Deceptive Dynamics in Drug Addiction PDF

853299 research-article2019 JODXXX10.1177/0022042619853299Journal of Drug IssuesCaputo Article...

853299 research-article2019 JODXXX10.1177/0022042619853299Journal of Drug IssuesCaputo Article Journal of Drug Issues Deceptive Dynamics in Drug 2019, Vol. 49(4) 575­–592 © The Author(s) 2019 Article reuse guidelines: Addiction and Their Role in sagepub.com/journals-permissions DOI: 10.1177/0022042619853299 https://doi.org/10.1177/0022042619853299 Control Beliefs and Health journals.sagepub.com/home/jod Status Reporting: A Study on People With Substance Use Disorder in Treatment Andrea Caputo1 Abstract This study aims at exploring deceptive dynamics (i.e., impression management [IM], self- deception, and emotional manipulation [EM]) and their role in control beliefs and health status reporting in a sample of people treated for substance use disorder. Seventy-eight participants following drug rehabilitation treatment were recruited, who provided background information and completed measures of social desirability responding, EM, locus of control, and health- related status. Moderated-regression analyses and t tests were performed. The results highlight that self-deception is associated with not reporting the use of secondary substances and being in treatment for a shorter time period. IM appears as the main deceptive tendency able to account for internal control beliefs and better mental health. Some interaction effects emerge among the examined deceptive tendencies, which suggest to deepen the role of EM as a risk factor for drug relapse and treatment success. Keywords drug addiction, health status, control beliefs, self-deception, impression management, emotional manipulation In drug addiction–related research, there is a strong consensus about the presence of deceptive dynamics in individuals with substance abuse because drug addiction is often conceptualized as a disorder that is characterized by self-delusion, denial, mystification, and dishonesty (Ferrari, Groh, Rulka, Jason, & Davis, 2008; Latkin, Edwards, Davey-Rothwell, & Tobin, 2017). Several reasons may be advocated to explain the use of deceptive dynamics in individuals with substance abuse, such as maintaining the delusion of control over their dependence to preserve a good self- view, denying or justifying their problem to continue with substance abuse, and defending 1Sapienza University of Rome, Italy Corresponding Author: Andrea Caputo, Department of Dynamic and Clinical Psychology, Sapienza University of Rome, Via degli Apuli 1, Rome 00185, Italy. Email: [email protected] 576 Journal of Drug Issues 49(4) themselves from shame and guilt feelings to avoid facing reality, which in turn contribute to lying to themselves, the loved ones, and health care professionals. Specifically, two different deceptive tendencies have been particularly investigated that refer to impression management (IM) and self-deception (Paulhus, 1991). IM refers to the tendency to intentionally present oneself favorably, by self-attributing virtuous characteristics so as to appear good in the eyes of others. Instead, self-deception relies on an unconscious tendency to think of one’s self in an overly positive way, by exaggerating desirable qualities so as to enhance self- concept. Therefore, although IM involves a voluntary manipulation of others’ perception and aims at denying one’s own socially unacceptable behaviors, self-deception does not represent a deliberate manipulation—as the subject firmly believes in his or her descriptions—but mostly aims at fostering self-esteem and adjustment (Lanyon & Carle, 2007; Paulhus, 1991; Paulhus, Bruce, & Trapnell, 1995). With regard to IM, previous studies highlighted that such a tendency is associated with alco- hol and drug severity (Zemore, 2011) and that people with a strong motivation to impress may minimize their involvement in addiction-related problems by downplaying what is perceived as a deviant or unattractive behavior (Davis, Thake, & Vilhena, 2010). Indeed, drug users may rely on IM by creating symbolic boundaries to distinguish themselves as functional users from other less desirable types of users depicted as dysfunctional ones (Copes, 2016; Morris, 2018). Besides, due to the strong stigmatization of drug addiction, substance users try to manage their “spoiled identity” through controlling information about themselves and attempting to influence how they are perceived so as to restore a more positive