Introduction
Nonattendance at initial therapy appointments is a pervasive problem, yet the reasons remain poorly understood. This study draws on an account of the motivational processes underlying nonattendance, termed the loss of valued control (LVC) model (Schauman and Mansell, Reference Schauman and Mansell2012) based on perceptual control theory. The LVC model conceptualizes nonattendance as ambivalence due to approach-avoidance goal conflicts between motives to access therapy, whilst anticipating that therapy will conflict with another goal. The model also highlights that conflict with avoidance goals that are part of the psychological problem itself, such as a need to avoid further negative emotional experiences, are most problematic. This study aimed to identify approach and avoidance attitudes, beliefs and goals (hereafter referred to as attitudes) associated with attendance and is the follow-up to a pilot study (Murphy, Mansell, Craven, Menary and McEvoy, Reference Murphy, Mansell, Craven, Menary and McEvoy2013). It was hypothesized that positive and negative attitudes towards therapy would predict increased and decreased attendance respectively, and that an interaction effect would emerge based on competing approach-avoidance tendencies. Informed by the LVC model, anxiety was hypothesized to amplify the effect of negative attitudes on nonattendance, due to an interaction with anxiety-related avoidance goals. Depression was hypothesized to moderate the relationship between approach attitudes and attendance, due to depression-related deficits in approach motivation.Footnote 1
Method
Design and setting
A prospective study was conducted in a low-intensity cognitive behavioural therapy service established under the Improving Access to Psychological Therapies programme in Salford, England. The service inclusion criteria were age 16 years or over and mild to moderate anxiety or depression. Participants were referred from nine general practices from 16 June 2011 to 31 March 2012, mainly serving urban Central Salford, selected on the basis that they were agreeable to participation. Each practice participated for varying periods, with an average of 3.8 months. Receptionists were instructed to offer all patients a questionnaire pack at the time of GP referral to the service. Consenting participants completed the forms in the waiting room and returned them via reception. The research was approved by the North West National Research Ethics Committee.
Measures
The Initial Appointment Questionnaire, IAQ (Mansell, unpublished) measures positive and negative therapy attitudes. It is rated on visual analogue scales from 0 to 100. Previous versions were piloted in student and clinical samples (Murphy et al., Reference Murphy, Mansell, Craven, Menary and McEvoy2013). Scales were also administered to measure depression (Patient Health Questionnaire-9), anxiety (Generalized Anxiety Disorder Assessment-7) and functioning (Work and Social Adjustment Scale). Further data were collected on age, gender, ethnicity, employment and benefits status, Index of Multiple Deprivation and waiting time. Outcome data on first appointment attendance were collected from electronic patient records.Footnote 2
Analysis
Principal components analysis was used to identify the IAQ subscales. Item ratings were standardized and examined as univariate predictors (logistic regression). Standardized IAQ subscale scores were then examined as predictors. Moderation analysis was used to test the interactions.
Results
Sample characteristics
Ninety-six people completed the questionnaires (60% female, 96% of white ethnicity, 54% employed). Sixty people attended the first session; thus the nonattendance rate was 37.5%. No demographic (age, gender, ethnicity, employment status, Index of Multiple Deprivation) or distress measures (PHQ-9, GAD-7, WSAS), nor waiting times differed significantly by attendance status.
IAQ items
One item predicted attendance (OR 0.5 (95% CI 0.3, 0.8), p = .007, inverse association): “I would feel vulnerable if I disclosed something very personal I had never told anyone before to a therapist”. After controlling for age, employment and distress the finding remained significant (OR 0.5 (95% CI 0.3, 0.9), p = .01).
IAQ subscales
The IAQ items loaded onto two subscales, thus supporting the intended factor structure. Cronbach's alpha for the subscales were 0.90 and 0.89. The subscale scores did not predict attendance (positive attitudes OR 1.1 (95% CI 0.7, 1.7), p = .58; negative attitudes OR 0.7 (95% CI 0.5, 1.1), p = .14) and there was no interaction between them (OR 0.9 (95% CI 0.6, 1.4), p = .56).
