^A"You can't control and measure outcomes if you don't control the input."Dr. Lawrence WeedFirst Annual National Conference & Exposition on Computerization in Behavioral Healthcare Services Feb 23-26, 1994 Orlando, FlaHWB (Hippocrates-Winslow-Babbage) FOUNDATION MEETING - ANNUAL MEETING BOARD DIRECTORS - 2/17/95MINUTES Dr. William Burman, Foundation President, called to order the second Annual Meeting of the Board of Directors of the HWB Foundation on February 17, 1995 ,1:15 PM at :Bayhill Suite 1Peabody H otel Orlando, FloridaAttendance was recorded as follows :Dr. Kenneth Johnson, Director, Professor, Department of Orthopaedics, Vanderbilt University School of MedicineDr. Bruce Browner, Director, Professor and Chairman, Department of Orthopaedics, Un iversity of Connecticut School of MedicineDr. Edward Yang, Director, Associate Professor, Department of Orthopaedics, Mount Sinai School of MedicineDr. Henry Mankin, Director, Professor and Chairman, Department of Orthopaedics, Harvard School of Medic ineDr. Jeffrey Mast, Director, Associate Professor, Department of Orthopaedics, Wayne State University School of MedicineDr. David Karges, Director, Assistant Professor, Department of Orthopaedics, Henry Ford HospitalDr. William Burman, PresidentDr. Dj oldas Kuldjanov, HWB Regional Director of former Soviet Republics, Visiting Professor of Orthopaedic Surgery, Wayne State University School of MedicineDr. Elizabeth Ouellette, Chief of Hand Surgery, University of Miami School of MedicineDr. Corn elius Stamp, Emergency Physician, Union Memorial HospitalMr. Steven Taylor, Director of Academic Computing, Hospital for Special SurgeryMr. Rory Gleadhill, Director of Trauma Products, Smith-Nephew Richards Manufacturing Corp.Business opened with a readin g of the mission statement of the HWB Foundation :"The mission of the HWB foundation is to collect uniform and well specified clinical data in the form of text and graphics from reliable, university-affiliated sources and make that data, in quantities of statistical significance, available in the public domain. There, in an electronic bulletin board database format, collected data may be evaluated and re-evaluated by any party - particularly university affiliated research groups." "The foundation endeav ors to establish a new pattern of research whereby instead of the basic data being available only in small samples within the purview of a select few determined to make a specific case - an open database deriving input from multiple sources is created to permit a larger sample size with equal access from all points of view. Thus, enhanced validation of the reporting is possible."A full financial disclosure was made by Dr. Burman of all Foundation transactions since incorporation September 23, 1994.Financ ial ReportIncome Statement (see attachments for detailed expenditures)11/23/93 through 01/28/95Consultation - 600.00 Contribution +10,440.34 Dues - 329.00 Hardware - 4,775.78 Legal fees - 687. 34 Library - 47.96 Maintenance - 40.00 Office Eqpt - 18.18 Software - 708.30 BBS Line -490.94 ÒÒÒÒÒÒÒÒÒÒ Current Balance 2,742.84 HWB DONATIONSCASH Contributors 11/23/9 3 - 01/28/9511/23/93 Dave Greenblatt 5,000.00 03/31/94 Gwion Davies 2,925.34 10/04/94 Henry Sprague 1,015.00 10/11/94 Bob Day 500.00 10/11/94 Dan Willett 500.00 01/17/95 Rich Ghillani 500.00 Total Inflows 10,440.34DONATED SERVICESMair Burman Voice Line ($100/month), $5000 incorporation feesRick McDowell - IRS & State of NY AccountsKenneth Johnson - arrangements for 2nd annual HWB meeting, Peabody Hotel, Orlando, FLA.DONATED EQUIPMENTDavid Karges - Outbound Portable Notebook Computer ( $1000 value)Kenneth Johnson - Outbound Portable Notebook Computer ($1000 value)Edward Yang - HP Inkjet Printer ($350 value) Proposed Budget 1995AMIA Dues - 350.00 Ha rdware - 5000.00 Legal fees - 500.00 Library - 100.00Maintenance - 100.00Office Eqpt - 100.00Software - 5000.00 BBS Line - 500.00Internet Line - 1000.00TOTAL 12,650.00Report on Foundation Activities & Plans :A review of the Foundation Clinical Data Base was presented by Dr. Burman. HWB 95 DATA INPUTCenter CasesElmhurst Hospital Center 3256Detroit Re ceiving 2207Vanderbilt 1229Norfolk & Norwich 633MIEMSS 168University of Geneva 46TOTAL 75 39Approximately 30-40% of the records submitted have direct input from attending or senior level surgeons and are therefore classified as well specified. Dr. Edward Yang indicated that attending input was approaching 100% at Elmhurst and Dr. Kenneth Johns on from Vanderbilt noted that widespread availability of the data input interface over the Vanderbilt Token Ring network would significantly increase the number of cases receiving high level entry at that site. Dr. Bruce Browner said that records were bei ng created at Hartford Hospital with only junior-level input at this time. Outcomes Analysis:Both Dr. Edward Yang and Dr. Jeffrey Mast expressed concern about the lack of outcome data in the database at this time. Dr. Burman acknowledged the deficiency and cited several reasons for the delay in the development of this