8 Keys to Understanding a Patient Registry

Know your users, define your data

Patient registry
Jochen Sand/Digital Vision/Getty Images.

A patient registry is broadly defined as an organized system that uses observational methods to collect data for assessment of patient outcomes. Patients in a registry share a common health-related characteristic, such as a disease (e.g. diabetes), health condition (e.g. pregnancy), exposure (e.g. to a drug), or health service (e.g. hernia surgery).

A patient registry can exist in any format. The simplest registry would be a stack of index cards.

Registries are more commonly organized on electronic spreadsheets or complex online databases.

Ask yourself these questions if you are seeking to understand the benefits of a specific registry, configuring an existing registry package, or building a registry from scratch.

1) What is the purpose of the registry?

Decide why you need a patient registry. What problem will the registry solve? What will it allow you to do that you can’t do now? A registry can serve multiple purposes, but each should be defined and well-understood. 

Common purposes of registries include:

  • Monitor health status of defined populations
  • Facilitate disease management
  • Understand variations in treatment and outcomes 
  • Observe the natural history of a disease
  • Evaluate clinical effectiveness or cost effectiveness of health care services
  • Measure quality of health care
  • Monitor safety or harm
  • Public health surveillance

2) Who does the registry track?

There should be clear criteria on which patients will be entered into and tracked in the registry. Consider characteristics such as health status, age, geographic location, and affiliation with health care entities. For example, a registry could track children 5 to 12 years old seen at public hospitals in New York City for asthma.

Another registry could track women who have been diagnosed with breast cancer in Colorado.

3) Who uses the registry?

Potential users of a registry include physicians, nurses, case managers, social workers, researchers, and health care administrators. Somebody who uses the registry to track health services delivery may need different features than a researcher who is monitoring potential side effects of a drug.

4) What do the users do with the registry?

The functions of the registry should match known or anticipated needs of its users. For example, a case manager may need to identify the patients in a diabetes registry who have not had a flu shot, and follow up with them to remind them to get vaccinated. The registry would need to acquire or receive data on vaccination status, possibly from multiple sources, and may need to display the patient’s contact information.

5) What data does the registry track?

Determine which data elements are essential and which are desirable but not essential. Prioritize accordingly.

For example, a registry of patients with congestive heart failure would likely need to track the use of certain medications (e.g. beta-blockers, diuretics), lab results, and findings on echocardiography (ultrasound of the heart). Efforts should be concentrated on obtaining such data for the registry before targeting other data. 

6) Where does the data come from?

A registry can include primary and/or secondary data. Primary data are collected for the purposes of the registry itself. For example, patient survey responses or research data entered directly into a registry could be considered primary data. 

Secondary data are those that were first collected for other purposes but are later made available to the registry. Examples of secondary data sources include electronic health records, health insurance claims data, birth or death records, census databases, and other existing registries. 

7) Does the data mean what you think it means?

If a registry extracts data from an electronic health record, then the data should be evaluated to determine if the information was originally collected and recorded in a manner consistent with the registry’s needs. The reason is that the electronic health record contains information generated during the course of routine health care, but these data may not be subject to quality assurance processes as required by the registry. For example, if the registry is designed to track patients who have rheumatoid arthritis as a diagnosis in the electronic health record, then errors in documentation could result in the wrong patients being enrolled in the registry.

Some registries extract and re-classify data from an electronic health record and display it in a fashion that is easy to interpret by its users. The registry output would need to be validated to ensure that the underlying data assumptions are valid. For example, suppose that a registry is designed to identify women who have had any type of screening mammogram in the previous year. The electronic health record may have several different names for screening mammograms, and the registry would need to capture all tests with those names and label them as screening mammograms.

8) What is the duration of observation for patients in the registry?

Registries track patients for a given length of time, depending on the purpose of the registry and the nature of the underlying disease or health condition. A registry of focused on short-term mortality after stroke may only track patients for 30 days after the occurrence of the stroke. A chronic disease registry may need to monitor patients with diabetes for several years.


Gliklich RE, Dreyer NA, editors. Registries for Evaluating Patient Outcomes: A User's Guide. 2nd edition. Rockville (MD): Agency for Healthcare Research and Quality (US); 2010 Sep. Accessed on May 22, 2014

Continue Reading