Cares data

The International Cardiac Arrest REsearch consortium (I-CARE) Database includes baseline clinical information and continuous electroencephalography (EEG) recordings from 1,020 comatose patients with a diagnosis of cardiac arrest who were admitted to an intensive care unit from seven academic hospitals in the U.S. and Europe. Patients were monitored with 18 bipolar EEG channels over hours to days for the diagnosis of seizures and for neurological prognostication. Long-term neurological function was determined using the Cerebral Performance Category scale.

The complete data set of annual utilization data reported by primary care clinics contains basic clinic identification information including community services, clinic staffing data, and patient and staff language data; financial information including gross revenue, itemized write-offs by program, an income statement, and selected capital project items; and information on encounters by service, principal diagnosis, and procedure codes (CPT codes). These products provide trend utilization information for primary care clinics in the form of tables and pivot tables. The primary care clinic trends resource includes information on the number of clinics by type, the number of patients (by race, ethnicity, gender and age), the number of encounters by payer source; and revenues by payer source including the average revenue per encounter.

This database is part of the National Medical Information System (NMIS). The National Health Care Practitioner Database (NHCPD) supports Veterans Health Administration Privacy Act requirements by segregating personal information about health care practitioners such as name and social security number from patient information recorded in the National Patient Care Database for Ambulatory Care Reporting and Primary Care Management Module.

The purpose of the collection of outpatient health statistics is to monitor, evaluate and plan curative and preventive health care at the primary and secondary level of health care system.

Data on outpatient statistics are an important source of information for population health monitoring indicators
and accessibility of outpatient health care activities in Slovenia. Health care providers collect data for each individual contact of the patients with the health service. It is reported by public and private healthcare providers.

Outpatient health statistics record contacts and services at general practicioners and specialist outpatient activities at the secondary level.

ONC uses the SK&A Office-based Provider Database to calculate the counts of medical doctors, doctors of osteopathy, nurse practitioners, and physician assistants at the state and count level from 2011 through 2013. These counts are grouped as a total, as well as segmented by each provider type and separately as counts of primary care providers.

PATRON is a human ethics approved program of research incorporating an enduring de-identified repository of Primary Care data facilitating research and knowledge generation. PATRON is a part of the 'Data for Decisions' initiative of the Department of General Practice, University of Melbourne. 'Data for Decisions' is a research initiative in partnership with general practices. It is an exciting undertaking that makes possible primary care research projects to increase knowledge and improve healthcare practices and policy. Principal Researcher: Jon EmeryData Custodian: Lena SanciData Steward: Douglas BoyleManager: Rachel CanawayMore information about Data for Decisions and utilising PATRON data is available from the Data for Decisions website.

The Medical Care Cost Recovery National Database (MCCR NDB) provides a repository of summary Medical Care Collections Fund (MCCF) billing and collection information used by program management to compare facility performance. It stores summary information for Veterans Health Administration (VHA) receivables including the number of receivables and their summarized status information. This database is used to monitor the status of the VHA's collection process and to provide visibility on the types of bills and collections being done by the Department. The objective of the VA MCCF Program is to collect reimbursement from third party health insurers and co-payments from certain non-service-connected (NSC) Veterans for the cost of medical care furnished to Veterans. Legislation has authorized VHA to: submit claims to and recover payments from Veterans' third party health insurance carriers for treatment of non-service-connected conditions; recover co-payments from certain Veterans for treatment of non-service-connected conditions; and recover co-payments for medications from certain Veterans for treatment of non-service-connected conditions. All of the information captured in the MCCR NDB is derived from the Accounts Receivable (AR) modules running at each medical center. MCCR NDB is not used for official collections figures; instead, the Department uses the Financial Management System (FMS).

