Journal of Affective Disorders 174 (2015) 150–156
Contents lists available at ScienceDirect
Journal of Affective Disorders journal homepage: www.elsevier.com/locate/jad
Research report
Beliefs of people taking antidepressants about the causes of their own depression John Read a,n, Claire Cartwright b, Kerry Gibson b, Christopher Shiels a, Lorenza Magliano c a
Institute of Psychology, Health and Society, University of Liverpool, Liverpool L69 3GB, UK School of Psychology, University of Auckland, New Zealand c Department of Psychology, Second University of Naples, Italy b
art ic l e i nf o
a b s t r a c t
Article history: Received 14 August 2014 Received in revised form 4 October 2014 Accepted 6 November 2014 Available online 28 November 2014
Background: The beliefs of people receiving treatment about the causes of their own mental health problems are researched less often than the causal beliefs of the public, but have important implications for relationships with prescribers, treatment choices and recovery. Method: An online survey on a range of beliefs about depression, and experiences with antidepressants, was completed by 1829 New Zealand adults prescribed anti-depressants in the preceding five years, 97.4% of whom proceeded to take antidepressants. Results: Six of 17 beliefs about the causes of their own depression were endorsed by more than half the sample: chemical imbalance, family stress, work stress, heredity, relationship problems and distressing events in childhood. There were some marked differences in content, structure and level of conviction of beliefs about one's own depression and the sample's previously published beliefs about depression in general. There were also significant differences between the beliefs of demographic groupings. Regression analyses revealed that self-reported effectiveness of the antidepressants was positively associated with bio-genetic causal beliefs. The quality of the relationship with the prescribing doctor was positively related to a belief in chemical imbalance as a cause and negatively related to a belief in unemployment as a cause. Limitations: The convenience sample may have been biased towards a favourable view of bio-genetic explanations, since 83% reported that the medication reduced their depression. Conclusions: People experiencing depression hold complex, multifactorial and idiosyncratic sets of beliefs about the causes of their own depression, apparently based at least in part on their own life experiences and circumstances. Exploring those beliefs may enhance the doctor–patient relationship and selection of appropriate treatment modality. & 2014 Elsevier B.V. All rights reserved.
Keywords: Depression Causal beliefs Antidepressants Stigma Prognostic pessimism
1. Introduction In sooth, I know not why I am so sad: It wearies me; you say it wearies you; But how I caught it, found it, or came by it, What stuff 'tis made of, whereof it is born, I am to learn. (Shakespeare, 1597) In New Zealand the number of people prescribed antidepressants [ADs] increased by 35% from 304,530 to 412,631 between 2006/07 and 2011/12 (PHARMAC, personal communication, 2012), in a population of 4.4 million, of whom 3.7 million are aged 16 or older. Thus one in nine of the adult population (and approximately one in six women) receive ADs every year. In the USAADs had n
Corresponding author. þ 44 7923 892 140. E-mail address: [email protected] (J. Read).
http://dx.doi.org/10.1016/j.jad.2014.11.009 0165-0327/& 2014 Elsevier B.V. All rights reserved.
become the most frequently prescribed drug type by 2005, with one in ten people over the age of six being prescribed ADs annually (Olfson and Marcus, 2009). In England, prescriptions for ADs increased 10% per year between 1998 and 2010 (Ilyas and Moncrieff, 2012). In 2013 there were 53 million prescriptions, in a country with a total population of 52.6 million (Health and Social Care Information Centre, 2012, 2014). These increases occur despite evidence that their efficacy was initially overestimated (Kirsch et al., 2008; Pigott et al., 2010) and their adverse effects minimised. For example, eight adverse effects were reported by over 50% of the large sample of AD recipients used in the current study, including sexual difficulties, emotional numbing, drowsiness, weight gain and withdrawal effects (Read et al., 2014a). Concerns about over-prescribing are often framed within a broader critique about the ‘medicalisation’ of distress (Dowrick and Frances, 2013; Ilyas and Moncrieff, 2012; Read and Dillon, 2013). Therefore the opinions and beliefs of the millions of people who take ADs are important. A recent review of AD trials found,
J. Read et al. / Journal of Affective Disorders 174 (2015) 150–156
however, that none reported patients' illness beliefs (Hughes et al., 2011). A review of the small body of qualitative research that has sought the views of AD recipients concluded that patients have ambivalent, but predominantly negative, views of ADs and that they tend to prefer psychotherapy (Gibson et al., 2014). One set of beliefs that has been studied extensively, is beliefs about the causes of depression, largely because causal beliefs are considered central to stigma. Contrary to the hopes of the designers of destigmatisation programmes based on the ‘mental illness is an illness like any other’ notion, which promote biogenetic causal explanations, such illness-type beliefs have been consistently shown to increase fear and prejudice and decrease expectation of recovery (Angermeyer et al., 2011; Kvaale et al., 2013; Read et al., 2006, 2013; Schomerus et al., 2014). This body of research also tells us what the public actually believes about the causes of depression. Despite decades of wellresourced programmes promoting a bio-medical perspective (often funded by drug companies), the public, internationally, continues to prefer psycho-social explanations (Angermeyer et al., 2013; Jorm et al., 2005; Munizza et al., 2013; Pilkington et al., 2013; Schomerus et al., 2014). For example, the German public endorses ‘family/ partnership problems’ (81%) and ‘work stress’ (80%) more than ‘brain disease’ (38%) or ‘heredity’ (40%) as a cause of ‘major depression’ (Angermeyer et al., 2013). In Australia, although belief in ‘inherited or genetic’ increased from 49% in 1995 to 65% in 2011; the most frequently endorsed causes remained ‘day to day problems’ (97%), ‘death of close friend/relative’ (97%) and ‘problems in childhood’ (96%) (Pilkington et al., 2013). Findings about the causal beliefs of depressed people (about depression in general, rather than about their own depression), confirm that, like the general population, they prefer psychological and social explanations to biological or medical ones. An English study of generic GP patients found that those with experience of depression were more likely to endorse ‘medical’ causes of depression (‘hormones’ and ‘chemical imbalances in the brain’) than those who had never experienced depression. Nevertheless, both the depressed and non-depressed groups endorsed ‘psychological’ causes (stress, life events, unhappy childhood) slightly more strongly than medical causes (Ogden et al., 1999). A recent German study found that depressed people are even more likely than other members of the public to endorse ‘psychosocial stress’ items (job stress, family strain, and critical life event) and ‘conditions of socialisation’ (broken home, lack of parental affection and sexual abuse in childhood) (Mnich et al., 2014). A 2008 review of studies of the beliefs of depressed people about the causes of their own depression found that most are consistent with the general public's beliefs (Prins et al., 2008). A study of 340 people with ‘major depression’ found that ‘psychological’ items were far more strongly endorsed than ‘biological items’ as causes of their own depression, with by far the most frequently endorsed being ‘stress and negative life experiences’ (Bann et al., 2004). A Swedish study, of 303 depressed primary care patients, found that 69% endorsed ‘current life stressors’, 32% ‘past life events’ and 4% ‘biological/heredity’ (Hansson et al., 2010). The causal belief most frequently endorsed by 102 depressed Canadian outpatients was ‘stressful and negative life experiences’ (prompting the researchers to call for ‘proper psychoeducation’) (Srinivasan et al., 2003). So, experiencing depression does not seem to reduce the public's psycho-social understanding of depression. It remains possible, however, that interacting with medical professionals and receiving a treatment explicitly based on a bio-medical causal model, might affect one's causal beliefs. This would be important because these beliefs have been shown to be related to the relationship with the prescribing doctor, including the potential of that relationship to shape ideas about depression and about
151
appropriate treatments (Karp, 1993; Schofield et al., 2011; Turner et al., 2008). Little research has been undertaken targeting the causal beliefs of people taking ADs. In a review of studies of the causal beliefs of ‘affected individuals’ (Lebowitz, 2014) only one reports the causal beliefs of AD recipients. This study, of 191 primary care patients in the USA. found that ‘stress’ (81%) was the only cause endorsed by more than half the sample as a cause of their own depression (Brown et al., 2007). Subsequently, the causal beliefs, about depression in general, of 1829 AD recipients in New Zealanders (the same sample as the current study) has been reported. The most commonly endorsed causes were: family stress (90.8%), relationship problems (89.9%), loss of loved one (87.5%), financial problems (86.9%), isolation (86.3%), abuse or neglect in childhood (85.4%), and chemical imbalance (84.8%) (Read et al., 2014b). The negative effects of bio-genetic beliefs found in studies of the public are replicated in people who are depressed or taking ADs, particularly in terms of pessimism about recovery (Lebowitz, 2014; Lam and Salkovskis, 2007). In the large New Zealand sample of AD recipients, bio-genetic causal beliefs were significantly correlated with the belief that ‘people can't get better by themselves even if they try’ (Read et al., 2014b). No study, to the best of our knowledge, has compared the causal beliefs of depressed people, or AD recipients, about depression in general with the causal beliefs of the same sample about their own depression. Therefore, the three aims of the current paper were: 1. To report the beliefs of a large sample of AD recipients about the causes of their own depression. 2. To analyse those beliefs in terms of demographics and in relation to prognostic pessimism, perceived efficacy of the ADs, quality of relationship with the prescriber, length of AD use and beliefs about placebo effects. 3. To compare the content and factorial structure of those beliefs with the previously published beliefs of the same sample about the causes of depression in general.
2. Method 2.1. Instrument and procedure The study used an online questionnaire with 47 questions covering: the prescribing process; perceptions of the effectiveness of the ADs; side-effects; benefits; and beliefs about the causes of depression. Responses to the question ‘There are many theories, and lots of debate, about what causes problems like depression. Please indicate the extent to which you agree or disagree that the following factors are causes of depression IN GENERAL’ have been published (Read et al., 2014b). This question was immediately followed by the key question for the current paper: Please indicate the extent to which you agree or disagree that the following factors are causes of YOUR OWN problems’. Both questions were followed by a list of 17 items, based on the research literature and including bio-genetic (e.g. Chemical imbalance, and Heredity (genes)) and psycho-social causes (e.g. Family stress and Isolation) (see Table 1 for complete list), and five point Likert scales from ‘strongly disagree’ to ‘strongly agree’. Participants were also invited to identify ‘other’ causes, of both depression in general and their own depression. Criteria for participation included having been prescribed ADs in the last five years and being at least 18 years old. A webpage (which was not linked to other websites) provided the participant information for the study and a link to the online questionnaire. The study
152
J. Read et al. / Journal of Affective Disorders 174 (2015) 150–156
Table 1 Percentage agreement, mean scores, and demographic relationships for 17 causal beliefs.
Chemical imbalance Family stress Work stress Heredity (genes) Relationship problems Other unhappy or distressing childhood experiences Isolation Financial problems Hectic pace of modern life Abuse or neglect in childhood Loss of loved one Unconscious conflict Disorder of the brain Unstable personality Unemployment Drug or alcohol abuse God's will or fate
Agreea (%)
Meanb (s.d.)
