 Good afternoon, and welcome to this public meeting of the United States Consumer Product Safety Commission Today we're or this afternoon I should say since this morning. We had a hearing on priorities This afternoon we're going to focus on data and the importance of data to this agency's work We like to say we're a data-driven agency. We like to be a data-driven agency and More and more as we go forward data is only going to be more critical. I do have to point out though from a Ligistic standpoint. We do have a hard stop at five o'clock and that is because Dramatic pause commissioner Adler is being Feted downtown and he's going to receive a lifetime achievement award for his work in the consumer industry And so I do want to give him a round of applause So we do want to make sure we have a chance to honor our colleague So we're going to follow a process that we followed this morning We've got three panels and we're going to give each of the panelists 10 minutes to testify and the commissioners are Going to stick to five minutes per question. I do want to note a couple of other things before we turn it over to the first panelist that This really is a critical hearing for us and we wouldn't be doing this But for the leadership of a couple of commissioners and revert call to my right, but most importantly To my left is to my far left is Commissioner Robinson from the moment that Commissioner Robinson came to the agency almost two years ago to the day We're about a week's shy of that She has been a champion of data and of the agency getting better at using its data and finding better sources of data and Her work and the work of her staff Boas Green, Heather Bramble and Doddy Lee They've inspired this commission really to have this hearing and to focus that much more on data And they've really been inspiration behind at least my two critical goals in this area Which are as I mentioned Trying to do better with the data that we get and also trying to find better data Sources and better data analytics and we've been very creative and very open-minded about where we look for that information About a year ago for sports fans sports illustrated had a cover piece that talked about how the Houston Astros are going to win the World Series in 2017 and that led us actually to Houston to go visit the Astros and to bring with us some of our experts To have an expert to expert discussion about what they do because we felt like there were a lot of similarities And it turns out we were on to something because lo and behold the st. Louis Cardinals allegedly were hacking into the same system So that actually validated our trip. I felt good about that considering the success that the Cardinals have had We've also had extensive discussions with the white with the chief technology offer officer who is at the White House and her staff and We're in a unique position now where we announced actually a little while ago a vacancy to fill our chief information officer So it's very well-timed. We're eager and hopeful that we'll get the type of applicants That will allow us to really see through the vision that we all have and to carry out The goals and the purpose that Commissioner Robinson has tried to bring to this issue So with that I'm going to turn to the panel. We have dr. Kieran Quinlan again from AAP We have dr. Brian Rudolph From he was an assistant professor pediatrics at the Albert Einstein College of Medicine Children's Hospital of Montefiore We have dr. Dan Budnitz and he is from the Centers for Disease Control and Prevent in medical safety program division of health care quality promotion and Mr. John Myers chief of the surveillance and field investigations branch division of safety Research at NIOSH. So thank you very much gentlemen and dr. Quinlan if we can begin with you, please Well, good afternoon again Chairman Kay and commissioners Adler Burkle or more. I knew I was going to stumble on Mohorovic and Robinson As I mentioned this morning, my name is Kyren Quinlan and I'm here today on behalf of the American Academy of Pediatrics a non-profit professional organization of 64,000 primary care pediatricians pediatric medical subspecialists and pediatric surgical specialists dedicated to the health Safety and well-being of infants children adolescents and young adults I'm a pediatrician and currently serve as the chair of the American Academy of Pediatrics Council on injury violence and poison prevention executive committee as well as an associate professor of Pediatrics and preventive medicine at the Rush University Medical Center in Chicago The AAP appreciates the opportunity to make recommendations to the US Consumer Product Safety Commission on data sources and consumer product related Incident information and how such data can be used to inform CPSC's work on hazard identification risk management and regulatory enforcement pediatricians especially those involved in Child injury Epidemiology and prevention utilize all of CPSC's data sets and value them greatly as they work to better understand Treat and prevent injuries in infants children and teenagers The significance of these information sources in protecting children's health and safety cannot be overstated Specifically I and my AAP member colleagues use the following CPSC information and data sets first the National Electronic Injury Surveillance System and the nice AAP the all-injury program We found access to the nice AAP Simplified by the web-based Injury query and reporting system the whisker system Through the partnership between CPSC and the CDC At and the National Center for Injury Prevention and Control at the CDC the nice and nice AAP have served remarkably well To provide descriptive epidemiology identify emerging issues and evaluate effectiveness of interventions While I was working at CDC some years back. I helped Lianist sort of evaluate the expansion of the all-injury program and I was fascinated going out to the various hospitals at the nimbleness of that system how economical the data is And how useful it's been over these years. It's an outstanding data system and CPSC should really be congradulated for administering it for all these years Second the medical examiners and corners alert project Also the safer products.gov public complaint database CPSC in-depth investigation reports CPSC recall data and CPSC injury and fatality data for all terrain vehicles My colleagues and I use these sources of information for both case specific and aggregated data Purposes these data help pediatricians to understand injury trends Obtained statistics measure the efficacy of various product related interventions protections and regulations We were asked to to look at or explain talk about non CPSC data sources that we use and in the world of pediatric injury prevention injury epidemiology and prevention We also utilize data sources compiled by organizations other than CPSC including Vital statistics system of the National Center for Health Statistics often also through the whiskers Website it's very easy access to US fatality data the national poison Data system Also the health care cost and utilization product project kids inpatient database The H cup kid which was established by a fate federal state Industry partnership sponsored by the Department of Health and Human Services Agency for healthcare research and quality the H cup nationwide emergency data sample which produces national estimates about emergency department visits across the country and National inpatient sample which is the largest publicly available all-payer inpatient care database in the US Containing data and more than seven million hospital stays each year. These are also Data sources we use both the Neds and the NIS are a part of the health care cost and utilization project created by a hrq We also very frequently use Various databases from the national highway traffic safety administration Nitsa such as the fatality analysis reporting system the crash worthy worthiness data system and the general