Talk:Screen for child anxiety related disorders

Ages

IS this really for 8-11 year olds? Or is it for "children and adolescents"?

The rubric "For children ages 8 to 11, it is recommended that the clinician explain all questions..." implies that it is intended to be used by older children as well.

All the best: Rich Farmbrough, 21:37, 25 August 2016 (UTC).

Scoring

It seems unlikely that a lesser score is needed to establish an anxiety condition than if one of the specific anxiety conditions is established. If this is the case, and I can imagine justifications, then the reason should be made clear. All the best: Rich Farmbrough, 17:40, 26 August 2016 (UTC).

Valid and reliable

"The child and parent Screen for Child Anxiety Related Emotional Disorders (SCARED) both yielded five factors: somatic/panic, general anxiety, separation anxiety, social phobia, and school phobia. For the total score and each of the five factors, both the child and parent SCARED demonstrated good internal consistency (α = .74 to .93), test-retest reliability (intraclass correlation coefficients = .70 to .90), discriminative validity (both between anxiety and other disorders and within anxiety disorders), and moderate parent-child agreement (r = .20 to .47, p < .001, all correlations)."

"Test-retest reliability" does not seem a good synonym for "reliability". Nor is it (without access to the paper) clear that discriminative validity is with respect to DSM, rather than the statistical factors.

All the best: Rich Farmbrough, 17:43, 26 August 2016 (UTC).

Serious cleanup and update needed

I have done quite a bit of editing to remove an external link farm, original research of unpublished COI and to note a recent secondary review at PMID 30081293 from which the entire article should have been written. Besides being incorrectly sourced (not to a recent secondary review), a good deal of this article's content was in breach of WP:NOT (a webhost or a How-To manual) and WP:ELNO. Cleanup and rewrite to the recent secondary review is needed, and PubMed also reveals much more recent validation of the instrument in other languages. SandyGeorgia (Talk) 15:21, 20 June 2020 (UTC)

This is a diff to all of my edits; @Ongmianli and Eyoungstrom: I suggest stepping through each one of my edits, beginning here, to begin a list of issues that should be addressed in all of the article created by this course, and to review the explanations given in each edit summary.

The most significant issues-- besides avoiding WP:COI editing and citing to unpublished COI research-- is to keep in mind what Wikipedia is and is not. It is WP:NOT a webhost or a how-to manual, articles should be encyclopedic in nature and for bio and health info, based on the WP:MEDRS compliant secondary reviews, rather than primary or commercial sources, and biographies should be based on what secondary sources say about a person, not what their self-published bios say about them, except in limited circumstances (eg, if birthdate or alma mater is self-published, we use that). Almost everything created by these courses will need these kinds of improvements. It will be most helpful to first teach students how to locate the most recent secondary review in PubMed, and then write articles around that, rather than doing their own searches for primary studies. Hope this helps, SandyGeorgia (Talk) 15:28, 20 June 2020 (UTC)

Reliability

Rubric for evaluating norms and reliability for the Screen for Child Anxiety Related Disorders

Criterion	Rating (adequate, good, excellent, too good*)	Explanation with references
Norms	Good	Multiple research studies of large relevant clinical samples.
Internal consistency (Cronbach's alpha, split half, etc.)	Good	Alphas ranging from .74 to .89
Inter-rater reliability	Not published	No published studies formally checking inter-rater reliability
Test-retest reliability (stability	Adequate	r = .86 over a median of 5 weeks
Repeatability	Not published	No published studies formally checking repeatability

Validity

Evaluation of validity and utility for the Screen for Child Anxiety Related Disorders

Criterion	Rating (adequate, good, excellent, too good*)	Explanation with references
Content validity	Excellent	Covers DSM diagnostic symptoms for anxiety disorders and specific phobias.
Construct validity (e.g., predictive, concurrent, convergent, and discriminant validity)	Excellent	Shows convergent validity with other symptom scales. SCARED significantly positively correlated with CBCL.
Discriminative validity	Good	AUC of .67, able to discriminate between children with anxiety versus non-anxiety disorders in clinical settings, as well as individual types of anxiety disorders.
Validity generalization	Good	Used in clinical settings for children and adolescents ages 9–18. Reliable across genders and ethnicities. SCARED has been translated into multiple languages with good reliability
Treatment responsivity	Good	Studies show sensitivity to treatment effects when children took the SCARED pre and post treatment suggesting the assessment is effective in both clinical and research settings
Clinical utility	Good	Free (public domain)

