activity pacing

Pacing, pacing, pacing…

If there’s one pain management and rehabilitation strategy that keeps me awake at night, it’s pacing. Living with persistent pain, I loathe the idea of pacing because I know everyone “booms and busts” from time to time, and few people like the idea of planning every single aspect of every single day as they come to grips with modifying their daily routines. BUT it’s one of the most popular strategies in textbooks, self-help books, and in treatment so there must be something in it, right?

Vexed definitions

One of the problems with the whole pacing concept is defining what we mean by it. I like Nicole Andrew’s approach: Nicole acknowledges that defining pacing is difficult, so when she talks about her research into pacing, she’s clear about the definition she’s using in that piece of work.

Various definitions abound. As a broad concept, pacing refers to organising daily activities in such a way that a specific end is achieved. The difficulty arises when we begin to determine the end goal of pacing (pain reduction? maintaining consistent activity levels? completing important tasks? avoiding a flare-up? reducing the relationship between pain fluctuations and activity? increasing overall activity levels over time?) and the means used to achieve these ends (time as a guide? activity intensity as a guide? importance and values as a guide? “spoons” of energy as a guide?). You can see how complex this concept is…

Nielson, Jensen, Karsdorp & Vlaeyen (2013) discussed this and identified two treatment goals (they weren’t considering the spontaneous use of pacing, nor the use of pacing outside a treatment context). “Whereas the operant approach seeks to improve function (decrease disability), the energy conservation approach is designed to reduce symptoms (pain, fatigue).”

Fordyce developed the operant conditioning approach, viewing pain behaviours as reinforced by other people – or by avoiding negative consequences such as a pain flare-up. His approach involved establishing a quota – a certain number, or a certain time in which people maintain activity irrespective of pain flucuations. In a clinical setting, this is the approach I mainly use, though there is an art to setting the “minimum” a person does (setting a baseline) and to nudging the activity levels up.

Sternbach, another influential pain management person from around the late 1970’s, followed a similar approach – but instead of simply establishing a baseline, he advised people to anticipate the point at which they would increase their pain and to stop the activity just before then. This is also a popular approach in pain management rehabilitation today – but has the unfortunate effect of reinforcing a pain avoidance (and pain contingent) approach, if not done very carefully.

Occupational therapists have frequently advocated the “5 p’s”. Pacing, positioning, posture, persistence and problem-solving. This approach was based on energy conservation, and while I can’t find the original papers from which this approach was developed, it was introduced to me as part of rheumatology practice, and in conditions where fatigue is a problem such as multiple sclerosis. I can see it being used today as part of the popular “spoons” meme where people are thought to have a fixed number of “spoons” of energy, and need to allocate their energy accordingly. My main criticism of this approach is that it doesn’t allow for people to increase their capabilities over time, either through “training” effects, or habituation.

Now, how about some evidence for any of these approaches?

Well therein lies a problem – there is very little research to support activity pacing despite its popularity. This is why I was so interested when I spotted a pilot study published in Journal of Pain, testing the energy conservation approach to activity managing (aka pacing) against an operant conditioning approach in a group of people with fibromyalgia. This group of people provides us with a useful population to test both approaches because fatigue is thought to be a prominent feature of fibromyalgia, and energy conservation has some degree of face validity for managing fatigue.

The design of the study involved four groups, two immediately treated using either an operant conditioning variant of pacing, or the energy conservation variant, and two groups with delayed treatments, again with the two versions (these groups acted as the control groups for this study). 178 participants were involved, with confirmed diagnoses of fibromyalgia given by occupational therapists using the American College of Rheumatology’s 2010 FMS diagnostic criteria. If the occupational therapist had doubts about the individual’s diagnosis, or the person wasn’t able to provide formal documentation confirming the diagnosis, the study rheumatologist assessed the potential participant for inclusion. This is an important procedure in studies of people living with fibromyalgia, given there is no definitive diagnostic test such as a blood test or imaging result.

The two treatment approaches were documented in treatment manuals to establish consistency, and it’s interesting to note that the approaches were applied across all activities in a day rather than just exercise, as often happens. For full descriptions of each of the ten treatment sessions, the article should be referred to, and the treatment manuals are available at management

Cutting to the chase, what did they find?

Well… to quote the authors “Inconsistent with the study’s primary hypothesis, neither treatment was effective in reducing average pain or usual fatigue symptoms. However, analyses of secondary outcome measures suggest the possibility that OL-based activity pacing treatments might be more effective than EC-based treatments in improving patient function.”