estimates system The FARS the fatality analysis reporting system is a census of all fatal crashes in the United States The CDS has detailed data on a representative random sample of thousands of minor serious and fatal crashes and the GES began in 1988 and Data from G for GES come from a nationally representative sample of police reported motor vehicle crashes of all types from minor to fatal And we also use various national surveys such as the medical expenditure panel survey MEPS is also administered by a hrq and is a set of large-scale surveys of families and into an individuals They're medical providers and employers across the United States MEPS is the most complete source of data on the cost and use of health care and health insurance coverage So now for recommendations for improvements to CPSC Datasets the AAP believes there are a number of improvements that can be made to the excellent content of the CPSC's data sets First we urge the CPSC to continuously create new codes within within its databases to better track emerging emerging poisoning hazards from items such as laundry detergent packets packets for food choking and Another example of amusement related injuries Second we urge CPSC to consider sampling urgent care center data nationally Given the proliferation of urgent care centers around the country and consumers increasing use of them CPSC should find a way to incorporate their data into its other emergency care databases One aspect that I would like to draw some attention to as well is I know that the CPSC has had a long history of dealing with safe sleep for infants But I'd like to draw mention something about the recent changes in the epidemiology of Child deaths in this country that I think should draw that to be even more of a priority There's been tremendous successes with graduated drivers licensing In the effectiveness on decreasing teen crash deaths in this country and teen crash deaths have come down significantly in the last five to seven years because of this It is true that right now. There are more babies that die from Sleep-related infant death or if you want to call it sudden unexpected infant death that includes SIDS Includes accidental suffocation or strangulation in bed and the other category of unknown That combination kills more children than all of the teen crash deaths for 15 to 19 year olds per year in this country and All those infants about 90 to 95 percent of those infants die in the first six months of life So if you look at rates, it's a towering difference between those two This is a huge issue. It's truly the primary threat To the life of a healthy infant in this country throughout their childhood In terms of recommendations for improvements to dissemination of CPSC data sets Pardon me In addition to the above improvements to data collection We urge CPSC to make changes to how it disseminates this information to the public Data compilations such as the annual ATV reports should be released in a timely fashion In addition the commission should make nice occupational injury related data Which currently go to the National Institute for Occupation of Health and Safety Safety and Health available to the public This same the same should be done for nice all injury program data, which currently go to the CDC So to conclude the data for the data the CPSC gathers and provides to the public is critical to preventing future injuries and fatalities Pediatricians in particular rely on this information to treat their patients and to help reduce child injury and death rates We are grateful for the opportunity to comment and look forward to assisting the Commission in protecting the health of all children And I'm happy to answer any questions when the time is ready Thank You dr. Quinlan dr. Rudolph So thank you so much for the opportunity to speak here today My name is dr. Brian Rudolph find assistant professor of pediatrics the children's hospital one if you're Albert Einstein College of Medicine in New York I'm also here today as representative of the North American Society for pediatric gastroenterology Hepatology and nutrition or NASP again So NASP again represents pediatric gastroenterologists and related practitioners in the United States of North America Several years ago my colleagues many within our organization rather began identifying children injured by high-powered Magnet ingestions as a response NASP again conducted a member survey to better understand the ingestion incidents Alarming survey results led to several publications of epidemiologic data as well as NASP again strong support of the consumer product safety Commission's effort to protect children from dangerous magnets Including a new safety standard for high-power magnets magnet sets that took effect this year Today, I'd like to briefly share our experience with you all specific Specifically focusing on how we use CPSC databases and others to care for patients It is my sincere hope that this statement offers useful information as you seek to improve your methods of data collection NASP begins concerned with high-powered magnet sets began in early 2012 After a pediatric gastroenterologist posted a case of a child with a poor outcome from a magnet ingestion to an online forum He was curious if other physicians had had similar cases The response was overwhelming and it quickly became apparent that these magnets were adversely affecting children from across the United States Was seemingly high frequency NASP again leadership subsequently were supported a member-led epidemiologic survey from June to October of 2012 Which confirmed a high incidence of ingestion and a significant rate of overall morbidity Dr. Mazena boss at all published the first epidemiologic data on high-powered magnet ingestion shortly thereafter in July of 2013 in this trend analysis the authors demonstrated a 75% annual increase in emergency department visits for magnet ingestion from 2012 to 2011 this data was obtained from the National Electronic Injury Injury Surveillance System or NICE by using quote ingested object and quote in the Queries section the authors and manually reviewed each narrative with the word magnet total 678 to more readily identify cases of high-powered Neodymium magnet ingestions those that were listing magnets as round With the word spherical ball pebble shaped and or small Such as less than one centimeter pebble small or tiny All an important finding this study was unable to provide any outcome data, which is a limitation of the NICE database in November of 2013 a separate group of researchers retrospectively analyzed 72 cases of magnet ingestion Collected from two other CPSC sources the injury potential injury in incident or ID I I am the in-depth investigation or ID I databases in Contrast to dr. Amos's study this article is able to provide information on patient outcomes, but was limited by both sample size and selection bias We all know the conclusion to this story, and I once again like to thank the commission for its role in regulating high-powered Magnet sets, and I sincerely believe that your actions will save children's lives So thank you very much for that But as you can imagine this experience presumably for all of us was not easy As a physician and a researcher it was often difficult to obtain and analyze the data Second much of the information used to form treatment and risk stratification Guidelines namely outcomes data is not readily available within nice The nice database is extremely useful, but has multiple limitations first nice only contains records for patients treated in emergency departments By excluding outpatient physician visits and or acute care centers the actual number of injuries for any given hazard is likely underestimated Second documentation of emergency department visits in the nice data database often does not provide a sufficient level of detail In the case of magnet ingestions for example Researchers were unable to differentiate suspected from actual magnet ingestions Third the