identity (Goffman, 1968; Hughes, 2007). The effects of IM have been demonstrated in terms of underreporting substance consumption and related harms by about 50% (Davis et al., 2010), especially in structured residential recovery set- tings (Groh, Ferrari, & Jason, 2009). Indeed, socially undesirable behaviors—such as those refer- ring to drug use—tend to be generally misreported (Sloan, Bodapati, & Tucker, 2004), especially the use of the most socially stigmatized drugs (e.g., cocaine; Jackson, 2012). With regard to self-deception, previous pieces of research on drug or alcohol addiction showed that such an unconscious tendency may represent a way to deny the harmful consequences of substance use that may undermine one’s positive self-concept (Davis et al., 2010; Richards & Pai, 2003). In this regard, substance-abusing individuals show lower self-deception than non– drug-using peers (Richards & Pai, 2003); in addition, it is demonstrated that self-deceptive indi- viduals tend to evaluate their addiction-related problems more favorably (Ericsson & Simon, 1993) and not to acknowledge the negative effects (Davis et al., 2010), in turn contributing to relapse (Donovan, 1996; Marlatt & Donovan, 2005). Besides this, previous research highlights that recovering abstainers show lower self-deception over the course of treatment (Mellor, Conroy, & Masteller, 1986; Strom & Barone, 1993), moving from denial and distorting defenses toward honesty and realism (McLellan et al., 1992), with consequent positive impact on readi- ness to change (DiClemente, Schlundt, & Gemmell, 2004). Overall, these two social desirability biases seem to be associated with higher internality in control beliefs of drug users, where positive beliefs about controlling their dependence represent attempts to maintain a positive self-concept or downplay socially unattractive behaviors, thus resulting to be defensive rather than genuine (Ferrari et al., 2008; Strom & Barone, 1993). In addition, drug users who tend to give more socially desirable responses may underreport depres- sive symptoms and mental health issues (Latkin et al., 2017; Rusch, Kanter, Manos, & Weeks, 2008)—also due to the effect of inflated control beliefs on well-being reports found in addictive populations—thus overestimating their health-related status (Heidari, Ghodusi, Bathaei, & Shakeri, 2018). Another relevant tendency that is strongly intertwined with the previously discussed deceptive dynamics refers to emotional manipulation (EM), meant as a deliberately cynical attempt to emotionally manipulate others (Austin, Farrelly, Black, & Moore, 2007). Despite this construct Caputo 577 being less examined in addiction-related research, several authors have claimed that manipula- tion represents a fundamental component of deception (Moral Jiménez & Sirvent Ruiz, 2014) as it relates to mystification aimed at achieving one’s own manipulative goals (Spidel, Herve, Greaves, & Yuille, 2011). Lying and dishonesty are, indeed, quite frequent in people with drug addiction (Ferrari et al., 2008; Sher & Epler, 2004), who often tend to manipulate others close to them to continue their substance use because they are jeopardized by their addiction and intense craving. Some common manipulation tactics refer to making empty promises, playing the victim, making excuses for irresponsibility, making others feel uncomfortable or guilty with the aim of satisfying unreasonable requests, threatening to self-harm, and so on. In this regard, previous research has confirmed that substance use disorders appear as connected with more deviant and antisocial relational tendencies, as well as with psychopathic personality traits (Kimonis, Tatar, & Cauffman, 2012; Phillips, Meek, & Vendemia, 2011). Background and Aim of the study Based on the discussed rationale, three deceptive tendencies seem to emerge in substance use literature, respectively, referring to IM (as conscious tendency to underestimate one’s own socially deviant behaviors for avoidance of negative evaluations), self-deception (as uncon- scious tendency to exaggerate one’s own positive qualities for self-enhancement), and EM (as deliberate cynic tendency to emotionally manipulate others for self-gain). However, the lack of previous research studies about the interrelations among such tendencies does not allow a deeper understanding of their adaptive function for people with drug addiction, within a com- prehensive and ecological framework that looks at the complexity and