66.6 65.6 60.4 60.3 56.2 50.5 48.7 43.5 40.4 38.2 35.9 32.9 28.8 23.5 18.6 17.7 2.1
3.78 3.70 3.53 3.50 3.41 3.18 3.18 3.02 2.99 2.80 2.77 2.85 2.72 2.45 2.07 2.02 1.36
(1.25) (1.30) (1.33) (1.32) (1.40) (1.51) (1.42) (1.40) (1.32) (1.59) (1.54) (1.31) (1.35) (1.30) (1.35) (1.33) (0.78)
Gender
Age
Female nn Malennn Female n
Old
Income
nn
Young nn Old nnn
Female
Male Male Male
n
n nnn
Old
nnn
Low Low Low
n
Low Low
nnn
nnn n
nnn
Young nn Young nn Old nnn
nnn
High
nnn
Rank for beliefs about depression in generalc 7 1 9 12 2 8 5 4 15 6 3 16 13 14 10 11 17
a
‘Agree or ‘strongly agree’. 1¼ Strongly disagree and 5¼ Strongly agree c Same sample: (Read et al., 2014a, 2014b). n po 0.01. nn p o0.001. nnn p o0.0001. b
was further publicised in the New Zealand media via media releases, interviews with the researchers and advertisements. The study was approved by the University of Auckland Human Participants Ethics Committee. 2.2. Participants Of the 2171 people who started the survey, 295 stopped before question 20 and their responses were not analysed. Of the remaining 1876, 45 cited medications other than ADs. Two respondents whose scores were nearly identical to those of someone else with the same Internet Protocol address (indicating probable use of the same computer) were excluded. This left 1829 for analysis (97.4% of whom had taken the ADs when prescribed them). The number of responses to each question varied as not all participants responded to all questions. The sample has been described in detail elsewhere (Read et al., 2014a). Females constituted 76.6% of the sample. The modal age group was 36–45 (24.2%); 92.1% identified as ‘New Zealand/ European’. In terms of education, 49.6% had a university degree, 26.1% gained a diploma or certificate after high school, 17.2% completed high school, and 7.1% did not complete high school. Annual income (in NZ dollars) ranged from less than $10,000 (15.0%) to more than $100,000 (7.7%). The modal income was $40,000–$59,999 (22.1%). (The median income of the NZ population in 2012 was $29,000 (Statistics New Zealand, 2012)). 2.3. Data analysis Associations between the causal beliefs and demographic variables were tested using t-tests (two-tailed) and spearman rank correlation coefficients. Age was recorded as 18–25, 26–35, 46–55, 56–65, 66–75, 76–85 or 86 þ. Education categories were: ‘Did not complete High School’, ‘Completed High School’, ‘Diploma/Certificate after High School’ ‘Undergraduate Degree’ and ‘Postgraduate Degree.’ For the main analysis, seven survey items were transformed into dichotomous (Yes/No) outcome variables: perceived reduction in depression during antidepressant use; perceived improvement in quality of life; reported use of antidepressants for three years or more; ‘good’ relationship with prescribing doctor; felt that the doctor
understood their problem; belief that the effect of antidepressant was 450% placebo; and 50% agreement with the statement, “People can’t get better by themselves, even if they try”. Univariate analysis, using 2 2 cross-tabulations and a chisquare test for statistical significance, was conducted in order to investigate associations between the seventeen causal belief items and the seven binary outcomes. Beliefs found to be significantly associated (p o0.05) with outcome were entered into subsequent multivariate models. Seven logistic regression models were constructed and run in order to estimate the independent effect of beliefs about causes of own depression on each of the seven outcomes. Only the causal belief items that attained significance (po0.05) in the respective univariate analysis were entered as covariates in each model. For each independent variable, the Odds Ratio, 99% Confidence Interval and p-value were generated. In all logistic regression models the estimates of the effect of causal beliefs on outcome were adjusted for respondent gender, age, annual income and whether religious or not. A more stringent criterion of significance (po0.01) was used in interpretation of effects in the regression analysis. In order to allow a comparison with the data structure in our previous study looking at general beliefs about depression, using the same sample and methodology (Read et al., 2014a, 2014b), responses to the 17 causal belief questions were entered into an explorative principal-component (varimax rotation) factor analysis. Factors were identified using an eigenvalue of Z1 as criterion. The relationship between the two sets of causal beliefs (about own depression and depression in general) was assessed using a Spearman rank correlation coefficient.
3. Results 3.1. Beliefs of AD recipients about the causes of their own depression 3.1.1. Responses to questionnaire items Table 1 summarises responses to the 17 causal beliefs presented to participants in the questionnaire. Six items were endorsed (‘agree’ or ‘strongly agree’) by more than half of all respondents as causes of their own depression: Chemical imbalance (67%), Family stress (66%),
J. Read et al. / Journal of Affective Disorders 174 (2015) 150–156
Work stress (60%), Heredity (60%), Relationship problems (56%) and Other unhappy or distressing childhood experiences (51%). 3.1.2. ‘Other’ causal beliefs The invitation to identify ‘other’ causes elicited one or more from 324 respondents. Those reported by more than five people are listed in Table 2. The most common type involved a range of medical conditions (47), ten of which involved ongoing pain and six of which were identified as head/brain injuries. Twenty three participants identified difficulties either with giving birth or caring for a new born baby. For example: Birth of a child, adjustment to motherhood, poor family support after having children. Lack of sleep; feeling of isolation/change in identity with a screaming newborn. Six of these 23 described their difficulties as ‘post natal depression’. Psychological difficulties were identified by 22 respondents, including low self-esteem, negative thinking, perfectionism, oversensitivity and a lack of ways to cope with either emotions or stressors. For example: Not knowing how to deal with stressful or confusing situations. Over-sensitivity and inability to deal with stress and anxiety, and a mindset that creates unnecessary anxieties in the first place. Low self-esteem, unhelpful ideas. Overwhelming emotions – over sensitive. Anxiety caused by a bunch of really self-depricating core beliefs. Seventeen identified specific mental health problems as a cause of their depression, most commonly ‘post natal depression’ (6) and ‘PTSD’ (5). Violence in adulthood was reported as a cause by 15 people, mostly within intimate relationships. For example: Domestic violence …….. Physical, emotional and mental abuse over sixteen years and continuing after separation through alienating the children from me. Prolonged abusive relationship with husband. Fifteen either reported ‘trauma’ (10) or ’PTSD’ (5) without specifying the nature of the trauma. Eleven identified a range of adverse experiences in childhood, for example: Depressed mother during infancy and childhood. Sustained bullying in childhood. Sexual abuse as a child. Table 2 ‘Other’ causal beliefs. n Medical conditions Childbirth/newborn baby/‘post natal’ Psychological difficulties Violence in adulthood Trauma/PTSD Negative childhood experiences Study stress Medication side effects Societal injustice Anxiety Others' illness/mental health problem Earthquakes Parenting stress No direction/hope Immigration stress
47 23 22 15 15 11 9 7 6 6 6 5 5 5 5
153
Six wrote about societal injustice as a cause of their depression: There is an inextricable relationship between inequality and injustice and my personal views of myself. The environment affects me. The hopelessness of modern capitalism and the state of our society in which social injustice, poverty and cruelty are both tolerated and accepted by most people. Frustration at greater problems, “weight of the world” depression. Having the perception to consider huge issues but feeling insignificant and ineffectual when it comes to solving them. Other responses were not easily categorised but seem worthy of citing, to demonstrate the diversity of views: Not one thing but a gradual build up of many things. I don’t think I was ‘Depressed’ I think I was struggling to cope with my circumstances and I needed to be taught coping methods…not labelled and put onto pills. I don't know about hereditary, but I feel I was taught depression from my mother and older sister, by not being shown how to deal with my emotions and anger properly. Neighbours excessive noise. I have these issues, which make me depressed BUT feel that a lot of these issues are caused by the way I am after 9yrs of medication. Anti depressant drugs magnify the symptoms. Side effects from medication, loss of health. sense of injustice, Lack of support, funding, Poverty, Abuse and neglect of Mental health system, and domestic abuse. In my case, I have no reason for depression, I was born with it and was depressed as a child. I am not at all sure what caused my depression, so feel I can't really answer the above. Sorry. We all face problems…its called ‘life’. 3.2. Relationships of causal beliefs with other variables 3.2.1. Demographics Table 1 shows that seven causal beliefs were significantly related to age, with by far the largest correlation being between Loss of loved one and older age (r ¼0.17, p o0.0001). There were significant gender differences on the mean scores of seven of the 17 items (with 5 representing ‘strongly agree’ and 1 representing ‘strongly disagree’). The largest differences were that men more strongly endorsed Work stress (mean 3.81, 95% CI 3.69–3.93 vs mean 3.45, 95% CI 3.36–3.52 ), Unemployment (mean 2.42, 95% CI 2.26–2.58 vs mean 1.95, 95% CI 1.86–2.02) and Drug or alcohol abuse (mean 2.33, 95% CI 2.16–2.46 vs mean 1.92, 95% CI 1.84–2.00); while the women more strongly endorsed Family stress (mean 3.76, 95% CI 3.68–3.84 vs mean 3.50, 95% CI 3.36–3.64). The strongest relationships with income were a positive correlation with Work stress (r ¼0.16, p o0.0001), and a negative correlation with Isolation (r ¼0.19, p o0.0001). There were no significant relationships with level of education. The only differences in means in terms of religious belief were that those who described themselves as religious were significantly more likely to endorse God's will or fate (mean 1.57, 95% CI 1.47–1.67 vs mean 1.27, 95% CI 1.23–1.31) and Abuse or neglect in childhood (mean 2.96, 95% CI 2.82–3.11 vs mean 2.71, 95% CI 2.62–2.81). Although Chemical imbalance was the most frequently endorsed belief in the sample as a whole, some demographic groups most strongly endorsed psycho-social beliefs. Family stress was the most endorsed belief by people aged 36–45 and 66–75, and by people with an income less than NZ$40,000. Work stress was the most endorsed cause by men, by people aged 46–55, and by people with an income of between $NZ80,000 and 100,000. (Although the number of Maori participants (45) was too small to include in analyses, we note that
154
J. Read et al. / Journal of Affective Disorders 174 (2015) 150–156