portion of the Trauma Register program.1) To date, there has been insufficient access to a data collection interface via networked or integrated systems of computers in most recording hos pitals. The actual deployment of the Trauma Register Program in the clinic, even as an ancillary medical record, has been limited. At Vanderbilt, patient information has been printed out from the TR program and placed in outpatient medical record - but no mechanism for direct input from the clinic with respect to outcome is yet available. At Elmhurst Hospital Center, patient files have been copied onto a portable computer which has served in the clinic as a read-only patient reference database - providin g an alternate source of information when the patient's xrays or charts are either missing or insufficient.In order to gain consistent clinic input the computer interface needs to be as available and easy to use as the dictation or conventional charting s ystem. Rapid access to existing patient files is necessary over a network to increase the utilization and perceived and actual value of that patient information source. At several sites of HWB data collection the hardware for the capture of follow-up info rmation is becoming available but as of yet is not fully in place.2) The appropriate level of outcome analysis has not been decided. Should AMA Permanent Partial Impairment Scores be generated ? Should detailed scoring such as the Hip Society's method of scoring results of Total Hip Replacement Arthroplasty be used? Should SF 36 and Sickness Impact Profile or a host of newer, briefer, unvalidated patient satisfaction profiles such as HSQ-12 be derived. Dr. Bruce Browner indicated that Dr. Marc Swiontkows ki is about to release a new, unvalidated, system of outcome measurements this summer. Clearly, direction is needed as to the minimal acceptable parameters of the outcome analysis before significant amounts of time are expended in programming an efficient mechanism for the real-time capture of such data.3) The question of who should input the outcomes data is not completely decided. Input derived from therapists may be of better quality and credibility than that derived from orthopaedic surgeons. Generall y, a therapist probably engages in a more regular, consistent and reproducible methodology when recording limb motion and function. Because of an overall greater time expenditure with the patient may gain a better sense of that patient's true capability. The distance from the surgical procedure and/or surgical decision making process with its potential conflict of interest and bias, may give rise to greater objectivity and patient candor. The use of the therapist as a seasoned analyst of outcome will also help to better distribute the input load. A design decision must be taken as to the optimal agent in the record of outcome before significant amounts of time are expended in programming an efficient mechanism for the real-time capture of such data.4) As opposed to the collection of well specified starting-point, injury data, the value of outcome data matures with the passage of time. As demonstrated by Ignacio Ponsetti, the longer the interval before the determination of outcome, the more respected the d ata. The ability to capture injury severity and bone and joint destruction of a particular case will not improve in time. The accuracy of a determination of true residual impairment should increase in time. Publishing HWB Data in the Public Domain :"The m ission of the HWB foundation is to collect uniform and well specified clinical data in the form of text and graphics from reliable, university-affiliated sources and make that data, in quantities of statistical significance, available in the public domain ."Dr. Burman presented the following proposal for the meeting the HWB mandate of providing public access to HWB data.
An HWB internet node would be established providing access to Mac, DOS, Windows and Unix operating systems. A file server would be ava ilable to run on-line searches on a read-only (tamper proof) CD ROM data repository. Search results producing data files of 500 K or less could be directly downloaded via the interNet. Larger files would require mailings of floppy or compact discs to interested parties.With respect to the establishment interNet connection, Dr. Henry Mankin recommended that the Foundation consult with daughter Allison (Mankin@isi.edu). James Karton, her husband, a meteorologist may also be a valuable resource with r espect to the organization of public data bases. .Dr. Elizabeth Ouellette indicated that her husband through his connections with meteorological databases would also be a source of information and assistance to the Foundation.A lengthy discussion and cons ideration of protection of patient confidentiality was undertaken. Encryption of all demographic and location data was decided. An off-line key to the encryption should be maintained either at Foundation headquarters or at the contributing institutions - should the validity of the data be called into