Yellow-flag Screen

Status: stable Category: fear_avoidance Added: 2026-04-20 Last reviewed: 2026-04-21

Summary

The Optimal Screening for Prediction of Referral and Outcome — Yellow Flag screen (OSPRO-YF) is a brief presence/absence screen for psychosocial risk factors in musculoskeletal pain. Trinity implements a 10-item binary version: each item is yes/no, and the total count maps to three bands (low, moderate, high).

Unlike most other instruments in the catalog, OSPRO-YF's score is a flag count, not a continuous-scale severity.

⚠ Binary response format

Every item here accepts only 0 (no) or 1 (yes). Trinity will score other values per the rubric's math (summing whatever integer is provided), but the band semantics assume binary responses. Do not send Likert-style responses to this endpoint — use kinesiophobia, fear_avoidance_beliefs, depression_screen, or anxiety_screen for Likert-scaled constructs.

Resource identity

Field	Value
Category	fear_avoidance
Resource	yellow_flag_screen
Instrument ID	yellow_flag_screen
Rubric version	1.0.0
Path prefix	/api/fear_avoidance/yellow_flag_screen

Operations

Operation	Method	Path	Contract
Questions	GET	/api/fear_avoidance/yellow_flag_screen/questions	patterns-questions.md
Score	POST	/api/fear_avoidance/yellow_flag_screen/score	patterns-score.md

Items

Ten binary presence/absence items.

ID	Prompt (paraphrased)
item_1	Marked pain-related worry present.
item_2	Depressed mood present.
item_3	Anxious / tense mood present.
item_4	Low self-efficacy for recovery.
item_5	High pain catastrophizing.
item_6	High fear of movement / reinjury.
item_7	Low recovery expectation.
item_8	Somatic complaints beyond index pain region.
item_9	Sleep disturbance reported.
item_10	Active substance-use concern.

Answer choices per item:

Value	Label
0	No
1	Yes

Scoring

Field	Value
Method	simple_sum over all ten items.
Score range	0 – 10 (count of flags present).

Bands

Band	Range (inclusive)	Clinical meaning
low	0 – 2	Low psychosocial-flag burden.
moderate	3 – 5	Moderate flag burden; consider targeted follow-up.
high	6 – 10	High flag burden; multi-domain follow-up typically warranted.

Bands reflect the count of "yes" flags — more flags means broader psychosocial risk footprint.

Required items

All ten items are required for a strict /score call.

Worked example

All ten flags raised. See quickstart.md for per-language scaffolding and patterns-score.md for the shared /score contract.

Send:

POST /api/fear_avoidance/yellow_flag_screen/score HTTP/1.1
Host: engine.tripod-health.com
apikey: $ENGINE_API_KEY
Content-Type: application/json
X-Request-ID: docs-yellow_flag_screen-score-basic

{
  "responses": {
    "item_1": 1, "item_2": 1, "item_3": 1, "item_4": 1, "item_5": 1,
    "item_6": 1, "item_7": 1, "item_8": 1, "item_9": 1, "item_10": 1
  }
}

Receive (200 OK):

{
  "instrument": "yellow_flag_screen",
  "category": "fear_avoidance",
  "finding": {
    "algorithm_id": "yellow_flag_screen",
    "algorithm_version": "1.0.0",
    "score": 10,
    "band": "high",
    "raw": {
      "rubric_version": "1.0.0",
      "responded_items": 10
    },
    "context": null,
    "imputed": false,
    "sd": null,
    "prior_source": null
  },
  "confidence": {
    "overall": 1,
    "tier": "high",
    "breakdown": {
      "input_completeness": 1,
      "imputation_variance": null,
      "model_certainty": null
    }
  },
  "recommendations": null,
  "depth": "basic",
  "catalog_version": "0.13.0",
  "request_id": "docs-yellow_flag_screen-score-basic",
  "timestamp": "2026-04-21T23:15:00.089456+00:00"
}

Computation: sum of 0/1 flags = 10 → high band. Zero flags → 0 → low; three flags (for example item_1, item_3, item_6) → 3 → moderate (3–5).

Calling in your language

curl

curl -sS -X POST "https://engine.tripod-health.com/api/fear_avoidance/yellow_flag_screen/score" \
  -H "apikey: $ENGINE_API_KEY" \
  -H "Content-Type: application/json" \
  -d '{"responses":{"item_1":1,"item_2":1,"item_3":1,"item_4":1,"item_5":1,"item_6":1,"item_7":1,"item_8":1,"item_9":1,"item_10":1}}'

Python (requests)

# Uses the call() helper from quickstart.md.
responses = {f"item_{i}": 1 for i in range(1, 11)}
result = call("POST", "/api/fear_avoidance/yellow_flag_screen/score",
              {"responses": responses})
print(result["finding"]["score"], result["finding"]["band"])

Node.js (native fetch)