HUD CARES Act Allocation Data

  HUD CARES Act supplemental allocation amounts - cities and counties plus non-entitlement About HUD CARES Act Allocation Data: Links to several different datasets related to CARES ACT supplemental funding, including Treasury Dept and HUD allocations for cities, counties, tribal communities and non-entitlements. An additional dataset contains allocations for HHS Provider Relief Fund COVID-19 High-Impact Payments to individual providers by city and state. Click for more detail.
  Geography Level: State, City or CountyItem Vintage: Not Available
  Update Frequency: N/AAgency: HUDAvailable File Type: Excel (Links goes to same HUD CPD dataset link as listed in HUD Housing datasets listed above) 

  Return to Other Federal Agency Datasets Page

The complete data set of annual utilization data reported by primary care clinics contains basic clinic identification information including community services, clinic staffing data, and patient and staff language data; financial information including gross revenue, itemized write-offs by program, an income statement, and selected capital project items; and information on encounters by service, principal diagnosis, and procedure codes (CPT codes). These products provide trend utilization information for primary care clinics in the form of tables and pivot tables. The primary care clinic trends resource includes information on the number of clinics by type, the number of patients (by race, ethnicity, gender and age), the number of encounters by payer source; and revenues by payer source including the average revenue per encounter.

Networks are able to update on an ongoing basis data originally added to the Veterans Health Administration Physician Productivity and Staffing initiative to ensure that it reflects current conditions. This data access link function is restricted to a limited number of Network representatives. All the available facility, Network, and National Primary Care Staff and Room Utilization reports are available. In addition key guidance documents are available to people without edit access.

States report information from two reporting populations: (1) The Served Population which is information on all youth receiving at least one independent living services paid or provided by the Chafee Program agency, and (2) Youth completing the NYTD Survey. States survey youth regarding six outcomes: financial self-sufficiency, experience with homelessness, educational attainment, positive connections with adults, high-risk behaviors, and access to health insurance. States collect outcomes information by conducting a survey of youth in foster care on or around their 17th birthday, also referred to as the baseline population. States will track these youth as they age and conduct a new outcome survey on or around the youth's 19th birthday; and again on or around the youth's 21st birthday, also referred to as the follow-up population. States will collect outcomes information on these older youth at ages 19 or 21 regardless of their foster care status or whether they are still receiving independent living services from the State. Depending on the size of the State's foster care youth population, some States may conduct a random sample of the baseline population of the 17-year-olds that participate in the outcomes survey so that they can follow a smaller group of youth as they age. All States will collect and report outcome information on a new baseline population cohort every three years.

Units of Response: Current and former youth in foster care

Type of Data: Administrative

Tribal Data: No

Periodicity: Annual

Demographic Indicators: Ethnicity;Race;Sex

SORN: Not Applicable

Data Use Agreement: https://www.ndacan.acf.hhs.gov/datasets/request-dataset.cfm

Data Use Agreement Location: https://www.ndacan.acf.hhs.gov/datasets/order_forms/termsofuseagreement.pdf

Granularity: Individual

Spatial: United States

Geocoding: FIPS Code

The Child Care and Development Fund (CCDF) provides federal money to states and territories to provide assistance to low-income families, to obtain quality child care so they can work, attend training, or receive education. Within the broad federal parameters, States and Territories set the detailed policies. Those details determine whether a particular family will or will not be eligible for subsidies, how much the family will have to pay for the care, how families apply for and retain subsidies, the maximum amounts that child care providers will be reimbursed, and the administrative procedures that providers must follow. Thus, while CCDF is a single program from the perspective of federal law, it is in practice a different program in every state and territory. The CCDF Policies Database project is a comprehensive, up-to-date database of CCDF policy information that supports the needs of a variety of audiences through (1) analytic data files, (2) a project website and search tool, and (3) an annual report (Book of Tables). These resources are made available to researchers, administrators, and policymakers with the goal of addressing important questions concerning the effects of child care subsidy policies and practices on the children and families served. A description of the data files, project website and search tool, and Book of Tables is provided below: 1. Detailed, longitudinal analytic data files provide CCDF policy information for all 50 states, the District of Columbia, and the United States territories and outlying areas that capture the policies actually in effect at a point in time, rather than proposals or legislation. They capture changes throughout each year, allowing users to access the policies in place at any point in time between October 2009 and the most recent data release. The data are organized into 32 categories with each category of variables separated into its own dataset. The categories span five general areas of policy including: Eligibility Requirements for Families and Children (Datasets 1-5) Family Application, Terms of Authorization, and Redetermination (Datasets 6-13) Family Payments (Datasets 14-18) Policies for Providers, Including Maximum Reimbursement Rates (Datasets 19-27) Overall Administrative and Quality Information Plans (Datasets 28-32) The information in the data files is based primarily on the documents that caseworkers use as they work with families and providers (often termed "caseworker manuals"). The caseworker manuals generally provide much more detailed information on eligibility, family payments, and provider-related policies than the CCDF Plans submitted by states and territories to the federal government. The caseworker manuals also provide ongoing detail for periods in between CCDF Plan dates. Each dataset contains a series of variables designed to capture the intricacies of the rules covered in the category. The variables include a mix of categorical, numeric, and text variables. Most variables have a corresponding notes field to capture additional details related to that particular variable. In addition, each category has an additional notes field to capture any information regarding the rules that is not already outlined in the category's variables. Beginning with the 2020 files, the analytic data files are supplemented by four additional data files containing select policy information featured in the annual reports (prior to 2020, the full detail of the annual reports was reproduced as data files). The supplemental data files are available as 4 datasets (Datasets 33-36) and present key aspects of the differences in CCDF-funded programs across all states and territories as of October 1 of each year (2009-2022). The files include variables that are calculated using several variables from the analytic data files (Datasets 1-32) (such as copayment amounts for example family situations) and information that is part of the annual project reports (the annual Book of Tables) but not stored in the full database (such as summary market rate survey information from the CCDF plans). 2. The project website and search tool provide access to a point-and-click user interface. Users can select from the full set of public data to create custom tables. The website also provides access to the full range of reports and products released under the CCDF Policies Data

