The Databases in Clinical Practice:
Advantages and Disadvantages
1. Introduction
The field of medicine is an area of dynamic growth. As the population increases there are greater demands on the delivery of effective healthcare. Individual practitioners cannot keep up with the growing body of clinical information which is reported around the world. There is a growing need to create an effective way to keep all clinicians current in their medical practices. Although the practice of medicine is typically a one-on-one relationship between patient and doctor, the computer has become a useful element in the organization of patient records and the dissemination of clinical results. Every modern hospital and doctor’s office now has a computer to organize patient records. Medical journals usually distributed in print are now transformed into electronic communications with subscribers. The transformation to an electronic world has facilitated an understanding that information can be made available to everyone instantaneously. The use of databases is on the forefront of current medical practice. Databases are beginning to facilitate the dissemination of clinical data and future research and will become increasingly more important in the future.. In the end this will provide better treatments and outcomes for patients and their attending physicians.
A database in the medical field is a computerized collection of clinical and diagnostic information easily accessible to the practitioner, researcher, and patient in a useful understandable format. Formally, a database is the data that comprises all information related to the entity or disease under study. The data within the database, therefore, is the sum total of what has been collected through clinical and experimental experience. The database is the culmination of work of multiple individuals and institutions working together. There is no easy way by which this can be accomplished. The form with which the data is arranged is examined in two situations and is the subject of this paper. The advantages and disadvantages of databases in the clinical practice setting will be explored.
A database is dependent on the accuracy and accessibility of the sources of data. Many disease entities are collected slowly over time in diverse clinical settings. Most clinical information may come from experience over centuries. The formation of a definitive database can help formulate the best treatments and outcomes. It is imperative that decisions be made to include as many parameters as possible to make the database is useful as possible. This paper takes the disease of the uveal melanoma as an example in demonstrating a database with and without a computer.
The uvea is defined as the parts of the eye consisting of the iris, ciliary body, and choroid. The most common intraocular tumor arises from the pigment cells, melanocytes, of these structures. The melanocytes of the uvea are distinct from the epithelial pigment cells in the retina; and do not form melanomas. It is fortuitous that the onset of this disease is readily visualized on direct physical examination.
The first sign of the disease in the majority of patients may be the appearance of a pigmented nodule in any part of the uveal structures found on routine ophthalmological examination. Many patients may be totally unaware that they have a problem until they are seen on routine examination by their doctor. Patients may present with some loss of visual acuity, however, most have no presenting symptoms. Can ophthalmologist and optometrists can instantly recognize the problem and initiate further diagnostic and therapeutic treatments. A well designed database assists the practitioner in diagnosis and facilitates the choice of the latest advances in clinical treatment. In the past such references were available only in print and through countless hours of personal research. The computer has changed how this process now works.
2. Materials and Methods
Practical understanding of diseases is built upon experience through direct observation and treatment. By collecting data through the history and physical examination characteristics can be assigned for correlation in the future; added to this are the ever increasing array of diagnostic tests and procedures. A database is built in exactly the same way as clinical experience. The facts must be separated into categories which create a cogent picture of the disease. Building a database requires a working knowledge of the disease under investigation.
Beginning with the basics understanding is facilitated by dividing the clinical observations into categories. These same categories are essential in setting up a frame of reference, sometimes called research keys, on the computer. They are as follows:
categories (entities) (these are the basic references for lookup)
attributes (these are the specific values looked at by the clinician)
keys (these are specific words which will lead to the desired information)
The beginning of forming a usable search engine is to define specific categories; values define what a category is; and keys are words which will specifically direct the investigator to an area of specific interest quickly.
Database design must begin with an appreciation of the underlying disease entity. As for uveal melanoma, ocular neoplasia becomes category and keywords directing the investigator to the area of interest. Once within the database other parameters further subdivide and direct the investigator. The database for uveal melanoma will contain basic information about the patient in addition to the various clinical manifestations seen. All the data of necessity comes from clinical experience over a considerable length of time. Sometimes the observed results may be biased by the clinician in charge of a particular case. With enough cases this tendency toward bias may be minimized statistically. The following is a tabular list of categories and attributes in such a database:
personal information (age, sex, national origin, race)
history (family history, drugs, allergies, sun exposure)
physical examination (general and regional)
slit lamp examination
ultrasound
cat-scan/MRI
biopsy
The building of a database begins with defining categories and attributes before compiling the data. The literature is scanned for the appropriate published collections of data on a specific subcategory; in this specific illustration uveal melanoma. This review can be augmented by the inclusion of personal inquiries to investigators and clinicians in the field. A review of the various articles leads to understanding of common elements which can bring them all together into one useful document. Not all articles will contain the same parameters. Personal biases must be identified and recognized as potential problems. This presents a conundrum in deciding whether clinical information was either omitted or not relevant. In the end, a unified table of descriptive and diagnostic criteria can be constructed. The database need not omit the biased data. This data can be annotated for consideration on a personal level by each individual practitioner.
