JazakAllaah khayr. The Springer article is not free to download :(

I have gone through the WHO document but that does not address my question.

I have calculated all domain scores. I am just not sure if the total overall score is calculated by averaging the domain scores or by adding them. I tried the internet but failed. 

Hi Inaamul,

There is no total score - just the domains. you need to report these as is. There are the two global questions that are also reported separately. 

Thanks a lot Tracy. That was great help.

Thankyou all for the information

Two years too late. Sorry.

 http://www.who.int/mental_health/media/en/76.pdf

Nothing is late when you learn. Thanks Renee.

Here is the complete manual!

http://apps.who.int/iris/bitstream/10665/77932/1/WHO_HIS_HSI_Rev.2012.03_eng.pdf

And this is a very useful article, for a cut-off point for the questionnaire.

https://www.ncbi.nlm.nih.gov/pmc/articles/PMC4203073/

Great! THX

when you get your answer plz tell me

Tracy has given us the answer Rubi. There’s no total score.

i am even more confused i am planning to calculate in overall score, how can i do that and is there any cutt off for poor moderate or good quality of life?

We have use median of total scores of all domains as cut point recommended by advisor and up on publishing on BMC, it raises question why median is used. so, is there any one who can help me to justify why median is used as poor and good quality of life below and above median respectively?

kefyalew Dagne

This document would help.

Hi there, If anyone would like an Excel sheet calculator to transform WHOQOL-BREF domain scores into the 0-100 scores outlined in the manual, I have created a document to do so here on Google drive:

https://docs.google.com/spreadsheets/d/e/2PACX-1vQeTjqyf20hdTAIVix3Qg2z7sOzqRo-3NvUDC14CWHt1XVMqCYhUhypnwQDQGcopW1xAwWiT2bYB4Ju/pub?output=xlsx

Hopefully it can help someone out there!

Dave

Dear Dave Skvarc,

Thank you for your helpful answer. It is very kind that you created the calculator to help others.

Kind regards,

Nina

My pleasure Nina, I hope it is of some use to you and others! I had to perform these transformations myself recently on a large dataset, and really didn't want to have to individually match up each score :)

Dear Dave Skvarc,

thank you for your wonderful answer

Kind regards,

fatma

http://www.who.int/mental_health/media/en/76.pdf

I have seen papers where they report the overall total scores and it wasn't clear to me. Tracy A. Commans makes more sense

Hi Nudra,

No problem, the instructions can certainly be confusing at first. I'm taking this info straight from here: http://www.who.int/mental_health/media/en/76.pdf

As you know there are 26 questions on the WHOQOL-BREF, each scored from 1-5, comprising 4 domains. The domain scores are not averages, they are the sum total score for each question within the domain. Multiplying the mean by 4 is used to transform the WHOQOL-BREF scores into the longer form WHOQOL-100. This is done in circumstances where researchers who use the WHOQOL-BREF want to easily compare their data with another researcher who has used the longer form.

3 of the questions are negatively phrased and so are reversed scored when calculating the domain scores - a score of 5 becomes a 1 and vice versa, a score of 4 becomes a two and vice versa etc. This is performed on questions 3, 4, and 26. You do this before calculating any domain scores.

The first domain is Physical Health and comprises Q3, Q4, Q10, Q15, Q16, Q17 & Q18. To get the Physical Health domain score, just sum all the values for each of these questions (after correcting the reverse coded questions).

Psychological domain is Q5, Q6, Q7, Q11, Q19, & Q26

Social Relationships is Q20, Q21 & Q22

Environmental is Q8, Q9, Q12, Q13, Q14, Q23, Q24 & Q25.

If you only want to talk about the WHOQOL-BREF without transformation, then that is as far as you need to go. If you'd like to transform these domain scores into the WHOQOL-100 scores, you take the mean of each domain score (summing as instructed above), and then multiply by 4. This value then corresponds to a value on tables from page 13 of the manual (use the column labeled "Transformed scores 4-20). Alternatively, you can skip the mean and multiplication step and using the same tables directly transform the domain score - I've made an Excel file in a previous answer on this page that will do this for you, feel free to use.

Cheers,

Dave

No worries Nudra, good luck! :D

Once we have the domain scores, how are they interpreted? (poor, moderate QOL, etc.). Thanks!

Read this article. It gives an answer that I found plausible.

