Review the article in depth and discuss some healthcare policies that could be adopted to overcome these nonfinancial barriers to accessing healthcare.
hpassignment_1_article.pdf
Unformatted Attachment Preview
Health Services Research
© Health Research and Education Trust
DOI: 10.1111/j.1475-6773.2011.01308.x
SPECIAL ISSUE: BRIDGING THE GAP BETWEEN RESEARCH AND HEALTH
POLICY-INSIGHTS FROM ROBERT WOOD JOHNSON FOUNDATION CLINICAL
SCHOLARS PROGRAM
Nonfinancial Barriers and Access to
Care for U.S. Adults
Jeffrey T. Kullgren, Catherine G. McLaughlin, Nandita Mitra,
and Katrina Armstrong
Objective. To identify prevalences and predictors of nonfinancial barriers that lead
to unmet need or delayed care among U.S. adults.
Data Source. 2007 Health Tracking Household Survey.
Study Design. Reasons for unmet need or delayed care in the previous 12 months
were assigned to one of five dimensions in the Penchansky and Thomas model of
access to care. Prevalences of barriers in each nonfinancial dimension were estimated
for all adults and for adults with affordability barriers. Multivariable logistic regression
models were used to estimate associations between individual, household, and insurance characteristics and barriers in each access dimension.
Principal Findings. Eighteen percent of U.S. adults experienced affordability barriers and 21 percent experienced nonfinancial barriers that led to unmet need or
delayed care. Two-thirds of adults with affordability barriers also reported nonfinancial barriers. Young adults, women, individuals with lower incomes, parents, and persons with at least one chronic illness had higher adjusted prevalences of nonfinancial
barriers.
Conclusions. Nonfinancial barriers are common reasons for unmet need or delayed
care among U.S. adults and frequently coincide with affordability barriers. Failure to
address nonfinancial barriers may limit the impact of policies that seek to expand
access by improving the affordability of health care.
Key Words. Access to care, nonfinancial barriers, health reform
The recently enacted Patient Protection and Affordable Care Act (PPACA)
seeks to increase access to health care for U.S. adults by improving the affordability of health services (Patient Protection and Affordable Care Act 2010).
To achieve this goal, the law requires private health insurance plans to allow
young adults to remain as dependents on their parents’ plans and eliminate
cost-sharing for evidence-based clinical preventive services. It will also
462
Nonfinancial Access Barriers
463
expand eligibility for Medicaid and provide lower income individuals with
subsidies for health insurance premiums and cost-sharing.
While the affordability of health care has long been recognized as a central element of access, many patients may face barriers that extend beyond
their ability to pay for services (Ahmed et al. 2001; Ngo-Metzger et al. 2003;
Barr and Wanat 2005; Fairbrother et al. 2005; Grol, Giesen, and van Uden
2006; Pathman, Ricketts, and Konrad 2006; Yang et al. 2006; Devoe et al.
2007; Probst et al. 2007; Clemans-Cope et al. 2008; Colwill, Cultice, and
Kruse 2008; Pitts et al. 2010). These nonfinancial barriers have significant
implications for the implementation of PPACA. For example, the identification of and development of plans to address common nonfinancial barriers—
particularly those that co-exist with problems affording care—could maximize
the likelihood that substantial investments in improving the affordability of
care will translate into true gains in access. On the other hand, policy makers’
inattention to prevalent nonfinancial barriers could potentially lead to adverse
consequences. Reductions in only affordability-related access barriers could
perpetuate—if not exacerbate—access disparities if certain groups disproportionately experience nonfinancial barriers. Public support for health reform
could wane among individuals who are required to purchase health insurance
but are unable to effectively access care due to remaining nonfinancial barriers.
Although nonfinancial barriers have important ramifications for the
success of PPACA and health services researchers have long recognized their
conceptual importance (Andersen and Newman 1973; Aday and Andersen
1974; Aday 1975; Penchansky and Thomas 1981; Thomas and Penchansky
1984; Friedman 1994; Andersen 1995; Gold 1998; McLaughlin and
Wyszewianski 2002), there has not been an analysis of patient-reported data
on the current extent of these barriers that policy makers would need in order
to reduce them. In this study, we sought to address this need by estimating
the prevalence of nonfinancial barriers that lead to unmet need or delayed
Address correspondence to Jeffrey T. Kullgren, M.D., M.S., M.P.H., Robert Wood Johnson Foundation Clinical Scholars, Philadelphia Veterans Affairs Medical Center, University of Pennsylvania, 1303B Blockley Hall, 423 Guardian Drive, Philadelphia, PA 19104; e-mail: kullgren@mail.
med.upenn.edu. Jeffrey T. Kullgren, M.D., M.S., M.P.H., and Katrina Armstrong, M.D., M.S.C.