The National Diabetes Audit (NDA) is part of the National Clinical Audit and Patient Outcomes Programme (NCAPOP) which is commissioned by the Healthcare Quality Improvement Partnership (HQIP) and funded by NHS England. The NDA is managed by NHS Digital in partnership with Diabetes UK. The NDA measures the effectiveness of diabetes healthcare against NICE Clinical Guidelines and NICE Quality Standards, in England and Wales. The NDA collects, analyses and reports data for use by primary care and specialist services, local and national commissioners to support change and improvement in the quality of services and health outcomes for people with diabetes. This data release includes the care process and treatment target measurements for 2019-20 (1st January 2019 – 31st March 2020). Data were collected during May and June 2020. The national report, scheduled for 2021, will contain commentary on the audit findings and recommendations. We will communicate to users when the publication date for this report has been finalised. GP practice participation in England and Wales has increased from 98.0 per cent in 2018-19 to 99.2 per cent in 2019-20. Diabetes specialist service participation stands at 98 services in 2019-20. For NDA 2019-20, Diabetes Eye Screening (DES) data has been collected directly from DES providers for the first time. All but one DES provider in England (Liverpool) successfully submitted data, although three providers made partial submissions. For Liverpool, eye examination information secondarily recorded in Primary Care systems has been used, which is likely to be incomplete. The new 'Retinal Screening' care process measure appears in the care process and treatment targets worksheets and also feeds into the new 'All Nine Care Processes' measure, which is reported in addition to the longstanding ‘All Eight Care Processes'. Please note that there is a potential issue with the SNOMED codes used to identify if a person has had their serum creatinine care process check. Two serum/plasma creatinine codes were removed from the NDA creatinine code set during the universal SNOMED code refresh. This has affected the measurement of creatinine care process completion in a small number of health economies, and thereby has the potential to influence the all eight/nine care process percentages for organisations/areas that still use these codes. To resolve the issue, the NDA business rules are currently being amended to add these codes back into future NDA data extractions.

The International Cardiac Arrest REsearch consortium (I-CARE) database includes baseline clinical information and continuous electroencephalography (EEG) recordings from 1,020 comatose patients with a diagnosis of cardiac arrest who were admitted to an intensive care unit from seven academic hospitals in the U.S. and Europe. Patients were monitored with 18 bipolar EEG channels over hours to days for the diagnosis of seizures and for neurological prognostication. Long-term neurological function was determined using the Cerebral Performance Category scale.

The Home Based Primary Care (HBPC) database receives and compiles data from local Hospital Based Home Care (HBHC) sanctioned programs at Veterans Affairs Medical Centers (VAMCs) that run home care programs under the Home Based Primary Care program. The primary purpose is to provide HBPC management with case mix, case load, and other performance information. The HBPC information system is referred to as HBC at the VA Austin Information Technology Center and as HBHC at the local level. The HBHC automated a paper-based system of reporting home care episodes. When an admission form is completed, an episode is opened and input into HBHC for a potential home care patient. The patient is evaluated and accepted to or rejected from the program. When a patient leaves the program for any reason an episode is closed and a discharge form completed and input into HBHC. HBHC runs a nightly extract of information within the Veterans Health Information Systems and Technology Architecture. Extractions include information on all Patient Care Encounters (PCEs) with the patient and home visits made by home care providers. Details of which provider(s) made the visit, the date, any diagnosis and any procedures performed are included. Each local application sends its data to the Austin HBC database on a monthly basis. A monthly report is prepared based on this information identifying the active cases at each VAMC. A more detailed quarterly report is produced that includes national comparisons among sites.

