DSM-III-R and persons with dual diagnoses: conceptual issues and strategies for future research.

Sturmey (1995) looked at how DSM-III-R labels mental health problems in people with severe or profound intellectual disability. The author argued that the manual's questions do not fit people who cannot speak or show feelings in typical ways.

The paper is a narrative review. It pulls together earlier studies and points out where the standard checklist breaks down. It ends with a call to test new, ID-friendly questions before using them for real diagnosis.

The review found clear mismatches. For example, DSM items like 'feeling guilty' or 'decreased concentration' require language skills many clients with profound ID do not have. Using the items anyway can lead to missed or wrong diagnoses.

The author urged future researchers to collect social validity data first. That means asking caregivers and clinicians if the new questions make sense for this population.

Gustafsson et al. (2005) and Oliver et al. (2002) answered the call. Both teams tested ID-specific screeners, the Swedish PIMRA and the Swedish RSMB. Each tool showed positive overall findings, but both warned that single-item ratings were shaky. The lesson: total scores work better than one-off questions, echoing P's warning about misfit items.

Fujiura (2012) pushed the idea further. That review showed people with mild to moderate ID can give valid self-reports if you adjust the interview style. Together the papers create a bridge: start with caregiver tools for severe ID, then move to adapted self-report when the client can participate.

Goodwin et al. (2012) added a narrow but useful twist. They showed that even 'simple' behaviors like self-injury need clear, ID-specific definitions. Without them, researchers count different actions under the same label, just as P predicted would happen with mood or anxiety criteria.

Before you write 'major depression' or 'generalized anxiety' in a report, pause. Check whether the DSM questions match your client's communication level. If not, borrow from tools like PIMRA or RSMB, or break items into behavior-based indicators. Run them by the team for social validity. This small step cuts misdiagnosis and keeps treatment plans grounded in what you can actually observe and measure.