P. 1
Public Health Informatics How Infomration Age Technology Can Strengthen Public Health

Public Health Informatics How Infomration Age Technology Can Strengthen Public Health

4.67

|Views: 207|Likes:
Published by xizst999

More info:

Published by: xizst999 on Aug 18, 2008
Copyright:Attribution Non-commercial

Availability:

Read on Scribd mobile: iPhone, iPad and Android.
download as PDF, TXT or read online from Scribd
See more
See less

02/01/2013

pdf

text

original

Annual Reviews www.annualreviews.org/aronline Annu. Public Rev. Health. 16:239-52 1995.

Annu. Rev. Public. Health. 1995.16:239-252. Downloaded from arjournals.annualreviews.org by EAST CAROLINA UNIVERSITY on 03/20/07. For personal use only.

PUBLIC HEALTH INFORMATICS:How Information-Age Technology Can StrengthenPublic Health*
Andrew Friede InformationResources Management Office, Centers for DiseaseControl and Prevention, 1600Clifton Road Atlanta, Georgia30333 NE, Henrik L. Blum Universityof California, 570University Hall, Berkeley,California94720 Mike McDonald 2600TenthStreet, Suite 400, Berkeley,California 94710 KEYWORDS: computers, communications, education, informatics, publichealth, information systems, software

ABSTRACT The combinationof the burgeoning interest in health, health care reform and the advent of the Information Age, represents a challenge and an opportunity for public health. If public health’s effectiveness and profile are to grow, practitioners and researchers will need reliable, timely informati6n with which to makeinformation-driven decisions, better ways to communicate,and improved tools to analyze and present new knowledge. "Public Health Informatics" (PHI) is the science of applying InformationAge technologyto serve the specialized needs of public health. In this paper we define Public Health Informatics, outline specific benefits that mayaccrue from its widespread application, and discuss whyand howan academicdis*The Governmentthe rightto retaina nonexclusive, US has royalty-free licensein andto any copyright covering paper. this 239

Annual Reviews www.annualreviews.org/aronline 240 FRIEDE, BLUM& MCDONALD

cipline of public health informatics should be developed. Finally, we make specific recommendations actions that governmentand academiacan take for to assure that public health professionals have the systems, tools, and training to use PHIto advancethe mission of public health.

INTRODUCTION
Annu. Rev. Public. Health. 1995.16:239-252. Downloaded from arjournals.annualreviews.org by EAST CAROLINA UNIVERSITY on 03/20/07. For personal use only.

The combination of the burgeoning interest in health, combinedwith health care reform and the advent of the Information Age, represent a challenge and an opportunity for public health. If public health’s effectiveness and profile are to grow,practitioners and researchers will needreliable, timely information with whichto makeinformation-driven decisions, better waysto communicate, and improved tools to analyze and present new knowledge. "Public Health Informatics" (PHI) is the science of applying Information-Age technology to serve the specialized needs of public health. Systemsdeveloped with PHIknowledge help a county health office staff use surveillance data can to estimate the number doses of measlesvaccine neededfor an outbreak; make of it easier for a regional planningcommission integrate census, surveillance, and to hospitaliz,ation data to project the occupancy AIDS of hospicebeds for a city, and preparerel~ited fundingrequestsfor the state legislature; assist an epidemiologist in collecting and analyzing data to study malnutrition among homelesschildren and the impactof nutrition programs different regions of the US;and facilitate in a health educator’s use of multimediato communicate, a compellingfashion, in newknowledge professional and lay audiences. to In 1990, Greenes &Shortliffe, in the Journal of the American Medical Association, presented an overviewof MedicalInformatics (10). In this paper we define Public Health Informatics, outline specific benefits that mayaccrue from its widespread application, and discuss whyand howan academic discipline of public health informatics should be developed. Finally, we make specific recommendations actions that governmentand academiacan take for to assure that public health professionals have the systems, tools, and training to use PHI to advancethe mission of public health.

