Select Page

Last month I was invited to write a guest blog on Plinths and Platforms about load monitoring and rehabilitation.  It included an overview of Banister’s Fitness-Fatigue concept, the acute:chronic workload ratio and how it can be incorporated into the Return to Play (RTP) decision.  You can find that post HERE but much more importantly I would recommend reading the following two BJSM articles if you have not already:

 

The training-injury prevention paradox: should athletes be training smarter and harder? (Gabbett, 2016) Br J Sports Med doi:10.1136/bjsports-2015-095788

Has the athlete trained enough to return to play safely? The acute:chronic workload ratio permits clinicians to quantify a player’s risk of subsequent injury (Blanch and Gabbett, 2015) Br J Sports Med doi:10.1136/bjsports-2015-095445

 

I wanted to follow on from my Plinths and Platforms article by applying some of the analysis discussed to real world data.

To give some background to this case study: the athlete had suffered a groin injury and once ready to advance to working outdoors, underwent an 8 week progressive on-field rehabilitation programme.  Unfortunately after approximately 3 weeks of the return to training and playing, a reinjury occurred.

 

This is not intended to question the rehabilitation programme itself but to use an applied example to critically reflect on the load monitoring data.

 

Original Analysis

This data was collected prior to literature being published on the acute:chronic workload ratio but I used combinations of acute and chronic perspectives to visualise the load progression, as shown by the graph below.  The daily load is represented by a compound metric incorporating volume and intensity of acceleration, deceleration and change of direction outputs, colour coded for rehab, training and games.  The pink and green area series relate to the secondary axis and show rolling weekly and 3 weekly total for each day.

 

Original Analysis

Figure 1: Daily and Weekly Load

Visually you can see the jump in load in the daily bars from the rehab setting to training, but this is further emphasised when you view the rolling one week total and even more so a rolling three week total.

 

I wanted to revisit this case study to specifically apply the acute:chronic workload ratio to this data.

 

Weekly Acute:Chronic Workload Ratio

Day 1 is taken from the first outdoor session and thus prior to that date there were no measures of this specific metric, although it goes without saying “load” in other formats would have been applied throughout early stage rehabilitation.  Consequently with this outdoor-specific metric we see a particularly high acute:chronic workload ratio in the first few weeks.

Week to Week AC

Figure 2: Weekly Acute:Chronic Workload Ratio

From week 4 onwards a more perhaps “realistic” acute:chronic workload ratio is established.  I have included a threshold of 1.5 as per Hulin and colleagues, who found a significantly higher risk of injury above this ratio, both in elite fast bowlers (2014) and rugby league players (2015).

 

We observe that the ratio came close to this threshold as the volume of outdoor rehabilitation progressed at both weeks 5 and 7, influenced at week 7 especially by a number of offload days given during week 6.  The athlete returned to team training during week 9 and interestingly we witness a ratio of 1.7 the following week, the first (realistic) instance above the threshold from the literature.  During the week that the reinjury occurred, the acute:chronic workload ratio was 1.4.

 

Rolling Acute:Chronic Workload Ratio

It just so happens in this example that the length of time between the first outdoor session and the reinjury fell neatly into exactly 12 weeks.  However, sport does not always fall specifically on a weekly pattern or run according to the Monday to Sunday calendar.  Consequently, limiting the ratio on a week to week basis may not tell the whole story.  Therefore I have also plotted the rolling acute:chronic workload ratio for each day.

Daily AC

Figure 3: Daily Acute:Chronic Workload Ratio

In the previous week to week analysis the highest ratio after week 4 was 1.7 (week 10) but when we plot daily rolling ratio we also observe a ratio of 1.9 and three days of 1.8.  If we restrict our analysis to a week to week basis we are only capturing 1 out of every 7 data points.  Now clearly the weekly analysis was able to highlight week 10 as a high week and a ratio greater than the 1.5 threshold but this highlights there is more detail in the rolling analysis.  This suggests our load monitoring system in applied practice should be structured to establish rolling workload analysis on a daily basis.