Maori endorsed Family stress, Relationship problems and Work stress more frequently than Chemical imbalance). 3.2.2. Perceived efficacy of antidepressants Those answering ‘yes’ to ‘Did the antidepressants reduce your depression?’ (82.8%) were significantly more likely to believe that Heredity and Chemical imbalance were causes of their depression, and significantly less likely to endorse Abuse or neglect in childhood, Isolation, Financial problems and Unstable personality (see Table 3). Among those who did not think the ADs reduced their depression, the six most endorsed causes were all psycho-social; most frequently Family stress (69.1%), Relationship problems (62.4%) and Isolation (61.4%), with Chemical imbalance (46.9%) ranked seventh. In a regression analysis (controlling for demographics and the nine other causal beliefs related (po0.05) to reduced depression in the univariate analysis) only Heredity and Chemical imbalance remained significantly related to reduced depression. Results for the other measure of perceived efficacy, reported improvement in Quality of Life [QoL], were almost identical (see Table 3). 3.2.3. Interaction with prescribing doctor Participants were asked to respond, on a five-point Likert scale, to the question ‘How would you describe your relationship with your Table 3 Unadjusted and adjusted associations between causal beliefs and key outcomes. Variable Causal beliefs
Univariate analysis (%)
Reduced depression Heredity Chemical imbalance Abuse or neglect in childhood Isolation Unstable personality Financial problems Improved quality of life Chemical imbalance Heredity Abuse or neglect in childhood Isolation Unstable personality Distressing childhood experience Taken ADs for more than 3 years Heredity Brain disorder Chemical imbalance Distressing childhood experience Abuse or neglect in childhood Drug or alcohol abuse Good relationship with doctor Unemployment Distressing childhood experience Abuse or neglect in childhood Isolation Doctor understood problems Chemical imbalance Distressing childhood experience Abuse or neglect in childhood Isolation Drug or alcohol abuse Unemployment Belief that effect is 450% placebo Chemical imbalance Distressing childhood experience Heredity
Yes, No 88.7, 76.6nnn 89.1, 75.5nnn 78.9, 86.3nnn 79.3, 87.5nnn 77.5, 85.4nn 80.6, 86.1nn Yes, No 91.3, 76.6nnn 90.2, 79.3nnn 81.2, 88.4nnn 81.8, 89.4nnn 80.7, 87.4nn 83.4, 88.6nn Yes, No 58.7, 42.3nnn 64.0, 47.0nnn 56.9, 42.2nnn 56.8, 47.9nnn 57.6, 48.6nn 60.5, 49.6nn Yes, No 69.4, 77.4nn 72.5, 80.5nnn 71.5, 79.0nn 72.7, 77.9nn Yes, No 66.3, 53.9nnn 56.0, 67.4nnn 55.0, 65.6nnn 55.4, 67.6nnn 54.2, 63.6nn 54.4 63.4nn Yes, No 12.8, 29.9nnn 21.2, 16.1n 15.3, 23.9nnn
99% CI of OR Adjusted Odds ratioa
2.04 2.03 0.73 0.79 0.64 0.81
1.28–3.04 1.27–2.99 0.44–1.02 0.47–1.12 0.41–1.08 0.54–1.24
2.65 1.76 0.71 0.71 0.66 0.82
1.62–4.03 1.18–2.95 0.37–1.06 0.45–1.12 0.40–1.10 0.49–1.42
1.66 1.53 1.30 1.20 1.15 1.39
1.21–2.31 1.07–2.23 0.93–1.85 0.86–1.73 0.78–1.62 0.93–2.09
0.65 0.78 0.74 0.99
0.39–0.94 0.52–1.15 0.50–1.12 0.67–1.36
1.75 0.72 0.78 0.80 0.76 0.80
1.29–2.40 0.51–1.03 0.53–1.11 0.56–1.04 0.49–1.12 0.52–1.17
0.37 1.59 0.74
0.23–0.52 1.06–2.35 0.48–1.10
Multivariate: Significant (P o0.01) effects in bold. Univariate. a Adjusted for other belief covariates that were significantly related (p o 0.05) at the univariate level, and for age, gender, income and whether religious. n po 0.05. nn p o0.01. nnn p o0.001.
doctor?’. Those who answered ‘good’ or ‘very good’ (75.4% of the whole sample) were significantly less likely to endorse four psycho-social beliefs: Unemployment, Abuse or neglect in childhood, Other unhappy or distressing childhood experiences and Isolation. Only Unemployment remained significant in the regression analysis (OR ¼0.65; po 0.01) When asked ‘How well do you think your doctor understood your problems?’ 60.1% of the responded ‘quite a lot’ or ‘a lot’. Participants who believed in five of the psycho-social causes were significantly less likely to feel understood (see Table 3). However, the only relationship remaining significant in the regression analysis was the positive relationship with Chemical imbalance (OR¼ 1.75, p o0.001). 3.2.4. Duration of use of antidepressants Six causal beliefs (see Table 3) were significantly related to whether a participant had taken ADs for more than three years (51.7% of the total sample), but only the Odds Ratios for Heredity (1.66; po 0.001) and Brain Disorder (1.53; p o0.001) remained significantly related in the regression analysis. 3.2.5. Prognostic pessimism 28.9% of participants agreed with the statement ‘People can't get better by themselves even if they try’. Only Chemical imbalance was related to respondents’ agreement with this statement in the regression analysis (OR ¼1.30; p¼ o0.05) but this did not reach the level of significance set for this study. 3.3. Comparisons with the beliefs of the same sample about causes of depression in general 3.3.1. Beliefs The rank order of participants’ beliefs about the causes of their own depression was correlated (r¼0.57) with that of their beliefs about the causes of depression in general (Read et al., 2014a, 2014b) (see Table 1), but only at the po0.05 level of significance. The two types of beliefs correlated for each of the 17 items at the o0.0001 level, but the strength of the correlations varied from 0.10 (Drug and alcohol abuse) and 0.23 (Loss of loved one) to, at the other extreme, 0.61 (Chemical imbalance), 0.62 (Heredity), and 0.66 (God's will or fate). Among the most marked differences in rank order were that Chemical imbalance and Heredity were 1st and 4th respectively for beliefs about one's own depression but 7th and 12th for beliefs about depression in general; whereas Loss of loved one was 3rd for beliefs about self and 11th for general beliefs. 3.3.2. Degree of certainty There was a marked difference in the overall level of agreement with items. Whereas sixteen of the seventeen items in relation to beliefs about depression in general were endorsed by more than half the sample (Read et al., 2014a, 2014b), this was the case for only six beliefs about one's own depression. The median frequencies of agreement were 79.8% for beliefs about depression in general and only 38.2% for beliefs about self. 3.3.3. Factor analysis The principal component factor analysis produced five orthogonal factors with eigenvalues 41. No factor, however, accounted for 25% or more of the variance or included more than three items with a loading of 0.5 or more. The factor accounting for the most variance (24.9%), and with the largest eigenvalue (4.2), was comprised of Relationship problems (0.77), Isolation (0.67) and Financial problems (0.65). The second factor (11.9% of the variance; eigenvalue 2.0) included Chemical imbalance (0.80), Disorder of the brain (0.76) and Heredity (0.75). The third factor (8.5%; 1.5) included Drug or alcohol
J. Read et al. / Journal of Affective Disorders 174 (2015) 150–156
abuse (0.71), Unemployment (0.69) and God's will or fate (0.69). The fourth (7.5%; 1.3) included Abuse or neglect in childhood (0.88) and Other unhappy or disturbing childhood experiences (0.87). The final factor (6.0%; 1.0) was comprised of Work stress (0.84) and Hectic pace of modern life (0.75). The remaining five items (see Table 1 for list of all items) did not load (at the 0.5 loading level) on any factor. One factor (3) had no obvious coherent meaning. Only one of the four factors with any obvious meaning (the two-item factor 4) met the usual criteria for good internal reliability of having a Chronbach alpha of 0.80 or above (0.82), with the others ranging from 0.62 to 0.72. Furthermore, to have based further analysis on the four meaningful factors would mean excluding the effects of the seven of the original 17 items which did not load on to any of the four factors. Therefore, further analyses were conducted at the individual item level rather than at the factor level.
4. Discussion 4.1. Causal beliefs about own depression This large sample of AD users places somewhat greater emphasis on the ‘chemical imbalance’ notion in understanding their own depression than the studies, summarised in the Introduction, of the public's beliefs (depressed or not), or the current sample (Read et al., 2014b), does about depression in general. The current sample also differed from the 191 primary care patients in the USA, for whom ‘stress’ was the only cause endorsed by more than half the sample as a cause of their own depression (Brown et al., 2007). Nevertheless, despite decades of attempts to persuade the public that feeling depressed is a medical type illness with predominantly bio-genetic causes, the public continues to hold a multifactorial model and to place great weight on life events such as stress at work and at home, relationship problems and loneliness. They do so even when experiencing depression, even when meeting with medical practitioners about that depression, even when receiving a biologically based treatment, and even when they experience that treatment as helpful. The range of ‘other’ causes demonstrates the breadth of explanations used by AD recipients, some of which are not captured by studies that use only pre-determined closed questions. 4.2. Relationships between causal beliefs and treatment variables Bio-genetic explanations of one's own depression were significantly related (independently of demographics and other beliefs) to both the perceived efficacy of the ADs and the length of time taking the medication. Both these findings also occurred in the study of beliefs about depression in general. While it is possible that these beliefs influence one's perceptions of the dugs’ efficacy, it is also likely that experiencing alleviation of depression or improvement in one's quality of life while taking a biologically based treatment will enhance a biological explanatory model (Karp, 1993). Of course, both these processes may be operating, in a reciprocal manner. Conversely, believing that one's depression is caused primarily by one's life circumstances may reduce believe in the drug's efficacy and/or result from, or be strengthened by, the drug's failure to make one feel better. Cognitive dissonance may be operating, in that a behaviour (in this case taking ADs) elicits beliefs consistent with that behaviour. It is somewhat easier to determine the direction of causality (if any exists) when it comes to the patient–doctor relationship. Since a poor relationship with a doctor does not cause unemployment, child abuse or isolation, one can assume that believing that these sorts of life events have made you feel depressed might hinder the development of a therapeutic relationship, especially, perhaps, if the doctor
155
