Tuesday, April 16, 2013

U of Penn study of Metzitza:The story gets weirder and weirder!

The issue of whether metzitza is an inherent risk factor for herpes and therefore should be banned - has been debated for a while now. See previous post and its comments. The Aguda in its defense of metzitza against the NYC Department of Health cited a metastudy from the U of Penn which analyzed previously done studies and which the Aguda  claimed said that no causal relationship was found between metzitza and herpes in infants. Now the U of Penn has issued a statement - bizarrely through the Forward and not directly - criticizing the use of the study which they said in fact concluded the opposite of what the Aguda said it did. They also added the study was not only not published  it was not even peer reviewed and was meant purely for internal use at the U of Penn. The study was not released and not even doctors can get copies without some sort of protexia. At this point it is very unclear why the U of Penn did the study in the first place. It is highly unlikely that U of Penn was interested in knowing whether they should allow metziza in their medical facilities - so what purpose did it serve. They are not making the study available - so how did the Aguda get a copy? How could the Aguda conclude the opposite of what the U of Penn claims the study says? At this point all one can conclude from the information funneled through the Forward - is that the U of Penn is claiming that it is possible to get herpes through metzitza - a point that no one contests anyway! The only relevant question is whether metzitza done by a mohel following proper procedures causes herpes. In short the whole incident is weird, makes no sense and  even smells funny. It does absolutely nothing to resolve the issue.   

Update 4/16/13 Here is an Israeli study of complications in mila  IMA 7:368 Ben Chaim et al 2005

Update 4/16/13  New study indicates mohel is significant source of infection

Update 4/16/13 Israeli Rabbinate's directives for metzitza

Update 4/16/13 U of Penn just sent me a copy of the report in response to a simple request for one
=============================
Forward   The authors of a research report on metzitzah b’peh say their study is being distorted by defenders of the controversial ritual circumcision practice who claim that the procedure poses no risk of neonatal herpes to infants.

In an April 9 press release headlined “Ivy League Study Casts Doubt on Claims that Jewish Tradition Leads to Herpes in Infants,” a public relations agency representing Agudath Israel of America and several other ultra-Orthodox groups sought to debunk the public health consensus on metzitzah b’peh, or MBP. The press announcement claimed that a study conducted at the University of Pennsylvania “found little evidence to support the claim that circumcision ritual is infecting infants” with herpes simplex virus.

The announcement noted that the UPenn study was submitted as evidence in an appeal filed by the ultra-Orthodox groups, who are plaintiffs in a suit opposing regulation of the practice by New York City’s Department of Health and Mental Hygiene.

“We have been saying for years that the evidence attacking this religious practice is highly dubious, and now we have world class doctors agreeing with us.” Rabbi Gedaliah Weinberger, chairman emeritus of the board of trustees of the Aguda, said in the press release.

But in a statement released Monday, the University of Pennsylvania’s Center for Evidence-based Practice termed it “regrettable that our evidence review was manipulated for purposes other than advising physicians of important clinical risk factors for newborns.”

“The unpublished report was used without our knowledge or consent and importantly, without proper context,” the center’s statement said. “Further, a subsequent press release mischaracterized our review by implying that there is no causal relationship between circumcision performed with oral suction and the transmission of neonatal herpes simplex virus (HSV) when the full report on the existing evidence concluded this link does exist.” [...]

38 comments :

  1. I suspect the authors, in collaboration or as a favor to Agudah initiated a study improperly (since this is a question not pertinenent to the applied mandate of the center). Once Penn caught wind they disowned it and internally disciplined the folks involved but prefer not admitting their internal processes were corrupted.

    Though they would have loved to have the story disappear it was too late because Reuters had picked up the Agudah press release. So Penn disavowed the Agudah conclusion and is trying to deep six the story.

    ReplyDelete
    Replies
    1. Why is it listed on their web site

      http://www.uphs.upenn.edu/cep/reports/index.html

      Don't understand - haven't seen the study - but it wasn't an experiment that generated new data.They just analyzed previous existing studies. Why is it listed by the U of Penn? If they were doing the Aguda a favor they don't need the U of Penn. Or does that mean that the authors wanted to give it more serious weight as a U of Penn study and are working to support the Agudas position?

