Comments on Induced Abortion and Mental Health for the Royal College of Psychiatry

Dr. Philip Ney

17/09/11

Introduction.
I am glad to be able to comment on this draft report partly because there is such murkiness to this very contentious topic but also because it has been a major interest of mine for over 4 decades. Although my main area of expertise is child abuse and neglect, I have written 31 published papers which directly or indirectly deal with this area, 5 books, and many expert opinions for courts and committees. I have taught in 5 medical schools in 3 countries and have been the chairman of psychiatric services and an academic department. I have discovered there are quite a few elephants in living rooms but none like abortion. Therefore I welcome any sane debate and honest research on abortion. Both of these are hard to find. Sadly this draft is no exception. I write this from a scientific not a moral perspective.

General Considerations

1. Primary object. This matter is considered with the understanding that in the UK abortion is permitted for women who require this treatment to "prevent grave permanent injury to their physical or mental health" Since it is acknowledged abortion is seldom required for medical or surgical reasons, this report must address the psychiatric, psychological or social indications for abortion. It does not.

2. Abortion by choice. This report assumes there are only 2 possibilities to a pregnancy outcome and that a woman is within her rights "to choose" either to terminate the pregnancy or bring it to full term. Thus the report is written with the belief that abortion is a choice but very little about any constraints to that choice such as being fully informed. Nor does this report consider informed consent an important variable to include in the analysis of factors that may bear on the mental health of the woman who chooses to abort.

3. Abortion by medical indication. The law in the UK creates the impression that performing abortion is still a medical matter requiring: indication, benefit, few side effects, less invasive and more reversible therapies tried first, done with clear conscience, informed consent etc. Nothing in this draft makes mention of these issues which are bound to influence the rate on mental insult following an abortion. No author has the temerity to ask whether or not abortion as treatment is effective in preventing mental illness, mainly I suspect because there is no such evidence.

4. Living conditions There are very few comments in this report regarding conditions in a woman's life that bear on her mental health after an abortion compared to those after giving birth. The research generally assumes that life for the aborted woman and the woman with a baby to raise, sometimes with little spousal or community support, are equal. Yet they are so different that comparisons of mental health for women in these two groups are essentially meaningless.

5. Other pregnancy outcomes. This report acknowledges that adoption is an option but cites no reliable evidence comparing the effect on mental health to those who give up their baby to another woman or place the infant in temporary foster care or at least 6 other options with those who abort. These options are becoming more readily available and are more frequently used.

6. "Unwantedness" Almost all the consideration in this draft are based on "unwanted pregnancies" or "unwanted babies" or "unintended pregnancies" without defining what these terms mean or how these most important variables are measured. The terms are usually used interchangeably but they are not identical.

7. Wantedness. Since almost everyone is ambivalent almost all the time on almost every issue, it is likely that almost every woman will have mixed feelings of wanting a child or wanting to be pregnant. Her feelings will fluctuate daily depending on mood, quality of relationship, finances, employment, health, and 20 to 30 other variables not considered in the studies quoted by this draft.

8. Unintended pregnancy. There are very few couple who carefully calculate the exact time and conditions to "make a baby" with intent. Most are more intent on multiple orgasms. Even those who appear to be harmoniously intent have doubts and second thoughts, "do you think we did the right thing?" Humans after all are not red and green marbles.

9. Change in wanting" The amount of wanting a child changes during the pregnancy. Our research shows that it is reasonably high before the pregnancy, dips in the first trimester than climbs to its highest point after birth. I suspect this is seldom explained to a woman contemplating an abortion and time, duration of pregnancy, as a variable is not included in the calculations of the studies cited in the draft.

10. Incidence and prevalence. Prevalence is usually defined as the rate of some disorder in the community whether or not it is detected by some diagnostic device. In this draft prevalence is used almost interchangeably with incidence.

11. Bias. Although the authors ostensibly ruled out bias, they paid little heed to their own which at times was so blatant as to discredit their whole work. Eg. Although Major states that her study was done in "2 free standing clinics and 1 physician's office" the authors of this draft felt it necessary to give the locus of abortions a better slant by stating they were done in "3 hospitals"

12. Four types of outcome. There is no marked distinction between those who are mentally unwell after an abortion and those who are not, partly because the are four types of reaction. a) Those who are less sensitive for whom life may be less valuable and pain less excruciating who appear to cope well but who have become more hardened. b) Those who are basically unstable who are pushed into a definable mental illness by the trauma of abortion. c) Those who are reasonably mentally healthy but because of their sensitivities, they are deeply hurt and develop psychiatric symptoms which a researcher defines as an illness. d) Those who choose abortion as the least worst case scenario. They may feel relief, at least for a few years.

13. Selection of articles. The author's bias is nowhere more apparent then in the selection of relevant articles; even in the selection of mesh headings. Eg. "abortion" etc but not "pregnancy outcome" which would have netted more articles with a wider range of outcomes. It should be noted that some authors preferentially use "pregnancy outcome" and "pregnancy losses" because it is easier to get their research published using these terms rather than with "abortion". Why not use "health" and "mourning" and "weight gain" and "bonding" and "death" unless of course one does not wish to entertain the possibility these are relevant to matters of mental health.

14. Unpublished research. An analysis of published vs unpublished drug studies shows a strong bias toward publishing that research that ostensibly shows positive results for the drugs in question. The authors did not attempt to ascertain whether or not there is a bias in the publication of articles showing abortion does not contribute to mental health problems. Comparing our publication rate for articles on child abuse and neglect (almost 100%) to the publication success rate for our research into the effects of abortion, which are better studies, the success rate is much lower. Ours is not the only experience like this.

15. Soliciting data not included in this draft. There is a very limited list of researchers ( 5 ) who were contacted for any unpublished data that could be considered in this analysis. There are far more who should have been contacted including Reardon, Coleman, Shuping, Cougle, Ney, Gissler, etc. all of whom are less likely to give evidence to support the authors unstated thesis the abortion isn't harmful to mental health. There is a definite bias in this limited list.

16. Arbitrary cut off date. There is no explanation given by the authors as to why 1990 was used as the cut off date for excluding research. It has little to do with the quality of the research but quite possibly much to do with avoiding unsupportive data.

