Printer Friendly

A database for abnormal morphology.

Supervisors often want to use slides to demonstrate certain diagnostic features to medical technology students and pathology residents. After I had observed their frustration for years as a bench level technologist, it occurred to me that a computerized database listing the contents of slides would ease their way.

Our laboratory, the oldest university-based hospital lab in the state, had had very little experience with computer programming. Since I had worked with dBase III + (Ashton-Tate Corp., Torrance, Calif.) for a couple of years, I decided to explore using that program for our database. In my research I derived many useful ideas from dBase IV books[1,2] and from technical sessions held in Houston by Ashton-Tate in late 1990.

All the features we needed were to be found in dBase IV, version 1.1: a customized data entry screen, a choice of print formats, and simple retrieval and printing. Best of all, we could achieve all this without programming too much on our own. With dBase IV, it would be easy to customize reports and data entry screens. Programming and debugging required no more than three or four hours a week for about six to nine months.

We had not previously computerized the data because of the difficulty involved and the lack of available software to meet our needs. Our database is used only rarely because not all data have been entered yet. Most of the people using the database have little experience with personal computing and dBase. A manual is available, but we all know what that means.

* Establishing the database. Our first project was to define what the database would contain. We decided to include pathologist comments, RBC morphology comments, notes, surface markers, source, and diagnosis. Field types (Figure I) are numeric, character, date, and memo, the last of which is unique to dBase IV 1.1. I included a source field because eventually I wanted to include not only peripheral blood smears but also bone marrow and body fluid abnormal smears. This area is still in progress because of the large number of fields and different formats in counts for bone marrows and body fluids.

After designing the database, I created several indexes to expedite retrieval: name, diagnosis, date, and name + date. The user presses F2 to enter the database and F10 to create a new index, then types in the fields to be included. The index for name + date is somewhat more complicated because the date must be converted to a character format with the DTOC (date to character) command.

The next step is to design the format for entry screens. The compact format shown in Figure II includes all necessary information on one screen.

A particularly useful feature of dBase IV is its memo field. Using this field allows the data entry person to type in comments pertaining to each case. The field starts as a small window, but if an entire screen is needed, pressing F9 will zoom to that size. If even this is not enough, the screen will scroll up a line at a time. The PACE UP and PAGE DOWN keys will move the cursor to the top and bottom of the screen for easy maneuvering.

Another feature that can be added dBase IV with the purchase of the Control Center Booster is the user-define function. (This booster is a powerful program from dBase for little cost - about $30 - that provides additional features.) With this function, I made a subdatabase of the diagnosis and source fields. By making another database of the most common diagnoses as listed in SNOMED (systematized nomenclature of medicine), I was able to standardize the entry of this field. The technologist enters CGL - standing for either chronic granulocytic leukemia or chronic myelocytic leukemia, since CML is not available in the database. A side benefit of standardized entry is to limit the possibilities for making spelling errors. Standardizing the entry of the source field helped determine the format to be used for the remainder of the fields.

Another functional feature of the Control Center Booster is in creating help screens for each field of the data entry screen. The help fields allow the data entry person to standardize the information in each field, particularly the memo field. A navigation line at the bottom of the screen tells the user how to access the memo field and how to obtain help. Using the surface marker fields let us include information from the flow cytometry lab. In the FAB field, we enter the French-American-British classification of leukemias and myelodysplastic syndromes.

With the data entry form in place, special report forms can be created. Ours include a patient demographic list, a WBC diagnosis list, an RBC diagnosis list, and a patient report format. These report formats, set up with the Report Create feature of dBase, use a simple layout-type screen (Figure III). The patient demographic list includes patient name, admission number, date of specimen, time of specimen, age, sex, race, and hospital ward.

The WBC and RBC diagnosis lists are similar in appearance. Both reports group test results by diagnosis (at the top of the screen), then list patients' names and relevant data (Figure IV). The patient report format is identical to the data entry screen.

* Application generator. To save technologist time and to avoid the worry of having to teach computerese, I pulled all the above areas together under one application. Using a dBase feature called the application generator, I set up a bar menu, a pop-up menu screen, and a title screen. I made the title screen and all subsequent screens red, white, and blue so that even the color deficient could read them easily.

