catia点云生成巧克力面:谁能帮我找到一篇关于信息管理方面的英文文章

Introduction to Information Management
Information Management

In 2000, the Institute of Medicine in collaboration with the National Academy of Science published a report titled: “To Err is Human: Building a Safer Health System”. The authors estimated that 98,000 people die annually as a direct result of medical errors occurring in hospitals. These deaths are not associated with the day-to-day risk of surgery and medical practice in outpatient clinics. Instead, a significant portion of the mortality is a direct result of poor design and utilization of medical records. Medical Records Kill. To put this number of deaths in perspective, more people die annually because of medical errors then as a consequence of car accidents, breast cancer, or AIDS. Other examples include:

*
One in 25 hospital admissions results in an injured patient.
*
Three percent of adverse effects cause permanent disabling injury; of these one in seven leads to a
patient death.
*
Preventable medical errors account for 12-15 percent of hospital costs.
*
About 23,000 hospital patients die each year from injuries linked to medication use.
*
80% of nurses calculate dosages incorrectly 10 percent of the time, and 40 percent of nurses
make mistakes more than 30 percent of the time.
*
Approximately 180,000 unnecessary deaths and 1.3 million injuries occur from medical treatment in
the United States.

Nearly all of these events are due to errors in data management or interpretation.

Most of this information is contained in a book titled, To Err is Human: building a safer health system. 2000. Edited by Kohn LT, Corrigan JM, Donaldson MS. National Academy Press Washington DC. A recent whitepaper refers extensively to the same ongoing problem.
Building a better system

Information is a broad term and includes all sources of facts and opinions that we use to make decisions (printed, heard, or seen). We are in the midst of a continuous stream of information and are being asked to assimilate and organize these data at a faster and faster rate. This challenge is even more severe to the professional fact finder. As scientists and public, production, and wildlife health practitioners you are active discoverers of facts and ultimately a source of information for the world. In those roles, you must organize resource information and original data so that it is correctly analyzed, and interpreted to be delivered to your eager and demanding consumers.

The health industry (of which we are card carrying members) has grappled with the issue of data integrity for decades and the discipline of Information Management has grown to deal with it . There are professional schools training people to work as specialists in this industry (see our main page). A professional organization, the American Health Information Management Association (AHIMA www.ahima.org ) is devoted to the challenges of collecting information in hospital settings. The United States Animal Health Association and American Association of Veterinary Diagnostic Laboratories have a joint committee devoted to Animal Health Information Systems. The United States Department of Agriculture has The Center for Economics and Animal Health and the National Animal Health Monitoring System completely devoted to the methodologies of collecting, organizing and analyzing animal health data. As this course progresses and as you proceed through the MPVM curriculum, you will take the lessons learned from these groups and their experience and apply them to smaller scale work. Please note that I have focused on monitoring efforts and not at all on research. From an epidemiologic perspective, these efforts are related but they are not identical and there are specifics to research that need to be addressed in a different manner than from monitoring efforts.

The Skills of Information Management

A survey of professionals (both employees and employers) identified 20 skills that were the most important for Information Management specialists. While some of the skills were important technology and computer skills many of them were skills directly related to understanding data. These skills included: defining data elements, understanding how the database or application will meet the needs of the end user, being able to integrate analysis into a database, perform data retrieval, ability to create and use if/then statements, ability to manage data quality, ability to create calculations within a data query, ability to communicate effectively, ability to listen and understand client requests, and an ability to create conceptual models.
Data Quality Management Model

Information or data is a fragile commodity and all steps in the data process need to be carefully mamaged. and the process AHIMA recently published a model for Data Quality Management (www.ahima.org/journal/pb/98.06.html). It is a simple model that will serve as our guide for the class.
Details of the Model

The steps of data quality are APPLICATION, COLLECTION, WAREHOUSING, AND ANALYSIS.

Although we often focus our efforts on the analysis and the subsequent reports it is absolutely, unequivocally essential that significant energy be devoted to obtaining and maintaining good quality data that will be used in analysis. Everyone has heard ‘garbage in – garbage out’. Translated, if the data going into the analysis is bad then the information resulting from the analysis -- no matter how sophisticated and no matter how clever -- will be wrong. This problem is the classic epidemiologic and statistical issue of "Bias".

Application--This defines the reason, approach, and the methods for collecting the data

Collection--This is the process of data collection

Warehousing--How the collected data will be stored and accessed

Analysis--Organizing, summarizing, and reporting the data

The Characteristics of High Quality Data

Accessibility—The data are available and useable.

Consistency—Over the time the data are collected and stored, the data were collected in the same manner. In practice this means that the definitions, formats, and storage media were the same or compatible.

Currency—The data has value to the target audience

Granularity—The dimensions or specificity of the data--crude to fine; an example is the difference between a weight category (crude) and the actual weight (fine)

Precision—The data are collected with the same measuring tool (biased or not) resulting in low variation data.

Accuracy—The data are collected with few biases and are a representation of the true state you are measuring.

Comprehensiveness—The data contain sufficient detail to answer the monitoring or research question.

Definition—The meaning and intent of the data are clearly defined

Relevancy—The data pertain directly to the question.

Timeliness—The data are temporally related to the question.

Application

You will hear the following many times in the course of the MPVM year, but understanding and clearly defining the data application is the key to successfully utilizing the information you collect. Application design is the process of carefully defining: 1) the purpose of the research or monitoring effort and 2) the specific monitoring or research question to be answered. Application design is often an iterative process that includes the following:.

1. Based on an observation or belief, pose a question that can be answered.

Examples include:

Antibiotic resistance in the human population is a direct consequence of the use of antibiotics in animal feeds

Increased levels of estrogen in the environment decrease the fertility of free-roaming ungulates

Removing horses from watersheds can decrease Cryptosporidial infections in people

The primary source of salmonellosis in humans is from meat and poultry

The incidence of catastrophic diseases observed in poultry has increased steadily in the last year

The incidence of debilitating heart conditions in household cats has diminished in the last 5-years

2. Investigate the problem/question by doing a literature and grant search. This will entail library work, web searches,or contacting colleagues or experts in the field.

3. Read the literature and summarize the information and test your question against published information

4. Restate the question or need to more specifically target the data application.

5. Choose an approach to answer the question

6. Define the specific data sets that will be needed to answer the question

While the basic steps of the process are clear cut, there are many ways to achieve the outcome. In many cases, the process is individual and may require little interaction with others. More frequently the process requires that you work within or part of a group. When this occurs, the success of your project lies not only with the quality of the idea but also with your ability to understand group dynamics and the processes of groups.

Carefully dealing with these first steps will define the nature of the other three steps in the data quality model and ensure that the correct data are collected and brought to bear on the question.