Help-Seeking Behavior and Illness Detection
Only a minority of people who have a diagnosable mental disorder receive professional help. This has been repeatedly shown across major epidemiological surveys (). In both the European Depression Epidemiological Survey (DEPRES) () and in the Ontario Health Survey (), fewer than 50% of detected cases of major depressive disorder in the community had been identified by a health care professional, and fewer than half of those identified had received psychiatric treatment. Similarly, only 28% of those in the Epidemiologic Catchment Area study who met DSM-III-R (American Psychiatric Association 1987) criteria for a psychiatric disorder during the previous year had sought help from a mental health professional, and only 21% of those diagnosed with a mental disorder in the last year in the National Comorbidity Survey had sought help (). Overall, a systematic literature review on treatment-seeking rates for individuals with various types of depressive illness ranged from 21% to 61% (), yet very few investigators have measured the effect of personality variables on treatment seeking. More often demographic and clinical variables have been examined in epidemiological surveys (). For example, being male, living alone, and living in a large urban community adversely affected help-seeking behavior. Among men, low neuroticism scores and an external attributional style have also been linked to a reduced rate of help-seeking behavior ().
Personality and Symptom Presentation
The relationship between personality dimensions and symptom presentation has also been evaluated. Using the TCI () to assess personality dimensions and the Center for Epidemiologic Studies Depression Scale (CES-D) () to evaluate depressive symptoms, Grucza and colleagues () examined relationships between various depression parameters and the four personality dimensions, including symptom severity, comorbidity, and suicidality (). In brief, Grucza et al. () reported that 1) high harm avoidance was associated with panic attacks and total severity of depressive symptoms; 2) high novelty seeking alone was associated with suicide attempts, inability to concentrate, and alcohol use; 3) high harm avoidance with high novelty seeking was associated with the number of psychiatric admissions; 4) high reward dependence with high persistence was associated with restless sleep; 5) high reward dependence with low persistence was associated with appetite loss and low energy; and 6) high harm avoidance with low novelty seeking was associated with a lack of positive affect.
Treatment Preference and Selection
Bedi and colleagues () addressed two issues relevant to treatment effectiveness for major depressive disorder in primary care. Although the study was primarily designed to compare the effectiveness of counseling and pharmacotherapy under randomized controlled conditions, a secondary aim was to compare outcomes in the group who agreed to be randomly assigned and the group who opted for personal selection of treatment. No differences were observed across patient variables in the self-selection and randomized groups, including scores on the Eysenck Personality Inventory. Similarly, Ward and Colleagues () found no differences in outcome between self-selection and randomized subjects in a comparison of nondirective counseling, cognitive-behavioral therapy, and usual general practitioner care for the treatment of major depressive disorder. In summary, although more research is needed, personality dimensions do not appear to influence treatment preference or selection.
Compliance With Treatment Recommendations
In a review of patient adherence or compliance during randomized controlled trials in the treatment of major depressive disorder, Pampallona and associates () concluded that one in every three patients did not complete the investigational treatment. Even higher dropout rates have been reported in other settings. Fifty-three percent of patients who were treated in primary care practices with antidepressant medication had discontinued treatment before the end of 12 weeks (). Rates of discontinuation and the reasons provided in this study were remarkably similar to those reported by Demyttenaere and colleagues () across family practice units in Europe in Table Reasons cited for antidepressant medication discontinuation:
|Feeling better||35%||Feeling better||55%|
|Adverse events||30%||Adverse events||23%|
|Other reasons||17%||Fear of dependence||10%|
|Lack of effect||15%||Lack of effect||10%|
Side effects are one reason for discontinuation. Using the Eysenck Personality Inventory, Davis and colleagues () investigated the influence of personality variables on side-effect reporting in a group of healthy volunteer subjects during a randomized placebo-controlled trial involving the reversible inhibitor of monoamine oxidase-A, moclobemide. The authors confirmed a positive relationship between neuroticism score at baseline and subsequent side-effect reporting. By the end of the study the relations between neuroticism and side-effect reporting had increased in the placebo group and declined in the moclobemide group. In summary, these results underscore the interindividual variability in side-effect reporting, which is usually neglected in clinical trials. The following case illustrates the importance of individual differences in side effect reporting:
Ms. A, a 25-year-old patient with major depressive disorder, panic disorder, and comorbid borderline personality disorder, was repeatedly unable to complete even a minimum trial of numerous antidepressants. After agreeing to participate in a randomized placebo-controlled trial of a novel antidepressant with potentially fewer side effects, the patient arrived at the emergency room 2 days later. She reported severe nausea and panic symptoms, which she attributed to the new medication. She discontinued the trial, and the double-blind medication turned out to be placebo. This experience was discussed with Ms. A in terms of her expectancy effects and fear of losing control when she took any medication. It became a defining experience and allowed her to be rechallenged with gradually increasing doses of a previously intolerable antidepressant. She responded to this medication.
