Understanding Specific and Non-Specific Effects

In episodes of PT Inquest, we have discussed the need to understand the specific effects of an intervention in contrast to the non-specific effects of an intervention. Since understanding the specific effects of a treatment is a main goal of any research on interventions, the hypothesis that all of the effects are non-specific is the null that must be rejected. But what exactly does that mean? Are all specific effects objective and measurable? Yes, absolutely. Are all objective changes specific effects? No, not at all. Confused? Then read on…

First, let’s talk about the word specific. What does that mean to you? Whenever you read the word specific I want for you to replace it with the word unique. So that means that the specific effects are the effects that are unique to the intervention. It is what that intervention does to the human body that nothing else can do. It is specific to that intervention. It also is predictable across all human subjects. Let me give you a brutally clear example:

If you strike someone’s femur with a baseball bat with a high enough load you will fracture their femur. The specific effect is “femur fracture”. The intervention is “blunt load”. The delivery system is “baseball bat”. There is no other way to fracture a healthy femur than a heavy enough load. That effect is specific to that intervention. Now, does it matter whether or not the load is applied via baseball bat, fall, hammer, or surgical grade two by four (Simpsons reference you might not get)? No. It does not need to be specific to the delivery system as long as you can find the common intervention.

Electro-Abs-5000 on hirsute male

Electro-Ab-Blaster 5000 on hirsute male

Another less brutal example would be resistance training. When you apply a certain load to healthy muscle tissue, it responds in a predictable fashion. If you apply a mechanical load to the hamstrings it will respond with an increase in measurable structural changes over time. There is no other way to do that (sorry Electro-Ab-Blaster 5000) other than through exercise. It is a specific effect. The exercise load is required whether it is through a hamstring curl or a Romanian deadlift (RDL) or one of a hundred other exercises.

Of course, clinical relevance of the specific effect is also important. I’ve written about this before.

Let’s explore non-specific effects for a minute. These are the effects of an intervention that are due to something that is not specific to the intervention. Most people think of the placebo effect as the main example. Although it is true that the placebo effect is an example of a non-specific effect, not all non-specific effects are placebo. Some can be nocebo or a perceived adverse effect from an intervention. So just as being compassionate and attentive can have a placebo effect, being an asshole can have a nocebo effect. Both are non-specific. This is sometimes called the interaction.

Now a therapeutic effect is always a combination of both specific effects and non-specific effects (ideally placebo, not nocebo). There is nothing wrong with this at all. There are known ways to enhance these effects and we should always use them. Here are some examples:

  • Face validity
  • Good listening habits
  • Confidence
  • Physical contact
  • Clean and professional presentation
  • Spending time with the patient
  • Validation of the patient’s concerns
  • Convincing explanation
Your friendly neighborhood physician (white lab coat included)

Your friendly neighborhood physician (white lab coat included)

Can an objective change be a non-specific effect? You bet it can! As long as it is not specific to the intervention. Imagine a patient who goes to their physician with neck pain and can only turn their head 10 degrees to the left. The physician is professional appearing in their white lab coat, sits down and listens attentively with compassion, palpates the patient’s neck, spends 30 minutes, and validates the patient’s concerns. Then the physician pulls out a sugar pill and tells them that it is a unique pill that will change the patient’s pain in about 15 minutes without the adverse side effects of narcotics. Now the physician returns 15 minutes after administering the sugar pill and asks how their pain is. “Much better!” Then the physician objectively measures that the patient now can turn their head 20 degrees to the left.

Now, was that objective change in their range of motion a specific effect of the sugar pill? Of course not. It was a non-specific effect due to the interaction. How do you feel about what the physician did? To me, knowingly giving a placebo and nothing more than a placebo without the patient’s complete informed consent is ethically dubious. In other words, you must say, “This is a sugar pill” or “This may be nothing more than a placebo”. Be crystal clear. Now, are you being clear in your billing? Sure, bill for the professional assessment but how much did that physician charge for the sugar pill itself?

Could an intervention be found later to have a specific effect? Sure. Just keep in mind that historically it doesn’t take decades to go from new intervention to known specific effects. As a matter of fact, most medical treatments are not even approved for use until the specific effects, good or bad, are known and validated. Also, keep in mind that if a treatment has “no potential side effects”, it probably has no specific effects either.

We always want the best therapeutic effect and should do everything we can to achieve that. Just try to make sure you are doing more than just placebo. The patient can go almost anywhere for a placebo.