DEPARTMENT OF PSYCHOLOGY AND SOCIOLOGY

1
DEPARTMENT OF PSYCHOLOGY AND SOCIOLOGY COURSE
NUMBER Psych 4395 COURSE TITLE Theory and
Research in Caregiving CREDIT Three Semester
Hours BULLETIN DESCRIPTION A research-based
discussion of key issues in the study of
caregiving, including similarities and
differences among caregiving populations,
cultural influences on caregiving, and the
interplay between professional and family
caregivers. PREREQUISITES PSYCH 1101 OR 1101H
OR PSYCH 2000 FACULTY Richard Birkel,
Ph.D. rbirkel_at_canes.gsw.edu Office
Rosalynn Carter Institute building Phone
928-1234 Website www.rosalynncarter.org Of
fice hours by appointment TIME AND
LOCATION Wednesdays from 600 P.M until
840 Rosalynn Carter Institute TEXTS This
class will use readings handed out in class each
week.
2
Social Science Research

• Goal is to create knowledge of human behavior
  that can be used to better understand and
  ultimately improve the human condition
• Researchers look for order, structure and
  predictability

3
A Field of Research

• Learning occurs in a community of scholars
• Researchers publish and go to conferences and
  share their work
• Research is cumulative and builds on what others
  have already done
• A field has a growing knowledge base
• Students are mentored by senior scholars and
  become the new leaders over time.

4
Is Caregiving a field?
5
Research begins with a Question

• Is there a relationship between two things I can
  observe (measure) about people that is important
  to my field?
• Field of Education GPA and Study Time
  Variables

6
GPA x Study time

• Can we measure them?
• How accurately? Is there variability in study
  time (e.g. week to week semester to semester)?
  How stable is GPA?
• What is our sample?
• Is there a statistically significant
  relationship between the two things?

7
Association Relationship

• Are a way of saying that when one is high or low,
  the other tends to be high or low also.
• But is it always a simple and direct
  relationship?

Study time
GPA
8
What if.
High Difficulty of Courses
Study Time
GPA
9
And also.
Low Difficulty of Courses
Study Time
GPA
10
So be careful about

• Simple explanations that are always true
• Assuming something is true because it is true for
  YOU
• The best research is that which helps us
  understand what is true for which groups of
  people under which conditions.

11
At best

• Social science research findings explain only a
  portion of what we observe.
• Even the most important variables have only
  partial effects.
• Human behavior is simply too complex to explain
  fully or predict with accuracy all the time or
  even a majority of the time.

12
There are three basic types of questions that
research projects can address 1. Descriptive.
When a study is designed primarily to describe
what is going on or what exists. Telephone
surveys and polls are a good example.. they seek
only to describe the proportion of people who
engage in some behavior or hold various opinions
and are primarily descriptive in nature. For
instance, if we want to know what percent of the
population provides a particular kind of
assistance to a family member, who the family
member is, their age and condition, how the
respondent feels about certain things, the help
they say they need, etc.., we are simply
interested in describing something.
13
Relational and Comparison. When a study is
designed to look at the relationships between two
or more variables, or compare two groups on one
variable. If we wanted to know whether women or
men were more likely to provide care to a family
member, or compare rural to urban caregiving, or
look at cultural differences, then we would
essentially be looking at the relation between
gender, or location, or cultural background and
certain features of caregiving.
14
Causal/ Experimental When a study is designed
to determine whether one or more variables (e.g.
a program or treatment variable, or some
situational or individual or group experience or
characteristic) causes or affects one of more
outcome variables. If we did a study to
determine whether a particular kind of help
offered to a caregiver helped them be less
stressed, then we would essentially be studying
whether the program or intervention caused a good
health outcome. (Same if we gave students
Ritalin in order to determine if it led to more
concentrated study and higher GPA)
15
Examples of Descriptive Surveys in Caregiving are
those conducted by National Alliance on
Caregiving (NAC) http//www.caregiving.org/pubs/da
ta.htm
16
(No Transcript)
17

• Key Findings Profile of Caregivers
• We estimate that there are 44.4 million
  caregivers in the U.S. age 18
• and older who provide unpaid care to another
  adult age 18 or older
• (21 of the adult U.S. population). They are
  present in an estimated
• 22.9 million households (21 of U.S. households).
  
• Although one profile does not fit all caregivers,
  a typical caregiver
• in the U.S. is a 46-year-old female, who has some
  college education,
• works, and spends more than 20 hours a week
  providing care to her
• mother.
• More caregivers are women (61) than men (39).
• Many caregivers work either full or part-time
  while providing care
• (59). More than six in ten (62) working
  caregivers say they have
• had to make some adjustments to their work life
  as a result of their
• caregiving responsibilities.