variety of emotional life. Besides this, such deceptive tendencies have been rarely investigated in drug addictive popula- tions in treatment, notwithstanding their potential relevance in the recovery process for both preventing relapses and guaranteeing treatment success (Caputo, 2019; Donovan, 1996; Marlatt & Donovan, 2005). Based on these premises, this study aims at examining such deceptive dynamics in an Italian sample of people with substance use disorder following rehabilitation. Among European coun- tries, Italy has one of the highest estimates of high-risk opioid use despite the prohibitionist drug policies. In addition, the number of cocaine-using first-time entrants to specialized drug treat- ment centers has been increasing since 2013 (European Monitoring Centre for Drugs and Drug Addiction [EMCDDA], 2016), thus recognizing drug addiction as a public health emergency. Specifically, the Italian drug treatment system, which is organized at the regional level and funded by the national government on a yearly basis, includes public drug addiction service units (SerTs), as part of the national health system, and social-rehabilitative facilities, mostly provided by private organizations. SerTs mainly carry out outpatient treatment by integrated interventions and reintegration programs, whereas social-rehabilitative facilities deal with both inpatient and outpatient treatment within residential or semi-residential therapeutic communities. Referral to therapeutic communities is made and paid for by the SerTs, based on the diagnostic criteria for substance use disorder. Overall, interventions include psychotherapy, psychosocial, and social service interventions; detoxification (which may be also carried out in general hospitals) in resi- dential settings; and vocational training in semi-residential ones. Italy has a long-lasting tradition of therapeutic community approach (Vanderplasschen, Vandevelde, & Broekaert, 2014), includ- ing two thirds of such social-rehabilitative facilities across Europe. The key feature of the pro- grams offered by therapeutic communities is the provision of a drug-free environment within small, family-type structures, by promoting responsible agency, sense of belongingness, and respect of norms and values for next reintegration into the outside society. Therefore, in line with Goffman’s (1968) thought, the role played by deceptive dynamics in rehabilitative environments seems to be quite relevant because individuals within therapeutic communities may feel ashamed, 578 Journal of Drug Issues 49(4) insecure, and subordinated due to their highly visible addictive status and the required compli- ance with drug-free prescriptions. This could trigger an intense need for IM and boundary con- struction about substance use during the recovery process (Copes, 2016; Morris, 2018). Indeed, people following treatment may attempt to overcome stigma renegotiating what normal means and divorcing themselves from a negative drug user identity (Neale, Nettleton, & Pickering, 2011; Radcliffe & Stevens, 2008). Therefore, self-presentation may be affected by self-enhance- ment strategies to repair their spoiled identity (Hughes, 2007), also involving manipulative tac- tics in relation to health care staff and other community users (Caputo, 2019). Specifically, three research goals are addressed in the present research study as follows: 1. Examine potential differences in IM, self-deception, and EM based on sociodemographic, drug abuse, and treatment-related variables; 2. Explore the potential role of IM, self-deception, and EM in accounting for control beliefs and physical and mental health status; 3. Test whether some interaction effects exist among IM, self-deception, and EM on control beliefs and physical and mental health status. Method Participants The participants were recruited within four different therapeutic communities for drug rehabilita- tion treatment in Central Italy. Inclusion criteria for enrollment into the study were as follows: being aged 18 to 65, being diagnosed with a substance use disorder, not having a co-occurring psychological disorder, and following a rehabilitation program. Overall, 78 individuals voluntarily participated in a cross-sectional research study, after having provided written and informed con- sent. The participants were mostly males (89.2%), on average aged 36.54 years (SD = 9.56), with a mean of 10.27 years of schooling (SD = 2.67). With regard to their history of drug abuse, 74.7% of