I didn’t expect pain reduction, or fatigue to be altered by an activity management approach: the relationship between movement and pain is highly variable, and there are many times we’ll be happy doing something and not experience pain simply because it’s something we enjoy. At the same time, I did hope to see a difference between the two approaches in terms of overall “doing” (function). My expectation was that pain may actually increase as people begin doing more, or alternatively, that people will feel more confident that they can achieve what’s important to them in a day, and that pain intensity becomes less of a guiding factor. The authors provide some explanations: perhaps the study numbers were too low to detect a difference (ie the study was under-powered); and perhaps a brief intervention isn’t intensive enough to help change over so many different aspects of a person’s life. Or perhaps, I want to add, neither approach is terribly great and while they both have intuitive appeal, persistent pain is too complex for any single activity management approach to make much of a difference. Maybe it’s something that needs other strategies to be incorporated such as exercise, mindfulness, medications, and even scheduling pleasant events.

So where does this leave us?

I guess for me, I like to think of activity pacing as one of many different tools in my toolbox. I bring it out when I’m attempting to increase my overall activity level – such as my walking programme, where I’m slowly but gradually increasing my capabilities without giving myself a whole two weeks of DOMs! I otherwise use a more flexible activity management approach: if something is important to me, and I think I can deal with the flare-up, I’ll do it. If it’s not as important to me, or I don’t think I can deal with the flare-up, I’ll probably modify my approach. Pacing, or activity management is only one tool…

Andrews, N. E., Strong, J., & Meredith, P. J. (2012). Activity Pacing, Avoidance, Endurance, and Associations With Patient Functioning in Chronic Pain: A Systematic Review and Meta-Analysis. Archives of Physical Medicine and Rehabilitation, 93(11), 2109-2121.e2107.

Nielson, W. R., Jensen, M. P., Karsdorp, P. A., & Vlaeyen, J. W. S. (2013). Activity Pacing in Chronic Pain: Concepts, Evidence, and Future Directions. Clinical Journal of Pain, 29(5), 461-468.

Racine, M., Jensen, M. P., Harth, M., Morley-Forster, P., & Nielson, W. R. (2019). Operant Learning Versus Energy Conservation Activity Pacing Treatments in a Sample of Patients With Fibromyalgia Syndrome: A Pilot Randomized Controlled Trial. Journal of Pain, 20(4), 420–439.

Pacing and avoidance in fibromyalgia

The recent emergence of study into ‘pacing’ or activity regulation in pain management is a welcome addition to our knowledge of this coping strategy. Although pacing has been described and included in many self-help books as well as clinical texts as an effective strategy for people with chronic pain to use, the research base for its use is pretty skinny (see Gill and Brown, 2009). McCracken and Samuels (2007) found that increased use of pacing was associated with higher disability and less acceptance, while Nielson and Jensen (2004) found that it was associated with lower disability in people with fibromyalgia.

The study I’m looking at today, by Karsdorp and Vlaeyen, looked in whether pacing specifically was different from ‘other behavioural strategies assessed with the Chronic Pain Coping Inventory (CPCI), such as guarding, resting, asking for assistance, relaxation, task persistence, exercise/stretch, seeking social support, and coping self-statements.’ The second part of this study looks more closely at whether ‘pacing was associated with physical disability when controlling for pain catastrophizing, pain severity and the other behavioural strategies as measured with CPCI.’

The methodology was pretty simple: a random sample of around 400 patients from the Dutch Fibromyalgia Association responded to being sent a set of questionnaires, a response rate of 68%. 388 women; mean age = 47.58 years, SD = 10.18, range 18–75 years. The mean duration of pain was 160 months (SD = 116.79). Thirty-four percent of the patients had a job and 39% of the patients received income from a disability income insurance.

The questionnaires used were the Chronic Pain Coping Inventory (Jensen, Turner, Romano & Strom, 1995). CPCI comprises 70 items measuring 9 behavioural strategies: Guarding (9 items), Resting (7 items), Asking for Assistance (4 items) , Relaxation (7 items), Task Persistence (6 items), Exercise/Stretch (12 items), Seeking Social Support (8 items), Coping Self-statements (11 items), and Pacing (6 items).
Pain intensity was measured using VAS, catastrophising was measured using the Pain Catastrophising Questionnaire (Sullivan, Bishop & Pivik, 1995), functioning was measured using a fibromyalgia specific Fibromyalgia Impact Questionnaire (FIQ-PH) (Burckhardt, Clark & Bennett, 1991), and the Pain Disability Index (Pollard, 1984) were used. Note: these were translated into Dutch for this study.