      Delete
    2. It's listed, not published, not printed, not available. It's like saying "I heard Rav Moshe said..." without any proof other than a chapter title somewhere.

      Delete
  2. I got a copy a few days ago directly without any protexia whatsoever.. Stop making up stories.. Most likely at this point that the anti-mbp anti-orthodx > read Forward < people are trying to cause a fuss about it they're not interested accommodating potential rabble-rousers by sending it to them..

    But more importantly, the points made in the study rebutting the "evidence" in the other "studies" have yet to be explained away..

    ReplyDelete
    Replies
    1. You simply asked the U of Penn for a copy and they sent it to you? Could you send me a copy? yadmoshe@gmail.com

      Delete
    2. Just sent a request to stephanie.dunbar@uphs.upenn.edu

      Stephanie Dunbar is the person listed on the U of Penn web page which lists the contested report.
      ===============================
      Evidence Reports - Report Type

      Please click on a category.

      To view completed guidelines, please visit our intranet site or call or email Stephanie Dunbar | 215-662-2463 |

      Each title shows the year, month, and date. For example, if a report was created on February 1, 2010, then the title would appear in the following way: 2010 02 01 | Title.


      Review

      2013 03 26 | Cognitive and procedural skills of aging physicians
      2013 03 01 | Intravenous or rectal acetaminophen for post-operative pain
      2013 03 01 | Disinfecting caps and antimicrobial locks for reducing blood stream infections associated with central venous catheters
      2013 02 06 | Relationship between hemoglobin A1c levels and complications after elective hip and knee surgery
      2013 02 05 | Esophageal temperature measurement to prevent complications of catheter ablation
      2013 01 21 | Risk of neonatal herpes simplex virus type 1 infection associated with jewish ritual circumcision

      Delete
    3. You see now. All you had to do was ask them for it. That simple. Not a big deal at all.

      Delete
  3. Nice, Harold; I'm still waiting for mine. The literature survey concludes, as the U Penn spokesperson said, that MBP is linked with neonatal herpes probably in large part due to the most recent of the articles surveyed, the one from the New York Department of Health which isn't too bad. I'd say the NYDOH authors may have learned from the weaknesses of previous articles.

    In any case, all of them are case reports or compilations thereof. None of them whether favoring the Agudah's spin or not (actually, if you're going to use scare quotes, "rebutting" would probably be the place to use them) are really studies or "studies" as generally meant, namely prospective with standardized diagnostic criteria, etc. Case reports are as good as we're likely to get for now, at least until the AI calls for a rigorous study. That type of support, in addition to finding funding would probably be necessary to get such a study done.

    Unfortunately, if R' Tzadok's take on matters is correct, the rabbinic authorities in the communities that insist on MBP in the US are protecting one or more mohels known to have active herpes and likely to be the source of most of the reported cases, so I wouldn't hold my breath waiting for such a study.

    ReplyDelete
    Replies
    1. Such a study was done in Israel. Ben Haim Et Al 2005, and printed in the Israeli Journal of Medicine. IMAJ 7:368.
      They studied over 19,000 males 83% were circumcised by a tradition mohel using MBP and 17% were circumcised by a doctor. They had a total infection rate(of all types) of 0.01%. The study included this line:

      חשוב לזכור כי למרות ביצוע מציצה מסורתית בפה, במהלך המאה האחרונה לא נצפתה עד העשור האחרון פגיעה בנימולים. כמו כן מחקרים שנעשו בארץ הראו רמת בטיחות שווה בין בריתות שבוצעו בידי מוהל מסורתי המבצע גם מציצה בפה לבין בריתות שנעשו על ידי רופא
      , הובא באברהם שטינברג שם,

      There is a roughly comparitive study(just on the risk of circumcision in general) done in hospitals in Washington(state). There they studied 30K males, had no mohelim, and had an infection rate of 0.008%. Now if I remember statistics, and Chem P<0.01 is the ideal, as it is considered statistically irrelevant.

      Delete
    2. I think I am missing something.

      1) Statistical significance is not a rate of infection of 0.01. It is the likelihood that it happened by chance less than 0.01. Or are you saying that the difference between an infection rate of .01 and .008 did not approach statistical significance?