17. Citation rate bias. Although Reardon et al have done larger studies and published more research, Major is cited proportionally more often. ( 93 to 89 times)

18. Variables not considered. Although the authors often complain that other factors influencing a woman's mental health are usually not included, they avoid referring to those which do.

a) Bonding and abuse. Our studies ( appendix A) on the greater difficulty post abortion women have in bonding to children of pregnancies subsequent to an abortion were not mentioned although the ongoing impact on children ( higher rates of neglect and abuse) and on mother's difficulty breast feeding and parenting. In one large study (unpublished) we found that breast feeding rates went from 87% to 18.5% after abortions became freely available in a certain country.

b) General health. The studies by Reardon et al and Ney et al showing the impact of abortion on HIV rates, cerebro-vascular and cardio vascular illness (Reardon) and general health ( Ney) were not used. The mind-body dichotomy is no longer valid in any consideration of health and should not be here. The study by Ney et al was dismissed as "inappropriate" although it had high correlations on physical and emotional health by physician, independent rater and patient. It: had a large nationally representative, unselected sample, included all pregnancy outcomes, considered many other relevant factors, and used pertinent statistical analyses.
c) Partner support. The authors never mentioned the impact of abortion on a woman's relationships or the influence of partner support. We found the amount of partner support (Ney et al submitted for publication) one of the most important determinants in a woman's "choice" to abort. We also found that men are less supportive of women who they suspect will have and abortion.
d) Children survivors. We have reliable evidence (appendix B) that children born following an abortion have difficult to resolve existential conflicts. Their struggles and behaviours tend to result in confusion and consternation in their mothers whose behaviours contribute to the child's difficulties in a complex vicious cycle. The authors of this draft either do not know or have disregarded this component of the mother's post abortion mental health partly because they are lumped together with those who wanted a pregnancy.
e) Men. Although their mandate does not stipulate that the only consideration of mental health should be for women only, these authors assume only women are affected. There is considerable evidence, not mentioned here, that men also suffer post abortion ill effects.
f) Some other factors. The following symptoms were not considered by these authors although clinicians encounter them frequently: poor sleep, nightmares, disinterest in sex, weight gain, prolonged grief, preoccupations, decline in work productivity, partner loss etc. In one unpublished survey we found that post- abortion, 78% of partner relations break up

19. Probability of false negative. Although there is comment by the authors of the probability of false positive findings, they fail to mention there is an equal probability of false negative results of research.

20. Poor measures. Currently there is considerable debate about the validity of DSM IV and V. It is increasingly apparent that almost every variable in nature is on a continuum. All dichotomous measures are arbitrary and distort the true assessment of almost everything, human diagnoses in particular. It is not at all surprising to find that each patient has some anxiety, a little less depression, more psycho-physiological disturbances etc. that make any one diagnosis a distortion of the patient's problems. A combination of diagnoses, (currently popular) still leave out many complaints and don't help a physician in deciding which medication to give. In our research, we have often illustrated that the most reliable and thorough measures are on visual analogue scales. The authors rely on outdated diagnostic indices and seem not to understand the change toward visual analogue scales. Almost all the studies using formal diagnoses should be discounted because they distort the true clinical picture of women post abortion.

21. Grading studies. Although the authors used the GRADE to help them select the best studies, they did not obtain a second opinion from experts not already committed to a particular point of view regarding abortion. They used as major criterion "unplanned" and "unwanted" which could not be less scientific, (see above). They also considered interviews as superior to self report but from long experience in data collection, I warrant that interviewers have subtle ways of imposing their bias.

22. Conclusions. Despite the above caveats, the authors make bold conclusions. They admit that due to heterogeneity, a meta analysis could not be performed. The problem is less with the data than it is with the biased interpretation of it. There is no scientific justification to any conclusion than, there is no evidence of benefit to mental health from having an abortion.

Comments on Specific Items in this Draft.

P 6. lines 4 to 44. This draft makes no attempt to answer the issues raised by the Abortion Act and its amendments. Indeed the authors make assumptions not contained in the Act that women have an unfettered right to choose an abortion. The act is clear that abortion is a medical matter and can only be performed if and when it is necessary to preserve a woman's health. The real question to be addressed is, what is the evidence of benefit Not what is the data of harm. This was studiously avoided by the authors of this draft. The College of Psychiatry erred in not making their mandate clear.
If a woman has a right to have an abortion when she so elects, then abortion is not a medical matter and should be performed by technicians If a woman has a right to good medical treatment that may include having an abortion on her physician's recommendation, then this paper is valid if it addresses the questions: a) is there a pathological process in pregnant women in general and this patient in particular that warrants having an abortion? It must be recognized that pregnancy is not a disease. b) what is the evidence that an abortion will benefit women with this disease (pregnancy) and this patient in particular? c) what are the adverse effects from an abortion and if there are some, do they outweigh the anticipated benefit? d) have all less invasive, more reversible treatments been offered, tried and failed before an abortion is recommended? e) is the abortion done in good faith? Has the abortionist carefully studied to relevant literature in order to practice evidence based medicine, honed his/her skills and performed a careful followed up on his/her ex-abortion patients to know personally that he/she will be providing good treatment? e) Has the physician facilitated all options to abortion of a truly unwanted child, ie, adoption, fostering etc. f) Has he/she made a clear recommendation to the patient with evidence to support that recommendation, options available, potential benefits and hazards, and shown the ambivalent woman the ultrasound of her fetus? Has he/she been given fully informed consent which requires the patient have full opportunity to ask questions, get a 2nd opinion and make a decision with enough time to do so and without pressure from mate, family, IPPF, physician etc..
It must be remembered that until any treatment is well proven, it must be considered as experimental and constrained as such. Moreover the burden of proof rests with the performing physician, his/her supporters and those who fund this activity.

P 7 lines 11 to 19. Although often touted, there is no evidence that an abortion is safer than a continued pregnancy. The life long benefits to having a child cannot be compared to the potential life long regret. Moreover the comparison is not valid because the duration of pregnancy is approximately on a 1 to 3 ratio. Any adverse event is more likely to happen when the time is longer.
Lines 21 to 26. The findings of the Rawlinson Report cannot be disputed by these authors because they did not address the issue of indication and benefit.

P 8 lines 11 to 24. There is no scientific way to measure the intent in a woman at the moment she conceives. Although she may be contracepting, she also knows that no method is fool proof. The data on the proper application of contracepting clearly indicates either the woman is very foolish or at some subconscious level there is a hope for a child. None of the studies approved by these authors indicate how they determined that the pregnancy was unintended or at what moment the woman felt she didn't wish to have a child. Women will tend to justify having an abortion with many phrases that reflect her momentary disposition at some points but whether or not they indicate her true feelings for any length of time is open to wide speculation.
We have good evidence that wanted children are more, not less, likely to be abused and neglected. (appendix c) This is partly because on them are heaped higher levels of expectation and therefore they are more likely to be considered a disappointment which parents try to correct, sometimes harsly.
Wantedness grows with the duration of the pregnancy after the 1st trimester but most women are not given the chance to experience this phenomena.
Before contracepting became so well known and accepted, the majority of pregnancies were unplanned, sometime resented. However almost all were eventually welcomed into the family and grew into productive citizens with their own well loved family.