question. Dr. Henry Mankin recommended that individual institutional Human Subjects clearance be obtained before the publication of any clinical data. Dr. David Karges recommended that other compilation s of clinical data such as those proposed by the Orthopaedic Trauma Association, the American College of Surgeons Committee on Trauma (TRACS), the Food and Drug Administration, the National Cancer Center, the Health Care Financing Administration, etc. b e studied as precedents and models of legal and ethical public dissemination of clinical data.Dr. Henry Mankin provided a quotation from the Talmud. A paraphrase is provided herein. "A physician who charges nothing is considered to have advice that is w orth nothing". A proposal was advanced that there be an access fee to HWB data. Dr. Burman expressed concern about the possibility of a fee schedule becoming a barrier to less well endowed investigators (e.g. from third world countries). Dr. Bruce Browne r cited the Health Care Finance Administration and National Library of Medicine as examples of publicly supported sources of information which charge access fees. It was generally agreed that the mechanism for public dissemination of clinical data should be financially self-sustaining and that a reasonable fee schedule for access should be established.Dr. Henry Mankin recommended that there be an advisory board to screen queries of HWB data. The major concern here is the danger of the citation of the HWB database in the delivery of invalid, erroneous conclusions improperly derived from HWB data. Dr. Elizabeth Ouellette suggested that the Foundation rather than imposing surveillance and screening on all queries of HWB data, provide a disclaimer to the ef fect that all interpretations derived from HWB data do not necessarily represent the view of the HWB Foundation and its Board of Directors. The raw clinical data from which any conclusion is drawn is open to any party for evaluation and re-evaluation at any time. The validity of raw data in the HWB database is equal to and possibly greater than that of any comparable database in that it is data directly provided by senior level physicians and is linked to billing procedures, which have a level scrutiny for insurance fraud. Dr. Burman also expressed concerns about the possibility of a politicized and biased query review board which would serve to undermine the public trust of HWB data. It was generally agreed that an advisory board designated to as sist rather than police the interpretation of HWB data would be very beneficial. Dr. Elizabeth Ouellette recommended a waiting period of 6 months to one year after presentation of clinical data to the Foundation to enable investigators at the contr ibuting institution to have first option in the analysis of that data. Immediate on-line publication of clinical data may not afford an adequate interval for contributors to meet their academic reporting requirements and lead to the possibility of a publishing preemption by another party. This could lead to a serious reluctance or resistance to the submission of data to the HWB Foundation. It was generally agreed that a suitable waiting period should be undertaken prior to the publication of fresh clinical data.Resolved by the HWB Board: There shall be encryption of the data to the degree that patient identification and location will be impossible to determine.Resolved by the HWB Board: There shall be an access fee sufficient to maintain the broa dcast of HWB dataResolved by the HWB Board : There shall be clearance by the Human Subjects Committees of participating hospitals and any other regional authorities concerned with maintaining confidentiality of patient records prior to the broadcas t of any and all HWB data.Resolved by the HWB Board : There shall be a mandatory waiting period prior to the broadcast of any and all HWB data to enable those submitting the data first option with respect to its analysis and reporting.HWB Data - Quali ty Control :The ultimate credibility of HWB Foundation depends upon a relentless vigilance with respect to the insurance of its data quality in terms of reliability and validity.Dr. David Karges asked if less specific data collected with instruments othe r than the Trauma Register (such as the OTA Trauma Registry) would be acceptable for combination with the HWB data pool. Some board members voiced concern about the dilution of the specificity of HWB data if this was permitted.During the first annual mee ting of the HWB Foundation, Dr. Jeffrey Mast raised concerns about interobserver variability and reliability as well as the validity of all classification systems as presented in the Editorial of the American Journal of Bone and Joint Surgery December 199 3. It was decided that in order to ensure consistency, minimum inter-observer error and completeness, computer assisted observation should be required. It was also decided that in accordance with the concluding statement of the Editorial of the America n Journal of Bone and Joint Surgery December 1993 "The Journal is reluctant