Talk:Paternal age effect

Paternal Age
Is there any part of the paternal age effect that belongs in Wikipedia? —Preceding unsigned comment added by Anniepema (talk • contribs)
 * It really needs to be peer-reviewed research which has been published in a reputable scientific journal. See No original research. -Fsotrain09 17:31, 27 December 2006 (UTC)

Large parts of this page are copied/pasted from this link: http://www.uthscsa.edu/opa/issues/new33-32/graduate.htm

Article claims that at least one X-linked condition is influenced by paternal age; no source cited for this questionable claim. —Preceding unsigned comment added by 72.224.135.176 (talk) 23:48, 2 August 2008 (UTC)

hemophilia
as far as i can tell, hemophilia is a genetic disorder that's inherited through two recessive genes. How the age of the father affects this, i don't know. How would his age affect a gene he had in his DNA from birth? —Preceding unsigned comment added by 167.7.17.3 (talk) 19:55, 27 August 2008 (UTC)

de novo mutations increase with advancing paternal age —Preceding unsigned comment added by 156.145.79.132 (talk) 03:13, 22 October 2009 (UTC)

Neutral tone
The tone of this article is strange for a medical article. Something like this "If my son or daughter was to ask..." isn't very scientific. Bhny (talk) 20:26, 24 December 2013 (UTC)

Suggestions
Definition


 * MedicineNet.com lists two types of paternal age effects. The two types are autosomalmutations and an indirect paternal age effect from mutations on the X chromosome.

Clinical implications


 * MedicineNet.com also states that there is no universal definition of advanced paternal age, but does suggest that in the realm of genetic counseling, all men 40 yrs and older at the time of conception meet the criterion.

Pathophysiology


 * Commenting on the study of 78 Icelandic families, Harry Fisch, MD, clinical professor of urology and reproductive medicine at Weill Cornell Medical College of Cornell University, suggests that now men too have a reference point in decisions of advanced paternal age and risk for genetic defects. The article cites that for women the age of 35 is a benchmark in determining the age of increasing escalation of genetic defect risks and that now men can assume a doubling of the mutation rate every 16 years.

Wilson.3308 (talk) 04:07, 30 September 2014 (UTC)

In the History section of the article, a more developed history of how the Paternal Age Effect was discovered, came about could be included. In the Notable Conditions and Diseases section, many diseases are listed at the bottom without explanation. If enough information is available, more common conditions and diseases should be elaborated on, namely cataracts, heart defects, hemophilia, Klinefelter's Syndrome. Go more in-depth in the Semen & Sperm Abnormalities section, taking more about the actual studies that were preformed and the results they had. Also include aneuploidy as potential abnormality.


 * I agree that Autism Spectrum Disorder and Bipolar Disorder are missing links to the Wiki pages in the Notable Conditions and Diseases section. Is a free full text available for your citation on the aneuploidy study? And caution, the word performed is not spelled correctly in your edit change. Wilson.3308 (talk) 04:01, 5 October 2014 (UTC)
 * I Thank you for the feedback. I've gone and corrected the error. I'm still trying to find the full text for the article.   — Preceding unsigned comment added by Collins.1128 (talk • contribs) 02:44, 15 October 2014 (UTC)

RASopathies such as achondroplasia and Noonan are not mentioned.

Image



 * I would like to add the following image to the article to help portray the meaning and improve upon the page.Wilson.3308 (talk) 02:57, 11 October 2014 (UTC)

Edit

 * As per my assignment, I have edited the previous "other conditions" section. It is now labeled (PAE) Disorders, mechanism, and other conditions. This edit lists the disorders collectively known as (PAE) disorders and briefly mentions the "Selfish Selection" mechanism as outlined by Goriely A, and Wilkie AOM (2013). Also, so as to not have the disorders listed twice, I have removed these (PAE) disorders from the long list of other conditions that follow. Wilson.3308 (talk) 00:49, 20 November 2014 (UTC)

Article is poor
Refs need to be improved per WP:MEDRS. Doc James (talk · contribs · email) 06:16, 12 December 2014 (UTC)

Remove sections "Paternal mortality" and "Fertility"?
The two subsections on "Paternal mortality before adulthood of child" and "Fertility" are not really paternal age effects, in the sense that these two effects relate primarily to the father, not to the offspring. Delete or move to some other more appropriate Wiki article? Suggestions? — Preceding unsigned comment added by 86.137.100.88 (talk) 09:55, 3 May 2015 (UTC)

Message to Doc James
Good morning Doc James. Today you slimmed down the article considerably by replacing detailed references with summary statements, and by restricting the definition of "paternal age effect" to a health problem rather than a general biological effect (see definition in the first sentence).

I can see the advantage in this new format for a time-pressed medical practitioner (this makes the Wiki-page a quick-reference source), but it is now less useful for the researcher trying to enter this increasingly active field of research. And there is nearly a complete lack of quantification - is the risk 1 percent? 40 percent? Who knows. The disappointed reader needs to return to PubMed, which undermines the purpose of Wikipedia, methinks.

