- Email: [email protected]
The implications for the biological and sociodynamic causal explanations of attitudes toward alcohol-dependent patients
Psychiatry Research 215 (2014) 766–770
Contents lists available at ScienceDirect
Psychiatry Research journal homepage: www.elsevier.com/locate/psychres
The implications for the biological and sociodynamic causal explanations of attitudes toward alcohol-dependent patients Annemarie Heberlein a,n, Rilana Schuster a, Yvonne Ziert c, Birgitt Opfermann b, Stefan Bleich a, Thomas Hillemacher a a
Department of Psychiatry, Social Psychiatry and Psychotherapy, Hannover Medical School, Germany Medical Service of the Health Funds Niedersachsen, Germany c Centre for Biometry, Medical Informatics and Medical Technology, Hannover Medical School, Germany b
art ic l e i nf o
a b s t r a c t
Article history: Received 4 January 2013 Received in revised form 15 July 2013 Accepted 22 December 2013 Available online 4 January 2014
This study tested whether sole neurobiological or sociodynamic explanations of alcohol dependence altered respondents0 attitudes toward alcohol-dependent patients. We investigated the effect of information leaflets on 444 participants: one group received an information leaflet with a biological explanation of AD; the other received a leaflet with a sole sociodynamic explanation of AD. A third, control group did not receive any leaflet. Afterwards, all three groups completed a questionnaire regarding their attitudes toward ADPs and their opinions of the underlying causes of AD. We found a significant group difference with regard to participants0 agreement with a neurobiological explanation of AD. Moreover, respondents in the neurobiological intervention group considered the characteristics of ADP to be significantly more positive than those in the sociodynamic group. Furthermore, they were significantly less likely to accept AD as a self-inflicted disease. Correlation analysis revealed associations between accepting the sociodynamic disease model and all of the stigmatization dimensions tested in our questionnaire. In summary, stigmatization toward ADP was closely associated with the agreement with sociodynamic origins of AD in this study. & 2014 Elsevier Ireland Ltd. All rights reserved.
Keywords: Alcohol dependence Disease models Framing Stigmatization Attitudes Neurobiological and sociodynamic explanations
1. Background Although Alcohol Dependence (AD) is a disease with both environmental and biological causal factors, public opinion varies widely regarding the origins of AD. In general, AD is widely accepted as a “controllable condition” characterized by an “instrumental rationality” (Pickard, 2011), rather than considered as a disease. Studies have consistently shown that Alcohol Dependent Patients (ADPs) provoke more social rejection and more negative emotions than those who suffer from psychiatric diseases that are distinct from AD (Schomerus et al., 2011). The origins of this strong stigmatization toward ADPs have been frequently identified with regard to the attribution of responsibility and self-infliction explanations of alcohol consumption: the so-called “attribution theory” states that an association exists between the notion that a condition is “controllable” and the attribution of responsibility for
Abbreviations: AD, alcohol dependence; ADP, alcohol-dependent patient; MD, mean difference; S.D., standard deviation; NB-G, neurobiological group; SD-G, sociodynamic group; CO-G, control group n Correspondence to: Center for Addiction Research (CARe), Department of Psychiatry, Social Psychiatry and Psychotherapy, Hannover Medical School, CarlNeuberg-Straße 1, D-30625 Hanover, Germany. Tel.: þ 49 511 532 0. E-mail address: [email protected] (A. Heberlein). 0165-1781/$ - see front matter & 2014 Elsevier Ireland Ltd. All rights reserved. http://dx.doi.org/10.1016/j.psychres.2013.12.040
that condition (e.g., psychiatric diseases are the patients0 faults; (Angermeyer et al., 2011; Read and Harré, 2001; Weiner, 1988)). For example, Weiner (1988) reported that physical disabilities (e.g., blindness) were perceived as uncontrollable and therefore elicited pity rather than blame. Surveys have consistently shown a high acceptance rate (up to 60 percent) regarding whether AD is a self-inflicted condition but much lower rates have been found regarding whether other psychiatric diseases such as eating disorders (34 percent agreement), depression, panic attacks or, schizophrenia (4–13 percent agreement) are self-inflicted (Crisp et al., 2005; Crisp et al., 2000). Several studies (Crisafulli et al., 2008; Lincoln et al., 2008) have tested the central hypothesis of the attribution theory: advocating the neurobiological origins of psychiatric diseases (thereby stressing uncontrollability and guiltlessness) engenders attitudes toward psychiatric diseases in line with attitudes toward somatic diseases (Boyle et al., 2009). Although the results do not support the general reduction of stigmatization toward patients who suffer from psychiatric diseases, advocating for a biogenetic explanation model partially supports the reduction of moral censure hypothesis (by disrupting the association between self-infliction and psychiatric disease) for patients who suffer from psychiatric diseases such as schizophrenia (Lincoln et al., 2008) and anorexia nervosa (Crisafulli et al., 2008).
A. Heberlein et al. / Psychiatry Research 215 (2014) 766–770
Our study sought to determine the degree to which the attitudes toward ADPs are affected by the support of neurobiological and sociodynamic models of AD. In particular, we tested the effects of neurobiological versus sociodynamic explanations of AD with regard to the following dimensions: (1) Agreement with neurobiological origins (supporting an intervention effect). (2) Agreement with sociodynamic origins (supporting an intervention effect). (3) Evaluation of ADP characteristics. (4) Agreement with AD self-infliction. (5) Agreement with ADP heteronomy with regard to self-treatment. (6) Estimation of a poor therapeutic prognosis with regard to AD.
767
Randomization was achieved by dividing the questionnaires (NB-G, SD-G, and CO-G) across the seating order of the participants. The study supervisor (RS) collected the surveys after 10–15 min. 2.3. Structure of the questionnaire One of the authors (AH) translated the original set of questions into German, and a native English speaker back-translated the German version. This protocol was conducted until the back-translated German version was identical to the original English version. The original questionnaire (Crisafulli and colleagues) was supplemented by several additional questions regarding attitudes toward ADPs and the presumed origin of AD. In particular, the supplement explored whether participants believe that ADPs are capable of consenting to treatment (i.e., whether they are selfdetermined with regard to therapeutic decisions and assumptions). Statements were worded unambiguously e.g., “ADPs should autonomously decide their own treatment”; “Doctors should decide the treatment regimen to be used”; and “ADPs should be forced into treatment”. The final questionnaire consisted of 75 questions. All questions were answered using a six-point Likert scale. 2.4. Data analyses
2. Methods The current investigation was conducted in accordance with the Declaration of Helsinki and approved by the Ethics Committee of the Medical School of Hannover.
