Assessment Bias and Diagnostic Overshadowing
a teaching sheet.

Learning outcomes

By the end of the session, participants should be able to:

• Define diagnostic overshadowing and its origin.
• Explain how somatic items in common measures can score a physical condition as low mood.
• Use a six-step check for a new symptom, from describing the change to recording uncertainty and a follow-up plan.
• Tell clinical enquiry apart from medical diagnosis, and know when to refer.

Before the session

• Participants need: no prior reading.
• You need: this sheet. The Assessment Bias and Diagnostic Overshadowing page has more depth if you want it.
• Preparation time: about 45 minutes the first time.
• Materials: the case, the worked example and the worksheet below, shared in advance in accessible formats.

Key ideas

• The term was coined in 1982 in research on people with intellectual disabilities (Reiss et al., 1982).
• In a UK vignette study, 223 clinicians identified fewer symptoms when the same man was described as having an intellectual disability (Mason & Scior, 2004).
• Items about fatigue, work and health can inflate depression scores in physical conditions such as MS (Mohr et al., 1997).
• Some measures have barely been tested with particular groups: one review found only one study of depression measures in autistic adults (Cassidy et al., 2018).
• Extensions to physical disability and chronic illness are mostly by analogy, not separately replicated.

Full background: Assessment Bias and Diagnostic Overshadowing.

A session plan

A suggested 60-minute outline. Adapt timings to your group. Say at the start that people can take a break or step out whenever they need to, and add a planned break if the group needs one.

• Opening (10 minutes): ask the group for an example of a new symptom being put down to a condition the person already has.
• Input (15 minutes): the key ideas above.
• Activity (20 minutes): the case below, in small groups.
• Discussion (10 minutes): the questions below.
• Close (5 minutes): one change each person will try in practice.

Activity: “That’s just her condition”

Fictional teaching example

Niamh, 47, has rheumatoid arthritis. Over two months her sleep has worsened and she has stopped seeing friends. Her GP letter says “fatigue and low mood consistent with RA”. In pairs: work through the four-step check; decide what else is still a live possibility; and draft one question you would ask Niamh.

Worked example: the six-step check with Niamh

A note that records uncertainty:

Facilitator commentary. The point is not to decide that the change is “really” physical, or “really” depression. Symptoms can have physical, psychological and interacting causes. The aim is careful enquiry, so no single explanation is chosen too early. On measures: don’t remove items or adjust scores informally. Use the validated measure as intended, and say in your interpretation and notes where items overlap with a physical condition.

Facilitator notes: likely answers

What good looks like

• keeps more than one explanation open;
• includes Niamh’s own view;
• names an appropriate referral or consultation, with consent;
• records uncertainty and a plan, with a review date.

Discussion questions

• Which of your routine measures include somatic items?
• How would you record that you have checked a new symptom, rather than attributed it?
• When might a client want their existing diagnosis to explain a symptom?
• How does overshadowing work in the other direction, for physical health in mental health services?

Work here, pause, change direction

Difficult moments

Short responses for moments that often come up in this kind of session.

Don’t ask disabled participants to argue for whether their own access needs are legitimate.

Common misunderstandings

• “So we should stop using measures.” No: check what they were tested on, and ask what answers mean. Don’t drop items or adjust scores informally; note the overlap when you interpret and record the score.
• “The therapist should work out the medical cause.” No: the therapist’s job is to notice, ask, record and refer, not to diagnose.
• “Every new symptom is separate from the condition.” Sometimes it isn’t; the point is to investigate rather than assume.

After the session

• Follow-up: each participant checks one measure they use for items that overlap with common physical conditions.
• Review: next session, ask what they found and how it changes how they record scores.

Running the session accessibly

• Share materials in advance, in accessible formats, and use captions for any video.
• Build in breaks, and allow people to take part by speaking, writing or listening.
• Assume some participants are disabled. Don’t ask anyone to disclose, or to speak for disabled people as a group.
• Agree how personal material will be handled before the activity.