I am not suggesting you revert things now, but please do not discourage future editors from adding some more flesh on the bones. — Preceding unsigned comment added by 81.131.171.56 (talk) 11:16, 28 May 2015 (UTC)
 * We need to use recent secondary sources. Not masses of 30 year old primary sources. This article has huges issues with its references. It needs to be updated per WP:MEDRS.
 * Health effects are what people care about. General biological effects may or may not be mechanisms of the resulting health effects Doc James  (talk · contribs · email) 11:37, 28 May 2015 (UTC)
 * I too care about health effects. But the field is not yet sufficiently advanced to be sure on most of these health effects. So far, only the biology is secure. The health effects must await further research. So it is imbalanced to make this article all about medicine, and put biology in the closet. Wait another 5 years and then you may be right, but it is too early now. — Preceding unsigned comment added by 81.131.171.56 (talk) 12:12, 28 May 2015 (UTC)
 * wikipedia content should be based on the most recent secondary sources we can find. you need a very good reason to use very old primary sources. Jytdog (talk) 12:29, 28 May 2015 (UTC)
 * The mechanism per WP:MEDMOS goes lower in the article and gets seperated from health effects as they are not. Yes we all agree this area has a poor amount of research thus the article should likely be shorter. Doc James  (talk · contribs · email) 12:31, 28 May 2015 (UTC)
 * Right - conclusions of research published in primary sources, that do not get picked up in reviews, are unlikely to be worth discussing in WP. Jytdog (talk) 12:34, 28 May 2015 (UTC)
 * I trimmed primary and very dated references--Ozzie10aaaa (talk) 12:57, 28 May 2015 (UTC)

I reverted Ozzie10aaaa's changes because in several cases he removed dated references (such as a 2003 review) while leaving the associated claim, now unreferenced. If all references are removed, the claim should also be removed. If not, a 2003 review is better than no reference at all. KateWishing (talk) 15:08, 28 May 2015 (UTC)

Doc James and Ozzie:
 * I confess I am curious how on earth you manage to make large numbers of changes within minutes. Do you have a team working for you?


 * Secondly, although I am not entirely happy with the strong medical vs biological emphasis, I think the article overall is better now - can you upgrade its rating?


 * And I think we can now remove the warning that the references are poor. — Preceding unsigned comment added by 81.131.171.56 (talk) 16:03, 28 May 2015 (UTC)
 * I type fast. There is no team :-) The article still needs work. I am okay with older reviews as this is not a well researched area. Hopefully they are within the last 15 years though. Doc James  (talk · contribs · email) 12:46, 29 May 2015 (UTC)

Ozzie referencing
I am concerned by Ozzie10aaaa's method of adding references, which seems to involve searching related words on Google Books. Often, the results are only vaguely related to the article text. Here are examples from his recent edits, with unsupported text bolded:

KateWishing (talk) 17:24, 28 May 2015 (UTC)
 * Doc James I believe the #1 and #3 references are fine...in regard to the #2 reference, the statement in question has been supported by two references, have therefore adjusted the text accordingly to the reference in question (please revert or delete should you believe inadequate)...(update) ...I had to delete the #2 ref due to an IP reverting the additional text --Ozzie10aaaa (talk) 23:05, 28 May 2015 (UTC)
 * The references and our article should convey exactly the same thing in different words. You can't just take a pre-existing sentence and source it to a vaguely related, but distinct statement in a book. None of these sources are saying the same thing as our article, and #2 and #3 are barely similar. These are not isolated examples, but a general pattern in your editing. You should be more careful to only add a source when it supports precisely what our article states.
 * You've attempted to reword the #2 material to match the source, but in taking it out of context you produced nonsense. What "changes?"
 * I'm sorry if I am being harsh. KateWishing (talk) 23:42, 28 May 2015 (UTC)

Recent review
May 2015  Doc James  (talk · contribs · email) 16:20, 30 May 2015 (UTC)

Doc James please comment on Diabetes type I
Hi Doc James,

On 28 May 2015 you removed the diabetes subsection (which cited research up to 2005), and replaced it with a 2010 textbook reference. I have now looked up the textbook, and it provides no data, only a reference to Peng and Hagopian 2006 (the abstract does not mention paternal age at all) and Baan et al. 2005 (an obscure Dutch governmental report I cannot access). As you made the change, can you please access those two papers and provide the risk factors mentioned there, if they exist? I would like to see the information included in the Wiki article. My suspicions/concerns are that (a) the textbook you cite is not a useful reference in this instance and (b) the weight of evidence suggests that there is potentially no paternal age effect for diabetes type I, and therefore the Wiki article as it stands is medically misleading. To underline point (b), see the Lammi et al 2007 (http://www.ncbi.nlm.nih.gov/pubmed/17943268) which finds no paternal age effect for diabetes type I. For your convenience, here is the section as it was before your replacement on 28 May 2015:

