Accepted Manuscript Patient Satisfaction and Perceived Quality of Care Among Older Adults According to Activity Limitation Stages Hillary R. Bogner, MD, MSCE, Heather F. de Vries McClintock, MSPH, MSW, PhD, Sean Hennessy, PharmD, PhD, Jibby E. Kurichi, MPH, Joel E. Streim, MD, Dawei Xie, PhD, Liliana E. Pezzin, PhD, JD, Pui L. Kwong, MPH, Margaret G. Stineman, MD PII:
S0003-9993(15)00485-2
DOI:
10.1016/j.apmr.2015.06.005
Reference:
YAPMR 56230
To appear in:
ARCHIVES OF PHYSICAL MEDICINE AND REHABILITATION
Received Date: 17 March 2015 Revised Date:
14 May 2015
Accepted Date: 2 June 2015
Please cite this article as: Bogner HR, de Vries McClintock HF, Hennessy S, Kurichi JE, Streim JE, Xie D, Pezzin LE, Kwong PL, Stineman MG, Patient Satisfaction and Perceived Quality of Care Among Older Adults According to Activity Limitation Stages, ARCHIVES OF PHYSICAL MEDICINE AND REHABILITATION (2015), doi: 10.1016/j.apmr.2015.06.005. This is a PDF file of an unedited manuscript that has been accepted for publication. As a service to our customers we are providing this early version of the manuscript. The manuscript will undergo copyediting, typesetting, and review of the resulting proof before it is published in its final form. Please note that during the production process errors may be discovered which could affect the content, and all legal disclaimers that apply to the journal pertain.
ACCEPTED MANUSCRIPT
Running Head: Patient Satisfaction Among Older Adults Patient Satisfaction and Perceived Quality of Care Among Older Adults According to
Hillary R. Bogner, MD, MSCE1,2
RI PT
Activity Limitation Stages
Heather F. de Vries McClintock, MSPH, MSW, PhD1,2 Sean Hennessy, PharmD, PhD2,3
Joel E. Streim, MD4
M AN U
Dawei Xie, PhD2
SC
Jibby E. Kurichi, MPH2
Liliana E. Pezzin, PhD, JD5 Pui L. Kwong, MPH2
Margaret G. Stineman, MD6
Department of Family Medicine and Community Health, University of Pennsylvania,
Philadelphia, Pennsylvania 2
TE D
1
Clinical Epidemiology and Biostatistics, Perelman School of Medicine, University of
3
EP
Pennsylvania, Philadelphia, Pennsylvania
Center for Pharmacoepidemiology Research and Training, University of Pennsylvania,
4
AC C
Philadelphia, Pennsylvania
Geriatric Psychiatry Section of the Department of Psychiatry, Perelman School of Medicine,
University of Pennsylvania, Philadelphia, Pennsylvania and VISN 4 Mental Illness Research Education and Clinical Center (MIRECC), Philadelphia VA Medical Center, Philadelphia, Pennsylvania 5
Department of Medicine, Medical College of Wisconsin, Milwaukee, Wisconsin
1
ACCEPTED MANUSCRIPT
6
Department of Physical Medicine and Rehabilitation, Perelman School of Medicine, University
of Pennsylvania, Philadelphia, Pennsylvania
RI PT
Disclosure of Interest: The authors have no financial or any other kind of conflicts of interest to declare.
SC
Acknowledgments:
This work was supported through a Patient-Centered Outcomes Research Institute PCORI
M AN U
Project Program Award AD-12-11-4567 and by the National Institutes of Health (R01AG040105 and R01HD074756). Dr. Bogner was supported by an American Heart Association Award #13GRNT17000021, a National Institute of Mental Health R21 MH094940, and a National
Corresponding author:
TE D
Institute of Mental Health R34 MH085880.
Hillary R. Bogner, MD MSCE
Department of Family Medicine and Community Health
EP
Center for Clinical Epidemiology and Biostatistics
AC C
928 Blockley Hall University of Pennsylvania Perelman School of Medicine Philadelphia, PA 19104 Email: [email protected] Phone: 215-746-4181
Reprints are not available from the authors.