Thank goodness for statistics! Multiple regression analysis was undertaken – two hierarchical regression analyses were conducted with physical functioning or disability as the dependent variables. The first step included gender, age, and education, pain intensity, and pain catastrophizing. At the second and third step, the 8 CPCI subscales and the pacing subscale were entered, respectively.

What did they find?
At step one in the regression analysis, the demographic variables, pain intensity, and pain catastrophizing explained a significant amount of variance in physical functioning.

Older patients, patients with more severe pain, and patients who tended to catastrophize about pain reported greater physical impairment and more disability.

At step two, the 8 CPCI subscales without the pacing scale reduced the effect of age and pain catastrophizing to zero and significantly explained an additional amount of variance in physical functioning.

At step 3, the pacing subscale did not explain a significant additional amount of variance in physical functioning, leaving, in the final model, patients who avoided physical activities and asked for assistance to manage their pain reported greater physical impairment and more disability, even when controlling for demographic variables, pain intensity, pain catastrophizing and the other behavioural strategies.

So, what does this mean?
Remember, the first question was whether pacing forms a separate scale within the Chronic Pain Coping Inventory, and it seems to – at least in this Dutch version of the CPCI. Pacing is different from guarding, asking for help, avoiding and so on.

The second question was whether pacing was associated with physical disability when controlling for pain catastrophizing, pain severity and the other behavioural strategies as measured with CPCI. The reason for testing this hypothesis was to ascertain whether ‘activity pacing is an adaptive behavioural strategy that could be taught in pain management programs to improve adjustment in FM.’

Zero-order correlations revealed that patients using more pacing strategies reported greater physical impairment and more disability as opposed to less physical disability, so perhaps not such a great strategy to use – but wait: regression analysis demonstrated that pacing did not significantly contribute to physical functioning and disability over and above demographic variables, pain severity, pain catastrophizing and other behavioural strategies reported in chronic pain.

This means that the present study suggests that helping patients to increase pacing strategies in pain management programs may not be a key element in diminishing disability in FM. The authors suggest that the context in which pacing is used may determine whether it’s adaptive or not. Some patients may use pacing as an avoidance strategy, while others may use it, along with graded activity, to increase their ability over time.

The two strongest predictors of disability in this study were guarding and asking for assistance, which are likely to be dysfunctional strategies in FM. This shouldn’t be surprising, as these findings fit with the fear-avoidance model of disability.

Now, some caveats when interpreting this study. Don’t ever confuse correlation with causation – this is a correlational study, so there could very well be some intermediate factor that moderates the effect of activity pacing and disability. The authors quite rightly identify this. Longitudinal and experimental designs are needed to explore the relationship between activity pacing and disability in much more detail. Observational studies are needed to counter the self-report nature of the study instruments in this study. (Of course, we know how difficult observational studies are to carry out!).

Nevertheless, the authors suggest (and I agree based on my experience) pain management programs targeting activity pacing or behavioural strategies in general may not be effective in FM. Instead, therapeutic interventions based on fear-avoidance models specifically targeting paradoxical safety behaviours are likely to be useful in FM. I’m not sure we do exposure therapy as well as we might – and I’m certainly not sure we identify safety behaviours well at all. Perhaps something to explore in the future?

If you’ve enjoyed this post, and want to read more you can subscribe using the RSS feed link above. You can also bookmark and just visit! I write most days, love comments and usually respond pretty quickly. If you do make a comment, remember your comment will be public, so be aware of this. If you’d rather contact me directly, go to my ‘About’ page and send an email from there.

Karsdorp, P., & Vlaeyen, J. (2009). Active avoidance but not activity pacing is associated with disability in fibromyalgia Pain DOI: 10.1016/j.pain.2009.07.019

J.R. Gill and C.A.A. Brown, Structured review of the evidence for pacing as a chronic pain intervention, Eur J Pain 13 (2009), pp. 214–216.

L.M. McCracken and V.M. Samuel, The role of avoidance, pacing, and other activity patterns in chronic pain, Pain 130 (2007), pp. 119–125.

W.R. Nielson and M.P. Jensen, Relationship between changes in coping and treatment outcome in patients with Fibromyalgia Syndrome, Pain 109 (2004), pp. 233–241.