has a strong bio-medical framework for depression, as is often – but not always – the case (Lester et al., 2005; Schumann et al., 2012). The relationship between reduced recovery expectations and bio-genetic beliefs (found in the same sample in relation to beliefs about depression in general, and in many previous studies) did not reach the 0.01 level of significance set. The fact that it reached the 0.05 level even after controlling for demographics and other causal beliefs suggests that there is no cause here to doubt the findings of the previous studies summarised in Section 1. 4.3. Comparison with causal beliefs about depression in general The three factors generated by the current participants' beliefs about depression in general (Read et al., 2014b) replicated the three emerging from the beliefs of the public in Germany (Schomerus et al., 2014). In the current study, however, the principal component factor analysis of beliefs about one's own depression did not produce the same three factors, despite using exactly the same 17 items. Although the factor analysis did produce two of the three factors (‘Bio-genetic’ and ‘Childhood Adversity’), the third factor, ‘Adulthood Stress’ was not replicated. It seems that one cannot assume that beliefs about our own problems and lives cluster in the same way that our beliefs about other people do. The markedly lower levels of endorsement of items about one's own depression, compared to beliefs about depression in general, can be at least partially explained by the fact that if a discrete event, such as abuse or unemployment, has not happened to you, it cannot be a cause of your depression; but you may still endorse it as a general cause. (This may also be a partial explanation for the different factors that emerged). This is illustrated by the fact that the four causes with the greatest reduction from beliefs in general to beliefs about self were all phenomenon experienced by only some of any population: Unemployment (80% 19% ¼61%), Drug or alcohol abuse (78% 18% ¼60%), Loss of loved one (87% 36%¼ 51%) and Abuse or neglect in childhood (85% 38%¼ 47%). Conversely, the smallest reductions were for causes that, if true, would be more likely to apply to all depressed people: Chemical imbalance (85% 67%¼18%) and Heredity (77% 60% ¼17%). Thus bio-genetic causal beliefs are at a distinct advantage when it comes to ranking based on beliefs about one's own problems. Nevertheless it also seems, given that even these bio-genetic beliefs decreased somewhat, people may be more reluctant to form views about the causes of their own depression. More than 25% of respondents ‘strongly’ agreed with thirteen causes of depression in general, but this was the case for only two causes in relation to their own depression. Causes identified in response to the open-ended ‘other’ question were broadly similar, with medical conditions being by far the highest type of cause in relation to both one's own depression and depression in general (Read et al., 2014b), with childbirth, violence, negative thinking and other psychological/mental health problems also featuring in both lists. The relationships between causal beliefs and demographics were different for the two sets of beliefs. The only two consistent relationships were between women and ‘Loss of loved one’ and between lower income and ‘Other unhappy or distressing childhood experiences.’ In the study of beliefs about depression in general women agreed significantly more than men on 15 of the 17 items. This was not the case in the current study. It seems that men's relative reluctance to express a view about the causes of other people's depression does not apply when it comes to explaining their own depression. 4.4. Limitations This self-selected, convenience sample, despite being the largest ever surveyed, was not, in some regards, representative of the New Zealand population. Maori, Pacific Islanders, men, older people, and
156
J. Read et al. / Journal of Affective Disorders 174 (2015) 150–156
poorer and less educated people were all underrepresented. The overrepresentation of women (77%) is not of great concern because women are prescribed ADs at approximately twice the rate as men internationally. Given that the majority (83%) reported that they believed the drugs had reduced their depression, a rate far higher than most conventional efficacy studies of ADs, it is possible the sample was unrepresentative in terms of positive attitudes towards ADs, and therefore, perhaps, towards bio-genetic explanations of depression. The study relied on self-report. Conventional studies however, also rely, to varying degree, on self-report. One concern is that some of the data is retrospective and therefore subject to the fallibilities of memory of past experiences, some of them several years in the past. The majority (69%), however, were still taking the ADs at the time of completing the questionnaire. The close juxtaposition in the questionnaire of the two sets of belief questions may have artificially inflated similarities between the two. If this occurred it was a minimal effect since some of the most important findings are, in fact, the differences between the two sets of beliefs, both in structure and content. The fact remains, however, that the differences might have been even greater if, for instance, the two sets of questions had been posed with a time interval between them. The questionnaire would have been more comprehensive had it included some of the causal beliefs most frequently identified by participants in response to the ‘other’ questions (e.g. medical conditions). Consideration should be given to including these additional causes in future studies. As noted earlier (4.2), the direction of causality of some of the correlational findings cannot be determined from this crosssectional study. Prospective studies are therefore required. 4.5. Conclusions People coming into contact with GPs or psychiatrists when feeling depressed have complex, varied and idiosyncratic combinations of beliefs, psycho-social and bio-genetic, about what is wrong and what they need help with. Taking a little time, where possible, to explore those beliefs and needs before deciding on a treatment plan, may not only enhance the therapeutic relationship (itself a powerful ‘placebo’ effect) but increase the chances of whatever treatment is eventually tried being effective.
Role of funding source The study was supported by the University of Auckland's Faculty Research Development Fund.
Conflict of interest No conflicts declared.
Acknowledgements We thank all those who took the time to complete the survey.
References Angermeyer, M., Holzinger, A., Carta, M., Schomerus, G., 2011. Biogenetic explanations and public acceptance of mental illness: systematic review of population studies. Br. J. Psychiatry 199, 367–372. Angermeyer, M., Matschinger, H., Carta, M., Schomerus, G., 2013. Changes in the perception of mental illness stigma in Germany over the last two decades. Eur. Psychiatry 29, 390–395. Bann, C., Parker, C., Bradwejn, J., Davidson, J., Vitiello, B., Gadde, K., 2004. Assessing patient beliefs in a clinical trial of Hypericum perforatum in major depression. Depress. Anxiety 20, 114–122. Brown, C., Battista, D., Sereika, S., Bruehlman, R., Dunbar-Jacob, J., 2007. Primary care patients' personal illness models for depression: relationship to coping behavior and functional disability. Gen. Hosp. Psychiatry 29, 492–500.
Dowrick, C., Frances, A., 2013. Medicalising unhappiness: new classification of depression risks more patients being put on drug treatment from which they will not benefit. Br. Med. J. 347, f7140. Gibson, K., Cartwright, C., Read, J., 2014. Patient-centred perspectives on antidepressant use: a narrative review. Int. J. Ment. Health Nurs. 43, 81–99. Hansson, M., Chotai, J., Bodlund, O., 2010. Patients' beliefs about the cause of their depression. J. Affect. Disord. 124, 45–59. Health and Social Care Information Centre, 2012. Anti-Depressant Prescriptions Account for Largest Annual Rise in Items Dispensed in the Community. HSCIC, London. Health and Social Care Information Centre, 2014. Anti-Depressant Prescriptions. HSCIC, London, pp. 2010–2013. Hughes, J., Gabbay, M., Funnell, E., Dowrick, C., 2011. Exploratory review of placebo characteristics reported in randomised placebo controlled antidepressant drug trials. Pharmacopsychiatry 44, 1–8. Ilyas, S., Moncrieff, J., 2012. Trends in prescriptions and costs of drugs for mental disorders in England, 1998–2010. Br. J. Psychiatry 200, 393–398. Jorm, A., Christensen, H., Griffiths, K., 2005. Public beliefs about causes and risk factors for mental disorders: changes in Australia over 8 years. Soc. Psychiatry Psychiatr. Epidemiol 40, 764–767. Karp, D., 1993. Taking anti-depressant medications: resistance, trial commitment, conversion, disenchantment. Qual. Sociol. 16, 337–359. Kirsch, I., Deacon, B., Huedo-Medina, T., Scoboria, A., Moore, T., Johnson, B., 2008. Initial severity and antidepressant benefits: a meta-analysis of data submitted to the Food and Drug Administration. PLOS Med. 5, 260–268. Kvaale, E., Haslam, N., Gottdiener, W., 2013. The ‘side effects’ of medicalization: a meta-analytic review of how biogenetic explanations affect stigma. Clin. Psychol. Rev. 33, 782–794. Lam, D., Salkovskis, P., 2007. An experimental investigation of the impact of biological and psychological causal explanations on anxious and depressed patients' perception of a person with panic disorder. Behav. Res. Ther. 45, 405–411. Lebowitz, M., 2014. Biological conceptualizations of mental disorders among affected individuals: a review of correlates and consequences. Clin. Psychol. Sci. Pract. 21, 67–83. Lester, H., Tritter, J., Sorohan, H., 2005. Patients' and health professionals' views on primary care for people with serious mental illness: focus group study. Br. Med. J. 330, 1122–1127. Mnich, E., Makowski, A., Lambert, M., Angermeyer, M., vondemKnesebeck, O., 2014. Beliefs about depression – do affliction and treatment experience matter? Results of a population survey from Germany. J. Affect Disord. 164, 28–32. Munizza, C., Argentero, P., Coppo, A., Tibaldi, G., Di Giannantonio, M., Picci, R., Rucci, P., 2013. Public beliefs and attitudes towards depression in Italy: a national survey. PlOS One 8, e63806. Ogden, J., Boden, J., Caird, R., Chor, C., Flynn, M., Khan, K., MacLurg, K., Swade, S., Thapar, V., 1999. ‘You're depressed’; ‘No I'm not’: GPs' and patients' different models of depression. Br. J. Gen. Pract. 49, 123–124. Olfson, M., Marcus, S., 2009. National patterns in antidepressant medication treatment. Arch. Gen. Psychiatry 66, 848–856. Pigott, H., Leventhal, A., Alter, G., Bowen, J., 2010. Effficacy and effectiveness of antidepressants: current status of research. Psychother. Psychosom. 79, 267–279. Pilkington, P., Reavley, N., Jorm, A., 2013. The Australian public's beliefs about the causes of depression: associated factors and changes over 16 years. J. Affect. Disord. 150, 356–362. Prins, M., Verhaak, P., Bensing, J., Van Der Meer, K., 2008. Health beliefs and perceived need for mental health care of anxiety and depression: the patients' perspective explored. Clin. Psychol. Rev. 28, 1038–1058. Read, J., Haslam, N., Sayce, L., Davies, E., 2006. Prejudice and schizophrenia: A review of the ‘Mental illness is an Illness like any other’ approach. Read, J., Cartwright, C., Gibson, K., 2014a. Adverse emotional and interpersonal effects reported by 1829 New Zealanders while taking antidepressants. Psychiatry Res. 216, 67–73. Read, J., Cartwright, C., Gibson, K., Shiels, C., Haslam, N., 2014b. Beliefs of people taking antidepressants about causes of depression and reasons for increased prescribing rates. J. Affect. Disord. 168, 236–242. Read, J., Haslam, N., Magliano, L., 2013. Prejudice, stigma and ‘schizophrenia’: the role of bio-genetic ideology. In: Read, J., Dillon, J. (Eds.), Models of Madness: Psychological, Social and Biological Approaches to Psychosis. Routledge, London, pp. 157–177. Schofield, P., Crosland, A., Waheed, W., Aseem, S., Gask, L., Wallace, A., Dickens, A., Tylee, A., et al., 2011. Patients' views of antidepressants: from first experiences to becoming expert. Br. J. Gen. Pract. 61, e142–e148. Schomerus, G., Matschinger, H., Angermeyer, M., 2014. Causal beliefs of the public and social acceptance of persons with mental illness: a comparative analysis of schizophrenia, depression and alcohol dependence. Psychol. Med. 44, 304–314. Schumann, I., Schneider, A., Kantert, C., Löwe, B., Linde, K., 2012. Physicians' attitudes, diagnostic process and barriers regarding depression diagnosis in primary care: a systematic review of qualitative studies. Fam. Pract. 29, 255–263. Shakespeare W., The Merchantmerchant of Venice, Craig W.J., (Eds.) The Oxford Shakespeare, 1597/ 1914, Oxford University Press; London. Srinivasan, J., Cohen, N., Parikh, S., 2003. Patient attitudes regarding causes of depression: implications for psychoeducation. Can. J. Psychiatry 48, 493–495. Statistics New Zealand, 2012. New Zealand in Profile 2012. 〈http://www.stats.govt.nz〉. Turner, K., Sharp, D., Folkes, L., Chew-Graham, C., 2008. Women's views and experiences of antidepressants as a treatment for postnatal depression: a qualitative study. Fam. Pract. 25, 450–455.
