Appendix B: PALS 2006 severity scale for children

View the most recent version.

1. Types of disability
2. Scores
3. Filter questions
4. Imputation of the 'undetermined'
5. Special cases
6. Construction of index
7. Creation of the severity classes

For PALS 2001, an index measuring the severity of the disability has been constructed based on answers to the filter and screening questions. This index was recalculated for PALS 2006 in order to compare results of both surveys. The methodology used to produce the severity scale for children is presented in this paper. Please note that the interviews about children activity limitation were conducted with the parent or guardian of the child.

1. Types of disability

For children as for adults, there are 10 different types of disability: hearing, seeing, speech, mobility, dexterity, learning, delay, developmental disability, psychological condition and chronic health condition. Table 9 below presents the different disability types and the age group(s) for which each type applies. This table also presents the different contexts in which disability questions are asked. It should be noted that only one functional limitation is measured for each disability type except for speech where two functional limitations are measured.

For adults, since the same questions were asked for all age groups, a single scale was derived and applied to all respondents. For children however, two different scales were constructed because some questions differ according to the age of the child. Two scales were constructed, therefore, one for the 0 to 4 years old and one for the 5 to 14 years old.

2. Scores

Some questions measure the intensity of the disability while others the frequency of the disability. Both types of questions are used in the calculation of the scores. Scores are assigned for each question in increasing order of severity. For example, the points given for a question measuring the intensity are as follows:

• "Some difficulty": 1 point
• "A lot of difficulty": 2 points
• "Completely unable": 3 points
• Other answer (no, refusal, don't know): 0 points

For a frequency question, points are assigned as follows:

• "Yes, sometimes": 1 point
• "Yes, often or always": 2 points
• Other answer (no, refusal, don't know): 0 points

When both frequency and intensity are available, the product of the two scores is used. For each disability type, a unique score is required. For speech disability, two different questions are asked: the ability to speak and the ability to make himself/herself understood. These two questions (two scores) have to be combined in order to obtain a unique score for this type of disability. Furthermore, for certain disability types, the same question is asked in different contexts. For example, for learning disability, Question CLFT_Q04 is asked for both the home and school situation as well as for play or recreational activities. These sub-questions also have to be combined in order to obtain a unique score for a given disability type.

When there is more than one question for a particular type, the scores are added and then standardized in order to obtain a score between 0 and 1:

(1a)

Whereis the score for disability type T, N is the number of different questions (functional limitations) for type T,is the maximum score for the i th question for disability type T andis the score obtained for the i th question for disability type T. In some cases,may be made up of more than one question. When the same question is asked in different contexts, the mean of the scores is calculated. For example, for speech disability, Question COFT_Q05 about the ability of the child to make himself/herself understood, is asked in three contexts: when speaking with family members, friends or other people. Since the same question is asked three times, the mean score is assigned to this type:

(2a)

Whereis the number of different contexts in which the i th question of disability type T is asked andis the score for question i and context j of type T.

Here is an example for speech disability ( T =SPEE). This type is composed of two different questions (), SPEE1 (difficulty speaking) and SPEE2 (difficulty to make himself understood when speaking). SPEE1 is asked in a general context () while SPEE2 is asked in three different contexts (family, friends and other) (). The maximum score for SPEE1 is 3 and the maximum score for SPEE2 is 2 (, ):

(1b)

Where is asked in a general context and where:

(2b)

The mean score of Question COFT_Q05 over the three contexts is calculated in expression (2b), while in expression (1b), the score for SPEE is standardized in order to obtain a score between 0 and 1.

3. Filter questions

For respondents with a score different than zero for the screening questions, no additional points are assigned for the filter questions. However, for respondents with no points for the screening questions, essentially the "yes-no" group (YES to the filter questions and NO to the screening questions), points are given based on the four filter questions. A global score for filter questions is computed the same way than the score attributed to each disability type, with expression (1a) and (2a). In this case, , , ,.

