Prevalence and the Cut-off Thresholds

It is important to note that there is no single standard measure of the general prevalence of disability. Furthermore, prevalence is a function of the inclusion criteria selected, and different criteria are appropriate for different objectives. The choice of inclusion criteria (the selection of a cut-off point on the continuum) determines not only the percentage of the population with disabilities (and therefore prevalence) but, consequently, also the characteristics of that group. A cut-off point that identifies a subpopulation with more severe limitations might be used to estimate the population requiring more intensive and focused services while a cut point that includes a subpopulation with milder limitations might estimate those who would benefit from universal design. Both estimations of prevalence would be valid and useful for the associated intended purpose.

Box 22. Cut-off thresholds of persons with disabilities

For making international comparisons, the WG recommends that those with disability are defined as those who answer, “a lot of difficulty” or “cannot do at all”. 

A few countries, such as Oman (2010), Qatar (2010) and Tunis (2014), had set their cut-off measure for the population with difficulty in at least one domain in order to include the three levels of those with “a lot of difficulty”, “cannot do it at all” or “some difficulty”.

 

Devising special indicators for persons with disabilities is unnecessary since standard indicators can be used. Although they must be reported separately from their disability status.

Box 23. WG classification of disability

The threshold for those with a disability are defined as those who answer the WG questions with “a lot of difficulty” or “cannot do at all.”

With disability:  1. A lot of difficulty. 2. Cannot do at all.

Without disability:  1. No difficulty. 2. Some difficulty.

 

However, the threshold for identifying persons with disabilities is important. For example, if different thresholds are used by different countries in reporting disability status then clearly their disaggregation results would be incomparable. 

Therefore, the WG recommends that for international comparisons, the threshold for those with a disability are defined as those who answer the WG questions with “a lot of difficulty” or “cannot do at all.”

However, countries that also apply the “some difficulty” threshold will include in their results those people who have minor difficulties and, thus, most likely will have a higher outcome of disability prevalence. Conclusively, the more severe the threshold for defining disability, the bigger the disability gap.

Box 24. Heterogeneity in national disability estimates

A single estimate of disability — like the 15.3 per cent prevalence from the World Report on Disability — masks a great deal of heterogeneity. The World Report points this out and notes that the rate of severe disability is closer to 3 per cent.  However, at the country level the cut-off between a person being disabled or not disabled depends on the reason for identification. For the purpose of providing permanent cash benefits, the cut-off might be quite high. A government might only want to provide such benefits to people for whom there is no possibility of work. 

 

Validity is given to viewing disability as a spectrum, which is also advantageous for analysing the prevalence of persons with varying degrees of disability. Although people who identified as only “some difficulty” are not categorized by the WG as having a disability, some researchers analyse the sub-population of people with “some difficulty” in the six domains related to activity functioning. While these people may not be identified as having a disability, they would most probably benefit from policies that make environments more inclusive.