The moderating role of substance abusers personal attributes in predicting relapse and partner persuasiveness
Addiction Research and TheoryOctober 2006, 14(5): 493509The moderating role of substance abusers personalattributes in predicting relapse and partnerpersuasivenessCARRIE J. CROPLEYDepartment of Communication, Santa Barbara City College, Santa Barbara, CA(Received in final form 15 February 2006)AbstractIn relationships where one partner is dependent on substances, the non-dependent, orfunctional partners use a variety of strategies to stop the abuse and prevent relapse.Inconclusive research results on the effectiveness of these strategies can be partiallyexplained by the failure to consider personal characteristics of the substance abusers thatmay make them more or less receptive to their partners attempts at assisting them in theirsobriety. This work explores the substance abusers anger, over-controlled hostility, andego-strength as moderating factors influencing both the types and amounts of persuasivestrategies used by the significant other. The sample was composed of 67 married orcohabitating couples with one functional and one substance-abusing partner (as assessedby the chemical use, abuse, and dependence (CUAD) scale). Participants personalityvariables were tested using the Minnesota Multiphasic Personality Inventory (MMPI),followed by videotaped interviews where they were independently asked about theirown and their partners behavior. Contrary to predictions, as the abusers anger andego-strength increased so did their partners use of punishment and the consistency ofthe substance-abusing behavior. Finally, as the abusers ego-strength increased and over-controlled hostility decreased the partner presented more reinforcement of the substanceabuse. These findings encourage a systems approach to the treatment of substance abuse,and provide insight into appropriate and effective strategies used by the functional partnerin this system.Keywords: Recidivism, ego-strength, hostility, anger, inconsistent nurturing as controlIntroductionPast research in drug and alcohol dependence and abuse has suggested that thefamily has a significant effect on the substance abusers behavior. ResearchersCorrespondence: Carrie J. Cropley, Department of Communication, 721 Cliff Drive,Santa Barbara, CA 93109. Tel: (805) 965-0581. E-mail: email@example.comISSN 1606-6359 print/ISSN 1476-7392 online 2006 Informa UK Ltd.DOI: 10.1080/16066350600636613Addict Res Theory Downloaded from informahealthcare.com by University of California Irvine on 10/31/14For personal use only.conclude that the communication of the non-dependent partner may encourageor discourage the drug and alcohol use and abuse, and more optimistically, mayhave the potential to help the abuser change the patterns of use and ultimatelybecome clean (Le Poire & Cope, 1999; Le Poire, Erlandson, & Hallett, 1998;Le Poire, Hallett, & Erlandson, 2000). These findings clearly provide a rationalefor studying communication patterns in relationships that include substanceabusers. However, simply knowing which strategies are more or less effective atcurtailing substance abuse is insufficient. Specifically, it is likely that substanceabusers are self-medicating some larger psychological problem (Gawin & Kleber,1986; Khantzian, 1985). To further compound the issue, the same psychologicalvariables that caused the dependence in the first place may actually result in thesubstance abusers resistance to others strategies for curtailing the abuse.This work thus examines whether the individual level variables of anger,over-controlled hostility, and ego-strength interfere with the significant othersability to assist in reducing the substance abuse.Overview of INC theoryNurturer-controllers (commonly known as codependents) find themselves ina unique position as they attempt to maintain a relationship with their substance-abusing partner. The challenge of utilizing relational maintenance strategiesis complicated by the substance-abuse behavior, such that the functionalpartners strategies are dominated by the attempt to alter their partnersbehavioral compulsions (see Le Poire, 1992, 1995 for review). Inconsistentnurturing as control (INC) theory attempts to explain the effectiveness of thestrategies non-abusing partners utilize in attempting to control their partnerssubstance-abuse.Skinner (1971) originally examined the differential effects of consistency andinconsistency in reinforcement and punishment of behavior. INC theory expandson Skinnerian behaviorism by outlining certain dynamics that are unique to thecontext of relationships that include one afflicted partner. Le Poire (1995) arguesthat several paradoxical injunctions exist in relationships that include afflictedpartners (e.g., drug abusers, aggressive individuals, depressed individuals, andeating disordered individuals) and that these paradoxes ultimately impactexpressions of control by the functional family member (i.e., the partner withno problem interfering with day-to-day functioning) in the relationship. In thecontext of a relationship where one person is functional and the other afflicted,the functional person is very likely to intermittently reinforce behaviors theyactually want to extinguish. This reinforcement occurs when functional familymembers first nurture the substance-dependent individual when they arein crisis, thus ultimately reinforcing the behavioral compulsion. Then, ascaregivers become resentful, as is likely to happen (Asher, 1992; Wiseman, 1991),they may fail to nurture the afflicted individuals, thus failing to reinforce thebehavior. The lack of care giving by the functional partner not only communicatesresentment, but also is likely an attempt to punish, or extinguish the undesirable494 C. J. CropleyAddict Res Theory Downloaded from informahealthcare.com by University of California Irvine on 10/31/14For personal use only.behavior of the dependent partner. However, INC theory argues thatthe intermittent nature of this punishing behavior will actually increase thebehavioral