Barbell Medicine - From Bench to Bedside

So you’ve been training, but have started experiencing pain. This is an extremely common experience that we get asked about on a daily basis. Essentially, the question is “Doc, what do I do?”

Before we begin, we should point out that this will not be a philosophical magnum opus on pain and the human condition — those discussions will be left to Drs. Ray and Amato. This will be a more practical discussion to equip you with knowledge and strategies to self-manage and return to your normal activities over time. With that said, some basic concepts about pain are necessary to lay the groundwork for what we’ll actually be doing in the gym.

Part 1: Education

Most people view pain through a mechanical lens, as if the body operates like a machine that is prone to breaking and “wearing out”. In this view, pain serves an accurate indicator of damage to their bodily tissues — the idea that “hurt” always indicates “harm”. The seemingly logical conclusion from this premise is that painful activities should be avoided. Bunzli 2017 While this view is commonly perpetuated in society and by healthcare professionals, the past several decades of research have shown it to be inaccurate. Setchell 2017 We are complex organisms with many differences compared to machines — principally, the ability to adapt. We now understand that pain is a complex experience that is more related to the perception of threat and a need for protection than active tissue damage. Cohen 2018 Furthermore, this experience can be influenced by a number of biological, psychological, and social / environmental factors, which we will refer to as biopsychosocial factors. Moseley 2007


Turk DC, Wilson HD. Fear of pain as a prognostic factor in chronic pain: conceptual models, assessment, and treatment implications. Curr Pain Headache Rep. 2010;14(2):88–95. 

A useful analogy here is the experience of hunger, which is another sensation that serves as a protective mechanism against starvation. Most readers will recognize that the experience of hunger can be influenced by biological factors (the number of calories and types of food eaten that day, hormone levels, circadian rhythms, seeing/smelling food, etc.), psychological factors (stress, anxiety, expectations, preferences, cravings, seeing the clock, etc.) and social factors (social gatherings, seeing others eating, surrounding environment, etc.). In other words, there is no single “hunger generator” that can be isolated from this complex experience. Additionally, it should be clear that the intensity of one’s appetite is not directly and singularly correlated with the amount of calories one needs.

Similarly, there is typically no single “pain generator” that can be isolated in the complex experience of pain. Our tissues can detect potential threats from mechanical (e.g., pressure, stretch), thermal (e.g., hot, cold), or chemical stimuli through a process called nociception, but this is distinct from pain itself. Every experience of pain is influenced by varying combinations of biological factors (nociception, inflammation, visual/auditory inputs, fatigue/training load, illness, sleep, etc.), psychological factors (fear/anxiety, expectations, prior experiences, stress, mood state, attention, etc.), and social/contextual factors (reactions from others, knowledge of others’ experiences, surrounding environment, etc.).

These factors combine and interact to result in the emergence of a pain experience, which then drives an individual behavioral response. And just as we observe a poor correlation between the intensity of hunger and one’s calorie needs, there is similarly a poor correlation between the severity of pain and the degree of tissue injury, if any at all. In other words, just because something hurts more intensely doesn’t necessarily mean that more severe damage is occurring in the tissues. Pain is thus a non-linear experience.

This is illustrated by the enormous variation in different people’s reports of pain intensity in response to an identical nociceptive stimulus. In an experiment, 321 adults each received a standardized 48 °C (118 °F) thermal stimulus and reported their pain ratings on a scale of 0-100. Fillingim 2017 If pain provided an accurate measure of what is happening in the tissues, we would expect all ratings to be the same in response to the same stimulus. This standardized stimulus produced pain ratings ranging from 4 to 100, suggesting that other factors influence individual pain experiences outside of the tissues alone, and that pain isn’t “measuring” what we think in the tissues. All of this is to say that “hurt” does not necessarily equal “harm”, but rather apprehension of threat.

This idea of threat is especially relevant given common public perceptions of lifting weights as inherently unsafe, or unsafe in the absence of “perfect technique”. The obsession with rigid technical perfectionism stems from the traditional “body-as-machine” narrative, rather than from an understanding that humans can adapt to an enormous variety of movements and positions when dosed and loaded appropriately. Consider the wide variety of movement styles we observe just among powerlifters, or all the other ways humans move across different sports. Consider: what would constitute “unacceptable technique” for a Cirque du Soleil performer?

