Tech Advisor: Computerized Medicine

John Luo, MD

Assistant Clinical Professor, Department of Psychiatry and Biobehavioral Sciences, University of California, Los Angeles

Background

There is an important difference between the computer as a medium for delivering patient care and the computer serving directly in the role of a healthcare provider. Telemedicine, which is the use of videoconferencing equipment to facilitate direct patient care, is an example of the computer mediating patient care, whereas computer-based care is an emerging trend which transforms the computer into the role of the medical assistant.

In 2001, Revere and Dunbar² composed a literature review of computer-generated outpatient behavioral interventions. Only 37 of the thousands of studies and reports were randomized controlled trials. Of these 37 trials, 34 had statistically significant or improved outcomes, such as improved medication compliance or reduced blood glucose levels. With a success rate of almost 92%, it seems surprising that these computer-generated interventions are not more commonplace. It is helpful to understand what types of computerized programs and tools are in existence to learn of their position in medical practice.

Assessment

Physicians have used paper-based screening tools to help with data gathering and diagnosis. With computerized versions of these tools, calculations and recommendations are done quickly. Baer and colleagues³ devised a computer-assisted telephone system using digitized speech to administer two rating scales for obsessive-compulsive disorder. They found >98% correlation for the scale findings using human administration of scales by telephone and paper-and-pencil scales returned by mail.

The Primary Care Evaluation of Mental Disorders (PRIME-MD), a screening tool for diagnosing psychiatric disorders in the primary care setting, was used in a randomized controlled study to screen patients at the University of California, Los Angeles emergency room (ER) who presented with vague or diffuse complaints, atypical behavior at triage, or whose complaints were greater than their physical findings.⁴ In this study, the PRIME-MD identified psychiatric disorders in 42% of the patients who participated, whereas only 5% of the ER physicians documented any psychiatric findings or requested psychiatric consultation. This study suggested that in the ER, the physicians’ primary focus on physical complaints perhaps led to a missed opportunity to address psychiatric co-morbidity.

Porter and colleagues⁵ used a patient kiosk in the ER to capture historical data that helps providers ask the appropriate questions for childhood asthma management. Gathering recent historical data of the asthma course was important to follow appropriate evidence-based guidelines for treatment. The mean time for 94% of the parents to complete the kiosk interview was only 11.8 minutes, and 95% of the patients’ parents felt that using the kiosk was a “good use” of their time and knowledge. Most notably, the authors discovered that the kiosk picked up indicators of chronic disease severity of which the ER physicians were unaware.

In comparison to diagnostic screening, automated computer interviews can be used to determine patient preferences for treatment of deep venous thrombosis.⁶ For this intervention, the computer illustrated the complications of deep venous thrombosis and used both a visual analog scale and standard gamble question to determine patient preferences for intervention decisions. Patients often had difficulty recognizing the potential difference in outcome with various levels of management, and this program assessed patient attitudes with less physician influence. Over 85% of the users found this system to be a very positive experience.

PsychDiagnoser is a software program that evaluates patients for depression, bipolar disorder, alcohol and drug abuse, and dangerousness in an outpatient practice.⁷ Patients fill out the interview questions on the computer, which then predicts the diagnosis. The program has an emphasis on sensitivity and less on specificity; however, in the private practice setting, there was >85% correlation with the diagnostic impression of the board-certified psychiatrist. More significantly, >90% of patients expressed satisfaction with the experience despite the average 70 minutes total time to complete the computer-based interview.

Education

Computer-based tools for patient education are more than just static Internet pages of information. They are more interactive and can customize their approach based on user need. The Patient Education and Activation System project at the University of Wisconsin Medical School developed a system called the Layman Education and Activation Form.⁸ This system⁹ expands upon mere filling in of medical information forms by providing educational activities to help patients understand medical terminology and healthcare-related issues. It is interactive, and via specialized computer scripts it automatically filters out irrelevant portions of history gathering which may be confusing. In addition, based upon the terminology and searches done by the user, it customizes the educational material it presents.

