Foundational Business Diagnostics Introduction
PLEASE NOTE THAT THIS REFERENCE DATABASE IS OUTDATED AND DOES NOT REFLECT INFORMATION GATHERED ON VALUE BASED PURCHASING IN 2015.
The employer-based health care system began proliferating in the 1940s and 1950s as an economic imperative to recruit and retain quality employees at a time when there were limits on salary. Since then, costs have risen in a manner that has placed significant financial stress on the employer community. This first occurred in the late 1970s. In order to control costs, benefits were reduced and utilization controls were put into place. This was also the era of the “Health Maintenance Organization” (HMO). Although these strategies slowed costs for a period of time, soon they again began to rise again. In the 1990s, similar economic pressures caused a significant change in strategy as payers turned to health and wellness programs in the hope that education and behavior modification would ease the financial burden associated with employer-provided health care.
The early 2000s brought international competition to the US employer market due to the global economy in conjunction with a renewed double-digit rise in health care costs. The health care cost pressures threatened many US corporations’ existence. In the hope of slowing health care utilization, most US businesses shifted greater cost-sharing, in the form of higher co-pays and co-insurance, to employees. These strategies were undertaken as broad-based actions with little precision. Rarely were they based on concrete data because most of the data was unavailable to the employer’s human resource team.
Today, employers and their employees are paying 56% of the national health care bill. Health care is now consuming more than 15% of many companies’ revenue. Hewitt Associates found that in 2011 employees will be contributing about 22.5% of the total health care premium. Average employee out-of-pocket costs jumped 12.5% in just one year. Employees’ share of medical costs—including employee payroll contributions and out-of-pocket costs—have tripled over the last decade. These costs have reached unsustainable levels for both employers and employees.
Data: A New & Necessary Tool For Employers
These circumstances have created the opportunity for a new strategy to take hold. The use of health care data has taken on a new role within the health care management world. Historically, health insurers utilized data primarily as an actuarial tool. Over the last ten years, however, employers are increasingly relying upon data to identify cost risks as well as clinical and quality opportunities within their companies. When asked, 86% of employers stated that they bear responsibility for not only costs but also quality of the health care that they offer their employees. This perspective is transforming the employer from a passive payer to an active purchaser and strategist in their health care benefit. The collection and analysis of data is an integral part of benefits management. Without it, a company risks making decisions that cost the company precious resources and that could have long term negative consequences.
Data can be used in a variety of ways within both an employer and employer coalition environment. As no two employers and coalitions have utilized their data warehouse (see definition below) in exactly the same manner, it is important to understand the potential opportunities that exist through the use of data. Yet, there are also a number of obstacles that need to be addressed in order to maximize data warehousing activities to ensure they translate into meaningful and instructive information.
Data collection and analysis can be time consuming, difficult and expensive. Employers need to have a clear understanding of what data mining can contribute to their decision-making and overall benefits strategy. Employers should not anticipate that the return on investment associated with data collection, analysis and warehousing will occur within one year.
Data Drives Design
Historically, data was used for actuarial purposes within insurance plans. As employers began to utilize health care data, they too initially did so to address cost issues. More recently, as value-based purchasing has become more prevalent, data is also being used to analyze and address issues surrounding quality. As value-based purchasing gains momentum, the need for data and data-based decision-making will continue to grow. Knowledge, commitment, and practical interventions are all necessary to propel positive change that improves both employee health and employers total healthcare costs. The knowledge in this case will come from the information accessed through health care data.
The information that can be gleaned from data can support initial decision-making, aid in plan implementation, and be utilized during ongoing assessments of benefit design’s impact upon participants. Comprehensive data collection and rigorous analysis will give the payer a deeper understanding of the true financial implications of chronic conditions on the productivity and health of the targeted population, as well as the impact of health care quality has on patient outcomes. Data analysis will help guide employers to better strategies to create better long-term health and economic value.
Data Aids All Health Care Stakeholders
The use of data should not be relegated to only the corporate benefits team. Its potential to affect cost or quality can be maximized by finding opportunities to share it in actionable ways with other constituencies—such as employees or health care providers. Over the last several years, more employers are working with their health care vendors to create transparency by sharing information related to cost and quality with their employees. This information sharing is necessary in order for the employee to make better health care decisions and become a savvy “health care consumer.”