The distinct advantage of the computer is the ability to branch out in an investigation with a double-click of the mouse. Tables can be developed listing findings and data. When a certain finding is noted in the table a double-click can give additional information specific to that finding. There are numerous easy-to-use software packages for the design and construction of databases for PCs, Macs, and the Internet. Databases may be stored on local computers or held in the cloud. The advantage to having the database on local computers is to have absolute control over content and use. Holding the database in the cloud adds an element of insurance and allows access to those who have passwords.
The National Cancer Data Base Report on Cutaneous and Noncutaneous Melanoma was a reported database in the journal Cancer in 1998 (Chang, 1998). The database comprised a summary of 84,836 cases diagnosed in the US between 1985 through 1994. The cases comprised cutaneous and noncutaneous melanomas. The percentages of melanomas were divided into various anatomic regions and an unknown primary category. It was specifically noted that ocular melanomas comprised the second largest group. While the largest group comprised 91.2%, ocular melanomas comprised 5.2%. Using statistical paradigms the analysis identified stage, histology, gender, age, and income as independent prognostic factors. It was noted that 85% of ocular melanomas were uveal, 4.8% were conjunctival, and 10.2% occurred at other sites. Generally for those patients with lymph node involvement the prognosis was poor. The tables within the report broke down the individual statistics for review. The call for data in 1994 included 1227 hospitals submitting 689,714 records representing an estimated 57% of all newly diagnosed cancer patients in that year. The attempt to gather and correlate this information remains a tedious and massive effort. The report concludes that those patients with cutaneous melanomas fared better than those with noncutaneous melanomas; and early diagnosis and treatment led to better long-term results.
The report on the database constructed by Ocular Oncology Unit of the University of Santiago Hospital Complex is based on 250 patients with information specific to uveal melanoma (Bande Rodriguez, 2012). This database although limited to this small group of patients provides a broad spectrum of data relating to each individual patient. The attention to detail allows the clinician to research specifics with ease. Unlike the database from the United States which comprise multiple institutions and is open to all clinicians this study and database is primarily used exclusively by the creators of the database according to this report. It is a model however which serves a useful function in bringing attention to the utility of adding multiple categories, active references, and multimedia to facilitate diagnosis and treatment.
The differences between these two studies are dramatic. The earlier study is static and dependent on input from multiple institutions over time. The Santiago Hospital study although small allows for interaction and addition of patients on a real-time basis. With this added ability to update information the most current and the most effective therapies are available to clinicians.
The computerized database immediately demonstrates its prime advantage: there is instant access to information to the user. This access is limited only by the availability of an Internet connection and computer. A database is therefore available in any location. The computer allows input of new data on a continuous basis which when coupled with automated statistical paradigms researches facilitated with updated prognostic and therapeutic modalities.
The prime disadvantage of a computerized database is that familiarity with the basic concepts which comprise a database must be understood. This further necessitates a familiarity with computer systems. Computer access to the date database must be available for it to be useful. In the instance of a private database user fees may be a consideration. These costs however should be reasonable.
3. Results
The database is designed to facilitate investigation by taking main categories and dividing them into subcategories. This process is akin to forming a branching tree, each limb leading to other subcategories and pertinent information needed. The computer facilitates the process by adding a dimension of automation to the process. As information is derived from the database, input can result in continual updating of statistical values and interpretations.
The following are general categories:
diagnosis/genetics
treatment/monitoring
outcome/prognosis
Diagnosis can be divided into subcategories:
physical exam
histopathology
bio-microscopy
ophthalmoscopy
additional tests
Treatment can be further sub categorized into the various modalities used; and monitoring before or after treatment is acceptable. The various outcomes are dependent on the treatment modalities used. Interpreting prognosis on the basis of clinical information in the database may be extremely helpful.