WHATIS PUBLIC HEALTH INFORMATICS?
Public Health Informatics is the application of informationscience and technology to public health practice and research. Specifically, this meansdeveloping innovative waysto use inexpensiveand powerfulcomputers, on-line databases, the capacity for universal connection of people and computers, and multimedia communications support the mission of disease prevention and health promoto tion. Practical PHIwork,ideally guided by PHIprofessionals who trained and are experiencedin both information technologyand public health, involves bringing

Annual Reviews www.annualreviews.org/aronline PUBLIC HEALTHINFORMATICS241 together specialists in both fields to conceptualize new ways of applying information technologyto solve public health problems.The practice of public health informatics goes beyondapplying known "computerscience." Rather, it involves synthesizing knowledge from both disciplines, whichis leading to new waysof thinking about and practicing public health. Software and hardwaredeveloped for laboratory science or business often lack features required for public health. For example,standardstatistical packages cannot easily be used to performstandardization, fit mathematicalmodels to disease patterns, or calculate samplesizes for case-control studies; commonlyavailable data-entry programscannot handle the very long or complex questionnaires common public health; general-purpose graphing programs in makebar and pie charts but not histogramsor county maps. Public health data files often contain many millions of records, and quickly accessing these very large files requires special database designs. Finally, there is an increasing need to search the full contents of massivetext databases such as publications, reports and recommendations, tables of summary data, information about the Healthy People 2000Objectives, etc. Typically, these documents not keyare worded, and their diversity, dynamic nature, and the unavailability of appropriate key words in the MeSH (Medical Subject Headings) system would preclude doing so; one of the challenges for public health informatics is to develop streamlined waysto access complextextual data. CDC other governmentagencies increasingly consider information disand seminationto be central to their mission. Theseactivities could be enhanced by building information systems specifically designed for public health data and various users: Synthesized knowledgecan be easily communicated the to public through specialized voice and video systems (in addition to standard radio and television); summary data can be provided to the manageror public inquiries specialist via on-line information systems; researchers maywish to use more sophisticated systems to manipulate complexdatabases. Developing diverse systems that meetthe full range of public health needsrequires developing a cadre of professionals with training and experience in both public health and information technology. Medicalinformatics has forged an analogous link betweenclinical medicine and information technology, focusing on hospital and clinical research informationsystems (automatingmedical charts, linking laboratory data to clinical data, etc), computerizeddiagnostic systems, biomedical engineering, patient and student education, and medical library automation(10). Clearly there overlap between medical informatics and PHI. However,because public health has its foundations in epidemiologyand prevention, public health informatics must focus on speeding and simplifying the conversion of hypotheses about the distribution and determinantsof diseases in populationsinto usable information, and help to disseminate new knowledgein ways that will support

Annu. Rev. Public. Health. 1995.16:239-252. Downloaded from arjournals.annualreviews.org by EAST CAROLINA UNIVERSITY on 03/20/07. For personal use only.

Annual Reviews www.annualreviews.org/aronline 242 FRIEDE, BLUM& MCDONALD

Annu. Rev. Public. Health. 1995.16:239-252. Downloaded from arjournals.annualreviews.org by EAST CAROLINA UNIVERSITY on 03/20/07. For personal use only.

public health practice. In addition, to the extent that clinicians of the twentyfirst centuryincorporate epidemiology public health principles into clinical and medicine, and cometo see the patient as part of a community, public health informatics will makeimportant contributions to the workof practicing physicians (11). Public health needs to makemore effective use of information technology to support its mission. Modem businesses use information technology to update centralized databases from hundredsof retail stores located aroundthe country several times a day, to predict and report elections as they are taking place, and to instantly makereservations on any airline in the word. Public health could makeuse of similar technology to enhancethe timeliness and efficiency of survey and surveillance systems; rapidly and effectively communicate scientific findings to the public; and offer services (such as public health clinic appointmentsfor childhood immunizations)to underservedpopulations. Public health leaders in academiaand at all levels of governmentwill thus be empoweredto better evaluate prevention strategies, and allocate resources for services and research. This will require applying currently available knowledge, and makingadvances in: (a) data and information systems; (b) communications; and (c) specialized tools.

IMPROVED DATA AND INFORMATION SYSTEMS FOR INFORMATION-DRIVEN DECISIONS Data Systems
Data systems are formal systems for data collection, collation, editing, and distribution. Theybenefit public health practice by standardizing data gathering and processing. Data systems produce data (i.e. raw observations or numbers). To generate reports (tables, graphs, or data subsets), the end-user must write computer programs. An example of a widely used data system is the NationalVital Statistics System,whichstate vital registrars use to provide data to CDC accumulationinto a national data bank. for Data systems could be improvedby makingmore effective use of technology to improve accuracy, timeliness, and confidentiality. Accuracy can be improved by more widespread use of computer-assisted personal interviewing, wherein the data are cross-checked as they are keyed (for example, if the interviewee is pregnant, then age = "73" cannot be entered). CDC written has and makesextensive use of programsthat provide this feature, including Epi Info, SURVEY, the data-entry modules of the AIDSReporting System. and Accuracycan also be enhanced by maintaining the data in databases during all phases of the data cycle, where records can be reviewed and edited upon receipt, and then can be used to update information systems. Timeliness can

