NHATS full description - sporedata/researchdesigneR GitHub Wiki

Introduction to NHATS

Background to NHATS

The world's population is aging. Older folks will outnumber children by 2035. The likelihood of impairment rises dramatically with age. NHATS was developed to improve the quality of life and minimize impairment in older people.

The goal of NHATS is to provide a platform for scientific inquiry to guide efforts to maximize functioning and improve quality of life in older people, as well as to support the study of general disability trends and dynamics in older people, and to create a data infrastructure to enable the study of disablement at older ages and its consequences for individuals, families, and the community.

Researchers led these investigations from the Johns Hopkins Bloomberg School of Public Health and the University of Michigan's Institute for Social Research, and Westat collected data. A conceptual framework created for the study drove the development of disability-related material for NHATS. This framework combines the terminology of the WHO internal categorization system and the Nagi model of disability.

The physiological, cognitive, and sensory ability of an individual to complete a specific activity in a controlled environment is referred to as capacity. The framework also identifies or alters their actions in response to changes in capability, such as executing an activity less frequently or with the assistance of someone. This, in turn, affects an older person's capacity to carry out adequate self-care and household chores, as well as their engagement level in constructive, generative, community, social, and civic life. The framework also acknowledges that the environment might have an impact on the entire disablement process.

Content Overview includes a section on

Health conditions ask about common diagnosis hospital stays and surgeries, and common clinical concerns.

The impairments and symptoms section asks about impairments and several body systems, pain and fatigue, and sensory impairments.

Physical capacity is measured with both performance-based and self-reported items, and cognitive capacity is measured with activities relating to memory orientation and executive function.

NHATS investigated the number of elderly people who got assistance with everyday tasks such as self-care, home, and medical care. Participation in meaningful activities, as well as well-being and Utilization of assistive technology and rehabilitative services Chronic conditions, symptoms, sensory impairments, transportation, subjective and economic well-being, and demographic characteristics are all covered in the NHATS. A final month of life interview, created by NHATS, focuses on the quality of end-of-life care. Linkages to Medicare data and geographic information complement NHATS data.

NHATS Design Basics

Overview

Design issues

Participants for NHATS were drawn from the Medicare enrollment file. Medicare covers the great majority of elderly persons in the United States. Some older persons are not enrolled in Medicare, including those who delay joining, ineligible, and those who want to remain on commercial insurance after the age of 65.

NHATS 3-stage Sample Design

  • The sampling frame was first clustered into Primary Sampling Units (PSUs) based on individual counties or groups of Counties.
  • PSUs were then sampled from strata created based on a region of the country and variables reflecting. demographic composition. Within sampled PSUs, zip codes (or clusters of zip codes) were then sampled, creating the Secondary Sampling Units(SSUs).
  • Medicare beneficiaries of ages 65 and older within SSUs were sampled, with oversamples of older age groups and Black Individuals. Some of the effects of complex design include:
  • Clustering reduces the costs and increases the feasibility of data collection.
  • Oversampling increases sample sizes of some groups either by age or by Race.
  • Analyses using NHATS need to take into account these complex survey design features or estimates will be biased.

In General,

  • NHATS represents persons of ages 65 and older in the US.
  • NHATS can be used to assess both population-level trends(using repeat cross-sections) and trajectories(following one or more cohorts over time).
  • NHATS represents persons in all settings, including nursing homes and other residential care settings.
  • Sample persons residing in nursing are not administered the sample person interview, but only a facility questionnaire is completed.

NHATS and NSOC Sample Weights and Design Variables

NHATS Weights

In order to draw the NHATS, a complex design with oversampling was used. Because not everyone responded to the draw, each participant was allocated a weight for each round (taking into account different probabilities of selection and response). The use of survey weights is critical because it allows for conclusions to be made about the older Medicare population. The many types of weights and their applications are listed below.

  1. Base weights
    • Accounts only for differential probabilities of selection.
    • Useful for Analysts who wish to engage their own non-response adjustments.
  2. Tracker and Analytic Weights
    • Accounts for both differential probabilities of selection and non-response.
    • Appropriate for making national estimates.