      2) Infection rate per se is not of interest - it is rather infections that can cause permanent or life-threatening harm. Was the rate of life threatening infections different between the cases that had a mohel and those done by doctors?

      3) This study raises an interesting point - considering the likelihood of getting serious infections is higher in a hospital then elsewhere. It could be in fact that the likelihood of infection from a mohel with metzitza in a non-hospital situation is the same as infection from mila in a hospital without metzitza - and both have higher infection rates than mila without metzitza in a non-hospital setting. Is there any legislation being proposed to prohibit circumcision in a hospital?

      4) Did the Aguda utilize the Israel study to defend MP against the N.Y.C Department of Health? If they did - then who cares about the U of Penn study. If they didn't - why not?

      Delete
    3. Ben Chaim et al study is available in English here

      http://www.ima.org.il/FilesUpload/IMAJ/0/50/25254.pdf

      Delete
    4. 1) Statistical significance is not a rate of infection of 0.01. It is the likelihood that it happened by chance less than 0.01. Or are you saying that the difference between an infection rate of .01 and .008 did not approach statistical significance?
      The difference between an infection rate of .01 and .008 has no statistical significance.
      I would have to access the study again, but there were only two infections in total, so I don't know if HSV was even considered.

      Delete
  4. Consider the language being used. No one is suggesting banning metzitzah b'peh, only DIRECT metzitzah b'peh. You want to suck blood? Go ahead, just use a sterile intervening pipette.
    What's more, word out of Israel is that despite the safety of MBP that Rav Tzadok attests to, the Chief Rabbis are urging mohels to avoid direct MBP. Kofrim!

    ReplyDelete
    Replies
    1. Did they say the reason is fear of infection? Can you send me a copy of the request from the Chief Rabbis concerning this matter?

      Delete
    2. This the link to the article in Maariv

      http://www.nrg.co.il/online/1/ART2/459/447.html

      Delete
    3. What's more, word out of Israel is that despite the safety of MBP that Rav Tzadok attests to, the Chief Rabbis are urging mohels to avoid direct MBP. Kofrim!

      Problem with word on the street is that it is often wrong. I'm not going to get into what the certifying branch will or won't let you do. Or the proper answers you have to put on your test. Mostly because well, this is Israel and I live in Jerusalem, and those things are likely to be different say in Efrat... However the official position paper of the Rabbinut which is now on their website does not urge mohelim to avoid MBP. It urges Mohelim to follow their protocols and to let the parents decide.
      http://www.rabanut.gov.il/VF/ib_items/443/%D7%9E%D7%A6%D7%99%D7%A6%D7%94%20%D7%99%D7%A9%D7%99%D7%A8%D7%94%20%D7%91%D7%A4%D7%94%20%D7%91%D7%9E%D7%99%D7%9C%D7%94%20%E2%80%93%20%D7%A2%D7%9E%D7%93%D7%94%20%D7%A8%D7%A4%D7%95%D7%90%D7%99%D7%AA-%D7%94%D7%9C%D7%9B%D7%AA%D7%99%D7%AA.pdf

      Delete
  5. " There they studied 30K males, had no mohelim, and had an infection rate of 0.008%. Now if I remember statistics, and Chem P<0.01 is the ideal, as it is considered statistically irrelevant. "

    It sounds like you didn't remember statistics right.

    ReplyDelete
    Replies
    1. It sounds like you didn't remember statistics right.
      Maybe you are right. It would appear according to these three sights
      http://www.numberwatch.co.uk/significance.htm
      http://www.statsdirect.co.uk/help/basics/pval.htm
      http://en.wikipedia.org/wiki/P-value

      That the scientific threshold is actually P<0.05 for it to be considered statistically irrelevant.

      Delete
  6. New study from Israel: http://www.ncbi.nlm.nih.gov/pubmed/23334339
    None of the mothers had documented intrapartum visible genital HSV lesions or a previous history of genital herpes. Ritual circumcision was the source of HSV-1 transmission in 7 infants (31.8% of cases).