Annual Reviews www.annualreviews.org/aronline PUBLIC HEALTH INFORMATICS 243 be enhancedby electronic data transmittal; for example, states nowtransmit notifiable diseases data every Monday CDC publication in that Friday’s to for Morbidity and Mortality Weekly Report. They are later added to CDC WONDER,CDC’s on-line public health information system, where they are available for moredetailed analysis. Preserving confidentiality can be facilitated by makingwider use of computer programs that generate a pseudonymbased on the phonetics of an individual’s namecombined with other information (e.g. birth date). The AIDS Reporting System uses this methodto assure that information that could identify an individual is never transmitted to CDC, it allows a case to be yet tracked over time. Anotherwayto guarantee confidentiality is to transfer the data to information systems where the data are stored in aggregate (summary) form only, which is howseveral surveillance data sets are stored in CDC WONDER. INFORMATION SYSTEMS

Annu. Rev. Public. Health. 1995.16:239-252. Downloaded from arjournals.annualreviews.org by EAST CAROLINA UNIVERSITY on 03/20/07. For personal use only.

Information systems are computer systems that provide routine mechanisms for converting data into information(e.g. summary statistics, tables, graphs). In contrast to data systems, information systems can be accessed without any computerprogramming the user; rather, results are accessed via menusor by simple commands. Information systems benefit public health practice by providing easy and rapid access to information. One exampleof a widely used information systemis CDC’sAIDSReporting System used for AIDSsurveillance data, whichallows state and local health departmentsto input, verify, and prepare reports via menus. The reporting area, which maybe a state or county, can use menus quickly and easily prepare a profile of risk factors to among AIDS cases reported this year as comparedto last year. Because this systemintegrates data entry and reporting, it promotesboth the uniformity of the national database,and facilitates state-level analysis, therebyputting timely information in the hands of local decision-makers. A second example is CDCWONDER, which provides menu-driven access to more than 40 large on-line databases (7). CDC WONDER been used has to performtasks such as updatingreports in a few minutes(14), pulling together data on older Americans from a dozen different surveys and surveillance data sets in an afternoon (a task that wouldnormally require weeksof computer programming); searching the full text of the Morbidityand Mortality Weekly or ¯ Report for all articles during the past 10 years that mentionAIDS Washand ington State, an assignmentthat wouldbe virtually impossible unless the data were in a systemthat offered full-text searching. To take on newpublic health challenges,, such as tracking progress towards meeting the Healthy People 2000 Objectives, and understanding whygoals

Annual Reviews www.annualreviews.org/aronline 244 FRIEDE, BLUM& MCDONALD

are (or are not) being met, public health workersneedto acquire and synthesize many more data about the public’s health than are nowavailable~ information mustbe in formats that are directly usable for developingpreventionstrategies under frameworkssuch as those provided by the "Model Standards" (2). be useful for the Healthy People 2000 (6) and "ModelStandards" processes (1), this information must contain concise, up-to-the-minute, age-, race-, and geography-specific syntheses of key data, recommendations,and supporting materials. Background tables and supporting documents must be available, but to facilitate rapid action, the bottomline needsto be crystal clear. PHIcan be brought to bear on several aspects of data and information system development. Information systems could be improved by increasing the numberof data sets that are included, providingsophisticated facilities to help users identify pertinent data sources, and creating easier-to-use and moreflexible statistical and graphing tools. Information systems vary in their objectives: CDC WONDER contains many data sets, but statistical analysis facilities are limited (7). In contrast, BWTRGR data limited to Massachusettsbirths, but allows the has user to perform regression (19); EPIGRAM provides access to detailed data about a single state (Texas) (9). Both kinds of systems have important roles to play. Agenciesthat have had success in building public health information systems need to share their expertise. An unknown number data-collection and of information systems are currently in various stages of early consideration or planning, including a national system to track the immunization status of children, expandedcancer incidence surveillance, regional traumaregistries (17), and registries of people exposedto substances suspected of being hazardous to health. Making plans to develop similar systems widely known any wouldpromotecollaborations, increase data accuracy, and reduce duplication. Developingstandards for communications protocols, data elements, interfaces (the appearanceof the screen and keystrokes that a user needs to work a computerprogram), etc, will makeboth data and information systems easier to build and use. For example,if all public health systems used the samekind of modems, had compatible age groupings, and had screens that worked the same way, then encoding, transmitting, and accessing data wouldbe greatly speeded, users wouldbe less confused by programsthat do not look and work alike, and user support costs wouldbe reduced. With respect to interfaces, standardization would be promoted by moving towards the adoption of the emerging standard for commercial software (or using commercial products directly, whenpossible), whichuses a graphical user interface with drop down menus, an enabled mouse, etc (as is found in Microsoft Windows, X-Windows, and the Apple Macintosh). There are several systems created by various CDC departments to electronically collect, edit, analyze, and transmit laboratory, hospital-infection, human

