In this Issue...
For those of us who work with people in recovery, an attitude of gratitude and wonder is not limited to Thanksgiving. We are regularly blessed with the opportunity to witness miraculous transformations as people shed the overwhelming fear that is part of the fabric of substance dependence and emerge with a renewed capacity to experience serenity and joy.
All of us at The SASSI Institute are gratified by the comments you have shared with us regarding the SASSI-3 and the new support materials and by the expressions of appreciation we received. As we worked on it, we felt we were doing a good job; we were excited by the power of the research data and by the quality of the support materials we were developing with the help of our skilled editorial consultants. The real test, of course, was whether or not you found the new materials to be useful in your work. I was especially pleased that many of you told us that the video entitled "Connecting Clinician & Client: SASSI Interpretation and Feedback" has helped you in your efforts to get clients into treatment and in supporting ongoing recovery. Thank you for your questions, suggestions and expressions of appreciation.
Another ongoing source of gratitude is that in a climate of shrinking resources for treatment, correctional programs appear to be developing better identification and treatment for substance-dependent inmates. The fact that such efforts may be motivated by the recognition that treatment costs are more than offset by reduced recidivism doesn't change the fact that lives are being saved and transformed. I am also pleased to note the emergence of innovative programs for identification and treatment of substance misuse in hospitals and other medical settings.
Working with substance dependent people is taxing. Often, during the initial stages of the treatment process, before clients experience the profound relief that comes with "letting go," they approach treatment providers with mistrust and anger. Fear and setbacks are also part of the process of recovery. I am thankful that there are so many dedicated human service workers who recognize the need to stop the cycle of pain that is the product of substance abuse. I value the opportunity to support your work in identifying and treating substance dependent people.
A peculiar development in the evolution of health care plans is that substance abuse is singled out for treatment limitations that are not imposed on other conditions. This is particularly mystifying given that the presumed rationale for the policy is cost saving. Does it really save money to impose limits on treatment for substance dependence since substance dependence is a cause of other disorders such as pancreatitis, cirrhosis and cardiovascular disease?
There have been a number of studies that demonstrate that treatment for alcohol and other drug dependence actually reduces overall health care costs. The California Drug and Alcohol Treatment Assessment General Report (1994) presents the results of a follow up study of 1,900 persons in substance abuse treatment in 1992. Generally, treatment was found to be cost effective, yielding a $7 return for every dollar invested. There were specific cost benefits in the area of health care. Substance abuse treatment was associated with a 36% reduction in hospitalizations for physical health, a 58% reduction in hospitalizations for drug overdose, and a 44% reduction in hospitalizations for mental health. Addictions treatment reduced doctor visits (-14%), emergency room visits (-38%), and hospital days (-25%). Clearly, substance abuse is a factor underlying a wide range of disorders, and substance abuse treatment therefore reduces overall health care costs.
A second question arises: How much is saved in the short run by imposing special limitations on substance abuse treatment? A recent study, "Premium Estimates for Substance Abuse Parity Provisions for Commercial Health Insurance Products" (Melek and Pyenson, 1997) addressed that issue. The authors concluded that based on their estimates "… a full and complete substance abuse parity provision would increase 'composite' per capita health insurance premiums (aggregate premium increases across fee-for-service, PPO/POS and HMO/EPO plans) by 0.5% or less than $1 per member per month." Singling substance abuse treatment for special limitations not only does not make sense in the long run, it does not produce significant savings in the short run.
The costs of substance dependence are both broad and deep. Society as a whole pays the price in crime, health care costs, and lost productivity. Significant others pay a price in the emotional pain that accompanies living with and loving a dependent person. The dependent person pays the price of lost opportunity to experience the blessings of life. It is unfortunate that quality of life must be relegated to terms of a balance sheet, but it seems apparent that we could increase our "profits" and cut our "losses" by expanding the availability of substance abuse treatment.
Increased Accuracy of SASSI-3: The Importance of New Decision Rules
For the clinical interpretation in this issue of News and Reports, we will be using a SASSI-3 profile that was called in to our clinical consultation service. We thought it would be useful to select a case where the SASSI-3 decision rules yield a positive classification, and the SASSI-2 rules result in a test negative.
The profile is of a 37-year-old male parolee who was released from prison one month prior to the administration of the SASSI. The results indicate that he seemed to understand the items and responded in a meaningful way (RAP=0). There is some evidence, however, that he approached the assessment in a moderately defensive manner (DEF=9). Despite his defensive responding, the SASSI-3 decision rules classify the client as having a high probability of having a substance dependence disorder (Decision Rule 9). Notice that this decision rule combines the results of the face-valid scales (FVOD=8) and the subtle scales (SAT =3, DEF=9, SAM=8) resulting in a test positive. This decision rule was developed to maintain the high level of test identifications that were part of the SASSI-2 while minimizing the likelihood of false positives on the SASSI-3.
As we begin to examine the profile, it is important to recognize that this client's defensiveness may indicate some limitations in his ability or willingness to acknowledge symptoms relevant to substance misuse. His defensiveness may also be related to situational factors or circumstances surrounding the evaluation. However, his responses are significantly similar to substance dependent individuals who were instructed to minimize or conceal any evidence of alcohol or other drug abuse in their lives (SAM=8). There is some risk, then, that he may have understated or misrepresented the degree to which his substance misuse has led to behavioral problems. He may externalize blame to other people, places or things, characterize himself as a victim and have difficulty accepting responsibility for his behavior. Nonetheless, he does acknowledge enough current and/or past problems related to other drug use (FVOD=8) that, when combined with other scale scores, significantly contribute to the test-positive results.