Despite an increased awareness that noncompliance with antidepressant treatment is a major contributing factor to suboptimal therapeutic outcomes (), there has been very little empirical research conducted in the area of personality and relationship variables and compliance with antidepressant medication (). Sirey and colleagues (), using the Inventory of Interpersonal Problems () to screen for personality pathology, reported that antidepressant compliance was associated with the absence of personality pathology. However, the Inventory of Interpersonal Problems has not been validated against more frequently used measures. Ekselius and colleagues () used the Karolinska Scales of Personality () to investigate the relationship between personality traits and compliance with antidepressant medication. Ekselius et al. reported that medication noncompliance, defined as out-of-range plasma levels of medication, was associated with elevated scores on the sensation-seeking subscale of this instrument. Similarly, Wingerson and colleagues () found early discontinuation from clinical trials was positively associated with the novelty-seeking dimension on the Tridimensional Personality Questionnaire in patients with panic disorder and generalized anxiety disorder.
Although other researchers have proposed that personality factors may be associated with medication noncompliance in psychiatric populations (), there are virtually no evaluations. On the basis of the combined use of the Minnesota Multiphasic Personality Inventory (MMPI-2) clinical and content scales and the NEO PI, Stein and Hackerman () proposed that low conscientiousness scores and high neu-roticism scores, combined with the MMPI-2 content scale “work interference” and “negative indicators,” are indicative of the potential for poor treatment compliance. However, as these conclusions were drawn from a single case study, a prospective controlled approach is required to test the hypothesis.
Demyttenaere () also recognized anecdotal and theoretical literature addressing personality factors and compliance, including psychodynamic literature by Book (), who discussed the implications of projection, denial, identification, and transference in understanding medication compliance in psychiatric patients. Demyttenaere noted that noncompliance may reflect an individual’s attempt to deny having depression or to avoid identifying with another family member who may also have the disorder.
Cohen and colleagues () evaluated the relationship between personality (FFM) and compliance in depressed outpatients. Compliance was measured using an electronic bottle-cap device designed to record openings and closings of medication bottles, the Medication Event Monitoring System (MEMS). High extraversion was found to be a significant negative predictor of compliance to antidepressant medication. At the level of NEO facets, activity (extraversion) and feelings (openness) were negatively correlated with compliance, whereas modesty (agreeableness) was positively associated with compliance.
Although excitement seeking as measured by the NEO-PI-R was not significantly related to medication compliance, the elevation in extra-version is consistent with the findings of Ekselius and colleagues (), who found that noncompliance with antidepressant medication was associated with impulsive sensation seeking. The extraversion equivalent, novelty seeking, in patients with panic disorder and generalized anxiety disorder was also associated with early discontinuation in clinical trials ().
The feelings facet of openness, characterized by excitable, spontaneous, insightful, imaginative, affectionate, talkative, and outgoing attributes, was also inversely related to compliance. Taken together, the characteristics associated with feelings coupled with the characteristics associated with activity, particularly being energetic, hurried, quick, excitable, and spontaneous, are somewhat related to characteristics of excitement seeking. From a clinical perspective, this suggests that people who have elevated levels of extraversion and openness are more likely to experience early (although potentially transient) symptomatic improvement followed by a reduction in medication compliance. This is illustrated in the case of Mr. T.
Mr. T worked in the sales department of an office supply company. He was a particularly gregarious individual. Over a 6-year period, he had been seen for approximately 2 months at three different psychiatric clinics. On each occasion he presented with typical symptoms of major depression. For him, the most prominent symptom was losing interest in his normally active social life.
There was also a pattern to his antidepressant usage. On each occasion, he responded very quickly, often during the first week. He would regain interest in social activities and increase his success in sales. Unfortunately, this is also the time when he would skip his antidepressant medication for days at a time and eventually stop altogether.
Only after a detailed discussion about the potential negative relationship between his personality style and medication compliance did Mr. T agree to comply with antidepressant therapy and he has remained well for the past 2 years.