P 10 line 34. Almost all "validated tools to measure mental health" use questions that distort the reality of nature being always distributed on a continuum. This distortion invalidates the diagnostic tool.
Line 22. The authors make no mention of the extensive carefully considered criticisms of the APA review whose principal author was the same person who writes much "pro-choice" rhetoric and who is the 2nd most often cited author considered by the writers of this draft. The APA never mentions the women who because of the picketers changed their mind and now are able to contentedly hold a not aborted child.

P 12 line 30 to 34. The "ideal comparison" is impossible. Most women may change their mind daily if not hourly depending on mood, hormones, a fight with their mate or mother in law etc. After all people are not red or green marbles. Almost everyone is ambivalent about almost everything almost all the time. Pregnancies intensify their ambivalence because it is part of the intense psychological shift in thinking which is necessary to incorporate the child into her psychological frame of reference and include the child into all considerations of her family.
The glaring gap in this draft for any consideration of the effect of abortion on men and children, makes this report invalid if for no other reason than because what effects spouse and children will have a pronounced effect on the woman's mental health.

P 13 ;line 1-2 Stating that there is an important distinction between unplanned and unwanted pregnancies doesn't make it any easier to do research based on this supposed difference. I challenge these authors, who appear to have no experience in doing the research upon which they so pomposly adjudicate, to make that distintion in practice. I interviewed or treated or have done research with probably more women than anyone else, and I would not know where to begin. If I asked any post pregnant woman very quickly she would probably give me the answer she detected I wanted to hear. After I really got to know her she would honestly say, "I really didn't intent any of my children. We were just making love and knew in our hearts it could happen. I had mixed feeling about wanting children but as they grew inside me, I learned to love them. By the time they were about to arrive, I was excited to see who was coming next. When she/he popped out, we fell in love. Now I want that brat like a hole in the head but I also know he is loyal to his old mum and will stick with me when I can't see or hear. Intend him? yes and no. Want him? That depends on which day you ask me. During school days, mostly yes. On weekends when he is fighting with his sister, definitely no. So Ms author, how should I mark that one or the majority just like.
When any researcher forces arbitrary distinctions on some complex research question, they are bound to end up with junk data and do the subject a great disservice. That they then have the gall to publish it as fact, destroys credibility in themselves and all their findings. Yet these authors base all their conclusions on so called unwanted pregnancies. They should not have wasted their time and the government's money.

P 13, line 39-40 Mail back responses may introduce a bias just like every other way of collecting data. However there is no data to show it did. It may limit the amount one can generalize the findings but for that sample, they are correct. If it can be shown that the demographics of this sample are typical of the wider population as I have in our studies, then it is reasonably safe to generalize to the whole population being considered. Very few researchers bother to collect data on the same population in more than one way to see if there is any consistent difference.

P 13 line 46. There is well documented evidence of under-reporting of abortions in the UK. Ref. **** Gilchrist et al is a prime example of how this distorts the evidence they show. The BBC reports ( 20/4/11) that the UK Dept of Health challenged a decision the Information Tribunal handed down in Oct. 2009 saying freedom of information laws require all of the abortion statistics to be released. A group made the request 2 years ago because it was concerned rules on abortions were not being followed in order to allow for abortions on babies with minor medical issues like a cleft palate or club foot, that can be corrected with surgery. Dr. Evan Harris (member of BMA medical ethics committee stated that it was "hard to see why successive governments" had fought the Information Tribunal Decision. "Secrecy will only serve those doctors authorizing or performing abortions outside the terms of the law, which is already widespread practice" said a spokes-person for the group.

P 14. line 6 to 5 - 31. The authors do well in detailing the difficulties in doing post abortion research but seem to not let these bother them in deciding good from bad studies. The studies of Russo with a 35 % follow- up rate and Major with a follow-up rate at 2 years of 442/ 1043 or 42.4 % of those who were considered eligible, have so little reliable data that under normal circumstance their research would not be published. Her caveat of "lack of evidence of retention bias" says nothing about what this means or how it was determined. It is well known by abortionists that many or most of their patients tend to avoid them and their precincts. This distrust of aborting physicians with other factors is so extensive that approximately 35 % of Canadian women refuse to have any physician or midwife participate in their delivery (CMAJ 2011; 183:648 - 650) Sadly these are not normal times and anything that favors abortion like the very poor studies of Russo and Major are not only more likely to be published but they are highlighted in reviews like that of the APA and this one.
To collect data 1hr before an abortion when most women are very anxious and ambivalent and 1 hour after when they are experiencing a plethora of confused feelings (Major) does the women an injustice and is the best example of how to collect unreliable data that is possible. I am very surprised this study received ethical approval. In fact there is no evidence it did.

P 15 line 3 to 8. To this clinician it is important to understand how abortion contributes to nicotine dependance, partly because it is a good indication of how much more post abortion women smoke which contributes to a wide range of pulmonary and circulatory diseases which certainly have and impact on mental health. The statement by these authors negating the importance of Pedersen's research is another indication of their bias and their clinical naivety.

P 15 line10 to 16. Though the chance of finding something positive is always there, (that is partly why we use statistics) but it is no greater than the chance of a false negative finding which the authors don't mention. Almost no one believes correlation means causation but when the correlation is statistically significant and keeps on appearing, one can normally assume there is something worth noting and avoiding if possible.

P 15 lines 1 - 46 all of the difficulties surrounding doing research are important only if it is assumed the burden of proof lies with those who are skeptical about abortion to show its hazards. The real necessity of proof lies with those who assume abortion is beneficial or at least harmless. The authors neatly side step this issue.

P 15 line 41 -42. Our research shows (submitted) indicates that it is not the woman's level of education, or number of children or poverty that is most closely associated with the decision to abort but it is childhood mistreatment, the subject's mother's abortion and partner support. Since all of these relate to a woman's mental health and stability, the most important factor is not previous mental health, which is secondary, as concluded by these authors.

P15 line 44 to 46. The APA's conclusions were based on a very biased sample of the literature and did not consider our findings even though objective appraisers find it one of the best. The statement that the higher the quality of the study the less likely the findings are there is a greater relative risk of problems associated with abortion is a subjective, pejorative statement that has no business being in a scientific document but clearly indicates the bias of the APA review. It does not belong in this draft if it is to be taken seriously.