information needed To estimate product harm is often obtained only upon manual review of the data This process is laborious limiting to future researchers and likely leads inaccurate incidence estimates Fourth sample estimates provided by nice are reliant upon US Census Bureau historical estimates Fifth nice cannot provide further information on incidence changes over time That is whether an increase in neodymium magnet ingestions for example are due to increased exposure public awareness or increased health care access Sixth as mentioned nice does not contain outcome data In the case of magnet ingestions, we do not know which injections results in surgery or endoscopy prolonged hospital stay or long-term complications and or disabilities Lastly analysis of the outcome information in the IDI and IDI databases is limited by significant selection bias and Generally a small sample size These limitations significantly affect the ability of physicians to establish treatment protocols when foreign body ingestions occur Taking high-powered magnet sets as an example NASA begins treatment guidelines are largely based on expert opinion According to the anatomic position number ingested in timing of the magnet ingestion Further data is needed to identify predictors of poor outcome and or different Differentiate treatment modalities such as watchful waiting endoscopy or surgery Additional outcome data would also be helpful for other childhood ingestions including such current threats as button batteries or detergent pods Unfortunately, I'm unaware of any non CPSC data sets that contain consumer product related incident or outcome data with a high level of detail Two databases maintained by the Centers for Disease Control and Prevention in the CDC the National Hospital care Excuse me the National Ambulatory Medical Care Survey and the National Ambulatory Medical Care Survey are national probability sample surveys But cannot be used to obtain true incidents or prevalence estimates Further the data can be expensive to obtain is often difficult to navigate and has little detail in product specifics or Circumstances surrounding the injury Another CDC database the nice all injuries projects works with a subset of nice hospitals to obtain information on Emergency department injuries from all sources that is not just related to products over which the CDC has jurisdiction It is possible this database would have more information on some products or injuries than the nice data database Others such as the kids inpatient database, which was developed for the health care cost and utilization project and sponsored by AHRQ provides more comprehensive information on inpatient stays But also lacks much information on circumstances surrounding injuries It's also inadequate to generate incidents data The solution to any of these problems is not easy at Minimum might encourage the Commission to address two main database limitations discussed above First our patients could benefit from more robust data sets data sets with increased granularity Which would facilitate the ability to more easily detect injuries from novel products? Second we need additional outcome data one possibility Maybe to ask hospitals contributing to the nice to provide an additional report on any admitted patient Another possibility would be to partner with other governmental agencies such as the CDC to combine existing databases Ideally one with inpatient data or even electronic medical record companies for more seamless comprehensive data transfer There is no doubt that while the solutions these problems are difficult the mission is clearly important I applaud the Commission's effort to better capture product related injury information and once again Appreciate the opportunity to share my opinions with you today. Thank you. Thank you. Dr. Rudolf. Dr. Bednitz Good afternoon I'd like to thank all the commissioners for the opportunity to share some information on the use of nice for Pharmaceutical product safety monitoring I'd like to start with some background on Pharmaceutical product related harms or adverse drug events Which basically are when a patient is harmed rather than helped by taking a pharmaceutical product after they experience a side-effect an allergic reaction Naxanal overdose or a medication error They'll talk about a few things that make nice unique and useful And give some examples of how many are translating data from nice into impactful action And finally I'll conclude with some potential opportunities relating to enhancements So why is nice such an important source of information on adverse drug events? Well, it starts with Sutton's law named for this fellow here who's a prolific bank robber In the mid-1900s who when asked why do you rob banks replied because that's where the money is Now Sutton's law is the basis for a public health approach to medication safety Which can be trans which can be applied as where the greatest number of adverse drug events occur Therein lie the highest potential for overall harm reduction So how common are adverse drug events or ADEs and which drugs do cause the most harms on? 2001 we honestly didn't know on a national level In fact a GAO report from 2000 Concluded that the data on emergency department visits and hospital admissions Were insufficient for estimating overall adverse drug event frequency in 2004 a Collaboration between CDC the consumer product safety commission and the food and drug administration Established the nice cooperative adverse drug event surveillance project And it was inspired in fact by some work done by dr. Quinlan and just two years later We were able to quantify the frequency of hospitalizations and ed visits for adverse drug events And since then we've continued to use the CPSC CDC FDA collaboration and identify the drugs and circumstances That most commonly lead to ed visits and hospitalizations among Americans I want to point out some of the key features that make nice data so useful First is timeliness Well, nice data are available within months of the event occurring most other health services research databases are available years later Second the data are detailed Unlike many other systems. They're not based on ICD-9 or billing diagnoses But based on the actual clinical diagnosis recorded by the treating clinicians You find the narratives that are included nice rich and circumstances that can be used to help guide prevention Nice also is cost-efficient instead of recreating a system specifically for pharmaceutical products This this collaboration has been a model of how to leverage existing federal systems and resources. I Think you'll hear other testimony about how many academic publications result from Nice there've been over 30 publications related to Medication safety and these have been seminal articles and the most influence on journals including JAMA the New England Journal of Medicine and Pediatrics But the purpose of nice is not just to publish articles in the words of our CDC director Tom Frieden The purpose of public health surveillance is to turn data into actions that protect the Americans health Nice data is used by CDC For post marketing surveillance for all types of pharmaceutical products But post marketing surveillance I mean monitoring for safety concerns. They're not detected in the pre marketing stage before the drugs are approved This monitoring has led to labeling changes and specific drugs safety communications on a number of products I'll point out one specific example and that is cough and cold medicines for children Previously these medications were sold for infants in all ages, but based largely on nice data Just a few years ago These medicines were relabeled not to be used by children less than four again based on data From nice that most harms occurred in this age group Nice data has been used by CDC and CMS collaborations that CMS the Centers for Medicaid or Medicaid services In their patient safety initiatives one of these is the partnership for patients under