Abd El Latif FI, Abd El Wahid HA, Amina Ahmed Mohamed

AA, Farg HK, 2016. Physical and psychological health domains of

quality of life in type 2 diabetic patients in relation to clinical factors

of diabetes mellitus in Egypt. Int Res J Med Med Sci, 4(1): 7-16.

It reads:

"The following values of scores were extracted from the reviewed studies and were applied in the current study: score ≤ 45, low QOL; score 46 to 65, moderate QOL; and score > 65, relatively high QoL (Bani-Issa, 2011)."

Thanks so much Agnes.

how we transform 31 item of WHOQOL BREF into WHOQOL 100?

Hi Getandale,

Do you mean item 3? The WHOQOL-BREF only has 26 items.

Cheers,

Dave

Hi Getandale Zeleke

This should help, I hope

https://stats.stackexchange.com/questions/25894/changing-the-scale-of-a-variable-to-0-100

Hi, I'v been struggling a little with the WHO-QOL BREF scoring interpretation as well. While there are some studies that have identified certain cut off point (Low, medium, high), using those cut off points may not be accurate as they have been generated by that particular sample for the study. I'm looking to development my own cut off points, but I need guidance of how I can go about it. Moreover, simply calculating the mean and SD does not work for my study as I am using the individual scores as a baseline measure. Anyone have any suggestions?

Hi Sonali,

You could convert your domains into 0-100 scores (have a look for my Excel calculator mentioned above), and then request that your statistics program separate the data into divisions. Otherwise, you could just check the cumulative percentage. Depending on which statistics program you use, this might vary - in SPSS, run Analyse>Descriptives>Frequencies. Click "statistics" and request that SPSS provide percentiles for 3 equal groups (or however many divisions you'd like in your categories).

The first table in the SPSS output will tell you the value of your data at the 33rd and 66th percentiles - any valules below your 33rd value are considered "low" for your sample, between 33rd and 66th are moderate etc.

These values will of course be idiosyncratic to your sample - that is, if you have a skewed distribution you'll likely have skewed estimates of QoL. For example, my own data had quite a positive skew because we were rating QoL prior to major surgery - people were in pain and thus generally had low QoL. Thus, simply using the top third of the data wouldn't be sufficient in my case - some of the ratings where still quite low! So unless your data follows a relatively normal distribution, consider carefully specifying what you want your cut-off's for low/moderate/high to be.

Let me know if that helps :)

Dave

Thank you Dave Skvarc, I will surely give this a try.

Hi Nudra Malik,

There isn't a single QoL score obtained with the WHOQOL-BREF, just the four domains. This is reflect that a person might be doing quite well in one aspect (such as physical health) but quite poorly in another (such as environmental quality).

Depending on your analysis plan, there shouldn't necessarily be a problem with running either regression or SEM with four dimensions. For example, you could run four multiple regressions with each domain as a separate DV.

Could you tell us a little bit more about what you'd like to do?

Cheers,

Dave

Hi friends,I am really confused with calculation of WHO QOL Bref score.

I) some authors calculate each domain score in decimal points.

2) some authors calculate each domain score out of 100 which is in Transform form

which is the correct way out of above two methods ?

3) Can we take mean of all 4 domains & get final score ?

4) Is there cut off score in each domain in ordinal form like low,moderate,relatively high & high level of QOL ? ( Ref.study on diabetic 2 patients by Fatma Ibrahim Abd El Latif et al,International j of medicine & medical Sciences vol 4(1),March 2016,pp 7-16)

5) while calculating regression analysis ,Do i have to consider all independent vaiables of demographic data with each domain of WHO QOL Bref separately ?

6) In my study I need to correlate QOL score with DASH score which is out of 100. Do I need to compare it separately with each domain of WHO QOL Bref

7) how do I correlate WHO score with Brief- COPE score as again there is no total Brief score ?

Can any friend help me getting answers to above questions ? I need these queries to be solved for my PhD data analysis.

Regards

Hi Nudra Malik,

Have you considered a structural equation model, with the DV as a latent QoL variable using the the four domains? This would simplify your task quite a bit, but you'd lose some explanatory power in your model (i.e. it might be harder to say if sleep disorders were impacting upon one domain more than another). If you've got access to SPSS through your institution, you may also have AMOS which can handle the analysis fairly easily.