The title screen displays as the system loads. The next display is a horizontal bar menu listing the main menus, left to right: Records, Find, Print, Utilities, and Exit (Figure V). These menus, the heart of the system, pop up when the arrow key highlights the item in the bar menu.

The Record menu has three options: add new records, browse all records, and edit a record. Add new records accesses the data entry screen and displays it with all its features. Browse all records displays the data in a tabular format so that the user can scan the database quickly. Edit a record displays the data in a patient report format, which can be edited.

The Find menu includes three options: diagnosis, admission number, and name. Each is a basic program using "if" and "then" statements. If a technologist wanted to know the date of a patient's last CBC, the Find menu and Name option would ask the patient's admission number and name, directory style (last name, first name), and whether the date of the procedure was known. The computer would search the database for the information and display it on the screen or show a message stating that the information was not in the database. An additional command, SOUNDEX, assigns an alphanumeric code to the name, searches all names with the same approximate phonetic sound, and displays it on the screen with an option to print out the information. A printout of patient information is available for all Find requests.

Four formats are available for the print option of the main menu, which prints out all data rather than selective data for a single patient. With the Utilities option, the user can engage in such basic computer functions as changing the system date, backing up or copying the application to drive A, and repairing a corrupted index.

Changing the system date is useful because despite the battery backup, the wrong date or time is often displayed. Backing up and copying an application allows easy backup of the data to be made at the end of each data entry or at the end of the day, month, or other time span.

It is handy to be able to repair a corrupt index as well; if the index is accidentally contaminated, data can still be saved. Two copies of the database are kept in a locked floppy file cabinet in case of disk failure. The exit menu will send the user either back to the Control Center of dBase or out of dBase and back to DOS.

Patient confidentiality requires security for the database and all data in it. The Protect command of dBase allows the administrator to set up a system of passwords, permitting various levels of use. This security is implemented at the time one selects to run dBase IV from the main menu. To facilitate use by our technologists, I used identifications that were unique to each individual but used every day in the lab. Two identifications were needed: an identification number and a password. The identification number consists of the employee's computer identification number, while the password is the person's employee number at the university. All employees in the department belong to the group Hemapath.

To enhance security, I turned off certain features of dBase, such as the ability to design reports and format screens. I did not want any technologist to move a field to another area. That change would alter other areas in the database and could cause problems in running the database as a whole.

I am still working on some problems in running the database. I suggest that the person naming fields for the database, database file, and indexes make sure no field name is duplicated in any of these areas. To those who need a very simple database without customized screens and reports, I would recommend dBase III + because it is easier to learn and use than dBase IV. While the more advanced version can be intimidating at first, its outstanding features encourage one's creative energies to flow.

[1.] Simpson, A. "Understanding dBase IV." Alameda, Calif., Sybex, 1989. [2.] Prague, C., and Hammitt, J.E. "dBase IV 1.1 Programming." Blue Ridge Summit, Pa., Windcrest (Tab Books), 1990.

The author is a senior medical technologist in the department of hematopathology, University of Texas Medical Branch, Galveston.
COPYRIGHT 1992 Nelson Publishing
No portion of this article can be reproduced without the express written permission from the copyright holder.
Copyright 1992 Gale, Cengage Learning. All rights reserved.

Article Details
Printer friendly Cite/link Email Feedback
Author:Jackson, Valetta Carol
Publication:Medical Laboratory Observer
Date:Apr 1, 1992
Previous Article:AIDS precautions in the '90s: prudence or paranoia?
Next Article:Keeping things simple in a complex world.

Related Articles
How we automated our hematology lab; careful preparation enabled this laboratory section to computerize CBCs, differentials, and RBC morphology...
The clinical lab of the future.
An automated system for record keeping, test reporting, and QC in gyn cytology.
A redesign of RBC morphology reporting.
Streamline your automated hematology laboratory.
Streamline your automated hematology laboratory.
Giant platelets.
A public Internal Luminosity Function database for Gamma-Ray Bursts.
Answering your questions; poikilocytosis terminology.

Terms of use | Copyright © 2016 Farlex, Inc. | Feedback | For webmasters