It is difficult — if not impossible — to establish physical positions that are inherently “bad” across the population (note: cadaver studies are not helpful here, since corpses can’t adapt). Given the lack of evidence to support claims about inherently dangerous positions or movements, and given the global problems of physical inactivity and musculoskeletal pain, we would prefer not to perpetuate negative expectations and perceptions of threat associated with physical activity and exercise. Saraceni 2019 Nolan 2019 Imparting these sorts of beliefs with our words has an enduring harmful impact on people. Darlow 2013 Bunzli 2017

Therefore, when working with individuals who report pain with a particular position or movement we aim to engage the movement in a way that desensitizes the person rather than to perpetuate a learned response of avoidance. Bunzli 2017 For example, many of our rehab clients presenting with knee pain are fearful and avoidant of movements like high bar squats, front squats, lunges, or split squats because of misplaced fears about things like “shear forces” that they’ve learned from individuals who hold a position of authority. Similarly, people with back pain often demonstrate significant fear of spinal flexion and extension, despite these being normal human movements that are unavoidable during lifting and life. Lehman 2018 Caneiro 2018 We challenge these narratives and get people to work towards executing these movements over time with very gradual dosing, while providing reassurance and education about the meaning of pain (i.e., that symptoms do not reflect active/ongoing tissue damage).


From Woby et al‘s 2007  European Journal of Pain article, Self-efficacy mediates the relation between pain-related fear and outcome in chronic low back pain patients.

Of course, we recognize that there are absolute limits to tissue tolerance under external loads, but these relate to injury more frequently in the context of high-velocity, unpredictable movements. This is what we observe in field sports, where injury rates are indeed much higher compared with the more controlled gym environment where we can adapt to move in a variety of ways (including with spinal flexion!). Furthermore, when injuries in the gym do occur, they are generally not serious, and most often resolve on their own within days to weeks. Even herniated intervertebral disks are extremely common among people who have no back pain symptoms or dysfunction at all, and a majority of these heal on their own over time despite their reputation as a catastrophic injury. Brinjiki 2015 Zhong 2017 Chiu 2014

So it is reasonable to coach a particular movement style for the purposes of performance and efficiency, but we deliberately avoid pairing our movement cues with unnecessary messages (either overt or subliminal) of danger or threat. In the same way we criticize irresponsible healthcare professionals who warn those with back pain that “One wrong move and you’re paralyzed,” we similarly criticize coaches who perpetuate the belief that “If you let your technique slip, you’ll get injured,” since these ideas induce unnecessary hyper-vigilance and threat around exercise.

With all this said, many of us have experienced pain that lingers or that is particularly distressing, especially when it interferes with the training or other activities we’d like to do. So, with this new understanding of pain in mind, what should our initial approach in the gym look like?

Part 2: Exercise

One approach we typically do NOT recommend for most routine aches and pains in the gym is absolute rest. While recovery is certainly an important component of training adaptation and rehabilitation, absolute rest is typically unnecessary and comes with costs as well. Specifically, it results in a loss of fitness adaptations, leaving the individual progressively more detrained the longer they spend away from training. This often results in a situation where an athlete returns to training with a strong desire to handle workloads they are no longer adapted to handle, increasing the risk of recurrent symptoms and further psychological distress given their now-decreased tolerance for loading. Additionally, we have evidence in several contexts that absolute rest often results in either no improvement (e.g. in tendinopathy) or worse outcomes (e.g., in low back pain).

Step 1: Finding the entry point

Our first and most important step is to find our “entry point” into the rehabilitation process. The usual approach for determining this is to find a type and dose of exercise stimulus that results in either improved symptoms or stable symptoms over the subsequent 24-48 hours after training. A marked increase in symptoms during or after training reflects a dose of stimulus that is likely too high in terms of intensity, volume, or both.

Of course, there are a multitude of training variables we can choose to manipulate, including external intensity (absolute weight), internal intensity (e.g., effort level or RPE), volume (sets & reps), frequency of exposure, exercise selection and tempo, and many others. There is no one “best” way to approach this process for all situations. However, we can provide a general framework that you can use as a starting point for experimentation with the goal of finding your entry point.

Decreasing external intensity (absolute load, or “weight on the bar”) while maintaining other variables constant is our typical initial intervention. Situations where symptoms are minimal or non-existent at very low loads, but progressively worsen as loads increase, have a very good prognosis. For example, an individual who currently experiences pain while squatting 365 lbs but has minimal or no symptoms while squatting 95 lbs is likely to do very well with temporary reduction in loading, and a gradual progression back to normal levels.

