Healthcare Use Case

Health Use Case

Psychological Assessment Report

Tag can be used for document automation in the Health sector, to create many different kinds of documents. This use case describes how a psychologist can use Tag to automate complex client assessment reports.

Background

One of the jobs of psychologists is conducting psychological assessments for their clients. This involves interviewing the client and administering a series of tests. The tests provide a method to assess an individual’s cognitive abilities, academic achievement, behavior, personality, mental health, and several other domains. This assessment is the mental equivalent of a physical examination.

At the end of an assessment, a report is prepared. These reports can be complex and the content varies widely according to the client’s reason for referral, background history and test results.

The challenge is made bigger by the nature of some of the tests. For example, the WIAT-III (Wechsler Individual Achievement Test) alone has more than 150 data fields which can inform clinical judgment. Each assessment can include several tests of comparable complexity.

For experienced psychologists, the writing process involves some degree of copy/paste reuse from a personal collection of templates, text fragments, and even old client reports. Typically, the psychologist relies on memory to recognize patterns of scores that they have seen before, and then looks for previous templates and reports that discuss that pattern. If a relevant fragment is found in an old client report, care must be taken to update client name, he/she references, and so on. This process is familiar to many different professions.

Writing these reports is time consuming. For a comprehensive assessment, it can take more than 8 hours to produce a 20+ page report.

Goals

  • Dramatically save time creating reports
  • Organize and reuse hundreds of small text fragments
  • Eliminate copy/paste errors
  • Automate data entry where possible
  • Keep it easy

How is Tag used?

The following lists the main steps of this use case.

The starting point for most smart content projects is the desired output. What should the generated document look like, and what kinds of text fragments will be used to create it. For this use case, a cross section of previous reports would be helpful.

There are common elements in these reports. Header/footer, title, table summarizing client information, introduction and signature block. Although some of these elements contain embedded data, for the most part they look the same in all reports.

There are also sections that correspond to information gathered during the assessment. For example, a behavioral observations section includes notes about the client (e.g., appearance, mood during assessment, effort-level). There may also be background information received from the client, a guardian or a teacher. This can be gathered through clinical interviewing, or by providing forms for someone to fill out.

Most of the body of the report relates to the tests administered. The author evaluates test scores and provides interpretations for what is discovered.

A typical assessment report may also include sections for diagnostic probabilities, summary/comments, recommendations and score tables. These sections can be challenging and time-consuming to write because they synthesize hundreds of data points into a cohesive narrative that can have thousands of different iterations.

The above discussion reveals several distinct groups of data.

  • Client information
  • Behavioral observations
  • Background information
  • Data for multiple tests
  • Diagnoses
  • Recommendations
  • Score tables

Client information is pretty standard (e.g., name, school, profession). It also includes the important gender data field, which allows you to automate personal pronouns like he/she/they.

Behavioral observations can often be described by true/false (was client anxious) or select from a list (was client rate of speech too fast, too slow, or within normal limits) data fields.

When data is gathered through clinical interviewing or direct observations, it can be manually entered directly in Tag’s auto-generated forms. When data is gathered by forms (e.g., Google Forms), it can be mapped to data fields in Tag and imported with the click of a button.

Data for the tests may be manually entered, or imported from another system wherever possible. For example, Tag can import client test results that have been exported from the Q-Global or Q-Interactive scoring systems in a CSV file.

Some assessment reports are diagnostic in nature. Cognitive and mental health disorders are diagnosed by the clinician, and must correspond to public codes recognized by their profession (e.g., The Diagnostic and Statistical Manual of Mental Disorders (DSM-5) or The ICD-10 Classification of Mental and Behavioural Disorders (ICD-10)). Diagnostic codes which are used regularly could be selected from a dropdown list and each code could generate its own text fragments in the report.

Many psychologists struggle to keep a well-sorted and up-to-date recommendations bank. In Tag, collections of recommendations addressing various client needs can be stored in easily sorted and reusable template files. These files can be automated using true/false data fields and if/then logic to populate recommendations sections that are highly tailored to the client.

For relatively complex reports like this, it helps to separate content into multiple template files which can then be mixed and matched in various ways to create an integrated report. For example, a psychologist may have a psychoeducational report template that calls the templates for the WISC-5 and the WIAT-III. They may also have a neuropsychological report template that calls templates for the NEPSY-II and the CPT-3.

Having content separated into multiple files makes integrated reports easier to maintain because changes or edits only need to be done in one template file rather than in all integrated report versions. It also increases the opportunities to share content with other reports or with colleagues.

The main report file (e.g., the integrated Psychoeducational Assessment Report or the integrated Neuropsychological Assessment Report) is where headers, footers, preamble, signatures and other report-level content are defined. This file imports all the other files (e.g., observations, tests) to make those templates visible. At the various points where major sections should go, simply call the desired interpretation fragment. Formatting is also set at this level and the selected font and style will be propagated through all the templates that are called.