The addition of color photographs and graphic records in the database makes for a better understanding of the clinical picture. Although no cases are identical, there may be enough similarities to allow accurate diagnostic judgment. The database with selective menu bar, drop-down menus, checkboxes, and radio buttons add a dimension of depth and flexibility which no previous database published can offer.
4. Discussion
The interface of a database must be intuitive. The investigator must be able to find the necessary finding without additional instructions or training. For the purposes of illustration a database from the published literature was examined and compared to the database within a private account of the Ocular Oncology Unit of the University of Santiago Hospital Complex (Bande Rodriguez, 2012).
The earlier literature database presents with a static collection of information relating to cutaneous and noncutaneous melanoma (Chang, 1998). There are facts in table form which are clearly useful to the epidemiologist but of little clinical significance to the practitioner. A table lacks the dynamism available from a computer database. However, a tabular database is instantly usable and understandable. No special training or equipment is necessary; and the information is completely portable.
The database of the Ocular Oncology Unit of the University of Santiago Hospital Complex is a well thought out dynamic tool. A review of the journal article presenting the database clearly demonstrates the broad flexibility and versatility of this program. Illustrations show the levels of attributes and values within categories. There is a large menu from which to select. In addition the database has been constructed so as to include references to current literature. The diagnostician does not have to spend additional time chasing down the latest reports and recommendations.
Neither published or computer database forces conclusions of fact on the user. There is always an element of clinical judgment that is left to the interpretation of the physician. This flexibility is further enhanced with the computerized database. The computerized database is not a static entity. It can be constantly monitored in its development. The authors of the computerized database claim the flexibility of their program will enable research which will contribute to a better understanding of uveal melanoma. Using computerized databases improves clinical work and basic experimental medicine with an increased appreciation for both.
5. Conclusions
Purpose: A database has become an important tool in the arsenal of diagnosis and treatment of many diseases. The computer has revolutionized the access, completeness, and usefulness of clinical and experimental medical findings. The ease with which the practitioner can provide the best care possible has risen tremendously.
Method: A database is designed from the collection of numerous published databases in the literature. An example is the database from the Ocular Oncology Unit of the University of Santiago Hospital Complex in which uveal melanoma is the category.
Results: The uveal melanoma database has been reported successful in the clinic with diagnosis treatment and development of therapies in the Ocular Oncology Unit of the University of Santiago Hospital Complex. This is a good example to follow when a database needs to be built to search for multiple clinical parameters and to do statistical research at the same time.
Conclusion: The database in the computer has transformed clinical and research practice by making an efficient way of compiling, accessing, sorting, and storing medical information. A database stresses the necessity of a commonality of instructions to deliver the best possible information to its users. In a dynamic world in which data is continuously being collected there has to be a universal communication to coordinate and correlate everything into a useful form for all clinicians and researchers. Without the computer databases progress will continue however at a slower pace than necessary. It is imperative that the latest technology be utilized to help as many patients as possible. Although the total cost may initially seem enormous in terms of time and money, the end result will result in better treatments, better prognoses, and lower costs. The system created at the Santiago Hospital Complex is a good model to follow. The system should be adopted by other institutions in the interest of good patient care.
In summary the following are advantages of a computerized database:
Instant access to users
Available in any location w
Ongoing update of therapies
Continuous statistics
Research facilitated
Dynamic portability
In summary the following are disadvantages of a computerized database:
Familiarity with databases essential
Necessity of computer access
User fees
The databases in clinical practice are the future of good medical practice. Although there is no substitute for direct patient physician contact, the addition of the computer with databases will facilitate better understanding during and after treatment. Understanding is by far the greatest asset a physician can possess when tackling diseases such as uveal melanoma. The rapid diagnosis and treatment of an aggressive carcinoma translates into lives saved.
Works Cited
Bande Rodriguez, M.F., Santiago Varela, M., Blanco Teijeiro, M.J., Mera Yanez, P., and Pardo Perez, M. “Diseno de una base de datos informatizada para la gestion clinica y basica del melanoma uveal.” Arch Soc Esp Oftalmol. (2012):89(9):278-283. Print.
Chang, A.E., Karnell, L.H. and Menck, H.R. “The National Cancer Data Base report on cutaneous and noncutaneous melanoma.” Cancer (1998):83(8):1664-1678. Print.
