Regression rarely happens overnight. Before someone fully loses a skill or abandons a routine, there are often warning signs—patterns that may signal things are starting to slip.

The challenge is that these early indicators are subtle. They don't trigger alarms. They get lost in the noise of daily operations. By the time regression becomes obvious, you're not preventing decline—you're doing damage control.

This is where leading indicators come in. Unlike lagging indicators (outcomes that tell you what already happened), leading indicators are observable patterns that sometimes appear before decline becomes obvious. They give you a window to investigate and intervene—before full regression occurs.

The Warning Sign Window

Research on workers with autism found something important: at the six-month mark post-placement, many needed increased support to maintain their progress. Those who received that additional support tended to maintain their jobs and often advanced. Those who didn't were more likely to lose their positions.

The insight here isn't just that support matters. It's that the data suggests a window—often around the six-month mark—where increased support can make a meaningful difference. Early signs of struggle are a signal to investigate and act, not to wait and see.

Seven Leading Indicators to Track

Based on observations across employment, behavioral skills, and mental health research, these are patterns that have been associated with later regression. None of them guarantees decline—but each warrants attention and investigation.

1. Declining Routine Completion

A person who was consistently completing their morning routine starts missing steps. A participant who always checked in on time starts arriving late or skipping days.

What this may indicate: Routines are often the scaffolding that holds other skills in place. When routine completion drops, it's worth exploring whether something underlying is shifting—stress, environmental change, or fading motivation.

What to do: Don't wait for complete routine breakdown. A pattern of declining completion—even partial decline—warrants a check-in conversation to understand what's happening.

2. Increased Prompting Needs

Someone who was performing a skill independently starts needing verbal reminders. A participant who had reduced to minimal support suddenly requires more assistance.

What this may indicate: Skills may be weakening. This could stem from lack of practice, environmental changes that make the skill harder to execute, or emerging challenges that are draining cognitive resources.

What to do: Investigate the cause. Is the skill no longer being practiced regularly? Has something in the environment changed? Is there stress or distraction competing for attention?

3. Reduced Engagement

A participant who was actively engaged becomes passive or withdrawn. Someone who used to ask questions or seek feedback stops communicating.

What this may indicate: Withdrawal can be an early sign of struggle. People sometimes disengage before they visibly fail—especially if they're embarrassed about declining performance or don't want to ask for help.

What to do: Reach out proactively. Create a safe space to discuss challenges. Reduced engagement may be a signal for connection, even if it doesn't look that way.

4. Escalating Incidents

Minor incidents start occurring more frequently—small behavioral issues, minor workplace problems, low-level conflicts. Nothing dramatic, but a pattern of escalation.

What this may indicate: Increased incidents can reflect declining coping skills or mounting stress. The incidents themselves might be minor, but the pattern is worth examining.

What to do: Look for the underlying cause rather than just addressing each incident individually. What might be driving the pattern? Environmental factors? Skill gaps? Unmet support needs?

5. Avoidance Behaviors

Someone starts avoiding situations they previously handled. A participant skips social interactions they used to navigate. An employee avoids tasks they once completed.

What it signals: Avoidance is a coping mechanism for situations that feel overwhelming. When it increases, it suggests that previously manageable challenges are becoming harder.

What to do: Identify what's being avoided and why. Is the skill truly lost, or is confidence declining? Sometimes a brief confidence-building intervention is enough to reverse the pattern.

6. Changes in Mood or Affect

Persistent changes in how someone presents—increased irritability, visible anxiety, flat affect where there was previously engagement, emotional volatility.

What it signals: Mood changes often precede behavioral changes. Someone struggling internally will show emotional signs before skill regression becomes apparent.

What to do: Address the whole person, not just the skills. Mood tracking (if you're doing it) becomes especially valuable here. Even if you're not formally tracking, staff observations of affect changes should be taken seriously.

7. Disrupted Communication

Someone who was responsive becomes hard to reach. Check-in calls go unanswered. Communication becomes inconsistent or one-directional.

What it signals: Disconnection is a warning sign for regression across multiple domains. People who are struggling often pull back from contact—sometimes to avoid having to admit difficulties, sometimes because they're overwhelmed.

What to do: Persistent outreach. Make it clear you're available to help, not to judge. If communication breaks down entirely, consider whether alternative contact methods or a different person reaching out might help.

Building Detection Systems

Knowing what to look for is only useful if you're systematically observing. Here's how to build detection into your workflow.

Document with trajectory in mind. Progress notes should capture not just what happened today, but patterns over time. Is routine completion trending up, down, or stable? Is prompting level increasing or decreasing? Tracking trajectory across multiple observations reveals what single data points miss.

Create threshold alerts. Determine what level of change warrants closer attention. Three missed routine steps in a week? Two unanswered check-in attempts? Define your thresholds and investigate when they're crossed.

Train staff on what to notice. Frontline staff often see changes first. Make sure they know that declining engagement, increased prompts, and mood changes are important observations worth documenting and escalating.

Use follow-up touchpoints strategically. The 30-60-90 day playbook provides natural windows for warning sign detection. Build these indicators into your check-in questions.

From Observation to Action

Noticing early signals is only valuable if it leads to action. Here, ‘early signals’ means what staff can observe and document consistently—not predictive scoring or automated forecasting.

When you spot leading indicators:

Act early. The value of tracking these patterns is the time it creates for investigation. Don't wait to confirm that decline is "real enough" before checking in. By then, you've lost the advantage of early action.

Be curious, not punitive. These patterns are data points, not failures. Approach with genuine interest in understanding what's happening, not judgment about declining performance.

Match response to cause. A routine breakdown caused by an environmental change needs a different response than one caused by skill fade. Use what you learn to target your intervention appropriately.

Document your response. What you tried and what happened creates valuable data for understanding what works—both for this individual and for your program overall.

The Bottom Line

"We didn't see it coming" should become less common in programs that track leading indicators. Regression often shows early patterns. Those patterns become visible when you know what to look for and build systems to detect them.

The goal isn't to predict the future—it's to notice what's happening now, investigate thoughtfully, and respond before small changes become big setbacks.

For risk factors to assess before discharge, see The Regression Risk Framework.

‍