to accept manuscripts about studies in which the investigator attempted to correlate clinical outcomes with the classification of a patient population, if that classification sch eme has not been shown to be a valid tool by appropriate investigation of intraobserver repeatability and interobserver reliability. Proper caution can prevent us from cluttering our clinical literature with studies that relied on poor tools."new systems of classification, including that of the recently introduced by Orthopedic Trauma Association, will not exclusively satisfy the Foundation's data formatting requirements. In other words, data returned to the Foundation in unsubstantiated formats (such a s those of the Orthopaedic Trauma Association) without any other qualifiers (such as ICD codes, CPT codes, SNOMED codes, natural language, etc) will not be deemed as acceptable for entry into the Foundation Data Base and not sufficient for the purposes of an HWB grant or assistance.Although it has been assumed that computer assisted observation provided by the Trauma Register would produce greater consistency, minimum inter-observer error and completeness of records, there has not yet been any proof of t his hypothesis. Interobserver Reliability Project :Dr. Burman proposed that there be an evaluation of the inter-observer reliability of the Trauma Register program starting with an evaluation of xrays and clinical photographs of ankle fractures. This is a common injury and generally only requires plain xrays for classification. Poor interobserver reliability and intraobserver reproducibility for the Lauge-Hansen and Weber classification of ankle fractures has been reported in the past by :1) Nielsen JO, Dons-Jensen H, Sorensen HT; Lauge-Hansen classification of malleolar fractures. An assessment of reproducibility in 118 cases; Acta Orthop. Scandinavica, 61 385-387, 19902) Thomsen NOB, Overgaard S, Olsen LH, Hansen H, Nielsen, ST; Observer variation in the radiographic classification of ankle fractures, J. Bone and Joint Surg., 73-B(4) 676-678, 1991Each center contributing HWB data would provide 30 sets of xrays for analysis by independent observers both with and without a computer-assisted observation instrument. Results would also be matched against historical controls provided by the aforementioned studies.Graphic Data Acquisition Project :The images required for this test of interobserver reliability would be archived and distributed for interpreta tion on CD ROM. Dr. Kenneth Johnson, citing widespread dissatisfaction with past and present classification systems urged that the HWB Foundation begin acquiring graphical data expressed serious doubts about the longevity of these frames of reference. Th e only reliable archive of information may well be graphical. Mr. Steven Taylor enumerated several major obstacles to the collection of graphical data. At this time, only one hospital in the U.S. (VA Hospital in Washington, D.C.) has a full digital imag e storage and retrieval capability. Image capture, if not systematically undertaken by the institution in the form of digitized xrays, presents a time-consuming, technical and logistic problem. Dr. William Enneking of the University of Florida, Gainesvi lle, has been archiving for over five years greater than 16,000 digital images pertaining to the diagnosis and treatment of musculoskeletal tumors and diseases. This work is supported by a major grant from Bristol-Myers and requires an FTE completely d edicated to image capture. The results of image capture whether from flatbed or slide-scanner are variable and often require enhancement with sophisticated image manipulation software such as Adobe Photoshop.Quality control of this process usually require s clinical input with respect to selection of stored image magnification, cropping and contrast/color balance. Unique identifiers and uniform natural language labels for storage and retrieval need to be assigned. Massive storage such as that provided by a CD ROM or magneto-optical juke-boxes is required and generally, a broad band, high speed fiber optic network is needed for multicentric retrieval (such as would be frequently required in the outpatient clinic when previous films are missing). Unless the re are multiple viewing monitors available, the possibility to juxtapose xrays in order to determine interval changes (e.g. progress towards union or lack thereof) is negated. Eventually simplified, low cost solutions to these problems of routine image s torage and retrieval will be at hand. Toward that end, there was a general consensus that the HWB Foundation should participate in the research of optimum methodologies for quality archiving of selected images. A comparison of plain xrays and image captur e gained by digital camera, slide scanner, Kodak Photo-CD and flatbed scanning should be undertaken as part of the computer-assisted classification of ankle fracture study. Images would be presented to both radiologists and orthopaedists for a relative de termination of diagnostic quality and clinical utility. Dr. Henry