Contents lists available at ScienceDirect
Journal of Affective Disorders journal homepage: www.elsevier.com/locate/jad
Research report
Beliefs of people taking antidepressants about the causes of their own depression John Read a,n, Claire Cartwright b, Kerry Gibson b, Christopher Shiels a, Lorenza Magliano c a
Institute of Psychology, Health and Society, University of Liverpool, Liverpool L69 3GB, UK School of Psychology, University of Auckland, New Zealand c Department of Psychology, Second University of Naples, Italy b
art ic l e i nf o
a b s t r a c t
Article history: Received 14 August 2014 Received in revised form 4 October 2014 Accepted 6 November 2014 Available online 28 November 2014
Background: The beliefs of people receiving treatment about the causes of their own mental health problems are researched less often than the causal beliefs of the public, but have important implications for relationships with prescribers, treatment choices and recovery. Method: An online survey on a range of beliefs about depression, and experiences with antidepressants, was completed by 1829 New Zealand adults prescribed anti-depressants in the preceding five years, 97.4% of whom proceeded to take antidepressants. Results: Six of 17 beliefs about the causes of their own depression were endorsed by more than half the sample: chemical imbalance, family stress, work stress, heredity, relationship problems and distressing events in childhood. There were some marked differences in content, structure and level of conviction of beliefs about one's own depression and the sample's previously published beliefs about depression in general. There were also significant differences between the beliefs of demographic groupings. Regression analyses revealed that self-reported effectiveness of the antidepressants was positively associated with bio-genetic causal beliefs. The quality of the relationship with the prescribing doctor was positively related to a belief in chemical imbalance as a cause and negatively related to a belief in unemployment as a cause. Limitations: The convenience sample may have been biased towards a favourable view of bio-genetic explanations, since 83% reported that the medication reduced their depression. Conclusions: People experiencing depression hold complex, multifactorial and idiosyncratic sets of beliefs about the causes of their own depression, apparently based at least in part on their own life experiences and circumstances. Exploring those beliefs may enhance the doctor–patient relationship and selection of appropriate treatment modality. & 2014 Elsevier B.V. All rights reserved.
Keywords: Depression Causal beliefs Antidepressants Stigma Prognostic pessimism
1. Introduction In sooth, I know not why I am so sad: It wearies me; you say it wearies you; But how I caught it, found it, or came by it, What stuff 'tis made of, whereof it is born, I am to learn. (Shakespeare, 1597) In New Zealand the number of people prescribed antidepressants [ADs] increased by 35% from 304,530 to 412,631 between 2006/07 and 2011/12 (PHARMAC, personal communication, 2012), in a population of 4.4 million, of whom 3.7 million are aged 16 or older. Thus one in nine of the adult population (and approximately one in six women) receive ADs every year. In the USAADs had n
Corresponding author. þ 44 7923 892 140. E-mail address: [email protected] (J. Read).
http://dx.doi.org/10.1016/j.jad.2014.11.009 0165-0327/& 2014 Elsevier B.V. All rights reserved.
become the most frequently prescribed drug type by 2005, with one in ten people over the age of six being prescribed ADs annually (Olfson and Marcus, 2009). In England, prescriptions for ADs increased 10% per year between 1998 and 2010 (Ilyas and Moncrieff, 2012). In 2013 there were 53 million prescriptions, in a country with a total population of 52.6 million (Health and Social Care Information Centre, 2012, 2014). These increases occur despite evidence that their efficacy was initially overestimated (Kirsch et al., 2008; Pigott et al., 2010) and their adverse effects minimised. For example, eight adverse effects were reported by over 50% of the large sample of AD recipients used in the current study, including sexual difficulties, emotional numbing, drowsiness, weight gain and withdrawal effects (Read et al., 2014a). Concerns about over-prescribing are often framed within a broader critique about the ‘medicalisation’ of distress (Dowrick and Frances, 2013; Ilyas and Moncrieff, 2012; Read and Dillon, 2013). Therefore the opinions and beliefs of the millions of people who take ADs are important. A recent review of AD trials found,
J. Read et al. / Journal of Affective Disorders 174 (2015) 150–156
however, that none reported patients' illness beliefs (Hughes et al., 2011). A review of the small body of qualitative research that has sought the views of AD recipients concluded that patients have ambivalent, but predominantly negative, views of ADs and that they tend to prefer psychotherapy (Gibson et al., 2014). One set of beliefs that has been studied extensively, is beliefs about the causes of depression, largely because causal beliefs are considered central to stigma. Contrary to the hopes of the designers of destigmatisation programmes based on the ‘mental illness is an illness like any other’ notion, which promote biogenetic causal explanations, such illness-type beliefs have been consistently shown to increase fear and prejudice and decrease expectation of recovery (Angermeyer et al., 2011; Kvaale et al., 2013; Read et al., 2006, 2013; Schomerus et al., 2014). This body of research also tells us what the public actually believes about the causes of depression. Despite decades of wellresourced programmes promoting a bio-medical perspective (often funded by drug companies), the public, internationally, continues to prefer psycho-social explanations (Angermeyer et al., 2013; Jorm et al., 2005; Munizza et al., 2013; Pilkington et al., 2013; Schomerus et al., 2014). For example, the German public endorses ‘family/ partnership problems’ (81%) and ‘work stress’ (80%) more than ‘brain disease’ (38%) or ‘heredity’ (40%) as a cause of ‘major depression’ (Angermeyer et al., 2013). In Australia, although belief in ‘inherited or genetic’ increased from 49% in 1995 to 65% in 2011; the most frequently endorsed causes remained ‘day to day problems’ (97%), ‘death of close friend/relative’ (97%) and ‘problems in childhood’ (96%) (Pilkington et al., 2013). Findings about the causal beliefs of depressed people (about depression in general, rather than about their own depression), confirm that, like the general population, they prefer psychological and social explanations to biological or medical ones. An English study of generic GP patients found that those with experience of depression were more likely to endorse ‘medical’ causes of depression (‘hormones’ and ‘chemical imbalances in the brain’) than those who had never experienced depression. Nevertheless, both the depressed and non-depressed groups endorsed ‘psychological’ causes (stress, life events, unhappy childhood) slightly more strongly than medical causes (Ogden et al., 1999). A recent German study found that depressed people are even more likely than other members of the public to endorse ‘psychosocial stress’ items (job stress, family strain, and critical life event) and ‘conditions of socialisation’ (broken home, lack of parental affection and sexual abuse in childhood) (Mnich et al., 2014). A 2008 review of studies of the beliefs of depressed people about the causes of their own depression found that most are consistent with the general public's beliefs (Prins et al., 2008). A study of 340 people with ‘major depression’ found that ‘psychological’ items were far more strongly endorsed than ‘biological items’ as causes of their own depression, with by far the most frequently endorsed being ‘stress and negative life experiences’ (Bann et al., 2004). A Swedish study, of 303 depressed primary care patients, found that 69% endorsed ‘current life stressors’, 32% ‘past life events’ and 4% ‘biological/heredity’ (Hansson et al., 2010). The causal belief most frequently endorsed by 102 depressed Canadian outpatients was ‘stressful and negative life experiences’ (prompting the researchers to call for ‘proper psychoeducation’) (Srinivasan et al., 2003). So, experiencing depression does not seem to reduce the public's psycho-social understanding of depression. It remains possible, however, that interacting with medical professionals and receiving a treatment explicitly based on a bio-medical causal model, might affect one's causal beliefs. This would be important because these beliefs have been shown to be related to the relationship with the prescribing doctor, including the potential of that relationship to shape ideas about depression and about
151
appropriate treatments (Karp, 1993; Schofield et al., 2011; Turner et al., 2008). Little research has been undertaken targeting the causal beliefs of people taking ADs. In a review of studies of the causal beliefs of ‘affected individuals’ (Lebowitz, 2014) only one reports the causal beliefs of AD recipients. This study, of 191 primary care patients in the USA. found that ‘stress’ (81%) was the only cause endorsed by more than half the sample as a cause of their own depression (Brown et al., 2007). Subsequently, the causal beliefs, about depression in general, of 1829 AD recipients in New Zealanders (the same sample as the current study) has been reported. The most commonly endorsed causes were: family stress (90.8%), relationship problems (89.9%), loss of loved one (87.5%), financial problems (86.9%), isolation (86.3%), abuse or neglect in childhood (85.4%), and chemical imbalance (84.8%) (Read et al., 2014b). The negative effects of bio-genetic beliefs found in studies of the public are replicated in people who are depressed or taking ADs, particularly in terms of pessimism about recovery (Lebowitz, 2014; Lam and Salkovskis, 2007). In the large New Zealand sample of AD recipients, bio-genetic causal beliefs were significantly correlated with the belief that ‘people can't get better by themselves even if they try’ (Read et al., 2014b). No study, to the best of our knowledge, has compared the causal beliefs of depressed people, or AD recipients, about depression in general with the causal beliefs of the same sample about their own depression. Therefore, the three aims of the current paper were: 1. To report the beliefs of a large sample of AD recipients about the causes of their own depression. 2. To analyse those beliefs in terms of demographics and in relation to prognostic pessimism, perceived efficacy of the ADs, quality of relationship with the prescriber, length of AD use and beliefs about placebo effects. 3. To compare the content and factorial structure of those beliefs with the previously published beliefs of the same sample about the causes of depression in general.
2. Method 2.1. Instrument and procedure The study used an online questionnaire with 47 questions covering: the prescribing process; perceptions of the effectiveness of the ADs; side-effects; benefits; and beliefs about the causes of depression. Responses to the question ‘There are many theories, and lots of debate, about what causes problems like depression. Please indicate the extent to which you agree or disagree that the following factors are causes of depression IN GENERAL’ have been published (Read et al., 2014b). This question was immediately followed by the key question for the current paper: Please indicate the extent to which you agree or disagree that the following factors are causes of YOUR OWN problems’. Both questions were followed by a list of 17 items, based on the research literature and including bio-genetic (e.g. Chemical imbalance, and Heredity (genes)) and psycho-social causes (e.g. Family stress and Isolation) (see Table 1 for complete list), and five point Likert scales from ‘strongly disagree’ to ‘strongly agree’. Participants were also invited to identify ‘other’ causes, of both depression in general and their own depression. Criteria for participation included having been prescribed ADs in the last five years and being at least 18 years old. A webpage (which was not linked to other websites) provided the participant information for the study and a link to the online questionnaire. The study
152
J. Read et al. / Journal of Affective Disorders 174 (2015) 150–156
Table 1 Percentage agreement, mean scores, and demographic relationships for 17 causal beliefs.
Chemical imbalance Family stress Work stress Heredity (genes) Relationship problems Other unhappy or distressing childhood experiences Isolation Financial problems Hectic pace of modern life Abuse or neglect in childhood Loss of loved one Unconscious conflict Disorder of the brain Unstable personality Unemployment Drug or alcohol abuse God's will or fate
Agreea (%)
Meanb (s.d.)
