dimension of a model. The Annals of Statistics. 1978;6:461–464. [Google Scholar]
  58. Sobell LC, Maisto SA, Sobell MB, Cooper AM. Reliability of alcohol abusers’ self-reports of drinking behavior. Behavior Research and Therapy. 1979;17:157–160. doi: 10.1016/0005-7967(79)90025-1. [DOI] [PubMed] [Google Scholar]
  59. Sobell LC, Sobell MB. Timeline follow-back: A technique for assessing self-reported alcohol consumption. In: Litten RZ, Allens JP, editors. Measuring alcohol consumption: Psychosocial and biochemical methods. Toronto, Canada: Humana Press; 1992. pp. 41–72. [Google Scholar]
  60. Sobell LC, Sobell MC. Alcohol timeline followback users’ manual. Toronto: Addiction Research Foundation; 1995. [Google Scholar]
  61. Sokol RJ, Delaney-Black V, Nordstrom B. Fetal alcohol spectrum disorder. The Journal of the American Medical Association. 2003;290(22):2996–2999. doi: 10.1001/jama.290.22.2996. [DOI] [PubMed] [Google Scholar]
  62. Sood B, Delaney-Black V, Covington C, Nordstrom-Klee B, Ager J, Templin T, Sokol RJ. Prenatal alcohol exposure and childhood behavior at age 6 to 7 years: I. dose-response effect. Pediatrics. 2001;108(2):e34–e34. doi: 10.1542/peds.108.2.e34. [DOI] [PubMed] [Google Scholar]
  63. Stotts AL, DeLaune KA, Schmitz JM, Grabowski J. Impact of a motivational intervnetion on mechanisms of change in low-income pregnant smokers. Addictive Behaviors. 2004;29:1649–1657. doi: 10.1016/j.addbeh.2004.02.063. [DOI] [PubMed] [Google Scholar]
  64. Streissguth AP, Barr HM, Sampson PD, Bookstein FL. Prenatal alcohol and offspring development: the first fourteen years. Drug and Alcohol Development. 1994;36:89–99. doi: 10.1016/0376-8716(94)90090-6. [DOI] [PubMed] [Google Scholar]
  65. Streissguth AP, Sampson PD, Olson HC, Bookstein FL, Barr HM, Scott M, Mirsky AF. Maternal drinking during pregnancy: Attention and short-term memory in 14-year old offspring – A longitudinal prospective study. Alcohol Cliniccal and Experimental Research. 1994;18(1):202–218. doi: 10.1111/j.1530-0277.1994.tb00904.x. [DOI] [PubMed] [Google Scholar]
  66. Velasquez MM, Gaddy-Mauer G, Crouch C, DiClemente CC. Group treatment for substance abuse: A stages-of-change therapy manual. New York: The Gilford Press; 2001. [Google Scholar]
  67. Velasquez MM, Ingersoll KS, Sobell MB, Floyd RL, Sobel LC, von Sternberg K. A Dual-Focus Motivational Intervention to Reduce the Risk of Alcohol-Exposed Pregnancy. Cognitive and Behavioral Practice. 2010;17(2):203–212. doi: 10.1016/j.cbpra.2009.02.004. [DOI] [PMC free article] [PubMed] [Google Scholar]
  68. vonSternberg K. Comparing the factorial structure, invariance, and predictive validity of Transtheoretical model constructs for alcohol use across restricted and unrestricted settings. Dissertation Abstracts International: Section B: The Sciences and Engineering. 2005;66:3094. [Google Scholar]
  69. Williams JF, Smith VC. Fetal alcohol spectrum disorders. Pediatrics. 2015 doi: 10.1542/peds.2015-3113. Advanced online publicaiton. [DOI] [PubMed] [Google Scholar]
  70. Wilson GT, Schlam TR. The transtheoretical model and motivatinal interviewing in the treatment of eating and weight disorders. Clinical Psychology Review. 2004;24:361–378. doi: 10.1016/j.cpr.2004.03.003. [DOI] [PubMed] [Google Scholar]
  71. Yin H, Prochaska JO, Rossi JS, Redding CA, Paiva AL, Blissmer B, Kobayashi H. Treatment-enhanced paired action contributes substantially to change across multiple health behaviors: secondary analyses of five randomized trials. Translational Behavioral Medicine. 2013;3:62–71. doi: 10.1007/s13142-013-0193-4. [DOI] [PMC free article] [PubMed] [Google Scholar]

RESOURCES