P15 & 16 Section 1.3.3 The summary of key findings have all the problems of this document and more. Prof David Fergusson, commenting on the Task Force Chair, Brenda Major's conclusions, states, "The APA report, in fact draws a very strong and dogmatic conclusion that cannot be defended on the basis of the evidence. A better logic is that used by the critics of the (tobacco) industry: since there is suggestive evidence of harmful effects it behooves us to err on the side of caution …..before drawing strong conclusions. History showed which side had the better arguments"

P 16 line 4-6. If based on the determinations of unplanned pregnancies, the APA's findings are as useful as this ephemeral unscientific quality; not much. Especially since no consideration is given to such groups as "planned and aborted" as if they do not exist.

P 16 line 13. If "prevalence" is used correctly then the only way to determine this is a proper epidemiological study of whole populations. The only one I know of was done in Denmark.(***) It showed the prevalence of at least one abortion is 70 % of women by the end of their reproductive life. Since conditions regarding abortion are similar in the UK, the prevalence will be similar. No study of mental health has used the true prevalence rates. The pseudo prevalence definition used by these authors will likely find correspondingly low rates of mental health problems.

P 16 line 19-21 Why would these authors use the very biased APA review to "build upon" unless they wished to build into their review the same biases. Their very brief comment on the Rawlinson Report although well conducted with succinctly stated findings gives further evidence of their predeliction for skewed evidence.

P 18 lines 6-22. Not only did the Steering Group ask the wrong questions, they constructed their approach to find nothing of significance. The most useful and most salient questions, given the existing legislation in the UK on abortion, must be: a) "Does the present arrangement of elective abortion help prevent mental illness in any patient.
b) is there evidence that abortions provide the country any health benefits?" Surely they must realize that even if the harmful effects are a small percentage, there are so many abortions, there will be a significantly larger expenditure of health care funds. If there is evidence of preventing mental illness, then many woman will be the better and the country will not have to curtail health services as at present.
Using the base criteria of "an unwanted pregnancy" which is impossible to honestly determine, ensured the research could not determine whether or not abortion contributed to poor mental health. Given this situation, the Group can conveniently assume and state as fact, their preconceived belief, it does not.

p. 18 lines 26 plus. Using these mesh headings, the authors avoided having to consider all those studies of the effects of abortion on health in general. This is a major mistake because many authors who realizing their results implicating abortion are unlikely to be published, consciously used "pregnancy losses" and "health" instead of "abortion" and "mental health". Our study (appendix a) was a large cross sectional study using a random sample, with well validated measures of physical and emotional health, for all the possible pregnancy outcomes for the woman's entire reproductive life at the time of data collection, using reality based visual analogue measures, including many other relevant factors, especially partner support, statistically analyzed by a professor of statistics in a good university but considered as "inappropriate" with no explanation. One can only assume the authors were too lazy to read it or found it inappropriate for their preconceptions.

P 18 (table 1) As stated in their introduction, the authors are clear that abortions in the UK are "legal" if certain conditions are met. Since most frequently abortionists do not indicate for what legitimate reason he/she if performing the abortion, there is no way for the authors to know whether or not the abortions considered by various authors would be legal in the UK. Therefore even if the studies were well done, the authors cannot use their findings in the UK.

P 19 (table) Although the authors insist that mental health had to be assessed at least 90 days after the abortion, they make no comment on the more important constraint that the assessment cannot be earlier than 2 days after the abortion because of the patient's very vulnerable state. It would be provocative intrusion. Using that criterion, Major's study would not qualify and the authors would not want that so they didn't bother with this vital condition.

P 19 line11. Using their oft stated criteria that only those studies of abortions for "unwanted pregnancies were included" works to invalidate all the studies they included because "unwanted" is such a momentary, undefinable quality. No reputable researcher would attempt to use it unless they measured it daily for a month at different times on the day.

P19 line 25-28 The authors compound the above problem by assuming "that all abortions were due to unwanted/unplanned pregnancies" This is an unwarranted assumption. In addition inspite of their claim that unwanted was not the same as unplanned, they lump them together because they can't know and nor could the researcher. So why use this criteria at all.

P 19 line30-31. Almost every country does not have unconstrained abortions. They are legal, including the US and the UK under certain conditions. It is obvious that the authors ignore these conditions for their own purposes.

P19 line 36 to 38. The authors place no life span time limit on previous mental health problems so it is safe to assume they include psychological problems when the subject was a child but there is no evidence, this was included in their inclusion criteria although it is probably more important than an adult psychiatric illness in the post abortion reaction. Most children would like to and do forget their previous problems. There is no standardized way previous mental health problems are determined. Is it a one time consultation with a psychiatrist? Is it a self assessment using a questionnaire found on some website? Is it receiving psychiatric medication, (used by many insurance agencies as the defining criteria) even when the distress is normal and self limiting such as grief. Since there is no standard criteria among psychiatrists, the authors are in no position to use this nebulous criteria to assess the quality of any research.

P 20 The authors indicate they used previous systematic reviews but made no mention of one by Ney et al. (8) Granted it was beyond their time limit but it would have enlightened them to the state of research to that point.

P 21 line 2- 8. I don't expect anyone reading this report was convinced of the objectivity in selecting which articles to review when 2 reviewers resolved their disagreements by discussion. It is surprising that the authors did not use the standard technique to assess their determination of quality, ie obtain and write in this report what was the inter-rater reliability. Are they so embarrassed they don't wish to share such a finding when they have no hesitation of requiring this of others.

P 21 line 42-45. We found Visual Analogue Scales compared to rating scales more accurate and more often answered because it is so much easier to self rate. But these authors make no reference to their use in research.

P 21 line 47-48. The authors stating many researchers did not investigate " prevalence rates per se" Apart from the Danish and the Fergusson studies, I know of no study that did a proper epidemiological investigation to determine the prevalence of abortion and related mental health problems.

P22, line 16-24. This reports reliance on the DSM IV and V in light of recent heavy criticism mainly by its principal originator, puts all the conclusions in question. A much better method is to recognized almost all observable phenomena are on a continuum and use a Visual Analogue Scale which can pose the full range of possibilities between 2 extremes.

P 23, line 9-11. Rating scales are not continuous outcome measures. Only Visual Analogue Scales are.

P 23 line 26-36 On the basis of the criteria stated in this report there was no reason to exclude our study on the "Effects of Pregnancy Loss on Women's Health ( )

P 24 line36-37 It would be very informative to have a list of the experts the authors consulted. I might have been wiser to consult international experts who can represent a wider variety of opinions.