Shegog and colleagues¹⁰ developed a CD-ROM for pediatric self-management education. Watch, Discover, Think, and Act (WDTA) is one of many computer-aided instruction programs developed for children with asthma. This program uses a motivational approach to teach asthma management skills to children. It has a simulation of real-world activities, tutorials to learn specific skills, and a game treatment to motivate learning. The program was developed using social cognitive theory change methods to improve the child’s knowledge, self-efficacy, and attributions such as verbal reinforcement, guided practice with feedback, persuasion, goal setting, incentives, and symbolic modeling. Children using WDTA scored significantly higher on questions about steps of self-regulation, prevention strategies, and treatment strategies.

Student Bodies, a computer-assisted health education program on eating disorders, has been found to be effective in reducing body dissatisfaction and decreasing disordered eating attitudes and behaviors in college women.¹¹ The key components of the program are psychoeducational readings on body image, nutrition, exercise, and eating disorders; a body image journal to track thoughts and feelings; and a newsgroup. Compliance with the program has been a significant issue impacting participant success.¹² The advantage of using an Internet-based education program is that compliance is directly tracked versus relying on patient self-report.

Another advantage of an Internet-based program is that modifications to the program are easily made to incorporate improvements based on compliance data, further research, and feedback from participants. Celio and colleagues¹² found that a more structured intervention with facilitator reminders and homework had 85% compliance with the tasks compared to 53% with self-report.

Treatment

A number of studies demonstrate the effectiveness of computer-administered therapy. In 1990, Selmi and colleagues¹³ compared a six-session, computer-based version of cognitive-behavioral therapy (CBT) with six sessions of therapist-administered CBT and a wait-list control group for patients with depression. The computer program was accompanied by an experimenter to assist with computer operation, and provided homework at the end of each ses sion. The therapist-administered treatment followed an identical treatment model and homework as well. Both treatment groups were significantly improved on the Beck Depression Inventory and the Automatic Thoughts Questionnaire with no difference between the treatment modality. One caveat is that the sample size was small (12 patients each), and the population consisted of young, well-educated Caucasian participants. However, this was the first study to provide evidence of effectiveness.

Wright and colleagues¹⁴ also demonstrated efficacy of computer-assisted cognitive therapy with a more updated version of computerized treatment of depression. Selmi and colleagues¹³ used only a text-based program, whereas Wright and colleagues¹⁴ used multi-media such as interactive video for a variety of self-help skills. In this study, patients were in three groups, which included those who received sessions with a therapist but were immediately followed up with computer sessions, those who received standard cognitive therapy without computer sessions, and a wait-list control group. The standard therapy group had 50-minute sessions, but the computer-assisted group had 25 minutes of cognitive therapy followed by 20–30 minutes of computerized treatment. Improvements in the Hamilton Rating Scale for Depression and the Beck Depression Inventory were found in both computer-assisted and standard cognitive therapy. However, computer-assisted cognitive therapy was found to be associated with greater improvement in dysfunctional attitudes than no therapy, whereas standard cognitive therapy did not.

In 2003, Lenert and colleagues¹⁵ used an Internet-based behavioral program to promote healthy behaviors, namely smoking cessation. This program used a multi-contact model to deliver cognitive-behavioral treatment, using both Web-based and E-mail–based interventions. Eight different modules covered topics including monitoring of moods, replacing the positive reinforcement that smokers received from nicotine with other pleasant activities, and “talking back” to harmful thoughts. Each module included the text of the lesson, exercises illustrating lesson materials, and tools for self-monitoring of behaviors to help participants internalize behavioral changes and coping strategies taught in each lesson. In the pilot study of 49 smokers, 26 respondents in follow-up indicated that 34% had either quit or had a 50% reduction in smoking. However, only an average of two modules of the eight had been completed, and participants only returned to the Web site a median of two times.