The most common error in this situation is a reluctance to reduce external intensity enough to mitigate symptoms. An individual who has pain under 365 lbs might reduce the load to 315 lbs and express frustration or hopelessness because they still have symptoms, but remains reluctant to reduce the load to 225 lbs, 135 lbs, or even to the empty 45 lb barbell. We would generally prefer to “under-shoot” when seeking out the entry point rather than continually “over-shoot” as demonstrated in the example above.

Another strategy to force a reduction in external intensity is to 1) increase the number of reps per set (upwards of 12-15 repetitions), and 2) add a “tempo” component to the lift (e.g., 3-count tempo on the eccentric plus 3-count tempo on the concentric, or an isolated 6-count tempo on the eccentric). For example, the individual who normally squats 365 lbs x 5 reps may only be able to handle 85 lbs for a set of 15 repetitions with a 3-0-3 tempo, and they may find they are able to do this with tolerable symptoms that do not markedly worsen during or after training. This indicates that we have successfully found an entry point into the rehab process.

Now, some individuals may have persistent, intolerable symptoms across all loading ranges for a particular exercise, even down to very low loads. At this point we can adjust the stimulus via 1) range of motion alteration based on tolerability, or 2) changing the exercise altogether.

For example, an individual who experiences back pain immediately off the floor when deadlifting may temporarily substitute an elevated deadlift from blocks or from pins in a squat rack. An individual who experiences knee or hip pain in the bottom of a squat may temporarily substitute a box squat or a pin squat to a different depth. An individual with shoulder or elbow pain in the bottom of the bench press or overhead press may temporarily substitute a board press, floor press, or pin press. It should be emphasized, however, that the external load should still be reduced in this situation; in other words, just because the range of motion is reduced in a rack pull being used for rehabilitation purposes does not mean the exercise should automatically be loaded heavier than the individual’s typical deadlift weights.

Alternatively, if the individual experiences persistent or worsening symptoms during/after training with this approach, a temporary change in the exercise altogether may be needed. There are no “right” or “wrong” exercises in this situation, but rather exercises that should target the affected area in a way that is tolerable to the individual. This may involve minor changes, such as swapping low bar squats for front squats, or more significant changes may be required. This may even involve unilateral work like substituting barbell squats for weighted lunges, or deadlifts for single-leg RDLs.

In practice we often combine these approaches, including both a VARIETY of bilateral and unilateral exercises directed at the affected area, given that many individuals exhibit movement compensation to off-load symptomatic areas (e.g., shifting away from a painful side in the squat). This step can sometimes present challenges for individuals who have strong psychological attachments to particular exercises, but we would prefer to build physical and psychological resilience to tolerate a wide variety of movements and movement styles without fear or perception of threat.

The last variable to consider is the frequency of exposure. When establishing our entry point we would similarly prefer to under-shoot this target rather than to continually over-shoot it and risk progressive symptoms. For this reason, we prefer to start out with 2-3 exposures per week on non-consecutive training days, leaving at least one day off in between exposures for recovery and to assess tolerance to the dose of stimulus.

So, what might a generic template look like for someone who is experiencing non-specific knee pain in the squat? We may begin with 2-3 non-consecutive days per week of training:

Day 1:
  • 3-0-3 Tempo Low Bar Squat: 12 reps @ RPE 6, 12 reps @ RPE 7, 12 reps @ RPE 8
  • 3-0-3 Tempo Split Squat: 15 reps @ RPE 6, 15 reps @ RPE 7, 15 reps @ RPE 8
Day 2:
  • 3-0-3 Tempo High Bar Squat: 12 reps @ RPE 6, 12 reps @ RPE 7, 12 reps @ RPE 8
  • 3-0-3 Tempo Weighted Lunge: 15 reps @ RPE 6, 15 reps @ RPE 7, 15 reps @ RPE 8

Of course, if the individual is unable to tolerate regular squatting without a significant increase in symptoms, this template may need to be regressed further. For example, replacing barbell squatting with other movements entirely (leg press, Spanish squat, goblet squat, or any number of other movements), reducing external load, or reducing total volume further until a tolerable entry point is established.

Step 2: Progression and Expectation management

Once this entry point is found, our goal is to begin stringing together a series of small victories in order to build positive physical and psychological momentum. In this way, the nature of progression becomes critically important, as overly aggressive jumps in loading dosage increase the risk of symptom exacerbations.