the ACA That based on nice data focused hospital's attention on the drugs identified in nice as the most common cause of emergency department visits and hospitalizations And finally CDC uses nice data in its efforts to reduce antibiotic overuse and prevent antimicrobial resistance to these valuable medications They use nice data on the harms from ad From allergic reactions that may be prevented by unnecessary use overuse Nice data has also led to departmental actions across all HHS agencies Last year the national action plan for adverse drug event prevention was released that aligns medication safety initiatives across all departments of HHS It focuses efforts on three drug classes that were identified by nice Data as the most common cause of serious harms. These are blood thinners Diabetes drugs and opioid analgesics and again this Plan was inspired by nice data Finally nice data are used for set goals for the health of the nation It would be healthy people 2020 objectives the four drug safety objectives that have measurable data are all based on nice And finally nice data are leading to voluntary private sector action as well. I Like to highlight some data on pediatric adverse drug events that is currently driving private sector innovation for medical product safety First companies are now Implementing innovative child safety packaging called flow restrictors to provide safety above and beyond that typically required by child resistant packaging These were flow restrictors now implemented voluntarily on all infant acetaminophen products Based on nice data that identified a new Increase from essentially zero medication Ingestions in children from the from a drug buprenorphine to a thousand over a thousand a year Manufacturers are now starting to implement unit dose packaging for this product details contained in the nice narratives has led to efforts To update our educational messages on safe storage of medications and how to dose medications for example using milliliters To prevent confusion between tablespoons teaspoons and other measures And finally, I'll cite an example from Walmart, which I call their triple play for prescription liquid medication safety Not through regulation, but they voluntarily at Walmart were convinced by the data and nice to now Deliver all their liquid medications from their pharmacies with flow restrictors To include milliliter dosing devices and to include educational messages on safe storage So when I present nice data the questions I most commonly asked about is how do we get more? and I'll go over some of these ways specifics about more and The leading one right now has to do with pharmaceutical pharmaceutical product abuse CDC has noted that there's actually an epidemic going on right now of prescription drug abuse based on death data But in terms of morbidity data Emergency to visits and hospitalizations. It's almost like it's 2001 all over again We do not have national data on ED visits due to prescription drug abuse But because of nice's flexibility there's currently work By CDC FDA and CPSC to expand the definition of the adverse drug event special study to include prescription drug abuse as well Another question that I'm commonly asked is is there a way to increase the number of hospitals to try to develop Even more estimates that might be regional or state of course to get high quality data requires Enhanced good training of the coders to deliver that data So that was always an opportunity for enhancement And finally the data collected by nice are from a probability sample And so it does take some expertise in interpreting these data The last opportunity I will talk about briefly is correlation with other data sources As I mentioned earlier I see nine codes and most commonly used codes for many health services resource health services research research data sources and Correlation with the nice data not to determine if one it may be right or wrong There certainly are different purposes for each but correlation may be helpful for the future Thank you. Thank you. Dr. Butnitz. Mr. Myers Thank you The National Institute for Occupational Safety and Health is pleased to have the opportunity to provide this testimony in Response to the Consumer Product Safety Commission's Federal Register notice of a hearing entitled data sources and consumer product related incident information NIOSH supports the continued product related emergency department based data collection efforts of the CPSC While this data provides useful information to inform product related prevention efforts It also provides a mechanism to collect surveillance data for several other areas of public health importance Including work related areas a priority area for NIOSH CPSC captures all product related injuries treated in EDs using the National Electronic Injury Surveillance System Medical record abstractors review all ED records at a sample hospital and abstract data from all records identified to be product related The data abstracted for a product related surveillance include demographics diagnosis injured body part and Narrative explanation of the injury NIOSH relies on occupational on an occupational Supplement to the nice known as nice work to collect data on non-fatal work related injuries treated in EDs Nice work is funded in part by NIOSH through an interagency agreement with CPSC With data collected from a hospital sample that represents about two-thirds of the overall nice sample The nice medical record abstractors captured nice work data Collecting the same core variables collected in the product related data as well as nine additional variables specific to work related cases including industry and occupation The partnership between NIOSH and CPSC via the nice enables NIOSH to collect surveillance data that would not be otherwise possible NIOSH our NIOSH work is one of two major sources of nationally representative Occupational non-fatal injury data the other source the Bureau of Labor Statistics survey of annual of occupational Injuries and illnesses or soy Collects data from a sample of private industry and state and local establishments based on record-keeping requirements of the occupational safety and health administration Certain types of establishments are not included in the soy because OSHA record-keeping requirements do not apply to all establishments or all work situations specifically self-employed workers private household workers certain public employees and Agricultural establishments with less than 11 employees are not subject to OSHA record-keeping requirements In addition soy only captures detailed data on occupational injuries that result in days away from work While nice is limited to ED injuries It does not face the same establishment work situation or injury severity limitations that are inherent in the soy Consequently although neither nice work nor soy comprehensively captured work injuries they offer complimentary Perspectives that contribute to better understanding of occupational injuries in the United States Nice work has consistently been collected by CPSC for NIOSH since 1998 Its value in utility has been demonstrated through numerous publications products and data requests These data have been used in a multitude of publications covering injuries in specific occupations such as emergency responders and correctional officers and Injuries in specific industries such as construction and fast food Publications have also addressed specific diagnoses such as traumatic brain injury and burns and Also covers specific demographic populations such as younger and older workers NIOSH makes de-identified nice work data available on a publicly accessible quarry site that is used by researchers Accom additions and occupational safety and health professionals to provide national estimates of work related injuries Nice work data are periodically requested for the purpose of providing justification for a proposed prevention effort such as eye and face protection Finally nice work data