# AI Predictive Adherence Counseling (South Africa)

### Countries: <img src="/files/vzKLa7aqZasHIs3mjpE5" alt="" data-size="line">South Africa

### Intervention Description &#x20;

Model that can predict whether a patient will miss their next scheduled appointment. Patient gets a tailored intervention based on the results.

Presenter: Tom Compton, Right to Care; Kieran Sharpey-Schafer, Palindrome

[DUC Meeting Recording](https://archive.org/details/2021.09.14-duc-community-meeting-recording)

### Intervention Details

|                                                                                                           |                                                                                                                                                                                                                                    |
| --------------------------------------------------------------------------------------------------------- | ---------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------- |
| <img src="/files/ZnQzrtFUZBTR1t9FjJV5" alt="" data-size="line">Data Elements                              | <p>Viral load </p><p>Demographics </p><p>Clinical data (age, gender, viral load history, visit pattern)</p>                                                                                                                        |
| <img src="/files/QxebJwCqcBiRpZ37yN1M" alt="" data-size="line">Evidence                                   | <p>Compared to adherence scorecard Model performance:</p><p>Loss to follow up: </p><p>Accuracy – 66% </p><p>Specificity- 94% </p><p>Sensitivity -61% </p><p>Graphs were presented on tracking LTFU at different stages of care</p> |
| <img src="/files/RrTZsAks3vukmm5dKSYl" alt="" data-size="line">Technology Requirements / Interoperability | <p>Machine learning and AI </p><p>Patient risk score are used to assist workers in nudging towards intervention.</p>                                                                                                               |
| <img src="/files/KU9EhZrpfw5twoqOs3Og" alt="" data-size="line">Calculations / Algorithms                  | <p>Machine Learning Algorithm </p><p>Total Adherence Score</p>                                                                                                                                                                     |
| <img src="/files/SbcBa0OpnFf4xQTFDVhs" alt="" data-size="line">Factors to Scale                           |                                                                                                                                                                                                                                    |
| <img src="/files/LlIOE3Kb7FWAuwFgE8QV" alt="" data-size="line">Implementation Considerations              | What are the impacts of telling someone they’re high risk?                                                                                                                                                                         |
| <img src="/files/jamtpgsEuuE7Fhh3MXm6" alt="" data-size="line">Governance Considerations                  |                                                                                                                                                                                                                                    |


---

# Agent Instructions: Querying This Documentation

If you need additional information that is not directly available in this page, you can query the documentation dynamically by asking a question.

Perform an HTTP GET request on the current page URL with the `ask` query parameter:

```
GET https://guides.ohie.org/duc/hiv-treatment-continuity-technology-intervention-framework-tif/during-the-visit/proactive-adherence-counselling-interventions/ai-predictive-adherence-counseling-south-africa.md?ask=<question>
```

The question should be specific, self-contained, and written in natural language.
The response will contain a direct answer to the question and relevant excerpts and sources from the documentation.

Use this mechanism when the answer is not explicitly present in the current page, you need clarification or additional context, or you want to retrieve related documentation sections.