AIM Data Collection Process


Hospice Aim logo

Overview


This section describes the major components to collecting data. 

While it is highly recommended that an agency collects data for all of the quality measures, it is possible for an agency to collect quality measure data on just a few of the measures.  In this case, you may not wish to work through each of the components from start to finish.  However, there are essential steps that must be completed regardless of how many quality measures you decide to focus on.

Ten Essential Steps to Implementing the AIM Process


Select the Abstractors/Team

Whether you use one person to collect the data or convene a team of abstractors will depend on the size of your agency and the availability of resources.  We recommend that your team consists of at least 2 people regardless of how small your agency is. However, what is most important is that someone takes responsibility for one or more of the following roles:

Abstraction Supervisor;
Clinical Record Abstractor(s);
Data Entry Specialist;
Clinical Advisor.
   
Abstraction Responsibilities

Return to top

Conduct Abstraction Training

The Abstraction Supervisor or another member of the Quality Team should lead the abstraction training with all of the abstractors/team members including the Clinical Advisor. 

It is extremely important that all clinical record abstractors understand exactly what each question on the AIM Data Collection Tool is asking.  The AIM Data Dictionary offers definitions and clarifications about each of the items on the AIM Data Collection Tool.

Training details are found in the AIM Abstraction Process Slide Presentation and the Abstraction Facilitator Script.  At first, abstraction can take as long as 45 minutes per clinical record to complete.  However, depending on the structure of your clinical record system and the ability to automate your data collection process, it could take as little as 15 minutes.

The Importance of Accurate Data Collection
AIM Data Collection Tool Abstraction Process (slide presentation)
Abstraction Facilitator Script
AIM Data Collection Tool
AIM Data Dictionary
Frequently Asked Questions and Answers


Select Clinical Records for Abstraction

The following information presents the rationale for choosing which clinical records to abstract  and why it is important to follow a process while selecting clinical records.

The most important factor when deciding how to select the clinical records for abstraction is to ensure that the method you use will give you diversity in both your patient population and settings of care.  The AIM Quality Measures have only been tested in the adult population.  For patients to meet the criteria for having the clinical records abstracted, they must have been:

It is recommended that the greater of 10% or 10 records per month be abstracted to provide adequate data.  Small agencies may not have 10 discharges per month.  If this is the case, all clinical records should be abstracted.  For very large agencies, resource availability may limit how many records can be abstracted.

Agency Abstraction Calculation Program (Excel Program)

Return to top

Capture Data

The 12 quality measures are captured in two ways: data at the individual patient level and aggregated data (agency level).

List of Quality Measures by Data Source

Individual Level Data

There are four tools available to help capture the individual level data:
  1. Excel Data Entry Tool;
  2. Data Entry Reporting Form;
  3. Abstraction Evaluation;
  4. FEHC and Adverse Event Log Calculation Form.

AIM Data Entry Excel Tool

The AIM Data Entry Excel Tool uses the data collected with the AIM Data Collection Tool for generating numeric and graphic monthly reports for 9 of 12 quality measures.  The AIM Data Entry Excel Tool allows you to monitor your improvement progress over time.  The AIM Data Entry Excel Tool is designed to look exactly like the AIM Data Collection Tool to make entering the data easier. Abstractors or other staff must be designated to enter the data into the Excel Tool so the quality measures can be calculated.  As abstractors get more comfortable with collecting data, the data may be entered directly into the AIM Data Entry Excel Tool during data abstraction, eliminating the additional step of entering the data from the paper tool into the Excel Tool.

AIM Data Entry Excel Tool (Excel Program)
Using the AIM Data Entry Excel Tool for Excel 2003
Using the AIM Data Entry Excel Tool for Excel 2007

Data Entry Reporting Form

Faulty data collection or errors in data entry are very common especially when first learning how to use a new tool.  There are two sources of errors that you will most likely encounter: transient personal factors such as health, fatigue, motivation, and situational factors such as relations with colleagues or distractions.  However, you can improve error detection by putting in place a defined communication plan between the clinical record abstractor and the data entry person.

Using the Data Entry Reporting Form will allow you to record errors or questions that arise during data entry.  Using a "kick-back" system, the Data Entry Reporting Form is filled out by the data entry person and given to either the Abstraction Supervisor or Abstractor.  The form requires that either the Abstraction Supervisor or the clinical record abstractor reviews the issue and notes a resolution (either corrects the information or notes that no correction is needed).  It is important to record and review problems encountered and the corrective steps taken so that you can discover what happened and why.  This is especially important if you need to conduct additional abstraction training or train new abstractors.

Data Issue Reporting Form

Abstraction Process Evaluation

Feedback from your abstractors can be very important for evaluating how well the data abstraction process is working at your agency.  The Abstraction Process Evaluation Form can be used to identify systemic issues that need addressing (complete only one form per abstraction team) such as the need to streamline a documentation process, or to identify individual issues (each abstractor completes a form) such as the need for additional training on the AIM Data Dictionary.  Because the information gathered on the Abstraction Process Evaluation Form is used only for learning purposes, evaluations should be customized to fit the needs of your agency.

It is recommended that evaluations are conducted every three months for the first year of data collection and then every six months thereafter.

Data Abstraction Process Evaluation


Agency Level Data

Three of the quality measures are calculated on the entire population of patients served by the agency.

Quality Measure 10 is measured using data from two of four domains (Provide Information about Symptoms and Inform & Communicate about Patient) found in the Family Evaluation for Hospice Care (FEHC); a post-death survey that assesses the family's perceptions of the quality of care provided by the agency.

Quality Measures 11 and 12 are calculated on data collected in adverse event logs recorded by the agency for all patients served within a given period of time.  Only when agencies use the same definitions and specifications for the adverse events can agencies compare data among agencies.

Adverse Events and FEHC Data Collection Tool Instructions
Family Evaluation for Health Care Data Collection Tool
Adverse Events Data Collection Tool
Recommended Definitions for Adverse Events -

Assess Reliability

Reliability is of particular concern with data collection.  Reliability looks at whether multiple people can obtain the same values using the same questions and the same clinical record.  Testing for reliability is an important and needed step in the data collection process.  Without reliability testing, you will never know if your data are accurate or "true".

The Importance of Testing for Reliability

AIM Reliability Program: Instructions for AIM Reliability Program


References and Resources



Return to top
Page last modified: November 18, 2010
IPRO Staff Access