compulsion. Thus, functional family members unintentionallystrengthen the likelihood of the substance-abuse behavior through intermittentreinforcement and intermittent punishment.Application of INC theory to the substance abuser-significant other relationshipResearch on INC theory thus far indicates that family members of substance-abusive individuals are indeed inconsistent in their use of reinforcement andpunishment. In an investigation of drug users and their functional partners,Le Poire et al. (2000) found that partners typically cycled from reinforcing topunishing communication strategies following labeling of their partners assubstance abusive. This first study of INC theory hypothesized and found thatfunctional partners changed their strategy usage over time, such that (a) theyreinforced substance-dependent behavior more before their determination that thebehavior was problematic than after; (b) they punished substance-dependentbehavior more after they labeled the drinking/drugging behavior as beingproblematic, than before; and (c) they reverted to a mix of reinforcing andpunishing strategies, resulting in an overall pattern of inconsistent reinforcementand punishment. This cycling is central to the inconsistent nature of reinforcingor punishing communication strategies postulated by INC theory. In addition tothe findings afore-mentioned, a qualitative analysis of the strategies used byfunctional partners of substance abusers (Le Poire & Addis, 2000) also supportsthe cycling pattern. Functional partners used several macro-level strategies whichincluded both reinforcement and punishment including verbal abuse, makingrules pertaining to the addiction, punishment, getting a third party involved,threats, avoidance, ending the relationship, expressing personal feelings, with-holding something from the partner as a punishment, supporting abuseby participation, demanding the partner stop/active involvement, and confront-ing. The use of these strategies will strengthen the tendency to engage in alcoholor drug use, according to the hypotheses posited by INC theory and theconfirmatory research. Although this patterning in and of itself was not found tobe more predictive of greater relapse, patterns of reinforcement and punishmentwere linked to persuasive outcomes. Specifically, partners who were moreconsistent in punishing substance abuse and reinforcing alternative behaviors(e.g., encouraging attendance at AA meetings) had substance-abusive partnerswho relapsed less. Moreover, more successful partners also reported lessdepression than those with partners who relapsed more (Le Poire et al., 2000).This is important for two reasons. First, partners of substance-abusing individualscan help reduce their partners recidivism. Second, this assistance can alsotranslate into better mental health outcomes for the partners.The findings thus far have clearly supported the underlying principles ofINC theory. However, the utility of the theory extends beyond describing thebehavioral patterns of the functional family member. INC theory makes specificPersonal attributes and persuadability 495Addict Res Theory Downloaded from informahealthcare.com by University of California Irvine on 10/31/14For personal use only.recommendations that will allow the family members of substance-abusingindividuals to engage in communication behaviors that will assist in recovery.Specifically, INC theory asserts that to make a difference in the recovery of thenon-functioning partner, the functioning partner should adopt consistent com-munication strategies in the form of consistent punishment of substance abuse.Furthermore, functional members should refrain from the behaviors thatintermittently reinforce the substance abuse between punishments. The key tosuccessfully implementing these strategies requires that the functioning partnerprovide reinforcement of alternative behaviors for the non-functioning partner(Le Poire et al., 2000). Marlatt (1985) concurs, writing social factors areinvolved both in the initial learning of the addictive habit, and in the subsequentperformance of the activity once the habit has become firmly established (p.10).It appears, however, that in viewing the relationship and the addiction holistically,there are a multitude of other variables to consider. Although the communicationpatterns of the functioning partner play a significant role in aiding recovery andpreventing relapse, the personal attributes of the substance abuser may mediatethe success of these strategies.Personal attributes of the substance-abusingpartner affecting recidivismAnger, over-controlled hostility, and ego-strengthA consistent finding is that between 30 (Sheehan, 1993), and over 50% ofsubstance-dependent individuals are susceptible to dual-diagnoses (e.g., Bryant,Rounsaville, Spitzer, & Williams, 1992) including antisocial personality disorder(Lehman, Myers, Thompson, & Corty, 1993). Thus, it is important to considerthe impact such personality disorders may have on communication behavior andon communication outcomes in the relationship. Specifically, it is likely that thereare personal characteristics that make it less likely that substance abusers will bepersuaded by their significant others. This work, therefore, focuses on personalcharacteristics with interpersonal ramifications. It is highly likely that substanceabusers with problematically high levels of anger and expressions of over-controlled hostility are likely to be less susceptible to their significant otherspersuasive attempts. Moreover, this anger and over-controlled hostility mayactually lessen the amount of attempts a significant other may communicate thefirst place. Further, those same individuals with greater ego-strength are also lesslikely to be persuaded by their significant others as their opinions are rated asmore important than their partners. Thus, there are several personal attributevariables that may make it less likely that an individual would be persuaded bytheir significant other to use less alcohol/substances. In addition, met