A management information system to evaluate black infant health care.

ABSTRACT

Health Agencies are increasingly faced with the need to develop data systems capable of both generating reports for funding and assisting providers with managing patient care (U.S. Government, 1990). Therefore, these health care agencies now require infrastructures and capabilities in information technology that can help them justify funding from state and local agencies and to evaluate patient care.

1. INTRODUCTION

To assist in the development of the technological infrastructure, California State Department of Health Services authorized the development of a Management Information System (MIS) to help 16 Healthcare Jurisdictions sites manage and evaluate client care in their Healthcare Projects. Evaluating the effectiveness of health care services is a high priority of federal agencies, particularly for public funded care; therefore technology supported infrastructure has become an essential evaluation tool (Raskin, 1991).

The objective of this paper is to discuss the challenges in developing a MIS for a multi-user and multi-service system to evaluate Healthcare Models and the development of the MIS infrastructure that facilitated the evaluation. The MIS was designed to meet the needs of health care management, quality assurance, program planning, resource development, and information sharing.

The Management Information System (MIS) developed is currently used by the health care providers in 16 Health Agencies throughout California to assist in evaluating, managing, and standardizing their health care services. Because of the confidential and complex nature of the healthcare components in the health services, this article will not discuss the details of the healthcare model development or the healthcare services or health agencies.

1.1 Health Care Model Background

To better understand the nature, motivation, and incentive of this project, here is a brief history of what lead to the development of the MIS.

Low Black infant birth weight in African American women is the chief risk associated with pregnancy and is highly predictive of the eventual disability, illness, or death of newborns (State of California Health Services, 1995). Consequently, the most common measure of health outcomes of African American infants is birth weight (Korenbrot, 1984).

Therefore, the Maternal and Child Health Branch of the California Department of Health Services authorized an assessment of healthcare interventions to increase Black infant birth weights. Their assessment data revealed that early-sustained healthcare interventions assure better-to-best birth outcomes, and that a coordinated uniform approach across health programs would most likely yield the maximum impact in improving higher Black infant birth weights and the health of their mothers (Prenatal Care Network, 1994). This assessment results lead to the design, development, implementation and evaluation of uniform Best Practice Health Intervention Models (BPHIM) in the areas of:

Case Management                Prevention
Health Behavior Modification   Role of Men
Outreach                       Social Support and Empowerment

These six BPHIM were identified by Gates-Williams (1992) as the adversity factors that had an impact on the coping responses of high-risk African American pregnant women. The coping styles are especially important because no matter how much health service is provided, high risk African American women cannot be protected from all of the adversity in their environments (Gates-Williams, Jackson, et al 1992). Consequently, the Best Practice Health Intervention Models provide care plans to:

* protect high risk African American women from the adversity,

* remove them from the adversity, or

* modify or extinguish the adversity (Focus Group July 1995).

The BPHIM is now adopted within the State by the 16 Health Jurisdictions identified with the highest Black infant mortality rates.

1.2 Problem

Before adopting the Best Practice Intervention Models (BPHIM), the 16 California Health Jurisdictions did not have a standardized method of measuring the impact of services provided to their clients. All 16 Health Jurisdictions offered similar services, but had different data collection methods and evaluation procedures for their Health Care Programs. Therefore, with the wide range of data collection methods evaluation and analysis of the BPHIM to reduce low Black infant birth weight could not be evaluated.

2. MIS DEVELOPMENT

To facilitate this evaluation, the State Health Department authorized the development of a Management Information Systems (MIS) to support the Black Infant Health (BIH) Evaluation Project. This section describes the methods used in the development, and the next section describes the implementation methods used in the development of the MIS.

2.1 MIS End Users

The MIS development focused on an end-user-centered development process based on requirements of three end-user groups:

* State Department Health Services--These end users were the State Department Research Program Specialists, the Health Program Specialists of Prenatal Care, and an epidemiologist; all were in the Maternal and Child Health Branch of the State Department of Health Services.