 

Critical Reflection

At the time of this case study, the data enabled us to visualise the potential jump in load that may have played a part in the reinjury and allowed us to shape an objective loading programme going forward.  Of course it was disappointing to provide this data only in retrospect and not have been more proactive during the process to potentially influence the outcome.  Having revisited this example again and applied new analysis to the data, I was able to critically reflect on acute:chronic workload in this case.  As well as what I have already discussed, these points also came to mind:

 

  • Timeframes, thresholds, metrics… how do we know what are the right measures to use? For instance, was it incorrect to originally use a 3 week rolling total, rather than a 4 week average? I would argue not because this still conveyed the story of what had happened. It is important to consider that within different datasets there will be different relationships with injury risk, whereby the stronger relationships may be with different acute and chronic exposures and/or ratios.  Tim Gabbett, Billy Hulin and their colleagues have expertly demonstrated the relationships within their datasets, such as defining spikes as an acute:chronic workload above 1.5 with 1 week representing the acute timeframe and 4 weeks the chronic timeframe.  That is not to say there will not be relationships with say 5 days as acute and 3 weeks as chronic for example, in other environments.  Furthermore, if our data reveals a ratio above the threshold of significance this certainly does not indicate certain injury!
  • Best practice would be to model acute and chronic workloads in your own environment to establish the time lengths, metrics and thresholds most relevant to your own setting. But that does require a large, clean data set, time and statistical knowledge to carry out, therefore applying these findings to your own setting (whilst still maintaining a critical awareness) is certainly a good starting point.
  • Different metrics can represent different ratios for the same athlete over the same timeframes, potentially painting different stories of ‘fitness/freshness’ or ‘fatigue’. I have some data on this that perhaps I’ll share in another blog post!  I think this is another reminder to use common sense and context – for instance, the acute:chronic workload ratio for high speed running may be of most importance during rehabilitation from a hamstring injury and/or to monitor the workload of a fit athlete with a history of such an injury.
  • Often with load monitoring we are actually only capturing part of the picture when it comes to ‘load’. Rehabilitation in particular is complex; there are mixed methods of applying load (on pitch, in the gym, in the pool etc) and there are multifactorial milestones/targets to achieve throughout the process before RTT/RTP (read more HERE).  As with the RTP assessments discussed in that article (and HERE specifically for hamstrings), the data helps to add some objectivity but cannot take away from the clinical presentation and common sense – always remember the context of real life!  For instance, can we really not allow an offload day because of the effect on the acute:chronic load going forward, if clinically an offload day is needed!
  • If we believe the high acute:chronic ratio in week 10 (1.7-1.9) did influence the reinjury two weeks later in this case study, this may add evidence to the research that there exists some sort of delay between high workloads and injury. For example, Dr John Orchard demonstrated an increased risk of injury up to 3 to 4 weeks after a high acute workload (http://www.ncbi.nlm.nih.gov/pubmed/19346405).  This would have implications for practitioners in terms of how we analyse our training load data and consider the injury risks associated with acute:chronic workloads.

 

I am a big believer in analysing training loads from an acute and chronic perspective.  I concur with Hulin and colleagues when they state “monitoring the comparison of acute and chronic workloads should be mainstream practice in elite sport” (2015).  Whilst the guidelines that are proposed in the literature can be used as a framework to utilise in the applied world, I think it is important for practitioners to not be guided solely by the data; to maintain a critical eye on their specific data as well as remembering the context of what is occurring in real life.  As always the challenge for the Sport Scientist is to balance the art and the science of the data.

 

Jo Clubb

 

References

Hulin BT, Gabbett TJ, Blanch P, et al. (2014) Spikes in acute workload are associated with increased injury risk in elite cricket fast bowlers. Br J Sports Med; 48: 708-12.

 

Hulin BT, Gabbett TJ, Lawson DW, et al. (2015) The acute:chronic workload ratio predicts injury: high chronic workload may decrease injury risk in elite rugby league players. Br J Sports Med; Published Online First: 28 Oct 2015. doi:10.1136/bjsports-2015-094817

 

Orchard JW, James T, Portus M, et al. (2009) Fast bowlers in cricket demonstrate up to 3- to 4-week delay between high workloads and increased risk of injury. Am J Sports Med; 37:1186-1192.