serve as a source for monitoring injury trends as exemplified by the use of Nice in monitoring us health the US Department of Health and Human Services healthy people 2020 objectives Now I ask has also used nice to conduct follow-back studies Contacting injured workers identified in the nice work data and requesting their participation in telephone interview surveys This extension of the routine nice date our nice work data collection allows capture of detailed injury and Worker information and has led to publications related to blood-borne pathogens exposures and work related violence Results from additional completed follow-back studies related to workplace injury reporting and injuries to Emergency medical services workers are included in manuscripts in process that will be published shortly while NIOSH work While NIA nice work publications are primarily targeted at worker safety Some of the publications are focused on injuries related to products that are also used by consumers outside of the workplace For example recent publications have addressed falls from ladders and injuries from nail guns These types of publications off often have the potential to address a broader audience Including risk and prevention strategies that apply to both workers and non workers The NIOSH fatality assessment and control evaluation program is another data system maintained by NIOSH That looks at work related fatalities that contains consumer product information The face research program is designed to prevent occupational fatalities fatalities across the nation by identifying work situations at high risk for injury identifying fatalities in these areas and Formulating and disseminating prevention strategies Interventions have included fatalities related to products that also Apply to to consumers such as all-terrain vehicles ladders and paint strippers The recommendations found in face reports can often be implemented by both consumers and workers to prevent additional product related fatalities in Conclusion NIOSH depends on CPSC collection of NICE for sustainability of our own occupational injury surveillance system NICE work is a critical component to understanding and preventing National occupational injuries with more than 15 years of funding and collaboration NIOSH supports the continued collection of NICE by the CPSC Recognizing this system is crucial to inform prevention efforts related both to consumer products and the safety and health workers Thank You mr. Myers, and thank you again to the panel for your testimony We're certainly grateful in particular to hear the value of NICE It's something that we can appreciate from a CPSC standpoint, but it's also phenomenal to hear Outside of CPSC across the government and beyond the value of it And we're incredibly fortunate to have an extremely Dedicated and talented division of epidemiology and our director of that division is sitting in the room as Kathleen Strauka And so and mr. Tom Schrader her deputy. I'm sure that they both appreciate Hearing this they're the ones on a day-to-day basis and their staffs who are making this work and keeping this up and running and continually looking for ways to improve it and so it's valuable to us and hopefully valuable to them as well We're now going to turn to five minutes of questions per commissioner and in recognition of Commissioner Robinson's Role on this issue. I'm actually going to yield my five minutes to her for her to use So she has extra time so yielding to Commissioner Robinson Thank you so much chairman K when I first came with this agency and realized how critically important our data are to virtually every product safety Decision being made in the world I started to focus on it and I came to know personally the Outstanding work that Tom Schrader and Kathleen Strauka have done in this area But we have been trying to focus in my office and my personal staff on other ways in which we can make this data better Given the fact that we now know that 90% of the data in the world Came into existence during my two years on the Commission, which is a little frightening and We know that our act systems are excellent, but they're old I mean the nice system is close to if not over 40 years old and given that we're in 2015 what we can do to try to improve that I was absolutely Delighted thrilled beyond belief when chairman K said that he thought that this hearing was a good idea And I went back to my staff and they were all over it. They were crazy about it I have the most energetic Brilliant staff Heather Bramble and Boas Green and Dottie Lee and the three of them really Spearheaded getting in touch with the agencies that use our data people who have ideas about how we can improve it and Several government agents who are going to be talking to us about the ways in which they're improving their data to hopefully give Us some ideas so let me just start quickly With this this area and what I'm going to do is I'm going to spill these questions out and then just turn to you Since I have limited time even though chairman K has been kind enough to give me as five minutes I'd like to first of all talk about urgent care centers because we know that there are holes in our data And we know that that's one of them and many of you have mentioned Most of you have mentioned that you do not have data from this and I've heard this before What we've tried to do in my office is trying to contact urgent care Oh The the organizations of professional organizations the large hospital groups to see if we can figure out how to do this I know I'm from Michigan and when somebody says that there's increasing use of urgent care That's certainly my personal experience that people are going more and more to urgent care facilities instead of emergency departments but my question of all of you and You can just answer me as and we can continue this conversation Afterwards, but do you have any ideas for us on how we might be able to access that data? That is in the urgent care facilities and if you don't do you have any? Relationships with people who might be in the large hospital groups that would have urgent care facilities or the professional Organizations that we might be able to tap into to start getting some of that data Dr. Rudolph you look like you're first thing with something to say. No, I was trying to avoid the question So the short answer is no, I don't have any Great ideas as to how to get that date that that data a lot of the urgent care centers to my understanding are Private businesses or for-profit businesses Although a lot of academic medical centers do have their own urgent care centers So you may be able to capture some of the market in that way some of the hospitals They are already collecting nice data from Priors You're struggling with the same issue that Nyosh is struggling with I'm not surprised and Because I think a major concern what we have Even though, you know nice has been very useful to us is that there is a shift in how health care is being done and Because of that we have made a decided effort to get involved in the electronic health record area and While that doesn't have the perfect solution to CPSC's problem I think it's an area that CPSC probably should be looking into because it is going to happen Whether we want it to or not and we need to make sure that the data we need to Prevent injury in the United States is captured. Well in those systems so Dr. Budnitz did you want to add to that? I was just going to add that we've had some of the same thoughts about how do we capture urgent care Data data from drugstore clinics even but is somewhat reassuring that still the most serious cases will end up being referred to emergency departments and if our focus is on common and serious Injuries, I think there's some reassurance that we should focus on continuing to capture the emergency department data as well as we can I Concur with dr. Budnitz about that. I think the most serious cases if they come to where I work If you know, we're in