Annu. Rev. Public. Health. 1995.16:239-252. Downloaded from arjournals.annualreviews.org by EAST CAROLINA UNIVERSITY on 03/20/07. For personal use only.

Annual Reviews www.annualreviews.org/aronline PUBLIC HEALTHINFORMATICS 245 Table Recommendationsdata and information 1 for systems Time frame Shortterm (1-2 Years) Recommendations Data systems: Implement widespread more electronic collectionand data transmittal; to information link systems Information systems: Increase number data sets; enhance of analytical and reporting featuresto support Healthy People Objectives 2000 process Datasystems: Standardize elements equipment data and Information systems: Standardize interfaces; combine duplicative systems

Annu. Rev. Public. Health. 1995.16:239-252. Downloaded from arjournals.annualreviews.org by EAST CAROLINA UNIVERSITY on 03/20/07. For personal use only.

Long term (3-5 Years)

immunodeficiencyvirus (HIV) and AIDS,and notifiable disease data from local and state health departments to CDC. These systems reduce the burden for reporting agencies and make data available sooner and in standardized form, which makesit easier to prepare timely reports and update information systems (3). However, they do not share common interfaces or communications protocols. France’s computerizednetworkfor infectious disease surveillance has demonstratedthat standardized equipment and data elements can greatly speed disease surveillance and feedback(18). Just as public health laboratorians have developed standard ways to manufacturevaccines, and the Advisory Committee on Immunizations Practices recommendsdoses and schedules, public health professionals are starting to hammer proposedstandards for out technical features and implementationplans for data and information systems (12; Table1).

IMPROVED COMMUNICATIONS FOR DYNAMIC ACCESS TO INFORMATION
Improved communicationsencompasses the technology and practices of enhancing, speeding, and simplifying the flow of information, whether that information is held by a human being or a machine, and whether the information is a number, picture, a sound, or anotherrepresentation of a fact or idea. a PHI should enable the public health community communicatemore effecto tively in three domains: (a) human-to-human; human-to-machine; (b) and machine-to-machine. The application of new human-to-human communicationstechnologies has already benefited public health by improvingwork efficiency. Electronic and voice mail eliminate telephone tag and makeit easy to send the same message or computerfile to manyindividuals simultaneously. For example, The Director of CDC electronic mail to transmit to all CDC uses staff the text of selected speeches and the minutes of executive staff meetings, thereby helping to unify

