The American Psychiatric Association (APA) has updated its Privacy Policy and Terms of Use, including with new information specifically addressed to individuals in the European Economic Area. As described in the Privacy Policy and Terms of Use, this website utilizes cookies, including for the purpose of offering an optimal online experience and services tailored to your preferences.

Please read the entire Privacy Policy and Terms of Use. By closing this message, browsing this website, continuing the navigation, or otherwise continuing to use the APA's websites, you confirm that you understand and accept the terms of the Privacy Policy and Terms of Use, including the utilization of cookies.

×
ColumnFull Access

Clinical Computing: The Computerized Lab Alert System for Patient Management in Clinical Care

Published Online:https://doi.org/10.1176/ps.50.7.869

In the name of efficiency and quality control, computers have been introduced into many aspects of organized health care. Most modern facilities boast computerized clinical record management (1) and data banks (2). The health care industry is a gold mine for software developers, who offer state-of-the-art programs for everything from basic medical texts (3,4) to diagnostic codes (5) and Clinical Outcome and Resource Monitoring (CORM) (6), now networked and available in English and Dutch, and currently being translated into German (7).

In psychiatry, computer self-help is now available by screen or telephone in interactive voice-response programs. For depression management, computer interviews, diagnostic and assessment programs, and even computer therapies (8) have made their debut and have proven helpful in monitoring treatment response and determining outcomes. BT STEPS is an interactive voice-response program that helps individuals with obsessive-compulsive disorder pinpoint, record, and rate triggers of their rituals and obsessions to prepare for subsequent treatment (9).

However, integrating these disparate elements is very difficult. Standard coding systems and broadly accepted protocols for communication with local computer systems are needed. Web-based HTML programs may provide the solution (10).

Results of laboratory examinations are often provided instantaneously on computer printouts that can be directly recorded in computerized patient records. But how well do physicians use significant laboratory data that have been so expertly and efficiently stored in cyberspace? Do abundant computer printouts actually facilitate improvement of patient care, or do they add to the attending physician's already daunting workload and cluttered desktop?

The Computerized Lab Alert System

The Computerized Lab Alert System (CLAS) for patient management provides a simple but efficient solution for highlighting relevant data that may influence decision making in clinical care (11). CLAS is a single module in a larger computerized clinical records program called CLICKS (Clinical Records), which emphasizes traditional chart formats and routine work habits in the hospital ward (12). CLAS alerts the attending staff to pathological or missing laboratory results on a daily basis.

All conscientious physicians scan laboratory reports and medical records regularly. Many medical errors may be due to physicians' intrinsic limits as information processors in carrying out routine clinical activities, rather than to correctable human deficiencies or flaws in physicians' fund of knowledge (13). However, when routine test results appear to go unnoticed, it is difficult to determine whether physicians are making human errors by overlooking these data, or whether they have examined the results and are accurately discriminating between trivial and important information. Are the physicians burying their mistakes or oblivious of making them?

Seemingly trivial information may ultimately have a significant impact on patient care, and it is important to ensure that physicians do not fail to notice results of tests. The CLAS module deals with the reality that medical errors stem from a variety of sources, and it offers a solution.

Benefits of CLAS software

In 1996 CLAS was implemented in the 34-bed psychogeriatric ward at Geha Psychiatric Hospital, a public teaching hospital in Petah Tiqva, Israel (12). The department is a computerized work environment that uses the CLICKS program. Patient data are fed directly into personal computers at work stations on the ward, which are linked to additional hospitalwide work stations (12). The CLAS module cross-checks laboratory values with treatment or clinical information, identifies contraindications, and initiates alert messages. CLAS ensures that physicians remain aware of the daily progress of all patients on the ward.

The amount of computerized data available to attending physicians may be extensive. Thus highlighting new or pathological developments enhances the efficiency of the attending staff. A physician may disregard laboratory results that do not seem clinically significant or may inadvertently overlook more important findings. CLAS provides a report method for ensuring that laboratory results will not escape the scrutiny of clinicians. All alert messages continue to reappear until they are dealt with by the physician.

The physician has a number of options for dealing with a message. He or she either repeats the laboratory test or addresses the concern raised in the message. Use of CLAS may reveal an unacceptable level of physician unawareness or forgetfulness that may have to be investigated.

Pilot study

A pilot study was conducted on the psychogeriatric ward for 33 days during October and November 1996 to examine the contribution of a computerized alerting system to patient care (11). During the pilot study, CLAS generated a total of 864 messages. More than 7 percent of the messages initiated treatment decisions, and 15 percent led to repeat laboratory tests. Before making daily rounds, physicians were required to review all laboratory reports. At this time they were also alerted by the daily CLAS messages. However, although all physicians confirmed having reviewed all CLAS messages, direct questioning later each day showed that physicians were unaware of 50 percent of the day's CLAS alerts.

The study revealed that all CLAS messages not dealt with were associated with nonurgent chronic physical illnesses. Thus it would be incorrect to conclude that the physicians did not see the messages or that they ignored their contents. They may have read certain messages and decided, based on clinical judgment, not to deal with the contents. The study indicated that a message not acknowledged the first day it appeared was likely to be dealt with the following day. On any given day, 12 percent of the messages were new and 88 percent were rotating messages that had appeared for at least two days.

Experience with CLAS

Our experience shows that the CLAS module contributes significantly to safe patient management. CLAS is suitable for any active clinical unit in which automated daily screening of laboratory results is of major importance. It significantly improves the quality of care for patients in need of frequent laboratory follow-up, such as psychiatric patients who receive clozapine. For example, laboratory reports of a clozapine-treated patient showed that a granulocyte count was reduced to 2,000 per ml. Even though this is a borderline value and is in compliance with Ministry of Health regulations, CLAS automatically printed a danger alert with the granulocyte count next to the administered drug.