66.6 65.6 60.4 60.3 56.2 50.5 48.7 43.5 40.4 38.2 35.9 32.9 28.8 23.5 18.6 17.7 2.1
3.78 3.70 3.53 3.50 3.41 3.18 3.18 3.02 2.99 2.80 2.77 2.85 2.72 2.45 2.07 2.02 1.36
(1.25) (1.30) (1.33) (1.32) (1.40) (1.51) (1.42) (1.40) (1.32) (1.59) (1.54) (1.31) (1.35) (1.30) (1.35) (1.33) (0.78)
Gender
Age
Female nn Malennn Female n
Old
Income
nn
Young nn Old nnn
Female
Male Male Male
n
n nnn
Old
nnn
Low Low Low
n
Low Low
nnn
nnn n
nnn
Young nn Young nn Old nnn
nnn
High
nnn
Rank for beliefs about depression in generalc 7 1 9 12 2 8 5 4 15 6 3 16 13 14 10 11 17
a
‘Agree or ‘strongly agree’. 1¼ Strongly disagree and 5¼ Strongly agree c Same sample: (Read et al., 2014a, 2014b). n po 0.01. nn p o0.001. nnn p o0.0001. b
was further publicised in the New Zealand media via media releases, interviews with the researchers and advertisements. The study was approved by the University of Auckland Human Participants Ethics Committee. 2.2. Participants Of the 2171 people who started the survey, 295 stopped before question 20 and their responses were not analysed. Of the remaining 1876, 45 cited medications other than ADs. Two respondents whose scores were nearly identical to those of someone else with the same Internet Protocol address (indicating probable use of the same computer) were excluded. This left 1829 for analysis (97.4% of whom had taken the ADs when prescribed them). The number of responses to each question varied as not all participants responded to all questions. The sample has been described in detail elsewhere (Read et al., 2014a). Females constituted 76.6% of the sample. The modal age group was 36–45 (24.2%); 92.1% identified as ‘New Zealand/ European’. In terms of education, 49.6% had a university degree, 26.1% gained a diploma or certificate after high school, 17.2% completed high school, and 7.1% did not complete high school. Annual income (in NZ dollars) ranged from less than $10,000 (15.0%) to more than $100,000 (7.7%). The modal income was $40,000–$59,999 (22.1%). (The median income of the NZ population in 2012 was $29,000 (Statistics New Zealand, 2012)). 2.3. Data analysis Associations between the causal beliefs and demographic variables were tested using t-tests (two-tailed) and spearman rank correlation coefficients. Age was recorded as 18–25, 26–35, 46–55, 56–65, 66–75, 76–85 or 86 þ. Education categories were: ‘Did not complete High School’, ‘Completed High School’, ‘Diploma/Certificate after High School’ ‘Undergraduate Degree’ and ‘Postgraduate Degree.’ For the main analysis, seven survey items were transformed into dichotomous (Yes/No) outcome variables: perceived reduction in depression during antidepressant use; perceived improvement in quality of life; reported use of antidepressants for three years or more; ‘good’ relationship with prescribing doctor; felt that the doctor
understood their problem; belief that the effect of antidepressant was 450% placebo; and 50% agreement with the statement, “People can’t get better by themselves, even if they try”. Univariate analysis, using 2 2 cross-tabulations and a chisquare test for statistical significance, was conducted in order to investigate associations between the seventeen causal belief items and the seven binary outcomes. Beliefs found to be significantly associated (p o0.05) with outcome were entered into subsequent multivariate models. Seven logistic regression models were constructed and run in order to estimate the independent effect of beliefs about causes of own depression on each of the seven outcomes. Only the causal belief items that attained significance (po0.05) in the respective univariate analysis were entered as covariates in each model. For each independent variable, the Odds Ratio, 99% Confidence Interval and p-value were generated. In all logistic regression models the estimates of the effect of causal beliefs on outcome were adjusted for respondent gender, age, annual income and whether religious or not. A more stringent criterion of significance (po0.01) was used in interpretation of effects in the regression analysis. In order to allow a comparison with the data structure in our previous study looking at general beliefs about depression, using the same sample and methodology (Read et al., 2014a, 2014b), responses to the 17 causal belief questions were entered into an explorative principal-component (varimax rotation) factor analysis. Factors were identified using an eigenvalue of Z1 as criterion. The relationship between the two sets of causal beliefs (about own depression and depression in general) was assessed using a Spearman rank correlation coefficient.
3. Results 3.1. Beliefs of AD recipients about the causes of their own depression 3.1.1. Responses to questionnaire items Table 1 summarises responses to the 17 causal beliefs presented to participants in the questionnaire. Six items were endorsed (‘agree’ or ‘strongly agree’) by more than half of all respondents as causes of their own depression: Chemical imbalance (67%), Family stress (66%),
J. Read et al. / Journal of Affective Disorders 174 (2015) 150–156
Work stress (60%), Heredity (60%), Relationship problems (56%) and Other unhappy or distressing childhood experiences (51%). 3.1.2. ‘Other’ causal beliefs The invitation to identify ‘other’ causes elicited one or more from 324 respondents. Those reported by more than five people are listed in Table 2. The most common type involved a range of medical conditions (47), ten of which involved ongoing pain and six of which were identified as head/brain injuries. Twenty three participants identified difficulties either with giving birth or caring for a new born baby. For example: Birth of a child, adjustment to motherhood, poor family support after having children. Lack of sleep; feeling of isolation/change in identity with a screaming newborn. Six of these 23 described their difficulties as ‘post natal depression’. Psychological difficulties were identified by 22 respondents, including low self-esteem, negative thinking, perfectionism, oversensitivity and a lack of ways to cope with either emotions or stressors. For example: Not knowing how to deal with stressful or confusing situations. Over-sensitivity and inability to deal with stress and anxiety, and a mindset that creates unnecessary anxieties in the first place. Low self-esteem, unhelpful ideas. Overwhelming emotions – over sensitive. Anxiety caused by a bunch of really self-depricating core beliefs. Seventeen identified specific mental health problems as a cause of their depression, most commonly ‘post natal depression’ (6) and ‘PTSD’ (5). Violence in adulthood was reported as a cause by 15 people, mostly within intimate relationships. For example: Domestic violence …….. Physical, emotional and mental abuse over sixteen years and continuing after separation through alienating the children from me. Prolonged abusive relationship with husband. Fifteen either reported ‘trauma’ (10) or ’PTSD’ (5) without specifying the nature of the trauma. Eleven identified a range of adverse experiences in childhood, for example: Depressed mother during infancy and childhood. Sustained bullying in childhood. Sexual abuse as a child. Table 2 ‘Other’ causal beliefs. n Medical conditions Childbirth/newborn baby/‘post natal’ Psychological difficulties Violence in adulthood Trauma/PTSD Negative childhood experiences Study stress Medication side effects Societal injustice Anxiety Others' illness/mental health problem Earthquakes Parenting stress No direction/hope Immigration stress
47 23 22 15 15 11 9 7 6 6 6 5 5 5 5
153
Six wrote about societal injustice as a cause of their depression: There is an inextricable relationship between inequality and injustice and my personal views of myself. The environment affects me. The hopelessness of modern capitalism and the state of our society in which social injustice, poverty and cruelty are both tolerated and accepted by most people. Frustration at greater problems, “weight of the world” depression. Having the perception to consider huge issues but feeling insignificant and ineffectual when it comes to solving them. Other responses were not easily categorised but seem worthy of citing, to demonstrate the diversity of views: Not one thing but a gradual build up of many things. I don’t think I was ‘Depressed’ I think I was struggling to cope with my circumstances and I needed to be taught coping methods…not labelled and put onto pills. I don't know about hereditary, but I feel I was taught depression from my mother and older sister, by not being shown how to deal with my emotions and anger properly. Neighbours excessive noise. I have these issues, which make me depressed BUT feel that a lot of these issues are caused by the way I am after 9yrs of medication. Anti depressant drugs magnify the symptoms. Side effects from medication, loss of health. sense of injustice, Lack of support, funding, Poverty, Abuse and neglect of Mental health system, and domestic abuse. In my case, I have no reason for depression, I was born with it and was depressed as a child. I am not at all sure what caused my depression, so feel I can't really answer the above. Sorry. We all face problems…its called ‘life’. 3.2. Relationships of causal beliefs with other variables 3.2.1. Demographics Table 1 shows that seven causal beliefs were significantly related to age, with by far the largest correlation being between Loss of loved one and older age (r ¼0.17, p o0.0001). There were significant gender differences on the mean scores of seven of the 17 items (with 5 representing ‘strongly agree’ and 1 representing ‘strongly disagree’). The largest differences were that men more strongly endorsed Work stress (mean 3.81, 95% CI 3.69–3.93 vs mean 3.45, 95% CI 3.36–3.52 ), Unemployment (mean 2.42, 95% CI 2.26–2.58 vs mean 1.95, 95% CI 1.86–2.02) and Drug or alcohol abuse (mean 2.33, 95% CI 2.16–2.46 vs mean 1.92, 95% CI 1.84–2.00); while the women more strongly endorsed Family stress (mean 3.76, 95% CI 3.68–3.84 vs mean 3.50, 95% CI 3.36–3.64). The strongest relationships with income were a positive correlation with Work stress (r ¼0.16, p o0.0001), and a negative correlation with Isolation (r ¼0.19, p o0.0001). There were no significant relationships with level of education. The only differences in means in terms of religious belief were that those who described themselves as religious were significantly more likely to endorse God's will or fate (mean 1.57, 95% CI 1.47–1.67 vs mean 1.27, 95% CI 1.23–1.31) and Abuse or neglect in childhood (mean 2.96, 95% CI 2.82–3.11 vs mean 2.71, 95% CI 2.62–2.81). Although Chemical imbalance was the most frequently endorsed belief in the sample as a whole, some demographic groups most strongly endorsed psycho-social beliefs. Family stress was the most endorsed belief by people aged 36–45 and 66–75, and by people with an income less than NZ$40,000. Work stress was the most endorsed cause by men, by people aged 46–55, and by people with an income of between $NZ80,000 and 100,000. (Although the number of Maori participants (45) was too small to include in analyses, we note that
154
J. Read et al. / Journal of Affective Disorders 174 (2015) 150–156