18
Caregiving in the U.S. Appendix A Detailed
Methodology Appendix A Detailed Methodology
The national study contains the results of
6,139 interviews among adults in the U.S., of
whom 1,247 were caregivers. A total of
approximate 200 caregiver interviews each of
African American, Hispanic, and Asian caregivers
were obtained.
19
E. Definition of a Caregiver For the purposes
of this study, a caregiver was defined using two
mechanisms self report from a screening
question, and verification through reporting
assistance with at least one ADL or IADL. The
screening question used for the randomly selected
respondent to identify the presence of a
caregiver in the household is as follows In
the last 12 months, have you or anyone in your
household provided unpaid care to a relative or
friend 18 years or older to help them take care
of themselves? Unpaid care may include help
with personal needs or household chores. It
might be managing a person's finances, arranging
for outside services, or visiting regularly to
see how they are doing. This person need not
live with you. (IF YES IS THAT YOU OR SOMEONE
ELSE IN THE HOUSEHOLD?) If a caregiver did
not report providing help with at least one ADL
or IADL in our series, the respondent was
thanked for her/his participation and the
interview was terminated. For all analytic
purposes in this study, these self-reported
caregivers are included as non-caregivers in the
base of the 6,139 interviews.
20
In the last 12 months, have you or anyone in
your household provided unpaid care to a
relative or friend 18 years or older to help them
take care of themselves? Unpaid care may
include help with personal needs or household
chores. It might be managing a person's finances,
arranging for outside services, or visiting
regularly to see how they are doing. This
person need not live with you.
21
(No Transcript)
22
Question is

• Do all these people call themselves caregivers?
• Is it necessary?
• Arent these just things that family members and
  friends do?
• What is the advantage of counting so MANY
  caregivers?

23
Definitions are Important

• And can lead to different results and very
  different pictures of what is really going on.

24
Estimating the Prevalence of Informal Caregiving
in the USErin Rand-Giovannetti , Health Policy
and Management, Johns Hopkins University,
Baltimore, MDInformal caregivers are a critical
part of the long-term care system in the US.
Public health surveillance of informal caregivers
in the United States is necessary to promote
healthy aging in the community, inform public
support systems and research the economic cost of
long-term care. This study systematically reviews
approaches that national surveys have used to
measure informal caregiving to community-dwelling
disabled older adults and examines the
implications of these different approaches for
prevalence estimates of caregiving. Review of six
national surveys indicates that published
prevalence estimates of caregiving to disabled
older adults range from 3.5 million to 33.8
million. Two major methodological sources of the
variation in estimates were identified. First,
the use of activities of daily living (ADLs) and
instrumental activities of daily living (IADLs)
to define the disabled older care-recipient
yielded reduced prevalence estimates in three
surveys compared with surveys which used a
less-defined definition of disability. Second,
indentifying caregivers through the disabled-care
recipient yielding lower prevalence estimates
than surveys in which caregivers were asked to
self-identify. The identification of a disabled
care-recipient and their caregiver are
inextricably interwoven, and attention to the
measurement of these concepts is needed. Survey
designers should consider asking questions about
caregiving and disability across the
caregiving/care-receiving dyad. In addition,
policy makers who use estimates of caregiving
from national health surveys may want to consider
how the methods for identifying caregivers and
disabled care-recipients affect our understanding
of how many caregivers are available to provide
care and at what cost.
25
Review of six national surveys indicates that
published prevalence estimates of caregiving to
disabled older adults range from 3.5 million to
33.8 million. Two major methodological
sources of the variation in estimates were
identified. First, the use of activities of daily
living (ADLs) and instrumental activities of
daily living (IADLs) to define the disabled older
care-recipient yielded reduced prevalence
estimates in three surveys compared with surveys
which used a less-defined definition of
disability. Second, identifying caregivers
through the disabled-care recipient yielding
lower prevalence estimates than surveys in which
caregivers were asked to self-identify.
26
Also

• In addition, policy makers who use estimates of
  caregiving from national health surveys may want
  to consider how the methods for identifying
  caregivers and disabled care-recipients affect
  our understanding of how many caregivers are
  available to provide care and at what cost.

27
What about Caregiver Health, Well Being and
Happiness?
28
(No Transcript)
29
The CVS/pharmacy and Caring Today 2007 Caregiver
Insights Study Well- Being and Care Management
was conducted to better understand the impact of
caregiving on caregivers' lives. The survey
examined many aspects of family caregiving,
including relationships, impact on health and
emotional well- being, available resources and
medication management. According to the study,
the vast majority of participants (76 percent)
reported that, in general, they enjoyed the tasks
associated with being a family caregiver. In
addition, 54 percent indicated that they have
bonded with their care recipient more than they
had anticipated.
30
The survey indicates that caregivers' concerns
vary significantly before compared to after
becoming a caregiver. Overall the differences are
positive, with caregivers initially
underestimating the degree of bonding with the
care recipient underestimating both how
rewarding and enjoyable the caregiving experience
can be, and over-anticipating the degree they
would feel overwhelmed.
31
Survey MethodologyThe CVS/pharmacy and Caring
Today 2007 Caregiver Insights Study conducted by
Anderson Analytics is an online survey of 514
qualified adults 18 and over fielded between
April 26, 2007 and May 20, 2007. The survey
participants are from a proprietary panel and
were prescreened to qualify for the study. Margin
of error of /-4.3 at the 95 confidence
level.Question What is a proprietary panel?
32
Proprietary Panels

• Companies specializing in conducting online
  surveys over time use the same group of people
  (research panel) for one client. The survey
  sample agrees to participate in a study about the
  same or different topics over time, and usually
  receives an incentive for participation.