sort of the acute care part of our clinic You know, we're gonna be sending them over to the ER One one thing I would like to toss out here is You know seeing patients is an absolute joy the EMR is not and I've heard that EMR is something that has created basically homework for every doctor in this country that sees patients It's new couple hours of data entry for every clinical session it's it's kind of a bummer and and I would be really careful about It may seem from some vantage points to be an easy way to get data but I Really believe that the quality of the data that you will get will be dependent largely on The funding that is provided for the data entry process If you're asking a very busy clinician to do your data entry and hope that you can mine that data later I I think you'll have unhappy doctors and lousy data and I think it really should be carefully thought about rather than just thought to be a place where you can get the data And let me just let me turn from that to the electronic health records Obviously the you I'm not telling you anything you don't know that these are going to be required from all medical providers urgent care doctor visits and hospitals And what we've done on this and I know mr. Schrader worked with the ICD 10 Committee to try to see if there was any way that we could capture data that way but but and I'm speaking for you Tom Sorry, but I think that it was it was something that was very frustrating to us because it's mostly focused on billing And so we weren't able to get the kind of information that we wanted And the other direction that I went was actually to go to a company who were named main nameless that is one of the largest medical Electronic medical record providers They do the software for the hospitals and I talked to people at the top of that company and at first They were really excited and then about helping us and then they realized that it would be adding a data field So and I that was my first appreciation about what you're speaking about dr. Quinlan of adding a data field to these Electronic records, so I guess I guess some what what I would ask of you is whether you have any ideas on how we might be able to get These electronic records to signal if an if an incident was related to a consumer product So that we could follow up on it if that was the case and before you answer dr. Quinlan my time unfortunately has expired So I believe commissioner Robinson will pick this up again after commissioner and they're going to turn to commissioner Adler Thank you. Well first let me join in thanking commissioner Robinson and her staff for really raising an important issue and sometimes when you have a historical perspective which I do You can think back in time and the one thing that I do recall was the moments of terror that I had when we had a group of commissioners that would that wanted to abolish nice and The reason was that one of the commissioners had Insisted that we needed to regulate gas cans during the height of the energy crisis And the staff said it's not a problem the only real problem we have is not gas cans But people trying to siphon gasoline from their neighbor's car and we have a lot of poisoning increases and the commissioner was so upset that The commissioner said well, let's just abolish nice and proceed with anecdotal data And I'm obviously not going to mention names, but it is such a joy to see the strength of nice and to see so much Good work going into nice I I did want to address a couple of points that have raised because they all seem to boil down to resources One is why doesn't the commission create new codes as we see new hazards emerging and If you talk to Kathleen and Tom they they have some strong answers It's not that we never create new codes, but if you're going to create new codes It isn't enough just to create the new codes internally you have to go out to every hospital in the system and explain to them And we're not paying them very much to do it how they've got to change the way They record codes and then you tend to lose historical data so that it's not that it's impossible But it's a real problem because I'm always asking and I'm always getting that answer and it always makes sense to me The other point with respect to urgent care centers again the historical perspective back in the day The Commission in response to this very same concern didn't go out to urgent care centers because we didn't have that many around But we had physicians and so we actually Spent a lot of money doing a physician office survey to see what it was that we were missing and my best recollection is and I think several of you have addressed the dr. Budnitz is that and this was to me surprising that The actual patterns of injuries were not that different you'd get the more severe ones going to the emergency rooms But it wasn't the case that you know We were seeing paper cuts at physicians offices, and we weren't seeing them in emergency rooms They just didn't seem to be that dramatic a difference, so that was a reassuring point not that I'm opposed in any way to Surveying urgent care centers, but my gosh, yeah, again, it's the expense and the limited resources So I guess my question to all of you And this would be an impressionistic point. Do you have any sense? That the injury patterns that are going to urgent care centers today as opposed to the years ago And we did it are showing up in a different pattern than the pattern of injuries that we're seeing in emergency rooms And just an impressionistic response if you if you have any thoughts I would guess that they that they are only because of the proliferation of Urgent care centers and how Common they are and you know in New York City where I am every other block. There's an urgent care center so I think that That you probably this is just anecdotal You know maybe opinion, but I probably think that if you looked at it today, you may see different And since we're picking your intuition and impression Can you hazard even the speculative thought about what the differing patterns would look like? No, okay. No, that that's that's certainly fair. Dr. Quinlan It does seem like some of these urgent care centers specifically market themselves as a place for casting of fractures and so they may particularly try to draw Children who have had playground related injuries say or something like that and fall on an extended hand and a fracture of the forearm or wrist and and Families may feel like that is the place that they could go for that type of injury rather than going to the emergency department and that may be particular to you know a region or a Certain area geographically it may be different in one place than another and the nice thing about that is that you might even try Testing that because that would be a much narrower and more doable kind of survey So that that's really an excellent thought. Dr. Budnitz or mr. Myers any thoughts? So I Still believe the most serious adverse events will come to the emergency department and they might be a little bit different flavor Again, just using examples Typically, I can just speak about adverse drug Effects and reactions like the drug rash might be more commonly treated in an emergent and an urgent care center and not need for their Follow-up so you'll might see a little bit less rashes but still the cases where a drug induces for example bleeding or Hypoglycemia that leads to person passing out. I think we'll still go to the emergency departments anecdotally the only thing that I'm aware of is that there is a tendency for employers to use urgent care for medical clearance much more than they Used to rely on emergency departments and physicians now. They're basically running all of that through urgent care That's not really impacting the kind of data. We're getting through nice because we're looking at that first Interaction between the patient and a doc whereas You know these clearances are actually the follow-on after they've already been treated making sure they're