Annual Reviews www.annualreviews.org/aronline 246 FRIEDE, BLUM& MCDONALD

CDC.Another exampleis CDC’sVoice Information System, which the public uses to access prerecorded information on hundreds of topics, including Chronic Fatigue Syndrome,Lyme disease, vaccination requirements for travelers, and AIDS.Since its inception in 1988, this system has been accessed millions of times; its benefits include giving moreconsistent information on a 24-hour basis, which frees CDC staff to answer more complexqueries from health professionals. Increased benefits will accrue by connectingmorepeople to electronic mail; integrating electronic mail with bulletin boards and other informationservices; and by makingwider use of video teleconferencing (which allows participants at both ends to see each other on a television screen) and satellite-based instruction (wherein the instructor appears on a television screen at remote sites). Video teleconferencing is used betweenCDC other PHSoffices in and Atlanta, Washington, DC, and Cincinnati to reduce travel time and expense for meetingsand classes; this technologyalso allows morepeople to participate, and thus speeds decision-making and reduces miscommunications. At the state level, Alabama’s Department Public Health is currently piloting of satellite televised instruction to local health departments; early results are encouraging. Improvedhuman-to-machine communications have facilitated public health workby makinguser-friendly interfaces more widely available, a result of the availability of microcomputers, which are usually easier to use than mainframes. Increased benefits to public health will accrue during the next 2-4 years as moremainframeinterfaces are supplanted by cooperative processing, wherein a microcomputer mainframework jointly on a single task. Under and this system, the user has access to the powerand massive storage of a mainframe computer(where the data are centrally stored and updated), but can manipulatedata via the user-friendly microcomputer, which can offer a mouse, windows, color selection, and access to higher-resolution printers (8). Finally, because systems that look and workin a consistent fashion are easier to learn and use, human-to-machinecommunicationswill be promoted by standardizing interfaces. The impact of messagesdirected at professional and lay audiences can be strengthened by more sophisticated use of sound, video, and modem graphics. Duringthe next few years, putting large amountsof information in easy-to,use form into the hands of all public health workerswill be facilitated by multimedia workstations .(which integrate sound, video, and computerized databases). Thechallenge to public health leaders will be to channelthis evolving technologyfor public health practice, by training staff to think in PHIterms, providing up-to-date workstations and software, and makingit clear that new technology mayrequire abandonmentof old work patterns (and maybesome software and hardware that has not evolved).

Annu. Rev. Public. Health. 1995.16:239-252. Downloaded from arjournals.annualreviews.org by EAST CAROLINA UNIVERSITY on 03/20/07. For personal use only.

Annual Reviews www.annualreviews.org/aronline
PUBLIC HEALTHINFORMATICS Table 2 Recommendations communications for Timeframe Short term (1-2 Years) Recommendations Make widespread of e-mail, bulletin boards, andvoice informamore use tion systems Develop cooperative processing publichealth, videoteleconferencing, for andinvest in better workstations communications and lines Develop multimedia public health applications for public health Make cooperativeprocessing widespread Develop implement and standardsfor interfaces andequipment 247

Annu. Rev. Public. Health. 1995.16:239-252. Downloaded from arjournals.annualreviews.org by EAST CAROLINA UNIVERSITY on 03/20/07. For personal use only.

Long term (3-5 Years)

Improvements in machine-to-machine lic

communications have benefited

pub-

health by providing increasedreliability andreduced costs for data transmissionand machine-to-machine connections.This is a result of the widespread availability of faster and less expensive modems, fiber-optic communications, andsophisticated electronicnetworks. the future, machine-to-maIn chinepublic health communications benefit fromfaster data transmissions will between collection andsynthesis/analysis data points, andreducing errors, the labor, andlosses associated mailing with disks andtapes. Clearly,the internet, although currentlynot suitable for transmittingsensitiveinformation, have will an importantrole to play. PHIis being used to developnewsurveillance systems, wherein computers receive data aroundthe clock and automatically update the database, send an electronic mail message notifying both sender andrecipient if the data contained errors (or failed to arrive), andreturn edited report to the sender(8). These systemswill speedthe availability accurate data for inclusion in truly timely informationsystemsand reports (Table2). BETTER TOOLS FOR PUBLIC HEALTH PRACTICE Public health informaticshas beenused to developtools that serve public health’s special needsfor population-based informationandcommunications. Examples demonstrate utility of such systems that the include: ¯ Patient Flow Analysis, a statistical analysis andgraphing systembasedon software-driven simulations, usedin familyplanning is clinics to optimize schedules the order of service delivery, therebyreducing and waitingtime andincreasingproductivity; ¯ Epi Info, a microcomputer databaseand statsfics program used by epidemiologists collect, verify, andanalyze to epidemic in the field (rather data

Annual Reviews www.annualreviews.org/aronline 248 FRIEDE, BLUM& MCDONALD

¯ ¯

¯
Annu. Rev. Public. Health. 1995.16:239-252. Downloaded from arjournals.annualreviews.org by EAST CAROLINA UNIVERSITY on 03/20/07. For personal use only.

¯

than sending the data to a central office for analysis), thereby allowing quicker formulation of recommendations control (4); for SUDAAN, statistical analysis system provides programsto analyze data a from complexcluster surveys; SETS, a system to distribute CDC survey data and specialized analysis software on CD-ROM, which facilitates data analysis and saves mainframe resourcesfor its users; Arkansas’ Early Intervention Services System, which links several databases and integrates remote communications,facilitating the delivery of public health services (18); An "electronic" extramural course in epidemiology developed in Canada, which uses computer conferencing for distance-based education, allows students to continue their public health education without needing to move or change work schedules (16).

