Project Overview This dataset comprises 5,223 unique documents published online on the official websites of governors’ or health departments’ offices across all U.S. states in relation to the COVID-19 vaccination program. The dataset covers the timeframe from December 2020, when states began preparing for Phase 1a of the COVID-19 vaccination allocation program, to September 2021, when COVID-19 vaccines were widely available to all adults and frequently mandated. It is a collaborative effort between the Yale School of Medicine and Yale's Tobin Center for Economic Policy. Our aim is to archive publications from State Governors and Departments of Health across 50 U.S. states and the District of Columbia, which researchers can utilize to assess the efficacy of communication strategies employed during this period. Ultimately, we aim to support policymakers in making more informed decisions. Data and Data Collection Overview This collection comprises 5,223 unique publications from the governors’ office and the State Department of Health from 50 states and the District of Columbia, released during the period from December 1, 2020 to September 30, 2021, and collected by the research team (specifically AM) between September 2021 and July 2024. Data were collected from the respective states’ Governor’s and Department of Health websites, using search with a custom data range, in week-long increments (e.g., 12/01/2020-12/07/2020), and key words . Search results were reviewed to satisfy the following inclusion/exclusion criteria: Inclusion criteria: Publications from state-run websites ending in .gov that relate to the COVID-19 vaccination program (except New Mexico’s Department of Health which ends in .org and Minnesota’s Department of Health which ends in .us). Exclusion criteria: Publications from county-level organizations, universities, and other organizations not related to state government branches or health sectors (e.g., .org, .com); videos with no transcription posted by the source; publications with no text; publications that refer to other than COVID-19 vaccines; publications not in the English language. The included publications are organized by sources → month → week of the publication. Next, the publications were organized by the publication type (classification done by BV, FFB, PW, LVDM, AG, and AM): information from the Governor and other state officials, policy from the Governor or other state officials, information from the State Department of Health or other state health officials, policy from the State Department of Health or other state health officials, flyers (1-2 pages with primarily visual information), and milestones (publications of quantitative patterns in the form of tables or graphs only). AM and HP conducted the final quality control. The number of publications from each state’s officials, by type and by month from December 2020 to September 2021 are also listed as documentation (see file named Number_of_publications_by_state_by_month_December_2020_September_2021.csv). Selection and Organization of Shared Data The top-level-organization of all 10,446 primary files is by state, using conventional two-letter acronyms. Additionally, each item is classified both by time of publication (in folders labeled “raw”) and by type (in folders with self-explanatory labels, “Policy”, “Flyer”, “Info” and “Milestones”). Thus, each unique item appears more than once in the full deposit. A full inventory of the items is also shared in both Excel and CSV formats, containing a full list of publications with their upload dates, as well as the number of publications by state and by type, organized by month from December 2020 to September 2021. Additionally, the documentation includes this Data Narrative and an administrative README file.

To facilitate effective targeted COVID-19 vaccination strategies, it is important to understand reasons for vaccine hesitancy where uptake is low. Artificial intelligence (AI) techniques offer an opportunity for real-time analysis of public attitudes, sentiments, and key discussion topics from sources of soft-intelligence, including social media data. In this work, we explore the value of soft-intelligence, leveraged using AI, as an evidence source to support public health research. As a case study, we deployed a natural language processing (NLP) platform to rapidly identify and analyse key barriers to vaccine uptake from a collection of geo-located tweets from London, UK. We developed a search strategy to capture COVID-19 vaccine related tweets, identifying 91,473 tweets between 30 November 2020 and 15 August 2021. The platform's algorithm clustered tweets according to their topic and sentiment, from which we extracted 913 tweets from the top 12 negative sentiment topic clusters. These tweets were extracted for further qualitative analysis. We identified safety concerns; mistrust of government and pharmaceutical companies; and accessibility issues as key barriers limiting vaccine uptake. Our analysis also revealed widespread sharing of vaccine misinformation amongst Twitter users. This study further demonstrates that there is promising utility for using off-the-shelf NLP tools to leverage insights from social media data to support public health research. Future work to examine where this type of work might be integrated as part of a mixed-methods research approach to support local and national decision making is suggested.

💉 COVID-19 Vaccine Adverse Events (2020-2025): VAERS Real-World Surveillance Data This dataset offers a critical, large-scale look into the real-world safety surveillance of COVID-19 vaccines, sourced from the Vaccine Adverse Event Reporting System (VAERS). Maintained by the CDC and FDA, this collection spans the unprecedented period of mass vaccination from 2020 through 2025, providing an invaluable resource for pharmacovigilance, public health research, and regulatory decision-making.