withresistance from their substance-abusing partners, functional partners may noteven present their substance-abusing significant others with the same amounts ofstrategies had their partners been less hostile and angry. Thus, it is possible thatfunctional partners may present less punishment (and consequently more496 C. J. CropleyAddict Res Theory Downloaded from informahealthcare.com by University of California Irvine on 10/31/14For personal use only.reinforcement) of substance abuse, and less consistency of strategies when theirpartners report more anger, over-controlled hostility, and ego-strength. Thus, it ispossible to hypothesize that:H1: As anger, over-controlled hostility, and ego-strength of the substance-abusing partner increase, the amounts of punishment of the substance-abusebehavior by the functional partner will decrease.H2: As anger, over-controlled hostility, and ego-strength of the substance-abusing partner increase, so too will the amounts of reinforcement of thesubstance-abuse behavior by the functional partner.H3: As anger, over-controlled hostility, and ego-strength of the substance-abusing partner increase, the amount of reinforcement of alternative behaviorby the functional partner will decrease.H4: As anger, over-controlled hostility, and ego-strength of the substance-abusing partner increase, the amounts of consistency in the functional partnerspunishment of the substance-abuse behavior will decrease.Given that the existence of anger, over-controlled hostility, and ego-strength ofthe substance-abusing partner may lessen the types of strategies used by theirfunctional partner, it is also possible that functional partners will be rated as lesspersuasively effective by their non-functional partners. This, in combination withthe fact that substance abusers with anger, hostility, and ego-strength should bemore resistant to the strategies that are presented by their functional partnerswould seem to indicate that it is even more likely that substance abusers will ratetheir functional partners as less effective the more they report anger, over-controlled hostility, and ego-strength.H5: The effectiveness of the functional partners persuasive strategies will bemoderated by the anger, over-controlled hostility, and ego-strength of the non-functioning partner, such that substance-abusing partners who are high in anger,over-controlled hostility, and ego-strength will report that their partners are lesspersuasively effective.To take this one logical step further, it is similarly unlikely that a substanceabuser with anger, over-controlled hostility, and ego-strength will be successfulin their recovery as a result of their significant others help. In addition, since theymay be attempted to self-medicate their issues with anger and hostility, it is alsopossible that it is more difficult for them to recover with or without their partnerspersuasive attempts. It is therefore also possible to hypothesize that:H6: Substance-abusive partners who are higher in anger, over-controlledhostility, and ego-strength are more likely to report higher amounts of relapse.MethodThis study was part of an omnibus investigation testing the viability ofINC theory. For a full delineation of the methods, see Le Poire et al. (2000).Personal attributes and persuadability 497Addict Res Theory Downloaded from informahealthcare.com by University of California Irvine on 10/31/14For personal use only.A recapitulation of the methods as relevant to the current investigation will bepresented here.ParticipantsParticipants (N 134) were married or cohabitating individuals and werereimbursed $50.00 for their time (in cash or in grocery store certificates). Theywere secured through a newspaper story and short news report on a localtelevision station, an advertisement in a weekly free paper, and by obtainingreferrals through a drug treatment agency in southern California. Eligibility forthis study was based on the assessment procedures used by the treatment agencywhen assessing clients positively for need of substance-abuse treatment, as well asan additional positive assessment made on the chemical use, abuse, anddependence (CUAD) scale by the principal investigator and her researchassistants.Screening. To differentiate between couples where only one member wassubstance-dependent and couples where both members were dependent, both theclient and the partner were screened for their drug use, abuse, and dependence.Trained raters (principal investigator and research assistants) made independentassessments of all participants on the CUAD scale. This scale allowed for theassessment of use, abuse, and dependence on substances and alcohol (for a fullexplication, see Le Poire et al., 2000).Drug-dependents ranged in age from 18 to 56 years and were usuallymale. Alcohol, cocaine, and heroine were the most common drugs of choice,though more than 80% of the sample was addicted to some combination of twoor more drugs. On the severity score of the CUAD, all users were evaluated asdependent on their drug of choice, and 98% were also rated as abusive of thedrug. About half (52%) of the users in the sample are or had been in treatment fortheir habit. Functional partners ranged in age from 18 to 80, and were primarilyfemale. The majority (68%) of functional partners also used substances. Abouthalf of these partners had been dependent on some drug in the past, but werecurrently sober. The other half was currently using but not dependent on orabusive of any drug (as rated on the CUAD). Only seriously involved romanticcouples were used in this study. Couples had been living together (or married) forbetween 6 months and 37 years. The majority of couples were heterosexual (over95%). A substantial portion of these couples (50%) had a combined annual grossincome below $20,000.Procedures and independent variablesSixty-seven couples (N 134 individuals) were interviewed (with each memberof the dyad interviewed individually) concerning the functional partners use ofreinforcing and punishing communication strategies of control/influence (datafrom 52 couples were analyzed because they met the criteria of not being dual-abusing couples). Interviews were conducted at the participants homes498 C. J. CropleyAddict Res Theory Downloaded from informahealthcare.com by University of California Irvine on 10/31/14For personal use only.