(now moved to article page)


 * The source says "other possible risk factors include stress, higher maternal or paternal age at birth" which I have summarized as "A higher paternal age is a possible risk factor for type 1 diabetes" Which is basically what all the primary sources said but in fewer words. We do not need to follow up on the sources that this book is using. Doc James  (talk · contribs · email) 09:06, 31 May 2015 (UTC)
 * "We do not need to follow up on the sources that this book is using." - Are you referring to Wiki guidelines here? If so, which one please? — Preceding unsigned comment added by 86.137.100.104 (talk) 09:31, 31 May 2015 (UTC)


 * There are apparently no dedicated reviews about paternal age and diabetes, but this 2010 meta-analysis of maternal age says: "Although children with older mothers are more likely to have older fathers, there is no clear association between paternal age at delivery and type 1 diabetes (10,11,19,28,34)." This 2011 diabetes textbook states results for paternal age have been inconsistent. So the current wording of "possible risk factor" is accurate, but should probably reflect more of the uncertainty in the literature. KateWishing (talk) 14:25, 31 May 2015 (UTC)
 * I added these sources to the section. KateWishing (talk) 16:59, 31 May 2015 (UTC)
 * Thanks Kate - grateful for all your professional detective work on this article. On a related matter - would you be interested in mediating between me and Jytdog (see below)? If not, can you recommend a mediator? From my point of view, all the mediator has to do is access the two articles within the 2010 book, and search them with a word processor for "Paternal age effect", which will take milliseconds. — Preceding unsigned comment added by 86.137.100.104 (talk) 17:16, 31 May 2015 (UTC)
 * The 2011 book says "paternal age have been less consistent". Am happy with the clarifications by User:KateWishing Doc James  (talk · contribs · email) 03:02, 1 June 2015 (UTC)

further comment
Hi again Doc James, On 28 May 2015 you removed the "Diabetes" subsection, and replaced it with a 2010 textbook reference. As mentioned above, I looked up the textbook, and it provides no data, only a reference to Peng and Hagopian 2006 (the abstract does not mention paternal age at all) and Baan et al. 2005 (an obscure Dutch governmental report which also contains no mention of paternal age). Thus the 2010 book reference is a "fake" with regards to diabetes and the paternal age effect. In case you cannot access the Baan et al. 2005 report yourself, I am pasting the contents below, so you can see its irrelevance for yourself.

I am therefore reinstating the old diabetes section, and urge you in future to read references carefully before you cite them on Wikipedia.

Baan et al. 2005 Contents:

Summary 1. Introduction Part I The Chronic Disease Model and diabetes 2. The Chronic Disease Model (CDM) and diabetes 3. Adaptations to the CDM with respect to diabetes Part II Diabetes input data in the CDM 17 4. Incidence, prevalence and mortality of diabetes 17 4.1 Introduction 4.2 The Dismod method 17 4.3 Incidence and prevalence data 18 4.4 Mortality data 5. Diabetes health care utilization, costs and quality of life 29 5.1 Diabetes health care utilization 29 5.2 Diabetes costs 5.3 Diabetes Quality of life 33 Part III Prevention of diabetes 35 6. Risk factors for diabetes incidence and primary prevention 35 6.1 Introduction 6.2 Methods 7. Risk factors for diabetes incidence 39 7.1 Body Mass Index 39 7.2 Physical inactivity 42 7.3 Smoking 7.4 Alcohol 7.5 Combination of risk factors 49 7.6 Conclusion risk factors for diabetes incidence 50 8. Primary prevention 8.1 Lifestyle interventions 51 8.2 Pharmacological interventions 52 8.3 Primary prevention conclusions 53 Part IV Prevention of diabetes complications 55 9. Macrovascular complications of diabetes and tertiary prevention 55 9.1 Introduction 9.2 Methods 9.3 Prevalence of macrovascular complications in individuals with diabetes 59 10. Risk factors for macrovascular complications in individuals with diabetes 61 10.1 Overweight 10.2 Physical inactivity 65 10.3 Smoking 10.4 Total cholesterol 70 10.5 Hypertension 10.6 HbA1c 11. Tertiary prevention 79 11.1 Interventions to reduce bodyweight 79 11.2 Strict control of blood glucose 80 11.3 Blood pressure control 80 page 6 of 145 RIVM report 260801001 11.4 Lipid control 11.5 Conclusions tertiary prevention 81 12. Discussion and conclusions 83 — Preceding unsigned comment added by 86.137.100.104 (talk) 13:10, 31 May 2015 (UTC)