Mankin suggested that the various vendors of the image capture equipment be contacted for voluntary participation of their products in this clinical test. The Orthopaedic Research and Educ ation Fund as well as the Orthopaedic Trauma Association should be considered as funding sources for this project. Dr. Henry Mankin indicated that his daughter-in-law, a radiologist at the Brigham Hospital would be an appropriate contact person with resp ect to establishing a computer-savvy radiologist's perspective on this project.HWB BBSThe HWB BBS (914-774-7103) opened Thu, Dec 23, 1993 at 9:27 PM. There have been 367 Log Ons for e-mail and software revisions.The HWB BBS will eventually reside on the interNet and permit direct dial-up access as well.Plans for Fund RaisingDuring this critical time in the restructuring of the U.S. Health Care System, it is resolved that at the HWB Foundation under the initiative and leadership of its Board of Directors , will coordinate an appeal to all parties - the medical and pharmaceutical industrial sector, the whole field of health care professionals, Government and private funding sources, and the general public- to make the maximum tax deductible contribution in order to ensure an abundance of valid outcome data with which to guide public policy and medical practice toward the improvement of patient care. Possible funding sources are listed below. Grants - AHCPRContributions - focusing Insurance and Orthopaedic equipment companiesE-CLAIMS - see AttachmentsHospital data and physician data are sent to an electronic clearing house (ETS) which formats the claim for any of 150 insurance companies. There is same day claim rejection/acceptance. ETS charges 39 cents/ claim. The Foundation receives either 7 cents per claim and/or claims data.The second annual HWB Foundation Meeting was adjourned at 3:00 pm. Feb 17, 1993APPENDIX A : HWB 95 Annual Meeting Notes - Dr. Kenneth Johnson Minutes of the Annual Meeting 199 5 of the HWB Foundation (Hippocrates,Winslow & Babbage)The meeting took place February 17th, 1995, 1:00 p.m. in the Bay Hill 1 Room of the Peabody Hotel in Orlando, FL. The meeting was opened by President William Burman and the meeting was brought to orde r. There were no minutes to read, this being the original meeting. Financial Report was given by Dr. Burman, going over income and donations for the year as well as expenses for the final balance in the neighborhood of $2,700. Potential for fundraising activities in 1995 were discussed, including donations, grant support, as well as fundraising strategies. This meeting might include the ability of the HWB Foundation to acquire clinical and cost data by assisting with electronic claims. This mechanism and the cost as such was discussed during the meeting and felt to be a reasonable goal. Warrant further investigation which will be performed by Dr. Burman and other members of the Board of Directors. Data dissemination was discussed at great length. Significant discussion took place about whether or not the HWB Foundation should place 6,000 to 7,000 clinical cases on public domain as in on-line clinical database via a live node on the Internet. There was considerable concern regarding confidentiali ty at this point. It was determined that all links to the patient and treating physician should be obliterated. This would include the patient's name, hospital number, location of the patient admission, physician and treating residents. There would, ho wever, need to be a link back through the HWB Foundation to those pieces of specific information to allow correlation of the public domain data with a real person, should that problem become necessary. At this point, additionally, considerable concern wa s discussed about the lack of an outcome module on the current software package. This was discussed at length, basically regarding the form of an outcome node. Because of significant concern of the Board of Directors, an outcome node will be developed a s soon as possible. This will be necessary for complete data collection. Additionally, concern was discussed regarding whether or not funds should be paid for acquisition to this public domain data. It was the general and majority feeling of the Board of Directors that there should be some charge for access and copies of the data present on the public domain. Dr. Henry Mankin suggested that a research committee be formed that would be able to allow people who access the data to use it to its fullest i ntent. In other words, a research committee would help to formulate plans and ideas as to how the data might be best used to serve the accessor's purpose. This service could be available for brief consultations at no charge, but for additional charge fo r further in-depth consultation. This service was felt by the Board of Directors to be a significant one, and one which should be researched further. Additionally, Dr. Henry Mankin offered his daughter, who has significant expertise with the Internet, a s well as his son-in-law, as a service to the HWB Foundation. The