Discussion

See also discussion at WT:MED SandyGeorgia (Talk) 02:33, 28 June 2020 (UTC)

I added these tables back and removed the insufficient citations and cited an up to date Assessment Book by Hunsley & Mash 2018 that uses this table's format etc. Emmagch Emmagch, 23:27, 25 June 2020 (UTC).

Emmagch please review WP:COPYVIO. Wikipedia cannot use someone else’s table without a copyright release. This table previously was cited as: [1] Table from Youngstrom et al., unpublished, extending Hunsley & Mash, 2008. This indicates the table is taken from someone else’s work (unpublished) and a copyright release is needed. SandyGeorgia (Talk) 00:13, 26 June 2020 (UTC)

Also, why did you repeat the text on this talk page? SandyGeorgia (Talk) 03:55, 26 June 2020 (UTC)

The table is similar but not identical. In no way did this book talk about the measure or anything in the table. I am sorry for the confusion SandyGeorgia, this table is not the table in the book but it is inspired by the book based on its format and what to include when talking about evidence based assessment. Emmagch Emmagch, 23:24, 27 June 2020 (UTC).

SandyGeorgia, I added it to the talk page so we could have a discussion about it in case you were not pleased with me adding it back and for the sake of transparency. The reason why we originally cited Eric and the 2008 version of Hunsley and Mash book is because they were examples of clinical psychologists using this format to report information about psychometric properties of psychological tests, measurements, and questionnaires in a digestible way. I added this new citation because it is more up to date example of a way of delivering information. Also, it is a more reputable source and a reference that majority of clinicians and researchers would be aware of since it is based on a collection of many journal articles and works. The book is not a "primary source" in the general sense of one. Emmagch Emmagch, 23:43, 27 June 2020 (UTC).

SandyGeorgia, in case this was not clear, this is not my measure and it was not made by Eric Youngstrom. It is a free measure and so there are no royalties connected to it in any way and I am not a student of Eric Youngstrom's but I used to be but I was a student of a lot of psychology professors. I work in the neuroscience field now and my colleagues studying anxiety and the brain use this measure with their youth participants. I am also familiar with the SCARED due to reading and studying evidence-based assessments and because it is so widely used in the small child and adolescent clinical psychology community. There is no conflict of interest for me here and the measure. I just want to help those deciding on what measures to use for practice or research in a helpful way so this chart helps people compare measures. Emmagch Emmagch, 00:20, 27 June 2020 (UTC).

Understood ... in the big picture, that means you still have a WP:COI, though. I am also concerned that this Table methodology is using primary sources, when Wikipedia articles and medical content should be based on secondary reviews. We should not be advancing new formulations, which is WP:OR and WP:SYNTH. It is also a breach of WP:NOT to use Wikipedia to host novel presentations (not a webhost). Since this affects multiple articles, it is under discussion at WT:MED ... I will explain more there when I am not typing on an iPad, and send you the link. SandyGeorgia (Talk) 00:28, 28 June 2020 (UTC)

So I have a COI because I am familiar with the SCARED-R? When reading Wikipedia articles, I love it when they present the information in an "attractive" form and even go as far as making templates for presenting information in a certain way. This presentation is not novel but common and comes in multiple variations. It would be a disservice to readers to delete this table and then write its contents in a long paragraph when it could be represented in a table that provides the same information but more concisely. Emmagch (talk) 00:47, 28 June 2020 (UTC)

Hey SandyGeorgia, The Wikipedia article, Adderall, uses several tables especially the ones near the end that looks like a novel presentation to me but it is not one even though they never cited where they got the way of organizing information from:

Look at the "Summary of addiction-related plasticity" chart and the Adderall#Comparison to other formulations table. Emmagch (talk) 01:02, 28 June 2020 (UTC)