P 24 line 46. As the authors note, randomized trials cannot be done on humans to determine if abortion is good treatment but it can be done on animals. There is no comment on this obvious gap. Randomized studies on animals should have been required many years previously if abortion was considered, as it should have been, an "unproven treatment" If that had been done many of the questions the authors attempt to address would have been answered. A good example would be to determine whether randomized abortions done at various stages of a rats pregnancy has any effect on its parenting. This could be quickly determined using a T maze. I have written a protocol but have been unsuccessful in obtaining research funding for the study.

P 26 line 4. I am quite sure the authors understand that bias is always present and will effect observation. It is particularly prevalent in interview collected data. If researcher will not be honest about their bias and try to control for its effect, their research should be discounted. This is especially true for authors like Major who play many roles in promoting the idea that abortion is not significantly harmful. The authors know this but have not used that information in evaluating the quality of her research.

P 27 line1. I note that the authors of this draft are seeking comments from researchers and reviewers. They should also obtain opinions from practitioners who see many post abortion women. I fill all these roles and considered an expert if reviewing article for publication and requests for expert opinion mean anything. Our 1989 review ( )
is months under the cut off but could be helpful in this report.

P27 line 29-31. If studies that made measures less than 90 days post abortion were excluded, why was Major's study included when she made measures of outcome, less that 90 minutes post abortion?

P 28 line12, Gissler was not the only record linkage study.

P28 line 16, the authors are confused in their use of prevalence and incidence.

P 28 Table 3 There is no explanation for why Gissler (1996) is rated very poor especially when his study is often quoted by experts and has all the characteristics the authors were seeking.

P 29 Table 3 The authors declare their bias often but none so great as when they repeatedly refer to Major's study occurring at "3 hospitals, US" when Major states "2 free standing clinics and 1 physicians office." Their attempts to make their more favored researchers look good while using pejorative descriptors to make others, with findings they don't appreciate, look not so good, if nothing else should discredit their conclusions.
There is no reason given as to why Reardon's study of 186,000 (2002) should not be included and his study (2002) of 293 be rated very poor. The death study is the largest and should be include for that reason alone.

P 30 line 32 We controlled for previous hospitalization because that is a far more definable criteria. Mental illness, at least in Canada, can have a wide variety of meanings.

P 33. line 34. The authors over estimate Major's follow-up rate.

P 33 line 49-50 I believe we are the only researchers that measured the amount on wantedness before the pregnancy, at different stages and after birth. It is a check mark shaped curve illustrating the mark drop in the rate in the early stages of a pregnancy at which point most abortions occur. Many of these women would have developed a desire for the child if the pregnancy had not been terminated.

P34 table 4 As far as I can detect, the authors are using "prevalence" where they should state incidence. Dorland's Medical Dictionary (25th ed) defines prevalence as: "the total number of cases of a specific disease in existence in a given population at a certain time.". This is not what any of these researchers determined.

P 34 table 4 There is no significant difference in the studies and that the designation of fair etc is arbitrary. Reardon (2002) did control for mental illness using record matching for psychiatric admission. That criteria is much more précis than any other such as, psychiatric contact, psychiatric diagnosis, emotional complaint etc.

P38 table 5. The Steinberg study (rated as very good) not only did not do follow-up but used interviews to collect information on mental health. Having taught interview techniques to medical students and counselors for many years, I assure the authors this is the most unreliable way to collect this data unless the interviewer has months of training and unless they have no bias regarding matters of abortion. Both those conditions are unlikely. Once again it reveals that their designation of quality is arbitrary and biased.

P 40 line 5-14 The authors note that Steinberg used post abortion assessments "from a few months to 20 years" In that time period there are so many intervening variables that no conclusion can be made regarding any association.

P 41 line10-16 Although it is true that some women with a psychiatric illness may not make a claim from Medicaid, the probability is that this are women who are coping less well, have less education, are not as desperately poor. These are those who would be most vulnerable to the effects of abortion. Thus if there is any influence, it would be in finding less of an effect, not more.

P42 table 6 There is much variability in the recorded "prevalence" rates which adds credence to my assertion that the assessment devices and procedures had very little in common. It must be remembered that the conditions that psychiatrists practice in Denmark are very different that the USA.

P 43 line 8-9 The authors use the term "elective" abortions in the USA and UK there is no such thing because even though the law or court decision is not enforced, in both countries there are prerequisite conditions. The term therapeutic abortion is also used very loosely for there is no evidence that women benefit from having there handicapped child terminated. Quite the contrary, there is evidence of very strong reactions post abortion that are hardly evidence of improvement.

P 44 19-21 and 30-32, These statements accurately illustrates that as treatment to improve or prevent mental health problems, abortion is not effective. It also means that because they are more vulnerable to abortion, woman with psychiatric illness must be screened out. Therefore abortionists will need to learn psychiatric skills. It also means the medical profession should repeatedly inform the public that psychiatric illness is not an indication for an abortion but a contra-indication.

P 45 table 7. All these so called prevalence rates are incident rates because they are only counted when the have been identified at some treatment facility.

P 45 lines 10-16. Small sample size is not a draw back if it is representative of the population being considered. It is much easier to obtain statistically significant data with a large sample and to find small difference and rare events. In that regard a small representative sample is more likely to discover phenomena with large effects. Our relatively small sample (submitted) found that the lack of partner support increased the association with abortion by a factor of 6 and with miscarriage by 2.5

P 47 line 37 Russo's study had a follow-up rate of only 35% of the original sample and should not be considered research.

P 51 line 36 After repeated formal requests Major has not made her data available to other researchers as is required by the APA.

P 51 line 38 The authors have still got it wrong. Major states 2 free standing clinics and 1 physicians office where the abortions were performed. The patient populations are likely to be very different because those attending a clinic probably could not afford the fees of the relatively benign conditions of the private physician.

P 53 line 33. These findings of Major are very suspect because they are taken from only 42% of the original sample. Women who return to the "clinic" for follow-up, even though paid to do so, are not representative of the average US American. They are much more likely to represent that segment of the USA population that personally favor a "prochoice" legal atmosphere and therefore more likely to report good effects &/or suppress reporting harmful effects of abortion.

P 55 line 1-4. The authors do not report whether or not the African-Americans reporting for follow up made up the same % as were in the original sample and whether or not the original group were of the same % as in the local population. I strongly suspect they were not.

P 56 line 5-8. Marital status has little bearing on whether or not the partner was present at different stages of the pregnancy and delivery and whether or not the partner was supportive. We found very large differences

P 57 line 1-4. Religious affiliation is barely related to the importance to any individual that their faith is. It is not surprising the researchers found no difference.

P 57 line 26-29 Measuring any attitude "at the time of the procedure" when most women will usually be very anxious is not only bad science it is unethical.

