| Healthcare Indicator Project |
|
|
 |
| (HIP) |
|
The Healthcare Indicator Project (HIP):
Measuring and Assessing Primary Care Service Delivery in Oregon
Over a year ago the Office for Oregon Health Policy and Research (OHPR), working with stakeholders on the Oregon Safety Net Policy Team, began the process of understanding capacity and demand in Oregon’s health care system by gathering and analyzing the following information:
- Demand for health care services, defining particular indicators, community characteristics, and measures of outcomes.
- Capacity of providers, focusing on services provided, hours of operation, organizational structure, and other information.
To continue with that work, OHPR, working collaboratively with the Office of Health Systems Planning and other key informants and stakeholders, will develop a set of indicators measuring primary health care capacity, access and outcomes. OHPR is using funding it has received for this project through Oregon’s State Planning Grant (SPG) program administered by the Health Resources and Services Administration (HRSA). The overall goal underlying this grant’s work is to provide data that would help to inform Oregon policymakers to incrementally expand health care coverage to all Oregonians.
To assist with the development of the indicators and to improve measurement of capacity and demand, OHPR will focus initially on strategies for updating urban primary care service areas (PCSAs). These PCSAs will provide a meaningful unit of analysis and are needed to complement the rural PCSAs developed by the Office of Rural Health (ORH). HRSA has already defined PCSAs nationwide by aggregating Zip Code Tabulation Areas (ZCTAs). However, the HRSA PCSAs may be too highly aggregated to meet the State’s needs for assessment of primary care access. For example, the HRSA PCSAs split Portland into just two service areas. This level of aggregation does not allow for sufficient community-level analysis of potential differences in access to primary care across the city.
|
Major Urban Areas
|
Bend
|
Corvallis (includes Albany)
|
Eugene (includes Springfield)
|
Medford
|
Portland/Tri-county metropolitan area
|
Salem (includes Keizer)
|
The first step for the urban PCSAs work is to bring together some technical experts in Oregon’s delivery systems to discuss the potential strategies and reach consensus. Once the urban PCSAs are updated, OHPR will be better positioned to determine both the supply of primary care health services and the demand for these services within each urban PCSA, much as ORH has previously done for rural PCSAs. By using these access and capacity indicators, the State can then better assess the level of unmet need for primary care health services Statewide. This will help to inform the work of the Oregon Safety Net Advisory Council and the Oregon Health Policy Commission as well as for individual communities.
Once the urban PCSAs are updated, the next step for the technical workgroup will be to discuss definitions of primary care outcome measurements. This work will also be brought to other groups for input including the Safety Net Advisory Council and the Health Policy Commission. Examining outcomes will allow the State, informed by the expertise of the technical workgroup and collaboration with other key stakeholders, to develop primary care performance benchmarks. This work will place the State in a stronger position to design and implement data-driven healthcare policy that best utilizes scarce funding resources.
HIP Process (Tri-County Metro Area)
|
Since its inception the HIP has worked diligently to involve key State and community partners in decision-making. The HIP has convened two major groups of key stakeholders from the Tri-County urban area. The first group is the Tri-County Policy Workgroup, which functions achieve consensus on key policy issues confronting the HIP in the Tri-County Metro Area. One concept clarified at the initial meeting is that the HIP should focus on what can be accomplished now using existing data sources. This Workgroup then nominated members to the Tri-County Technical Workgroup, which was charged with developing a list of proposed indicators, including data sources, and updated PCSA boundaries. There was much more discussion than anticipated regarding the PCSA boundaries, requiring that the Tri-County Technical Workgroup convene for two extra meetings before consensus was achieved.
Primary Care Service Area Boundaries (Tri-County Metro Area)
|
The Primary Care Service Areas were first developed in 1986 (see map in Appendix AA ). At that time, the following criteria were used:
- Areas are within 30-40 minutes travel time of primary care services.
- Areas are not smaller than a single ZIP code and aggregations of ZIP codes are geographically contiguous.
- Areas contain a population of generally more than 25,000 but not more than 100,000.
- Areas contain at least 75 percent of the population considered “urban” by the Census.
- Areas are a “rational” medical trade area considering geographic and neighborhood boundaries as well as market patterns.