@Emmagch: So, it seems the content of the tables is not taken from the book, only the format. That clears up the copyright question. The problem then shifts to the source of the content for the tables. Let me ask a concrete question by way of example: in the last row of the 'Validity' table, the rating for "Clinical utility" is given as "Good". Where did that rating come from? Knowing the answer to that may help us unlock the means to moving forward. --RexxS (talk) 01:49, 28 June 2020 (UTC)

Emmagch, my apologies for the piecemeal response: I have arthritic fingers and an essential tremor, so prefer to read from my iPad, but typing there is difficult. I make many typos, but can type better when I move to a real computer, although I have other pain when I sit at a computer because a tree attached to a hammock I was lying in fell on me, tried to kill me, and left me half an inch shorter but alive and with a bad back. Again, my apologies. On a real computer with a keyboard now, but you may find the number of typos I make to be quite frustrating :) :)

Your potential WP:COI is because of your relationship with Eyoungstrom and that these articles may be furthering his work. I am not saying this is a problem per se; Eyoungstom has demonstrated himself to be very willing to take on board feedback, work to understand Wikipedia's policies and guidelines, and "work with us". We're trying to work that out here ... and what is in play is potential WP:SYNTH. That is, synthesis of primary sources to draw conclusions not published by secondary reviews ... and in this case, in ways that may be advancing the work of Eyoungstrom. That is what we are trying to determine, and I believe we can do it in such a way that you can continue editing productively. But we have to work through these concerns.

As examples, you give the tables at adderall. First, be aware of the futility of using WP:OTHERSTUFFEXISTS arguments on Wikipedia. Because "anyone can edit" you can find lots of stuff that isn't necessarily policy compliant. But in the case of adderall, you will see that everything in the first table (I encountered-- I did not check them all) is sourced to one secondary review. That is summarizing a secondary source. The problem here is not about presenting information in table format. In the case of adderall, information that came from a WP:MEDRS-compliant source was presented in table form. That's fine. It wasn't multiple primary sources, not previously reviewed by secondary sources, presented in table form to reach a previously unpublished conclusion.

Here is another place where information is presented in table form: Management of Tourette syndrome#Practice guidelines. All of that information is paraphrased and summarized from a MEDRS-compliant secondary source. Presenting it in table format is not the issue. The question here is, are you synthesizing information from multiple primary sources to present novel conclusions that have not been previously published by secondary sources? Or stated alternately, is Wikipedia being used as a webhost to present previously unpublished conclusions. And ... potentially ... to advance the work of one group. Which is why RexxS asked the question he asked.

Admittedly, we got off on the wrong foot with this course's previous work because of the way the information was originally sourced, as unpublished work of Eyoungtrom. But he has demonstrated a helpful attitude, such that I believe we can get to the bottom of this. But the question at this point is WP:SYNTH, and it is important that all of you understand this before proceeding forward with these templates. Regards, SandyGeorgia (Talk) 02:29, 28 June 2020 (UTC)

@SandyGeorgia -- thank you for the Tourette's example -- that is very helpful. I will study it, too, and that is a great idea about looking at the practice parameters. They get updated more slowly in psychiatry than other areas of medicine, and even more slowly in psychology (because psychotherapy trials take longer to run, and they have fewer sources of grant support: Pharma has zero interest in paying for psychotherapy trials or including psychotherapy as a comparator), so please do not be shocked if some of the dates on the parameters look, ummm, more "super-annuated" than would be ideal. I promise that I get the gist of looking for the most current reviews (and Cochrane, Campbell, and TRIP Database are tools of choice for that). The difference in size and speed of updates for Cochrane versus Campbell gives a sense of context for the constraints we are working within. Much appreciated! Prof. Eric A. Youngstrom (talk) 15:49, 28 June 2020 (UTC)

Example of Original research/SYNTH concerns

In the case of this topic (I have not checked all other articles by this group on assessment), there is a recent secondary review (meta-analysis) at:

• Runyon K, Chesnut SR, Burley H (November 2018). "Screening for childhood anxiety: A meta-analysis of the screen for child anxiety related emotional disorders". J Affect Disord. 240: 220–229. doi:10.1016/j.jad.2018.07.049. PMID 30081293.