Maori endorsed Family stress, Relationship problems and Work stress more frequently than Chemical imbalance). 3.2.2. Perceived efficacy of antidepressants Those answering ‘yes’ to ‘Did the antidepressants reduce your depression?’ (82.8%) were significantly more likely to believe that Heredity and Chemical imbalance were causes of their depression, and significantly less likely to endorse Abuse or neglect in childhood, Isolation, Financial problems and Unstable personality (see Table 3). Among those who did not think the ADs reduced their depression, the six most endorsed causes were all psycho-social; most frequently Family stress (69.1%), Relationship problems (62.4%) and Isolation (61.4%), with Chemical imbalance (46.9%) ranked seventh. In a regression analysis (controlling for demographics and the nine other causal beliefs related (po0.05) to reduced depression in the univariate analysis) only Heredity and Chemical imbalance remained significantly related to reduced depression. Results for the other measure of perceived efficacy, reported improvement in Quality of Life [QoL], were almost identical (see Table 3). 3.2.3. Interaction with prescribing doctor Participants were asked to respond, on a five-point Likert scale, to the question ‘How would you describe your relationship with your Table 3 Unadjusted and adjusted associations between causal beliefs and key outcomes. Variable Causal beliefs
Univariate analysis (%)
Reduced depression Heredity Chemical imbalance Abuse or neglect in childhood Isolation Unstable personality Financial problems Improved quality of life Chemical imbalance Heredity Abuse or neglect in childhood Isolation Unstable personality Distressing childhood experience Taken ADs for more than 3 years Heredity Brain disorder Chemical imbalance Distressing childhood experience Abuse or neglect in childhood Drug or alcohol abuse Good relationship with doctor Unemployment Distressing childhood experience Abuse or neglect in childhood Isolation Doctor understood problems Chemical imbalance Distressing childhood experience Abuse or neglect in childhood Isolation Drug or alcohol abuse Unemployment Belief that effect is 450% placebo Chemical imbalance Distressing childhood experience Heredity
Yes, No 88.7, 76.6nnn 89.1, 75.5nnn 78.9, 86.3nnn 79.3, 87.5nnn 77.5, 85.4nn 80.6, 86.1nn Yes, No 91.3, 76.6nnn 90.2, 79.3nnn 81.2, 88.4nnn 81.8, 89.4nnn 80.7, 87.4nn 83.4, 88.6nn Yes, No 58.7, 42.3nnn 64.0, 47.0nnn 56.9, 42.2nnn 56.8, 47.9nnn 57.6, 48.6nn 60.5, 49.6nn Yes, No 69.4, 77.4nn 72.5, 80.5nnn 71.5, 79.0nn 72.7, 77.9nn Yes, No 66.3, 53.9nnn 56.0, 67.4nnn 55.0, 65.6nnn 55.4, 67.6nnn 54.2, 63.6nn 54.4 63.4nn Yes, No 12.8, 29.9nnn 21.2, 16.1n 15.3, 23.9nnn
99% CI of OR Adjusted Odds ratioa
2.04 2.03 0.73 0.79 0.64 0.81
1.28–3.04 1.27–2.99 0.44–1.02 0.47–1.12 0.41–1.08 0.54–1.24
2.65 1.76 0.71 0.71 0.66 0.82
1.62–4.03 1.18–2.95 0.37–1.06 0.45–1.12 0.40–1.10 0.49–1.42
1.66 1.53 1.30 1.20 1.15 1.39
1.21–2.31 1.07–2.23 0.93–1.85 0.86–1.73 0.78–1.62 0.93–2.09
0.65 0.78 0.74 0.99
0.39–0.94 0.52–1.15 0.50–1.12 0.67–1.36
1.75 0.72 0.78 0.80 0.76 0.80
1.29–2.40 0.51–1.03 0.53–1.11 0.56–1.04 0.49–1.12 0.52–1.17
0.37 1.59 0.74
0.23–0.52 1.06–2.35 0.48–1.10
Multivariate: Significant (P o0.01) effects in bold. Univariate. a Adjusted for other belief covariates that were significantly related (p o 0.05) at the univariate level, and for age, gender, income and whether religious. n po 0.05. nn p o0.01. nnn p o0.001.
doctor?’. Those who answered ‘good’ or ‘very good’ (75.4% of the whole sample) were significantly less likely to endorse four psycho-social beliefs: Unemployment, Abuse or neglect in childhood, Other unhappy or distressing childhood experiences and Isolation. Only Unemployment remained significant in the regression analysis (OR ¼0.65; po 0.01) When asked ‘How well do you think your doctor understood your problems?’ 60.1% of the responded ‘quite a lot’ or ‘a lot’. Participants who believed in five of the psycho-social causes were significantly less likely to feel understood (see Table 3). However, the only relationship remaining significant in the regression analysis was the positive relationship with Chemical imbalance (OR¼ 1.75, p o0.001). 3.2.4. Duration of use of antidepressants Six causal beliefs (see Table 3) were significantly related to whether a participant had taken ADs for more than three years (51.7% of the total sample), but only the Odds Ratios for Heredity (1.66; po 0.001) and Brain Disorder (1.53; p o0.001) remained significantly related in the regression analysis. 3.2.5. Prognostic pessimism 28.9% of participants agreed with the statement ‘People can't get better by themselves even if they try’. Only Chemical imbalance was related to respondents’ agreement with this statement in the regression analysis (OR ¼1.30; p¼ o0.05) but this did not reach the level of significance set for this study. 3.3. Comparisons with the beliefs of the same sample about causes of depression in general 3.3.1. Beliefs The rank order of participants’ beliefs about the causes of their own depression was correlated (r¼0.57) with that of their beliefs about the causes of depression in general (Read et al., 2014a, 2014b) (see Table 1), but only at the po0.05 level of significance. The two types of beliefs correlated for each of the 17 items at the o0.0001 level, but the strength of the correlations varied from 0.10 (Drug and alcohol abuse) and 0.23 (Loss of loved one) to, at the other extreme, 0.61 (Chemical imbalance), 0.62 (Heredity), and 0.66 (God's will or fate). Among the most marked differences in rank order were that Chemical imbalance and Heredity were 1st and 4th respectively for beliefs about one's own depression but 7th and 12th for beliefs about depression in general; whereas Loss of loved one was 3rd for beliefs about self and 11th for general beliefs. 3.3.2. Degree of certainty There was a marked difference in the overall level of agreement with items. Whereas sixteen of the seventeen items in relation to beliefs about depression in general were endorsed by more than half the sample (Read et al., 2014a, 2014b), this was the case for only six beliefs about one's own depression. The median frequencies of agreement were 79.8% for beliefs about depression in general and only 38.2% for beliefs about self. 3.3.3. Factor analysis The principal component factor analysis produced five orthogonal factors with eigenvalues 41. No factor, however, accounted for 25% or more of the variance or included more than three items with a loading of 0.5 or more. The factor accounting for the most variance (24.9%), and with the largest eigenvalue (4.2), was comprised of Relationship problems (0.77), Isolation (0.67) and Financial problems (0.65). The second factor (11.9% of the variance; eigenvalue 2.0) included Chemical imbalance (0.80), Disorder of the brain (0.76) and Heredity (0.75). The third factor (8.5%; 1.5) included Drug or alcohol
J. Read et al. / Journal of Affective Disorders 174 (2015) 150–156
abuse (0.71), Unemployment (0.69) and God's will or fate (0.69). The fourth (7.5%; 1.3) included Abuse or neglect in childhood (0.88) and Other unhappy or disturbing childhood experiences (0.87). The final factor (6.0%; 1.0) was comprised of Work stress (0.84) and Hectic pace of modern life (0.75). The remaining five items (see Table 1 for list of all items) did not load (at the 0.5 loading level) on any factor. One factor (3) had no obvious coherent meaning. Only one of the four factors with any obvious meaning (the two-item factor 4) met the usual criteria for good internal reliability of having a Chronbach alpha of 0.80 or above (0.82), with the others ranging from 0.62 to 0.72. Furthermore, to have based further analysis on the four meaningful factors would mean excluding the effects of the seven of the original 17 items which did not load on to any of the four factors. Therefore, further analyses were conducted at the individual item level rather than at the factor level.
4. Discussion 4.1. Causal beliefs about own depression This large sample of AD users places somewhat greater emphasis on the ‘chemical imbalance’ notion in understanding their own depression than the studies, summarised in the Introduction, of the public's beliefs (depressed or not), or the current sample (Read et al., 2014b), does about depression in general. The current sample also differed from the 191 primary care patients in the USA, for whom ‘stress’ was the only cause endorsed by more than half the sample as a cause of their own depression (Brown et al., 2007). Nevertheless, despite decades of attempts to persuade the public that feeling depressed is a medical type illness with predominantly bio-genetic causes, the public continues to hold a multifactorial model and to place great weight on life events such as stress at work and at home, relationship problems and loneliness. They do so even when experiencing depression, even when meeting with medical practitioners about that depression, even when receiving a biologically based treatment, and even when they experience that treatment as helpful. The range of ‘other’ causes demonstrates the breadth of explanations used by AD recipients, some of which are not captured by studies that use only pre-determined closed questions. 4.2. Relationships between causal beliefs and treatment variables Bio-genetic explanations of one's own depression were significantly related (independently of demographics and other beliefs) to both the perceived efficacy of the ADs and the length of time taking the medication. Both these findings also occurred in the study of beliefs about depression in general. While it is possible that these beliefs influence one's perceptions of the dugs’ efficacy, it is also likely that experiencing alleviation of depression or improvement in one's quality of life while taking a biologically based treatment will enhance a biological explanatory model (Karp, 1993). Of course, both these processes may be operating, in a reciprocal manner. Conversely, believing that one's depression is caused primarily by one's life circumstances may reduce believe in the drug's efficacy and/or result from, or be strengthened by, the drug's failure to make one feel better. Cognitive dissonance may be operating, in that a behaviour (in this case taking ADs) elicits beliefs consistent with that behaviour. It is somewhat easier to determine the direction of causality (if any exists) when it comes to the patient–doctor relationship. Since a poor relationship with a doctor does not cause unemployment, child abuse or isolation, one can assume that believing that these sorts of life events have made you feel depressed might hinder the development of a therapeutic relationship, especially, perhaps, if the doctor
155