them had initiated drug use before the age of 18 years; 39.5% used multiple primary substances and 69.2% used further secondary substances. About treatment-related variables, 67.9% of partici- pants had already received previous detoxification and entered the therapeutic community on a voluntary basis (71.8%) or referred by others (28.2%), following a residential (75.6%) or semi- residential (24.4%) program since a mean of 10.84 months (SD = 8.76). The participants com- pleted a paper-and-pencil questionnaire—including background information and measures of health-related status, locus of control, social desirability responding, and EM—that was adminis- tered at a single time point. All procedures followed were approved by the local ethical review board and were in accordance with the Helsinki Declaration regarding human research. Measures Background information. Sociodemographic characteristics (gender, age, education, employment, and relational status), history of drug abuse (time from first drug use, use of multiple primary substances, and use of secondary substances), and treatment-related variables (previous detoxifi- cation, motivation for treatment initiation, type of treatment program, and treatment duration) were assessed. Dependent Variables Health-related status. This construct was evaluated by means of the Italian version of the 12-item Short Form Health Survey (SF-12; Apolone et al., 2005; Kodraliu et al., 2001) measuring physical Caputo 579 and mental health status. It evaluates the subjective perception of the individual in relation to the concepts of health and wellness during the previous month. SF-12 provides two indexes referred to as physical component summary (PCS) and mental component summary (MCS), respectively, with highest scores corresponding to better health-related status. Cronbach’s alpha for the pres- ent study was.77 and.78 for PCS and MCS, respectively. Locus of control. This construct was assessed through the Locus of Control of Behavior (LCB; Craig, Franklin, & Andrews, 1984), a 17-item questionnaire measuring the person’s ability to take responsibility for maintaining new or desired behaviors using a Likert-type scale ranging from 0 to 5. Higher scores indicate a perception of external locus of control and lower scores indicate a perception of greater internal locus of control. Cronbach’s alpha for the present study was.72. Independent Variables Social desirability responding. This construct was assessed through the short version of the Paulhus Balanced Inventory of Desirable Responding (BIDR-6; Bobbio & Manganelli, 2011; Paulhus, 1991). It consists of 16 items measuring two different factors using a 6-point Likert-type scale: self-deceptive enhancement (SDE), meant as the unconscious tendency to provide honest but positively biased self-descriptions, and IM, indicating the deliberate tendency to overreport desirable behaviors and underreport undesirable ones. Cronbach’s alpha for the present study was.70 and.75 for SDE and IM, respectively. EM. This construct was measured through a six-item scale inspired by the EM subscale of the Emotion Manipulation Scale (EMS) developed by Austin et al. (2007) and adapted to Italian for the present study. The scale measures the extent to which an individual believes she or he can emotionally manipulate others (e.g., “If I want to, I know how to make another person feel uneasy” or “When I need it, I’m able to gain favors from others”) using a 5-point Likert-type scale ranging from 0 (strongly disagree) to 5 (strongly agree). The exploratory factor analysis on the current sample extracted a single factor explaining 42.63% of the overall variance, whereas the confirmatory factor analysis confirmed the goodness-of-fit indexes as follows, χ2 (12.74, df = 9), comparative fit index (CFI) =.949, Tucker–Lewis index (TLI) =.916, standardized root mean square residual (SRMR) =.054, and root mean square error of approximation (RMSEA) =.074, with all standardized coefficients being statistically significant. Cronbach’s alpha for this scale was.72. Data Analysis Preliminary descriptive statistics and Pearson’s correlations among the study variables were calculated. The t tests were computed to examine differences in IM, SDE, and EM based on sociodemographic, drug abuse, and treatment-related variables. Then, several moderated- regression analyses were performed, consistently with those suggested by Aiken and West (1991). To have a complete standardized solution, all of