These examples demonstrate that PHI’s function of applying information technologyto public health practice can facilitate workat Federal, state, and local levels. Proposednew PHIapplications include: better tools for authors of multimedia based training courses; public health clinic management software that emphasizes "walk-ins" and follow-up, and that is integrated with surveillance systems; mappingsoftware that can be used to correlate information about health effects and the location of putative environmental risk factors (e.g. contaminated ground water or sources of radiation); improvedsystems to identify sources of survey data and associated laboratory samples that may be available for further analysis; and quick waysto get expert consultations. It must be emphasized that the vast majority of public health software and hardwaresystems have been, and should continue to be, based on commercially available products. The best commercial products have large numbers users. of This large user base enables those manufacturers to provide customer support and upgrades, and to sponsor training and user groups. Large companiesthat stay in business tend to adhere to (or develop) emergingstandards. Economies of scale tend to control prices, especially as compared custom-built systems, to in whichmost of the actual cost maygo for specialized training or postpurchase modifications. PHIprojects should build on what is available and tested whenever possible. Newsoftware should only be developed when a commercial alternative is not available or is too expensive(especially for very large scale applications whenhundreds or thousands of copies maybe required), or when modifying existing software to suit public health needs is not feasible. Federal, state, and local agencies that have been successful in using information technology should nowextend technical assistance to locales that are just getting started, with a special emphasison promoting local systems development, which can be designed to suit specific local needs. For example,

Annual Reviews www.annualreviews.org/aronline

PUBLICHEALTH INFORMATICS 249 Table Recommendations tools 3 for better Time Frame Short term (1-2Years) Recommendations Develop applications public new for health practice: clinicmanagement software, mapping forenvironmental tools exposures, etc Develop capacity software for development andlocalhealth in state departments Develop implement and interface standards Assure PHI integrated all public that is into health activities

Annu. Rev. Public. Health. 1995.16:239-252. Downloaded from arjournals.annualreviews.org by EAST CAROLINA UNIVERSITY on 03/20/07. For personal use only.

Long term (3-5Years)

Washington State’s Department Healthhas developed systemto track its of a progresstowardsmeetingthe HealthyPeople2000Objectivesthat emphasizes the State’s concerns (8). One problem all this rich activity has beenthe with resultant diversity in interfaces, tools, andunderlying technology. systems If are too particularistic, theymay into disuse(10). Standardizing fall interfaces and underlyingtechnologiesmaypromote development tools that ha~e the of a lasting impact(17; Table3). THE DISCIPLINE OF PUBLIC HEALTH INFORMATICS

Thereare several reasonsto developa new academic discipline in informatics (10). First andforemost,it will help assure that new information technology andpractices are developed serve public health’s specific needs, andthat to technology an understanding how and of best to use it will evolvealong with public health itself, Second,havingPHIin an academic setting will promote basic research. Third, a discipline providesa locus of expertise andsupport for students who needto be trained in subjects not currentlywell represented in public healthfaculties. Finally, it providesa careerpath, thus attracting talented studentsandprofessionalswho help build the field. can Supporting triad of teaching,research,andservice is a powerful to the way encourage growthof a newscience. Thecomponents service have been the of outlined in this paper. Teaching would include developing instructional and degree-granting programs schoolsof public health, and offering seminars in andcoursesin public health agenciesandto academic collaborators.For those working government business, it means in or serving as adjunct faculty, preceptingstudentsin epidemiology, biostatistics, preventive medicine, nursing, informationscience, andengineering.CDC staff who workin this field teach an entire course at the Emory Schoolof Public Healthon using information systems, and one of Epi Info’s developersteaches it to the epidemiology students. CDC several engineeringstudents fromthe Georgia has Institute of