suggestion by Dr. Ouellette was that the data be placed on hold, either at the HWB Foundation or the local access site for a period of time. This would allow local individuals to assimila te the data and have first access to it for research in potential publications. After a significant period of time (one to two years) this material may be released to the public domain. This would ensure that the local individuals would have first acces s to the data. This would be similar to the National Weather Service's data, which is released for publicÏdomain after a certain period of time. It is of note that Dr. Henry Mankin's son-in-law does work with this weather data collection system on a reg ular basis, and knows the ins and outs of it.There was some concern that a disclaimer may be added to the data to prevent association of the HWB Foundation with misinterpretation of the data or mispublication of the data to avoid the negative impact that this might have. Additionally, the effect of hospital/patient confidentiality should be evaluated through the local human studies committee at each individual institution that submits data to the HWB Foundation. This may be necessary prior to allowing r elease of the data on the Internet. This should be researched locally by each institution that transmits data to the HWB Foundation. This suggestion was made by Dr. Henry MankinAt the request of Kenneth Johnson, it was determined that the purest data available to collect on this database would be original injury xrays. These would allow reclassification or redetermination of the patient's injury at any point in time if it were necessary. This means that if a classification system were to change with in five to ten years, those original data points would be available to allow reclassification of the individual and prevent loss of that data. The ability to store xrays in a digital form is now possible. This can be done by several methods, that is by transmitting slides to a CD ROM, using a digitizing camera could be used to photograph xrays giving a digital signal, or xrays can be scanned using a scanner with a transparency mode directly into a digital fashion. It was determined by the Board of Dire ctors that this was a very serious concern and each of these modalities should be very carefully evaluated. There was concern about which modality would be the highest quality, or more specifically, which modality would be the best quality at the most re asonable technique to collect the information. It was suggested that a multicentered study be developed. This would include the contribution of ten to thirty cases of ankle fracture by each participating institution. These xrays would be assimilated as well as their original fracture classification. These xrays would then be copied into digital form using the three methods previously described. The digital methods would then be transported to the contributing institutions for reclassification of each fracture, as well as quality evaluation of the three methods of digitization. This was felt to be an extremely worthwhile study that would be worthy of national publication. A suggestion was made to obtain the imaging equipment directly from each manuf acturer, as it would be commercially an excellent demonstration of the ability of their product, and therefore they should be willing to donate this equipment. If this is not possible, funds should be obtained through a research grant from perhaps the OR staff or OTA. A large amount of enthusiasm was present within the Board of Directors for this project.The meeting was adjourned at approximately 3:00 p.m. after significant and enthusiastic discussion. It is of note that a quorum of the Board of Direct ors was present. Those in attendance included, William Burman, M.D., Bruce Browner, M.D., Kenneth Johnson, M.D., David Karges, D.O., Henry Mankin, M.D., Jeffrey Mast, M.D., Edward Yang, M.D., Ann Ouellette, M.D., Joldas Koldjanov.There were additionally two representatives from Smith Nephew Richards present and other interested observers.APPENDIX B : HWB 95 Annual Meeting Notes - Dr. David KargesDate: 2/20/95 7:55 PMFrom: DEKTo: SysopSubject: Foundation Dear Bill, 2/20/95 I believe the foundat ion has an interesting road ahead of itself provided we remain focused with the same basic endpoint. By this I mean continuing to collect our data, diagnoses and eventually outcome information, yet I also believe going "On-line" on the Internet is essenti al. As you mentioned, the key human resources are certainly found among the Board of Directors. I believe the "Confidentiality" issue is solvable if Ann Ouellette has not already provided us with a workable answer. Personally, we must succeed for if we le t this opportunity slip from our grasp others will not, and most likely It will be the Federal government. I am aware this is of no news to you but after Bruce Browner's keynote address Sunday the future consequences are becoming all the more clear to me. The HWB Foundation can truly allow the documentation to help Trauma surgeons