Patient-specific evidence-based care recommendations for diabetes mellitus: development and initial clinic experience with a computerized decision support system

https://doi.org/10.1016/S1386-5056(98)00110-5Get rights and content

Abstract

Background: adherence with evidence-based recommendations for chronic disease management is often suboptimal. Providing patient-specific reminders at the time of clinical encounters has the potential to improve this situation. A necessary prerequisite for providing such reminders, however, is to have an efficient means of acquiring patient information that can be matched to an underlying knowledge base. The decision support system: we have developed a computer-based, self-administered questionnaire for diabetes care. The questionnaire assesses numerous diabetes-related topics. Patients complete the questionnaire using a touchscreen interface, and their responses are then matched to evidence-based guidelines so that patient-specific care suggestions can be provided for both the patients and their health care professionals. The guidelines are derived from a database of abstracts of studies of diabetes care that are screened for scientific merit and clinical relevance, supplemented by recommendations from diabetes organizations. Evaluation: initial evaluation of the system included an assessment of the agreement of responses to the automated questionnaire with responses to similar questions administered during a structured, personal interview. Overall agreement was 92.5% and the majority of disagreements were minor. More recently, patients aged 18–69 years have been completing the automated questionnaire before appointments at a diabetes clinic. The average time required has been 10.9 min and a mean of 3.0 recommendations have been provided per patient. Patient and health care practitioner satisfaction with the questionnaire and the patient-specific feedback have been high. Conclusions: evidence-based patient-specific diabetes care recommendations can be provided using a self-administered computer-based questionnaire.

Introduction

Providing patient-specific reminders at the time of a clinical encounter has been shown to improve compliance with recommendations 1, 2, 3, 4. Benefits have been observed for preventive health care interventions, cancer prevention, vaccinations, and many other aspects of medical care. This is encouraging because poor adherence with interventions of proven effectiveness, a factor that contributes to increased patient morbidity and mortality, has been observed for many conditions. For example, in patients with acute myocardial infarction, a recent study noted that while 81% of all patients received aspirin and 72% received thrombolytic therapy when appropriate, these values fell to 74 and 49%, respectively, among older patients [5]. A similar situation exists with many chronic diseases. While patients with congestive heart failure have been shown to benefit from angiotensin converting enzyme inhibitors, it is estimated that only a third of patients with this condition are prescribed these medications [6]. Diabetes is another example of a chronic condition in which there is evidence of suboptimal compliance with therapeutic interventions that have been shown to be beneficial [7]. The reasons for this include lack of awareness of new treatments by physicians, delayed recognition of complications, and poor patient adherence [8]. The situation is compounded by the fact that many therapeutic or screening interventions are only appropriate for specific subgroups of patients, based on various patient characteristics.

Computer-based reminder systems could help improve the quality of care by linking patient-specific information from an electronic patient record with an underlying set of rules based on high quality medical evidence concerning appropriate laboratory investigations, medications or other efficacious treatments. Indeed, Lobach and Hammond have shown that a computer-based reminder system improved diabetes care in a primary care setting [9]. In their study, compliance with care recommendations increased from 16% in the control group to 32% in the intervention arm.

While this is encouraging, there are a number of reasons why such reminder systems are not more widespread. Not least among these is that computerized decision support systems often depend on a comprehensive electronic medical record that includes information on each patient’s past medical history, current medical problems, current medications, and laboratory testing and results. While such electronic records are likely to become ubiquitous in due course, they are not in widespread use at present [10]. Additionally, even when such systems become more uniformly available, issues relating to keeping them up-to-date with patient-specific medical information and patient behaviours pertinent to chronic disease management will remain [11].

To address some of these issues and concerns, we have developed an automated self-administered questionnaire for adult patients with diabetes that assesses numerous patient-specific characteristics, without the need for linkage to an electronic medical record and then links the patient’s responses to specific recommendations for their care based on current evidence regarding optimal screening and therapy.

This paper reviews the development, preliminary evaluation and initial experience with using the automated questionnaire in a diabetes clinic.