Annual Reviews www.annualreviews.org/aronline 250 FRIEDE, BLUM& MCDONALD

Technologyworkingpart-time on various projects; this experience has provided an exposureto public health that they wouldnot otherwise have had and has stimulated career interests in human services. Research activities should include investigating the applicability of new technology to public health via formal hypothesis generation and testing, and implementing new ideas and technologies. There is a need to insert basic research, rigor, and peer-reviewinto a field that otherwisecould become overly focused on meeting day-to-day needs. For example, CDC collaborating with is the University of Michiganto investigate new communications architectures for databases, and with the Georgia Institute of Technology networking. on Traditional academic activities will be key to building credibility for public health informatics as a discipline, and will also greatly promote external collaborations. The preparation of results for publication and presentations at professional meetings is as important for PHI as it is for other disciplines. Much the medical informatics literature is in paired articles: one for the of medicalresult, another for the informatics significance; this modelis directly applicable in public health. Public health informatics programs in government and academia should develop the capacity to carry out consultations and collaborations on subjects ranging from the choice of appropriate software to the development complex of systems. Historically, an importantengine of public health has been the injection of experts into new environments,such as Federal assignees to state and local health departments, schools of public health, and the World Health Organization; and university scholars comingto public health service agencies for sabbaticals. This results in cross-fertilization betweengovernment, academia, international agencies, and foreign governments. A logical extension of consultation would be the creation of centers of academicand applied excellence in PHI to assure the developmentand sharing of expertise (Table 4). This modelhas been used by the National Library Medicine, National Institutes of Health, which has funded academicmedical centers to develop enterprise-wide information management systems under the Integrated Academic Information Program ManagementSystem (15). This project has contributed to nationwide interest in using medical informatics in medical centers and has helped move many projects out of the medical informatics laboratory into clinical use. Public health is poised to emulatethis model, assumingsufficient funds are madeavailable. Oneway to develop centers of excellence in PHIcould be via promoting partnerships madeup of Federal, state, and local governments; private foundations and nonprofit agencies; and private for-profit corporations. TheseCenters could evaluate and develop PHIstrategies that wouldthen be widely shared. One highly successful examplehas been the Georgia Information Networkfor Public Health Officials (INPHO) project, funded by a $5.2 million grant from the Robert W. Woodruff Foundation.

Annu. Rev. Public. Health. 1995.16:239-252. Downloaded from arjournals.annualreviews.org by EAST CAROLINA UNIVERSITY on 03/20/07. For personal use only.

Annual Reviews www.annualreviews.org/aronline PUBLIC HEALTHINFORMATICS251 Table4 Recommendations disciplineof PublicHealth for the Informatics TimeFrame Recommendations

Annu. Rev. Public. Health. 1995.16:239-252. Downloaded from arjournals.annualreviews.org by EAST CAROLINA UNIVERSITY on 03/20/07. For personal use only.

Shortterm The publichealthcommunity establishacademic should programs centersin and (1-2 Years) excellence PHI in PHI professionals should involve themselves practice,teaching, research; in and andpublish present and their findings Academia, government, businessshoulddevelop and collaborativeand consultative activities Long term Consider establishing departmentsPHI schools public of in of health,and journal a (3-5 Years) Develop formal more programs the exchange scholars,sabbaticals,etc for of

Collaborators include the foundation; academia(EmoryUniversity, Medical College of Georgia, and GeorgiaInstitute of Technology);and government (the Georgia Departmentof Public Health, CDC).This project has developed new strategies for networking,disease surveillance, and disseminatinginformation, and represents a modelthat can be exportedto other states.

CONCLUSION
Gaps in public health information and weaknessesin communications will be only partly ameliorated by the development of new technology. Rather, important gains will depend more on bold long-range planning, concerted and well-led efforts at standardization, documenting contributions of PHI tothe wards the improvement public health, and garnering and properly aiming of new resources. This, in turn, will require that governmentaland academic public health leaders develop a commitment public health informatics. A to fully developed academic discipline must be the foundation stone of this initiative. To do less will makepublic health outmoded will place it at and risk. Public health informatics is viewedtoday in the samelight as were chronic disease epidemiologyand advancedstatistics 20 years ago: obscure, forbiddingly complex,somethingfor the specialist, impractical, expensive, of doubtful general utility. Just as these other advances had salutary and indeed revolutionary impacts on epidemiology,so too could Public Health Informatics have profoundbenefits on public health practice. ACKNOWLEDGMENTS Wewould like to thank Lisa S Rosenblum,Daniel A Pollock, Andrew Dean, G DonnaF Stroup, HowardWOry, and Jeffrey R Harris, of the Centers for

Annual Reviews www.annualreviews.org/aronline
252 Disease FRIEDE, Control BLUM & MCDONALD and Prevention, thoughtful and Truls Ostbye, of the University drafts. of West-

ern Ontario,

for their

comments on earlier

I
Annu. Rev. Public. Health. 1995.16:239-252. Downloaded from arjournals.annualreviews.org by EAST CAROLINA UNIVERSITY on 03/20/07. For personal use only.