 * this edit is not OK. The sources are old and PRIMARY and don't comply with MEDRS.  The content overall is the editor's WP:SYN, building an argument by stringing together primary sources.  We don't do that here - we instead summarize secondary or tertiary sources. Please don't edit war, but instead continue to talk.  Thanks. Jytdog (talk) 13:50, 31 May 2015 (UTC)
 * Jytdog, I continue to be happy to talk. You know from the above that the 2010 book reference is inappropriate, as it does not cite any paternal age effect study, agreed? I will wait an hour or so for you to respond, and then revert your mistake, hope that is fine with you. Alternatively, please delete the 2010 book and provide a genuine reference within the next hour, and then I do not need to revert. As it is a medical matter, I think it is urgent for all of us to contribute constructively in fixing this, so your expert input would be welcome.
 * There is WP:NODEADLINE here. Do not edit war.  All of us are volunteers and work on our own time.  I will try to respond soon. Jytdog (talk) 15:11, 31 May 2015 (UTC)
 * Great that you are on board. And please do not suspect edit wars around every corner - this is the second time you make the claim, which is impolite. Deadlines are sensible here because we are working across very different time zones, and medical matters are a human priority. Let me know within the hour when you will do it. Meanwhile I am commenting "citation needed", because I doubt you will find a dedicated review, possibly not even a pertinent textbook (see Kate above). But I live in hope. — Preceding unsigned comment added by 86.137.100.104 (talk) 15:46, 31 May 2015 (UTC)
 * restoring the content was edit warring already. promising to restore it in an hour regardless of discussion here, is a promise to edit war.  Please don't edit war. adding a cn tag is POINTy., but whatever.  nothing in wikipedia is urgent other than threats of personal harm, BLP violations and COPYVIO.   please slow down.  Thanks. Jytdog (talk) 15:51, 31 May 2015 (UTC)
 * actually removing the source and adding a cn was entirely inappopriate. The cited source says of T1D: "Other possible risk factors include stress, higher material or paternal age at birth...".  The content says "A higher paternal age is a possible risk factor for type 1 diabetes." The source supports the content and is a MEDRS source.  I have given you an edit war notice.  Jytdog (talk) 16:00, 31 May 2015 (UTC)
 * Hi Jytdog. So we need to go to dispute resolution. But before that, very briefly, please answer the following question with a simple yes or no: have you read the two references cited on p301 in the 2010 book? Just a simple yes or no please. — Preceding unsigned comment added by 86.137.100.104 (talk) 16:11, 31 May 2015 (UTC)

slow down. i have only just begun to discuss this with the other editors here. The book is a reliable source per WP:MEDRS. Per WP:MEDASSESS (part of MEDRS) we do not conduct peer review on reliable sources. Please read that link and confirm that you understand it. But i do agree that neither of the refs used mention paternal age. I went onto pubmed to look for other secondary sources on paternal age and risk of diabetes. (search is here)
 * is a false hit
 * is a false hit - doesn't mention diabetes risk of offspring, just the risk of the older men getting diabetes
 * (old, from 2001) is a false hit - it discusses many risks to the child, correlated with paternal age, but doesn't mention paternal age.

I also went and looked for other books. this one: discusses many risks of advanced paternal age, but doesn't mention diabetes.
 * Tournaye, H. Male reproductive ageing. Chapter 10 in Reproductive Ageing: Royal College of Obstetricians and Gynaecologists Study Group. Eds Bewley S et al. Cambridge University Press, 2009 ISBN 9781906985134

this one has good content. In a section on risks, it says: "Results for birth order and for paternal age have been less consistent" (compared to maternal age) and cites and.
 * Stene LC et al. Epidemiology of Type 1 Diabetes. Chapter 3 in Textbook of Diabetes, 4th Ed.  Eds. Holt RIG et al.  John Wiley & Sons, 2011 ISBN 9781444348064

i see that kate had already found the same book, and cited it, along with a primary meta-analysis source. perfect!!! I left that, but removed the PRIMARY. This is a lot of fuss over something that we just can't say much about, as of now. We may be able to say more later, but we cannot much now. Jytdog (talk) 17:13, 31 May 2015 (UTC) (amend Jytdog (talk) 17:39, 31 May 2015 (UTC))
 * I don't mind that you removed it, but I did not add any primary sources. This is a meta-analysis, and the relevant comment (quoted above) is from the Discussion section in any case. KateWishing (talk) 17:21, 31 May 2015 (UTC)
 * you got me again! you are very good, it is a pleasure to work with you. Jytdog (talk) 17:38, 31 May 2015 (UTC)


 * Hi Kate and Jytdog, glad that we have managed to resolve it without killing each other. Have a nice remaining Sunday. — Preceding unsigned comment added by 86.137.100.104 (talk) 17:26, 31 May 2015 (UTC)
 * great. Jytdog (talk) 17:38, 31 May 2015 (UTC)

review /2010 /
might want to try this reference, its a Diabetes mellitus type 1 reference under Paternal age and outcome of offspring section (should you have difficulty  use CTRL+f for word find) --Ozzie10aaaa (talk) 18:20, 1 June 2015 (UTC)

selfish gene
The following content was added today:

"These three genes are thought to be particularly prone to a paternal age effect due to selfish spermatogonial selection, whereby the influence of spermatogonial mutations in older men is enhanced because cells with certain mutations have a selective advantage over other cells (see § DNA mutations). sourced to: {cite journal"}

Two issues with this. First, the "three genes" mentioned in the prior paragraph are FGFR2, FGFR3, and RET. The Goriely source discusses FGFR2, FGFR3, but not RET. Instead, it discusses "several members of the RTK/RAS and associated MAPK (Mitogen Activated Protein Kinase) signalling pathways, such as PTPN11/SHP2 and HRAS".

