|
|
|
Index - Major Sections
Site Map
Product and Services _______________ Index - Same Level Subject
Index - Child Subjects |
Introduction There are a number of factors that affect the Public's Health such as newly emerging and re-emerging diseases (H1N1, SARS, Tuberculosis, see links below), health care reform, new legislation, Universal Patient Health Record, operating both effectively and efficiently, and the evaluation of new technology. Public Health informatics is an area of increasing concern for authorities around the world. The need for comprehensive disease surveillance has never been greater, and there is the continuing requirement to track possible epidemics and pandemics. Health monitoring linked to disease rates and possible causative effects continues to be high on the public health agenda. Many of today's systems "rely relatively little on computers and telecommunications, and has been reporting only on the diseases that are internationally obligatory." By the time the data is recorded, collected, coded, entered into a information system, analyzed....it is too late! The World Health Organization (WHO) recently wrote "In order to properly plan, manage and monitor any public health programme, it is vital that up-to-date, relevant information is available to decision-makers at all levels of the public health system. As every disease problem or health event requires a different response and policy decision, information must be available that reflects a realistic assessment of the situation at local, national and global levels. This must be done with best available data and taking into consideration disease transmission dynamics, demographics, availability of and accessibility to existing health and social services as well as other geographic and environmental features" SurveillanceThe CDC defines surveillance as the following:
In general, surveillance systems today presupposes certain events or conditions. In general, Surveillances that are undertaken today by many governmental and non-governmental organizations presupposes certain events or conditions. Although the primary focus of national surveillances have been on episodic diseases such as HIV, cholera, dengue fever, TB and others, surveillances also can monitor and should monitor all treats to population health. These may include malnutrition, obesity, smoking, substance abuse and others. Surveillance is one of the most important uses of the Clinical Computer Information System. Since all data is forwarded in real time to a Data Warehouse, analysis by Data Mining can be performed and patterns recognized immediately. The key words in the definitions of surveillance relates to "on-going", or "continuous." Without the ability to analyze data continuous and immediately, patterns can not be detected and surveillance cannot be carry out. The traditional methods of existing surveillance systems: data collection, data analysis, data interpretation, and presentation, is too lengthy to effect timely solutions. It is not enough to "address specific program needs." There should be one system in place that gathers data for all programs and every possible problem. Example: One example is the delay of the detection of arsenic poisoning in India and Bangladesh. Although many cases were being reported throughout the area of patients with syndromes that would indicate arsenic poisoning, no method was used whereby these cases were correlated. The primary methods for data collection in many countries such as registries and sentinel surveillance do not give enough detail to be able to detect discernable patterns, especially previously unknown diseases. Since emerging diseases represent themselves in a variety of difference forms, as much data as possible needs to be collected. A few "key" indicators are not enough. In order to analysis data properly, trained professionals with sophisticated tools must be employed at the highest level. The ability for lower level staff to be able to use statistical tools is not required and is not desired. It is a waste of their time to be trained in such detail procedures and it is too expensive to provide them tools. The only means to do this is with very large databases located in Data Warehouses that are fed continuous with detail clinical data.
Surveillance systems are very important for the detection of emerging diseases and “rapid” determination of health care problems that occur in the selected area of concern. However, surveillance systems are so complex and spread-out that monitoring is a costly exercise and requires a high level of technical capacity within government. Therefore, while most countries have surveillance systems, they are often ineffective at providing the data in a timely manner. Also surveillance systems are often “invisible” and thus funds may be cut for their operation. Without strong internal controls and process measures, the first sign of a failing surveillance system is frequently a major outbreak—too late for corrective action. As an example, in one country, there are approximately 13 sentinel villages established in very poor areas. The theory is that if the there is a health or nutrition problem it will first occur in these very poor areas. The process is as follows: (1) complete data is collected on the population in that village by the health care worker, (2) the data is picked up once each month by a health department personnel, (3) the data is taken to the surveillance department of the national government. The date is read, coded, and keyed into a database, (4) The data is analyzed…and maybe two months later they have the results….but by then everyone in the village may be dead! Surveillance Systems need to have a system of reporting that is instantaneous. In order for health programs to create efficiency and effective levels of health, large volumes of data must be availability not only on health conditions but also social conditions and economical conditions. The complexity of data makes it impossible to use systems that are in place today. A new solution must be offered. Although it has been stated that the quality of data can only be improved by improving the accuracy of diagnosis and improving the accuracy of the codification of causes of death, this is not enough. It is well and good that we know what a person died from, but it is even better if we had the data to tell us why someone died and what we can do to minimize the risk in the future
WHO list the following on their Global Alert and Response (GAR) network-
Anthrax
and we do not even know about the ones we do not know
about...!
Solution Surveillance Systems need to collect massive amounts of data, in real time, and analyzed immediately thought trend and clustering data mining programs. Sampling does not work. Surveillance Systems should be designed to detect the "abnormal" and this cannot be done in random sampling. Every data case is important...especially if it is the exception. Disease PreventionDiseases that were once though were eliminated are back in the news. Many of these diseases are back because of a failure to track and provide vaccines to those at risk...and then there were those that received multiple vaccinations because the healthcare worker was being paid by the number of persons vaccinated.. Many vaccinations were ineffective because there was not good tracking and management of the Cold Chain Process. Basic Package of Health Services
While health care organizations agree that the population should be provided with universal access to a “basic package of health services”, they cannot agree on what services are needed.
The misunderstanding here is caused by the attempt to define the “basic package” for the “whole population.” Since no one group of individuals or country has the same needs, the most cost effective and equitable way is not to define one basic package but rather to define many “packages” based on local requirements.
Experience has been bad with trying to get poorly trained health care workers (even physicians) to implement more than a small number of interventions with any degree of quality.
"A better way to go is likely to be to keep the intervention package small in any given geographical area, but to vary the package from area to area, and change it over time, so that for each local population the package always focuses on the four or five most important local health needs." (Heaver 1995)
This "local package" approach requires the disaggregated of data based on local epidemiological information. Disease surveillance systems for health services targeting therefore need to become more sophisticated. Local disease occurrences should be compared to other local areas, state, and federal levels. If local area values are out of range for any selected variable, then service level changes can be made to correct those variables.
While the data can be collected automatically at the point of service and requires little training, the analysis of variables will have to be performed at the central level and made available to the local leaders.
Links
|
|
|