help themselves in their plight to continue to provide better care, save their jobs, and possibly augment their bargaining power with respect to increasing physician and hospi tal reimbursements. The Meeting at the Peabody was encouraging to me. We discussed the HWB monies and '95 budget, number of cases entered to this date, and a number of key issues. With respect to dissemination and availability of information, I felt tha t the Board members agree the data should be accessible to the public at a cost. Regarding the extent of what data is available to the general user should requires professional consultation from an outside source possibly the fellow in Princeton, NJ (Rein holdt?). I spoke with Marc Swointkowski and the initial draft of their outcome project should be available to the public this summer. Basically, it sounds like the outcome questionnaire is not going to be long, attending physicians will be filling out the forms. I suspect we can scan the data and put it on-line. Initially, we could start with a small project such as an out come study on tibia fractures. To assist us on the confidentiality issue I will speak with Frank Lewis and indirectly will attempt to learn how the ACS deals with this, and as already mentioned I think Ann had some very good ideas. Henry Mankin's relative who works for the Internet may also be of assistance. Next, I believe that the board must perform some specific functions for thi s to be a successful foundation. Ken Johnson has just passed on the Presidency to Al Levine but it appears that Ken will continue to be very active in the OTA which has one million dollars in its account. Bruce Browner is politically astute, and seems to be one individual that might notify the proper people at the federal level as to our noble intentions. Henry Mankin is connected to everybody. Jeff is also well connected both in North America and Europe. Ann Ouellette is a true diamond. It was a good m eeting and I think the emotional support is all there but somehow we need to raise about a million dollars which is very concerning. Talk to you soon. DaveAPPENDIX C : HWB 95 Annual Meeting Notes - Dr. Elizabeth OuelletteFebruary 22, 1995Bill Burman, M.D.604 Stage RoadMonroe, NY 10950-3209Dear Bill,There are several issues which I feel are very important for you to address. The first is ethics of information and how you will "control" or "not control" the data. The second is just the acquisition o f the data which I feel you have in hand. The third is performing the research and analysis of this data. Ethics:1) Input must be anonymous but the central control people must be able to access who put in what data.- The patients privacy must be protec ted- The physician's privacy must be protected- Your privacy must be protected- You must be able to purge the system of bad data and if you find out that a physician has not been honest in their data collection, you need to be able to cleanse the data ban k of poor or inaccurate data.- Legally, who owns this data, define extent of their control and create a contract of participation that delineates the responsibilities of the foundation towards this data and the responsibilities of the person collecting th e data. You definitely need this clear because you can count on law suits in the future. Once again, read the NSF, NOAA contracts that NCAR uses to release data.2) Data acquisition: Philosophically, you need to cleanse that data of all opinions. In o ther words, no classification systems, just descriptive terms. In order to do this initially, you could take classification descriptions for people to choose from menu style but not the classification type itself.As much information as possible which is derived on opinion and judgment.3) Performing the research:The research ideas will come to your automatically. You will have ideas, others will have ideas and they will be good.If you build it, they will come - you don't need to worry about finding the projects or ideas. The problem will be crediting the foundation appropriately for it's involvement.Several levels of involvement:1) The foundation initiated the idea, carried the research through and published it. No problem here, except that you may n eed to think of a way to give credit or recognition to all those who have donated their data.2) A donator / member of the foundation carrying out their study, again, no problem.3) Someone who has never donated data wanting to perform a study. Mucho pro blemo because they weren't as emotionally attached as donators to the data and therefore, don't have the sense of protection that they should. You need to build in protection mechanisms.Sincerely,E. Anne Ouellette, M.D.Associate ProfessorChief, Division of Hand Surgery(305) 585-7152 clinic(305) 324-7658 faxAPPENDIX D : E-Mail relating to the HWB World Wide Web****************From: jw milton/ESC Message-Id: <199501161518.AA02229@ayrton.eideti.com>Subject: data on serverTo: 70403.140 0@compuserve.comDate: Mon, 16 Jan 1995 10:18:56 -0500 (EST)Cc: rmhoerin@midway.uchicago.eduX-Mailer: ELM [version 