By Les Morgan
The following reports were produced as deliverables for our project “Aggregating Care Plans to Manage Supportive Care Services for Elders” (Joanne Lynn, M.D., Principal Investigator).This project is funded by the Gordon and Betty Moore Foundation through Grant GBMF5662 to Altarum Institute. Dr. Lynn will provide a more detailed report on the project as a whole in a following blog post.
We are releasing these reports now particularly to help some of the community groups we are working with on data projects. The first three reports cover technical details of how care plans can be structured, stored, and shared in electronic information systems. Those three reports taken together provide useful information that can be used by community groups seeking insight into eldercare through the use of data aggregation methods.
Communities that are seeking to improve eldercare need trustworthy data to use in setting priorities and in monitoring improvements. One appealing approach is to aggregate existing health-related records and analyze the data for key indicators of how their local system is functioning. However, for most of the communities we have worked with, on a practical level the aggregation of this sort of data seems to be very difficult or impossible, given important concerns over privacy and the serious penalties associated with breaches of privacy obligations. Our hope is that these reports will help overcome some of the obstacles that stand in the way of improving care for some of our most vulnerable citizens.
To read the full reports in PDF format, click on the report name you wish to view.
Joanne Lynn and Les Morgan. October 20, 2017
Effective multidisciplinary clinical teams know their patients but document only a skeletal summary of the case. That summary generally includes diagnoses, basics of the living situation, medications, treatments, and supportive services in use or recommended. The various biases, omissions, and lack of long-term perspectives in the clinical documentation are substantial, as summarized in the report.
Our scan of forms and data formats for care plan documentation shows a wide variety of approaches. All made heavy use of free-text narrative elements. The most complete and up-to-date set of care plan records that we have seen was done longhand on a typewritten sheet by a single care coordinator responsible for over one hundred high-need patients seen at one hospital.
The electronic record systems that we reviewed used only to carry the minimal information needed for a C-CDA transmission mostly limited to immediate and near-term needs. None of the electronic records we examined were being used to record the caregivers with their roles, the prognosis, or an advance directive. None used the existing FHIR data format standards for complete care plans.
Les Morgan. October 20, 2017
This is a high-level Information Technology (IT) architectural review discussing major variations in system design approaches for care plan data aggregation, based on a content analysis of interviews with vendors and some key users. A series of architectural diagrams is included to classify approaches. Our interviews with technology vendors are ongoing, and this preliminary summary of methods will evolve as we see more examples of working systems.
Les Morgan and Joanne Lynn. March 2018
This report gives an overview and assessment of regulatory, ethical, and privacy issues specifically related to the use of aggregated care plan and related health data for analytical use across defined geographical catchment areas. The literature covering regulation and privacy of healthcare data is extensive and often contentious. Our review here highlights some major sources for authoritative guidance, then moves on to the specific situation of using aggregated and de-identified data sets for geographic analysis. Regulations that are specific to requirements for maintaining care plans were covered in our prior project report (Interim Report on the Variety and Merits of Care Plan Templates and Regulations in Use, Including Implications for Information Technology) and are not repeated in this report.
A key conclusion of our review is that using aggregated and de-identified data on a broad geographic basis is probably not restricted as much as many of our community stakeholders and leaders think. Some uses of such data are specifically exempted from key privacy laws when appropriate protections are in place. Some regional health care payer systems and Health Information Exchanges (HIEs) have mechanisms to enable such research now, using very large data sets they have already compiled. This means that it is feasible to carry out pilot projects to demonstrate practical methods for data aggregation and analysis for a community or region in many parts of the United States.
Meghan Hendricksen and Joanne Lynn. February 2018
The overall purpose of this report is to provide an initial, alpha-version, benchmark of the workforce within geographic communities to use for illuminating comparisons with their own workforce estimates.
This baseline benchmark will provide a tool for consideration, and then feedback for improving the tool. In essence, we are creating the start of a learning system for optimizing the workforce in a geographic community. Once the estimates and ranges become substantiated in multiple communities and improvement activities, the maturing benchmarks will provide a trustworthy source of guidance for communities, where evidence is currently lacking, on what the priority needs are for the workforce that is serving older adults living in the community and also will provide a tool for monitoring improvements. Measuring the local workforce capacity also helps engaged community leaders to envision their whole system and its products, even though that “system” is owned and operated by many different entities that are not necessarily coordinated.