The client is likely to view himself as dissimilar to substance dependent individuals who acknowledge behavioral problems commonly associated with addictions (OAT=3). He also appears to be responding in a fashion that is different from that of people with extensive criminal histories (COR=3). Although the COR scale is not a measure of criminal activity, the risk for ongoing legal problems or involvement in the criminal justice system may be low.
The results of the SASSI indicate that this individual is likely to have a substance dependence disorder. Since the client may be limited in his ability or willingness to acknowledge the full nature of his substance misuse, it will be important to rely on external data in addition to the SASSI results in formulating a diagnosis and treatment plan.
Initial treatment efforts should be directed at helping the client to increase awareness of how substance use is creating problems in his life. This is best done in the context of a trusting relationship where the client feels supported and validated. For example, if historical information conflicts with the client's self-report on the face-valid items, it may be useful to join the client in exploring these discrepancies. This can be done by inviting the client to share a more detailed accounting of the experiences he acknowledged on the FVOD items. By assuming an active listening role and encouraging him to openly discuss problems related to his drug use, the full nature of his substance abuse problems may become clearer to him. This process may also help establish greater rapport and trust, thus leading to decreased defensiveness and increased ability and willingness to accept responsibility for the problems he is experiencing.
Effective treatment planning should also include cognitive, didactic interventions that provide the client with accurate information regarding substance dependence. Random toxicological screening should also be considered as a way of ensuring abstinence in an objective, non-judgmental manner. Although the client may not respond well to community self-help groups initially, it would be useful to encourage him to keep an open mind. This would be an excellent way for him to gain support in the process of building a social network that is more conducive to living a drug-free life style. Suggesting that he attend several open meetings during a period of time and then processing the experience may be an effective way to provide encouragement.
The accuracy of an assessment tool is typically described in terms of the percent of people who are correctly identified. It is not possible to be absolutely certain that a specific test result is "correct." For example, the SASSI agrees with independently derived assessments approximately 93% of the time. However, a given SASSI result may be one of the 93% overall correct identifications or it may be one of the 7% overall errors in classification. Therefore, it is important to learn as much as possible about factors that might affect the accuracy of any particular SASSI result
The procedures that were used in developing the SASSI were designed to minimize the impact of extraneous variables such as age, gender, ethnicity and drug of choice. The SASSI items were selected from a pool of thousands of items by administering approximately two hundred at a time to a series of small criterion groups. In other words, a set of items was administered to one group of dependent people and one group of controls. The items that differentiated the two groups were retained. Additional test items were then added to the ones that were found to be effective, and the new set of items was administered to a new criterion group of dependent people and a new group of controls. Again, items that discriminated between the two groups were retained, and additional new items were added. This procedure was followed until there was an adequate supply of items that had gone through repeated testing with a variety of criterion groups and a variety of control groups. This laborious procedure produced a final set of items that was effective in differentiating a broad array of substance abusers from controls. Additionally, the research participants who were recruited to provide normative data were recruited from census tracts in a large metropolitan community that approximated characteristics of the total population. The impact of extraneous data was further minimized by adjusting the decision rules on the basis of gender and by developing an adolescent version separate from the adult version.
The efforts to minimize the impact of extraneous variables on the effectiveness of SASSI items continue to pay off as the SASSI is used and updated. An examination of the possibility of racial bias on the SASSI, the MAST, and the CAGE revealed there were no effects of race on any of those instruments (Knack, 1993). The available data on the SASSI-3 indicate that gender, race/ethnicity, occupational status, marital status, age, and education do not have a significant impact on accuracy. Table one (below) reveals consistent accuracy rates across racial/ethnic groups.
Generally, the SASSI is robust to differences in demographic variables and drugs of choice. The SASSI Institute continues to work with SASSI users to gather data on the accuracy and utility of the SASSI across a wide a range of research participants. We are particularly cognizant of the importance of cultural factors and their impact on clients' responses to the entire assessment process.
SASSI-3 Accuracy In Identifying Substance Use Disorders as a Function of Ethnic Group
|
Data Source |
Prevalence of Disorder |
Sensitivity |
Specificity |
PPP |
NPP |
Overall Accuracy |
|
Combined Sample (n = 839) |
80 |
94 |
94 |
98 |
80 |
94 |
|
Caucasian (n = 428) |
76 |
92 |
96 |
99 |
79 |
93 |
|
African American (n = 151) |
86 |
93 |
90 |
98 |
68 |
93 |
|
Hispanic (n = 50) |
78 |
92 |
91 |
97 |
77 |
92 |
|
Other/Unknown Ethnicity (n = 210) |
81 |
98 |
92 |
98 |
92 |
97 |
The available data indicate that the accuracy of the SASSI-3 is not significantly affected by ethnicity.
Note. All figures are percentages.
A logistic regression analysis revealed that racial/ethnic group did not have a statistically significant impact on the accuracy of the SASSI-3
(Improvement X2 = .13, p = .99)