    April 3 2013 alert from NYC DOH citing it: https://a816-health29ssl.nyc.gov/sites/NYCHAN/Lists/AlertUpdateAdvisoryDocuments/2013%20ALERT%20mpb%20HAN.pdf

    ReplyDelete
    Replies
    1. Ritual circumcision was the source of HSV-1 transmission in 7 infants (31.8% of cases).
      A rather misleading statement on your part considering the study concluded
      The incidence of NHSV infection in Israel was found to be similar to the lower part of the scale reported in the United States, however higher than the incidence reported in Canada or in Europe.
      Meaning that Israel had lower incidence than most of the US, and slighly higher than Europe. Which according to their study has a total infection rate of 0.008 and an MBP infection rate of 0.0028%
      To put this in perspective the fatality rate of automobile occupants is 0.014%. Meaning that your child is five times more likely to die in a car crash on the trip home from the hospital than he is to contract HSV-1 from MBP. Are you going to want to outlaw auto travel for infants now?

      Delete
    2. Not misleading at all. 1/3 of a lower incidence -- when preventable -- is still too many. 70% of those surviving the infections will have systemic, including CNS damage; by this study that would statistically mean ~5 babies. That's 5 families with heartbreak ahead. Will you be making a house call on their families to explain that their baby boy had to take one for the team and how it's a big zechus?

      What exactly do you mean by "fatality rate of automobile occupants?" Please cite the source so I can understand what you're talking about. My cursory scan shows the WHO listing the USA as having 8.5 fatalities per billion vehicle km and Israel as having 5.9.
      That sounds like a very long way back from the hospital.

      Delete
    3. 70% of those surviving the infections will have systemic, including CNS damage
      Where are you coming up with those numbers?

      Fatality rates according to the NHTSA.

      Delete
    4. Recent birth statistics for Israel showed a bit over 121,000 (ken yirbu!) Jewish babies. Koren et al observed a neonatal herpes incidence rate of 8.4 per 100,000 live births. That would be around five baby boys a year who wind up with neonatal herpes. Let’s make sure they all have a Brit; at current utilization rates of MBP, that’s 1.6 cases of neonatal herpes per year due to MBP per Koren et. al. With a 70% complication rate of neonatal herpes, that would mean just about one baby damaged per year by not putting a glass tube between the mohel’s mouth and the fresh wound.
      One in about 60,000. I make that about 0.0016% which is in your ballpark. So your figures and the new ones are pretty similar.

      But again, 0.014%. per what? I'll get out my napkin again.

      Here's from the NHTSA: "2009 saw the lowest fatality and injury rates ever recorded: 1.13 deaths per 100 million vehicle miles traveled in 2009, compared to 1.26 deaths for 2008." Israel's roads are actually a bit safer(!) according to the WHO. But I'll use the US figures.

      You are contrasting a .0028% infection rate per Brit with MBP to a .00000113% chance of death per mile. Still sounds like a long car trip home from the hospital to make your statement work.

      Delete
    5. I was using rates per 100,000 on the road. It have more correlation to the matter at hand.

      Delete
  7. I find it misleading for Agudah to say no evidence = no risk (which in and of itself is not true and is what Penn is objecting to here) and at the same time failing to disclose WHY there is no good evidence, which is because ALL of the suspected mohelim have refused to provide the very evidence the investigators would need, a cheek swab under forensic conditions (in the presence of an investigator).

    ReplyDelete
  8. Despite Rav Tzadok's fine statistical analysis we're missing a point here:
    Halacha doesn't always care about statistics!
    Like I mentioned a few posts back we clear the bread knife from the table because of one incident almost 2000 years ago. We don't say "Well 0.0000005% of Jews bentching with a knife in front of them will stab themselves out of grief for the loss of the Temple so let's leave the knife where it is".
    Standard MBP fits the HSV mode of transmission. There is no denying this. There are multiple studies as noted in this and the last post on this subject showing a higher infection rate of HSV after MBP. This is a completely preventable issue that can be performed with a pipette and accomplish the exact same end without risking the infant. Yet one study from Israel which does show a small infection rate and a now nearly-totally discredited study from Penn oppose this and get waved around as the authoritative words on the subject. Why is there such resistance to facts? Why is there such an insistence on putting infants in harm's way?