The Flipkart Healthcare and Food Products Dataset offers a comprehensive look into consumer behavior, purchasing trends, and popular products. Extracted from Flipkart, this dataset is perfect for businesses seeking insights into food and healthcare product markets. It allows for in-depth analysis of customer preferences, brand popularity, and product performance.

Applications:

• Market Research: Understand consumer behavior and popular products.
• Product Development: Analyze trends for innovation.
• Marketing Strategy: Optimize campaigns based on consumer data.

Data for seismic compliance in general acute care hospitals is grouped by counties. All general acute care hospital buildings are assigned a Structural Performance Category (SPC) which measures the probable seismic performance of building structural systems. Building structural systems include beams, columns, shear walls, slabs, and foundations. SPC ratings range from 1 to 5 with SPC 1 assigned to buildings that may be at risk of collapse during a strong earthquake and SPC 5 assigned to buildings reasonably capable of providing services to the public following a strong earthquake. State law requires all SPC 1 buildings to be removed from providing general acute care services by January 1, 2020, unless an approved extension has been granted, and all SPC 2 buildings to be removed from providing general acute care services by January 1, 2030. A hospital facility meets the January 1, 2030 requirements if all the general acute care buildings on campus are SPC and NPC compliant. 2030 compliant SPC ratings are either SPC 3, 4, 4D, or 5. 2030 compliant NPC ratings is NPC 5. Data is provided for both hospital facilities and hospital buildings. Data is updated approximately every two weeks.

Digitization of healthcare data along with algorithmic breakthroughts in AI will have a major impact on healthcare delivery in coming years. Its intresting to see application of AI to assist clinicians during patient treatment in a privacy preserving way. While scientific knowledge can help guide interventions, there remains a key need to quickly cut through the space of decision policies to find effective strategies to support patients during the care process.

Offline Reinforcement learning (also referred to as safe or batch reinforcement learning) is a promising sub-field of RL which provides us with a mechanism for solving real world sequential decision making problems where access to simulator is not available. Here we assume that learn a policy from fixed dataset of trajectories with further interaction with the environment(agent doesn't receive reward or punishment signal from the environment). It has shown that such an approach can leverage vast amount of existing logged data (in the form of previous interactions with the environment) and can outperform supervised learning approaches or heuristic based policies for solving real world - decision making problems. Offline RL algorithms when trained on sufficiently large and diverse offline datasets can produce close to optimal policies(ability to generalize beyond training data).

As Part of my PhD, research, I investigated the problem of developing a Clinical Decision Support System for Sepsis Management using Offline Deep Reinforcement Learning.

MIMIC-III ('Medical Information Mart for Intensive Care') is a large open-access anonymized single-center database which consists of comprehensive clinical data of 61,532 critical care admissions from 2001–2012 collected at a Boston teaching hospital. Dataset consists of 47 features (including demographics, vitals, and lab test results) on a cohort of sepsis patients who meet the sepsis-3 definition criteria.

we try to answer the following question:

Given a particular patient’s characteristics and physiological information at each time step as input, can our DeepRL approach, learn an optimal treatment policy that can prescribe the right intervention(e.g use of ventilator) to the patient each stage of the treatment process, in order to improve the final outcome(e.g patient mortality)?

we can use popular state-of-the-art algorithms such as Deep Q Learning(DQN), Double Deep Q Learning (DDQN), DDQN combined with BNC, Mixed Monte Carlo(MMC) and Persistent Advantage Learning (PAL). Using these methods we can train an RL policy to recommend optimum treatment path for a given patient.

Data acquisition, standard pre-processing and modelling details can be found here in Github repo: https://github.com/asjad99/MIMIC_RL_COACH

Information on OCFS regulated child care programs, which includes program overview information and violation history.