Rather than report the conclusions of that secondary review in accordance with Wikipedia guidelines, the article uses a table of parameters with measurements drawn from primary studies selected by the authors of the article, and those primary study conclusions do not always agree with the secondary review. And, based on a 12-year-old rubric from one set of authors, which may or may not (???) be the same rubric used by modern secondary reviews. (That is, we don't know if the differences in rating are because the rubric is outdated, or the primary sources used are outdated.)

But one thing that is clear is that everything cited in the tables is very old, with dates ranging from 1997 to 2008, while we have a 2018 review that draws different conclusions. Wikipedia should not be advancing its own ratings of assessment instruments, and particularly not when those ratings differ with recent reviews.

This demonstrates the problem with using original research to write articles. And reinforces the concern that, in these cases, these assessment articles are hosting the work of one group on Wikipedia in ways not in accordance with Wikipedia's policies and guidelines. Emmagch what say ye about the secondary review I listed above and added to the first sentence of the article? What reasoning can you provide for citing Reliability and Validity parameters to 1997 and 1999 studies, when a 2018 meta-analysis is available? SandyGeorgia (Talk) 13:58, 28 June 2020 (UTC)

Hi, all! I just got caught up on this thread. I am downloading the Runyon meta-analysis, and I will read it and compare. The SCARED was reviewed and recommended in multiple other narrative reviews, including some that are 2020 or in press, so we are likely to be able to address that vis the SCARED. For other measures that are less studied, it may not be feasible to use a table and summarize like Hunsley & Mash and others have done, because the citations in the cells are primary sources. I agree with the thread above -- the pattern of the table is not copyrighted; the citations are intended to show that the *structure of the table* and the rubric are not original research on Wikipedia (and also to give readers a sense of how this connects to the way the field is thinking about things. But having to populate cells of the table with secondary sources on Wikipedia will require a different way of looking at things, and won't be possible for some of the measures. That's fine. Better to do it right, and we can move more of the detail about the underlying literature onto pages on Wikiversity for the people who are teaching assessment or are interested in the granularity. So it looks like we are getting things in better focus. I appreciate all of your help with this! It may take 48-72 hours for me to weigh back in -- I need to see if Runyon disagrees with Beidas et al., Becker-Haimes et al., and the Kendall chapter substantively. I am also working on a team grant (Conte Center application, my part focused on assessment of negative valenced RDoC domain in humans; some of the other projects are mouse model), and that deadline looms, so I am forced to divide my attention. The Wikipedia work remains a high priority -- I am only trying to provide context so that my slower cadence is not interpreted as a lack of interest or engagement. :-)

Best to all!Prof. Eric A. Youngstrom (talk) 15:48, 28 June 2020 (UTC)

It sounds like you are following the issues correctly ... and it looks like the citation to Hunsley & Mash may not even be necessary (??), since we aren’t actually citing anything to them? Much of the original confusion here over copyvio came from how the table was initially cited, and I agree that we are making progress here. Regards, SandyGeorgia (Talk) 15:55, 28 June 2020 (UTC)

I agree that the contents and information of this article should be updated with the best summation of trustworthy sources. I imagine the other psychological assessments that we edited years ago need to be updated too in some capacity. We want the most helpful and up to date information on Wikipedia that reflects what the research body is saying. I can help Eric with the literary review for the SCARED-R so that this table is up to date and reflecting better sources. Thank you for the guidance and patience! I am planning on taking the GRE in August but I have not studied much and so today I am trying to make head way on that. Monday, I can talk to Eric about what needs to be done and divide work up for this table and go from there. I have more access to articles than most since UM Libraries subscribes to a lot of journals and so I think my literary search will have less barriers. RexxS SandyGeorgia Eyoungstrom Emmagch (talk) 16:30, 28 June 2020 (UTC)

There is no hurry; what is important at this stage is that you all understand the concerns. Good luck with the GRE! SandyGeorgia (Talk) 16:35, 28 June 2020 (UTC)