Letters

In Reply: We thank Mr. Kanter and Professor Liberman for their thoughtful letters. We agree with most of their comments, but we provide a few clarifications.

Kanter questions our assertion that SSA beneficiaries with severe mental illness have "excellent potential to return to competitive employment." He concludes that "this rate of employment hardly demonstrates that the services they received were able to overcome the disincentives for employment in the SSA disability system." We offer these two clarifications: the goal of individual placement and support is competitive employment, not an exit from the disability rolls, and the most beneficial outcomes from this intervention are typically concentrated in a subset of persons enrolled in these services.

The main point of both letters concerns the huge barriers preventing SSA beneficiaries from leaving the disability rolls. We certainly agree on this point, as we tried to stress in our article. Kanter observes that avoidance of SSA benefits makes competitive employment more "achievable." Clearly receipt of SSA benefits makes competitive employment less likely. Thus in our discussion we note the value of early intervention programs aimed at preventing individuals from ever entering the disability system as the best way to avoid the "benefits trap." We also agree with Kanter on the importance of benefits counseling. Benefits counseling is a core principle of the individual placement and support model ( 1 ).

Both letters emphasize the fact that the actual gains from supported employment reported in the article are modest. To clarify our findings, we offer these supplemental statistics. First, the average time to first competitive job in individual placement and support studies is five months ( 1 ). Thus the follow-up period for judging job duration for our combined study sample is roughly 13 months. Probably the best way to characterize our findings is that a subgroup benefited most from individual placement and support by maintaining at least six months of competitive employment during the study period: 25.9% of SSI beneficiaries receiving the intervention compared with 5.5% of the SSI control group, and 34.5% of SSDI beneficiaries compared with 4.4% of the SSDI control group. Thus these comparisons suggest an approximately five- to sevenfold advantage for individual placement and support. On the basis of findings from long-term follow-up studies that showed a sizeable proportion of individual placement and support clients with steady employment after a lifetime of sporadic employment ( 2 ), we speculate that this best-outcome group started a journey toward a lifelong pattern of employment. Long-term outcomes for individual placement and support clients are much better than the inert and dismal employment rates for most individuals who do not receive evidence-based vocational services.

Both letters point to the need for further improvements in the individual placement and support model, and we wholeheartedly agree ( 2 ). Among other things, we need more effective methods to reach persons who are disenchanted with the service system and not hopeful of ever working again, which probably includes a majority of persons currently receiving SSDI and SSI ( 3 ).

It is commonly believed that the barriers to implementing evidence-based supported employment preclude broad dissemination, but we are cautiously optimistic, even in the face of underfunded service systems and other intractable problems. Liberman cites a Dutch study of individual placement and support to illustrate implementation problems with the model. However, U.S. findings for implementation are more encouraging. A recent national study showed an 89% success rate in implementing high-fidelity supported employment in new sites, mostly within one year ( 4 ). The spread of supported employment through a learning collaborative in nine states and the District of Columbia ( 5 ) has shown that creativity can often overcome the systems barriers Liberman notes.

Most psychosocial interventions show modest improvements in this population. It is critical to recognize the significance of those that have large and tangible influences. Readers can draw their own conclusions as to whether the findings are clinically meaningful and warrant the promotion of supported employment.