(to alleviate any transportation problems), or if the participants preferred, inprivate university offices. Using a memory enhancing time-line procedure, similarto McCrady et al. (1986), each couple was prompted by questions about whenthey first met, when they committed to one another, and when they moved intogether or got married. Following this prompting, each member of the couplewas interviewed separately, and was asked when the substance use was first seenas problematic by the functional partner (labeling), what the functional partnerdid before the labeling (during drugging episodes), what the functional partnerdid after the labeling, and what the functional partner did at some point whenthey felt their current strategies were ineffective (after frustration with currentefforts; modified from Wiseman, 1991).The interview schedule was semi-structured, in that participants were asked toprovide information with specific time periods in mind. (These time periods wereestablished by the investigators based on the time-line of the relationship providedby the couples.) Both partners were asked questions about what the functionalpartner did surrounding the dependents drug use. For instance, What did youdo first when your partner used drugs? (or, What did your partner do when shefound out you were using drugs?). After answering, they were again prompted:And what did you do after that? (or, And what did she do after that?). Thisquestion was re-asked until all the strategies used in that time period werereported. The investigators wrote down all answers so that the data could beeasily accessed for future reference and analysis.Reinforcement and punishment. Macro-strategies (strategies reported over thelong term of the relationship) were placed within a time frame as closelyas possible (i.e., First, I tried to hide his booze, then when that didnt work,I started to drink with him). Once the strategies were recorded, trained codersevaluated each behavior in terms of its reinforcing or punishing nature.Reinforcing behaviors were defined as those that encourage the behavior theyfollow. An additional orthogonal delineation was made with regard to the typeof behavior being reinforced, so that behaviors examined included reinforcementof the substance-abusing behavior and reinforcement of alternative behaviors.Punishing strategies were those that the substance-dependent partner should findaversive (and should diminish the likelihood of the behavior in the future). Thus,four delineations resulted: (1) reinforcement of substance-dependence (e.g.,I drank with him), (2) reinforcement of alternative behavior, (e.g., Aftera hard day I took him to a movie instead of pouring him a drink), (3) punishmentof substance-dependence (e.g., I hid the booze), and (4) punishmentof alternative behavior (I criticized the project he was doing to distract himselffrom drinking).With these ratings, a proportional estimate was calculated for each independentvariable. For instance, the number of reinforcing of substance-dependencebehaviors reported, divided by the total number of strategies used, generated apercentage of reinforcing substance-dependence strategies for any particularperson. This same procedure was utilized for punishing strategies.Personal attributes and persuadability 499Addict Res Theory Downloaded from informahealthcare.com by University of California Irvine on 10/31/14For personal use only.Consistency. Once a behavior was rated sequentially in terms of its punishingor reinforcing nature, it was further rated as consistent or inconsistent with thebehavior directly preceding and adjacent to it. Thus, a punishing behaviorfollowed by a punishing behavior (or a reinforcing behavior followed by areinforcing behavior) was rated as consistent, although a reinforcing behaviorfollowing a punishing behavior (or a punishing behavior following a reinforcingbehavior) was rated as inconsistent. Thus, in the example aforesaid, hiding hisbooze was rated as punishing, whereas drinking with him was rated asreinforcing. Additionally, drinking with him (reinforcing) was rated asinconsistent with hiding his booze (punishing). Similar to punishment andreinforcement estimates, a proportional estimate of consistency was calculated.For consistency, however, the strategy differs slightly, as the total number ofconsistent attempts was divided by the total number of strategies 1, since thefirst attempt cannot be deemed consistent with the previous component of itsadjacency pair that does not exist (this method is necessary only for the firststrategies used, as the strategies used after the labeling can be rated forconsistency with the last behavior utilized before the behavior was labeled asproblematic). These methods provided a proportional rating of consistencyranging from 0 (no consistency) to 100 (with each strategy being consistent withthe strategies that preceded it).Trained coders. Raters of strategic behavior were six undergraduates withexpertise in communication that received approximately 30 h of training.Approximately 3 weeks of training were devoted to teaching coders to recognizethe influence tactics as reinforcing or punishing. Instruction included presenta-tion of conceptual and operational definitions of the tactics, review of samplesof specific tactics, practice coding of pilot study interviews, and reviews of ratinginconsistencies among coders.Inter-rater and inter-item reliabilities. Reinforcement or punishment ratings andsubsequent consistency ratings were made for each strategy reported. Frequencycounts were summed for each strategy type (reinforcement/punishment), withpercentages of each type of strategy derived by dividing that frequency by the totalnumber of strategies used by that partner. Inter-rater Cronbach alpha