2
ACCEPTED MANUSCRIPT
1
Patient Satisfaction and Perceived Quality of Care Among Older Adults According to
2
Activity Limitation Stages
AC C
EP
TE D
M AN U
SC
RI PT
3
1
ACCEPTED MANUSCRIPT
4
ABSTRACT
5 OBJECTIVE: To examine whether patient satisfaction and perceived quality of medical care was
7
related to stages of activity limitations among older adults.
8
DESIGN: Cross-sectional study.
9
SETTING: Medicare Current Beneficiary Survey (MCBS) for calendar years 2001-2011.
RI PT
6
PARTICIPANTS: A population-based sample (n= 42,584) of persons 65 years of age and older
11
living in the community.
12
INTERVENTIONS: Not applicable.
13
MAIN OUTCOME MEASURE(S): MCBS questions were categorized under 5 patient
14
satisfaction and perceived quality dimensions: care coordination and quality, access barriers,
15
technical skills of primary care physicians, interpersonal skills of primary care physicians, and
16
quality of information provided by primary care physicians. Persons were classified into a stage
17
of activity limitation (0-IV) derived from self-reported difficulty levels performing activities of
18
daily living (ADLs) and instrumental activities of daily living (IADLs).
19
RESULTS: Compared to older beneficiaries with no limitations at ADL Stage 0, the adjusted
20
odds ratios (OR) (95% confidence intervals (CI)) for Stage I (mild) to Stage III (severe) for
21
satisfaction with care coordination and quality ranged from OR = 0.85 (95% CI: 0.80-0.92) to
22
OR = 0.79 (95% CI: 0.70-0.89). Compared to ADL Stage 0, satisfaction with access barriers
23
ranged from OR = 0.81 (95% CI: 0.76-0.87) at Stage I (mild) to a minimum of OR = 0.67 (95%
24
CI: 0.59-0.76) at Stage III (severe). Similarly, compared to older beneficiaries at ADL Stage 0,
25
perceived quality of the technical skills of their primary care physician ranged from OR = 0.87
AC C
EP
TE D
M AN U
SC
10
2
ACCEPTED MANUSCRIPT
(95% CI: 0.82-0.94) at Stage I (mild) to a minimum of OR = 0.81 (95% CI: 0.72-0.91) at Stage
27
III (severe).
28
CONCLUSIONS: Medicare beneficiaries at higher stages of activity limitation although not
29
necessarily the highest stage of activity limitation reported less satisfaction with medical care.
30
KEY WORDS: Disabled persons; Medicare; Patient satisfaction; Elderly
AC C
EP
TE D
M AN U
SC
RI PT
26
3
ABBREVIATIONS
32
activities of daily living (ADLs)
33
instrumental activities of daily living (IADLs)
34
Medicare Current Beneficiary Survey (MCBS)
35
Primary Care Physician (PCP)
36
odds ratios (ORs)
37
confidence intervals (CIs)
SC
31
RI PT
ACCEPTED MANUSCRIPT
AC C
EP
TE D
M AN U
38
4
ACCEPTED MANUSCRIPT
39
INTRODUCTION
40 41
Patient satisfaction is an important indicator of quality of care1 and has been demonstrated to provide useful insights for delivering efficient care that meets patient needs.2
43
In some prior studies, satisfaction has been related to a single type of disabling impairment (i.e.,
44
hearing impairment, mental disability, and other chronic conditions).3-6 Other work has examined
45
the relationship between dissatisfaction and counts of limitations of activities of daily living
46
(ADLs). These investigators have found that patients were more likely to report dissatisfaction
47
with the overall quality of their health care as their number of activity restrictions increased after
48
adjusting for sociodemographic, behavioral, and system characteristics.7
SC
M AN U
49
RI PT
42
However, counts of ADL or instrumental activity of daily living (IADL) limitations do not specify which activities are limited.8-10 In contrast, we sought to examine ADL and IADL
51
limitations separately by defined stages that specify the activities older persons must be able to
52
do without difficulty.11,12 Stages represent both the severity and types of limitations experienced
53
and specify clinically meaningful patterns of increasing difficulty with self-care skills. By
54
distinguishing activities older people are still able to do without difficulty from those that they
55
find difficult, stages enhance opportunities for discourse about specific strategies for reducing
56
disparities. No research to date has examined perceptions of care according to ADL and IADL
57
activity limitation stages.