Data analyses were performed using SPSS 20. Missing data were replaced by multiple imputation (via the multiple regression method). The 75 questionnaire items comprised the following seven subscales (two causality scales and five stigma scales):
2.1. Participants A total of 444 people (312 women and 125 men) participated in this study. Participants were recruited at the Medical School of Hannover and the University of Hildesheim, Germany. Table 1 provides an overview of the study population. The survey was completed before the beginning of classes and during a monthly staff orientation at each university. All participants were informed of the study goals of the questionnaire. They were instructed to first read the portion regarding the development of AD and then answer all of the questions spontaneously and honestly. No additional information was provided.
2.2. Questionnaire The questionnaire, including the portion on the development of AD, was adapted from Crisafulli et al. (2008) who investigated the effect of disease models on the attitudes of nursing students toward patients with anorexia. We designed two information leaflets by adapting the structure of Crisafulli and colleagues0 original leaflet. The leaflets provided background information that suggested either neurobiological or sociodynamic causes of AD. The diagnostic criteria of AD according to the ICD-10 were at the top of the survey, followed by citations of studies regarding the neurobiological (or sociodynamic) causes of AD. We altered the original study by Crisafulli et al. (2008) by adding a control group (CO-G) who answered the questionnaire without information regarding the origins of AD. This protocol allowed us to evaluate the effect of the two leaflets on general attitudes toward ADPs. Participants were randomly assigned to one of three groups: (a) Participants assigned to the NB-G received information that explained AD as a neurobiological condition based on the biological characteristics of the central nervous system. (b) Participants assigned to the SD-G received information that explained AD as a consequence of sociodynamic factors such as lifestyle, drinking habits within one0 s family and peer group, the availability of alcohol, and acquired personality traits. (c) Participants assigned to the CO-G did not receive prior information regarding the origins of AD.
Table 1 Participant demographics by group: graduated participants were classified as “academics”. Groups were not significantly different regarding education, age and sex. Group
Control group Sociodynamic group Neurobiological group
Sex
Education level
Age
♀
♂
Academics
Non-academics
Mean/S.D.
78 98 111
36 27 52
30 31 52
76 93 105
26.21/6.82 25.78/6.46 26.41/6.59
Causality scales (1) Neurobiological causes of AD (Cronbach0 s α ¼ 0.879) (2) Social causes of AD (Cronbach0 s α ¼ 0.763) Stigma scales (3) Characteristics of ADPs (Cronbach0 s α ¼ 0.870) (4) Dangerousness of ADPs (Cronbach0 s α ¼ 0.784) (5) Self-infliction of AD (Cronbach0 s α ¼ 0.733) (6) Heteronomy regarding therapeutic decisions (Cronbach0 s (7) Poor prognosis of AD (0.825)
α ¼0.800)
The reliabilities of the extracted subscales were tested with Cronbach0 s α. Questions that reduced the α-level of the subscales were removed from the analysis (five questions on the characteristics scale and two questions regarding surgical approaches to AD). A confirmatory factor analysis (i.e., a principal component analysis with Promax rotation) was applied to verify the major topics of the questionnaire. An oblique factor analysis was used because it allows correlations (which are common in the social sciences) among the factors extracted and therefore provides more consistent and replicable results. To assure the consistency of the extracted factors, we applied the principal component and maximum likelihood methods to extract factors (Costello and Osborne, 2005). A group-to-group comparison was conducted using an analysis of covariance (ANCOVA). The seven subscales were set as dependent variables, and the three groups (SD-G, NB-G and CO-G) were the fixed factors. Age, sex and profession were considered covariates. Effect sizes were calculated using partial η2. Bonferroni0 s post hoc test was used to compare the three sub-groups. We applied a significance level of α ¼0.007 due to repeated testing (i.e., a correction factor of 7). Pearson0 s correlation analysis was applied in order to investigate associations between the two causality scales and the five stigma scales.
3. Results 3.1. Characteristics of the respondents and the questionnaire A total of 444 questionnaires were included in the final analysis. Table 1 shows the characteristics of the respondents. 3.2. Group-to-group analyses We found a significant group difference with regard to the biological origins of AD (F¼17.649, p o0.001, η2 ¼ 0.054): a significant increase in the acceptance of the neurobiological disease model was found among the NB-G compared with the CO-G (p o0.001) and the SD-G (po 0.001) groups. Moreover, the NB-G respondents rated the typical characteristics of ADPs to be significantly more positive (F¼8.471, po0.001, η2 ¼0.054) than the SD-G (po0.001) and the CO-G (po0.001) respondents.
768
A. Heberlein et al. / Psychiatry Research 215 (2014) 766–770
A significant increase in the acceptance of self-infliction of AD (F¼ 2.860, p ¼0.014, η2 ¼0.018) was observed among the SD-G compared with the NB-G (p ¼0.003). A significant group-to-group difference was found regarding the evaluation of ADPs0 competence to freely decide their treatment schedule (F¼4.341, p¼ 0.001, η2 ¼ 0.026): CO-G respondents showed more acceptance of the heteronomy regarding therapeutic decisions compared with the NB-G respondents (p ¼0.001). Though, a significant effect of age was also observed (p¼ 0.002, η2 ¼ 0.12), which even outweighed the group effect. Moreover, a trend toward a group difference was observed with regard to an AD prognosis (F ¼2.404, p ¼0.035, η2 ¼0.015): SD-G respondents rated AD prognoses to be significantly more positive than CO-G (p ¼ 0.003) and NB-G (p ¼ 0.002) respondents. Table 2 shows the results of the ANCOVA and its post-hoc tests.
reason for the lack of an effect of the sociodynamic information leaflet among the SD-G might be their high pre-test agreement with the sociodynamic foundation hypothesis of AD. Consistent with this explanation, we found a mean of 4.21 points (using a sixpoint Likert scale) on the sociodynamic scale compared with a mean of 2.94 on the neurobiological scale among the CO-G. Moreover, we found an even stronger agreement with the sociodynamic (mean 4.21 points) but not neurobiological (mean 3.40 points) explanations of AD among the NB-G. This observation supports the hypothesis that the sociodynamic explanation of AD is strongly supported in Germany, upon which stigma-related attitudes might be grounded.
