1 The Data: A Trend That Was Already Moving

Look at the screening rates below. The program launched in 2019, and rates kept rising afterward. But notice something important: they were already rising before the program started. (Data are simulated for illustration.)

Screening Rate Over Time (Before and After Program)

Before Program (2014–2018)
After Program (2019–2024)

2 The Threat Explained: What Is Maturation?

Maturation means outcomes change naturally over time—independent of any program. If you only compare "before" to "after," you might credit your program for changes that would have happened anyway.

📈 What Causes Natural Trends?

Many things drive outcomes to change over time, even without intervention:

  • National health campaigns and awareness
  • Changes in medical guidelines
  • Technology improvements
  • Economic growth or decline
  • Demographic shifts in the population
  • Participants simply getting older

🎯 Why It's a Threat

If your outcome was already improving, and it continues improving after your program:

  • You might claim credit for natural change
  • Your "effect size" is actually zero
  • Resources spent on a program that added nothing
  • Decision-makers misled about what works

Think of It Like This

Imagine you start watering a plant in spring. By summer, it's grown taller. Did your watering help, or was the plant going to grow anyway because of the season? Without knowing what would have happened without watering, you can't tell.

The Key Question

Did the rate of improvement change after the program—or just continue at the same pace it was already on?

3 What To Do: Addressing Maturation

You can't eliminate maturation threats entirely, but you can recognize them and use study designs that account for natural trends.

1. Look at Pre-Program Trends

Before claiming your program worked, check if the outcome was already changing. Plot several years of baseline data.

"Screening rates rose 2 points/year before the program and 2 points/year after. The program didn't accelerate the trend."

2. Use a Comparison Group

Find a similar group that didn't receive the program. If both groups improve at the same rate, the improvement isn't your program's effect.

"Both the program county and the comparison county saw screening rates rise equally. The program added nothing beyond the regional trend."

3. Extend Your Baseline

The more years of pre-program data you have, the better you can estimate the natural trend and test if it was stable.

"With 5+ years of baseline data, we can fit a reliable trend line and project what would have happened without the program."

4. Report Honestly

If your data shows the trend didn't change, say so. Null findings are valuable—they prevent wasted resources on ineffective programs.

"The program launched, but the rate of improvement didn't change. We found no evidence the program accelerated outcomes."

Takeaway

Maturation is one of the most common threats in before-after studies. Always ask: "Was this outcome already moving in this direction?" If yes, you need a comparison group or trend analysis to separate program effects from natural change.