Second, I think we need to treat the Goriely source gingerly. Although it is classified by both PUBMED and MEDLINE as a "review", if you read it, what they do is review "epidemiological studies that the offspring of older fathers have an increased risk of neurodevelopmental disorders such as schizophrenia and autism" and then "present a novel mechanism that may contribute to this association." - namely, the "selfish spermatogonial selection" hypothesis that is the focus of the edit. The source is PRIMARY for that hypothesis and per MEDRS and NOT (which says that articles should describe "accepted knowledge") we should not include content about hypotheses like this unless they are discussed in review articles - for real. Hope that makes sense. ( this is exactly like the article we discussed at Finasteride btw.) Jytdog (talk) 23:11, 7 July 2015 (UTC)
 * I am not as concerned about lower quality sources being used to support mechanisms rather than effects. This should generally occur in a mechanism section though. Doc James  (talk · contribs · email) 04:00, 9 July 2015 (UTC)


 * My suggestions: 1. Move the "single-gene disorders" section to the top of the list. Because this is the first paternal age disorder ever characterised (1955), and furthermore it seems to be the disease with the most pronounced and uncontroversial effect. 2. Remove the GorielyWilkie2013 reference from here, and instead link the technical term "selfish spermatogonial selection" to the DNA mechanisms section which explains the selfish mechanism along with the GorielyWilkie2013 ref. Admittedly these changes do not solve the RET reference question, but let us do one thing at a time. I am implementing these changes in a minute. Revert if you disagree. — Preceding unsigned comment added by 86.137.103.77 (talk) 10:59, 9 July 2015 (UTC)
 * Please slow down. That seems to be the opposite of what Doc James just recommended. Jytdog (talk) 11:29, 9 July 2015 (UTC)


 * Selfish spermatogonial selection is discussed in 4 of the 6 reviews I've been consulting (Malaspina 2015, Kovac 2013, Ramasamy 2015, Abbas 2015). I don't think it needs to be mentioned in the "Health effects" section, though. Details about the mechanism belong in the Mechanism section, and selfish selection has been proposed to play a role in more than just single-gene disorders (schizophrenia, autism, cancer). We could add a mention of the FGFR genes to the Mechanisms section if that's the problem (RET is probably affected as part of the Ras pathway, but the secondary sources don't mention that). KateWishing (talk) 12:39, 9 July 2015 (UTC)


 * Kate, your suggestion: "We could add a mention of the FGFR genes to the Mechanisms section if that's the problem." Exactly. Please do so. I am not keen however on deleting the brief mention of the mechanism in the Health Effects section - some other Health Effects subsections also make brief mention of mechanistic aspects. The article would become unreadable if health effects were entirely separated from mechanisms, without any cross-referencing.
 * I don't care much whether it's mentioned in the Effects section. In the interest of resolving the current dispute, perhaps we could cite this review in the Effects section and change the text to match: "The FGFR genes are possibly particularly prone to a paternal age effect due to selfish spermatogonial selection, whereby the influence of spermatogonial mutations in older men is enhanced because cells with certain mutations have a selective advantage over other cells." The paywalled review states: "Interestingly, there is a discrepancy between the frequency of these FGFR mutations in sperm DNA and the effect of advanced paternal age on these syndromes, possibly due to selfish spermatogonial selection (32). Although harmful to embryonic development, these mutations might be paradoxically enriched because researchers suggest that they confer a selective advantage to the spermatogonial cells in which they arise (33, 34)." The evidence seems to be thinner for RET, anyway; this primary source says: "The quantitative data supporting selfish spermatogonial selection have emerged from studies of FGFR2 and FGFR3 mutations, but the same principles are likely to extend to the other PAE disorders." KateWishing (talk) 23:31, 9 July 2015 (UTC)
 * Great, Kate.