4. Imputation of the 'undetermined'

For children for whom enough information is available to determine that they have activity limitations, but the intensity and/or the frequency are unavailable, a flag of "Undetermined" is assigned and their score for that specific disability type is imputed. A fairly simple method was used for the imputation. This technique consists of looking for a group of respondents with the same answers to specific questions as the respondent with incomplete information. The average score of this group of respondents is then imputed for the respondent with the missing information. Here is an example:

A respondent has an answer of "Yes, sometimes" to Question CMFT_Q01 about mobility disability, but there is no answer to CMFT_Q02 on the intensity of the disability. All respondents who gave the same answer as this respondent to CMFT_Q01 are obtained, then the average score for mobility is calculated and this value is imputed for the "Undetermined" respondent.

This method of imputation is justified by the fact that there is a correlation between the frequency and the intensity. People answering "Yes, often" to the frequency question are more likely to answer "Completely unable" or "A lot of difficulty" than people answering "Yes, sometimes".

5. Special cases

There are certain disability types for which the following is asked:

A) If the condition reduces the amount or the kind of activity the respondent can do (the frequency question).

And then, if the answer is "Yes":

B) How many activities does this condition usually prevent the respondent from doing (at home, at school, at play or recreational activities) .

A respondent is considered limited if he/she answered "Yes" to A. However, it is possible to answer "None" to all four contexts in Question B. This situation is not changed by the edit rules that are used for both the adult and the child questionnaire. Since the points are assigned based on the product of the score to Question A and the score to Question B, no point is assigned in this case (score to B = 0). However, the respondent is considered limited for that specific disability type. For example, a respondent can be limited globally for four disability types, but points could be assigned for less than four types (for example, two types). In particular, a respondent could be considered limited with no point at all.

It was decided to assign a minimal score to these people for the specific disability type. This was done by giving one point to all people who answered "Yes" to A, and then by calculating the score as originally planned based on Question B. For example, if the maximum score for a given disability type is six (frequency (2) X intensity (3)), this maximum score now becomes seven with this modification. Respondents with " Yes" to A and "None" to B would obtain a score of 1. In summary:

"Yes" to A) and "None" to each question in B): 1 point

"Yes" to A) and at least one answer to B): 1 point + points assigned to B)

Respondents with the answer "None" to B will therefore necessarily be the ones with the smallest score as they will only have the point for A.

For some disability types, a respondent is considered limited (and points are assigned) if a disability is reported, even if there is no limitation. So, in these cases, points are always assigned. This is the case for learning disability and developmental disability. The list of these particular disability types and the questions where an additional point is given for a "Yes" are presented in Table 10.

6. Construction of index

There is a strong relation between learning disability and developmental disability. For most of the respondents reporting a developmental disability, a learning disability is also declared. However, from a subject-matter point of view, it could be argued that these two disability types are exclusive. In order to avoid double counting of points, respondents with a developmental disability are not given any additional points for reporting a learning disability.

The global score is computed by taking the mean over all the standardized scores of each disability type. As mentioned earlier, two different scales were derived for children, one for the 0 to 4 years old and one for the 5 to 14 years old:

(3a)

(3b)

Where

These two scales are computed only for those with a positive answer to the screening questions (the "yes-yes" group and the "no-yes" group). For the "yes-no" group ("Yes" to the filter questions but "No" to the screening questions), only the filter questions are taken into account and this group forms an additional type of disability (unknown):

(4a)

(4b)

The reason why we did not use the filter questions also in (3a) and (3b) was to avoid double counting of the same information. For example, a child with a mobility disability probably answered "Yes" to the filter questions thinking about his/her mobility limitation and answered "Yes" also to the mobility questions.

Individuals in the "yes-no" group probably answered "No" to the screening questions because the PALS questionnaire could not measure their disability or simply because they are too mildly disabled to report a positive answer to the screening questions. This is the reason why this group was treated separately. A very small disability score was given to that group.

Some results for the global score are presented in Table 11 and Table 12 below. Table 11 ( Table 11a for children 0 to 4 and Table 11b for children 5 to14) shows the number of children with one to four disabilities and the number of children with no disability or with points coming only from the filter questions. For each group, the mean, the minimum value, the maximum value and the standard deviation of the global score are presented. For example, there are 13 respondents aged between 0 to 4 with four disabilities (the maximum value for this age group). The mean score is 0.65137 with a standard deviation (STD) of 0.17398, a minimum value of 0.34524 and a maximum value of 0.91667.