Early on the Tri-County Technical Workgroup looked at these criteria and decided that they should not be changed. Additionally there was direction from the Tri-County Policy Workgroup that the PCSAs should not conflict with existing rural service areas (as defined by the Oregon Office of Rural Health).
The first step was to consider possible additions to the 1986 service areas (see map in Appendix BB). The Workgroup considered the possible consequence of including additional ZIP codes into the overall Tri-County urban area. Those ZIP codes where significant population centers were within ten miles of the center of a major city were included. Additional ZIP codes in one county were included based on recommendations from a key informant from that county’s health care administration. This resulted in proposed additions to the 1986 PCSAs (see map in Appendix CC ).
The second step was to check to see if the PCSAs meet all five criteria listed above. The travel time standard is easily met in all PCSAs except Forest Grove, where travel time by car from the furthermost populated place to the nearest primary care services is approximately 34 minutes (estimation from MapQuest®). In the vast majority of areas the travel time standard can be met by public transit. At this point in the process all PCSAs were aggregations of multiple contiguous ZIP codes and contained over 100% of the population considered urban by Census 2000.
The populations of some 1986 PCSAs did exceed 100,000. In the third step, the PCSAs were reconfigured to meet the population standard while attempting to retain identification with local neighborhoods. The resulting PCSAs are depicted in Appendix DD, and their populations range from approximately 32,000 to 79,000. Four PCSAs are now single ZIP codes while the remainder are aggregations of multiple contiguous ZIP codes.
The fourth step was to map data to the PCSAs. These maps are displayed in Appendix DD. Evaluation of the PCSAs will be ongoing. Development of PCSAs in the remaining urban areas of the State will be less time-consuming since these areas consist of much smaller populations than the Tri-County urban area and, thus, much fewer ZIP codes. Stakeholders from these communities are being contacted to pursue this work.
Primary Care Capacity and Demand Indicators (Tri-County Metro Area)
|
The HIP Tri-County Technical Workgroup by consensus developed a matrix of indicators of primary care capacity and primary care demand (see Appendix EE ). The matrix provides definitions, limitations, and data sources for both the numerator and denominator. Some indicators have been included even though the data sources have not yet been identified; this was done just in case a data source becomes available in the future.
The capacity indicators are a mix of provider workforce and clinic measures while the demand indicators are population and geographic measures. Although this Workgroup consisted only of stakeholders from the Tri-County urban area, it is anticipated that a core group of these indicators will be identified by all of the State’s urban areas. Again, stakeholders from the remaining urban communities have been contacted to pursue these discussions.
Two key challenges have emerged from the Tri-County Technical Workgroup’s discussions of capacity and demand indicators:
- Assessing the provider workforce is extremely difficult. It is well known that all Oregonians do not have access to all providers. However, it is extremely difficult to accurately determine which providers are accepting new patients, which providers are accepting OHP patients, and which providers are accepting Medicare patients. This information becomes obsolete very quickly. Providing reliable measures of access to providers requires a comprehensive source of data that is updated and disseminated at least weekly.
- Assessing outpatient clinic capacity is extremely difficult. There is no comprehensive data source for clinic measures. Local clinics may also have different administrative procedures that make comparisons untenable. For example, one measure of clinic capacity is the wait time for the next open appointment. Clinics vary on how far out appointments are booked; some do not make appointments at all.
Two key lessons learned can be identified from the experiences of the HIP. The first is to use a stakeholder-driven process. The community and State stakeholders are vital to the success of the HIP. Their input and concerns are of paramount importance. The process of developing indicators and units of analysis will not be successful unless the stakeholders are comfortable with the process and the outcomes. The stakeholders will let the HIP know when it is appropriate to proceed. An addendum to this lesson is to allow ample time for input. It would be far more efficient to sit in a vacuum and dream up indicators and units of analysis, but there would be no stakeholder buy-in to the products of such an exclusive process.
The second key lesson learned is that the focus should remain on what we can do today. Early discussions often strayed into aspirations to develop new data sources or to geo-code mountains of existing data. The HIP clearly does not have the capacity to take on huge data collection or geo-coding projects and this type of discussion was not allowed to deter progress on developing indicators from data sources that are currently available.
|
|
|
|
