Overall, the study findings reaffirm previous research showing MA dependent individuals have high rates of psychiatric symptoms, particularly women. This paper adds to the current literature in two respects. First, we document symptoms from a wider variety of disorders that have generally been understudied or underspecified. These include somatoform disorders, such as somatization and hypochondriasis, bulimia and even some subtypes of anxiety disorders (e.g. agoraphobia, social phobia and panic disorder). The importance of investigating these disorders is particularly evident in the large proportions of women who met their screening criteria. For example, the PDSQ scales that had the largest proportion of women meeting the screening criteria was social phobia (72%) followed by somatization (64%), both of which have not been well studied among MA dependent women.
A second way this paper adds to the current literature is by studying the correlates of MA dependence outside the context of formal treatment settings. Examining the long term course of drug use disorders and recovery outcomes is receiving increasing emphasis in recent years. SLHs are examples of resources that can be used to support sustained recovery in the community for individuals who have attended treatment programs as well as those looking for alternatives to formal treatment. However, these facilities need to be studied more thoroughly because residents there may have distinct differences from their counterparts in treatment.
Psychiatric Symptom Differences by Gender
The findings from our study are consistent with previous reports documenting that women with MA dependence have more significant problems with psychiatric symptoms than men.19–22, 26
On 10 of the 11 PDSQ scales that we examined women had higher proportions meeting the screening criteria than men. (See the discussion below on the unexpected findings for depression). Eight of the 11 scales were statistically significant (p<.05) or trended (p<.10) in the direction of more symptoms reported by women.
The differences in psychiatric severity by gender were for the most part specific to those with MA dependence. It was interesting that most gender differences in psychiatric symptoms disappeared when we compared them among non-MA dependent residents. Thus, there is something about MA dependence, not drug dependence generally, that is important in relation to psychiatric symptom differences between men and women residing in SLHs. The reasons and causal mechanisms for these differences are important areas for further research.
The study findings also indicate that psychiatric symptom differences between MA dependent and non-MA dependent individuals appear to be more prominent for women. There were few significant differences in psychiatric symptoms between men with and without MA dependence. The one exception was depression, which trended in the opposite direction of what we expected. Men with no MA dependence had larger proportions meeting the cutoff than men with MA dependence. We also found that larger proportions of men without MA dependence met the cutoff for anxiety disorders, although the difference did not reach a statistical trend.
The findings for women were different. Psychiatric symptoms were generally more prominent among those with MA dependence. For example, somatoform disorders and psychosis had trends for larger proportions meeting the cutoff among MA dependent women. Although they did not reach statistical significance, women with MA dependence had larger proportions meeting the cutoffs for anxiety and bulimia disorders than women with no MA.
It is unclear why both men and women who were not dependent on MA trended toward more symptoms of depression than those who were MA dependent. It is also unclear why there were no differences between MA dependent men and women on the proportion meeting the cutoff for depression. Both of these finding are not consistent with previous research. 20–22, 26
One factor could be the SLH setting, which is different from the treatment settings where most of this research has occurred. The point in the addiction to recovery cycle that individuals enter a SLH may result in a different presentation of depression relative to treatment programs. Most of our sample had been in treatment during the 6 months prior to entering the SLH. It is possible that those experiences resulted in more sustained reduction of depression symptoms than other symptoms, such as anxiety or psychosis. Another factor could be the use of the PDSQ scale to assess depression among MA dependent individuals. To the best of our knowledge the PDSQ has not been used to assess depression among MA dependent sample and might have less validity than other measures. Finally, the most common substance use disorder in in the non-MA comparison group was alcohol, which has been shown to be associated with dysthymic mood. That might account for the large proportion on non-MA residents meeting the cutoff for depression.
Our study found that residents with MA dependence entering SLHs had high rates of psychiatric symptoms despite the fact that nearly all had received drug treatment services at some point in their lives and 68% had been admitted to a residential treatment program over the past 6 months. One implication of the widespread psychiatric problems that we found among SLH resident is that the treatment services they received earlier did not substantially mitigate their psychiatric issues. It was noteworthy that relatively few (about 12%) had attended any type of outpatient psychiatric treatment the 6 months before entering the SLHs although somewhat more (a third) had received some type of psychiatric medication.
Residence in SLHS among this population should therefore address how residents will manage psychiatric symptoms which could threaten sustained sobriety. In part, this could be achieved by referrals to psychiatric treatment services. However, the SLHs themselves might develop ways of decreasing or managing these symptoms. This might be achieved through workshops for residents about psychiatric symptom management; additional training for house managers and operators; or greater collaboration with 12-step recovery groups that address both addiction and mental health issues (i.e. dual anonymous groups).
Because types of psychiatric symptoms varied by gender it might be wise to consider gender specific interventions. While relatively large proportions of both men and women reported symptoms related to social phobia and generalized anxiety disorder, there were large differences in other areas, such as somatization disorder and bulimia. Symptoms related to bulimia was nearly absent among men, yet nearly a quarter of the women reported some symptoms of bulimia. Somatization disorder among women was more than double the proportion among men and had the second highest proportion meeting the PDSQ screening criteria. For men somatization was fifth in terms of proportion meeting the screening criteria.
There are a number of limitations that are important to note. First, as a screening instrument the PDSQ does not identify DSM IV disorders. It only indicates the existence of some of the symptoms related to disorders and suggests the need for further assessment. Second, we do not know how these disorders play out over time. The assessments were conducted as a cross sectional analysis at the time of entry into the SLH. Third, the PDSQ is not designed to tease out the extent to which psychiatric symptoms are a cause or consequence of MA use. Finally, our N was small, especially for the group of MA dependent women (M=25). Thus, larger studies of this population are needed.