P 60 lines 13-16. Making these statements the authors need to put them into focus by indicating the follow-up rate was only 42% and these were probably not representative of the whole sample and that there were different scores at follow-up between those who responded in person and those who mailed in a response.

P 62 Line 41-43 Major's statement "the lack of evidence or retention bias in the final sample." (Page 783) could hardly be called "providing statistical analysis" as written by the draft's authors.

P 63 line 35-36 The authors statement that the list of potential risk factors here is not exhaustive" is an understatement. Their choice of factors reveals their bias. The lack of interest in the effect of abortion on the mother's ability to bond with a subsequent child is a glaring neglect.

P 64 line1-4. The authors make declarations of their very questionable findings with no hesitation or riders. Most scientists would write eg. "Of all the factors we considered, it appears the one with the closest association to poor mental health is…"

p.66 line 32. As a academic child and family psychiatrist who has assessed thousands of post abortion women and families I have found there is no method or diagnostic test that approaches the reliability and sensitivity of Visual Analogue scales. We have also used them in combination with other measures with more definitive end points such as employment, charged with some crime etc. As previously noted, dichotomous measures, even five point scales, constrain and distort the continuum on which reality is distributed. Moreover when people are rating themselves they tend to feel annoyed that they must choose between mild and moderate when they feel they are somewhere in between. Because they are "ticked off" by the scale, they are less cooperative.

p.68 table 12. Although Reardon is rated as only fair, he used the only definable end point with a high degree of inter-rater agreement, the death certificate. The authors make no mention of whether or not the researchers checked the instruments they used for diagnosis to determine how valid and reliable they were in their hands.

P 69 lines 39-43. There are other plausible explanations such as the women with pre-abortion mental health problems are more likely to feel relief following and abortion because they experience a sense of relief from turmoil which accompanies attempting ot make a very difficult decision ("I don't know if I made the right decision but at least I made a decision and that feels good").She may feel "better" because her partner after making threats over an extended period, finally left.

P69 line 36. Because the "psychiatric contact" varies so greatly from one country to another, it is not possible to equate these studies. In nations like Canada with universal medical coverage there is generally ready access except there are long waiting times, 6 to 12 months for an initial consultation. In the USA, psychiatric referrals may be seen much more quickly if the patient can afford it and if private insurance or cash is limited, there may be very long waits. The time interval from the event (abortion) to the onset of the "illness" probably relates more to the availability of treatment than it does to the extent of the trauma.
Those who deliver a child are under more financial and emotional stress, partly because deliveries cost considerably more than an abortion and because child care can be stressful, especially for those who have had a previous abortion and have consequently more difficulty bonding. It is not clear from this review, how many researchers controlled for the number of previous abortions a woman had before she delivered a child. Our data makes the difference clear. ( )

P71 line 12, The authors do not credit Reardon with 3 way record matching studies but here describe exactly that.

P71 lines 17-22. While there is some validity to the author's criticism of limited duration to pre-abortion mental illness indicators, there is logical time limit. From a child psychiatrist's point of view, the mental health of a woman as an indication of her vulnerability to the adverse impact of abortion should start in her infancy when her character resilience to adversity is established.. For that reason we were not surprised to find that one of the closest association to the decisions to abort was that the subject's mother had one or more abortions.
Since recall, financial and social conditions exponentially vary the further removed from the event in question, a 1 year cut off is probably the best compromise.

P 73, table 14 The studies ( not identified) in this table were presumably give poor very poor ratings because they did not indicate whether of not the pregnancies were wanted. As described earlier in this comment, intendedness and wantedness are such unreliable criteria to judge any event, that they are virtually useless. People are not marbles red and green. Ambivalence affects everyone's mood and decision making almost every hour, every day. Throughout history, this ambivalence could not be greater than in this era, because for good or ill, there are more choices to make and more social and political pressures to balance in making those choices
I estimate (unpublished study) there are 53 factors that the average woman must consider and resolve in order to make a rational decision. Since there is not sufficient time to conclude all the internal debate on these issues, it must be concluded that very few decisions to abort, deliver, adopt, foster etc are rational.
From our studies ( , , ) it is evident that a person's unresolved conflicts from mistreatment in childhood, become reenacted in adult life. Thus it is likely that a woman who was neglected in childhood, particularly by her father, will find a mate who is not very supportive. Under most circumstances he threatens to abandon the pregnant partner if she doesn't abort. She too readily acquiesces, aborts her child only to find he abandons her anyway. She would normally grieve but is so angry at him, she cannot feel sorrow. In time that anger would fade and complicated grief would develop except for the fact her feminist sister help stoke her anti-male antipathy and irritation. If she can find reasons to keep being enraged she does not need to feel the pain of mourning.

P 77, lines 29-36 The authors seem to have inadvertently contradicted themselves. If, as they state a less than 1 OR indicates increase anxiety then that is what Fergusson found, OR = 0.55 p<0.05. Whereas they authors write "no more likely to experience anxiety" Is this another indication of awkwardly rushing to confirm their preconceptions?
These affirmative findings ( abortion isn't good for most women) should always be considered and described in the light of very great difference in the woman' s life post delivery to the woman who is post abortion, as almost every woman would attest. That any evidence points to a greater anxiety or depression in post abortion women is surely most remarkable considering the much higher level of stress to women with small children.

P 77 line 45 to 48 The authors confuse the reader by interpreting Fergusson's findings one way and then in the opposite manner in these 3 paragraphs. The only thing which appears to be consistent here is that either way, they find support for their bias.

P 79 lines 3-6. The authors have a persistent tendency to state that the confidence intervals are wide if they don't like the evidence presented. The rules of research state that one sets what will be accepted as significant probability and confidence levels as part of the design prior to collecting any data. Once those levels are reached, the researcher must acknowledge the findings are significant whether or not he/she likes it. Thus if the confidence levels are sufficiently narrow, the finding is significant. Yes it is arbitrary but so is almost everything in this field of research.

P. 81 lines 4- 10 The authors are inconstant in their use of unwanted/unintended. When they wish, they use them interchangeably to give higher grading to findings they approve. When they don't like the evidence, they make a point of again writing, "an unplanned pregnancy is not the same as an unwanted pregnancy". Can they not understand, there is no scientific validity to this concept?