fit to go back to work Thank you. Thank you commissioner out there commissioner Robinson Mr. Chair Just in just following up a little bit on this electronic medical records situation I know this is this is new enough that you the none of you may know the answer to this but Do you know for example if we're going to be able to use these electronic medical records to get outcome data? You should I mean it if I see a patient In the hospital everything is documented from the time they step foot in the emergency department to the time they're discharged so every surgery every Follow-up visit with a specialist Medications prescribed everything is captured within that electronic medical record system. It's a matter of how you access that data and get it in a packaged form that Isn't overly cumbersome, but yet still effective and useful before I move to a more micro subject Let me just throw it out there for you that if you if you have any ideas at all in the ways Of which we could talk to we could get electronic medical records to signal if an injury was related to a consumer product And how we could collect that and get outcome data from it it would be huge and we're working on it But I just plant that seed with the with the four of you Have any of you had Dr. Rudolph you may be the one who can answer this the best But if you had any Challenges or as your group had any challenges in reviewing or manipulating the CPS a data due to technical formatting such as API I Technical formatting Actually, I don't know if you can if you have any experience with this so personally I'd say no as I as I mentioned, however Some of the data is not necessarily there that we need and it's a lot of manual review That's because of the lack of coding granularity. Exactly I would say no, I would say that it's very accessible and very easy to manipulate We're currently doing a study with with nice data And it's it's been very Very easy to work with and and it's a yeah, no problem. Glad to hear that Dr. Brunitz and and mr. Myers Do you do either of you use unstructured data like Twitter Facebook anything for purposes of gathering information at your agencies? So I don't know if I can speak for the entire agency, but what I can't speak to is for drug safety in general and I think some of those Non-traditional data sources can be useful for Signal detection and hypothesis generation, but they still require a little bit more traditional methods for Strengthening that signal or validating if that you know Twitter Is accurate tweet is accurate From my Arsha standpoint, I'm not aware of us using it to collect data We do use it to get information out quickly whenever we identify a new issue So that's been the primary purposes for And and turning to our nice data for a minute as I as I said, we know it's the gold standard and we know it's excellent But we also know it's 40 years old and if we were going to design a nice 2.0 We've talked about outcome data and granularity But do any of you have any thoughts on whether it would help you if we had more than a hundred hospitals? involved in this in terms of the ways in which you've used the information I Know that there is the the strata of children's hospitals within it and the expansion of that would be very welcome I think it would be really nice to have more Child specific data Although I know that the majority of children who go to eds go to non children's hospital eds are cared for and non children's hospital eds I do think that there's a certain Select group of Metropolitan areas that that's it's a robust source of data and it may be there may be some special value to that I Think this is an epidemiologic and a statistical Question, right? So if you have a nationally representative cohort, I think that's your most cost-effective means of obtaining the data and I As long as we're obtaining that information from children's hospitals and the other places that we've mentioned I think that's sufficient I'm just about out of time. So I'm just going to take one minute for mr. Mr. Myers your face data That I you probably don't have time to organize this but this is very interesting to me because we do have Problems as somebody pointed out this morning with our death data And I know that you have that data and I will follow up with you after this since my time is just about expired But I would like to hear about that. Thank you so much and thanks again to all of you for coming today Thank you for commission. Thank you commissioner Robinson. I do want to reiterate Fortunate we are to have actually two commissioners who have shown great leadership in this area Commissioner Birkel I'm about to turn to is the only one of us that I'm aware of who's actually been on the other end and it was a nurse Working in a hospital and brings incredible Real-life experience to this issue and so with that commissioner Birkel. Thank you, mr. Chair And again, thank you to all of you for being here We really appreciate appreciate your expertise and your insights into really what I consider one of our biggest priorities here At CPSC and that's data the collection of data and not only collecting them of being able to use it to Accomplish our mission of safety. So your insights are really greatly appreciated I Mean there's a couple of reasons why data has become important to me first enforcement foremost because as the chairman said earlier We are data driven and we need to have Relevant data we need to have accurate data. We need to have data that is Really on the front of what's going on and how we can identify safety issues. So That that's really kind of the first thing and I think In terms of the credibility of this agency making sure that data is robust and accurate is very important beyond that And another reason why data became interesting and important to me is a program We had called retailer reporting the retailer reporting was where the retailers would share a few of them would share information with us some of the agency and Being able to manage that just like being able to manage any other data source is very important And so it became an issue With regards to how we're going to handle that data coming in it was a lot of data And so I would be interested to hear your insights and dr. Quinlan I'll start with you because in particular you're talking about the use of many sources of data That you refer to you with along with nice and other sources. So When this data comes in do these systems talk to each other or how do you? Take the information from this data set and apply it to this data set because one of our issues would be taking all of our Sources of data and looking at it have some way to look at all of it and see what matches up And what doesn't to identify a number of things, but I'd be interested to your do your systems talk to each other Do they how do you share data? I? Think there are a few instances where there have been data linkage Efforts and those are seem very specific birth and death data being linked Based on a number of variables matching up perfectly and then you assume it's the same record same assume as the same individual Those are those are very particular. I don't know any general sort of suggestion their recommendation that comes out of that I don't know if I have any other Great insights into it because there's so many different data systems, and they're all different slices You know there's mortality and morbidity and mixtures of those two It's it's a such a variety Does anyone have anyone else have anything to share on that? to Myers I Know that at NIOSH we worked with a couple different agencies to link data sets looking at you know basically worker fatalities And it's not as easy as it appears even when you have Information that supposedly From similar sources trying to get it to match perfectly does not always work because of a lot of reasons Even something as simple as a typo can give you a you know two records make them look as if they're not a match so Again, this is something that