the variables were standardized in advance (z scores). Afterward, the standardized scores of IM, SDE, and EM were multiplied in pairs to form the interaction term. The moderation effect was then tested in different steps using hierarchical regression analysis: In the first step, each pair of predictors was regressed on the criterion (locus of control, physical, and mental health status, respectively), and in the second step, the interaction terms were finally included in the regression equation so as to test a poten- tial significant increment in the explained variance. Therefore, three multiplicative interaction terms were separately tested (IM × SDE, ΙΜ × EM, and SDE × EM) for each outcome (locus 580 Journal of Drug Issues 49(4) Table 1. Descriptive Statistics, Intercorrelations, and Scale Reliabilities of the Study Measures (N = 78). 1 2 3 4 5 6 1. Impression management — — — — — — 2. Self-deceptive enhancement.10 — — — — — 3. Emotional manipulation −.26*.14 — — — — 4. Locus of control −.28*.01 −.05 — — — 5. Physical health status −.11.15.21 −.11 — — 6. Mental health status.28*.10.08 −.27* −.23* — α.75.70.72.72.77.78 M 26.85 28.47 26.84 21.62 50.02 41.35 SD 9.79 8.22 5.55 10.87 7.92 11.07 Note. *Statistically significant at.05 level. of control and physical and mental health status). The interactions were examined graphically by plotting the standardized slopes of the regression on the other predictor at high (one standard deviation above the mean) and low (one standard deviation below the mean) values of the mod- erator. Correlations between the predictor and the criterion were calculated at high and low levels of the moderator, and differences between correlation coefficients were then tested. Given our small sample size (78 participants), further independent variables were not included in the model at this stage (in addition to each pair of predictors plus the interaction term), according to the formula (N > 50 + 8m, where N is the sample size and m is the number of independent variables) suggested by Tabachnick and Fidell (2013) to ensure adequate statistical power in a multiple regression equation. However, to test the robustness of the detected effects and increase the accuracy of the findings, further moderated-regression analyses were performed by adding the following covariates in a preliminary step of the estimation procedure: Gender, use of mul- tiple primary substances, use of secondary substances, and type of treatment were included as dichotomous variables, whereas age and duration of treatment were included as continuous variables. This was done to examine whether the interaction effects still persisted even control- ling for some relevant sociodemographics, drug abuse, and treatment-related variables of the study participants. As multiple analyses (t tests and moderated-regression analyses) were performed to address our research goals, which may result in inflation of Type I error (false positives), we further con- sidered the effect size (and relative confidence intervals [CIs]) of the observed relationships apart from the statistical significance to provide a more comprehensive method of statistical results interpretation (Garamszegi, 2006; Lee, 2016). According to what was suggested by Cohen (1988), to detect small, medium, and large effect sizes we used the following cutoffs: r values of.1,.3, and.5 to test the strength of associations (Pearson’s correlations and standardized regres- sion coefficients), d values of.2,.5, and.8 to test pooled standardized mean differences (indepen- dent samples t tests), and f2 values of.02,.15, and.35 to test the incremental variance due to interactions (hierarchical multiple regressions). Results Table 1 shows descriptive statistics, intercorrelations, and scale reliabilities of the study mea- sures. Overall, some small-to-medium-sized correlations were detected among the examined variables: IM was negatively associated with both EM (r = –.26, p <.05, 95% CI = [–.46, –.04]) and (external) locus of control (r = –.28, p <.05, 95% CI = [–.47, –.06]) and positively associated with mental health status (r =.28, p 35) 56.9 28.03 (9.42) — 29.14 (8.75) — 25.60 (5.76) — Education level    Lower (up to 8 years 51.3 26.32 (10.15) −0.48 27.49 (8.89) −1.06 26.90 (5.99) 0.09 of education)   Higher (>8 years of 48.7 27.41 (9.51) — 29.47 (7.46) — 26.78 (5.13) — education) Drug use variables Age of first drug use    Not legal age 74.7 26.47 (10.16) −0.48 28.38 (8.29) −0.44 27.66 (5.23) 1.95 (