Misplaced sections?
Off-topic, but what is the section "Paternal mortality before adulthood of child" doing in the "Health Effects" section? When a father dies, that is primarily a health outcome for the father, not for the offspring. And if the unknown contributor intended paternal death to be considered a health risk for the child, then suitable literature should be cited. The cited table is informative in its own right, so I do not suggest it should be deleted, but should be put into appropriate context/moved to a new/relevant section. — Preceding unsigned comment added by 86.137.103.77 (talk) 11:22, 9 July 2015 (UTC)
 * I was considering renaming that section to just "Effects" so that I could move "Social associations" there, which does not really belong in the Mechanisms section. KateWishing (talk) 12:39, 9 July 2015 (UTC)


 * I disagree. I did not write the Social section, so can only guess what the author intended, but it seems to me that the author was trying to point out confounding factors when seeking biological explanations for paternal age effects. To give a completely hypothetical example, low IQ might correlate with very young parental age; the naive biological explanation might be that very young parents have poor quality DNA. Or, as this section may be trying to caution, very young parents generally have excellent DNA, but dumb kids are more likely to fool around sexually and produce more dumb kids. Hence giving the false impression to the unwary researcher that young parents generally have bad DNA. If you agree to my interpretation, any ideas how to clarify this? — Preceding unsigned comment added by 86.137.103.77 (talk • contribs)
 * The source presents those factors as advantages of delayed parenthood rather than confounding variables, so your implication would be synthesis for now. I plan to do the following unless someone objects or I'm struck down by laziness:
 * Rename "Health effects" to "Effects"
 * I remain unkeen on this, as the currently well-defined Health Effects section might then re-expanded by other contributors in the future to include a hotch-potch of psychological, social, etc effects (you can tell I am not a social scientist!). Can you instead introduce a new section "Social Effects" or similar? That could then probably include the paternal death table.


 * Merge "Social associations" into "Paternal mortality before adulthood of child" under "Effects," renaming the section "Parenting quality"


 * See above.


 * Look for sources that discuss confounding variables explicitly and create a new "Confounders" section under "Mechanisms"
 * KateWishing (talk) 23:31, 9 July 2015 (UTC)
 * Excellent. But you should not go to great lengths just because I made a mistake.

Long telomeres ≠ longevity
I've removed the sentence "Some research even indicates a longevity advantage for offspring of older fathers. " from the introduction. That research refers to longer telomeres. Longer telomeres aren't the same as longevity and in fact several studies show a mortality increase with paternal age. I hope to edit these in in the near future, but have removed the misleading sentence for now. --Ruben (talk) 23:17, 8 September 2015 (UTC)
 * Hi Ruben. I have resurrected the Eisenberg reference in the Introduction, but reworded it to accommodate your criticism. Eisenberg et al. suggest "a mechanism by which humans could extend late-life function as average age at reproduction is delayed within a lineage." I believe this Eisenberg speculation is a useful counterpoint to the (equally unfounded) Crow speculation in the introduction. It shows the reader just how much disagreement there is in the young research field of aging. — Preceding unsigned comment added by 86.137.100.56 (talk) 11:35, 27 September 2015 (UTC)
 * I'll dig up the references when Im back from Travel (eg Gavrilov, more recently adam haywards 2015). So Crow's speculation may be exaggerated bit not unfounded while Eisenberg just shows something completely different. --Ruben (talk) —Preceding undated comment added 09:30, 28 September 2015 (UTC)
 * back from travel now. Some references documenting an increase in child mortality (some also on suicide, all-cause mortality) with paternal age attached (those factors lower longevity, postulating an opposite and stronger effect on longevity is not parsimonious). Even though these are primary sources (based on very large datasets though), I think it suffices to show why Eisenberg's speculation (also a single study, not a review) should not be in the introduction. Do you agree?
 * Ruben (talk) 09:18, 2 October 2015 (UTC)
 * Ruben (talk) 09:18, 2 October 2015 (UTC)
 * Ruben (talk) 09:18, 2 October 2015 (UTC)
 * Ruben (talk) 09:18, 2 October 2015 (UTC)
 * Ruben (talk) 09:18, 2 October 2015 (UTC)
 * Ruben (talk) 09:18, 2 October 2015 (UTC)
 * Ruben (talk) 09:18, 2 October 2015 (UTC)

External links modified
Hello fellow Wikipedians,

I have just added archive links to 1 one external link on Paternal age effect. Please take a moment to review my edit. You may add after the link to keep me from modifying it, if I keep adding bad data, but formatting bugs should be reported instead. Alternatively, you can add to keep me off the page altogether, but should be used as a last resort. I made the following changes:
 * Attempted to fix sourcing for http://ender.bu.edu/~tgardner/be209/lectures/12/articles/Crow.JF_NatRevGen_00.pdf

When you have finished reviewing my changes, please set the checked parameter below to true or failed to let others know (documentation at ).

Cheers.—cyberbot II  Talk to my owner :Online 13:27, 31 March 2016 (UTC)

Updating citations
I'm trying to update this page with references I've accumulated in my own research on paternal age. I haven't made many substantial contributions to Wikipedia in the past, but strive to do more in the future, so please tell me where/if I violate tacit or explicit norms. Here's some notes on decisions that I felt I might have to justify.