Hopefully and you know Something like the electronic health record where some of these medical records are supposed to have common identifiers Linking certain things such as hospital discharge and emergency department data might become easier I do know states that do that and some of them do rely on the rudimentary electronic records they have Um, but I don't think we're there yet. Thank you. Dr. Button. It's did you have anything to add to that? I Was just going to add we attempted to do something like this with vaccine reporting adverse events and with Vaccines adverse events reported in nice and again similar lots of challenges But a key one is that one's a probability sample and so would not of course expect to have Adverse events from a larger system And so I think that's something to keep in mind that you're not going to have you know full matching when you match a probability system from another source Yes, can I add one other thing? the in in reading a variety of literature Studies not just injury studies Mainly from from Europe, it's it's a totally different thing over there in terms of the ability to have population based studies you know in in many countries it appears that they have you know the the ability to know That they've captured every single case of whatever you want to capture Because they have a nationalized data system and I don't it's such a patchwork here in this country. We've got such a Change in our the way health care is paid for in this country right now. I don't know the implications of this but Taking a trip to Sweden might be very interesting. I'm going to ask the chairman right here in public Thank you all very much Thank You Commissioner Birken and I was just in Sweden a few weeks ago if I'd only known I think that We are going to want to follow up with you on that probably because through our work with OECD We're working on trying to create more of a global nice effort And so I think that would be a good conversation to continue to have commissioner mohawk. Thank you, mr. Chairman I'm surprised mr. Chairman that right before commissioner Birkel asked that you didn't remind her her time has expired And she wouldn't be able to get that off, but I similarly appreciate the leadership. That's Consistently demonstrated by my colleague commissioner Robinson and like you mr. Chairman I'd like to offer to yield my time to commissioner Robinson if she has additional lines of discussion She'd like to pursue with the panel. Thank you so much commissioner mohawk. It's very much appreciated And I have a daughter in Sweden, so I'll take you up on that too The face data mr. Myers could we turn back to that for a minute? And I'm just gonna I think it'll be More efficient if I just tell you the what I'm looking at question wise and then you can decide how to organize Your response, but I'm looking at how you organize that fatality data Is it open? How frequently is it updated? How is it accessed and when you identify trends in that data? Do you contact other agencies? Okay Okay, the the face data itself is not a complete Set of fatality data It's a selected case series of fatality investigations and the way the investigations are conducted is we identify a Specific fatality that we have an interest in we go to the work site we Identify as much information as we can from the employer and from witnesses of that fatality and we try and identify What were the circumstances occurring? Prior to the fatality event During the fatality event and then after the fatality event how quickly was medical care given and so forth and we try and piece that all together in a public health approach of not putting a Finger of blame But to actually just look at all the circumstances and said and saying if you had tried this intervention here It would have stopped the chain and would have prevented the fatality. Okay, so What we try and do is collect a specific number of different types of deaths That have a common theme We've done a lot with electrocutions. We are right now doing quite a few with firefighter fatalities that's been a Very large program that we've had since the mid 90s And a lot of this is basically Using that that approach to try and come up with recommendations that we can give back to the employer and to workers to say These small changes can prevent the deaths. Okay. Thank you so much for that And this next one I'd like to address to both dr. Budnitz and mr. Myers as the CDC and NIOSH people here You're the substantive experts on dealing with our nice and other data and we're looking at our data right now in fact we're Transitioning as the chair mentioned in our IT department and I'm curious if you have advice from your agencies as As to how we might deal in turn how you deal internally with your technology experts I don't know if you have See a CIO CTOs CDOs or how you how they interact with each other But if you could make a few comments on what your agency has that would be helpful Mines real quick ours is a bare-bones operation Primarily we rely so than ours Yeah, we rely heavily on CPSC to give us the data in good shape and then we take that and do what additional coding we need to do to get it Into an electronic format that we make it can make available on our public query site but to be honest with you the bulk of the IT and Data structuring we rely heavily on CPSC and they do such a good job of it. There's very little We have to follow up. Okay Expendence Well, I can't say that I'm an IT expert in this area. So I'm happy to End user, right? Yes. Yes. I'm happy to connect you with folks that do do more the data management Duties at CDC because certainly we do manage all types of yeah systems from states from local health authorities and from other sources You do you do have anything to do within CDC of overlaying different data sets? So I can't speak to that certainly there are folks that do do that CDC but I'm happy to put you in touch with us and Do do any of you have any ideas for us or or can you tell us how you are your agencies when you have information? That you need to get out to consumers How you do that? We have a network of state partners in the state health departments We have right now like 23 states that we have Surveillance grants with and that's probably our most power powerful Network for getting information out We also use our regular dissemination such as Twitter and Facebook and and those types of things but One of the bigger things we try to take advantage of is that State-level impact that we can get from the health departments So I think the the communications method that that is used depends on the audience of course CDC does rely heavily on its partners with partnership with the state health departments But also a clinician networks. So if you're trying to reach clinicians, they're definitely are CC has developed a Network of clinicians and then of course with the public using other public-facing Opportunities from from Facebook to Twitter to to other social media. Thank you very much my time though Thank you so much to the chair and commissioner moharovic Thank You commissioner Robinson and thank you again to the gentleman on the panel We're really grateful that you came and spent the time with us and of course for the work that you do There's two examples sitting in front of us at least from a child protectives and three actually Protecting children and Dr. Rudolph you and I spoke about this a few months ago on the telephone about the incredible work on The magnets rule that you all did and how that saved lives and dr. But it's I think you're modest about the contributions that you've made to put prescription overdose and Parents don't realize that when time and all is changing in dosage or flow restrictors are put in There's actually a real-life human being and that's you who's behind that effort and changing Children safety around the country. So thank you again. This concludes the first panel We'll take a two-minute break and we'll resume with panel two. Thank you