4.2. Association between stigmatiferous attitudes and the sociodynamic explanation of AD
3.3. Correlation analysis The sociodynamic scale was significantly associated with all five STIGMA subscales that measured the various dimensions of stigma in this study (see Table 3 for details). No association was found between the neurobiological scale and these subscales.
We found that the sociodynamic subscale was strongly associated with the five stigma subscales. In particular, a strong acceptance of the sociodynamic causes of AD was negatively associated with the evaluation of positive characteristics and positively associated with the acceptance of heteronomy of ADPs regarding therapeutic decisions, the assumed dangerousness of ADPs and the acceptance of AD as a self-inflicted condition.
4. Discussion Table 3 Correlation of the Causality scales with the five stigma subscales.
4.1. Effect of intervention Reading the neurobiological information leaflet significantly increased agreement with the proposed neurobiological causes of AD among the NB-G compared with the other groups (see Table 2 for details). A similar effect was not observed among the SD-G, who did not show a significantly greater agreement with the sociodynamic causes of AD compared with the other groups. One
Sociodynamic subscale r ¼ 0.174, p o 0.001 r ¼0.223, p o 0.001 r ¼0.210, p o 0.001 r ¼0.129, p o0.001 r ¼ 0.139, p o 0.001
Characteristics subscale Self-infliction subscale Dangerousness subscale Heteronomy subscale Poor prognosis subscale
Table 2 Group-to-group differences regarding the seven subscales of the questionnaire.
Neurobiological origins
Sociodynamic origins
Positive characteristics
Dangerousness
Self-infliction
Acceptance of heteronomy regarding treatment decisions
Poor prognosis
CO-G
SD-G
NB-G
F¼ 17.649, po 0.001, η2 ¼0.054 SD-G NB-G 0.162 0.463 p ¼0.121 p o 0.001
CO-G 0.162 p ¼ 0.121
NB-G 0.625 p o 0.001
CO-G 0.463 po 0.001
SD-G 0.625 p o 0.001
F¼ 0.955, p ¼0.445, η2 ¼ 0.006 SD-G NB-G 0.029 0.102 p ¼0.581 p ¼0.039
CO-G 0.029 p ¼ 0.581
NB-G 0.073 p ¼ 0.130
CO-G 0.102 p¼ 0.039
SD-G 0.073 p ¼0.130
F¼ 8.471, po 0.001, η2 ¼0.054 SD-G NB-G 0.151 0.070 p ¼0.004 p ¼0.159
CO-G 0.151 p ¼ 0.004
NB-G 0.220 p o 0.001
CO-G 0.070 p¼ 0.159
SD-G 0.220 p o 0.001
F¼ 2.479, p ¼0.084, η2 ¼0.006 SD-G NB-G 0.032 0.128 p ¼0.701 p ¼0.125
CO-G 0.032 p ¼ 0.701
NB-G 0.159 p ¼ 0.038
CO-G 0.128 p¼ 0.105
SD-G 0.159 p ¼0.038
F¼ 2.860, p ¼0.014, η2 ¼0.018 SD-G NB-G 0.122 0.063 p ¼0.072 p ¼0.327
CO-G 0.122 p ¼ 0.072
NB-G 0.185 p ¼ 0.003
CO-G 0.063 p¼ 0.327
SD-G 0.185 p ¼0.003
F¼ 4.341, p ¼ 0.001, η2 ¼ 0.026 SD-G NB-G 0.183 0.306 p ¼0.060 p ¼0.001
CO-G 0.183 p ¼ 0.060
NB-G 0.123 p ¼ 0.175
CO-G 0.306 p¼ 0.001
SD-G 0.123 p ¼0175
F¼ 2.404, p ¼ 0.035, η2 ¼0.015 SD-G NB-G 0.234 0.002 p ¼0.006 p ¼0.975
CO-G 0.234 p ¼ 0.006
NB-G 0.231 p ¼ 0.003
CO-G 0.002 p¼ 0.975
SD-G 0.231 p ¼0.003
A. Heberlein et al. / Psychiatry Research 215 (2014) 766–770
The association between the acceptance of the social causes of AD and AD-related stigma was mirrored by the group-to-group differences among the three intervention groups: the NB-G (p o0.001) and CO-G (p ¼0.004) rated the typical characteristics of ADPs as significantly more positive than the SD-G, whereas the NB-G showed significantly less acceptance of AD as a self-inflicted disease (p ¼0.003) and with the putatively poor prognosis (p ¼0.003) of AD compared with the SD-G (po 0.001). 4.3. The stigma of AD compared with the stigma of further psychiatric diseases
769
distinguishes the stigma of AD from that of other psychiatric diseases. Alcohol consumption is common in society, which supports the teleological usefulness of stigmatizing consumption beyond social boundaries (Phelan et al., 2008). Compared with other psychiatric conditions that are commonly accepted as diseases, alcoholism is often not associated with a disease. This common belief in alcohol consumption to be a controllable condition rather than an un-controllable disease might explain the close relationship between the sociodynamic disease model and the different aspects of AD-related stigma found in our data. 4.4. Limitations
As recent research (Pescosolido et al., 2010) has shown, the increased acceptance of the neurobiological disease model of psychiatric diseases such as schizophrenia (91 percent in 2006) and depression (72 percent in 2006) is associated with increased desires for social distance (between 72 and 74 percent of respondents) as well as increased perceptions of violent patients (increased from 78 to 79 percent) and detrimental characteristics (increase from 49 to 65 percent). Therefore, the lack of effects (Angermeyer et al., 2011, 2013; Pescosolido et al., 2010; Rüsch et al., 2010) or even the negative effects (Angermeyer et al., 2011, 2013) of the neurobiological disease model regarding the stigma of psychiatric disease have been reported. Unlike other psychiatric diseases, the results of this study indicate a close association between the acceptance of the sociodynamic disease model of AD and the various dimensions of AD-related stigma investigated in the questionnaire: supporting the central hypothesis of attribution theory, we found a strong negative association between the estimation of a putatively poor prognosis of AD and the acceptance of sociodynamic causes of AD. The supposed controllability of AD, which is mirrored by the assumed possibility of a successful prognosis, might engender stigma-related attitudes with regard to the typical characteristics of ADPs and their potential danger. This hypothesis is consistent with the teleological interpretation of AD-related stigma originally proposed by Phelan et al. (2008): in her interpretation AD-related stigma is understood to secure important social norms by clarifying the boundaries of acceptable social behavior and demonstrating the consequences of nonconformance (Schomerus et al., 2011). Therefore, believing that AD is an uncontrollable condition by presenting a sole neurobiological disease model might have loosened the connotation between moral