* Health Jurisdictions--These end users are the 16 Health Jurisdictions with the highest Black infant mortality rates in the State.

* Black Infant Health Evaluation Project--These end users were the Director of the Maternal Child Health Regional Institute, Program Coordinator, an epidemiologist, a research assistant, and an administrative assistant, all contracted to evaluate the BPHIM.

2.2 Requirements

With the multiple end users for the MIS, the development process required an approach that would make the system flexible enough to meet the needs of all three of the end user groups. The system had to be flexible enough to adapt to the end users health care growth in the use of the BPHIM and their changing operations.

2.3 Needs Assessment

After establishing the data needs for the BPHIM, a needs assessment was conducted to determine the input and output requirements for each of the 16 Health Jurisdictions. Face-to-face interviews and meetings were held with representatives from each group to assess the needs from the MIS for the components, users, and relationships.

To further define the needs of each Health Jurisdiction, sample client intake forms used at each of their sites were collected on clients entering the Black infant Health Care program. The data collected included prenatal information, client demographics, client history, family/living situation, support systems, current prenatal care information health pregnancy related conditions, family planning information, psychosocial factors, assessment of needs, referral services recommended, postpartum information, and birth outcome.

2.4 Data Set Analysis

Each of the 16 Health Jurisdictions had different information intake forms and methods plus they did not all collect the same information. Therefore, an analysis was conducted to ascertain any common data sets and classifying items into a small number of mutually excusive groups.

The results of the data set analysis indicated five distinct areas: history, tracking prenatal, tracking postpartum, contact, and psychosocial. History contained eight data elements; tracking prenatal contained two data elements, tracking postpartum contained five data elements, contact contained two data elements; and psychosocial contained eight data elements. Data collected produced a cluster of data sets that the 16 Health Jurisdictions were collecting.

Assessment from the data set analysis and requirements from the three end users groups and from the data set analysis revealed the need for an information system to provide a standard way of collecting data and measuring outcomes from the 16 Health Jurisdictions to meet the following requirements:

* standardized input data and allow them to apply the intervention models

* provide a tool to generate necessary reporting

* provide necessary client/patient tracking

* provide a source for documenting client/patient history

* provide a vehicle to monitor client/patient outcomes (tickler file)

* provide a source to examine, manipulate, document, and report Model data

2.5 Design Development

The data sets and model links established the main directory for the MIS data screens. Each data set would be represented as an end-user screen in the MIS. This element of the system design provided the match of the data screens for the BPHIM and for the end-users so they could maintain their current data collection in the MIS.

A prototype of the information system was developed for a user-centered environment. Use of a prototype was the most appropriate method for the end users to have something in hand to relate to their task of client health care service. They also needed to have something that they could provide feedback in relationship to their data needs.

The backbone of the MIS development is a data base management system (DBMS) written in Microsoft FoxPro for Windows. At the request of the State Health Department, FoxPro was the DBMS used for development because it was the one used in other health services databases. They wanted the compatibility for linking with other health service programs in the future. Relational tables with user screens for the MIS were established from the intake cluster data sets and model links. The relational tables were the vehicles for developing and navigating the screens for each of the BPHIM screens. The function of the system allows users to track the client and their baby in relationship to how the client entered the Black Infant health (BIH) program, the BPHIM, and the care outcome.

2.6 Design Considerations

In designing a self-guided end-user centered system for the three end user groups, the MIS Development Team had to find an intuitive and logical flow for the design of each screen. The intuitive and logical flow for the end user would be the major consideration in developing the format and layout of each screen.

To obtain the information for the format and layout of the screens, the performance of the end users were observed. The observations of performance were made while the end user used a prototype with dummy data that represented the type of data they would enter into the system. This prototype had been developed based on the requirements for the system and the user-centered environment. The observation participants were the members of the Data Work Group because they represented a cross-section of the three end user groups.