The use of cost and quality information can also be an effective change agent when shared with health care providers. To date, information sharing with the providers has occurred more often through data pooling within coalition activities. However, it has also occurred through a small number of employers. Companies such as Gulfstream have aligned incentives for both employees and their providers in a program that rewards the use of evidence-based medicine standards for a number of medical conditions. Quality data was shared with both stakeholder groups in order to create a successful model.
Definition of a Data Warehouse or Integrated Health Care Data Management System (IHDMS)
These two terms are often used interchangeably. In either case, it is a centralized set of databases, often from multiple sources. These data sources are then analyzed in order to produce information that can help employers or coalitions to make informed decisions. Through the use of analytical algorithms, details can be elucidated and relationships identified.
Just as employers need to decide whether to conduct data analysis independently or in conjunction with a coalition, employers and the employer-based health care coalitions must decide whether to create their own data warehouse or to utilize the services of a third party.
Overcoming Barriers: Potential Challenges to Data Collection
With the potential power that data has in informing decisions within an organization, some may ask why it has taken so long for it to be used. The broad answer to this question is that potential obstacles need to be overcome in order to collect, integrate and analyze health care data. Although none of the barriers are insurmountable, many employers are reticent to take on the task.
1. Health Care Data Is Unique. One of the most basic issues is that the transactional data found within health care is not as clear-cut as can be found in other areas. For example, care for the same diagnosis can occur in the physician’s office, the emergency room or a hospital’s outpatient facility. Even if the care and associated coding are the same, costs may be significantly different based upon where the care was given. This can cause a lot of confusion on initial review.
2. The Acquisition of Data Is Challenging. Employers have often stated that one reason that they do not utilize data is that they have difficultly acquiring it. One of the most common reasons for this is the large number of health vendors that they utilize. It was not that long ago that a large employer may have offered up to 20 different health plans to its employees. This situation has been changing over the past few years, as the trend seems to be toward consolidating the number of health care vendors. The reduced number of health care vendors makes it a bit easier to accumulate data from all of them.
As previously noted, many health care vendors are unable or unwilling to share health care data for a variety of reasons, including the real or perceived loss of income associated with the data, “ownership” issues that translate to “data is power,” the real or perceived privacy issues, and the legitimate cost incurred passing the data to another party in a usable format. Questions persist as to who owns the health care data. Is it the patient, the health care provider or the purchaser of the health plan? Depending on the purpose for which the data is being utilized, it may be all three. In some cases, state and federal law may dictate data ownership. For the purpose of plan design and health care payment, the payer does have a right to receive and utilize health care data.
Small and large organizations should put data clauses in all of health care vendor contracts. These clauses should include:
- What data will be shared;
- How often the data will be shared;
- Cost of sharing the data; and
- Quality of the data shared.
3. Data Collection Can Be Costly. Another common barrier encountered by employers is the cost of data collection. This is an issue that cannot be underestimated. The start-up costs of collecting and maintaining a data warehouse with standard specifications can vary. Nevertheless, cost can be broken down categorically by acquisition, storage and analysis.
The initial cost involves acquisition of the data. Many health care vendors will place a financial charge on the employer to receive data on any kind of regular basis; the more frequent the transfer of data, the greater the cost. There are also implications around decisions such as appending the last file or overwriting it. There are some vendors that will not place a charge on transfers of data that occur only a limited number of times throughout the year. Other variations on cost include the number of data feeds being combined and how the data are organized (e.g., flat file, data dictionary, non standard requests, number of vendors).
The second area affecting cost is storage. The data will need to be stored and protected in a way that complies with federal and state privacy and security laws. This will add an extra expense, with storage space being relatively inexpensive while security may be costlier. The expense associated with storage will depend on the amount of data stored as well as whether it is housed in a data warehouse built internally or maintained by a third party.
The third area with cost implications is the analysis of the data. Again, this cost can be due to either use of internal resources or payment of a third party vendor. This issue is not irrelevant as the cost–benefit analysis is done. It will also help to frame implementation decisions that are necessary early on in the process.
An argument can be made that the cost of not utilizing data may be greater than the costs of collecting and analyzing data. However, it is important that one not minimize time or costs involved or oversell the potential cost or quality improvements that may be achieved.
4. Data May Be Incomplete. One area of data that is often missing is that of employees’ dependents. If the dependent has not had a medical claim, they are often invisible to the employer. Employers utilizing health risk assessments (HRA) for their populations may also lack this information because health fairs that often collect the HRA and biometric data usually take place at the work site and therefore do not include dependents. The lack of data for this population may skew future costs and risks.