reliabilitieswere as follows for each type of rating: reinforcement of substance abuse, 0.75;punishment of substance abuse, 0.86; reinforcement of alternative behavior, 0.55;and consistency, 0.72. (There were too few punishment of alternative behaviorstrategies to include in analyses.)Dependent variablesChemical dependency relapses. Relapse has been defined inconsistently in theliterature. Some use any drinking behavior as an indicator of relapse (e.g.,Burling, Reilly, Moltzen, & Ziff, 1989; Sussman, Rychtarik, Mueser, Glynn, &Prue, 1986), others use re-admittance to a drug treatment center (e.g., Kivlahan,Walker, Donovan, & Mischke, 1985), some use drivers history files and arrest500 C. J. CropleyAddict Res Theory Downloaded from informahealthcare.com by University of California Irvine on 10/31/14For personal use only.blood alcohol content forms (e.g., Wells-Parker, Pang, Anderson, McMillen, &Miller, 1991), and still others use more subjective family reports (Watson, Jacobs,Pucel, Tilleskjor, & Hoodecheck, 1984). In an attempt to be more tolerantof slips, the current work will not define remission as strict abstinence, but willinstead utilize the more face-valid operational definition utilized by Moos, Finneyand Chan (1981). Their criteria for relapse include one or more of the followingincidents: (1) re-hospitalization, (2) missed work due to drug-addiction,(3) consumption of more than 5 ounces of ethanol per drinking day (or drug-equivalent) in the last month, (4) consumption of more than 3 ounces of ethanol(or drug-equivalent) per day, on average, in the last month, and (5) problemsassociated with drugging behavior (excluding family arguments) in the lastmonth. The presence of one or more of these behaviors classifies one as havingrelapsed.Persuasive effectiveness scale. A second measure of success of the influencestrategies has to do with the perceptions of the addicted partner. Such successdepends in part on whether the addicted partner changes his or her druggingbehaviors. Using a slightly modified scale developed by Newton and Burgoon(1990), the addicted participants were asked to evaluate the degree to which theirfunctional partners were successful in influencing the substance-dependentbehavior. This was indexed by ten 7-interval Likert-type items including itemslike: I was strongly persuaded by my partners attempts to get me to stopdrinking/drugging, and my partner raised some compelling arguments I hadnot thought of before. The reliability of this scale was 0.76.Personal characteristics assessment. The specific personality variables of thenon-functioning partner examined in the present study included ego-strength,over-controlled hostility, and anger, as assessed by the Minnesota MultiphasicPersonality Inventory (MMPI). The ego-strength scale is a measure ofadaptability, resiliency, personal resourcefulness, and effective functioning(Hathaway &McKinley, 1989, p. 40) in addition to being a good overall indicatorof the respondents general psychological health. Moderate scores on ego-strength indicate the ability to cope with stress and to recover from problems.However, high scores on ego-strength, especially when combined with narcissism,are often correlated with feelings of invincibility and the expression of anger andaggression (Papps & OCarroll, 1998).The original assessment for ego-strength includes 52 questions. For reliabilityin the present study, 20 of these questions were used ( 0.82). Example MMPIquestions for assessing ego-strength include Some people are so bossy that I feellike doing the opposite of what they request, even though I know they are right(question 98).The over-controlled hostility scale originally contained 28 questions. Thisparticular scale measures an individuals ability to experience frustrations withoutretaliation. For reliability in this study 15 questions were eliminated resultingin 13 questions 0.69. Example questions that assess hostility includePersonal attributes and persuadability 501Addict Res Theory Downloaded from informahealthcare.com by University of California Irvine on 10/31/14For personal use only.My hardest battles are with myself (question 89). The final personality variabletested was anger, which assesses the tendency to lose self-control. This scale wasoriginally composed of 16 questions but reduced to 14 for reliability in the presentstudy ( 0.76). Questions that assess anger include those such as At timesI feel like smashing things (question 37).ResultsThe first set of hypotheses predicted that as anger, over-controlled hostility,and ego-strength of the substance-abusing partner increased, the amounts ofpunishment of substance abuse, reinforcement of alternative behavior, andconsistency of persuasive strategies used by the functional partner woulddecrease, while the amount of reinforcement of substance abuse would increase.These hypotheses were tested through a series of four stepwise regressionequations of anger, over-controlled hostility, and ego-strength on (1) punishmentof substance abuse, (2) reinforcement of substance abuse, (3) reinforcement ofalternative behavior, and (4) consistency (across time).H1 was not supported. The regression equation for anger, hostility, and ego-strength on punishment of substance abuse was significant, but in the oppositedirection as predicted, F(3, 48) 2.7, phostility decreases, reinforcement of substance abuse increases. The weightsshow no significant effect for anger on reinforcement.H3 was not supported. The effect of anger, over-controlled hostility, andego-strength on reinforcement of alternative behavior was not significant,F(3, 48) 0.86, p 0.46, adjusted R20.008.H4 was partially supported. The regression equation for consistency wassignificant, F(3, 48) 4.55, pand ego-strength, as moderating factors influencing both the types and amountsof strategies used by significant others to attempt to get their partners to stopabusing substances. Additionally, the impact of these variables on persuasiveeffectiveness of the functional partner and relapse of the substance-abusingpartners was