The proportion of children with a certain type of disability by number of disabilities is presented in Table 12. As for Table 11, results are presented separately for the 0 to 4 children (Table 11a) and for the 5 to 14 children (Table 11b).

Other approaches were considered in order to limit the redundancy of the information contained in the severity scale. In addition to the strong relationship between developmental disability and learning disability, there are other significant correlations between some disability types in the scale. In order to remove redundancy of information, an unequally weighted scale was considered. Instead of having a weight of 1, disability types that are strongly correlated would have a smaller weight in the global score. Since it is difficult to justify the use of unequal weights in the scale, this option was rejected.

7. Creation of the severity classes

In order to create severity classes, the distribution of the global severity score was examined. The distribution has been separated into deciles. The first decile corresponds to the 10% of children with activity limitation with the lowest scores. The second decile corresponds to the next 10% of children with activity limitation with the lowest scores, etc. The average score was calculated for each decile and a plot of this average score as a function of the decile was produced in Chart 4. As can be seen in Chart 4, no obvious cut-off points in the global severity score distribution exist. Several techniques were considered in order to create the severity classes. Because of the smoothness of the distribution in Chart 4 and because of the requirement to use a simple method for data users, an intuitive graphical method was used to determine the cut-off points.

After discussion with some data users, it was decided that the severity scale for children 5 to14 would be separated into 4 severity classes, Class #1 being the less severe and Class #4 the most severe. The creation of the classes has been done in two steps.

In a first step, an attempt was made to identify a "natural cut-off point" in the scale. Although this is not obvious, one can note that the beginning of the distribution is fairly linear up to 50th or 60th percentiles and then, the slope starts to increase more and more rapidly. This cut-off point in the trend of the distribution seems to correspond to a score around 1/8. This particular score corresponds to the score of someone with the maximum score for one type of disability and no points for the other types. Many such cases were found in the sample. Of course, there is a number of ways to obtain a score of 1/8. Because of the particular interpretation of this point, the cut-off was chosen to be exactly 1/8. This cut-off creates two groups:

The "less severe" group:				<	1/8
The "more severe" group:	1/8				1

For example, someone with the maximum score for seeing () but no other disability, falls into the "more severe" group. Anyone with the maximum severity score for one disability type will, therefore, automatically be assigned to the "more severe" group.

The second step was to separate each of the two groups into two subgroups. The most logical cut-off points to do this were half and twice the maximum score for one disability type (cut-offs of 1/16 and 1/4). The point 1/16 corresponds to someone who would have half the maximum number of points for one disability type and nothing else. The cut-off of 1/4 corresponds to someone with the maximum number of points for two disability types and nothing else. Therefore, respondents with a score smaller than 1/16 fall into Class #1, those with a score between 1/16 and 1/8 fall in Class #2. Those with a score between 1/8 and 1/4 fall in Class #3 while those with a score larger than 1/4 fall in Class #4.

Class 1:				<	1/16
Class 2:	1/16			<	1/8
Class 3:	1/8			<	1/4
Class 4:	1/4				1

This type of classification has the advantage of being easy to understand and interpret by all users. Because of the subjective aspect of this classification, it is desirable not using specific terms in order to avoid misinterpretation. The classes are relative to each other, that is Class #1 is less severe than Class #2, which is less severe than Class #3, which in turns is less severe than Class #4. However, a nomenclature has been assigned to those classes for practical purposes. The terms "mild", "moderate", "severe" and "very severe" are used to refer to classes 1 to 4.

For children 0 to 4, a similar approach was used. It was decided that the severity scale for children 0 to 4 would be separated into two severity classes rather than four. Since there are four disability types for that group, the cut-off is 1/4:

Class 1:				<	1/4
Class 2:	1/4			<	1

The terms "mild to moderate" and "severe to very severe" are used to refer to classes 1 and 2.

---

You need to use the free Adobe Reader to view PDF documents. To view (open) these files, simply click on the link. To download (save) them, right-click on the link. Note that if you are using Internet Explorer or AOL, PDF documents sometimes do not open properly. See Troubleshooting PDFs. PDF documents may not be accessible by some devices. For more information, visit the Adobe website or contact us for assistance.