Another area of data that is often missing is productivity data, which includes disability costs, workers’ compensation, wage replacement for hiring a temporary employee or paying overtime to an existing employee, absenteeism and presenteeism. Some companies may not distinguish between vacation time and sick leave, making it more difficult to track lost time at work attributable to health issues.
In addition, there may be gaps in data due to the fact that some data is documented as text found only in the medical record and is therefore not easily captured. It is the hope of many that the increasing utilization of electronic medical records will at least partially address this issue.
5. The Quality of the Data Is Difficult to Assess. It is important that the accuracy and quality of the data being collected is assessed. Data integrity problems can create situations where the information taken from the data can lead to incorrect business decisions. In order to assure data accuracy, issues such as common identifiers to appropriately link information, aligned definitions associated with data elements, coordinated time periods, and collection methodology must be discussed and agreed upon by those participating in the data warehouse.
Once the data elements are agreed upon, a data summit with all parties should occur. During this meeting, all the issues associated with data accuracy and quality should be discussed and agreed upon. This will reduce time and effort in the future. A quality data guarantee should also be considered during contract negotiations.
6. Linking Data Is Cumbersome. Although time consuming, administrative claims data, especially medical and pharmacy claims, can usually be accessed and acquired. Furthermore, the rules and regulations that dictate its use are well documented. However, linking administrative claims data to other data elements, especially data from employee surveys, is often very difficult to accomplish. In addition to the challenges associated with the linking of these sources, the survey might require individual waivers from employees to share the data. These same challenges will exist with worker’s compensation, paid time off (PTO) or other productivity-related data as source system variations often make these data difficult to link internally.
More complexity occurs when trying to link medical and pharmacy administrative claims data from either different entities or different source systems. Among the challenges is uniquely identifying individuals and their claims history. A possible solution would be to have the payer provide a dataset where this is accomplished in-house and de-identified.
7. Myths about the Collection of Race/Ethnicity Data. Some people believe that collecting data on race and ethnicity is illegal under federal law. It is not. There are no federal statutes that prohibit the collection of this data, which may be vitally important to ensuring the success of a value-based benefit design. Certain races, religions and ethnicities have their own beliefs regarding illness, wellness and medical care. Therefore, the cultural communications and benefits designs may depend upon understanding who should receive the varying components based on their culture, background and preferences.
Although federal law does not prohibit collection of such data, patient self-reporting of race and ethnicity has its own challenges. Employees may question why a payer needs to collect this data or why an employer would want to study different impacts and utilization by race/ethnicity. In addition, many individuals will not give this information for fear of it being utilized in a discriminatory manner. If such data is being collected from medical providers or patients and their families, the data collector should make sure that there is a set of questions that are asked in a uniform manner. Privacy and security will be a key component of education for participants. Have a written rationale for why the patient is being asked to provide the information as well as a written security procedure to ensure information will not be used in a discriminatory way.
8. Privacy Is Paramount. A survey done by the California HealthCare Foundation found that there is significant concern over the use of health care data by both health plans and employers. The survey showed that 54% of those polled felt that the replacement of paper medical records by electronic medical records posed a threat to their privacy. They also voiced a concern that the data collected could potentially be used against them in employment matters or place their health insurance at risk.
Protection of patient data is of paramount importance. The Health Insurance Portability and Accountability Act (HIPAA) includes stringent privacy and data security regulations that have created both real and perceived data challenges. Changes to HIPAA in 2009 extended many regulations related to electronic health data to additional organizations and vendors. It will be necessary to create business associate agreements or other methods to ensure compliance with federal and state privacy law. The data summit should include individuals very familiar with current rules and regulations of data sharing as well as changes that might impact the process. There will need to be documented rules for storage, use and disclosure of any data on behalf of the payer. Further information on privacy regulations appears in the “Legal Considerations” section of the VBBD chapter.
In this chapter, we provide an overview of strategies associated with the use of data as a foundational tool in health benefits activities, which empower companies to employ proactive rather than reactive strategies. In some circumstances, employers find that they do not have the resources to collect and analyze their health care data. In those cases, they may consider partnering with others such as employer coalitions.
This chapter addresses:
what constitutes data;
- different types of data;
- barriers to data collection and analysis;
- case studies showcasing employer experience with data warehousing;
- relevant research and reference tools;
- steps to building a data warehouse; and
- legal considerations.