[Futuregen] Cohort paper [revised version]

Thu Nov 25 18:24:35 CET 2021

Dear Stefan et al.,

I read all your responses, the beautifully composed revised manuscript, and
others' comments. All issues have already been covered and I have actually
nothing more to add except a few notes that you might (or not) want to add
to responses. Just to convince the reviewers that their concerns are well
understood and addressed.

1) You clarified in the response that you indeed used average marginal
estimates but this is not in the porse except when you talk about the
'margin' command of STATA. A good idea to include this in the manuscript as
well.

2) A very good idea that you removed "robustness of the estimates" from the
manuscript, a vague term anyway. You may want to clearly say this in the
response and also in the manuscript that  collapsing of countries into
regions was to increase sample size and hence enhancing statistical power.

3) I hear you and understand the hesitancy for use of an OLS regression for
ordinal variables. Generally speaking it is not wrong to do so but
interpretation of the coefficients is a little tricky because the variable
is in fact not continuous. One way to go around this issue is just to
mention the results of the sensitivity analysis in one sentence and add
(data not shown, available upon request). And tell the reviewer that you
did not want to make the appendix longer unnecessarily.

Fantastic work, I am very positive soon you'll have a publication!

Cheers!

Afshin .

On Wed, Nov 24, 2021 at 4:29 PM Ricardo Rodrigues <rodrigues at euro.centre.org>
wrote:

> Dear Stefan, dear Johan,
>
> I read through the changes and replies to the reviewers and I really did
> not have anything to add, except a couple of suggestions on the "if you
> really need to" basis below.
>
> Susan mentioned the need for a reference for the use of the square root
> equivalization: OECD (2011), Divided We Stand – Why Inequality Keeps
> Rising, Paris.    (www.oecd.org/social/inequality.htm /
> www.oecd.org/fr/social/inegalite.htm ).
>
> I think the reviewers were quite positive on their assessment of the paper
> in the first place and your replies are to the point and very good.
> You argued well enough for using a dichotomous variable. Reviewer 1 was
> the one who raised this issue and she/he was also the one who thought the
> results section was already too long: if you want, you can also mention
> this (text and annexes already too long, as pointed by the said reviewer
> 😉) to reinforce the use of the dichotomous option only. I would keep the
> OLS out.
> Reviewer 2 mentions the inclusion of complete cases only as a sensitivity
> analysis - and I think you purposely (and wisely) steered away from opening
> that can of worms and left this more or less unanswered. 🙂 One argument
> I could think of, for not using only complete cases (apart from
> exacerbating attrition), is the fact that we are interested in cohort
> differences rather than exclusively on longitudinal developments.
>
> It looks very good!
>
> Best wishes,
> Ricardo
>
>
> ------------------------------
> *From:* Futuregen <futuregen-bounces at euro.centre.org> on behalf of Susan
> Phillips <phillip at queensu.ca>
> *Sent:* Tuesday, November 23, 2021 9:00 PM
> *To:* Stefan Fors <stefan.fors at ki.se>
> *Cc:* LIST: FutureGEN <futuregen at euro.centre.org>
> *Subject:* Re: [Futuregen] Cohort paper [revised version]
>
> Hi Stefan,
>
> Your response to reviewers is really clear and covers everything well. I
> have made a few language edits and a comment or two in the manuscript
> (attached). I don’t feel like the wisest one to decide about the inclusion
> of the sensitivity analyses. My sense, from what you have written would be
> to not include.
>
> Thanks for including me.
>
> Susan
>
> On Nov 22, 2021, at 8:59 AM, Stefan Fors <stefan.fors at ki.se> wrote:
>
> Dear friends,
>
> I hope all is well with you! We have now tried to address the comments
> made by the reviewers for the cohort paper. Please find the revised
> manuscript, tables and figures, appendix, and responses to the reviewers
> enclosed. For those of you that are co-authors on the paper – read the
> revisions and rebuttal letter and let us know what you think (if you have
> time). The deadline for submission is Friday, so I would need any comments
> by Thursday in order to make any changes to the manuscript.
>
> One thing Johan and I have been discussing is whether we should include
> figures from our sensitivity analyses using ordinal indicators of ADL and
> IADL and OLS-regressions (see attached file) in the response to the
> reviewers (not for publication). On the one hand, this would serve as a
> response to their comments on the dichotomization of the outcomes and
> potential sensitivity analyses. On the other hand, we may invite a whole
> new host of comments (“why are these not mentioned in the paper?” “why are
> you using OLS-regressions for ordinal variables?” etc.). So, let us know if
> you have any opinions on this issue as well.
>
> All the best,
>
> Stefan
>
> *Stefan Fors* | Forskare
>
> Aging Research Center | Karolinska Institutet
> Widerströmska huset, plan 9
> Tomtebodavägen 18A  | 171 65 Solna
> +46 (0) 703 368 588
> *stefan.fors at ki.se <stefan.fors at ki.se>* | *www.ki.se/nvs
> <https://can01.safelinks.protection.outlook.com/?url=http%3A%2F%2Fwww.ki.se%2Fnvs&data=04%7C01%7Cphillip%40queensu.ca%7C6d67746ff5e541895e6a08d9adc07298%7Cd61ecb3b38b142d582c4efb2838b925c%7C1%7C0%7C637731867070702642%7CUnknown%7CTWFpbGZsb3d8eyJWIjoiMC4wLjAwMDAiLCJQIjoiV2luMzIiLCJBTiI6Ik1haWwiLCJXVCI6Mn0%3D%7C3000&sdata=jm2vg%2BUgZSGNDug4mujBhSRctbUPP83GYeGq9KIeWGk%3D&reserved=0>*
>  | *www.ki-su-arc.se
> <https://can01.safelinks.protection.outlook.com/?url=http%3A%2F%2Fwww.ki-su-arc.se%2F&data=04%7C01%7Cphillip%40queensu.ca%7C6d67746ff5e541895e6a08d9adc07298%7Cd61ecb3b38b142d582c4efb2838b925c%7C1%7C0%7C637731867070712606%7CUnknown%7CTWFpbGZsb3d8eyJWIjoiMC4wLjAwMDAiLCJQIjoiV2luMzIiLCJBTiI6Ik1haWwiLCJXVCI6Mn0%3D%7C3000&sdata=s2b229Ynz%2Fv4uZWXf3Igw%2Fh0KKs27SGq5809YAAjxIw%3D&reserved=0>*
> ______________________________________
> Karolinska Institutet - ett medicinskt universitet
>
>
>
>
> *När du skickar e-post till Karolinska Institutet (KI) innebär detta att
> KI kommer att behandla dina personuppgifter. *Här finns information om
> hur KI behandlar personuppgifter
> <https://can01.safelinks.protection.outlook.com/?url=https%3A%2F%2Fki.se%2Fmedarbetare%2Fintegritetsskyddspolicy&data=04%7C01%7Cphillip%40queensu.ca%7C6d67746ff5e541895e6a08d9adc07298%7Cd61ecb3b38b142d582c4efb2838b925c%7C1%7C0%7C637731867070712606%7CUnknown%7CTWFpbGZsb3d8eyJWIjoiMC4wLjAwMDAiLCJQIjoiV2luMzIiLCJBTiI6Ik1haWwiLCJXVCI6Mn0%3D%7C3000&sdata=%2FtKpdTCB%2F%2FaKzqjgbXK9iJQJ5xb%2FGFTg9%2FORkKI05Lk%3D&reserved=0>
> .
>
> *Sending email to Karolinska Institutet (KI) will result in KI processing
> your personal data.* You can read more about KI’s processing of personal
> data here
> <https://can01.safelinks.protection.outlook.com/?url=https%3A%2F%2Fki.se%2Fen%2Fstaff%2Fdata-protection-policy&data=04%7C01%7Cphillip%40queensu.ca%7C6d67746ff5e541895e6a08d9adc07298%7Cd61ecb3b38b142d582c4efb2838b925c%7C1%7C0%7C637731867070722568%7CUnknown%7CTWFpbGZsb3d8eyJWIjoiMC4wLjAwMDAiLCJQIjoiV2luMzIiLCJBTiI6Ik1haWwiLCJXVCI6Mn0%3D%7C3000&sdata=UxNXdC45HKKpHizf4rwTBy%2FscxV85TyK1OA4Yy5th2g%3D&reserved=0>
> .
> <Fors et al.R1.docx><Fors et al_Tables and figures_R1.docx><Fors et
> al_Supplementary materials_R1.docx><adl iadl OLS regressions.pdf>
> <Reply_reviewers_R1.docx>_______________________________________________
> Futuregen mailing list
> Futuregen at euro.centre.org
> https://lists.euro.centre.org/cgi-bin/mailman/listinfo/futuregen
>
>
>
> Susan Phillips MD, CCFP
> Research Director, Centre for Studies in Primary Care
> Rosser Research Chair, Family Medicine
> Director QuARMS
> Queen’s U.
>
>
>
>
>
>
>
-------------- next part --------------
An HTML attachment was scrubbed...
URL: <http://lists.euro.centre.org/pipermail/futuregen/attachments/20211125/d6e139c1/attachment-0001.html>