AnyAnnual Reviewchapter, as well as any article cited in an Annual Reviewchapter, maybe purchasedfromthe AnnualReviewsPreprints andReprints service. 1-800-347-8007; 415-259-$017;emaiharpr@class.org

Literature

Cited discipline and institutional priority. J. Am. Med. Assoc. 263:1114-20 Greenlick MR.1992. Educating physicians for population-based clinical practice. J. Am. Med. Assoc. 267:1645-48 Hughes JM, Teutch SM, Stroup NE, Friede AM,Garbe P, et al. 1991. Surveillance Coordination GroupRepbrt: Recommendations for Electronic Systems for Public HealthSurveillance. Atlanta, GA:Cent. Disease Control Kelso S. 1991. Developing a system to trace health objectives and indicators. See Ref. 5, pp. 534-38 Kleinman JC, Kiely JL. 1991. Infant mortality. Stat. Notes 1:1-11. Hyattsville, MD: Centers for Disease Control; USDept of Health, Education, and Welfare PUbl. 92-1237 Lindberg DA, West, RT, Corn M. 1991. IAIMS:an overview from the National Library of Medicine. Bull. Med. Libr. Assoc. 80:244-46 Ostbye T. 1989. An"electronic" extramural course in epidemiology and medical statistics. Int. J. Epidemiol. 18: 275-79 Pollock DA, McClain PW. 1989. Trauma registries: current status and future prospects. J. Am. Med. Assoc. 262: 2280-83 Valleron A-J, Bouvet E, Garnerin P, Menares J, Heard I, et al. 1986. A computer network for the surveillance of communicablediseases: the French experiment. Am. J. Public Health 76: 1289-92 Wartenberg D, Agamennone VJ, Ozonoff D, Berry RJ. 1989. A microcomputer-basedvital records data base with interactive graphic assessment for states and localities. Am. J. Public Health 79:1531-36

1. American Public Health Association. 1991. Healthy Communities 2000: Model Standards. Washington, DC: Am. Public Health Assoc. Work Group. 1991. Assess2. APEXPH ment protocol for excellence in public health. Natl. Assoc. CountyHealth Off. 3. Bean NH, Martin SM, Bradford H. 1992. PHLIS:an electronic system for reporting public health data fromremotesites. Am. J. Public Health 82:1273-76 4. Dean AG, Dean JA, Burton AH, Dicker RC. 1991. Epi Info: a general purpose microcomputer program for public health informationsystems. Am. J. Prev. Med. 7:178-82 5. Eatmon JM. 1991. Hardware and software needs for coordination of early intervention services (public Law99457). Proc. Public Health Conf. Records Stat., pp. 514-19, July 15-17. Washington, DC 6. Friede A, FreedmanMA,Paul JE, Rizzo NP, Pawate VI, Turczyn KM. 1994. DATA2000: computer system to link a HP2000objectives, data sources, and contacts. Am. J. Prey. Med.In press 7. Friede A, Reid JA, Ory HW. 1993. CDCWONDER: comprehensive ona line public health information system of the Centers for Disease Control and Prevention. Am. J. Public Health 83: 1289-94 8. Friede A, Rosen DR, Reid JA. 1994. CDCWONDF_/PC cooperative processing for public health informatics. J. Am. Med. Inform. Assoc. 1:303-12 9. Goldman DA. 1994. The EPIGRAM computer for analyzing mortality and population data sets. Public HealthRep. 109:118-24 10. Greenes RA, Shortliffe EH. 1990. Medical informatics, an emerging academic

11. 12.

13. 14.

15.

16.

17.

18.

19.

Annu. Rev. Public. Health. 1995.16:239-252. Downloaded from arjournals.annualreviews.org by EAST CAROLINA UNIVERSITY on 03/20/07. For personal use only.

Annu. Rev. Public. Health. 1995.16:239-252. Downloaded from arjournals.annualreviews.org by EAST CAROLINA UNIVERSITY on 03/20/07. For personal use only.

You're Reading a Free Preview

Download
scribd
/*********** DO NOT ALTER ANYTHING BELOW THIS LINE ! ************/ var s_code=s.t();if(s_code)document.write(s_code)//-->