assessed.The hypotheses for the variables affecting punishment of substance-abusingbehavior and consistency of strategies were formulated under the assumption thatwhen a substance-abusing partner exhibits high anger, ego-strength, and hostility,these attributes may discourage the functional partner from aggressively andconsistently punishing the problematic behavior of the abuser. In other words, theanticipation of an angry, discounting, or hostile reaction may discourage thefunctional partner from communicating his or her discontent in the form ofconsistent and punishing responses. Thus, in relationships where the substanceabuser has low anger, ego-strength, and hostility, it was assumed that thefunctional partner would be more likely to exhibit punishment and maintainconsistent strategies because he/she would not fear a negative, angry, orhostile reaction from the abusing partner. Contrary to the hypotheses, as angerand ego-strength increased, so did the partners use of punishment of thesubstance-abusive behavior. This finding indicates that the combination of highanger and high ego-strength of the substance-abusing partner result in a greaterlikelihood of punishing behavior from the functional partner. This relationshipwas identical for consistency in that functional partners who had substance-abusing partners high in anger and ego-strength presented punishment andreinforcement of substance abuse in consistent ways.There are two potential explanations for these findings. First, this tendency toreact to high levels of anger and ego-strength with punishment may occur becausein anticipating a negative reaction from the abuser, the functional partner mayfeel the need to try even harder to punish the problematic behavior in order to havean impact. A second and more likely explanation is supported by relationshipliterature addressing negative reciprocity. When one member of a couple engagesin communication that is perceived as negative, the other is likely to reciprocatein an equally negative or often more negative manner (Sabourin, 1996). Thistendency is especially pronounced in couples who are distressed (Gottman,Markman, & Notarious, 1977), in couples where there exists physical or mentalabuse (Cordova et al., 1993; Sabourin, 1996), and in couples where one or bothmembers are substance abusers (Leonard & Roberts, 1998). Research suggeststhat the unique sample used in this study may exemplify negative reciprocity evenmore dramatically because one of the dyad members may have been under theinfluence during the interaction that was observed. Negative reciprocity tends tobe even more overt when couples are observed interacting while one or bothpartners is intoxicated ( Jacob, Leonard, & Haber, 2001; Leonard & Roberts,1998).It was expected that as ego-strength and hostility increased, so would thereinforcement of substance-abusing behavior by the functional partner. It wasconfirmed that as ego-strength increased significant others presented more504 C. J. CropleyAddict Res Theory Downloaded from informahealthcare.com by University of California Irvine on 10/31/14For personal use only.reinforcement of substance abuse. This finding is partially supportive of thehypothesis in that ego-strength on the part of the substance abuser may have beenmet with reinforcement of the substance abuse because the functional partnermay be trying to improve interpersonal relations. It has been consistently foundthat partners of steady state drinkers consistently reinforce this drinking because itis conducive to the interpersonal functioning of the relationship (Le Poire &Cope, 1999). In other words, increased substance abuse may actually enhancefamily functioning especially if it reduces the amount of ego-strength the partneris bringing to the relationship. This explanation is consistent with the finding thatover-controlled hostility is lower when reinforcement of substance abuse ishigher. Contrary to predictions, however, as over-controlled hostility went down,reinforcement of the substance-abusing behavior increased. This finding can beexplained by the likelihood that the substance abuser is less hostile when thepartner reinforces the substance abuse, because they do not have to be resistantto persuasive attempts.Theoretical implicationsPrevious testing of INC theory has revealed that in relationships where onepartner is a substance abuser, functional partners who engaged in consistentpunishing of the abusive behavior and reinforcement of alternative behaviorshad substance-abusing partners who relapsed less frequently. Thus, it has beenestablished that partners of substance-abusing individuals can help reduce theirpartners recidivism and that this assistance can also translate into better mentalhealth outcomes for the partners, such as less depression. However, this researchtakes a necessary step further in gaining predictive power for outcomesin relationships where one partner is a substance abuser through considerationof individual level variables that may mediate the effectiveness of the persuasivestrategies of the significant other. In other words, it is important to address whatpersonality variables the abuser brings to the table before analyzing theeffectiveness of the functional partners communication, since these personalitycharacteristics can pre-dispose the abusive partner to persuasive resistance,no matter what types of strategies are employed.Although the communication patterns of the functioning partner and therelational context play a significant role in aiding recovery and preventing relapse,this investigation reveals that the personal attributes of the substance abuser canmediate the types and successes of these strategies used. Specifically, this studyfound that as anger and ego-strength increase, so does the presentation ofpunishment of substance abuse, possibly because the functioning partner employsmore heavy-hitting strategies in anticipation of a negative reaction from theabuser, or because they are engaging in negative reciprocity patterns typical ofrelationships afflicted by