A key point here is that being “pain free” is not typically a realistic short-term goal; symptoms are always a part of the rehabilitation process, and there will be ups and downs along the way. An increase or recurrence in symptoms may be related to a number of factors, including the dosage of stimulus (external intensity, volume, etc.), but also due to outside biopsychosocial factors such as anxiety, sleep disruption/restriction, life stressors, concurrent medical illness. This is an opportunity to reiterate that “hurt does not equal harm,” that symptoms are an expected part of the process, and that we have strategies to mitigate symptoms — typically by modifying the dosage and type of loading.

We suggest that initial loading increments be conservative. If symptoms remain constant (i.e., neither exacerbated nor improved), we may even keep the absolute load constant for a few of the initial sessions to begin observing a de-sensitization effect. Once we observe this effect, we can begin incrementing loads across non-consecutive sessions. There is no “optimal” increment to use in terms of absolute weight (e.g., 2.5-5 lbs at a time), or relative increases (e.g., 5-10% of load) from session to session, but we may give individuals who are looking for specific guidance these sorts of arbitrary suggestions. However, we qualify this advice since we do not want to imply that the process will be linear. Neither strength acquisition nor injury rehabilitation are straight-line, predictable affairs. Rather, we should maintain flexibility in our approach to allow for the “up” days where we may be able to increment a bit more, as well as the “down” days where we may need to adjust loading down or hold it constant depending on the nature of symptoms during and between sessions.

The most common errors made in this process involve overly aggressive increments of loading despite worsening symptoms, which often results in the process taking longer than necessary. This may be due to inappropriate expectations about a reasonable timeline for recovery. For example, full recovery from a tendinopathy or acute radiculopathy may take several months, whereas other non-specific aches and pains may resolve in a matter of a few days. Patience, acceptance, and psychological buy-in to the process are at least as important as the physical training — these are discussed in further detail in Charlie Dickson’s article.

As loads gradually trend upwards with stable or decreasing symptoms, the target repetition ranges will naturally need to decrease (e.g., from sets of 12, to sets of 10, 8, 6, etc.) and tempo will accelerate towards normal speed (e.g., from 3-0-3 to 3-0-0 to regular or paused movements). With an improvement in symptoms we can begin to consider increasing the volume of training to continue building capacity. This may involve starting by adding one back-off set to the regular programming, assessing tolerability and symptoms in the subsequent 24-48 hours, and continuing on from there until the individual has progressed back to their baseline level of function and training.

Step 3: When to seek assistance / consultation

Ideally, if the steps listed so far lead to a resolution of symptoms and a return to normal activities, no further assistance or consultation are needed. However, we recognize that this process comes with a number of challenges. Even among clinicians and athletes who are intimately familiar with this process, it can be difficult to remain objective with the experience of pain. The staff at Barbell Medicine frequently consult with one another on such matters for more objective and qualified third-party input.

If you are struggling with symptoms, are having difficulty finding an “entry point” into the process, or have tried to proceed through this process without success, it is reasonable to seek out assistance from a trusted clinician. This may involve someone local to you, or our own rehab clinicians are happy to help. 

For individuals who have been through the process successfully but then experience recurrent symptoms, it is essential to analyze the programming being used and the surrounding life circumstances (e.g., external stressors, sleep, etc.) at the time symptoms onset, and whether any patterns can be discerned. It is likely that the dosage of stimulus being used is inappropriate and a more permanent programming change in terms of intensity, volume, frequency, and/or exercise selection may be required. The most common errors we see on this front are in individuals who use overly rigid programming models involving excessive intensity, excessive volume, poor/no fatigue management strategies, and/or hyper-specific exercise selection (i.e., low or no exercise variation). In these situations our recommendations usually involve immediately instituting a load and fatigue management strategy such as RPE-based autoregulation and increasing exercise variation as initial steps.

If what you are doing now is not working, do not assume that repeating the same process again and again will provide a different result. Sometimes, wholesale changes are required. Recovery from injury, especially chronically painful conditions, may require that you adjust the goal posts and accumulate small victories over long periods of time.

Find your entry point, and embrace the process.


Thank you to Drs. Thomas Campitelli, DTFP, Michael Ray, DC, Michael Amato, DPT, Derek Miles, DPT, Jordan Feigenbaum, MD, and Loraine Baraki, MD for their assistance in reviewing this piece.

About Austin Baraki

Dr. Austin Baraki is a practicing Internal Medicine Physician, competitive lifter, and strength coach located in San Antonio, Texas. Originally from Virginia Beach, Virginia, he completed his undergraduate degree in Chemistry at the College of William & Mary, his doctorate in medicine at Eastern Virginia Medical School, and Internal Medicine Residency at the University of Texas Health Science Center in San Antonio.

Read More by Austin Baraki