{"id":6199,"date":"2024-11-01T02:36:00","date_gmt":"2024-10-31T18:36:00","guid":{"rendered":"https:\/\/darylchow.com\/frontiers\/?p=6199"},"modified":"2024-12-27T02:43:14","modified_gmt":"2024-12-26T18:43:14","slug":"when-the-pill-model-doesnt-work","status":"publish","type":"post","link":"https:\/\/darylchow.com\/frontiers\/when-the-pill-model-doesnt-work\/","title":{"rendered":"When The “Pill Model” Doesn’t Work"},"content":{"rendered":"\n

Note: This article was originally published in Substack on 01 Nov. 202<\/em>4<\/p>\n\n\n\n

\n\n\n\n

Best practice is not always best.<\/p>\n\n\n\n

Most studies are based on the average of a given group. For instance, when a treatment effect is found, it means that the average<\/em> treatment group is better off than the average<\/em> control group. In other words, not everyone benefits.<\/p>\n\n\n\n

But let\u2019s consider the following absurd example: if you take five 10 year old, and take another five 50 year old and gather all ten of them, tally the average age, and you do everything to cater for 30-year-olds, you\u2019d quickly realise that there is no such person in the room.<\/p>\n\n\n\n

Stepping out of the roomful of kids and mid-life audience, in Chapter 3: Client Factors<\/strong> of our edited book, The Field Guide to Better Results<\/em>, Joshua Swift and Jesse Owen highlighted a good example of when we fall into the trappings of ecological fallacy, that is, whenever inferences are made about an individual based on aggregate data from a group.<\/p>\n\n\n\n

\n

For example, a research study could be conducted with 100 female clients over the age of 70 with generalised anxiety disorder. After the administration of a specific 12-week treatment, 85% of them show clinically significant change.<\/strong><\/p>\n\n\n\n

Now, a 75-year-old woman with generalised anxiety disorder comes into a therapist\u2019s office seeking help, and the therapist decides to use that research-supported 12-week treatment. Some might guess that the chance of treatment success for this woman is 85%; however, her chance of recovery is actually unknown. [emphasis mine]<\/strong><\/p>\n\n\n\n

There is no way to tell if she had been in the research study on the treatment, if she would have been one of the 85% who recovered or one of the 15% who did not.<\/p>\n<\/blockquote>\n\n\n\n

This scenario usually stumps most therapists.<\/p>\n\n\n\n

Swift and Owen explains:<\/p>\n\n\n\n

\n

If the same therapist were to apply this treatment to 100 women over the age of 70 over the course of a year, we could assume that around 85% of them would recover and 15% would not. But, we would have no idea which individual woman would fall into which category.<\/p>\n\n\n\n

And so it is critical that we make decisions based on the individual client and not the client disorder, age, gender, ethnicity, sexuality, or any other demographic or group characteristic or any combination of such.<\/p>\n<\/blockquote>\n\n\n\n

As a word of caution, Nassim Taleb says,<\/p>\n\n\n\n

\n

Don\u2019t cross a river if it is (on average<\/em>) four feet deep.<\/p>\n<\/blockquote>\n\n\n\n

\n\n\n\n
\n
Check out Measured Course<\/a><\/div>\n<\/div>\n\n\n\n
\n\n\n\n

Pill Model vs Continuous Calibration Model<\/h2>\n\n\n\n

As spelled out in the video above, let\u2019s quickly look at the difference between two ways of thinking, what I call the Pill Model vs. The Continuous Calibration Model.<\/p>\n\n\n

\n
\"\"\/<\/a><\/figure><\/div>\n\n\n

Pill Model<\/strong><\/h3>\n\n\n\n

In the curriculum design of training mental health professionals, I believe we adopted a limiting mental model called the pill model.<\/p>\n\n\n\n

The pill model says:<\/p>\n\n\n\n