By Paul Doran
The importance of mobility scoring in lameness control programmes has been well established, with research having repeatedly demonstrated that in the absence of formalised mobility scoring, lameness detection tends to be very poor, with early cases going undetected and therefore untreated.
This is of great importance as it has become increasingly recognised that prompt identification and treatment of lameness is key to success. A 2016 study demonstrated 6-week cure rates as low as 25% when cows had been lame for at least 2 weeks at the time of treatment. This is contrasted to a similar study the previous year which found significantly higher cure rates in cows mobility scored every fortnight and treated within 48 hours of being identified as lame.
The cure rates in this study were highest when cows were treated with use of a block on the sound claw and a 3-day course of Ketofen in addition to therapeutic trimming.
However, mobility scoring is not without its drawbacks. Not only is it time consuming, especially in large herds but it is also subjective, with potential for inconsistency and variation between different scorers. It is also infrequent, as few farms would score more often than every 2 weeks and many score only quarterly or less.
Research has demonstrated that repeatability between scorers can be improved with training, which of course is the principle on which Register of Mobility Scorers (RoMS) accreditation is based.
Given its widespread adoption in the UK dairy industry, it’s perhaps surprising that until recently, there has been little work to investigate how accurate mobility scoring actually is.
However, there have been a couple of studies on this published in the last 2 years.
Siachos et al. 2025, found mobility scoring had an average sensitivity of 60% and a specificity of 78%, whereas a similar study in Ireland (Logan et al. 2024) found a sensitivity of only 35% but a higher specificity of 94%. This means that the scorers in the UK study missed fewer lame cows than the Irish study but in over 20% the cows they scored, no obvious cause of lameness was found in the feet.
This means that the scorers in the UK study missed fewer lame cows than the Irish study but in over 20% the cows they scored, no obvious cause of lameness was found in the feet.
There may be several reasons why the figures from these studies differ, but it is probably worth noting that whilst both used the AHDB mobility scoring system, in Siachos et al. the researchers were all RoMS accredited, meaning they have undergone specific training and passed the RoMS exam.
The Irish study makes no mention of RoMS training and RoMS accreditation is not an industry standard in the Republic of Ireland at present.
An exciting new development is the use of automated lameness detection systems based upon artificial intelligence analysis of camera footage such as CattleEye and HerdVision.
Although these systems are very new to the market, they have the potential to overcome many of the problems associated with human mobility scoring such as variability between scorers and infrequent scoring.
Of the systems currently on the market, CattleEye currently has the most substantial scientific evidence base to support its use. Interestingly, work to date has suggested that CattleEye is not yet quite as accurate as experienced RoMS scorers.
As the technology continues to develop, it would not be surprising if automated systems like this are able to out-perform human mobility scoring in the near future.
However, even if automated lameness detection were to remain less accurate than trained human observers, it may still offer advantages, such as the potential for continuous monitoring being able to track subtle changes over time and identify trends that would be impossible to detect with infrequent manual scoring.
Another very uniquely innovative technology to come to the market in recent years is the Hoofcount PediVue system.
PediVue consists of a camera situated behind a self-cleaning screen on the front of the foot bath to record passage of cows as they exit the bath.
PediVue’s potential for automated mobility scoring remains unclear however the system demonstrates an impressive ability to sensitively detect digital dermatitis, which under normal circumstances, is particularly challenging to detect as a majority of infected cows are not lame and therefore cannot be detected by mobility scoring alone.
For more information speak to our vets.
To read the full newsletter please go to our Monthly Newsletters page.