has a strong bio-medical framework for depression, as is often – but not always – the case (Lester et al., 2005; Schumann et al., 2012). The relationship between reduced recovery expectations and bio-genetic beliefs (found in the same sample in relation to beliefs about depression in general, and in many previous studies) did not reach the 0.01 level of significance set. The fact that it reached the 0.05 level even after controlling for demographics and other causal beliefs suggests that there is no cause here to doubt the findings of the previous studies summarised in Section 1. 4.3. Comparison with causal beliefs about depression in general The three factors generated by the current participants' beliefs about depression in general (Read et al., 2014b) replicated the three emerging from the beliefs of the public in Germany (Schomerus et al., 2014). In the current study, however, the principal component factor analysis of beliefs about one's own depression did not produce the same three factors, despite using exactly the same 17 items. Although the factor analysis did produce two of the three factors (‘Bio-genetic’ and ‘Childhood Adversity’), the third factor, ‘Adulthood Stress’ was not replicated. It seems that one cannot assume that beliefs about our own problems and lives cluster in the same way that our beliefs about other people do. The markedly lower levels of endorsement of items about one's own depression, compared to beliefs about depression in general, can be at least partially explained by the fact that if a discrete event, such as abuse or unemployment, has not happened to you, it cannot be a cause of your depression; but you may still endorse it as a general cause. (This may also be a partial explanation for the different factors that emerged). This is illustrated by the fact that the four causes with the greatest reduction from beliefs in general to beliefs about self were all phenomenon experienced by only some of any population: Unemployment (80% 19% ¼61%), Drug or alcohol abuse (78% 18% ¼60%), Loss of loved one (87% 36%¼ 51%) and Abuse or neglect in childhood (85% 38%¼ 47%). Conversely, the smallest reductions were for causes that, if true, would be more likely to apply to all depressed people: Chemical imbalance (85% 67%¼18%) and Heredity (77% 60% ¼17%). Thus bio-genetic causal beliefs are at a distinct advantage when it comes to ranking based on beliefs about one's own problems. Nevertheless it also seems, given that even these bio-genetic beliefs decreased somewhat, people may be more reluctant to form views about the causes of their own depression. More than 25% of respondents ‘strongly’ agreed with thirteen causes of depression in general, but this was the case for only two causes in relation to their own depression. Causes identified in response to the open-ended ‘other’ question were broadly similar, with medical conditions being by far the highest type of cause in relation to both one's own depression and depression in general (Read et al., 2014b), with childbirth, violence, negative thinking and other psychological/mental health problems also featuring in both lists. The relationships between causal beliefs and demographics were different for the two sets of beliefs. The only two consistent relationships were between women and ‘Loss of loved one’ and between lower income and ‘Other unhappy or distressing childhood experiences.’ In the study of beliefs about depression in general women agreed significantly more than men on 15 of the 17 items. This was not the case in the current study. It seems that men's relative reluctance to express a view about the causes of other people's depression does not apply when it comes to explaining their own depression. 4.4. Limitations This self-selected, convenience sample, despite being the largest ever surveyed, was not, in some regards, representative of the New Zealand population. Maori, Pacific Islanders, men, older people, and
156
J. Read et al. / Journal of Affective Disorders 174 (2015) 150–156
poorer and less educated people were all underrepresented. The overrepresentation of women (77%) is not of great concern because women are prescribed ADs at approximately twice the rate as men internationally. Given that the majority (83%) reported that they believed the drugs had reduced their depression, a rate far higher than most conventional efficacy studies of ADs, it is possible the sample was unrepresentative in terms of positive attitudes towards ADs, and therefore, perhaps, towards bio-genetic explanations of depression. The study relied on self-report. Conventional studies however, also rely, to varying degree, on self-report. One concern is that some of the data is retrospective and therefore subject to the fallibilities of memory of past experiences, some of them several years in the past. The majority (69%), however, were still taking the ADs at the time of completing the questionnaire. The close juxtaposition in the questionnaire of the two sets of belief questions may have artificially inflated similarities between the two. If this occurred it was a minimal effect since some of the most important findings are, in fact, the differences between the two sets of beliefs, both in structure and content. The fact remains, however, that the differences might have been even greater if, for instance, the two sets of questions had been posed with a time interval between them. The questionnaire would have been more comprehensive had it included some of the causal beliefs most frequently identified by participants in response to the ‘other’ questions (e.g. medical conditions). Consideration should be given to including these additional causes in future studies. As noted earlier (4.2), the direction of causality of some of the correlational findings cannot be determined from this crosssectional study. Prospective studies are therefore required. 4.5. Conclusions People coming into contact with GPs or psychiatrists when feeling depressed have complex, varied and idiosyncratic combinations of beliefs, psycho-social and bio-genetic, about what is wrong and what they need help with. Taking a little time, where possible, to explore those beliefs and needs before deciding on a treatment plan, may not only enhance the therapeutic relationship (itself a powerful ‘placebo’ effect) but increase the chances of whatever treatment is eventually tried being effective.
Role of funding source The study was supported by the University of Auckland's Faculty Research Development Fund.
Conflict of interest No conflicts declared.
Acknowledgements We thank all those who took the time to complete the survey.
References Angermeyer, M., Holzinger, A., Carta, M., Schomerus, G., 2011. Biogenetic explanations and public acceptance of mental illness: systematic review of population studies. Br. J. Psychiatry 199, 367–372. Angermeyer, M., Matschinger, H., Carta, M., Schomerus, G., 2013. Changes in the perception of mental illness stigma in Germany over the last two decades. Eur. Psychiatry 29, 390–395. Bann, C., Parker, C., Bradwejn, J., Davidson, J., Vitiello, B., Gadde, K., 2004. Assessing patient beliefs in a clinical trial of Hypericum perforatum in major depression. Depress. Anxiety 20, 114–122. Brown, C., Battista, D., Sereika, S., Bruehlman, R., Dunbar-Jacob, J., 2007. Primary care patients' personal illness models for depression: relationship to coping behavior and functional disability. Gen. Hosp. Psychiatry 29, 492–500.
Dowrick, C., Frances, A., 2013. Medicalising unhappiness: new classification of depression risks more patients being put on drug treatment from which they will not benefit. Br. Med. J. 347, f7140. Gibson, K., Cartwright, C., Read, J., 2014. Patient-centred perspectives on antidepressant use: a narrative review. Int. J. Ment. Health Nurs. 43, 81–99. Hansson, M., Chotai, J., Bodlund, O., 2010. Patients' beliefs about the cause of their depression. J. Affect. Disord. 124, 45–59. Health and Social Care Information Centre, 2012. Anti-Depressant Prescriptions Account for Largest Annual Rise in Items Dispensed in the Community. HSCIC, London. Health and Social Care Information Centre, 2014. Anti-Depressant Prescriptions. HSCIC, London, pp. 2010–2013. Hughes, J., Gabbay, M., Funnell, E., Dowrick, C., 2011. Exploratory review of placebo characteristics reported in randomised placebo controlled antidepressant drug trials. Pharmacopsychiatry 44, 1–8. Ilyas, S., Moncrieff, J., 2012. Trends in prescriptions and costs of drugs for mental disorders in England, 1998–2010. Br. J. Psychiatry 200, 393–398. Jorm, A., Christensen, H., Griffiths, K., 2005. Public beliefs about causes and risk factors for mental disorders: changes in Australia over 8 years. Soc. Psychiatry Psychiatr. Epidemiol 40, 764–767. Karp, D., 1993. Taking anti-depressant medications: resistance, trial commitment, conversion, disenchantment. Qual. Sociol. 16, 337–359. Kirsch, I., Deacon, B., Huedo-Medina, T., Scoboria, A., Moore, T., Johnson, B., 2008. Initial severity and antidepressant benefits: a meta-analysis of data submitted to the Food and Drug Administration. PLOS Med. 5, 260–268. Kvaale, E., Haslam, N., Gottdiener, W., 2013. The ‘side effects’ of medicalization: a meta-analytic review of how biogenetic explanations affect stigma. Clin. Psychol. Rev. 33, 782–794. Lam, D., Salkovskis, P., 2007. An experimental investigation of the impact of biological and psychological causal explanations on anxious and depressed patients' perception of a person with panic disorder. Behav. Res. Ther. 45, 405–411. Lebowitz, M., 2014. Biological conceptualizations of mental disorders among affected individuals: a review of correlates and consequences. Clin. Psychol. Sci. Pract. 21, 67–83. Lester, H., Tritter, J., Sorohan, H., 2005. Patients' and health professionals' views on primary care for people with serious mental illness: focus group study. Br. Med. J. 330, 1122–1127. Mnich, E., Makowski, A., Lambert, M., Angermeyer, M., vondemKnesebeck, O., 2014. Beliefs about depression – do affliction and treatment experience matter? Results of a population survey from Germany. J. Affect Disord. 164, 28–32. Munizza, C., Argentero, P., Coppo, A., Tibaldi, G., Di Giannantonio, M., Picci, R., Rucci, P., 2013. Public beliefs and attitudes towards depression in Italy: a national survey. PlOS One 8, e63806. Ogden, J., Boden, J., Caird, R., Chor, C., Flynn, M., Khan, K., MacLurg, K., Swade, S., Thapar, V., 1999. ‘You're depressed’; ‘No I'm not’: GPs' and patients' different models of depression. Br. J. Gen. Pract. 49, 123–124. Olfson, M., Marcus, S., 2009. National patterns in antidepressant medication treatment. Arch. Gen. Psychiatry 66, 848–856. Pigott, H., Leventhal, A., Alter, G., Bowen, J., 2010. Effficacy and effectiveness of antidepressants: current status of research. Psychother. Psychosom. 79, 267–279. Pilkington, P., Reavley, N., Jorm, A., 2013. The Australian public's beliefs about the causes of depression: associated factors and changes over 16 years. J. Affect. Disord. 150, 356–362. Prins, M., Verhaak, P., Bensing, J., Van Der Meer, K., 2008. Health beliefs and perceived need for mental health care of anxiety and depression: the patients' perspective explored. Clin. Psychol. Rev. 28, 1038–1058. Read, J., Haslam, N., Sayce, L., Davies, E., 2006. Prejudice and schizophrenia: A review of the ‘Mental illness is an Illness like any other’ approach. Read, J., Cartwright, C., Gibson, K., 2014a. Adverse emotional and interpersonal effects reported by 1829 New Zealanders while taking antidepressants. Psychiatry Res. 216, 67–73. Read, J., Cartwright, C., Gibson, K., Shiels, C., Haslam, N., 2014b. Beliefs of people taking antidepressants about causes of depression and reasons for increased prescribing rates. J. Affect. Disord. 168, 236–242. Read, J., Haslam, N., Magliano, L., 2013. Prejudice, stigma and ‘schizophrenia’: the role of bio-genetic ideology. In: Read, J., Dillon, J. (Eds.), Models of Madness: Psychological, Social and Biological Approaches to Psychosis. Routledge, London, pp. 157–177. Schofield, P., Crosland, A., Waheed, W., Aseem, S., Gask, L., Wallace, A., Dickens, A., Tylee, A., et al., 2011. Patients' views of antidepressants: from first experiences to becoming expert. Br. J. Gen. Pract. 61, e142–e148. Schomerus, G., Matschinger, H., Angermeyer, M., 2014. Causal beliefs of the public and social acceptance of persons with mental illness: a comparative analysis of schizophrenia, depression and alcohol dependence. Psychol. Med. 44, 304–314. Schumann, I., Schneider, A., Kantert, C., Löwe, B., Linde, K., 2012. Physicians' attitudes, diagnostic process and barriers regarding depression diagnosis in primary care: a systematic review of qualitative studies. Fam. Pract. 29, 255–263. Shakespeare W., The Merchantmerchant of Venice, Craig W.J., (Eds.) The Oxford Shakespeare, 1597/ 1914, Oxford University Press; London. Srinivasan, J., Cohen, N., Parikh, S., 2003. Patient attitudes regarding causes of depression: implications for psychoeducation. Can. J. Psychiatry 48, 493–495. Statistics New Zealand, 2012. New Zealand in Profile 2012. 〈http://www.stats.govt.nz〉. Turner, K., Sharp, D., Folkes, L., Chew-Graham, C., 2008. Women's views and experiences of antidepressants as a treatment for postnatal depression: a qualitative study. Fam. Pract. 25, 450–455.
