Dear Gabriel,

A very good day to you. With regards to your question, unfortunately there is no definite/absolute/exact value or cut off threshold when it comes to the context of prevalence due to several reasons

First, prevalence study regardless to any disease, is very much depends on the total number of disease cases in proportionate to the Total number of population at risk. So the key factor here is the Number of People involved.

Secondly, low back pain is a big umbrella term that comprises variety of diagnosed and undiagnosed case definitions as well as across a wide spectrum of disease or injury.

As the matter of fact, one should not arbitrarily report their findings by saying low-moderate-high prevalence, but with reference to the findings that show significance in impact or in difference via statistical analysis. If it is statistically significant even though the prevalence is perceived as "low or moderate", it should be treated with crucial or importance.

Another way to determine whether the prevalence is worrying us or not is by referring to the measurement tool used by the author and its associated scoring system used. Some questionnaire uses scoring system to evaluate or assess the patient low back pain status. The total score rated could determine the severity or seriousness based on cut off threshold value. For example if a study reported that the prevalence of low back pain is 25%, it could be interpreted as "low". But among those 25%, it was found out that majority of the low back pain patients were scored as severe or serious category, then that 25% prevalence is neither low or high in value but require great attention.

I've came across some studies that compares the prevalence rate across different geographical areas, the study author explicitly stated based on the distribution and categorized it using quintiles-- either 3 quintiles, quartile (Box-whiskers plot) or 5 quintiles to describe the prevalence rate with definition stated as well.

Some studies without any reference, by perceiving the prevalence rate simply just equally divided 100% into 3, 4 or 5 group. For example it was reported the rate of 35% as low because of by using 100% and be categorized equally into 5 group with 20% range increment for every order e.g. 0-20% for very low, 21-40% for low, 40-60% for medium and etc.

Dear Gabriel, 

I agree with the answers by  @Raymond and @Joanne. There is not any specific criteria in context to your question. However, you can look for this paper and I believe you will be able to get some idea on what you want to do

http://www.ncbi.nlm.nih.gov/pmc/?term=prevalence+back+pain+systematic+review

Dear Gabriel

Thanks for the question. Joanne and Raymond gave great answers and Aamir gave  reference to a paper that provides some good stats showing how prevalent low back really is. Life time prevalence of up to 80%.

So with common conditions such as low back pain which is normal over the lifetime we need a time frame.  So questions such as have you ever had low back pain will not be helpful. We have to look at point prevalence mean 11%, or month prevalence mean 23% (see Hoy 2012).

With 'portmanteau concepts' such as low back pain, you may need to identify if there is in fact a specific diagnosis. However this is difficult as frequently patients have radiology that shows some degeneration. As we know there is no relationship between Xrays and the cause of pain and no relationship between disc bulges on CT and the pain. So some diagnoses (degeneration or lumbar spondylosis) are misleading.

With conditions that can be very mild and common, a functional categorisation can be helpful based on restriction of activity or work and medication requirement as these can exclude minor aches and pains which are not the focus of your survey.

All these suggestions have limitations.  But as long as you are aware of the limitations and  discuss the limitations in the paper/report, that is the best anyone can do. We all have the same issues in research.

The answer to your question is not in finding a number or a percentage, but in demonstrating your understanding of the limitations of your data  and in fact all data.  

Hoy 2012 - http://www.ncbi.nlm.nih.gov/pubmed/22231424         

First of all, an exact definition of the targent event ("low back pain") is mandatory. Please look at the CIE10 for the diagnosis code.

What follows is to design a "one sample size" statistical procedure, provided you have a hypothetical of the prevalence.. You can apply to any of the statistics books to find out what is a very simple formula for this calculation. Then you will be able to specify the degre of confidence you want the assessment will have after the collection of individuals.

The last issue is somewhat elusive: whether the proportion observed after the sample size is completed is or is not "high" (may be "low!"), depends on the settings you are "probing" the condition, as wel asl on the criteria you have in mind to qualify a result as "high" or not .

One wonders why you selected detecting "high" prevalence. Doesn´t it bear a prejudice of yours?. With a reasonable hypothetical assumption of  low back pain in your tarjet population  you are already prepared to calculating a sample size.

I can handle you precise statistical information for doing it.

Cordially.

Dear Gabriel, Find below a link of a study conducted in Nigeria on  The Incidence of Low Back Pain among Theatre Nurses: A Case Study of University of Ilorin and Obafemi Awolowo University Teaching Hospital. i'm optimistic that the paper will be of immersed help to you. 

International Journal of Nursing Science 2012, 2(3): 23-28 

DOI: 10.5923/j.nursing.20120203.02

Best of luck .

It will depend of your starting point, in epidemiology any disease which prevalence is lower than 10% is consider a rare disease, of course, you will have to take into account some other variables, such as those mentioned by collaborators.

you can compare it with a similar population with the same characteristics which is your interest  and its statistic is known .