    ReplyDelete
    Replies
    1. Garenel you are missing a point which I have answered several times. The issue of a knife is a practise which I think is only a minhag - but even if were rabbinic there is no question that it would be eliminated even for a sofek of being life threatening.

      Mila is doreissa. Those who hold the metzitza is required can't so readily dismiss it. Mila itself is not entirely without risks and yet it is done. According to your argument - Judaism prohibits doing mitzvos which involve in a possible danger. Proof that your argument is wrong is that even in a family that a child has died because of mila - it doesn't exempt his brothers from having mila. It is only if a baby has two brothers who died from mila that he is exempt.

      Delete
    2. I'm not missing the point. Consider: we have a mitzvah to wave a lulal d'oraisa on the first day of Sukkos but because of the fear of carrying on Shabbos we miss doing it every few years. As you mentioned, a baby whose 2 brothers died during mila doesn't get one of his own. We don't say "shomer mitzvah lo yeda d'var ra" or something like that. We just skip the mitzvah.
      Now we have metzitzah, which according to many mephorshim isn't even a part of the milah but a preventative health measure to ensure the baby's safety after the procedure. It's a machlokes if it's even d'oraisa but saving lives and preventing harm to others certainly is.
      So we're going to skip lulav because of a risk of chilul Shabbos but we're not going to adjust (not stop, not eliminate, not ban, just do it slightly differently) MBP despite the mitzvah of preventing harm?

      Delete
    3. Or think of it this way: Let's say a new kind of fungus infects almost all palm trees that are sources of kosher lulav branches. This fungus causes a severe blistering rash on human skin in about 0.1% of people who touch it. Blistering, like it causes a painful rash that never goes away. Rarely, like in 0.01% of people who touch it, it causes an allergic reaction that is almost always fatal.
      Now, what would the halachic response be to this?

      Delete
    4. Why are you ignoring the case of one brother dying - and yet we give the baby a bris?

      If you hold like the many authorities who hold that it is not an essential part of bris milah - then there is no major problem of not doing it if there is a possiblity of danger.

      You can't combine views. Those who posken that it is essential don't count the one that say it is not essential

      Regarding your case of lulav and shofar - that is an established halacha. Not in every case do we change halacha because of a possiblity of harm. Or alternatively according to your logic I'll start with the case of mila and since we ignore the possiblity of danger presented by the death of one brother that "proves" that halacha is not modified for a sofek of danger.

      You are simply making a chulent out of halacha and ignoring the accepted principles. You can't generalize from one case and ignore all the other cases which contradict your views!

      Delete
  9. FYI,

    Anything with p-value < .05 is considered statistically significant, not insignificant. Basically the p-value can be viewed as a measurement of the probability that the difference you're measuring is noise, so if you have a p value of .05, there is a 5% chance that the difference you're measuring is noise, and a false positive.

    Interesting discussion though.

    ReplyDelete
  10. The problem is that the brief filed in the US Court of Appeals by Agudath Israel, et al., on April 1, 2013, cites, several times, to an internal, unpublished U. Penn. report, prepared in December 2012. This report, according to U. Penn., is only available doctor-to-doctor. It obviously wasn't part of the record in the lower court. How are the City's lawyers supposed to comment on the Penn report? How is the three-judge appellate panel supposed to read the report? They can't. And U. Penn., in their press release, is saying that Aguda et al. are distorting what the report actually says. And, strictly speaking, this press release is also not part of the record that the US Court of Appeals is supposed to review. I would not be surprised if the City of NY files a motion to strike the Aguda brief, and forces them to file a new brief which does not cite to the Penn report. I would not be surprised if the Court of Appeals grants the motion. A very basic rule of appellate practice is that briefs may not cite to facts dehors (outside) the record made in the court below.

    Here's one excerpt, among others, from the Aguda brief, pp. 8-9, citing the Penn report:

    ".....an independent review by Penn Medicine’s Center for Evidence-based
    Medicine recently concluded that the evidence in the Department’s report was
    'significantly limited' by the 'very small number of reported infections,'
    'incomplete data,' 'confounding factors,' and 'methodological challenges.' Joel
    Betesh & Brian Leas, Risk of Neonatal Herpes Simplex Virus Type 1 Infection
    Associated with Jewish Ritual Circumcision, PENN MEDICINE CENTER FOR
    EVIDENCE-BASED MEDICINE at 1 (Dec. 2012) ('Penn Review')."

