Collect, score, analyse and interpret WEMWBS

Collecting

The WEMWBS scales have been designed to be self-completed. Both digital and paper versions work well, if there is a need to help participants complete the scale by reading out items or filling in the questionnaire for them there is a risk that participants will respond more positively than if they were completing the questionnaire on their own. This could introduce bias into the study. As a rule of thumb studies need to include at least 30 people with data at two points in time or 30 people in each group to be compared. Some simple guidance on sample size calculation is available on page 38 of the user guide for researchers, user guides are available from the Resources pages of the website once you are registered to use the scale.

Scoring

The 14-item scale WEMWBS is very simple to score. The total score is obtained by summing the score for each of the 14 items. The scoring range for each item is from 1 – 5 and the total score is from 14-70. For more help with scoring WEMWBS please refer to the user guides available on the Resources Page of the website which is available once your have registered to use the scale.

The 7-item scale SWEMWBS is more complicated to score and it is important that it is scored correctly in order for comparisons to be made across different studies. SWEWMBS is a shortened version of WEMWBS which is Rasch compatible. This means the seven items have superior scaling properties to the 14 items, but in order to take advantage of this and to compare results with those of other studies using the 7-item scale SWEMWBS, raw scores need to be transformed.

Analysis

(S)WEMWBS scores approximate to a normal distribution, permitting parametric analysis. So the most statistically efficient approach to analysing data is to calculate means and standard deviations and compare results using Students T-test. UK Population norms have been published and can be used as comparators for your scores (WEMWBS and SWEMBS population norms Health Survey for England 2011).

Some investigators prefer to analyse their data using categorical approaches. Scores can be divided into high, average and low mental wellbeing using cut points. Two different methods can be used.

Statistical approach:

This puts the cut points at plus or minus one standard deviation. This approach puts approximately 15% of the participants into high wellbeing and 15% into low wellbeing categories.

Using this approach, UK population samples put score ranges as follows:

WEMWBS: the top 15% of scores range from 60-70 and the bottom 15% 14-42
SWEMWBS: the top 15% of scores range from 27.5-35.0 and the bottom 15% from 7.0-19.5

As background information

WEMWBS has a mean score of 51.0 in general population samples in the UK with a standard deviation of 7 (Tennant R, Hiller L, Fishwick R, Platt P, Joseph S, Weich S, Parkinson J, Secker J, Stewart-Brown S. The Warwick-Edinburgh Mental Well-being Scale (WEMWBS): development and UK validation. Health and Quality of Life Outcome 2007;5(63)) This means that 15% of the population can be expected to have a score of less than 42.3, so we have set the cut point at 42. Equally 15% of the population can be expected to have a score greater than 59.7, so we have set 60 as the recommended cut point for high wellbeing.
SWEMWBS has a mean of 23.5 and a standard deviation of 3.9 in UK general population samples (Ng Fat L; Mindell J, Boniface, Stewart-Brown Evaluating and establishing national norms for the short Warwick-Edinburgh Mental Well-being Scale (SWEMWBS) using the Health Survey for England Quality of Life Research 2016;26(5):1129-1144 ). This means 15% of the population can be expected to have a score >27.4 so we have set the cut point at 27.5 for high wellbeing. Equally 15% of the population can be expected to have a score <19.6, so we have set the cut point at 19.5.

Benchmarking approach:

Both WEMWBS and SWEMWBS have been benchmarked against well-validated measures of depression and it is possible to suggest scores that are equivalents to cut points for possible and probable clinical disease on both these scales.

WEMWBS has been bench marked on CES-D with which it is highly correlated (.84). For more information see (Donatella Bianca report).

Using CES-D =26 and CES-D = 16 as cut points, this analysis suggests that : -

a score of 41-44 is indicative of possible/ mild depression
a score of <41 is indicative of probable clinical depression.

These cut points can be used for analysis if it is important to focus on likely mental illness

SWEMWBS has been benchmarked on PHQ-9 and GAD-7 with which is it highly correlated (Neha Shah, Mizaya Cader, William Andrews, Sarah L Stewart-Brown Short Warwick-Edinburgh Mental Well-being Scale (SWEMWBS): performance in a clinical sample in relation to PHQ-9 and GAD-7 BMC Health and Quality of Life Outcomes (2021) 19:260 and also see cut points on SWEMWBS analysis)

Using the cut points of PHQ-9 = 5 and PHQ-9 =10 this analysis suggests that

a score of >18-20 is indicative of possible mild depression
a score of 18 or less is indicative of probable clinical depression

These cut points can be used for analysis if it is important to focus on likely mental illness

As there is no gold standard for measuring high mental wellbeing all cut points are by definition arbitrary. A cut-point of 60 and above can be used to identify the top 15% of scores on WEMWBS and 28 and above for SWEMWBS

Interpretation

Parametric approach: If you have used a parametric approach to analysis you need to compare your scores with population norms. Ideally results should be adjusted for differences in age and sex distribution. You compare mean scores and standard deviations before and after programme or intervention using Student’s T test. Link to calculation template: https://warwick.ac.uk/fac/sci/med/research/platform/wemwbs/using/register/resources/excel_spreadsheet_for_calculating_scores_with_wemwebs.xlsx

Categorical approach: If you are using a categorical approach you will have used one of the cut points recommended above and estimate the % of the population which falls into each category. You need to compare %s before and after your intervention or programme using a X² test. Link to calculation template: (https://warwick.ac.uk/fac/sci/med/research/platform/wemwbs/using/register/resources/excel_spreadsheet_for_calculating_scores_with_wemwebs.xlsx

Individual-level differences: Neither of the measures was developed for monitoring change at the individual level or in clinical settings, but both the 14-item scale and 7-item scale WEMWBS have been shown to be responsive to change at the individual level and some practitioners are using the scales to help clients and patients think about ways in which their mental health is changing. Different statistical approaches give different results with regard to minimally important levels of change. For the 14-item scale WEMWBS the methods give a minimum of 3 points and a maximum of 8 points (Maheswaran H Weich S Powell J Stewart-Brown S Evaluating the responsiveness of the Warwick Edinburgh Mental Well-Being Scale (WEMWBS): Group and individual level analysis BMC Health and Quality of Life Outcomes 2012, 10:156 (27 December 2012) for the 7-item scale SWEMWBS, a minimum of 1 point and a maximum of 3 points (Neha Shah, Mizaya Cader, William Andrews, Dilini Wikijsera, Sarah Stewart-Brown Responsiveness of the Short Warwick-Edinburgh Mental Well-being Scale (SWEMWBS) in a clinical sample Health and Quality of Life Outcomes Under Health Qual Life Outcomes. 2018; 16: 239