Though Tracker and Analytic weights look similar, they differ a little. Below is the comparison between both weights.

Types of Analytic Weights

  • Final analytic weights: Provides every round and always represent a specific cross-sectional round or year in NHATS
  • Final analytic cohort weights: Starting in 2015, when the first replenishment sample was added, it allowed inferences about survivals from an earlier cohort.

For each type of weight, two weights have been produced, namely:

Full sample weight When using Taylor series linearization estimation methods, the entire sample weight should be provided using design variables.

Replicate weights NB: When making use of full sample weights, design variables are much needed to ensure correct precision of estimates

USING THE OTHER PERSON (OP) FILES IN NAHTS

NHATS OP

OP is an abbreviation for "Other person," as recognized during the NHATS interview. Interviewers may choose someone whose information is already saved on a roster from either the current or a previous round.

What is collected for Ops?

  • Demographic Information (NHATS gathers age, gender, and relationship to the sample person, as well as education for spouses/partners, children, and household members. NHATS also collects marriage status, number of children, and under-aged children).
  • Helper Information (NHATS collects data on the helper's connection to the SP if help is provided regularly, the days and hours helped, and a derived hours per month variable, as well as whether the helper is compensated (source and amount of payment).

OP files and variables

OP Files. Each OP has one observation. There are public, sensitive, and limited OP files, and each round contains all Ops for everyone in that round. Once added to the roster, an OP file cannot be withdrawn. Special Labels.

  • Since most questions in NHATS are re-asked every round, there is principally no need for special labels
  • Special labels are required when:
    1. Variables combine data for contemporary and new OP roster members.
    2. Items are collected only when a person is added to the roster.
    3. Items are only periodically updated.

USING NHATS

Public use data files

NHATS organizes files into three data annual files

  • Tracker file.

    1. Makes provision information on the sample drawn from NHATS.
    2. Includes respondents, nonrespondents, and ineligible cases.
    3. Cumulative(i.e., all persons ever sampled for NHATS.
    4. Each record has a unique spid.

  • Sample Person File.

    1. Rounds here are specific and not cumulative.
    2. Each record has a unique spid that is the same across rounds.

  • Other Person (OP) File.

    1. Each record represents a different person that has been identified in the SP Interview.
    2. Each record has two identifiers (spid and opid) used together.

Sensitive Data Files

  1. Sensitive SP Data files

    • One record for each SP
    • Includes more detailed demographic information about SP, for example, ethnicity, types of cancer, and years served as veterans
    • A round is specific and not cumulative, with each record having a unique spid that is the same across rounds.

  2. Sensitive OP Data Files

    • One record for each OP
    • Includes more detailed demographic information about OP
    • Each record has two identifiers (spid and opid)
    • Round-specific: Ops for SP is participating in NHATS in the current round
    • Cumulative aspect: Includes all OPs ever added for that SP and all prior round variables for OP

NHATS COVID-19 Supplements

  1. COVID-19.

  2. Diagnosis/tests were considered in order to limit the spread of the Virus.

  3. NHATS implemented changes in:

    • Living situation.
    • Social activities, health care.
    • Coping behaviors.
    • Caregiving to others.
    • Finances.
    • Well-being.

  4. Help received with daily activities and reason for changes.

  5. NHATS had a record of family members/ friends who helped most during the pandemic.

NSOC

NSOC Sample Design

  1. The NSOC is intended to represent family and unpaired caregivers to NHATS participants who have limits in everyday activities. It is represented by two samples: cross-sectional (which represents current caregivers for people in the current round of NHATS) and longitudinal (which represents people in the previous cycle of NHATS) (introduced in 2017, represent the experience of a cohort of caregivers from an earlier round).
  2. NSOC makes use of a Two-stage eligibility process.
    • NHATS participants are eligible if they received help with healthcare, mobility, or household activities for health or functional reasons.
    • A family of unpaid caregivers who provided help to an eligible NHATS participant. If more than five caregivers are eligible for NSOC, only five are randomly selected/sampled