    In a manner of speaking, you can't cross-examine an internal report not part of the record below.

    ReplyDelete
    Replies
    1. I just got a copy of the report and while the above paragraph is definitely there - it in no ways gives metzitza the green light. It basically says that it is possible for a mohel to transmit disease through metzitza but there are not enough cases to say anything more definitive and that more research is needed.

      The Aguda should not have used this study in support of its position

      Delete
  11. Wolfson Hospital says MBP causes HSV in babies:

    http://www.vosizneias.com/128402/2013/04/17/jerusalem-israeli-study-shows-direct-link-between-metziza-bpeh-and-herpes

    ReplyDelete
    Replies
    1. I provided a link to this study above



      Update 4/16/13 New study indicates mohel is significant source of infection

      http://www.nrg.co.il/online/1/ART2/459/447.html

      Delete
    2. The American Academy of Pediatrics ("AAP'), the Infectious Diseases Society of America ("IDSA"), the Pediatric Infectious Diseases Society ("PIDS”), and the American Sexually Transmitted Diseases Association ("ASTDA") stated that direct oral suction increases the risk that a neonate will acquire herpes simplex virus ("HSV") and other communicable diseases. These groups represent hundreds , if not thousands of doctors. Every expert in infectious diseases, other than Daniel Berman, says that Metziza is not safe.
      Rav Moshe Feinstein, ZT”L, diplomatically wrote to someone who was under the impression that Metzitza was part of the Mitzvah of Milah that it must have been a “slip of the pen,” as the Gemarah states clearly that Metzitza is a medical procedure. If so, the medical procedure should be done in the safest way possible, i.e. with a sterile tube.

      Delete
  12. Infectious Diseases Society of America ("IDSA"), the Pediatric Infectious Diseases Society ("PIDS”), and the American Sexually Transmitted Diseases Association ("ASTDA") stated that direct oral suction increases the risk that a neonate will acquire herpes simplex virus ("HSV") and other communicable diseases.

    ReplyDelete
  13. The importance of the new study on neonatal HSV in Israel should not be underestimated. The study by Koren et al was published in the Pediatric Infectious Diseases Journal in February. It is a retrospective review of cases of neonatal HSV diagnosed in 5 hospitals in central Israel. There were 22 cases of neonatal HSV identified. Nearly 1/3 of the cases (7 cases) formed a unique subpopulation - all male, all following ritual circumcision with MBP, all occurring later than the other 15 cases, and all with lesions on the genitals. This compares with the other 15 cases that were evenly split between males and females, all occurring at an earlier age, and none with lesions on the genitals. In addition, two pairs of the 7 cases were each associated with one mohel (ie 4 cases associated with 2 mohels). Perhaps someone with more statistical training than me would like to comment on the likelihood of all this occurring by chance.

    I read through the article by Ben Chaim et al. It was interesting, though a couple points should be noted. One is that the incidence of MBP in the ritual circumcisions was not mentioned. Without that data, this study cannot be used to judge the safety of MBP. But my main question on this study is, how were cases of complications identified? The article is not clear on this point, but it sounds like these patients were identified as having a post circumcision complication by an urologist. Neonatal HSV is not treated by urologists. ALso, episodes of neonatal HSV may not have been identified as a post-circumcision complication. If the authors did not specifically look for cases of neonatal HSV as a complication of circumcision, they of course would not find any cases. Unless that information is available, this data cannot be used to show the safety of MBP vis-a-vis neonatal HSV.

    The bottom line is the article by Koren etal is in my opinion an insurmountable argument against those who would say that there is no connection between MBP and neonatal HSV, and validates the data in other studies (such as the NY study) that support that link. The only argument left to those in the pro-MBP camp is that a few dead or neurologically devastated babies every decade is really an acceptable casualty rate of the procedure.

    ReplyDelete

ANONYMOUS COMMENTS WILL NOT BE POSTED!
please use either your real name or a pseudonym.