Cheers,

Dave

Hi Chhaya Vijaykumar Verma,

1/2. The WHOQOL-100 is an alternative, longer form of the WHOLQOL-BREF. Both measure quality of life across four domains, and the WHO provide a transformation tool for converting scores off of the "BREF" into the "100" to make them more easily comparable across studies. For example, a person who scores a 30 on domain 1 using the "BREF" would have a score equivalent to 81 on the "100" measure. Either method is fine, it just depends on what you are trying to achieve with your data/analysis. There shouldn't be any decimals in the final scores at all, unless your participants gave decimal answers (i.e. someone responded "3.5" instead of a 3 or 4).

3. You could do this, but consider that you may lose some explanatory power of using the tool - you won't be able to detect differences between the domains (i.e. is someone's psychological QOL different to their environmental QOL).

4. There are no set guidelines from the WHO about cut-off scores for identifying low/moderate/high. categories on the domains. However, if you wanted to do this it would be fairly trivial in terms of setting it up as a comparison to the population - you could just say that anyone who got a "100" score above the 66th percentile had high quality of life etc. Have a look at one of my previous answers on this page for some instructions on how you might do this for a within-sample comparison. There are also a number of published cut-off scores like the one you have linked, but consider that it may not necessarily be applicable to your sample.

5+. These are all considerations that you will have to make in regard to your hypotheses. For example, it would be possible to run a multiple regression for some IV upon each of the four domains of the QoL using a sequential regression model - you would just add in all the demographic variables you wish to control for in the first step of the model, and the QoL domains at the second. With the BRIEF-COPE and the DASH you may need to consider also running the correlations for each of the domains with those measures, but I'm not familiar enough with those scales to guide you, unfortunately. Perhaps check if there are similar domains comparable to those on the WHOQOL.

Cheers,

Dave

Hi Dave Skvarc ,

Why am I getting domain scores in decimals and different from the original transformation table 0-100 given in page 13 of " http://www.who.int/mental_health/media/en/76.pdf " after using the SPSS syntax as given in page 106 of " http://apps.who.int/iris/bitstream/handle/10665/77932/WHO_HIS_HSI_Rev.2012.03_eng.pdf;jsessionid=5D329C9665DFF9C8F14A0BAC7F52F0A7?sequence=1" although the raw scores remain same ??

Hi Supratik Roy ,

I've had a bit of a muck around with this, and I think I've figured it out! The SPSS syntax is a little unclear (at least to me).

1. What you need to do is first calculate the raw score for the domain by summing all the values within the domain together (and accounting for any reverse coded items).

2. you take that value and work out your mean response for that domain.

3. Multiply the mean value by 4.

4. Subtract 4 from the result of step 3.

5. Apply the formula for conversion into percentiles (given in the syntax as "([DOMAIN]-4)*(100/16). This should then give you comparable results between the table and the syntax. Note here that we've already subtracted the 4 in the previous step, so we don't do it again.

For example, say I've got a Physical Health Domain raw score of 28 (step 1).

Step 2. That domain has 7 items, and so the mean response is 4.

Step 3. 4*4 = 16

Step 4. 16-4 = 12

Step 5. 12 * (100/16) = 75.

Let me know if that works :)

Cheers,

Dave

thanks a lot Dave Skvarc

and cheers!!

Hi Dave Skvarc

I have a doubt regarding statistical analysis applied on scores. Do we apply descriptive statistics on all 4 domains separately. For instance, i was looking at calculating standard effect size of intervention. For that do i have to compare domain(physical health score) of patients in control group with domain(physical health score) of patients in experimental group, then similarly for all three remaining domains separately? i.e in the end i will get the intervention effect sizes in four domains separately?

Regards,

Neha

Thanks @ Dave Skvarc

Your solution really helped.

Thank you so...much Dave.You have cleared all my doubts.It will sure help me in my statistical analysis.

You can find the manuals for WHOQOL here:

https://www.who.int/mental_health/publications/whoqol/en/

For WHOQOL - BREF there is even SPSS Syntax included for refining the raw data and calculating the domain scores.

Hi Dave Skvarc thank you so much for your Excel sheet and advice on this measure, it's been relally helpful. You don't by any chance have a similar Excel sheet for the 3-point (intellectual disability) version of the WHOQOL-BREF do you?Many thanks,

Annemarie

There is a confusion in scoring of WHO-BREF in which the data result from measurement cannot be processed and analysed straightforwardly but need complicated transformation by using statistical software in which researcher has to transform all of the raw data into scores range between 4-20, then continue with second transformation to convert domain scores into a 0-100 scale to meet comparable with the WHOQOL-100 items (WHO Geneva, 1995).

allow me to read the article

Dave Skvarc you are a lifesaver. That syntax was messing with my head.

