In contrast with earlier studies based on traditional OCD treatments, the findings from this study indicate that a treatment protocol developed from the cognitive behavioral model of compulsive hoarding can significantly reduce hoarding behaviors including clutter, excessive acquisition, and difficulty discarding. Effect sizes for specific hoarding measures were very large (η2p of .49 and .67); even the effect for global improvement ratings (η2p = .14) was considered large, although there was no significant effect of time for that variable (possibly due to the small sample size). The findings further suggest that poor outcomes observed in the treatment of hoarding are associated with problems completing homework.
In the present study, treatment differed from standard therapy for OCD in several ways. Exposure was de-emphasized, with greater focus given to motivational interviewing, skills training for organizing and problem solving, and cognitive modification of beliefs about possessions. Particularly unique to this treatment were the frequent off-site sessions, in which therapists guided patients through sorting, discarding, and resisting acquiring. Half of the patients appeared to be clear treatment responders on the CGI-I and 60% were clinically significantly improved on the SI-R. Although this represents an improvement over previous research, it is clear that outcomes in hoarding are attenuated compared to other conditions such as OCD. We note as well that the long-term outcomes of this treatment are still unknown; a larger waitlist controlled trial is currently underway that will provide more information on immediate and longer-term outcomes.
Perhaps more than related conditions such as OCD, compulsive hoarding presents several clinical challenges that must be addressed in CBT. One such challenge is limited insight into the severity of hoarding behavior. Patients with compulsive hoarding are typically rated as showing poorer insight than are patients with OCD (Frost, Krause, & Steketee, 1996
), and social service providers report that most elderly clients with serious hoarding show little insight into their problem (Steketee, Frost, & Kim, 2001
). Related to insight is the low motivation and resistance to treatment exhibited by many hoarding patients (Christensen & Greist, 2001
; Steketee et al., 2001
). Patients with compulsive hoarding may be more likely than are OCD patients to refuse or drop out of treatment prematurely (Ball, Baer, & Otto, 1996
; Mataix-Cols et al., 2002
), and many individuals with hoarding problems refuse treatment unless pressured by family, social service workers, or health department officials (Frost, Steketee, & Williams, 2000
). Even after treatment has been initiated, motivation appears to wax and wane (Hartl & Frost, 1999
), with a notable lack of adherence to treatment (e.g., completion of homework assignments) (Christensen & Greist, 2001
; Steketee et al., 2000
). This was apparent in the present trial: the median homework adherence rating was 4, indicating that the patient “did 26–50% of the homework or its equivalent.” Reports from therapists in this study indicated that nearly all of the patients had difficulty completing homework assignments, yet most had little difficulty working directly with the therapist on sorting and discarding tasks. Homework adherence was clearly associated with treatment outcome, with greater adherence predicting greater reductions in hoarding severity. Thus, it appears that maximizing adherence to homework assignments is a central challenge for clinicians treating hoarding patients. The principles and strategies of motivational interviewing (Miller & Rollnick, 2002
), which have been used successfully for treatment-resistant OCD patients (Maltby & Tolin, 2005
), may be particularly useful in treating patients who hoard, and are likely to be needed throughout treatment whenever motivation wanes.
The extent and effect of comorbidity also merits clinical consideration. In the present sample, all of our hoarding patients met DSM-IV-TR
criteria for at least one comorbid Axis I disorder; depression and anxiety disorders were particularly common. Axis II comorbidity has also been reported in the majority of hoarding cases in other samples (Seedat & Stein, 2002
). Thus, many hoarding patients may require additional pharmacological and cognitive-behavioral treatment strategies aimed at reducing comorbid symptoms, particularly when they exacerbate hoarding or interfere with its treatment (e.g., a depressed patient who is too fatigued to comply with homework assignments, a socially phobic patient who is too fearful to allow others into his/her home, contamination fears that affect ability to touch and sort possessions).
The duration of treatment also merits additional consideration. As seen in , SI-R scores appeared to decrease in a linear fashion over the course of treatment. Although the SI-R subscales decreased significantly by mid-treatment, the individual subscales did not decrease significantly until post-treatment, suggesting that a longer course of treatment might be helpful.
As an initial exploratory study, the present study is limited by its small sample size and open-trial design. A further limitation is the completion of clinical ratings by the therapist, a decision based in part on patients’ reluctance to allow multiple observers into the home. This concern is particularly noteworthy in the homework adherence ratings, in which therapists’ observations of progress could have influenced their subjective ratings of adherence. Because of the exploratory nature of the study, we allowed a great deal of flexibility in the treatment protocol (e.g., 7–12 month duration, “marathon” sessions for 2 patients). This is both a strength and a limitation: on one hand, flexibility in manualized treatment delivery has been cited as critical (e.g., Kendall, Chu, Gifford, Hayes, & Nauta, 1998
); on the other hand, such variability makes it difficult to ascertain necessary and sufficient parameters of the treatment, particularly with a small sample size. Such issues await clarification in a larger, controlled study. An additional limitation is the fact that all patients were women. The demographics of compulsive hoarding are unknown, although we note that in our larger, non-treatment study, women comprise approximately 75% of the sample (Frost, Steketee, Tolin, & Brown, 2006
Additional research is needed to determine whether CBT yields results that are superior to no treatment, wait-list, or control treatments. Furthermore, the treatment used in the present study might not be feasible for many clinicians, whose ability to travel to patients’ homes or acquisition sites is limited. Future dismantling studies should examine the necessity of out-of-office sessions in treating patients who hoard.