E., are with the Leonard Davis Institute of Health Economics, University of Pennsylvania, Philadelphia, PA. Catherine G. McLaughlin, Ph.D., is with Mathematica Policy Research, Inc., and
the Department of Health Management and Policy, University of Michigan School of Public
Health, Ann Arbor, MI. Nandita Mitra, Ph.D., is with the Department of Biostatistics and Epidemiology, University of Pennsylvania School of Medicine, Philadelphia, PA. Katrina Armstrong,
M.D., M.S.C.E., is with the Abramson Cancer Center and the Division of General Internal
Medicine, University of Pennsylvania School of Medicine, Philadelphia, PA.
464
HSR: Health Services Research 47:1, Part II (February 2012)
care among U.S. adults, assessing how frequently adults with affordability
barriers that lead to unmet need or delayed care also experience nonfinancial
barriers, and identifying groups of adults who most frequently face nonfinancial barriers that lead to unmet need or delayed care.
METHODS
Data Source
We conducted a cross-sectional analysis of data from the 2007 Health Tracking Household Survey (HTHS) Restricted Use File. The 2007 HTHS was
conducted between April 2007 and January 2008 by the Center for Studying
Health System Change with funding from the Robert Wood Johnson Foundation and is the successor to the Community Tracking Study Household Surveys that were conducted periodically between 1996 and 2003. The 2007
HTHS used random digit dialing to collect data by telephone from 17,797
people in 9,407 households in the contiguous United States. The survey was
administered in both English and Spanish and the household response rate
was 47.2 percent (2009).
In the 2007 HTHS, one adult in each randomly selected household
provided selected data such as household income, employment status, insurance coverage, and general health status for all family members. Each adult
in each sampled household then completed a set of survey questions about
his or her own access to health care, chronic health conditions, and other
information that could not be collected reliably by proxy. In this part of the
survey, each adult respondent was asked, “During the past 12 months, was
there any time when you didn’t get the medical care you needed?” Each
adult respondent was also asked, “Was there any time during the past
12 months when you put off or postponed getting medical care you thought
you needed?” The 15,197 adults who completed these questions comprised
the analytic sample for this study.
All adult respondents who reported either unmet medical need or
delayed care were asked, “Did you not get the medical care you needed or
have delays getting medical care you needed for any of the following
reasons?” They could select from a list of prespecified reasons or provide
additional reasons that were not a part of the prespecified list. Respondents
could select as many reasons for their unmet need or delayed care as were
applicable and were not asked to ascribe primacy to any of the reasons or
rank their relative importance.
Nonfinancial Access Barriers
465
Classification of Access Barriers
We assigned reasons for unmet need or delayed care in the previous
12 months to one primary dimension in the Penchansky and Thomas model
of access to care (Penchansky and Thomas 1981; Thomas and Penchansky
1984; Kullgren and McLaughlin 2010). In the Penchansky and Thomas
framework, access to health care consists of five distinct dimensions: affordability, accommodation, availability, accessibility, and acceptability. Affordability is
the relationship of prices of services to patients’ income, ability to pay, and
existing health insurance. Accommodation is the relationship between the
manner in which the supply resources are organized to accept patients as
well as the patients’ perceptions of the appropriateness of these systems
(e.g., appointment systems and hours of operation). Availability is the relationship of the volume of existing services and resources to patients’ volume
and types of needs (e.g., the adequacy of the supply of clinicians, clinical
facilities, and specialized programs). Accessibility is the relationship between
the location of services and the location of patients (e.g., transportation
resources and travel time). Acceptability is the relationship between patients’
attitudes about personal and practice characteristics of clinicians and facilities to actual characteristics of existing clinicians and facilities (e.g., clinician
gender or ethnicity, clinic neighborhood or type), as well as clinician attitudes about acceptable personal characteristics of patients. For this study, we
also created a measure where all accommodation, availability, accessibility, and
acceptability reasons for unmet need or delayed care were classified as a nonfinancial barrier.
Reasons for unmet need or delayed care that did not describe a true
access barrier were not assigned to any Penchansky and Thomas access
dimension. For example, “other problems related to the health system” and
instances when the respondents “didn’t think the problem was serious
enough” were not assigned to an access dimension. Overall, there were seven
reasons that were not assigned to an access dimension.1 All assignments of
reasons for unmet need or delayed care to one primary access dimension—or
to no dimension at all—were agreed upon by all authors.