Key Features & Challenge The dataset is a rich blend of structured and unstructured information detailing reported Adverse Drug Events (ADEs), which range from mild local reactions to severe, life-threatening complications.

Structured Data: Includes standardized symptom codes, offering a direct, quantitative view of reported reactions.

Free-Text Notes: Contains verbose, real-world symptom descriptions provided by reporters. This text is a "treasure trove" of granular context, including details on duration, intensity, and location of symptoms.

The Challenge: The structured entries are limited in scope. The free-text notes, while rich, are inherently noisy and lack standardized metadata such as clinical severity scores or age-specific pattern normalization.

Value to Data Scientists This dataset presents a significant Natural Language Processing (NLP) and Machine Learning (ML) challenge:

Extracting Context: Develop models to effectively extract critical clinical context (e.g., "headache lasting three days, severe") from the raw, non-standardized free-text notes.

Standardizing Severity: Create predictive models to assign standardized severity and age-specific risk patterns to ADEs.

Informed Decision Making: The ultimate goal is to generate actionable, timely insights for regulators, healthcare providers, and pharmaceutical companies, improving both vaccine safety monitoring and public trust.

Dive into this dataset to apply your skills in advanced data cleaning, feature engineering, and state-of-the-art NLP to solve a crucial, high-impact public health challenge.

IntroductionThe decision about vaccinating children is subject to their parents' decision. To inform strategies that support full vaccination coverage, it is important to understand the parents' vaccination attitude and tendency to act. This study aims to investigate the intention and the factors affecting parents' decision-making about vaccinating their children.MethodsA cross-sectional, self-administered online questionnaire was completed by parents of children aged 3–12 yeas in Macao between 7 March and 17 April 2022. The survey tool was informed by the Theory of Planned Behavior (TPB) which composes of the variable “intention” and three TPB constructs (Attitude, Subjective Norm, and Perceived Behavioral Control). Respondents rated their level of agreement on the construct statements using a 5-point Likert scale. Multiple linear regression analysis was used to determine if the TPB constructs were predictors of parents' intention.ResultsA total of 1,217 parents completed the questionnaire. The majority of participants were mothers (83.2%), aged 31–40 years (62.7%), having two or more children (74.1%), had at least one dose of COVID-19 vaccine (84.4%) and considered themselves knowledgeable about the vaccine (62.1%), all of which were significantly associated with the intention to vaccinate their children (all p < 0.05). Their intention varied from negative (19.1%), neutral (38.4%) to positive (42.5%). Respondents were mostly concerned about the serious side effects that the COVID-19 vaccine (mean = 3.96 ± 1.23), highly acknowledged the expectation by the school (mean = 3.94 ± 1.15) and the community (mean = 3.90 ± 1.19) of children vaccination, and rated highly the ease of making necessary arrangement (mean = 3.93 ± 1.25). In the multiple linear regression model which explained 63.5% of the variance in the intention-to-vaccinate their children, only Attitude (B = 0.52, p < 0.001) and Subjective Norm (B = 0.39, p < 0.001) were identified as strong predictors. The major reasons for not having intention were safety concerns (n = 646/699, 92.4%). Participants' most trusted local information sources were doctors (n = 682), government (n = 426) and healthcare professional organizations (n = 416).ConclusionsVaccinating children with COVID-19 vaccine is a complex decision-making for parents. A key to a successful COVID-19 vaccination program is effective communication about the safety profile and the usage experiences warranting the integration of reliable information sources across different healthcare sectors.

Context

Pakistan will receive half-a-million free doses of China's Sinopharm COVID-19 vaccine by January 31. I think we make a Covid vaccination distribution plan using this data. we have to find out the shortest route to commute the vaccine to minimize the spread of the COVID-19 and to save the tex pairs money because it needs specially designed,temperature-controlled thermal shippers, utilizing dry ice to maintain recommended storage temperature conditions of -70°C±10°C for up to 10 days unopened.

Content

Pakistan has 6,445 cities, towns, villages and administrative units that are divided among 1872 postal zip codes.

Acknowledgements

This Dataset is got from Pakistan post, Google and https://data.humdata.org/ and few private repos.

Inspiration

if anyone can combine it with other external sources to make it useable for startups and logistics companies to map their supply chain.