Research

Content Analysis of HIV Prevention Drug - PrEP (an NIH grant funded research study)

Problem Description:

This research study aims to conduct a comprehensive content analysis of the HIV prevention drug, PrEP, as discussed on social media platforms such as Instagram, Facebook, and Twitter. The goal is to understand public sentiment, awareness, and perception regarding PrEP, which can have significant implications for organizations like the National Institutes of Health (NIH) and the government in promoting HIV prevention strategies. The study is funded by NIH.

Approach:

Data Collection: The first step involves collecting data from Instagram, Facebook, and Twitter using specific keywords related to HIV, PrEP, and AIDS. A total of 1000 records are collected from each platform, comprising information such as user details, post characteristics, and engagement metrics. An example from where the data (instagram) is pulled is given above.
Data Preprocessing: Clean and preprocess the collected data, including text normalization, removing duplicates, and handling missing values.
Feature Engineering: Create additional variables based on the collected data. These variables include categorizing posts into different types (photo, infographic, video, etc.) and classifying accounts into various categories (individual, organization, healthcare provider, etc.).

Source Characteristics - Individual, Organizations (Eg., Individual - Mother, Doctor; Organization - Business, Non-profit Organization, Government, etc.)
Content - Information about PrEP (Eg., Define, Side effects, How PrEP works, How to get, Cost, etc.)
Content - Source Attribution (Eg., CDC info, Political info, Gilead info, Doctor info, WHO info, etc.)
Content - Personal Account (Eg., Personal, Personal age, Hashtags mentioned in the post, Personal MSM)
Other concepts (Eg., Popularity, Credibility, Trialability, Network size; Intention, Attitude, Self - Efficacy)

Exploratory Data Analysis (EDA):

Key analyses include:

- Distribution of post types, account types, and user demographics.
- Temporal trends in PrEP-related posts.
- Sentiment analysis of user comments and engagement.
- Identification of influential accounts and common hashtags.

Machine Learning and NLP Analysis (In progress):

Utilize Natural Language Processing (NLP) and machine learning techniques to conduct sentiment analysis on the collected social media data. The best library for sentiment analysis, such as NLTK, TextBlob, or VADER, will be chosen based on performance.

Sentiment Classification: Train sentiment classifiers to categorize social media posts into positive, negative, or neutral sentiment categories. This helps in understanding public sentiment towards PrEP.
Insights and Recommendations: Based on the sentiment analysis results, draw insights regarding how people perceive PrEP. This information can be crucial for NIH and government organizations in the following ways:
- Assessing public trust and perception of PrEP.
- Identifying areas where awareness needs to be increased.
- Addressing potential issues related to stigma.
- Tailoring awareness campaigns and strategies to the sentiment detected.

Expected Results and Outcome:

The study will provide several valuable outcomes:

Sentiment Analysis: The sentiment analysis will reveal whether public sentiment towards PrEP is predominantly positive, negative, or neutral. This information can be used to gauge the success of current awareness campaigns and identify areas for improvement.
Awareness and Trust: The research will provide insights into the level of awareness and trust in PrEP among the public. This can inform strategies to enhance trust and promote its use.
Stigma Evaluation: By analyzing mentions of stigma-related topics, the study can shed light on the extent of stigma associated with PrEP. Strategies can then be devised to reduce stigma.
Recommendations: The findings will help NIH and government organizations make data-driven decisions. They can tailor their interventions and awareness campaigns to address specific issues identified through sentiment analysis.
Other Applications: The research outcomes can also be useful for Gilead, the manufacturer of PrEP, to refine their marketing and awareness strategies. Additionally, the data can be shared with other relevant stakeholders such as healthcare providers and advocacy groups for informed decision-making.

In summary, this NIH-funded research project aims to provide a comprehensive understanding of how PrEP is perceived and discussed on social media platforms, with the ultimate goal of improving awareness, trust, and utilization of this critical HIV prevention tool.

Page updated

Report abuse