Statistical Analysis
We constructed nationally representative estimates by applying sample
weights provided by the Center for Studying Health System Change that
account for the sampling design and survey nonresponse. Using these
466
HSR: Health Services Research 47:1, Part II (February 2012)
weighted responses, we estimated the raw proportions (i.e., unadjusted for
any confounding factors that could influence the presence of a barrier)
and 95 percent confidence intervals of U.S. adults who in the last
12 months had barriers that led to unmet need or delayed care in each
of the five access dimensions and in any nonfinancial dimension. Next, we
estimated the raw proportion and 95 percent confidence interval of U.S.
adults with affordability barriers that led to unmet need or delayed care
who also had nonfinancial barriers that led to unmet need or delayed
care.
Finally, we used multivariable logistic regression to estimate independent associations between an a priori set of predisposing, enabling, and
need-related factors related to health care utilization and barriers in each
dimension that led to unmet need or delayed care in the previous
12 months (Andersen and Newman 1973). The main predictor variables
were age, gender, race/ethnicity, household income, employment status,
parental status, health insurance coverage, chronic illness, and health status.
Other model covariates included educational attainment, marital status,
U.S. citizenship, U.S. Census region, county metropolitan statistical area
(MSA) category, and county Primary Care Health Professional Shortage
Area (HPSA) status.
All predictor variables were operationalized as categorical variables
with mutually exclusive categories. Race and ethnicity data were collected in
categories similar to those used in the U.S. Census. Chronic illness was
defined as a respondent ever having been told by a doctor or health professional that he or she has diabetes, heart disease, chronic obstructive pulmonary disease, hypertension, cancer (other than skin cancer), depression,
asthma, or arthritis. County MSA category and Primary Care HPSA status
were obtained from the 2007 Area Resource File.
We estimated seven regression models. In the first five regressions—
one for each of the five individual access dimensions—the dependent variable was whether the respondent reported a barrier in that dimension that led
to unmet need or delayed care in the previous 12 months. The dependent
variable in the sixth regression was whether the respondent reported a barrier
that led to unmet need or delayed care in any nonfinancial dimension. In the
seventh regression, we sought to estimate associations between predictor
variables and nonfinancial barriers that led to unmet need or delayed care
among adults with affordability barriers. The dependent variable in this case
was also respondent report of a barrier that led to unmet need or delayed care
in any nonfinancial dimension. For all seven regressions, estimated parameters
Nonfinancial Access Barriers
467
are reported as adjusted prevalences where all other predictor variables are
fixed at their mean values (Graubard and Korn 1999).2 Stata 11 was used for
all statistical analyses (StataCorp 2009).
Sensitivity Analysis
We conducted a sensitivity analysis to test the robustness of our estimates of
the prevalence of barriers in each access dimension to our classification of
reasons for unmet need or delayed care by reassigning reasons that could
be classified into more than one access dimension to their next most plausible dimension. For example, we reclassified “doctor or hospital wouldn’t
accept health insurance” as an affordability barrier; “had to wait in the office
or clinic too long” as an availability barrier; “couldn’t get appointment soon
enough” as an accommodation barrier; and “caring for family members” as
no access barrier at all. After each reclassification, we then re-estimated the
prevalence of barriers for each access dimension and the nonfinancial
measure.
RESULTS
Prevalence of Barriers That Led to Unmet Need or Delayed Care in Each Access
Dimension
Table 1 shows the characteristics of the sample. Among these adults, 29.0 percent experienced unmet need or delayed care in the previous 12 months.
Table 2 presents the estimated unadjusted prevalences of reasons for unmet
need or delayed care and their correspondence to one of the five access
dimensions. Table 3 shows the estimated unadjusted prevalences of barriers
that led to unmet need or delayed care in each of the five access dimensions.