- In some cases, I have added primary research studies, but only as per WP:MEDRS when the studies are large and fairly conclusive on their own (such as epidemiological studies based on hundreds of thousands or millions). Perhaps more controversially, I've also sometimes done so where older primary sources were already being cited, even if I couldn't identify the reason why they were considered citable. Here, I assumed a tacit norm that these are okay to cite (as they revolved more about non-medical traits and epidemiological work, where single studies are often very large). In these cases, I did not find a review of the evidence, but leaving the older studies and not citing newer studies with sample sizes that were orders of magnitude larger and improved methods did not seem right. It was easier to find reviews for medical traits than for non-medical associations.

- I removed the sentence: "Regarding the increased risk at very young paternal ages, an international study indicates that the DNA mutation rate in very young fathers may also be elevated.[1]"

It's a primary source based on a much weaker method (mutations in a very small, very mutable sequence as opposed to whole genome sequencing and comparing parents and children). There are now multiple studies comparing parents' whole genomes and exomes with their children and they report no such thing (reviewed in Ségurel).

- I removed the sentences: "Because people with longer telomeres are at decreased risk for age-related diseases, higher paternal age may also be associated with certain health benefits.[1] This mechanism may have evolved because the environment of children born to older fathers is likely to have a higher expected age of reproduction.[2]"

This is already mentioned in the Health effects: Life expectancy section and repetitive (and feels in the wrong place) here. There it is placed in more context, namely that several health effects on mortality have been documented in the opposite direction (decreasing longevity with higher paternal age). In the telomere section, I have instead now cited a meta-analysis showing both high heritability and a paternal age effect on telomere length and refer to the problems pointed out by Gratten et al. (genetic propensities for long telomere length may predict parental reproductive timing). Ruben (talk) 13:20, 24 February 2017 (UTC)


 * User:Doc James removed the primary sources I had added in this edit, pointing me to WP:MEDRS and to the tag at the top that primary sources should be replaced with secondary sources. Since I'm new to this, I have some questions/quibbles.


 * The article currently cites a lot of outdated, superseded primary studies (e.g. Malaspina's work on intelligence, work finding a J-shaped relationship for bipolar disorder). These should be removed and be replaced by reviews if I understand correctly. I will proceed to do this, but would rather ask a few things for clarification first:


 * 1. Is this criterion relaxed for non-medical outcomes? I have trouble finding reviews on outcomes such as personality and intelligence (the previous literature is "reviewed" in the intros of newer articles).


 * 2. If I cannot substitute these sources with reviews should I remove the section on intelligence completely (fine with me)?


 * 3. I understood WP:MEDRS to allow e.g. large RCTs where they stand on their own. In genetics and epidemiology, there are large consortia or national-database-based studies that can comprise as many cases as the preceding literature added together. Don't they fall under the same rule when reviews don't exist?


 * 4. You removed the number of subjects from my summary of a meta-analysis. I understand this to mean that meta-analyses can be used? I don't understand why you removed the number of subjects here, but not for older studies (e.g. Malaspina "44,175 people from Israel"). Please clarify.


 * 5. You removed claims about primary studies that also appeared in reviews (e.g. the enumeration of psychiatric illnesses that appear in D'Onofrio et al., which is prominently discussed in the review. These can go back in, right?


 * 6. You removed a section (about mutations, citing Iossifov etc., but also Ségurel et al.). Ségurel et al. is a review. I assume you missed that.


 * 7. You added back the 2009 review. The more current reviews supersede it and cover the same ground as far as I can tell (I can only view the excerpts as I don't have access to the book via my uni). I don't see why it should remain.


 * Thanks!
 * Ruben (talk) 13:23, 24 February 2017 (UTC)
 * We tend to try to use reviews for health claims.
 * Yes a meta analysis can be used. We do not tend to state the numbers.
 * Yes missed that one. Am looking at "studies found consistent, but smaller effect sizes of paternal age, but also supported relatively smaller effects of maternal age. "
 * The paper says "Second, jointly considering the four whole-genome studies with age information (19, 71, 80, 103) would suggest little, if any, paternal age effect, and yet a marked effect of the father's age is seen in three of the whole-genome studies"
 * I guess the question is how should we summary this?
 * Per "If I cannot substitute these sources with reviews should I remove the section on intelligence completely" IMO yes Doc James  (talk · contribs · email) 14:08, 24 February 2017 (UTC)