Section snippets

Development

The initial development of the automated diabetes questionnaire has previously been described [12]. This involved creating a questionnaire and subsequently translating the paper questionnaire into an automated, computer version. Topics for the questionnaire were identified using numerous sources. These included a systematic selection of original and review articles in over 100 clinical journals in conjunction with the production of the evidence-based journals, ACP Journal Club and

Initial evaluation

To assess acceptance of the automated questionnaire, patients were initially asked to complete it using a hand-held lap-top device known as Health-Quiz [17]. This device had a small display screen and three buttons for inputting responses (yes, no, unsure). Patients found the questionnaire easy to complete and indicated few concerns regarding the content, terminology, or flow of the questionnaire.

Subsequently, a more formal evaluation of patient responses to the questionnaire was undertaken

Clinic experience

More recently, patients have been asked to complete the questionnaire before their physician or nurse appointments in the diabetes clinic at McMaster University. This has permitted us to assess patient satisfaction with the questionnaire and printouts, as well as health care practitioner satisfaction with the feedback forms that list the patient-specific recommendations. Consecutive patients scheduled to see a specific nurse or physician during a morning or afternoon clinic were invited by

Discussion

Computer-based reminder systems can improve the rate of adherence with appropriate interventions. Before such systems can be implemented in many settings, however, rapid, convenient and inexpensive approaches to acquiring patient-specific information are essential. For practices with electronic medical records already in place, linkage with pharmacies and laboratories will permit certain information to be acquired and updated on an on-going basis and thus address some of these requirements.

Acknowledgements

This work was supported by a grant from HEALNet, a Networks of Centres of Excellence Program, funded by the Medical Research Council of Canada and the Social Sciences and Humanities Research Council of Canada.

References (17)

  • M.E Johnston et al.

    Effects of computer-based clinical decision support systems on clinician performance and patient outcome: a critical appraisal of research

    Ann. Intern. Med.

    (1994)
  • S.M Austin et al.

    Effect of physician reminders on preventive care: meta-analysis of randomized clinical trials

    Proc. Ann. Symp. Comput. Appl. Med. Care

    (1994)
  • E.A Balas et al.

    The clinical value of computerized information services, a review of 98 randomized clinical trials

    Arch. Fam. Med.

    (1996)
  • S Shea et al.

    A meta-analysis of 16 randomized controlled trials to evaluate computer-based reminder systems for preventive care in the ambulatory setting

    J. Am. Med. Inform. Assoc.

    (1996)
  • T.J McLaughlin et al.

    Adherence to national guidelines for drug treatment of suspected acute myocardial infarction: evidence for undertreatment in women and the elderly

    Arch. Intern. Med.

    (April 8 1996)
  • R.S Stafford et al.

    National patterns of angiotensin-converting enzyme inhibitor use in congestive heart failure

    Arch. Intern. Med.

    (1997)
  • J.P Weiner et al.

    Variation in office-based quality a claims-based profile of care provided to medicare patients with diabetes

    J. Am. Med. Assoc.

    (1995)
  • R.B. Haynes, K.A. McKibbon, R. Kanani, M.C. Brouwers, T. Oliver, Interventions to assist patients to follow...
There are more references available in the full text version of this article.

Cited by (21)

  • Ventilator-associated pneumonia: Improving outcomes through guideline implementation

    2008, Journal of Critical Care
    Citation Excerpt :

    Computerized decision support systems for diagnostic tests were rarely effective, and few studies show improvements in physician performance [16]. The use of CDSS to implement a guideline for diabetes management demonstrated improvements in both the process and outcomes of patient care while overcoming barriers to guideline use [17]. A recent systematic review [9] assessed whether audit and feedback are effective in improving the performance of health care professionals and health care outcomes.

  • Guidelines in the intensive care unit

    2003, Clinics in Chest Medicine
  • Testing quality improvement interventions

    2012, Clinical Epidemiology: How to Do Clinical Practice Research: Third Edition
View all citing articles on Scopus
View full text