EP
AC C
58
TE D
50
Older adults with disabilities have difficulty obtaining healthcare services despite the
59
availability of efficacious treatment. As individuals advance in age, their functional capabilities
60
decline increasing the complexity of their medical care and influencing satisfaction.13 Our goal
61
was to examine patient satisfaction and perceived quality of medical care among community-
5
ACCEPTED MANUSCRIPT
dwelling persons 65 years of age and older with differing ADL and IADL activity limitation
63
stages using data from the Medicare Current Beneficiary Survey (MCBS), a systematic,
64
representative sample. The objectives of the present study were 1) to describe how older
65
Medicare beneficiaries assess satisfaction with care, access, and physician quality categorized
66
under 5 patient satisfaction and perceived quality dimensions: care coordination and quality,
67
access barriers, technical skills of primary care physicians, interpersonal skills of primary care
68
physicians, and quality of information provided by primary care physicians; and 2) to assess
69
whether satisfaction with care, access, and perceived physician quality were related to stages of
70
activity limitation. We hypothesized that persons 65 years of age and older at higher stages of
71
activity limitation would report less satisfaction with care coordination and quality, greater
72
access barriers, and lower perceived technical skills, interpersonal skills and quality of
73
information provided by primary care physicians compared to persons with no activity
74
limitations.
SC
M AN U
TE D EP AC C
75
RI PT
62
6
ACCEPTED MANUSCRIPT
76
METHODS
77
79
Study Sample from the Medicare Current Beneficiary Survey (MCBS) Our study is cross-sectional using MCBS questions. The study sample is derived from the
RI PT
78
MCBS, using data from calendar years 2001-2011 (n=42584). The MCBS is conducted by the
81
Centers for Medicare and Medicaid Services through in-person beneficiary or proxy
82
interviews.14-16 The majority of the sample responded for themselves, but 11.6% used proxies.
83
Reasons for proxy use included hospitalization, institutionalization (temporarily), language
84
problems, lack of mental or physical capability, absence of medical records, preference for the
85
proxy to answer, or unavailability. Respondents are sampled from the Medicare enrollment file
86
and the oldest old (80 and over) are oversampled to permit more detailed analyses of this sub-
87
population.17 The sample was weighted to be representative of the fee-for-service and Health
88
Maintenance Organization Medicare beneficiaries 65 years of age and older who were living in
89
the community. The MCBS applies a rotating panel survey design whereby a panel is followed
90
for 12 interviews. We used the Access to Care files at the time of their first interview. This study
91
was approved by the Institutional Review Board at the University of Pennsylvania.