33
(No Transcript)
34

• Fast Facts
• More than half (55) of all caregivers feel
  isolated. Feelings of isolation increase with
  duration of caregiving.
• Fifty percent of all caregivers feel that the
  burden of caregiving is frequently or
  sometimes too much to handle.
• Almost one in five (22) have been told by a
  health professional in the last 12 months that
  they have depression, nearly twice the rate in
  the general population.
• Almost half (47) of caregivers feel it is
  very or extremely difficult for the care
  recipient to tolerate the emotional problems and
  symptoms he or she experiences.

35
(No Transcript)
36
SURVEY RESPONDENTS A total of 110,574
invitations to participate in the chronic
condition and caregiver population surveys were
sent out. Based on information from the research
firm contracted to collect the data,
approximately 2.5 (n2764) of the email
addresses were not valid, resulting in a base
rate of 107,810 invitations. A total of 19,434
responded to the invitation (18.0) and 17,879
completed the survey (16.7). Table 1 provides a
breakdown of individual samples. Of those who
responded to the invitation, 11,432 did not
qualify for any condition (10.6).
37
from the one million person sample frame for
purposes of survey administration. E-mail
invitations and reminders were sent 3000 at a
time until target sample sizes were completed.
Each invitation included a link to the online
survey and a unique 5-digit access code that
ensured that only one survey was taken from an
individual computer, and that each invited
participant only took the survey once. Only
those respondents who completed at least 80 of
the survey items were considered in the analyses.
In order to have a dataset that is
more reflective of the general US population,
several datasets were weighted by age and sex
within that condition using estimates from the
1998 National Health Interview Survey. The
sample of caregiver parents to children with
asthma was also weighted by race of the child, in
addition to age and sex. The samples of
caregivers to terminally ill and chronically ill
patients were not weighted as no
nationally-representative standard data for such
caregivers could be found.
38

• Information was collected for a total of eleven
  on-line
• populations (N9400)
• 1. Teenagers (age 13-17)
• 2. A random sample of all adults, age 18 and over
  
• 3-8 Adult populations for each of the following
  chronic conditions Arthritis, Asthma,
  Coronary Artery Disease, Depression, Type 2
  Diabetes, Hypertension
• 9. Parents of Children with Asthma
• 10. Caregivers of people who recently passed away
• 11. Caregivers of people with Alzheimers or
  other
• serious chronic conditions

39
The on-line sample was drawn from approximately
one million individuals recruited to participate
in an on-line survey panel (70 recruited via
on-line sources, and 30 using random
digit-dialing and mailed requests). The
on-line panel was stratified into groups
according to age, gender, and education, based on
U.S. Current Population Survey distributions
(2000). Random samples were sequentially drawn
40
Fact In research, our sample often determines
our conclusion
41
So which is TRUE? CVS According to the
study, the vast majority of participants (76
percent) reported that, in general, they enjoyed
the tasks associated with being a family
caregiver. In addition, 54 percent indicated that
they have bonded with their care recipient more
than they had anticipated. OR.FACTS More
than half (55) of all caregivers feel isolated.
Feelings of isolation increase with duration of
caregiving. Fifty percent of all caregivers
feel that the burden of caregiving is
frequently or sometimes too much to handle.
42
Relational and Comparison. When a study is
designed to look at the relationships between two
or more variables, or compare two groups on one
variable. If we wanted to know whether women or
men were more likely to provide care to a family
member, or compare rural to urban caregiving, or
look at cultural differences, then we would
essentially be looking at the relation between
gender, or location, or cultural background and
certain features of caregiving.
43
From the NAC Study

• Overall, female caregivers are providing more
  hours of care and a higher level of care than
  male caregivers.
• Because female caregivers give more care, it is
  not surprising that women are more likely to
  report experiencing emotional stress as a result
  of caregiving than men (40 v 26 in the highest
  levels of stress rating).

44
(No Transcript)
45
Causal/ Experimental When a study is designed
to determine whether one or more variables (e.g.
a program or treatment variable, or some
situational or individual or group experience or
characteristic) causes or affects one of more
outcome variables. If we did a study to
determine whether a particular kind of help
offered to a caregiver helped them be less
stressed, then we would essentially be studying
whether the program or intervention caused a good
health outcome.
46
(No Transcript)
47
Developmental Effect or Cohort?

• The question is, whether aging causes people to
  like the music of Frank better the data
  suggests that
• It is likely, however, that this is an historical
  or cohort effect
• Different generations have different tastes in
  music based on what they grew up with.

48
Can Research Findings Be Intentionally
Manipulated?

• There are Lies, Damn Lies and there are
• Statistics

49
In Policy Formation and Politics

• Research findings are Bullets that can be and
  are used in a war of ideas
• The bullets themselves rarely win the day (if
  ever)
• But, if used effectively, they do play a role in
  determining the outcome.

50
Being a scholar means

• Weighing the evidence
• Looking carefully at samples and measures
• Understanding the limits of each study you read