Conclusion

These studies highlight that computer-based assessment, education, and treatments are efficacious in disease diagnosis and improving health behaviors. However, despite such evidence, these programs remain more “proof of concept,” and have not become mainstays in health care. For example, PsychDiagnoser appears to be no longer available since the Web site uniform resource locator is defunct. It may be that with increasing reliance on the Internet for access to information, both the public and healthcare professionals are ready to integrate computer-mediated programs in daily life. Improved ability to finding appropriate health information goes a long way toward increasing reliance on the Internet for information.¹⁶ The Massachusetts Institute of Technology Media Lab initiative of the $100 laptop per child¹⁷ and other similar initiatives with both reasonable computing power and inexpensive Internet access also may make computer-aided health care seamless and ubiquitous.

References

1. Luo JS. Electronic prescribing systems with computer decision support. Primary Psychiatry. 2006;13(6):19-21.

2. Revere D, Dunbar PJ. Review of computer-generated outpatient health behavior interventions: clinical encounters “in absentia”. J Am Med Inform Assoc. 2001;8(1):62-79.

3. Baer L, Brown-Beasley MW, Sorce J, Henriques AI. Computer-assisted telephone administration of a structured interview for obsessive-compulsive disorder. Am J Psychiatry. 1993;150(11):1737-1738.

4. Schriger DL, Gibbons, PS, Langone, CA, Lee S, Altshuler LL. Enabling the diagnosis of occult psychiatric illness in the emergency department: a randomized, controlled trial of the computerized, self-administered PRIME-MD diagnostic system. Ann Emerg Med. 2001;37(2):132-140.

5. Porter SC, Cai Z, Gribbons W, Goldmann DA, Kohane IS. The asthma kiosk: a patient-centered technology for collaborative decision support in the emergency department. J Am Med Inform Assoc. 2004;11(6):458-467.

6. Lenert LA, Soetikno RM. Automated computer interviews to elicit utilities: potential applications in the treatment of deep venous thrombosis. J Am Med Inform Assoc. 1997;4(1):49-56.

7. Zetin M, Glenn T. Development of a computerized psychiatric diagnostic interview for use by mental health and primary care clinicians. Cyberpsychol Behav. 1999;2(3):223-229.

8. McRoy SW, Liu-Perez A, Ali SS. Interactive computerized health care education. J Am Med Inform Assoc. 1998;5(4):347-356.

9. LEAF (Layman Education and Activation Form) Student Evaluation. Available at: http://tigger.cs.uwm.edu/~alp/LEAFV1.1/eval.html. Accessed August 3, 2006.

10. Shegog R, Bartholomew LK, Parcel GS, Sockrider MM, Masse L, Abramson SL. Impact of a computer-assisted education program on factors related to asthma self-management behavior. J Am Med Inform Assoc. 2001;8(1):49-61.

11. Winzelberg AJ, Eppstein D, Eldredge K, et al. Effectiveness of an Internet-based program for reducing risk factors for eating disorders. J Consult Clin Psychol. 2000;68(2):346-350.

12. Celio AA, Winzelberg AJ, Dev P, Taylor CB. Improving compliance in on-line, structured self-help programs: evaluation of an eating disorder prevention program. J Psychiatr Pract. 2002;8(1):14-20.

13. Selmi PM, Klein MH, Greist JH, Sorrell SP, Erdman HP. Computer-administered cognitive-behavioral therapy for depression. Am J Psychiatry. 1990;147(1):51-56.

14. Wright JH, Wright AS, Albano AM, et al. Computer-assisted cognitive therapy for depression: maintaining efficacy while reducing therapist time. Am J Psychiatry. 2005;162(6):1158-1164.

15. Lenert L, Munoz RF, Stoddard J, et al. Design and pilot evaluation of an internet smoking cessation program. J Am Med Inform Assoc. 2003;10(1):16-20.

16. Zeng QT, Crowell J, Plovnick RM, Kim E, Ngo L, Dibble E. Assisting consumer health information retrieval with query recommendations. J Am Med Inform Assoc. 2006;13(1):80-90.

17. One Laptop per Child. Available at: http://laptop.org. Accessed August 3, 2006.