94
M AN U
TE D
EP
93
Patient Satisfaction and Perceived Quality
AC C
92
SC
80
In the MCBS, a series of questions was used to measure satisfaction with care, access,
95
and perceived physician quality. These questions were categorized under the 5 patient
96
satisfaction and perceived quality dimensions listed below. Each dimension is comprised of
97
patient or proxy answers to several questions rated either as: “very satisfied, satisfied,
98
dissatisfied, or very dissatisfied” or as: “strongly agree, agree, disagree, or strongly disagree”
7
ACCEPTED MANUSCRIPT
99
during the patient’s baseline interview. Persons reporting “no experience” with an item were counted as a missing value for only that particular item. The patient satisfaction and perceived
101
quality dimensions and component questions address: 1: Care Coordination and Quality: 5
102
questions on overall quality of care, information given, follow-up care, concern for overall
103
health, and needs met at same location. 2: Access Barriers: 3 questions on availability of
104
night/weekend care, ease/convenience, and out-of-pocket costs. 3: Technical Skills of Primary
105
Care Physician (PCP): 3 questions about checking “everything”, understanding medical
106
history, and understanding what is wrong. 4: Interpersonal Skills of PCP: 4 questions about
107
whether or not the PCP is in a hurry, does not explain, does not discuss, and acts like doing a
108
favor. 5: Quality of Information provided by PCP: 4 questions about whether the PCP “tells
109
all I want to know,” “answers all my questions,” “has my confidence,” and dependence on
110
doctor(s) to “feel better both physically and emotionally.” Categorizing these questions under the
111
5 patient satisfaction and perceived quality dimensions is supported by factor analysis.18 Patient
112
satisfaction and perceived quality dimensions 1 and 2 were available for all beneficiaries, but
113
PCP patient satisfaction and perceived quality dimensions were available only for 95.1% of
114
beneficiaries who have regular PCPs. The responses to the three to five questions under each
115
patient satisfaction and perceived quality dimension, values ranging from 1 to 4, were used to
116
calculate an average summated score. The average summated score corresponding to the upper
117
quartile was considered highly satisfied for patient satisfaction dimensions and highly favorable
118
for perceived quality dimensions. For each patient satisfaction and perceived quality dimension,
119
the highest quartile was compared to the 3 lower quartiles similar to prior work.19 Figure 1
120
contains the questions about patient satisfaction with care, access, and perceived physician
121
quality categorized by the 5 patient satisfaction and perceived quality dimensions.
AC C
EP
TE D
M AN U
SC
RI PT
100
8
ACCEPTED MANUSCRIPT
122 123
Stages of Activity Limitation Stages of activity limitation have been described in detail elsewhere.20 Briefly, stages of
125
activity limitation were built to reflect the International Classification of Functioning, Disability
126
and Health. These patient-centered outcomes were derived from patient- or proxy-reported
127
answers to simple questions about difficulties performing basic activities. The ADL and IADL
128
stages are ordered based on the probability that people with various levels of difficulty (i.e.,
129
none, mild, moderate, severe, or complete) describe difficulty performing each activity.21 The
130
ADLs were eating, toileting, dressing, bathing/showering, getting in or out of bed/chairs, and
131
walking. The IADLs were using the telephone, managing money, preparing meals, doing light
132
housework, shopping for personal items, and doing heavy housework. For each of the ADLs and
133
IADLs, respondents are asked, “Because of a health condition, do you (or does the person you
134
are answering for) have difficulty with…?”22,23
135
137
Covariates
Sociodemographic characteristics included age (65-74, 75-84, and 85 and over24); gender;
EP
136
TE D
M AN U
SC
RI PT
124
race/ethnicity (non-Hispanic African American or Black, Hispanic, non-Hispanic White, and
139
other25); education (high school diploma or greater/less than high school diploma26); income
140
(≤$25,000/>$25,00027); supplemental insurance type (Medicare only, Medicare and Medicaid-
141
dual enrollee, private-supplemental, other (i.e., Champus, VA)); living arrangement (lives alone,
142
with spouse, with children, with other relatives or non-relatives, or in a retirement community);
143
and the presence or absence of accessibility features in the home. Health status was captured by
144
self-reported chronic conditions and impairments.