failings and AD in the NB-G while strengthening this connotation in the SD-G. Consistently, we observed a significant group to group difference regarding the stigmatiferous attitudes of the SD-G and the NB-G, which was not seen between the NB-G and the CO-G. Our results mirror results presented by Schomerus et al. (2013): they reported that the acceptance of a biogenetic explanation of AD was not associated with significant effects regarding the social acceptance of ADPs, although it decreased respondents0 acceptance of onset responsibility of AD. Consistently, the neurobiological information leaflet presented in our study did not significantly impact respondents0 attitudes regarding the five stigma subscales when compared to the CO-G. In conclusion, our results support the central hypothesis of attribution theory: the agreement with the sociodynamic causal explanations of AD was closely associated with all dimensions of stigma investigated in this study. However, the neurobiological disease model did not reduce the stigma-related attitudes of the NB-G respondents as was demonstrated by comparison of the attitudes of the NB-G and the CO-G, whereas the sociodynamic disease model vice versa worsened respondents0 attitudes regarding the acceptance of putative detrimental characteristics of ADP. The shared knowledge of ADPs might explain these results: All participants in this study were familiar with alcohol consumption and ADP. The experience of respondents might be a key point that
This study has several limitations: first, our study was restricted to the Hannover region of Germany. As discussed before, cultural differences exist with regard to the stigmatization of AD (Schomerus et al., 2011); thus, our sample is not representative of ADP stigmatization in general or in Germany specifically. Moreover, our sample consisted of students and stuff from the Medical School of Hannover and might therefore differ in their attitudes toward ADPs compared with the general public. However, similar results were found in a recent nationwide German sample (Schomerus et al., 2013) supporting the hypothesis that stigma related to AD differs from stigma related to further psychiatric diseases. Furthermore, this study is limited by its use of a non-validated questionnaire. Although we found that the subscales showed high reliabilities, our results should be confirmed by future studies with validated questionnaires. 4.5. Conclusion In summary, this study and earlier results illustrate that advocating the neurobiological origins is not a successful method of decreasing the stigma related to AD due to its lack of effect and because it does not provide accurate information regarding the currently known causes of AD. References Angermeyer, M.C., Holzinger, A., Carta, M.G., Schomerus, G., 2011. Biogenetic explanations and public acceptance of mental illness: systematic review of population studies. British Journal of Psychiatry 199, 367–372. Angermeyer, M.C., Mnich, E., Daubmann, A., Herich, L., Wegscheider, K., Kofahl, C., von dem Knesebeck, O., 2013. Biogenetic explanations and public acceptance of people with eating disorders. Social Psychiatry and Psychiatric Epidemiology 48 (10), 1667–1673, http://dx.doi.org/10.1007/s00127-012-0648-9. Boyle, M.P., Blood, G.W., Blood, I.M., 2009. Effects of perceived causality on perceptions of persons who stutter. Journal of Fluency Disorders 34, 201–218. Costello, A.B., Osborne, J.W., 2005. Best practices in explanatory factor analysis: four recommondations for getting the most from your analysis. Practical Assessment, Research and Evaluation 10, 1–9. Crisafulli, M.A., Von Holle, A., Bulik, C.M., 2008. Attitudes towards anorexia nervosa: the impact of framing on blame and stigma. International Journal of Eating Disorders 41, 333–339. Crisp, A., Gelder, M., Goddard, E., Meltzer, H., 2005. Stigmatization of people with mental illnesses: a follow-up study within the Changing Minds campaign of the Royal College of Psychiatrists. World Psychiatry: Official Journal of the World Psychiatric Association 4, 106–113. Crisp, A.H., Gelder, M.G., Rix, S., Meltzer, H.I., Rowlands, O.J., 2000. Stigmatisation of people with mental illnesses. The British Journal of Psychiatry: the Journal of Mental Science 177, 4–7. Lincoln, T.M., Arens, E., Berger, C., Rief, W., 2008. Can antistigma campaigns be improved? A test of the impact of biogenetic vs psychosocial causal explanations on implicit and explicit attitudes to schizophrenia. Schizophrenia Bulletin 34, 984–994. Pescosolido, B.A., Martin, J.K., Long, J.S., Medina, T.R., Phelan, J.C., Link, B.G., 2010. “A disease like any other”? A decade of change in public reactions to schizophrenia, depression, and alcohol dependence. The American Journal of Psychiatry 167, 1321–1330. Phelan, J.C., Link, B.G., Dovidio, J.F., 2008. Stigma and prejudice: one animal or two? Social Science and Medicine 67, 358–367. Pickard, H., 2011. The instrumental rationality of addiction. Behavioral and Brain Sciences 34, 320–321.
770
A. Heberlein et al. / Psychiatry Research 215 (2014) 766–770
Read, J., Harré, N., 2001. The role of biological and genetics beliefs in the stigmatisation of “mental patients”. Journal of Mental Health 10, 223. Rüsch, N., Todd, A.R., Bodenhausen, G.V., Corrigan, P.W., 2010. Biogenetic models of psychopathology, implicit guilt, and mental illness stigma. Psychiatry Research 179, 328–332. Schomerus, G., Lucht, M., Holzinger, A., Matschinger, H., Carta, M.G., Angermeyer, M.C., 2011. The stigma of alcohol dependence compared with other mental disorders: a review of population studies. Alcohol and Alcoholism 46, 105–112.
Schomerus, G., Matschinger, H., Angermeyer, M.C., 2013. Causal beliefs of the public and social acceptance of persons with mental illness: a comparative analysis of schizophrenia, depression and alcohol dependence. Psychological Medicine, 1–12 Weiner, B., 1988. Attribution theory and attributional therapy: some theoretical observations and suggestions. British Journal of Clinical Psychology 27 (Pt 1), 99–104.
