The Practice Health Atlas

Introduction

The Practice Health Atlas (PHA) is a process that creates a decision support tool. It has been developed by the Adelaide Western General Practice Network (AWGPN) over the past three years.

General practice data is collected by a trained representative of the Division and then analysed, synthesised with other data sets, and a comprehensive report is produced for the practice. The thrust of the PHA is towards managing patients with chronic disease through better utilisation of the chronic disease item numbers, as well as improving the quality of clinical data. The practice can then act on the implications of the PHA Report, with assistance from its Division. The PHA process is intended to be performed on an annual basis, as this time frame usually allows practices to see measurable changes in practice epidemiology and key financial indicators.

The PHA Process

Once a practice expresses an interest in having a PHA report conducted for their practice, the next stage involves a process of consultation to inform the practice exactly what information is collected and how this information will be used. The information flow is strictly between the practice and the Division. It is not shared with any third parties and, in fact, access to a practice’s PHA is strictly limited within the Division to those involved in its construction.

Privacy and confidentiality agreements between Division and practice are signed, as well as an official enrolment form. This states that the practice acknowledges that it is aware of what a PHA is, that they have read and understood the Division’s privacy policy, and that they agree to have the relevant data collected, aggregated, and analysed.

Data collection

Once the formalities have been completed, the requisite information is collected from the practice. This consists of three main data sets:

1. De-Identified Clinical Data

This is collected using an extraction tool especially built for the PHA by Pen Computer Systems. The output of this tool is a deidentified data set, which mainly consists of patients’ chronic disease diagnoses, most recent clinical measures (e.g. HbA1c, cholesterol), and demographics. The data is de-identified to the postcode level, meaning that only the patients’ postcodes of residence, their age, and their gender are collected in addition to this clinical data.

2. Billing Data

As the PHA is chronic disease-focussed, the billing pattern for the relevant chronic disease Medicare items numbers including GP Management Plans (GPMPs) and Team Care Arrangements (TCAs), are collected (i.e. the number of items and dollar amounts).

3. Marketing Information

This consists of a photograph of the practice (sometimes with all of the staff gathered out the front!) and information about the practice. This information includes the practice’s opening hours, names and professions of practice staff and special interests. As the PHA is done annually, this section is quite popular with practices who have had the PHA done a second time - they can compare to the previous version and then reflect on the changes over the year, similar to a yearbook!

The collection of the data is usually done with minimal effort required from the practice staff, so the PHA construction process is not considered an imposition on the practice’s time. Collection of the billing data is usually the most time consuming activity because there are a range of billing programs in the General Practice environment, some with advanced search capability; others with more primitive functionality.

Constructing The PHA

The construction of the PHA takes place at the Division and is done using the Microsoft Office suite (i.e. Access, Excel and Word) and MapInfo Professional GIS (Geographical Information Systems) software.

The AWGPN PHA training program provides the Divisions with the requisite skills and templates to produce the PHA. Basically, the practice data is entered into different parts of the template. “Fine tuning” of the maps and business modelling is required to ensure presentation quality is high. Construction historically took between half a day and a full day depending on the speed of the operator and the number of interruptions they face. With mapping automation recently introduced, this time should be reduced to a couple of hours. Some of those already trained can do a PHA in approximately 3 hours without the automation.

Other aspects of the PHA are now automated to the stage where the bulk of the PHA construction can be handed to an admin person. The fine tuning and analysis can then be done by someone with more specialised informatics skills.

PHA Analysis

The PHA consists of two main components:

1. Epidemiology And Mapping

In this section, the practice’s clinical data is synthesised with Census data and presented in graphical format. The practice can then see the postcodes from which their populations are drawn and how this compares with market share (i.e. the total number of people residing in those postcodes).

Practices can also see the socio-economic status of their patient population, this having important implications for billing and services delivered. In addition to the census comparison and exploration of the patient populations, the “Epidemiology and Mapping” section also maps the practice’s patient populations, i.e. the regular patient population and also the populations with chronic diseases. An example of a chronic disease map can be seen in Figure 1, a practice’s Cardio Vascular Disease (CVD) population. The range of major chronic diseases are mapped – Diabetes, CVD, Asthma, Diabetes, Mental Health, Coronary Obstructive Airways Disease (COAD), and Bone Diseases.

Practices find this section very interesting — often it matches their perception of the practice population, and sometimes it can enlighten or surprise them. Certainly in most cases this is the first time that the practice has viewed their population in such a way.

2. Clinical And Business Modelling

Following on from above, this section takes the chronic disease figures and other key information (for example, number of patients aged between 45 and 49) and, in consultation with the practice, a business modelling exercise is undertaken.

It poses the question: “If you were to apply the CDM item numbers to X percent of your patients identified within the clinical records with each chronic disease, what could be your opportunity income from care planning activities and improved patient outcomes / well being?”.

The modelling is applied to the available chronic disease item numbers such as GPMPs, TCAs, and Service Incentive Payment (SIP) Cycles of Care. A comparison between current and potential utilisation of these items is made. Based on this business modelling, the Division would then discuss with the practice various options open to them e.g. bringing on board a practice nurse, investing in infrastructure and the like.

The business modelling in this section is quite flexible, with each item (or group of items) able to be individually modified. This allows the user to tailor the modelling to:

  1. Suit the practice’s capacity to perform those items.
  2. Set the utilisation of given item numbers to a realistic level.

The aim of the business modelling is to be realistic and conservative, so the resulting potential income is actually attainable for the practice.

THE PHA and Patient Care

The PHA can assist the practice with their ongoing patient care. At the basic level, it indicates the quality of the data within the clinical database. By mapping the pattern of the various chronic diseases, as well as making comparisons with national benchmarks, the practice can see quite quickly whether they have more or less patients with a particular chronic disease than expected. If the numbers are less than expected, does this mean that the practice is not attaching diagnoses to the patients as part of the consultation?

In terms of enhancing the care of patients with chronic diseases, the PHA has the potential to have practices change their business and clinical system to be more “geared” towards applying the CDM item numbers to patients with chronic diseases and improving the quality of their clinical information.

Summary

The PHA is an evolving general practice decision support tool with multiple potential uses. Among the GPs and staff engaged to date, it has demonstrated its potential to develop a professional culture around quality health data, the utility of integrating and synthesising data with various other sources, and as a driver of innovation in health care service delivery. It requires little technical input, time or effort from the practice staff. A practice interested in having a PHA done should contact their Division for more information.

While the PHA is excellent at indicating the direction and potential of any business/clinical change, it is certainly not a “silver bullet”. Ultimately it is up to the practice to enact change to meet the implications and modelling of the PHA.

Posted in Australian eHealth

You need to log in to post comments. If you don't have a Pulse+IT website account, click here to subscribe.

Sign up for Pulse+IT eNewsletters

Sign up for Pulse+IT website access

For more information, click here.