The results of the observation were analyzed and modifications were made to the system from the results. The results were integrated into the screen designs for the flow and processing of the data. Figure 1 shows the screens for the data forms and the categories for each BPHIM.

[FIGURE 1 OMITTED]

3. IMPLEMENTATION METHODOLOGY

Following the analysis and modification to the system from the performance observations of the prototype, the implementation phase was started. In the implementation phase, pilot testing, installing hardware and software, and training were done. The end users were actively involved in each of these phases by providing feedback on their experiences with the MIS, and providing information on any errors they discovered in the system. This section contains the methods used for the pilot testing, installation, and training done in the implementation phase.

3.1 Pilot Testing

Seven health jurisdictions were selected for the pilot test, and the MIS pilot test had to also include testing the BPHIM. This process of combining pilot tests is not the norm; but to avoid any further delay the MIS Development Team had no choice. The lesson learned here was to make sure the contents that the MIS is to support have already been completed and tested.

The pilot test took three months and the results indicated a successful integration of the BPHIM and the MIS. The success was measured by the end users' feedback on the ease of use and their exclusive use of the system. As a result of the feedback analysis from the pilot test, a few modifications were necessary to the database.

3.2 Hardware & Software Installation

Implementation of the remaining Health Jurisdiction overlapped the pilot test and provided additional feedback for modification to the MIS. The overlap was necessary because the State Health Department mandated all the Health Jurisdictions to start using the MIS immediately. However, from a systems point of view, the overlap of the implementation and pilot testing was very unusual and made it difficult to evaluate and update the system from the pilot feedback because the testing had not been completed.

3.3 Training

The feedback from the pilot tests along with the installation overlap indicated that for the implementation of the MIS to be a smooth transition, training would have to be a key factor and an ongoing process.

The format of the training consisted of discussion on the features and concepts, hands-on using sample data, and a questions and answer session. Training took place at each Health Jurisdiction site conducted by two of the MIS Development Team members. The topics covered in the training were data forms, database and Health Jurisdiction integration, BPIM and the database integration, and user manual.

4. CONCLUSIONS

The development of the information system is complete and the project is in the evaluation and maintenance stage. Feedback from the users is constantly evaluated for system maintenance and revision.

The overall Black infant Health Evaluation Project is ongoing for a longitudinal assessment of the impact of each of the BPHIM in the 16 Health Jurisdictions throughout the state. The users of the system are effectively using the system to collect, analyze, and report information on factors that may contribute to high rates of Black infant mortality and low birth weights in the African American population. All 16 Health Jurisdictions now have a standardized way of measuring the impact of services provided to their clients.

5. REFERENCES

BIH Focus Group Response to Adversity for the BIH Project, May 3, 1995.

Gates-Williams, J; Jackson, Nieri; Jenkins-Monroe, V; and Williams, L. The business of preventing African-American infant mortality. Cross Cultural Medicine--A Decade Later, Special Issues. West J Med 157: 350-356, September, 1992.

Korenbrot, Carol. C. Risk reduction in pregnancies of low-income women: A comprehensive prenatal care through the OB Access Project. The Regents of the University of California for MOBIUS, Vol 4, No. 3, 34-43, 1984.

Prenatal Care Network, County of San Diego San Diego County Prenatal Care Network Program Evaluation, May 26, 1994.

Raskin, Ira. E. and Maklan, C. W. Medical treatment effectiveness research: A view from inside the agency for Health Care Policy and Research. AHCPR Pub #91-0025. Public Health Service, Washington DC, 1991.

State of California Department of Health Services: Birth and Death Records, Maternal and Child Health Program Data and Evaluation Section, February, 1995.

U.S. Government Printing Office, Public Health Service: Healthy People 2000: National health promotion and disease prevention objectives. DHHS Publication No. (PHS) 91-50212, Washington, DC, 1990.

Dr. Carolena Lyons Lawrence earned her Ph.D. at The Ohio State University and is currently an associate professor at San Diego State University in the College of Business Administration, Department of Information and Decision Systems.