Contents lists available at ScienceDirect
Psychiatry Research journal homepage: www.elsevier.com/locate/psychres
The implications for the biological and sociodynamic causal explanations of attitudes toward alcohol-dependent patients Annemarie Heberlein a,n, Rilana Schuster a, Yvonne Ziert c, Birgitt Opfermann b, Stefan Bleich a, Thomas Hillemacher a a
Department of Psychiatry, Social Psychiatry and Psychotherapy, Hannover Medical School, Germany Medical Service of the Health Funds Niedersachsen, Germany c Centre for Biometry, Medical Informatics and Medical Technology, Hannover Medical School, Germany b
art ic l e i nf o
a b s t r a c t
Article history: Received 4 January 2013 Received in revised form 15 July 2013 Accepted 22 December 2013 Available online 4 January 2014
This study tested whether sole neurobiological or sociodynamic explanations of alcohol dependence altered respondents0 attitudes toward alcohol-dependent patients. We investigated the effect of information leaflets on 444 participants: one group received an information leaflet with a biological explanation of AD; the other received a leaflet with a sole sociodynamic explanation of AD. A third, control group did not receive any leaflet. Afterwards, all three groups completed a questionnaire regarding their attitudes toward ADPs and their opinions of the underlying causes of AD. We found a significant group difference with regard to participants0 agreement with a neurobiological explanation of AD. Moreover, respondents in the neurobiological intervention group considered the characteristics of ADP to be significantly more positive than those in the sociodynamic group. Furthermore, they were significantly less likely to accept AD as a self-inflicted disease. Correlation analysis revealed associations between accepting the sociodynamic disease model and all of the stigmatization dimensions tested in our questionnaire. In summary, stigmatization toward ADP was closely associated with the agreement with sociodynamic origins of AD in this study. & 2014 Elsevier Ireland Ltd. All rights reserved.
Keywords: Alcohol dependence Disease models Framing Stigmatization Attitudes Neurobiological and sociodynamic explanations
1. Background Although Alcohol Dependence (AD) is a disease with both environmental and biological causal factors, public opinion varies widely regarding the origins of AD. In general, AD is widely accepted as a “controllable condition” characterized by an “instrumental rationality” (Pickard, 2011), rather than considered as a disease. Studies have consistently shown that Alcohol Dependent Patients (ADPs) provoke more social rejection and more negative emotions than those who suffer from psychiatric diseases that are distinct from AD (Schomerus et al., 2011). The origins of this strong stigmatization toward ADPs have been frequently identified with regard to the attribution of responsibility and self-infliction explanations of alcohol consumption: the so-called “attribution theory” states that an association exists between the notion that a condition is “controllable” and the attribution of responsibility for
Abbreviations: AD, alcohol dependence; ADP, alcohol-dependent patient; MD, mean difference; S.D., standard deviation; NB-G, neurobiological group; SD-G, sociodynamic group; CO-G, control group n Correspondence to: Center for Addiction Research (CARe), Department of Psychiatry, Social Psychiatry and Psychotherapy, Hannover Medical School, CarlNeuberg-Straße 1, D-30625 Hanover, Germany. Tel.: þ 49 511 532 0. E-mail address: [email protected] (A. Heberlein). 0165-1781/$ - see front matter & 2014 Elsevier Ireland Ltd. All rights reserved. http://dx.doi.org/10.1016/j.psychres.2013.12.040
that condition (e.g., psychiatric diseases are the patients0 faults; (Angermeyer et al., 2011; Read and Harré, 2001; Weiner, 1988)). For example, Weiner (1988) reported that physical disabilities (e.g., blindness) were perceived as uncontrollable and therefore elicited pity rather than blame. Surveys have consistently shown a high acceptance rate (up to 60 percent) regarding whether AD is a self-inflicted condition but much lower rates have been found regarding whether other psychiatric diseases such as eating disorders (34 percent agreement), depression, panic attacks or, schizophrenia (4–13 percent agreement) are self-inflicted (Crisp et al., 2005; Crisp et al., 2000). Several studies (Crisafulli et al., 2008; Lincoln et al., 2008) have tested the central hypothesis of the attribution theory: advocating the neurobiological origins of psychiatric diseases (thereby stressing uncontrollability and guiltlessness) engenders attitudes toward psychiatric diseases in line with attitudes toward somatic diseases (Boyle et al., 2009). Although the results do not support the general reduction of stigmatization toward patients who suffer from psychiatric diseases, advocating for a biogenetic explanation model partially supports the reduction of moral censure hypothesis (by disrupting the association between self-infliction and psychiatric disease) for patients who suffer from psychiatric diseases such as schizophrenia (Lincoln et al., 2008) and anorexia nervosa (Crisafulli et al., 2008).
A. Heberlein et al. / Psychiatry Research 215 (2014) 766–770
Our study sought to determine the degree to which the attitudes toward ADPs are affected by the support of neurobiological and sociodynamic models of AD. In particular, we tested the effects of neurobiological versus sociodynamic explanations of AD with regard to the following dimensions: (1) Agreement with neurobiological origins (supporting an intervention effect). (2) Agreement with sociodynamic origins (supporting an intervention effect). (3) Evaluation of ADP characteristics. (4) Agreement with AD self-infliction. (5) Agreement with ADP heteronomy with regard to self-treatment. (6) Estimation of a poor therapeutic prognosis with regard to AD.
767
Randomization was achieved by dividing the questionnaires (NB-G, SD-G, and CO-G) across the seating order of the participants. The study supervisor (RS) collected the surveys after 10–15 min. 2.3. Structure of the questionnaire One of the authors (AH) translated the original set of questions into German, and a native English speaker back-translated the German version. This protocol was conducted until the back-translated German version was identical to the original English version. The original questionnaire (Crisafulli and colleagues) was supplemented by several additional questions regarding attitudes toward ADPs and the presumed origin of AD. In particular, the supplement explored whether participants believe that ADPs are capable of consenting to treatment (i.e., whether they are selfdetermined with regard to therapeutic decisions and assumptions). Statements were worded unambiguously e.g., “ADPs should autonomously decide their own treatment”; “Doctors should decide the treatment regimen to be used”; and “ADPs should be forced into treatment”. The final questionnaire consisted of 75 questions. All questions were answered using a six-point Likert scale. 2.4. Data analyses
2. Methods The current investigation was conducted in accordance with the Declaration of Helsinki and approved by the Ethics Committee of the Medical School of Hannover.