AC C
138
9
ACCEPTED MANUSCRIPT
145 146 147
Analysis Our analysis proceeded in two phases. First, a descriptive analysis was applied to estimate the distribution of baseline ADL and IADL stages, sociodemographic characteristics,
149
and self-reported health conditions and impairments. Second, logistic regression models were
150
run to determine whether the five patient satisfaction and perceived quality dimensions differed
151
by ADL and IADL stages. ADL stages (Stage 0 as reference) and IADL stages (Stage 0 as
152
reference) were analyzed separately. For each patient satisfaction and perceived quality
153
dimension, we performed a logistic regression using the specific dimension as the dependent
154
variable and sociodemographic characteristics (age, gender, race/ethnicity, education, and annual
155
income), supplemental insurance, living arrangement, home accessibility features, proxy status,
156
self-reported health conditions and impairments as explanatory variables. Our measures of
157
association were adjusted odd ratios with 95% confidence intervals. We tested collinearity
158
between the variables in the final models applying criteria established by Belsley et al.28 and
159
illustrated by Mason.29 According to this criteria, harmful collinearity is found when a condition
160
index is larger than 20. P-values were two-sided, with statistical significance at p
S0003-9993(15)00485-2
DOI:
10.1016/j.apmr.2015.06.005
Reference:
YAPMR 56230
To appear in:
ARCHIVES OF PHYSICAL MEDICINE AND REHABILITATION
Received Date: 17 March 2015 Revised Date:
14 May 2015
Accepted Date: 2 June 2015
Please cite this article as: Bogner HR, de Vries McClintock HF, Hennessy S, Kurichi JE, Streim JE, Xie D, Pezzin LE, Kwong PL, Stineman MG, Patient Satisfaction and Perceived Quality of Care Among Older Adults According to Activity Limitation Stages, ARCHIVES OF PHYSICAL MEDICINE AND REHABILITATION (2015), doi: 10.1016/j.apmr.2015.06.005. This is a PDF file of an unedited manuscript that has been accepted for publication. As a service to our customers we are providing this early version of the manuscript. The manuscript will undergo copyediting, typesetting, and review of the resulting proof before it is published in its final form. Please note that during the production process errors may be discovered which could affect the content, and all legal disclaimers that apply to the journal pertain.
ACCEPTED MANUSCRIPT
Running Head: Patient Satisfaction Among Older Adults Patient Satisfaction and Perceived Quality of Care Among Older Adults According to
Hillary R. Bogner, MD, MSCE1,2
RI PT
Activity Limitation Stages
Heather F. de Vries McClintock, MSPH, MSW, PhD1,2 Sean Hennessy, PharmD, PhD2,3
Joel E. Streim, MD4
M AN U
Dawei Xie, PhD2
SC
Jibby E. Kurichi, MPH2
Liliana E. Pezzin, PhD, JD5 Pui L. Kwong, MPH2
Margaret G. Stineman, MD6
Department of Family Medicine and Community Health, University of Pennsylvania,
Philadelphia, Pennsylvania 2
TE D
1
Clinical Epidemiology and Biostatistics, Perelman School of Medicine, University of
3
EP
Pennsylvania, Philadelphia, Pennsylvania
Center for Pharmacoepidemiology Research and Training, University of Pennsylvania,
4
AC C
Philadelphia, Pennsylvania
Geriatric Psychiatry Section of the Department of Psychiatry, Perelman School of Medicine,
University of Pennsylvania, Philadelphia, Pennsylvania and VISN 4 Mental Illness Research Education and Clinical Center (MIRECC), Philadelphia VA Medical Center, Philadelphia, Pennsylvania 5
Department of Medicine, Medical College of Wisconsin, Milwaukee, Wisconsin
1
ACCEPTED MANUSCRIPT
6
Department of Physical Medicine and Rehabilitation, Perelman School of Medicine, University
of Pennsylvania, Philadelphia, Pennsylvania
RI PT
Disclosure of Interest: The authors have no financial or any other kind of conflicts of interest to declare.
SC
Acknowledgments:
This work was supported through a Patient-Centered Outcomes Research Institute PCORI
M AN U
Project Program Award AD-12-11-4567 and by the National Institutes of Health (R01AG040105 and R01HD074756). Dr. Bogner was supported by an American Heart Association Award #13GRNT17000021, a National Institute of Mental Health R21 MH094940, and a National
Corresponding author:
TE D
Institute of Mental Health R34 MH085880.
Hillary R. Bogner, MD MSCE
Department of Family Medicine and Community Health
EP
Center for Clinical Epidemiology and Biostatistics
AC C
928 Blockley Hall University of Pennsylvania Perelman School of Medicine Philadelphia, PA 19104 Email: [email protected] Phone: 215-746-4181
Reprints are not available from the authors.