2.4 PL20]Mime-Version: 1.0Content-Type: text/plain; charset=US-ASCIIContent-Transfer-Encoding: 7bitContent-Length: 1963 As far as connec ting your medical data to the internet, here is what i can say about it.We believe it is possible to have a public database that can be accessed by many accounts simutaneously using one of two methods.The first is to simple create a directory on an Inte rnet host machine that can be mounted remotely using TCP/IP/NFS. What this produces is a folder (on a mac) or a subdirectory (on DOS) that appears and acts like a local directory, but is actually remote. Database files can be stored there, and the da tabase applications can read these files as if the file was local.The second method is to install a SQL database server on an Internet host machine that is compatible with whatever database program you are using.There are advantages and disadvantages to both that we are stillinvestigating.Right now we are exploring the NFS option. We have installed OS/2 with NFS to begin our tests with. We will then test NFS on the mac and other platforms.If an acceptable solution if found, the next step is to get the information onto a server that can handle the capacity both in size and access. Whatever site you place you data on would need a T1 data line, to support highspeed access across a number of users. You would also need a dedicated drive (that is mirrored) .Costs to set up a beta site would run about $5000 for hardware & software. Monthly costs would depend on the service you are using and the traffic. At eideti.com, we would probably charge by the traffic on a per login/per thruput basis.Hopefully, our f irst NFS test will be later this month.I do not have any more details to report as there are too many questions we have yet to answer, but feel free to ask me any specifics.I hope this letter is helpful,Jeff Milton*************************It certainly see ms reasonable. The problem with the NFS solution is thateveryone is going to need Internet access. A front end to a SQL based systemcould be easily developed in FirstClass.Maury Markowitz CompuServe 70511,2065SoftArc Inc.***************************From : jw milton/ESC Message-Id: <199501181443.AA23032@ayrton.eideti.com>Subject: Re: data on serverTo: 70403.1400@compuserve.com (William Burman)Date: Wed, 18 Jan 1995 09:43:34 -0500 (EST)In-Reply-To: <950118031205_70403.1400_CHV58-1@Com puServe.COM> from "William Burman" at Jan 17, 95 10:12:06 pmX-Mailer: ELM [version 2.4 PL20]Mime-Version: 1.0Content-Type: text/plain; charset=US-ASCIIContent-Transfer-Encoding: 7bitContent-Length: 1341 > It certainly seems reasonable. The problem with the NFS solution is that> everyone is going to need Internet access. A front end to a SQL based system> could be easily developed in FirstClass.That's not really true. The NFS solution will work on any TCP/IP network. TCP is now available on just about every network out there. So if a hospital had one of the HWB CD's, it could be installed on a company wide server.But this is moot anyway as an application can either look to a local drive, an NFS mounted drive, a network drive, etc, depending on h ow the app is set up. If someone does not have NFS or Internet capability, then they would simply mont the drive locally, or however they need to.The key point is that the internet solution is simply providing SPACE. It is not dependant on what is in tha t space. It could be SQL data, or whatever you want. Certainly, the solution is not to go exclusively Internet. It is simply another way to make the data available.An SQL developer, such as First Class, does not need to concern themselves about NFS/Int ernet access as these details are managed by the network drivers.I hope this all makes sense.BTW, When you get a CD ready, I would like one as I have someone working on this project who claims he can get a MOC online via NFS in a matter of days.Thanks, jw *********************Received: by ayrton.eideti.com id AA18915 (5.67b/IDA-1.5 for rmhoerin@midway.uchicago.edu); Tue, 17 Jan 1995 19:15:29 -0500From: jw milton/ESC Message-Id: <199501180015.AA18915@ayrton.eideti.com>Subject: Re: dat a on serverTo: rmhoerin@midway.uchicago.edu (Ronald M. Hoering)Date: Tue, 17 Jan 1995 19:15:28 -0500 (EST)In-Reply-To: from "Ronald M. Hoering" at Jan 17, 95 03:07:11 pmX-Mailer: ELM [version 2.4 PL20]M ime-Version: 1.0Content-Type: text/plain; charset=US-ASCIIContent-Transfer-Encoding: 7bitContent-Length: 633 > How are you hooked up to internet presently?Yes. Eideti.com is an Internet Service Provider that currently sellaccounts to users and prov ides internet services to business.> Also, do you have an idea how much a T1 and accompanying service might cost?It depends on the area. Our local tarifs costs 80 per mile for a t1, but inmetropolital areas, it can cost very little. I have a NYC provide r who iswilling to provide space to eideti.com for this service if you'reinterested.It would be cheaper if you borrowed a service providers access. The costwould then drop significantly, depending on a number of things.jw