P. 83 lines 7-15. The strong built in bias of the authors keep showing in how they state the background and the evidence for this draft review. They reiterate that women "may elect" or "have the option" The law is clear. Physicians may recommend and may perform an abortion if the appropriate indications are present. The law is only concerned with parameters for what the legislators of that time deemed to be the correct practice of medicine. The law is not addressed to women.
By ignoring this fundamental constraint, the authors have disqualified themselves to write this review.
By stating "the presence of risk to either the mother or child" without any reference to studies of the effect of abortion on children, the authors can only mean the risk of being alive for a child with disabilities. If being alive post pregnancy is a risk, it is one every human shares.
It is remarkable that the authors do not consider any post pregnancy condition for the mothers who give birth as if financial, social etc well being have no bearing on the woman's mental health. Nor do they consider all of the many other pregnancy outcome such as adoption, fostering etc. as if these were never used or were equally conducive to health or illness.

P. 86 lines 1-28 Having written an earlier review of research on this matter ( ), I concluded the studies were good as far as they went or could go with the research funding, time and support available to them, except those who allowed their biases to affect their findings of which there were many. Sadly these authors have allowed their biases to compound the problems of biased research.
Some countries are notorious for not reporting abortions and the health effects of abortions notably Canada the USA and the UK In the UK the Dept of Health and Social Services was so reluctant to publish the data concerning post abortion maternal mortality it was necessary to table a parliamentary question.( ) Since the aborting conditions approximate those of Denmark where a good study was done, it is safe to assume the prevalence of abortions in these three countries is approximately 60 to 70% of all women by the age of 45 years. The authors misunderstand this and thus many of their statements are not based on this reality.

P 87 line 8-11 The differences of post abortion life and post partum life are so great, they make comparisons between the results of induced abortion and childbirth relatively meaningless but these difference are almost totally ignored by these authors who blithely continue to use data from studies where this comparison is made.

P 89 lines 13-16 Stating that their conclusions confirm the findings of badly biased APA review is the most self incriminating statement that they share the same blatant biases and conclude the same faulty findings

P 89 line 18-19 Finding that the mental health status for women post abortion and post pregnancy, though their post pregnancy lives are almost totally different, should have alerted the authors to the conclusion that the impact of abortion was much more harmful, but it didn't. This also indicates their determination to conclude what they precluded.

P 89 lines 20-26, The fatuous sop that all women should have "support" if needed post abortion indicates how little these authors understand the deep dehumanizing damages inflicted by abortion and how long and complicated the treatment must be to resolve the severest psychological conflicts known to human kind that arise from natures most unnatural behaviour, killing the young of your own species, yes and even killing your own child. ( DD). They may retort, if the damages are so severe, how is it possible to they are so difficult to detect. A very good question that they should have addressed from the beginning. There is at least a tentative explanation (see the summary to follow).

P 93 Appendix 2. Why such a limited list of researchers who were contacted? There are hundreds more who have addressed this subject, many indirectly.

P 95 Appendix 4. Despite the convincing evidence ( ) that valuable studies which are medically incorrect and/or reputationally unprofitable are less likely to be published, these authors made no serious attempt to find and use them. Surely a wise and learned person needs to listen most carefully to the messages he/she does not wish to hear.

Pages 95 to 99 I am reasonably sure I am not the only one to wonder why some very useful search items are so conspicuous by their absence. This list should include: Health, mourning, weight gain, bonding, partner support, child mistreatment, child abuse, child neglect, social support, employment, and many others. They weren't included not because they are irrelevant to a woman's mental health but because the authors biases betrayed them into thinking these were the most important.
Why did the UK Fellowship of Psychiatrists not include in the mandate of this study, an equal consideration of men, children and families. Surely they understand the health of one member of the family or species has a very large effect on the others. Are they totally unaware of the impact of abortion on men and children. Have they never read the studies on children who grow up in families where one or more of their siblings was aborted. Admittedly these isn't much to read, not because there is a lack of effort or data or articles submitted for publication in professional journals but because this is the last thing that those who insist on the idea that abortion is almost harmless want to read or hear about. Yet if anyone is attempting a systematic review, Post Abortion Survivor Syndrome ( ) must be the most important subject of inquiry just because it is so offensive if for no other reason.
Does the Fellowship still adhere to the ancient notion of the mind-body dichotomy. Surely not. Then why confine this study to mental health with not the slightest indication of interest in how physical ill health affects the rate of mental unwellness and visa versa. There are none so blind as those who refuse to look in the direction from which relatively new information is likely to come.

P 101 Section 2. The authors ask some potentially useful questions but do not provide us with the answers . And still many more questions should have been asked of researchers not only about their studies but about themselves to hopefully provide these authors with an indication of where their biases lay. Maybe it is well they didn't because that would have deepened the impact of their own (draft authors) predilections and prejudices.
I suspect these comments are felt to be offensive by some authors and researchers but when the stakes ( the health and welfare of millions of women, men and children) are so high, there is no room for polite medical and academic parlor games. So let us play roughly, as in hockey or seven a side rugby, with fair rules and a level playing surface ( equal access to research funding and publications) May the fur fly and the truth be known. For medical history shows truth and gravity always win. Oh the eventual embarrassment of those learned scholars and eminent physicians who adamantly and roughly insisted there was no such thing as a bacteria.

SUMMARY.

1. Wrong mandates
a) The authors of this draft report seem to believe or were instructed to assume that UK women are free to elect, or choose or have an legal option to request some physician to perform an abortion for any reason at any time. The 1967 law and modifications of it clearly indicate a physician may perform an abortion only if it is indicated. The legal indications are mainly if abortion will prevent a woman experiencing a worsening of her mental health. The proper mandate should have been, does abortion as practiced in the UK and elsewhere adequately treat or prevent mental illness. This question was completely ignored. Therefore this review is irrelevant and invalid.
b) Although the title "A systematic review of the mental health impact of induced abortion" implies it will deal with any beneficial or adverse consequence of abortion to any person, it only deals with women. The Fellowship of Psychiatrist must assume the abortion is only a woman's issue. This flies in the face of substantial evidence that men, children, families, abortionists etc are greatly impacted. This huge bias must also invalidate the findings of this review.