substance abuse. This also explains why, similar topunishment of substance abuse, as anger and ego-strength increase, so does thepresentation of consistent strategies for curtailing future substance abuse.However, when the abuser has high ego-strength but low hostility, the functionalPersonal attributes and persuadability 505Addict Res Theory Downloaded from informahealthcare.com by University of California Irvine on 10/31/14For personal use only.partner may reinforce rather than punish the substance-abusing behavior becauseit may be more functional to the relationship. It may be the case here thatreinforcing the substance abuse may actually function to facilitate smootherfamily interactions. Thus, a headstrong partner who is also controlling a largeamount of hostility may be presented with greater reinforcement of substanceabuse in an attempt to soothe this potentially problematic partner.The findings of this investigation suggest that in promoting strategies that willhelp curtail substance abuse and also prevent relapse, INC theory may benefitfrom recognizing how the personal attributes and dispositions of the using partnermay affect the communication strategies of the functional partner. It may behelpful for the functional partner to become aware of his or her tendencies as astarting ground for suggesting strategies that may be more helpful and effective.Limitations and future directionsThe present investigation was limited by various methodological challenges. First,it is difficult to solicit a large sample size for a study of this nature. Because thereis a social stigma attached to substance abuse issues in relationships, findingrespondents to interview was difficult. In addition, the inability to gender matchmade it impossible to examine gender differences in types of strategies employedand strategy effectiveness. In the current study, most substance abusers were malewith female functional partners. Balancing on this particular variable is difficultin examining the current theory, however, in that it is nearly impossible to finda large sample of female substance abusers with male non-abusing partners.Another potential shortcoming of this investigation is that in order to increase thereliabilities of the MMPI scales for anger, ego-strength, and hostility, questionswere eliminated that may have affected the validity of these variables as measuredin the current investigation. Finally, in any study where respondents were askedto report communication behavior retrospectively, the data has the potential to beincomplete or biased. Although several measures were undertaken to ensureaccurate time sequencing of strategies, it is possible that respondents mis-remembered or remembered strategies out of sequence, which could ultimatelyimpact the accuracy of the measure of consistency.Future studies may benefit from doing separate analyses on psychoactivesubstance abuse, as it may be differentially difficult to exert control over differenttypes of substance abuse. In addition, future studies examining the moderatingeffect of personal attributes may want to separate positive versus negativereinforcement behavior. Finally, future directions for gaining a better under-standing of communicative outcomes in relationships where one partner is asubstance abuser include not only personality variables of the abusive partner,but also relational variables. Two specific variables that may provide additionalinsight are evidence of family problems and marital distress. While it will beimpossible to determine whether family problems and marital distress are a causeor a consequence of substance abuse, it is likely that the existence of familyproblems and marital distress may promote abusing partners to be more or less506 C. J. CropleyAddict Res Theory Downloaded from informahealthcare.com by University of California Irvine on 10/31/14For personal use only.receptive to their non-abusing partners influence attempts. Including these twoconsiderations may increase the likelihood of predicting communication out-comes by drawing a more accurate picture of the relational context in which thecommunication takes place. In considering the relational context, future researchon this subject would benefit from using Gonzalez and Griffins (1997) latentvariable model for analyzing interdependent data. This approach allows forindividual-level, couple-level, and cross individual-level analyses of the date.Summary and conclusionsThis investigation extends INC theory by presenting a more completeexplanatory calculus of the types and effectiveness of strategies functionalpartners tend to employ depending on the disposition of the non-functionalpartner. Although the specific communication patterns of the functioning partnercan play a significant role in helping the non-functional partner recover, thepersonal attributes of the substance abuser can mediate the likelihood thatthe functional partner will use certain strategies and the perceived effectivenessof the strategies employed. Specifically, this study found that as anger andego-strength increase, the presentation of punishment of substance abuseincreases. Also, as anger and ego-strength increase, so does the presentation ofconsistent strategies for curtailing future substance abuse. A possible interpreta-tion of these patterns is that functional partners will be more consistentin their strategies when they anticipate overtly negative reactions from theabusing partner, or that they are engaging in the norm of negative reciprocity.Finally, as ego-strength increases and over-controlled hostility decreases,reinforcement of substance abuse increases. This finding suggests that highego-strength may encourage the functional partners reinforcement of thesubstance abuse as a form of maintaining positive interpersonal relations,especially if the reinforcement will prevent hostility from the abuser. In additionto the individual level variables investigated here, future research should bedirected at assessing relational variables, such as family problems and maritaldistress, in order