Tables can be used to display dynamic content (e.g., an identification table with client personal information). Tables can be inserted with as many rows and columns as needed. In each cell you can type static content, or insert the value of an expression (e.g., client name). For example, for these instructions:

You get this generated content:

Another feature that can be helpful in these reports is dynamic lists. Dynamic lists are created using widget templates, where each possible element in the list is wrapped by a true/false condition. When the list is generated, it only contains the elements with a condition that is currently true. Layout of the list can be inline (e.g., comma separated) text, or a numbered or bullet list.

For example, if you had a cutoff number for deciding if a score is in the “high average” range (e.g., standard score >= 110), you could write a summary fragment that inserts “Harry had high average scores in score-A, score-D and score-H”.

Tag creates forms automatically to match data setup files, so data entry can occur as soon as report content exists. By default, only fields that are used in report text are displayed in forms, however this setting can be changed if more data is being collected than is actually used in the report.

The form for observations data could look something like this.

Tag has support for common statistical score types including percentage, percentile, scaled score, standard score, T-score and Z-score. Statistical score types can be automatically converted into another type based on the standard bell curve (e.g., Tag can automatically convert a standard score into a percentile or descriptive level, etc., depending on report requirements). In this way, only one score needs to be manually entered.

There is also a level of data validation protection for these scores because typos in test score data could result in serious errors. Errors in numeric fields are detected in forms by validation rules. Below shows an error for a T-score.

Data can also be imported from other systems. This is common for the larger assessment tests. Importing data mapped to a Q-Interactive data export is shown below.

Data can also be imported from form systems, like Google Forms. Google Forms can be made HIPAA compliant, and this allows the psychologist to send a form to be filled out by anyone with web access.

Data mapping is required when you import data from external form systems. The mapping only needs to be set up once and Tag will remember all the data mapping rules for the report. The mapping panel, shown below, is point and click and easy to use.

Mapping rules can be made conditional, so that data is only copied if a specified condition is true. Transformation rules can also be used to alter or replace imported data. For example, the above image shows transformation rules for the “Your relationship to the child” CSV column. If the answer contains the word ‘homeroom’ it stores ‘true’ in the “homeroom teacher” field in Tag.

Once all the data is entered, it is time to generate. Simply click the magic wand.

Generation usually creates a new word processing document and saves it to a file. Since each report data file is client specific, most psychologists store the generated *.docx file in the same folder with other files about the client, but where the data and generated document is stored is up to the psychologist.

Tag is designed to read files from your computer/network, process them securely, and ensure no copies of your files are ever stored on a web server. Therefore the location of data/report storage is up to the individual psychologist or organization, in a way that protects personal data privacy.

The generate page also allows for export of data. Data can be exported from one or more clients. This could be useful when looking to analyze the assessment practice across different sets of clients (e.g., by year, by psychometrist, etc.).

Benefits and outcomes

These are the goals for this use case.

  • Dramatically save time creating reports
  • Organize and reuse hundreds of small text fragments
  • Eliminate copy/paste errors
  • Automate data entry where possible
  • Keep it easy

With reports this complex, the time savings provided by Tag can be game-changing for the practice of psychological assessment. Turnaround time for reports is dramatically reduced for clients and referral agents. Or, better yet, the time saved in report-writing can be spent on case conceptualization and consideration of other issues relevant for the client.

Organization of templates is also a major benefit. A good naming convention goes a long way towards making templates and text fragments discoverable and easy to use and share. Templates can easily be added or edited whenever new information about client populations, tests, diagnoses, or other relevant topics surface that should be addressed.

Copy/paste and similar errors will be gone. Considering how many times a personal pronoun is used when discussing an individual’s circumstances, eliminating the possibility of these copy/paste errors will clearly improve quality and relieve the psychologist from the burden of mundane editing tasks.

Note that these errors are not uncommon. One of the most common complaints to regulatory bodies for psychologists has to do with the wrong name or misinformation in sensitive mental health documentation. In the worst-case scenario, this could lead to a breach of privacy and confidentiality.

Automation of data entry also improves quality assurance. A score that is imported from a CSV will never be transcribed incorrectly. Similarly, data imported from other forms systems can let you benefit from other stakeholders’ typing, and reduce the note taking burden of gathering background information.

The end result should be easy to use, even when dealing with a significant number of assessment tests. Once a report is setup, the forms wizard guides you through a series of forms for each client which address all aspects of the assessment.

There may also be practice-level benefits when multiple psychologists use a standardized process. Shared standard templates, recommendations bank, branding, and formatting improves consistency in reports between psychologists, allows strong writers to help out colleagues who may have weaker writing skills, and facilitates knowledge transfer between psychologists with differing areas of expertise.

Once the psychological assessment reports are up-and-running, consider other office documents that would benefit from automation: consent forms, guardianship documents, coding applications, letters to physicians, invoices and more.

 

How do I learn more?

Contact us for more information about this use case.