Although ministry guidelines did not require cessation of clozapine, and the treating physician did not discontinue clozapine therapy, the psychiatrist who checked the CLAS messages canceled clozapine therapy. Nevertheless, neutrophils continued to drop and finally reached zero. Treatment was initiated, and after four weeks the patient recovered. In this case the computer was more accurate in predicting the risk for agranulocytosis than the attending physician, whose treatment decision may have been influenced by additional considerations.

Cross-checking modules other than CLAS in the CLICKS clinical records program may be used to screen other clinical data such as drug sensitivities or contraindications. Other CLICKS modules monitor the safety of drug administration and automatically recommend cessation of medication administered for periods considered undesirable. Warning signals appear when appropriate, and the word "danger" appears near the name of the drug on the chart of the susceptible patient.

In addition to the agranulocytosis alert for patients treated with clozapine or carbamazepine, the CLAS warns about electroconvulsive treatment combined with carbamazepine, lithium, or benzodiazepines, and about tricyclic medications combined with irreversible monoamine oxidase inhibitors.

Conclusions

Conscientious physicians invest tremendous effort keeping up with professional literature and medical developments. However, the same physicians do not seem as willing to invest energy in implementing automated activities, even though they may benefit patient management and care. More than 20 years ago, McDonald (13) concluded that the individual physician is not perfectible because the amount of data encountered per unit of time is more than a person can process without error. However, McDonald claimed that the system of care could be perfected by computers that have the capacity to augment the physician's capabilities and thereby reduce the error rate. Computerized systems have an advantage over humans in that they never forget, never get tired, and make no mistakes if they are correctly programmed.

Automated alert systems are now available but not yet sufficiently taken advantage of in routine hospital care. The physician needs tools that are user friendly and that can easily integrate separate sources of clinical information under one "canopy" where physicians can gain access to patient data when they need it (10). CLAS is one of many software modules available for clinical patient management. It can be implemented in any active clinical unit providing that the information data system is used concurrently in routine clinical practice. In the long run, this simple technique may save patients' lives.

Dr. Modai is director of Sha'ar Menashe Mental Health Center in Hadera, Israel, where Ms. Kurs is medical library supervisor and research coordinator. Dr. Modai is also associate professor of clinical psychiatry at the Bruce Rappaport Faculty of Medicine in Technion, Haifa. Dr. Sigler is senior psychiatrist in the psychogeriatric department at Geha Psychiatric Hospital in Petah Tiqva. Address correspondence to Dr. Modai at Sha'ar Menashe Mental Health Center, Mobile Post Hefer 38814, Hadera, Israel (e-mail, ). John H. Greist, M.D., is editor of this column.

References

1. Lavelle SM: The intelligent decision mapping patient record (IDMR). Computer Methods and Programs in Biomedicine 45:97-99, 1994Crossref, MedlineGoogle Scholar

2. Metnitz PG, Laback P, Popow C, et al: Computer-assisted data analysis in intensive care: the ICDEV project: development of a scientific database system for intensive care (Intensive Care Data Evaluation Project) International Journal of Clinical Monitoring and Computing 12:147-159, 1995Google Scholar

3. Fauci A (ed): Harrison's 14 CD-ROM version 1.1 Mac&Win: Principles of Internal Medicine. New York, McGraw-Hill, 1998Google Scholar

4. The Merck Manual Illustrated on CD-ROM. Portland, Ore, CMC ReSearch, Inc—Medical CD-ROMs, 1998Google Scholar

5. Lopez-Ibor JJ Jr, Ayuso Mateos JL, Robriguez Gamazo M, et al: PCP-ICD-10: A computer program and network for psychiatric intervention, in Proceedings of the Tenth World Congress of Psychiatry, Symposium on Information Data Systems. Edited by Vistosky H, Mak FL, Lopez-Ibor JJ. New York, World Psychiatric Association, 1996Google Scholar

6. Marks I: CORM: a computerized system for clinical outcome and resource monitoring, in Proceedings of the Tenth World Congress of Psychiatry, ibidGoogle Scholar

7. Marks I: Overcoming obstacles to routine outcome measurement: the nuts and bolts of implementing clinical audit. British Journal of Psychiatry 173:281-286, 1998Crossref, MedlineGoogle Scholar

8. Greist JH: Computer interviews for depression management. Journal of Clinical Psychiatry 59(suppl 16):20-24, 1998MedlineGoogle Scholar

9. Marks I, Baer L, Greist JH, et al: Home self-assessment of obsessive-compulsive disorder. British Journal of Psychiatry 173:406-412, 1998CrossrefGoogle Scholar

10. McDonald CJ, Overhage M, Dexter PR, et al: Canopy computing: using the Web in clinical practice. JAMA 280:1325-1329, 1998Crossref, MedlineGoogle Scholar

11. Modai I, Sigler M: A computerized laboratory alerting system for patient management in a psychogeriatric unit. MD Computing 15:95-99, 1998MedlineGoogle Scholar

12. Modai I, Valevski A: Computerized patient information system in a psychiatric unit: five-year experience. Journal of Medical Systems 17:309-316, 1993Crossref, MedlineGoogle Scholar

13. McDonald CJ: Protocol-based computer reminders: the quality of care and the non-perfectibility of man. New England Journal of Medicine 295:1351-1355, 1976Crossref, MedlineGoogle Scholar