// Uses the call() helper from quickstart.md.
const responses = Object.fromEntries(
  Array.from({ length: 10 }, (_, i) => [`item_${i + 1}`, 1]),
);
const result = await call("POST",
  "/api/fear_avoidance/yellow_flag_screen/score", { responses });
console.log(result.finding.score, result.finding.band);

TypeScript

// Uses the typed call<T>() helper from quickstart.md.
const responses = Object.fromEntries(
  Array.from({ length: 10 }, (_, i) => [`item_${i + 1}`, 1]),
);
const result = await call<ScoreResponse>(
  "POST",
  "/api/fear_avoidance/yellow_flag_screen/score",
  { responses },
);

Go

// Uses the call() helper from quickstart.md.
responses := make(map[string]int, 10)
for i := 1; i <= 10; i++ {
    responses[fmt.Sprintf("item_%d", i)] = 1
}
result, err := call("POST",
    "/api/fear_avoidance/yellow_flag_screen/score",
    map[string]any{"responses": responses}, 3)
if err != nil {
    panic(err)
}
finding := result["finding"].(map[string]any)
fmt.Printf("OSPRO-YF: %v (%v)\n", finding["score"], finding["band"])

Java

// Uses the call() helper from quickstart.md.
StringBuilder body = new StringBuilder("{\"responses\":{");
for (int i = 1; i <= 10; i++) {
    if (i > 1) body.append(",");
    body.append("\"item_").append(i).append("\":1");
}
body.append("}}");
String result = call("POST",
    "/api/fear_avoidance/yellow_flag_screen/score",
    body.toString(), 3);
System.out.println(result);

C# (.NET HttpClient)

// Uses the Call() helper from quickstart.md.
var responses = Enumerable.Range(1, 10)
    .ToDictionary(i => $"item_{i}", _ => 1);
var result = await Call(HttpMethod.Post,
    "/api/fear_avoidance/yellow_flag_screen/score",
    new { responses });
var finding = result.GetProperty("finding");
Console.WriteLine($"OSPRO-YF: {finding.GetProperty("score").GetDouble()} ({finding.GetProperty("band").GetString()})");

PowerShell

# Uses the Invoke-Engine helper from quickstart.md.
$responses = @{}
1..10 | ForEach-Object { $responses["item_$_"] = 1 }
$result = Invoke-Engine -Method Post -Path "/api/fear_avoidance/yellow_flag_screen/score" -Body @{ responses = $responses }
"OSPRO-YF: $($result.finding.score) ($($result.finding.band))"

Clinical interpretation

• What it measures. Breadth of psychosocial risk across multiple domains — worry, mood, anxiety, self-efficacy, catastrophizing, fear of movement, recovery expectation, somatic complaints, sleep, and substance-use concern.
• Flag count, not severity. The score measures how many domains show concern, not the severity within any domain. A subject with 3 mild flags and a subject with 3 severe flags have the same OSPRO-YF score; severity should be assessed with the domain-specific instruments.
• Companion to Örebro. Where orebro returns a weighted total and one risk band, OSPRO-YF returns a breadth count. They answer overlapping but distinct questions.

Intended use

• Rapid multi-domain screen when the caller wants to know how many psychosocial risk areas are elevated.
• Triage signal for deciding whether to administer the domain-specific instruments (PHQ-9, GAD-7, TSK-11, FABQ-W, recovery expectation).
• Feeding the CDRI composite as a complementary breadth measure alongside Örebro's weighted total.

Mistaken use

• Do not send Likert or scaled values. The rubric's scoring is a simple sum; a 3 on an item produces three times the band contribution of a 1, which breaks the flag-count semantics.
• Do not interpret flag count as severity within any domain. A moderate OSPRO-YF with a severe PHQ-9 is different from a moderate OSPRO-YF with a minimal PHQ-9. The OSPRO-YF alone does not distinguish the two.
• Do not skip the domain-specific follow-up on a moderate or high band. The screen points where to look, but the depth of evaluation in each flagged domain belongs to a dedicated instrument.
• Do not compare OSPRO-YF scores across populations without context. The meaning of "moderate" flag burden differs between acute and chronic pain populations.

Debugging

1. Capture request_id. Present in every response body.
2. Score substantially above 10. A Likert-style response (value 2 or higher) was sent on a binary item. Send only 0 or 1.
3. 400 ValidationError with missing_items. Send all ten items.
4. Band unknown in a response. Not possible under the current rubric with binary responses (0–10 covers the whole band space). If you see it, a non-binary value was summed into the score.

Related

• patterns-questions.md, patterns-score.md — shared contracts.
• instruments-orebro.md — Örebro; multi-domain screen with a weighted total score.
• instruments-depression_screen.md, instruments-anxiety_screen.md, instruments-kinesiophobia.md, instruments-fear_avoidance_beliefs.md, instruments-recovery_expectation.md — the domain-specific instruments a moderate or high OSPRO-YF typically triages into.
• composite-cdri-compute.md — composite.

Change history

Date	Change
2026-04-20	Initial publication.