Data analyses were performed using SPSS 20. Missing data were replaced by multiple imputation (via the multiple regression method). The 75 questionnaire items comprised the following seven subscales (two causality scales and five stigma scales):
2.1. Participants A total of 444 people (312 women and 125 men) participated in this study. Participants were recruited at the Medical School of Hannover and the University of Hildesheim, Germany. Table 1 provides an overview of the study population. The survey was completed before the beginning of classes and during a monthly staff orientation at each university. All participants were informed of the study goals of the questionnaire. They were instructed to first read the portion regarding the development of AD and then answer all of the questions spontaneously and honestly. No additional information was provided.
2.2. Questionnaire The questionnaire, including the portion on the development of AD, was adapted from Crisafulli et al. (2008) who investigated the effect of disease models on the attitudes of nursing students toward patients with anorexia. We designed two information leaflets by adapting the structure of Crisafulli and colleagues0 original leaflet. The leaflets provided background information that suggested either neurobiological or sociodynamic causes of AD. The diagnostic criteria of AD according to the ICD-10 were at the top of the survey, followed by citations of studies regarding the neurobiological (or sociodynamic) causes of AD. We altered the original study by Crisafulli et al. (2008) by adding a control group (CO-G) who answered the questionnaire without information regarding the origins of AD. This protocol allowed us to evaluate the effect of the two leaflets on general attitudes toward ADPs. Participants were randomly assigned to one of three groups: (a) Participants assigned to the NB-G received information that explained AD as a neurobiological condition based on the biological characteristics of the central nervous system. (b) Participants assigned to the SD-G received information that explained AD as a consequence of sociodynamic factors such as lifestyle, drinking habits within one0 s family and peer group, the availability of alcohol, and acquired personality traits. (c) Participants assigned to the CO-G did not receive prior information regarding the origins of AD.
Table 1 Participant demographics by group: graduated participants were classified as “academics”. Groups were not significantly different regarding education, age and sex. Group
Control group Sociodynamic group Neurobiological group
Sex
Education level
Age
♀
♂
Academics
Non-academics
Mean/S.D.
78 98 111
36 27 52
30 31 52
76 93 105
26.21/6.82 25.78/6.46 26.41/6.59
Causality scales (1) Neurobiological causes of AD (Cronbach0 s α ¼ 0.879) (2) Social causes of AD (Cronbach0 s α ¼ 0.763) Stigma scales (3) Characteristics of ADPs (Cronbach0 s α ¼ 0.870) (4) Dangerousness of ADPs (Cronbach0 s α ¼ 0.784) (5) Self-infliction of AD (Cronbach0 s α ¼ 0.733) (6) Heteronomy regarding therapeutic decisions (Cronbach0 s (7) Poor prognosis of AD (0.825)
α ¼0.800)
The reliabilities of the extracted subscales were tested with Cronbach0 s α. Questions that reduced the α-level of the subscales were removed from the analysis (five questions on the characteristics scale and two questions regarding surgical approaches to AD). A confirmatory factor analysis (i.e., a principal component analysis with Promax rotation) was applied to verify the major topics of the questionnaire. An oblique factor analysis was used because it allows correlations (which are common in the social sciences) among the factors extracted and therefore provides more consistent and replicable results. To assure the consistency of the extracted factors, we applied the principal component and maximum likelihood methods to extract factors (Costello and Osborne, 2005). A group-to-group comparison was conducted using an analysis of covariance (ANCOVA). The seven subscales were set as dependent variables, and the three groups (SD-G, NB-G and CO-G) were the fixed factors. Age, sex and profession were considered covariates. Effect sizes were calculated using partial η2. Bonferroni0 s post hoc test was used to compare the three sub-groups. We applied a significance level of α ¼0.007 due to repeated testing (i.e., a correction factor of 7). Pearson0 s correlation analysis was applied in order to investigate associations between the two causality scales and the five stigma scales.
3. Results 3.1. Characteristics of the respondents and the questionnaire A total of 444 questionnaires were included in the final analysis. Table 1 shows the characteristics of the respondents. 3.2. Group-to-group analyses We found a significant group difference with regard to the biological origins of AD (F¼17.649, p o0.001, η2 ¼ 0.054): a significant increase in the acceptance of the neurobiological disease model was found among the NB-G compared with the CO-G (p o0.001) and the SD-G (po 0.001) groups. Moreover, the NB-G respondents rated the typical characteristics of ADPs to be significantly more positive (F¼8.471, po0.001, η2 ¼0.054) than the SD-G (po0.001) and the CO-G (po0.001) respondents.
768
A. Heberlein et al. / Psychiatry Research 215 (2014) 766–770
A significant increase in the acceptance of self-infliction of AD (F¼ 2.860, p ¼0.014, η2 ¼0.018) was observed among the SD-G compared with the NB-G (p ¼0.003). A significant group-to-group difference was found regarding the evaluation of ADPs0 competence to freely decide their treatment schedule (F¼4.341, p¼ 0.001, η2 ¼ 0.026): CO-G respondents showed more acceptance of the heteronomy regarding therapeutic decisions compared with the NB-G respondents (p ¼0.001). Though, a significant effect of age was also observed (p¼ 0.002, η2 ¼ 0.12), which even outweighed the group effect. Moreover, a trend toward a group difference was observed with regard to an AD prognosis (F ¼2.404, p ¼0.035, η2 ¼0.015): SD-G respondents rated AD prognoses to be significantly more positive than CO-G (p ¼ 0.003) and NB-G (p ¼ 0.002) respondents. Table 2 shows the results of the ANCOVA and its post-hoc tests.
reason for the lack of an effect of the sociodynamic information leaflet among the SD-G might be their high pre-test agreement with the sociodynamic foundation hypothesis of AD. Consistent with this explanation, we found a mean of 4.21 points (using a sixpoint Likert scale) on the sociodynamic scale compared with a mean of 2.94 on the neurobiological scale among the CO-G. Moreover, we found an even stronger agreement with the sociodynamic (mean 4.21 points) but not neurobiological (mean 3.40 points) explanations of AD among the NB-G. This observation supports the hypothesis that the sociodynamic explanation of AD is strongly supported in Germany, upon which stigma-related attitudes might be grounded.
