Practitioners of the healing arts have documented care for various purposes for thousands of years. For example, the 3,500 year-old Ebers Papyrus contains herbs and recipes for the healers of that time.
However, paper based records have several challenges as a medium to store medical data. These are still true today:
- Paper records are not easily shared, and can only be used by one person at a time.
- Paper tales up a lot of space and it doesn’t scale easily. This means when you add a lot of data on paper, it becomes harder and harder to do anything useful with it.
- Paper data can sometimes be hard to read because of bad handwriting. It can also be hard to read because people can write things in any place on the page or any order they want (called format)
- Paper records are difficult to keep safe and secure while still allowing authorized people to easily get it. Backups, access control, and auditing are a challenge with paper.
Interest in healthcare data was renewed when electronic storage and computing was invented during the second half of the 20th century. However computer hardware was expensive, software was cumbersome, and healthcare providers had no compelling reason to stop working with paper the way they were taught.Hannah, K. J., et al. (2006). History of Healthcare Computing. Introduction to Nursing Informatics. New York, NY, Springer New York: 27-40.
While clinical data would remain ‘locked up’ in paper until a major 2009 law forced EHR adoption among healthcare providers, billing and insurance data were beginning to be managed electronically decades earlier. Until the 1980s, the efficiencies of computerization outweighed the costs only for large healthcare organizations and medical billing companies. By the 1990s, the cost of computers was relatively low and technical staff were more available, thereby increasing the proliferation of billing data. In 2003 Medicare ensured universal adoption of billing computers with a law requiring all providers to bill electronically(with exceptions only for very small practices). 1
However, the most significant impacts of the computerization of healthcare data result from allowing the computer to perform new software functions actions on the data, rather than mimic a static paper chart. For example, the quality of documentation can be improved through the use of auto-correct, structured templates, and decision support.2
Delivering More and Better Health Data
Data standards, the rules by which data are described and recorded, are essential if you want to trust your data and anyone else’s. Data standards such as DRG, ICD-9-CM, and ICD-10-CM allow data from organizations around the country to be used together and compared. Moreover, these data standards all but require significant computerization, adding pressure to healthcare providers to ditch paper records.
Public data sources such as the Healthcare Cost and Utilization Project ( HCUP) and the All Payer Claims Databases (APCD) have emerged in an effort for transparency and to spark innovation. Although these two are regulated at the state level, CMS also provides public data related to its programs and patients.
Private data sources, such as EHR companies and clinical research organizations is also available. One such organization is TriNetX, which is a network of healthcare organizations who have agreed to certain conditions of data sharing and use.
Laws & Policies
Some laws, such as HITECH(2009), MACRA, and the 21st Century Cures Act directly increase the amount and availability of health data. For example, they may require the use of specific IT software, establish consistent data standards, or even mandate new data for reporting.
However, many laws and policies achieve the same thing indirectly, -often by requiring a capability that is difficult to do with paper records. Patient -Centered Medical Home Certification is a good example of this. Nowhere does it explicitly require IT capability. However, it would be virtually impossible to reach certification without the help of electronic health records, population disease registries, and electronic case management. Two other examples of indirect incentives of Health IT/Data adoption are the Prospective Payment Program (aka, DRGs) and Pay for Performance. Two common reasons for any industry to improve it’s data analytics capability are efficiency and quality. DRGS and Pay for Performance financing require efficiency and quality, respectively.