According to the guideline, after you have transformed the score into 0 - 100, you can calculate total score for each domains but not to do grand totaling score because this instrument was abbreviated from WHO-Qol 100 items that conceptually developed based on PCS and MCS fragmented main concepts. Therefore, the instrument was not to measure level of Qol and doesn't allow to compare Qol in repeated measure studies. However, you can still compare domains of Qol descriptively.

I have one question regarding the transformation of WHOQOL-BREF to WHOQOL-100. In the guideline provided by WHO, we can transform the raw score by using a calculation formula as well as using a Table provided for transformation. However, these two methods give different transformed scores. For example, if the raw score of Domain 1 (Physical) is 8, the transformed score would be 3.57. But in the table (Method for converting raw scores to transformed scores, provided by WHO), the transformed score is 6. There is discrepancy across all the scores. Why there is a difference and which one is correct? Am I missing something here?

You could use this SPSS syntax replacing the variables used here by your data set variable names.

recode q301 q302 q303 q304 q305 q306 q307 q308 q309 q310 q311 q312 q313 q314 q315 q316 q317 q318 q319 q320 q321 q322 q323 q324 q325 q326 (1=1) (2=2) (3=3) (4=4) (5=5) (else=sysmis).

recode q303 q304 q326 (1=5) (2=4) (3=3) (4=2) (5=1).

compute DOM1=MEAN.6(q303,q304,q310,q315,q316,q317,q318)*4.

compute DOM2=MEAN.5(q305,q306,q307,q311,q319,q326)*4.

compute DOM3=MEAN.2(q320,q321,q322)*4.

compute DOM4=MEAN.6(q308,q309,q312,q313,q314,q323,q324,q325)*4.

count total=q301 to q326 (1thru 5).

filter off.

use all.

select if (total>=21).

execute.

many thanks all!!

I would like to assess the quality of life as a secondary outcome among the participants in my study (it is not the main aim of the study), can I use only two general questions to asses the quality of life of the participants?

Can someone please advise me, in some versions of the who bref questions 3,4 and 26 are scaled 5-1 in others they are scaled 1-5, which is correct? Thanks so much in advance,

Keira.

Yes Keira, you have reverse-coded items

Thank you Roberto it's just as you can see below in attached images some versions questions 3,4 and 26 are scored in 1,2,3,4,5 and on others I see 5,4,3,2,1 I'm confused as to which is the correct order, can you advise?

Here you have some files for further guidance.

The 2012 file has a spss syntax for scoring the scale.

There is no total score for WHOQOL-BREF rather you would have domain scores. Each of the six domain and general QoL have their own items within WHOQOL-BREF. Refer the WHOQOL-HIV BREF guideline for more.

Is there a grading for this score as in where we can lable it as satisfactory or not satisfactory or like wise

Mean and SD

Saumy Johnson,

WHOQOL-bref very rarely normally distributed. So run K-S before doing it.

Saumy Johnson how to categorize with Mean and SD?

I have read the manuals on WHO-BREF and read a few papers which have reported it. I have got my transformed scores etc but my study is analysing a 12-week exercise intervention on QoL (secondary to my main outcomes) , so its exercise vs control and there is pre, mid and post testing. Would I be safe to just run a repeated measures ANOVA on each individual domain (seen as there is not one specific score?) or would anyone recommend to do it differently?

HEY EVERYONE,

One question, regarding the QUALITY OF LIFE questionnaire (WHOQOL-Bref). I am kind of confused because there are 4 types of domains (physical, psychological, social and environmental health) and two separate questions (how would you rate your quality of life, and how satisfied are you with your health), which are not concluded into those 4domains. 

As I followed the manual for this questionnaire. I did mean score for each domain, but as the quality of life is my dependent varibale. Should I do a sum of those 4 domains with those 2 separate questions or should I do an overall mean score? or should I just one the one question (how would you rate your quality of life) because I want to find out their overall quality of life. 

Hi All,

After calculating the transformed scores of WHOQOL-BREF how to evaluate further? How to apply norms or measure the high/low quality of life? If there is no specific norms which statistical method I should use to calculate? Please suggest. I need this information as soon as possible.

Thanks!