Interactions between attitudes and distress
Negative attitudes and anxiety did not interact to predict attendance (OR 1.1 (95% CI 0.7, 1.9), p = .58), but positive attitudes and depression did interact (OR 0.3 (95% CI 0.1, 0.7), p = .007). Positive attitudes predicted attendance when depression was low. At low depression (median split), the positive attitudes OR was 4.1 (95% CI 1.5, 11.7, p = .008). At high depression, the positive attitudes OR was non-significant 0.6 (95% CI 0.3, 1.3, p = .18). Examination of items at low depression showed the largest effects for: “I am very interested in examining what I think about” (OR 6.6 (95% CI 1.9, 23.3), p = .003); “If I share my thoughts and feelings with another person it will help me to get to know myself better” (OR 3.9 (95% CI 1.4, 10.7), p = .009); and “It is important to me to try and understand what my feelings mean” (OR 3.5 (95% CI 1.4, 8.8), p = .009). Two of these items were adapted from the Self-Reflection and Insight Scale (SRIS) (Grant, Franklin and Langford, Reference Grant, Franklin and Langford2002).
Discussion
Main findings
Endorsement of an item measuring concern about self-disclosure predicted nonattendance. Positive attitudes towards therapy, particularly those measuring motives for self-reflection, predicted increased attendance among less depressed individuals.
Strengths and limitations
Few psychological studies have examined first appointments (Sheeran, Aubrey and Kellett, Reference Sheeran, Aubrey and Kellett2007), and the prospective design adds external validity. Data were not available for calculation of a true questionnaire return rate; however, a conservative estimate based on all referrals shows the proportion completing forms was 18%. Therefore the sample may have been selective and may not have been representative of the whole cohort, thus potentially limiting its generalizability. However, the service nonattendance rate of 35.3% is similar to the sample rate of 37.5%, and there were no age or gender differences (60 vs. 63% female, p = .6; 35 vs. 36 years, p = .6) suggesting some comparability between the sample and the overall cohort.
Findings in relation to previous studies
Self-disclosure concern
The item “I would feel vulnerable if I disclosed something very personal I had never told anyone before to a therapist” was adapted from the Disclosure Expectations Scale, which was previously associated with help-seeking intentions (Vogel and Wester, Reference Vogel and Wester2003). Also, anticipating feelings of shame, embarrassment and exposure predicted nonattendance (Sheeran et al., Reference Sheeran, Aubrey and Kellett2007), which may represent similar concerns.
Positive attitudes
Positive attitudes predicted attendance among less depressed people. The main items were those measuring self-reflection motives. This is consistent with the pilot study (Murphy et al., Reference Murphy, Mansell, Craven, Menary and McEvoy2013), which found a main effect for a self-reflection item. Therefore motives for self-reflection may capture a therapy approach-goal.
Conflict
Anxiety did not amplify negative attitudes on attendance as predicted. More precise interactions require examination in future. For example, self-disclosure concerns might interact with social anxiety more specifically, as the latter involves motives to avoid embarrassment. The hypothesis that positive and negative attitudes would interact, based on approach-avoidance conflicts, was also rejected. Conscious awareness of ambivalent attitudes might not measure lasting conflict. Perceptual control theory describes how awareness allows reorganization processes, such as a decision to prioritize a goal, whereas unconscious conflict is more problematic.
Depression
Highly depressed individuals who endorsed positive items were not more likely to attend. This may reflect deficits in approach motivation processes, e.g. difficulty in making goal-action plans, in disengaging from unmet goals, and pessimism (Trew, Reference Trew2011). Therefore if people with more severe depression have planning difficulties and are more pessimistic about achieving their goals, then their attitudes may not translate to improved attendance.
Clinical implications
Openly talking with clients about their disclosure concerns (Vogel and Wester, Reference Vogel and Wester2003) or providing brochures to allay misconceptions may help. Promotional material that highlights opportunities for self-reflection might encourage attendance. Regarding individual concerns, referrers and therapists could ask clients what they personally want to have control over in therapy (Schauman and Mansell, Reference Schauman and Mansell2012). For more depressed individuals, encouraging them to make clear plans or implementation intentions regarding attendance (Sheeran et al., Reference Sheeran, Aubrey and Kellett2007) or exploring pessimism about pursuing therapy goals may be of benefit.
Conclusion
These results show that self-disclosure concerns contribute to therapy avoidance and that having goals for self-reflection may represent approach-motivation for therapy, although the latter has less impact among more highly depressed people.
Comments
No Comments have been published for this article.