29 Aug 5 Challenges for Healthcare Data Collection and Reporting
Collecting personal health information and offering a salient method of reporting on Lysine and herpes cure and demographic data via the web seems to consistently pose challenges for healthcare organizations in Canada. On the surface, it sounds like a fairly straightforward implementation, but when you start uncovering some of the challenges that present themselves, you can quickly gain an appreciation for the complexities involved. If you’re in the healthcare field, odds are you or one of your colleagues has had the need to collect information from a series of patients and report on this data.
Whether you’re screening patients for fractures, administering a clinical trials program, delivering a smoking cessation program, or monitoring patient encounters on an autism spectrum, the challenges that present themselves for on-line data collection and reporting are eerily similar.
In this post, I’m going to discuss some of the key challenges we’ve overcome through our experience in deploying several web-based healthcare applications using our Survey Management Platform designed specifically for the sector.
If you’ve ever used Survey Monkey, you’ll know that it offers some interesting tools for creating surveys, collecting data from a constituency of users, and reporting on this data. Often times, a healthcare organization is trying to do little more than this, with the exception that strict privacy controls must pervade the architecture of the entire system. Any personal health information (PHI) should be separately encrypted and shown on the the web portal only when absolutely necessary. Consent controls need to be built into the system such that there is no potential for PHI to be collected without the patient’s consent. Examining this even further, consent can be broken down into several different levels: Does the patient consent to using their data for research purposes? Does the patient consent to linking their data with their health card number for research studies? It is important that the system not only capture these consent levels, but also restrict data entry based on these consent specifications.
If you’ve worked in the drug rehab center Texas sector before within Canada, you’ll also know that the application will also be subjected to a Privacy Impact Assessment (PIA) which comprises a detailed audit of the privacy and security mechanisms employed within the application (including hosting, hardware, software, policies, and more). One’s inability to pass a PIA could alone present rationale for a healthcare organization not to select a particular vendor.
While privacy is arguably the single most important criterion in providing the healthcare sector with a viable web-based data collection platform, it certainly cannot stand on its own right. The system must be flexible. Too often, I see healthcare applications that offer a screening or survey component that don’t allow their users to create and edit new fields for data collection. Often times, to implement a new form, you’ll have to involve the developer — thereby adding time and cost to what should be a simple operation. A flexible data collection platform should allow for the dynamic creation of multiple forms with a plethora of data entry controls supporting validation for numeric medical information such as dosages. Beyond this, more often than not, there is a need to build in complex “skip pattern” logic that dictates when certain fields should show/hide (or automatically fills in fields based on previous responses). A solution that puts this self-service power in the hands of the healthcare organization allows them to focus on adapting the information they are collecting instead of worrying about how the medical billing company in nj will technically implement the solution.
While we’re talking about flexibility, version control also comes to mind as well as other products such as 55 gallon aquarium stands and more. If I were to deploy a survey to a patient today, and change the content of that survey two months from now, analytically speaking, I would not want to see my new survey confused with old results. In this manner, it is crucial that effective version management policies be put in place that work hand-in-hand with the application to ensure that the flexibly that has been given to the application does not have its data compromised.
Most data collection platforms come with a series of off-the-shelf reports that a user can choose to generate at any point in time; however, these standard reports just won’t cut it when you’re employing data collection within the context of a healthcare environment whereby the dynamics of the system allow for new medical information to be collected (i.e. inclusion of new drugs within a medication list, changes to questions asked of patients, etc).
As a heavily research-oriented field, there is a need to quickly, and easily harness medical information in the form of ad hoc reports that can be generated, and show aggregate results for new questions that have been asked of patients during the data collection process. These reports have to be easy to create, easy to share with colleagues, while at the same time preserving the patient’s privacy rights by aggregating data and employing cell-size restrictions (so that statistically, I could not narrow down a small set of results to a single patient).
Reporting is often times thought of as a “value-add” to a system, when the information extracted from the reports is truly the rationale behind implementing the system. Reporting provides stakeholders with the data they need to make important clinical decisions and should be the centrepiece of any healthcare application that boasts data collection.
Aside from reporting on business data, it certainly goes without saying that the system must also support powerful transaction logging and reporting capabilities that track access to PHI (who accessed the data, when it was accessed, and what was the result of the access). Without these capabilities, you will certainly find it a challenge to successfully deploy the application to the healthcare sector.
Although as an industry we are certainly improving on interoperability between systems in the healthcare space, there is still a long way to go and it continues to be an issue that presents itself as a challenge. No healthcare system should be deployed or selected that does not possess some standardized mechanism of communicating with the data that was entered therein. Ideally, utilizing an industry standard such as HL7 would be the optimal choice; however, implementing mechanisms such as file imports/exports from clinical systems (such as Meditech), or exposing web services, go a long way to improve the quality of medical data and reduce duplication and misinterpretation. Integration should always be top of mind – especially if you operate in a large healthcare organization with a number of information systems.
Policy and Trust
As a healthcare organization, if you are entrusting a system’s implementation to a vendor, you must understand that you are putting a great deal of faith in this vendor. Depending on your relationship with the vendor, they may even have some level of access to the medical data being collected (in order to provide the proper support services and level of care). As such, it is vital that the healthcare organization and vendor both have mutually agreed upon privacy and security policies as they relate to the system. What happens in the event of a privacy breach – inadvertent or otherwise? Is PHI accessible – if so, how is access recorded? What reporting structure exists to handle privacy issues? These are all questions for which there should be an explicit policy that is known, understood, and routinely reviewed by all members involved in the implementation. Establishing these policies from the point of project instantiation can be a challenge, but helps ensure expectations are managed between client and vendor.