I reported the domain scores as it is. but for the regression analysis, I used the mean score. still, I don't know whether it is correct or not.

There is no total score. It is true to report it as it is

Score of each domain to be reported

You have to report the domain scores. There are two transformation method, one converts scores to range between 4-20, comparable with the WHOQOL-100. The second transformation method converts domain scores to a 0-100 scale.

Actual raw score” is the values achieved through summation. “lowest possible raw score” is the lowest possible value that could occur through summation (this value would be 4 for all facets), and “Possible raw score range” is the difference between the maximum possible raw score and the lowest possible raw score (this value would be 16 for all facets: 20 minus 4). This transformation converts the lowest and highest possible scores to zero and 100, respectively.

The two stand-alone questions related to QOL rating and Satisfaction with Health are to be analysed separately, as recommended by WHOQOL-BREF (1996) while we report each of the domain as they are stated. No total score for WHOQoL-BREF.

I want to find out over all quality of Life score. How I will be given the score interpreatation.

Thanks!

WHOQOL-BREF not a single index instrument, it is only a descriptive of Qol, without total score, can not be used to measure level of Qol (high/low) and did not proper for repeated measures in interventional study. However, domain total scoring still can be done. Many structured errors was happen in existing published research. Thank you.

Please take note there are two scoring methods for this questionnaire ( WHOQOL) with three scoring methods ( raw total score, raw mean score, re-scaled 0-20 scored and a re-scaled 0-100 score) these multiple scoring manuals/methods had confused people worldwide .

The Row total score is simply adding up the items comprising each facet.

The raw mean score is also simply averaging the items comprising each facet.

The re-scaled 0-20 score computes the mean score ( i.e., the raw mean score ) and multiplies it by 5 yielding a score bound between 4-20 for each facets . This latest 0-20 score can farther be re-scaled into a score bound between 0-100 via using the formula that can be found in (Page-9 out of 12 pages) in this below link:

http://depts.washington.edu/seaqol/docs/WHOQOL-BREF%20and%20Scoring%20Instructions.pdf

I have attached the second scoring manual for you.

Accept my wishes with a great day everyone.

Mohammad

Tukimin Sansuwito I do agree you hundred percent this measure of Qol has been misleading and have been misused so frequently by the Copy-Cat researchers that follow each others blindly . I work as a freelance statistical consultant and the core problem i always face with researchers is that they want to force their analysis in formats and tables that match previous published research even if previous researchers had produced a Hodge-bodge analysis.

Can't agree more and thanks also to @Fahad Saqib @ Tracy A Comans a for sharing the ideas that made sense .

warmest regards

Mohammad

@Tracy A Comans & @Fatma Zaghloul Regarding the total score for the WHOQOL-BREF : According to most studies that has analysed the measurement model of the WHOQOL With structural equation CFA methods the total score is not tenable. However, doing a CFA on multiple studies on my end yielded a Good fit between the four indicators ( Psychological, social, physical and environmental) with their 0-100 scores against a latent overall qol score the model showed great fit in different datasets that came from Saudi, Qatar, Oman, Jordan, and Kuwait. These four composite indicators are supposed to comprise one whole unity theoretically . I am Not sure why the authors did not try to test them the way I described , no theoretical excuse for not trying that way . Therefore an upper General Bifactor Model may as well demonstrate that the total score can be supported when using the four sub-latent factors as indicators of a higher model that can be named as ( WHOQOL BREF overall score ). Factor analysis based (weighted) overall score can be estimated for more accuracy . Note : do Not use the 24 raw measured indicators however in such analysis, use the four sub-scales instead as shown in the attached path model figure . You may try that.

score ≤ 45, low QOL; score 46 to 65, moderate QOL; and score > 65, relatively high QoL (Bani-Issa, 2011)

You can use either or even both Rubaiya, I have chosen to use both,

Hope that helps,

Keira.

I am using Whoqol-Bref in my research. As there are two transformation methods 4-20 and 0-100.Which one should be used for the analysis ?

I am confused between two transformation methods of whoqol-bref i.e. 4-20 and 0-100. which scores should I use in the analysis of results ? Thanks

Thanks a lot @Keira White.

The 0-100 transformation makes the bref comparable to the original WHO 100 scale,

Keira.

What is cut off value for low/ medium/ high quality of life according to WHO?

You may watch this video for clear explanation https://www.youtube.com/watch?v=OfPaRASExvs