to better understand the complex nature of persuasiveeffectiveness and recidivism in relationships where one partner is a substanceabuser.ReferencesAsher, R. M. (1992). Women with alcoholic husbands. Chapel Hill: University of NorthCarolina Press.Bryant, K. J., Rounsaville, B., Spitzer, R. L., & Williams, J. B. (1992). Reliability of dual diagnosis:Substance dependence and psychiatric disorders. Journal of Nervous and Mental Disease, 180,251257.Burling, T. A., Reilly, P. M., Moltzen, J. O., & Ziff, D. C. (1989). Self-efficacy and relapse amonginpatient drug and alcohol abusers: A predictor of outcome. Journal of Studies on Alcohol, 50,354360.Personal attributes and persuadability 507Addict Res Theory Downloaded from informahealthcare.com by University of California Irvine on 10/31/14For personal use only.Cordova, J. V., Jacobson, N. S., Gottman, J. M., Rushe, R., & Cox, Q. (1993). Negative reciprocityand communication in couples with a violent husband. Journal of Abnormal Psychology, 102,559564.Gawin, F. H., & Kleber, H. D. (1986). Abstinence symptomatology and psychiatric diagnosis incocaine abusers. Archives of General Psychiatry, 43, 107113.Gonzalez, R., & Griffin, D. (1997). On the statistics of interdependence: Treating dyadic data withrespect. In S. Duck (Ed.), Handbook of personal relationships: Theory, research and interventions(pp. 271302). New York, NY: John Wiley & Sons, Inc.Gottman, J. M., Markman, H., & Notarious, C. (1977). The topography of marital conflict:A sequential analysis of verbal and non-verbal behaviour. Journal of Marriage and the Family,39, 461477.Hathaway, S. R., & McKinley, J. C. (1989).MMPI-2: Minnesota Multiphasic Personality Inventory-2:Manual for administration and scoring. Minneapolis, MN: University of Minnesota Press.Jacob, T., Leonard, K. E., & Haber, J. R. (2001). Family interactions of alcoholics as relatedto alcoholism. Clinical and Experimental Research, 25, 835843.Khantzian, E. J. (1985). The self-medication hypothesis of addictive disorders: Focus on heroin andcocaine dependence. The American Journal of Psychiatry, 142, 12591264.Kivlahan, D. R., Walker, R. D., Donovan, D. M., & Mischke, H. D. (1985). Detoxificationrecidivism among urban American Indian alcoholics. American Journal of Psychiatry, 142,14671470.Lehman, A. F., Myers, P., Thompson, J. W., & Corty, E. (1993). Implications of mental andsubstance use disorders: A comparison of single and dual diagnosis patients. Journal of Nervousand Mental Disease, 181, 365370.Le Poire, B. A. (1992). Does the codependent encourage substance dependent behavior?Paradoxical injunctions in the codependent relationship. The International Journal of theAddictions, 27, 14651474.Le Poire, B. A. (1995). Inconsistent nurturing as control theory: Implications for communication-based research and treatment programs. Journal of Applied Communication Research, 23, 115.Le Poire, B. A., & Addis, K. (2000). Understanding the qualitative nature of strategies used bypartners of substance abusers in an attempt to reduce recidivism. A Senior Honors Thesis directed bythe first author at the University of California, Santa Barbara.Le Poire, B. A., & Cope, K. M. (1999). Episodic and steady drinkers partners reinforce and punishdrinking differentially. Alcoholism Treatment Quarterly, 17, 7990.Le Poire, B. A., Erlandson, K. T., & Hallett, J. S. (1998). Punishing versus reinforcing strategies ofdiscontinuance: Effect of persuaders drug use. Health Communication, 10, 293316.Le Poire, B. A., Hallett, J. S., & Erlandson, K. T. (2000). An initial test of inconsistent nurturing ascontrol theory: How partners of drug abusers assist their partners sobriety. HumanCommunication Research, 26, 432457.Marlatt, G. A. (1985). Relapse prevention: Theoretical rationale and overview of the model.In G. A. Marlatt & J. R. Gordon (Eds), Relapse prevention: Maintenance strategies in thetreatment of addictive behaviors (pp. 370). New York: Guilford.McCrady, B. S., Noel, N. E., Abrams, D. B., Stout, R. L., Nelson, H. F., & Hay, W. M. (1986).Comparative effectiveness of three types of spouse involvement in outpatient behavioralalcoholism treatment. Journal of Studies on Alcohol, 47, 459465.Moos, R. H., Moos, B. S., & Finney, J. W. (2001). Predictors of deterioration among patients withsubstance-use disorders. Journal of Clinical Psychology, 57, 14031419.Newton, D. A., & Burgoon, J. K. (1990). The use and consequences of verbal influence strategiesduring interpersonal disagreements. Human Communication Research, 16, 477518.Papps, B. P., & OCarroll, R. E. (1998). Extremes of self-esteem and narcissism and the experienceand expression of anger and aggression. Aggressive Behavior, 24, 421438.Sabourin, T. C. (1995). The role of negative reciprocity in spouse abuse. A relational controlanalysis. Journal of Applied Communication Research, 23, 271283.Sheehan, M. F. (1993). Dual diagnosis. Psychiatric Quarterly, 64, 107134.508 C. J. CropleyAddict Res Theory Downloaded from informahealthcare.com by University of California Irvine on 10/31/14For personal use only.Skinner, B. F. (1971). Beyond freedom and dignity. New York: Knopf.Sussman, S., Rychtarik, R. G., Mueser, K., Glynn, S., & Prue, D. M. (1986). Ecological relevanceof memory tests and the prediction of relapse in alcoholics. Journal of Studies on Alcohol, 47,305310.Watson, C. G., Jacobs, L, Pucel, J., Tilleskjor, C., & Hoodecheck, E. (1984). The relationshipof beliefs about controlled drinking to recidivism in alcoholic men. Journal of Studies on Alcohol,45, 172175.Wells-Parker, E., Pang, M. G., Anderson, B. J., McMillen, D. L., & Miller, D. I. (1991). FemaleDUI offenders: A comparison to male counterparts and an examination of the effects ofintervention on womens recidivism rates. Journal of Studies on Alcohol, 52, 142147.Wiseman, J. P. (1991). The other half: Wives of alcoholics and their social psychology. New York:de Gruyter.Personal attributes and persuadability 509Addict Res Theory Downloaded from informahealthcare.com by University of California Irvine on 10/31/14For personal use only.