 * Thanks for the quick responses.
 * Ségurel is a bit ambiguous in the sentence you quote, the text accompanying Fig. 2 clarifies it (there is an effect in all studies but heterogeneity). Probably or are the newest references that deal with the variability in effect size estimates from different studies (also reviews Wong, highly relevant new study). The Moorjani paper is classified as an essay, but it reviews the existing literature and is written by subject matter experts. Is this something I can use? And since this is PLoS, could I put in their Figure 4 here (with attribution)? The second one is open access, but Copyright just says "the authors", so their figures are not usable?
 * I'm still not that clear on non-health claims (IQ, fitness). There's fewer reviews in this area. So, wait until someone writes one and until then remove the sections with primary sources?
 * You did not specifically react to the question whether cohort studies with >2million individuals are not considered as definitive as large RCTs. I assume this means you think they should not be used. I'm okay with those as long as the smaller studies don't remain.
 * You left in a few primary studies that I used to make the theoretical points about confounding (points which are also pointed out in reviews in summaries, but it's nice to have examples). That's okay then?
 * I looked through the pages you quoted at me and couldn't find reference to omitting sample sizes. There are effect sizes quoted here. I wouldn't have guessed this, seems inconsistent, is this a tacit norm or can I look up such things?
 * Ruben (talk) 14:32, 24 February 2017 (UTC)
 * Need to sleep. The "couldn't find reference to omitting sample sizes" is more an interpretation that we are writing for a general audience. The sample size IMO is not that notable.
 * Feel free to remove the primary sources large and small. We should stick with reviews especially for such a controversial topic.
 * Yes you can use figures from PLOS.
 * What do you think the Ségurel study means? 1 study found no effect and three do? Doc James  (talk · contribs · email) 16:10, 24 February 2017 (UTC)
 * Regarding Ségurel, I wrote to her regarding this some time ago, she clarified she did not mean to imply that she's certain there's actual biologically driven differences in this effect between populations, but that we should be aware of technical considerations that affect the effect size estimate (think: the sample, the sequencer, pipeline used etc) and that she was hoping for larger studies. There have been much larger studies since (esp. Wong et al.) and there are reviews including those, so I'll use them for this aspect. I wasn't aware there were so many reviews on this topic, I thought finding them would prove more difficult. I'm a bit unhappy with the coverage in reviews of the non-medical stuff, but if that's the WP policy, so be it. One more question, would you say using figures from primary sources is okay, if they're from a study a review cites and support a point made in the review? Ruben (talk) 10:40, 25 February 2017 (UTC)
 * With respect to using figures from primary sources, they are sometimes okay as long as they do not contradict the reviews avaliable. Doc James  (talk · contribs · email) 13:22, 25 February 2017 (UTC)

Fundamental misunderstanding: DNA replication errors versus DNA mutagens
The wholescale removal of primary research citations has been counterproductive here, at least where the basic mechanisms are concerned. WPMED requirements (i.e. review/textbook requirements) should apply only to medical research, not to biological mechanisms. The latter research field (paternal age effect mechanisms, not medical outcomes) is too young and too small to have generated meaningful reviews let alone meta-analyses or textbooks.

The simplistic view implied here is that aging men accumulate DNA mutations as a result of DNA replication errors. And that these errors cause medical outcomes in the offspring.

The reality is that we are only just beginning to understand how many paternal DNA mutations are caused by environmental mutagens (and are thus not primarily related to age but to lifestyle etc.), and how many mutations are indeed introduced inevitably during cellular replication (and are thus a true aging hallmark).

The only case where we can distinguish this paternal aging effect from environmental effects is repetitive DNA mutations (telomers, minisatellites, microsatellites) and other insertion/deletion events (duplication leding to copy number variation, deletions), as these mutations can arise ONLY during replication. (The references for this simple fact have been deleted, presumably because they were not reviews).

The Goldmann et al. 2016 Nature paper for example has made this comparison, with revealing results. But again, it is not a review and presumably therefore was deleted.

Please go through the deleted sections and salvage material where appropriate, to make these points clear. Thanks. 65.153.177.214 (talk) 14:49, 22 August 2017 (UTC)

External links modified
Hello fellow Wikipedians,

I have just modified one external link on Paternal age effect. Please take a moment to review my edit. If you have any questions, or need the bot to ignore the links, or the page altogether, please visit this simple FaQ for additional information. I made the following changes:
 * Added archive https://web.archive.org/web/20100109185819/http://www.schizophreniaforum.org/for/curr/Malaspina/default.asp to http://www.schizophreniaforum.org/for/curr/Malaspina/default.asp

When you have finished reviewing my changes, you may follow the instructions on the template below to fix any issues with the URLs.

Cheers.— InternetArchiveBot  (Report bug) 20:48, 20 September 2017 (UTC)

Weasel words
The whole article is full of weasel-words. Its sounds like something important is described but its not. Words like "low" have no meaning outside of the experts that know what is normal. One of the few (only?) exception is this sentence "odds ratio for low birthweight was approximately 1.1 at a paternal age of 20 and approximately 1.2 at a paternal age of 50". That is useful information. But statement such as: (regarding mental illness) "a twofold risk for younger fathers and a threefold risk for fathers >50 years old.". It means absolutely nothing if the base risk is not stated. Is it 3 times 0,00001% or 3 times 10% ? — Preceding unsigned comment added by 2403:6200:8810:6239:942A:BD72:8D84:E6E1 (talk) 09:24, 22 October 2020 (UTC)

Good secondary source to integrate
I don't have time to integrate this at the time, but I think this is a good secondary source for the article: -Pengortm (talk) 22:18, 31 December 2020 (UTC)