2. Bad methodology
a) Although the major criteria for determining which research to include and/or grade as worthy of real consideration was whether or not the pregnancies were, unplanned or unintended or unwanted. No one provided a operational definition of these terms. No researcher made a serious attempt to measure these at different times under different circumstances. Humans are almost always ambivalent, especially about a pregnancy. The more intense the ambivalence the better the struggle to incorporate into her psyche and family the growing infant. Thus ambivalence is good for bonding and breast feeding and parenting. It is very unlikely that any woman is single minded about a pregnancy before or after conception all the time. These factors are impossible to determine and therefore of no *** value in research and of no value in this review. Many "unplanned" pregnancies become wanted, preferably welcomed, children under the increasing impact of hormones, imprinting, social acceptance and psychological receptivity. Our research uncovered the increase in wantedness during the pregnancy. This critical evidence was ignored by the authors.
b) Post partum, post abortion comparisons. All the studies cited assume a woman's post abortion state is equivalent to her post partum life when making comparisons of her mental health. Common sense and experience indicates they are very different. Going back to work and social life "as if nothing happened" cannot be compared to the much greater stress and joy of raising a child, too often without partner support, fewer funds and a restricted social network. Since these 2 conditions cannot be compared, all the research that make these comparisons are of no scientific value thus of no value for this review that relies for it's conclusions so heavily upon them.
c) Reality distorting measurements. Although insisting on "validated measures" like the DSM IV to evaluate a woman's mental health, the reviewers ignore the mounting criticism of the DSM and the fact that dichotomous measure are bound to distort reality almost all of which is spread on a continuum. None of the studies used Visual Analogue Scales which are easier for the subject and which can measure the full spectrum of possibilities between two extremes. Having badly distorted the reality of a person's symptoms or experience, these studies are of no empirical use nor is this report.
d) Not controlling for their own biases. It is obvious that not only the researcher but these writers of the draft were greatly influenced by their own biases. The biases were blatant in: the choice of research, grading of each study, criticism of research that did not coincide with their presumptions and praise they gave to the studies which supported their prejudice. They cited Major more often than any other study although by any unbiased assessment it would not have been published, in the main because her conclusions were based on the responses of follow-up group which had a racial skew and were only 42 % of the original sample. They even attempted to make Major's small study look better by repeatedly describing the locus of the abortions as "3 hospitals" when the author indicated the were "2 clinics and one physicians office". It would not have been difficult to control for or at least measure the extent of their bias by having truly independent check assessors and calculating the percentage of agreement in their ratings. The authors either assumed they were not biased or the influence of their bias could not sway their conclusions or were not detectable by readers or they knew what conclusions they would come to in any event.
e) Although the writers noted the lack other relevant factors to be measured or controlled for, these deficiencies seemed not to affect their ratings of research. Some obvious gaps they did not mention included: quality and quantity of partner support (which we found had a very large effect) sleep patterns, multiple sex partners, sexually transmitted diseases, quality of delivery, bonding and others that have a large effect on mental health.
f) There exists gross under-reporting of abortion especially in the UK probably 50 - 60%, the USA, 50%, Canada it isn't possible to know because Stats Can no longer collects any abortion statistics. Without a correct prevalence it is impossible to know how many women have unreported abortions and thus what percentage of them have post abortions effects. Since women who have a good reasonably good outcome from abortion are much more likely to return for assessment at follow-up while women who are well post partum see no need to spend money of a physicians visit post pregnancy rates of health are badly skewed. The only study that could address this problem is that of Fergusson which the writers did not rate highly. This is more a comment on the writers lack of objectivity than it is the quality of his research.
g) Lack of clinical sophistication. If the writers had clinically evaluated or attempted to treat with psychotherapy thousands of post abortion women, (I have) they would have been more aware of less reported symptoms and major unresolved conflicts. They would also know that clinical states cannot with any accuracy be diagnosed by "standardized tests". They would also know that many diagnoses are not exclusive and test cannot be relied upon to detect how much of each a patient has.
h) Human ecology. Although aborting women is the most frequent procedure in medicine and affects probably more than 100 million women each year, the authors ignore what effect this has on populations. From an ecological perspective, the wide-spread destroying of their own young is the most unnatural activity for any species, homo sapiens included. It is so unnatural that people in surprise or shock say of a young woman who has just had an abortion, "not her" "she must have been out of her mind" "now she has really lost it" etc. This is probably as accurate a description of abortion in some women as any. They experienced temporary insanity and now must fight their way back by dealing with harsh reality.

CONCLUSIONS
1. This "systematic review" is not systematic and is so deeply and extensively flawed that the authors conclusions must be completely discounted.
2. The only reliable conclusion anyone can make on the basis of this review is that there is no evidence that the current practice of abortions provides no benefit in treating or preventing mental illnesses.
3. The government must clearly decide whether or not abortion providers are performing a technical service for women who elect to have an abortion or providing an essential health service. The current law clearly indicates performing abortions is part of medicine.
4. If the UK government continue to rule that aborting patients and terminating the earthly existence of the preborn infant is legal only when there are proper medical indications, they must insist all the other medical guides and constraints for providing good medicine be applied.
5. Since there is no evidence of benefit to patients women, men and children, abortions must be treated as an unproven remedy.
6. This means it must be withdrawn from the market and not be charged to taxpayers until it is clearly established: a) what are the scientifically established indication, b Is it therapeutic, c) are the hazards less than the benefits, d) Are other treatments which are less invasive and more reversible tried first, e) Are 2nd opinions available and used, f) Is it done in good faith by the abortionists ( they are convinced they are providing good treatment based on their extensive knowledge of good and pertinent research and by a careful complete follow-up of their own patients. g) Are patients able to provide fully informed consent which includes seeing their infant on a good sonogram.
7. It also means that there must be more and proper research with equal access to funding and journal publications.
8. The government must make the above changes quickly because they are now imposing taxes on about ½ the population who do not wish to contribute to the killing of preborn infants and feel guilty for aiding and abetting this practice, especially as it does no one good.
9. Since no one appears to have an explanations for such discrepant findings and why if there is such a devastating effect on individuals, this effect so hard to find.in large populations. I hypothesize that this is because research is not differentiating 4 essentially different groups of women: a) those who are hardened or embittered or ideologically convinced that having an abortion is a woman's basic right. Some of these women deliberately conceive in order to abort and thus show themselves and the world how much power over matters of life and death woman have. For these women having an abortion no matter how painful is asserting themselves and they will deny to themselves and the world that they are tough and can take it is stride. Post abortion they will deny suffering and symptoms to any researcher and so be graded unaffected in the data.
b) Sensitive women who are healthy and whole. Because of their acute perceptiveness, they feel devastated by their decision to abort and cause suffering to an innocent unsuspecting infant no matter how small and even if they were responding to considerable duress. They may have many symptoms which are graded as an illness but they are still essentially normal. Paradoxically the more human and whole they are the more they suffer.
c) Vulnerable women who may of may not have had treatment prior to an abortion but who were damaged by childhood mistreatment. For them an abortion may be a reenactment by proxy of their mother's abortion. This would help explain why we find that one of the closest association to a choice to abort is the subject having a mother who aborted.
d) Resilient adaptive women who can take a great deal of traumatizing of almost any kind. These women appear in the short term to be unaffected by abortion but decades later present as clinically depressed. This may occur when aging and frailty weaken their ego defenses. When they are able to connect their symptoms to a much earlier abortion, they are usually much relieved.