4.2. Association between stigmatiferous attitudes and the sociodynamic explanation of AD
3.3. Correlation analysis The sociodynamic scale was significantly associated with all five STIGMA subscales that measured the various dimensions of stigma in this study (see Table 3 for details). No association was found between the neurobiological scale and these subscales.
We found that the sociodynamic subscale was strongly associated with the five stigma subscales. In particular, a strong acceptance of the sociodynamic causes of AD was negatively associated with the evaluation of positive characteristics and positively associated with the acceptance of heteronomy of ADPs regarding therapeutic decisions, the assumed dangerousness of ADPs and the acceptance of AD as a self-inflicted condition.
4. Discussion Table 3 Correlation of the Causality scales with the five stigma subscales.
4.1. Effect of intervention Reading the neurobiological information leaflet significantly increased agreement with the proposed neurobiological causes of AD among the NB-G compared with the other groups (see Table 2 for details). A similar effect was not observed among the SD-G, who did not show a significantly greater agreement with the sociodynamic causes of AD compared with the other groups. One
Sociodynamic subscale r ¼ 0.174, p o 0.001 r ¼0.223, p o 0.001 r ¼0.210, p o 0.001 r ¼0.129, p o0.001 r ¼ 0.139, p o 0.001
Characteristics subscale Self-infliction subscale Dangerousness subscale Heteronomy subscale Poor prognosis subscale
Table 2 Group-to-group differences regarding the seven subscales of the questionnaire.
Neurobiological origins
Sociodynamic origins
Positive characteristics
Dangerousness
Self-infliction
Acceptance of heteronomy regarding treatment decisions
Poor prognosis
CO-G
SD-G
NB-G
F¼ 17.649, po 0.001, η2 ¼0.054 SD-G NB-G 0.162 0.463 p ¼0.121 p o 0.001
CO-G 0.162 p ¼ 0.121
NB-G 0.625 p o 0.001
CO-G 0.463 po 0.001
SD-G 0.625 p o 0.001
F¼ 0.955, p ¼0.445, η2 ¼ 0.006 SD-G NB-G 0.029 0.102 p ¼0.581 p ¼0.039
CO-G 0.029 p ¼ 0.581
NB-G 0.073 p ¼ 0.130
CO-G 0.102 p¼ 0.039
SD-G 0.073 p ¼0.130
F¼ 8.471, po 0.001, η2 ¼0.054 SD-G NB-G 0.151 0.070 p ¼0.004 p ¼0.159
CO-G 0.151 p ¼ 0.004
NB-G 0.220 p o 0.001
CO-G 0.070 p¼ 0.159
SD-G 0.220 p o 0.001
F¼ 2.479, p ¼0.084, η2 ¼0.006 SD-G NB-G 0.032 0.128 p ¼0.701 p ¼0.125
CO-G 0.032 p ¼ 0.701
NB-G 0.159 p ¼ 0.038
CO-G 0.128 p¼ 0.105
SD-G 0.159 p ¼0.038
F¼ 2.860, p ¼0.014, η2 ¼0.018 SD-G NB-G 0.122 0.063 p ¼0.072 p ¼0.327
CO-G 0.122 p ¼ 0.072
NB-G 0.185 p ¼ 0.003
CO-G 0.063 p¼ 0.327
SD-G 0.185 p ¼0.003
F¼ 4.341, p ¼ 0.001, η2 ¼ 0.026 SD-G NB-G 0.183 0.306 p ¼0.060 p ¼0.001
CO-G 0.183 p ¼ 0.060
NB-G 0.123 p ¼ 0.175
CO-G 0.306 p¼ 0.001
SD-G 0.123 p ¼0175
F¼ 2.404, p ¼ 0.035, η2 ¼0.015 SD-G NB-G 0.234 0.002 p ¼0.006 p ¼0.975
CO-G 0.234 p ¼ 0.006
NB-G 0.231 p ¼ 0.003
CO-G 0.002 p¼ 0.975
SD-G 0.231 p ¼0.003
A. Heberlein et al. / Psychiatry Research 215 (2014) 766–770
The association between the acceptance of the social causes of AD and AD-related stigma was mirrored by the group-to-group differences among the three intervention groups: the NB-G (p o0.001) and CO-G (p ¼0.004) rated the typical characteristics of ADPs as significantly more positive than the SD-G, whereas the NB-G showed significantly less acceptance of AD as a self-inflicted disease (p ¼0.003) and with the putatively poor prognosis (p ¼0.003) of AD compared with the SD-G (po 0.001). 4.3. The stigma of AD compared with the stigma of further psychiatric diseases
769
distinguishes the stigma of AD from that of other psychiatric diseases. Alcohol consumption is common in society, which supports the teleological usefulness of stigmatizing consumption beyond social boundaries (Phelan et al., 2008). Compared with other psychiatric conditions that are commonly accepted as diseases, alcoholism is often not associated with a disease. This common belief in alcohol consumption to be a controllable condition rather than an un-controllable disease might explain the close relationship between the sociodynamic disease model and the different aspects of AD-related stigma found in our data. 4.4. Limitations
As recent research (Pescosolido et al., 2010) has shown, the increased acceptance of the neurobiological disease model of psychiatric diseases such as schizophrenia (91 percent in 2006) and depression (72 percent in 2006) is associated with increased desires for social distance (between 72 and 74 percent of respondents) as well as increased perceptions of violent patients (increased from 78 to 79 percent) and detrimental characteristics (increase from 49 to 65 percent). Therefore, the lack of effects (Angermeyer et al., 2011, 2013; Pescosolido et al., 2010; Rüsch et al., 2010) or even the negative effects (Angermeyer et al., 2011, 2013) of the neurobiological disease model regarding the stigma of psychiatric disease have been reported. Unlike other psychiatric diseases, the results of this study indicate a close association between the acceptance of the sociodynamic disease model of AD and the various dimensions of AD-related stigma investigated in the questionnaire: supporting the central hypothesis of attribution theory, we found a strong negative association between the estimation of a putatively poor prognosis of AD and the acceptance of sociodynamic causes of AD. The supposed controllability of AD, which is mirrored by the assumed possibility of a successful prognosis, might engender stigma-related attitudes with regard to the typical characteristics of ADPs and their potential danger. This hypothesis is consistent with the teleological interpretation of AD-related stigma originally proposed by Phelan et al. (2008): in her interpretation AD-related stigma is understood to secure important social norms by clarifying the boundaries of acceptable social behavior and demonstrating the consequences of nonconformance (Schomerus et al., 2011). Therefore, believing that AD is an uncontrollable condition by presenting a sole neurobiological disease model might have loosened the connotation between moral