Implementing health information technology to improve the process of health care delivery: a case study

From Clinfowiki
Jump to: navigation, search

Review by Mark Binstock MPH MD

SUMMARY This paper describes how two information technology systems relate at the Marshfield Clinic. The first system is an electronic medical record (EMR). This is an in-house developed EMR for capturing clinical data that was begun in the 1960s. The EMR does include a snapshot or “Dashboard” functions as well as reminder system for overdue services. The second system is a chronic disease management system (CDMS). Marshfield purchased this proprietary system, InformaCare from Pfizer Health Solutions Inc. The purpose of the CDMSs is to manage patients with chronic conditions usually done with a non face to face encounter by a chronic care team. This includes functions for risk stratification, care guidelines and care management protocols. In the case of the Marshfield clinic, this team consists of 11 RN care managers, 2 nurse supervisors, 1 data manager, 1 data analyst and 2 health service coordinators. They have managed 1000 patients with diabetes.

They conducted a provider satisfaction survey on the addition of care management including CDMSs . The response rate was 32% with an N of 46. 67% “strongly agreed or agreed” it was helpful. Additionally providers thought this care management improved the quality of care (50%), patient’s satisfaction (33%), saving provider time (22%) and saving staff time (26%). About 15% of the article text deals with one clinical vignette.

In their conclusion they state “…how the use of 2 distinct health information technology systems can be adapted by large, multispecialty health care system to potentially improve the overall efficiency and quality of care.” They also mention that a limitation is that the EMR and CDMS were not electronically linked

COMMENTARY This article is for the most part a description two health information technology systems at Marshfield Clinic: the electronic medical record (EMR) and chronic disease management system (CDMS). “Results” are limited to a provider satisfaction survey, reported in one table. No data is presented on quality measures, safety, timeliness or efficiencies.

This article deals with a problem now facing many who have implemented an EHR, namely that although an EMR may and should be effective for managing individual patients; EHRs now available do not lend themselves to population care management. Most providers using an HER in the course of time limited office encounters do not have time to provide comprehensive chronic disease management. Hence the advent of care management and disease management which were basically an outgrowth of disease registries like pap tracking, cancer registries, transplant registries. As the number of patients in these tracking systems grew, there became a need for information management tools, databases. The distinguishing feature of CDMSs is that they imbed ARTICLE electronically management guidelines and protocols.

From the description in the article, it does not appear that the Marshfield EMR includes computerized provider order entry (CPOE) or a Drug- Drug Interaction module. Although care management programs and CDMSs were conceived to manage patients with multiple chronic medical problems like diabetes, hypertension, hyperlipidemia, congestive heart failure, especially for patients who had multiple chronic health problems, at Marshfield the care management team is managing only the diabetes. What is striking from the article is how rich their care management staffing is (17 full time equivalents) caring for only 1000 patients with only one chronic disease (diabetes). It hard to see how this program could be considered economically viable especially after software purchase costs are considered. What is even more surprising is with the large clinical management staff devoted to the care management of these relatively few patients that the program did not register a more convincing impact on provider satisfaction.

What is needed is fully integrated disease management modules within an EMR.


Article Review

Follen, M., Castaneda, R., Mikelson, M., Johnson, D., Wilson, A., Higuchi, K. Implementing Health Information Technology to Improve the Process of Health Care Delivery: A Case Study. Disease Management. 2007, 10(4): 208-215.

Introduction & Purpose:

The authors discuss how the implementation of an electronic medical record (EMR) and a chronic disease management system (CDMS) could promote improved delivery of medical care. They discuss their experience at the Marshfield Clinic, a large, private, multi-specialty health care system in Wisconsin and provide a case study demonstrating how utilization of these systems can lead to improved care management.

Background:

The EMR at Marshfield Clinic has been in use for more than 20 years. Providers across a wide spectrum can access a large volume of clinical and administrative data to help coordinate care. Tools such as the “Medication Manager” and “Dashboard” provide prompts and reminders to track and monitor patient health information.

The Marshfield Clinic also utilizes a commercially available CDMS (InformaCare) to manage information as part of its Diabetes and Community Health Access programs. This system is used exclusively by nurse care managers. Care managers access a variety of tools (i.e. clinical decision support, a registry of patient data storing clinical and behavioral information, system alerts, online documents, a medication database, a problem-list) to achieve patient care goals.

The authors state that together both systems have “revolutionized the delivery of health care by providing real-time access to patient data” by facilitating care planning, care coordination, and communication amongst providers.

Results of Provider Satisfaction Survey:

67% “strongly agreed or agreed” with the 2 systems’ positive impact on care management. Specifically, 69.6% “strongly agreed or agreed” the service improved the self-care behaviors of patients; 71.8% “strongly agreed or agreed” the system resulted in timely and appropriate communication; 60.8% “strongly agreed or agreed” the service improved practice efficiency and clinical outcomes; 67.4% “strongly agreed or agreed” the service is helpful in managing patients.

Conclusion:

An EMR and CDMS can be integrated in a large, multi-specialty healthcare system to improve overall efficiency and quality of care by providing useful tools for monitoring clinical and behavioral data, assessing chronic disease symptoms, and promoting collaboration amongst providers.

The authors point out several operational issues with regard to implementation:

1) There was a small cohort of providers who were reluctant to adapt their clinical management styles initially. However, use of persistent communication and education helped to overcome this barrier. 2) At the onset there was lack of awareness among care managers about how to use the 2 systems efficiently. 3) The EMR and CDMS are not linked electronically. 4) An evaluation of the system to accurately measure effectiveness and clinical outcomes of the Marshfield Clinic operation is lacking but is underway; it will include preventative as well as disease specific metrics.

Commentary:

The authors provide a detailed description of how one large institution leveraged 2 separate clinical information systems to promote improved disease management and integration of care. The patient scenario provides a vivid depiction of how these systems are used to coordinate and improve care management of what would otherwise be a labor intensive case.

A key element to this study is the defined number of providers (11 nurse care managers and 2 nurse supervisors) who utilize the integrated system to manage a relatively small cohort of patients (~1000). The nurse care managers are also provided modest IT support. The scalability of this "dual" system model is certainly an issue, particularly from a financial and operational perspective.

The authors remark how the 2 systems operate independently, an obstacle preventing optimal utilization. Their comments also suggest that early and persistent education could ameliorate issues with regard to adoption and efficient utilization. In general it seems that better integration of disparate systems upfront would minimize many of these issues.

Lastly, although the authors provide data with regard to overall provider satisfaction, they do not provide additional "hard" metrics such quality indicators, readmission rates, reductions in redundant test ordering or adverse medication events, etc. They do mention that an evaluation is underway but do not offer details on which preventative and/or disease specific metrics will be utilized.

Anuj K. Dalal, M.D.