THANKS for all responses to the question. Actually i had carried out 2 prevalence study, one published already in a reputable journal. I just wanted to know whether thier exist any standard border line at which epidemiologist and public health researchers and clinicians conclude lowback pain as been of high prevalence in a group or community. I just came from the conference of the society for the study of pain Nigeria where someone presented on OA KNEE prevalence. He reported 18% as very high prevalence of OA in a rural community. a lot of questions were asked. I then took further interest to ascertain what scientific facts guide the choice of high or low prevalence in a prevalence study., especially in noncommunicable diseases. whatever choice a researched make, he or she should be able to rationally justified such. Learning continues.....

May I read one of your prevalence studies?. It would enable me to adjust my suggestions.

Thank yoiu in advance.

Hello Gabriel, before you can make conclusion regarding the prevalence of a disease especially a long term condition such as low back pain, you will definitely need to standardize your definitions of what you are actually speaking about, take into account the context where the study is been carried, consider other factors and most importantly compare it to a baseline value or previously reported in a similar or same setting if the data exist. However, if this is not possible, then you need to correct for other factors and medically consider the cut off you will consider high or low in your study.

From the answers I can get there are two issues here: one, to performa a study on the low back pain prevalence. Second, to deciding whther the figure obtained from the study is to be understood as "high" or not.

It is not easy to state categorically at what point prevelance can be said to be high if there are no cut-off ponits globally. I think for now just report your prevalence figure.

dear Gabriel

It is almost impossible to answer such question as frequency of an affection would depend on many variables : your population source, the way the sample has been selected, the tools you're using to define low-back pain, etc. To be able to confirm that you have a high prevalence you need a comparator. You can compare to a previous study, preferably performed in the same population as yours and using the same tool or create your own comparator, for example by selecting another sample in another place, etc.

With best wishes

Christophe

Prevalence is the existing cases of a condition in a given geographical region. It will be adjudged high or low depending on the circumstances of geography, occupation, age, gender and other variables. You must look at history, precedent and previous works done and then compare your value with that obtained in the past. Also dont forget that prevalence RATES are what you compare and not prevalence; denominator must be taken into consideration for any reasonable comparisons. Then you will now be guided to determine an appropriate cut off.

Hi,

The prevalence of LBP would depend on the definition you have of it.  LBP at least one day during the previous year or LBP with a specific consequence (such as seeking care, not going to work) would get different estimates. What is your definition?

As LBP is a recurring condition, the one-year and life-time prevalence estimates would be fairly similar, whereas of course the one-day prevalence would be considerably lower. 

There is no real difference between sexes, also age does not really play a role (LBP is established around twenty and starts to diminish around 60). There is really no compelling evidence that different geographical areas or different jobs bring forth higher or lower estimates.

If you want to know if your estimate is higher than what is to be expected, you need to search the literature for studies in the general population of sufficient sample size to prevent occasional abnormal values that used the same sort of definition as you did.

Does this help?

Hi

I don’t know of studies that establish what percentage can be considered as high, medium or low to quantify injury.

Studies of incidence or prevalence of lesions simply detail the existing percentage of each lesion or a certain specific injury in a sample or study population.

Usually it is an interpretation of the authors and readers the values are high, medium or low, but for scientific purposes , is detailed percentages and what were the most frequent injuries.

Dear Gabriel,

We 'struggled' with the same question.

May be the following two attached papers are of help.

Good luck!

Kind regards,

Paul Kuijer 

The Global Burden of Disease study gives regional and country level data showing the range of prevalence of low back pain.

see eg vizhub.healthdata.org/gbd-compare/  and papers on my RG entry

Dear Gabriel and friends

In a study about the descriptive epidemiology of low-back pain in Algiers in 1991, the overall prevalence rate among individuals aged 15 and over was 90.9 per thousand. The prevalence was 126 per thousand for those aged 25 and over.  The prevalence of low back pain increased markedly with age, exceeding 250 per thousand in women aged 50 to 59. The  annual incidence was estimated at 30 per thousand  in the elderly over 30 years.

According to surveys conducted in population, low back pain prevalence rates published varied from 140-450 for 1000. The differences in methodology from one study to another at least partly explain the disparity of results. A fact is constantly highlighted: the high frequency of low back pain among the population.

Other numbers are in my published article "Descriptive epidemiology of low-back pain in Algiers" which is available in my Researchgate profil.

Prevalence of over 70 per 1000 is high. The etiology should be properly diagnosed so as to prevent occurrence. Naturally as population ages coupled with use of soft seats in offices, low back pain is likely to be on the increase.

Actually it varies with regard to different areas and work environments;

Globally the annual prevalence of back pain ranges from 15% to 45%, with point prevalence averaging 30%. The one-year prevalence of LBP among Western societies is reported to be between 20% and 62% and in Africans ranged from 14% to 72%. So you can recognize in relation to your concern area.

You can not fix the cut off value for your study , you can only compare the percentage of your study with local , neigbouring and developed countries. From this comparison some hypothesis could be formulated for further testing .

THANKS TO EVERYONE FOR YOUR RESPONSES SO FAR. REMAIN BLESSED

Assuming that the study was well done, the best way to conclude whether the prevalence is high or not, is comparing this prevalence with other prevalence done in the same place with the same methodology. Taking into account the normal sample variation.  Do not forget the confidence intervals.