failings and AD in the NB-G while strengthening this connotation in the SD-G. Consistently, we observed a significant group to group difference regarding the stigmatiferous attitudes of the SD-G and the NB-G, which was not seen between the NB-G and the CO-G. Our results mirror results presented by Schomerus et al. (2013): they reported that the acceptance of a biogenetic explanation of AD was not associated with significant effects regarding the social acceptance of ADPs, although it decreased respondents0 acceptance of onset responsibility of AD. Consistently, the neurobiological information leaflet presented in our study did not significantly impact respondents0 attitudes regarding the five stigma subscales when compared to the CO-G. In conclusion, our results support the central hypothesis of attribution theory: the agreement with the sociodynamic causal explanations of AD was closely associated with all dimensions of stigma investigated in this study. However, the neurobiological disease model did not reduce the stigma-related attitudes of the NB-G respondents as was demonstrated by comparison of the attitudes of the NB-G and the CO-G, whereas the sociodynamic disease model vice versa worsened respondents0 attitudes regarding the acceptance of putative detrimental characteristics of ADP. The shared knowledge of ADPs might explain these results: All participants in this study were familiar with alcohol consumption and ADP. The experience of respondents might be a key point that
This study has several limitations: first, our study was restricted to the Hannover region of Germany. As discussed before, cultural differences exist with regard to the stigmatization of AD (Schomerus et al., 2011); thus, our sample is not representative of ADP stigmatization in general or in Germany specifically. Moreover, our sample consisted of students and stuff from the Medical School of Hannover and might therefore differ in their attitudes toward ADPs compared with the general public. However, similar results were found in a recent nationwide German sample (Schomerus et al., 2013) supporting the hypothesis that stigma related to AD differs from stigma related to further psychiatric diseases. Furthermore, this study is limited by its use of a non-validated questionnaire. Although we found that the subscales showed high reliabilities, our results should be confirmed by future studies with validated questionnaires. 4.5. Conclusion In summary, this study and earlier results illustrate that advocating the neurobiological origins is not a successful method of decreasing the stigma related to AD due to its lack of effect and because it does not provide accurate information regarding the currently known causes of AD. References Angermeyer, M.C., Holzinger, A., Carta, M.G., Schomerus, G., 2011. Biogenetic explanations and public acceptance of mental illness: systematic review of population studies. British Journal of Psychiatry 199, 367–372. Angermeyer, M.C., Mnich, E., Daubmann, A., Herich, L., Wegscheider, K., Kofahl, C., von dem Knesebeck, O., 2013. Biogenetic explanations and public acceptance of people with eating disorders. Social Psychiatry and Psychiatric Epidemiology 48 (10), 1667–1673, http://dx.doi.org/10.1007/s00127-012-0648-9. Boyle, M.P., Blood, G.W., Blood, I.M., 2009. Effects of perceived causality on perceptions of persons who stutter. Journal of Fluency Disorders 34, 201–218. Costello, A.B., Osborne, J.W., 2005. Best practices in explanatory factor analysis: four recommondations for getting the most from your analysis. Practical Assessment, Research and Evaluation 10, 1–9. Crisafulli, M.A., Von Holle, A., Bulik, C.M., 2008. Attitudes towards anorexia nervosa: the impact of framing on blame and stigma. International Journal of Eating Disorders 41, 333–339. Crisp, A., Gelder, M., Goddard, E., Meltzer, H., 2005. Stigmatization of people with mental illnesses: a follow-up study within the Changing Minds campaign of the Royal College of Psychiatrists. World Psychiatry: Official Journal of the World Psychiatric Association 4, 106–113. Crisp, A.H., Gelder, M.G., Rix, S., Meltzer, H.I., Rowlands, O.J., 2000. Stigmatisation of people with mental illnesses. The British Journal of Psychiatry: the Journal of Mental Science 177, 4–7. Lincoln, T.M., Arens, E., Berger, C., Rief, W., 2008. Can antistigma campaigns be improved? A test of the impact of biogenetic vs psychosocial causal explanations on implicit and explicit attitudes to schizophrenia. Schizophrenia Bulletin 34, 984–994. Pescosolido, B.A., Martin, J.K., Long, J.S., Medina, T.R., Phelan, J.C., Link, B.G., 2010. “A disease like any other”? A decade of change in public reactions to schizophrenia, depression, and alcohol dependence. The American Journal of Psychiatry 167, 1321–1330. Phelan, J.C., Link, B.G., Dovidio, J.F., 2008. Stigma and prejudice: one animal or two? Social Science and Medicine 67, 358–367. Pickard, H., 2011. The instrumental rationality of addiction. Behavioral and Brain Sciences 34, 320–321.
770
A. Heberlein et al. / Psychiatry Research 215 (2014) 766–770
Read, J., Harré, N., 2001. The role of biological and genetics beliefs in the stigmatisation of “mental patients”. Journal of Mental Health 10, 223. Rüsch, N., Todd, A.R., Bodenhausen, G.V., Corrigan, P.W., 2010. Biogenetic models of psychopathology, implicit guilt, and mental illness stigma. Psychiatry Research 179, 328–332. Schomerus, G., Lucht, M., Holzinger, A., Matschinger, H., Carta, M.G., Angermeyer, M.C., 2011. The stigma of alcohol dependence compared with other mental disorders: a review of population studies. Alcohol and Alcoholism 46, 105–112.
Schomerus, G., Matschinger, H., Angermeyer, M.C., 2013. Causal beliefs of the public and social acceptance of persons with mental illness: a comparative analysis of schizophrenia, depression and alcohol dependence. Psychological Medicine, 1–12 Weiner, B., 1988. Attribution theory and attributional therapy: some theoretical observations and suggestions. British Journal of Clinical Psychology 27 (Pt 1), 99–104.