Among all adults, barriers in the affordability dimension were the most common reasons for unmet need or delayed care (18.5 percent). However, 17.5
percent of adults experienced an accommodation barrier that led to unmet need
or delayed care and 8.4 percent experienced an availability barrier. Overall,
barriers in any nonfinancial dimension (21.0 percent) were more frequent reasons for unmet need or delayed care in the previous 12 months than affordability barriers. These estimates were robust to reassignment of reasons that
could be classified into more than one access dimension to their next most
plausible dimension.3
468
HSR: Health Services Research 47:1, Part II (February 2012)
Table 1:
Sample Characteristics (n = 15,197)
Characteristic
Age
18–25 years old
26–39 years old
40–54 years old
55 years old
Gender
Female
Male
Race/ethnicity
White non-Hispanic
African American non-Hispanic
Hispanic
Other non-Hispanic
U.S. citizenship status
Citizen
Noncitizen
Education
College or greater
High school
Less than high school
Household income
<$50,000
$50,000 to < $100,000
$100,000
Employment
Not working
Part-time
Full-time
Marital status
Married
Single
Parental status
Parent
No children
Insurance status
Medicare
Private health insurance
Medicaid
Military
Uninsured
Chronic condition*
1 chronic illness
None
Health status
Percent (95% CI)
14.0 (13.0–15.1)
24.7 (23.5–25.9)
29.7 (28.6–30.8)
31.6 (30.5–32.7)
51.8 (51.0–52.6)
48.2 (47.4–49.0)
68.5 (67.0–70.0)
11.9 (10.8–13.1)
13.7 (12.5–15.0)
5.8 (5.2–6.5)
91.8 (90.8–92.7)
8.2 (7.3–9.2)
25.8 (24.8–26.8)
58.6 (57.4–59.8)
15.6 (14.5–16.7)
48.9 (47.4–50.4)
31.8 (30.4–33.2)
19.3 (18.2–20.5)
43.8 (42.7–45.0)
16.8 (16.0–17.7)
39.4 (38.3–40.4)
64.0 (62.6–65.4)
36.0 (34.6–37.4)
41.3 (39.8–42.7)
58.7 (57.3–60.2)
19.5 (18.6–20.4)
55.6 (54.2–56.9)
7.0 (6.3–7.8)
1.5 (1.2–1.8)
16.4 (15.2–17.8)
54.4 (53.1–55.6)
45.6 (44.4–46.9)
Continued
Nonfinancial Access Barriers
469
Table 1. Continued
Characteristic
Percent (95% CI)
Fair or poor
Good/very good/excellent
County metropolitan
statistical area category
Not statistical area
Micropolitan
Metropolitan
U.S. Census region
Northeast
Midwest
South
West
County primary care health
professional shortage area status
None of county
Part of county
All of county
Access problems
Unmet need
Delayed care
Unmet need or delayed care
20.8 (19.8–21.9)
79.2 (78.1–80.2)
6.0 (5.4–6.7)
9.1 (8.2–10.0)
84.9 (83.8–86.0)
16.9 (16.0–18.0)
23.1 (21.9–24.3)
36.9 (35.4–38.5)
23.1 (21.8–24.4)
16.9 (15.9–18.0)
39.6 (38.1–41.1)
43.5 (42.0–45.0)
10.0 (9.2–10.8)
26.8 (25.7–28.0)
29.0 (27.8–30.1)
*Ever told by a doctor or health professional that has diabetes, heart disease, chronic obstructive
pulmonary disease, hypertension, cancer (other than skin cancer), depression, asthma, or arthritis.
Prevalence of Nonfinancial Barriers among Adults with Affordability Barriers That
Led to Unmet Need or Delayed Care
Two-thirds of adults (66.8 percent) who experienced an affordability barrier
that led to unmet need or delayed care in the previous 12 months also experienced a nonfinancial barrier (Table 3). Among adults with affordability barriers,
coexistent accommodation (54.3 percent) and availability (28.6 percent) barriers
were more frequent than acceptability (18.6 percent) and accessibility (15.6 percent) barriers.
Adjusted Prevalences of Barriers That Led to Unmet Need or Delayed Care in Each
Access Dimension
Our estimated adjusted prevalences of barriers in each access dimension varied by individual characteristics (Table 4). For example, there were statistically significant age group differences in prevalences of barriers that led to
470
HSR: Health Services Research 47:1, Part II (February 2012)
Table 2: Estimated Unadjusted Prevalence of Reasons for Unmet Need or
Delayed Care among U.S. Adults, by Access Dimension (n = 15,197)
Reason for Unmet Need or Delayed Care*
Affordability
Worried about the cost
Health plan wouldn’t pay for the treatment
Accommodation
Too busy with work or other
commitments to take the time
Couldn’t get there when the
doctor’s office or clinic was open
Couldn’t get through on the telephone
Had to wait in the office or clinic too long
Couldn’t get off work
Caring for family members
Availability
Couldn’t get an appointment soon enough
Didn’t know where to go/couldn’t
find doctor/couldn’t use doctor of choice
Accessibility
Took too long to get to the doctor’s office
or clinic from house or work
Transportation problems
Acceptability
Doctor or hospital wouldn’t accept health insurance
Negative attitudes with doctors,
or bad experiences in getting care
Estimated
Prevalence,% (SE)
17.0 (0.5)
6.9 (0.4)
13.9 (0.4)
7.0 (0.3)
3.9 (0.3)
0.3 (0.1)
0.1 (0.04)
0.05 (0.02)
8.2 (0.4)
0.3 (0.05)
4.2 (0.3)
0.4 (0.1)
3.8 (0.2)
0.2 (0.04)
*Respondents could list multiple reasons for why they experienced unmet need or delayed care,
even within the same access dimension.
unmet need or delayed care in each access dimension. As compared wi ...
Purchase answer to see full
attachment
Recent Comments