Why PT Exercises Stop Helping: Bird Dogs, Dead Bugs, and the Missing Progression
Somewhere in your house there is probably a printed sheet of exercises from a physical therapist: bird dogs, dead bugs, cat-cow, glute bridges, maybe pelvic tilts. You did them for a while. They helped a little, or seemed to. Months later you are still doing the same ten reps of the same drills, and your back is about where it was.
Nothing on that sheet is wrong. I want to be precise about what it is for, though, because the sheet has a limit built into it, and the limit is the reason so many people plateau on PT homework while believing they are doing everything right.
What those drills are for
Bird dogs, dead bugs, and cat-cow belong to a family clinicians call motor-control exercise. The goal is coordination and confidence: teaching the trunk to brace and move without guarding, waking up muscles that pain taught you to switch off, proving to a protective nervous system that gentle movement is safe. For the first weeks after a bad episode, that is exactly the right job, and these drills do it while asking almost nothing of you. The bird dog in particular earns its reputation: it loads the back extensors usefully with the spine in a neutral, well-tolerated position.
The evidence backs the family as a starting point. A Cochrane review of motor-control exercise for chronic low back pain found it works, about as well as other forms of exercise, and no better. That last clause matters more than it sounds.
The plateau built into them
Bodies improve on one condition: the demand keeps rising. Sports science calls this progressive overload, defined in the American College of Sports Medicine's guidelines as the gradual increase in stress placed on the body during training. Hold the demand constant and adaptation stops, at whatever level the demand reached.
Now look at the sheet. A bird dog loads your back with the weight of one arm and one leg. A dead bug loads less. Ten reps this month, ten reps next month, the same limbs weighing the same amount. These drills have no load lever: no place to add weight, no range to extend, no measurable number that climbs. Once they get easy, and they get easy fast, repeating them maintains a low level of capacity. It does not build more.
This is not a flaw in the exercises. It is a mismatch between the tool and the job. Control drills solve a control problem, and after the first weeks, most stubborn backs no longer have a control problem. They have a capacity problem: muscles that tire in minutes, a spine support system that cannot carry a normal day without complaining. Capacity problems only answer to rising load.
What the evidence says about loading
When researchers compare exercise types for chronic low back pain, loading keeps surfacing. A 2015 meta-analysis in Clinical Rehabilitation found strength and resistance programs outperformed other exercise types for pain reduction. A review of isolated lumbar-extension training, the kind of progressive loading a back-extension bench delivers, reported meaningful improvements in pain and disability across its included trials. Meanwhile the Cochrane finding holds: no exercise form is magic, and the common ingredient in programs that keep working is that they keep progressing.
Same movement, rising demand
That is the whole design behind the four-stage back-extension progression. The movement never changes; the demand never stops changing. Fifteen-second holds become longer holds. Holds become partial reps, partials become full range, full range takes on a plate held to the chest, and the plate gets heavier in small steps for as long as you care to follow it. Every session has a number, the numbers climb, and your next-morning pain readings decide the pace.
Compare a year of that against a year of ten bird dogs a day. One back has been asked to do slightly more a hundred times and has built the capacity to match. The other has repeated week one all year. Both people were consistent. Only one was progressing, and lasting pain reduction tracks the one whose back got stronger, because a back with capacity to spare stops treating ordinary life as a threat.
Keep the drills, add the engine
You do not have to abandon the sheet. Cat-cow makes a fine warm-up, bird dogs and dead bugs are useful movement on flare weeks and rest days, and if a clinician prescribed specific drills for a specific finding, follow them. Treat the drills as a warm-up and a fallback, and give the program an engine: one loaded movement, progressed in small measured steps, paced by how your back responds. The progression guide lays out the stages, and the bench guide covers the one piece of equipment it needs.
Sources
- Saragiotto BT, et al. Motor control exercise for chronic non-specific low-back pain. Cochrane Database of Systematic Reviews. 2016;(1):CD012004. link
- Searle A, Spink M, Ho A, Chuter V. Exercise interventions for the treatment of chronic low back pain: a systematic review and meta-analysis of randomised controlled trials. Clinical Rehabilitation. 2015;29(12):1155–1167. link
- Steele J, Bruce-Low S, Smith D. A review of the clinical value of isolated lumbar extension resistance training for chronic low back pain. PM&R. 2015;7(2):169–187. link
- American College of Sports Medicine. Position stand: progression models in resistance training for healthy adults. Medicine and Science in Sports and Exercise. 2009;41(3):687–708. link
The app
I built Low Back Pain Coach because this method lives or dies on pacing, and pacing is the part people quit. The app reads your pain logs, sets the exact holds and reps for each session, counts them out loud so your phone stays on the floor, and switches you to a recovery protocol when a flare needs rest instead of work. You can run the whole method from a notebook; the app is for making sure you still run it in week nine. The first week is free.