2
ACCEPTED MANUSCRIPT
1
Patient Satisfaction and Perceived Quality of Care Among Older Adults According to
2
Activity Limitation Stages
AC C
EP
TE D
M AN U
SC
RI PT
3
1
ACCEPTED MANUSCRIPT
4
ABSTRACT
5 OBJECTIVE: To examine whether patient satisfaction and perceived quality of medical care was
7
related to stages of activity limitations among older adults.
8
DESIGN: Cross-sectional study.
9
SETTING: Medicare Current Beneficiary Survey (MCBS) for calendar years 2001-2011.
RI PT
6
PARTICIPANTS: A population-based sample (n= 42,584) of persons 65 years of age and older
11
living in the community.
12
INTERVENTIONS: Not applicable.
13
MAIN OUTCOME MEASURE(S): MCBS questions were categorized under 5 patient
14
satisfaction and perceived quality dimensions: care coordination and quality, access barriers,
15
technical skills of primary care physicians, interpersonal skills of primary care physicians, and
16
quality of information provided by primary care physicians. Persons were classified into a stage
17
of activity limitation (0-IV) derived from self-reported difficulty levels performing activities of
18
daily living (ADLs) and instrumental activities of daily living (IADLs).
19
RESULTS: Compared to older beneficiaries with no limitations at ADL Stage 0, the adjusted
20
odds ratios (OR) (95% confidence intervals (CI)) for Stage I (mild) to Stage III (severe) for
21
satisfaction with care coordination and quality ranged from OR = 0.85 (95% CI: 0.80-0.92) to
22
OR = 0.79 (95% CI: 0.70-0.89). Compared to ADL Stage 0, satisfaction with access barriers
23
ranged from OR = 0.81 (95% CI: 0.76-0.87) at Stage I (mild) to a minimum of OR = 0.67 (95%
24
CI: 0.59-0.76) at Stage III (severe). Similarly, compared to older beneficiaries at ADL Stage 0,
25
perceived quality of the technical skills of their primary care physician ranged from OR = 0.87
AC C
EP
TE D
M AN U
SC
10
2
ACCEPTED MANUSCRIPT
(95% CI: 0.82-0.94) at Stage I (mild) to a minimum of OR = 0.81 (95% CI: 0.72-0.91) at Stage
27
III (severe).
28
CONCLUSIONS: Medicare beneficiaries at higher stages of activity limitation although not
29
necessarily the highest stage of activity limitation reported less satisfaction with medical care.
30
KEY WORDS: Disabled persons; Medicare; Patient satisfaction; Elderly
AC C
EP
TE D
M AN U
SC
RI PT
26
3
ABBREVIATIONS
32
activities of daily living (ADLs)
33
instrumental activities of daily living (IADLs)
34
Medicare Current Beneficiary Survey (MCBS)
35
Primary Care Physician (PCP)
36
odds ratios (ORs)
37
confidence intervals (CIs)
SC
31
RI PT
ACCEPTED MANUSCRIPT
AC C
EP
TE D
M AN U
38
4
ACCEPTED MANUSCRIPT
39
INTRODUCTION
40 41
Patient satisfaction is an important indicator of quality of care1 and has been demonstrated to provide useful insights for delivering efficient care that meets patient needs.2
43
In some prior studies, satisfaction has been related to a single type of disabling impairment (i.e.,
44
hearing impairment, mental disability, and other chronic conditions).3-6 Other work has examined
45
the relationship between dissatisfaction and counts of limitations of activities of daily living
46
(ADLs). These investigators have found that patients were more likely to report dissatisfaction
47
with the overall quality of their health care as their number of activity restrictions increased after
48
adjusting for sociodemographic, behavioral, and system characteristics.7
SC
M AN U
49
RI PT
42
However, counts of ADL or instrumental activity of daily living (IADL) limitations do not specify which activities are limited.8-10 In contrast, we sought to examine ADL and IADL
51
limitations separately by defined stages that specify the activities older persons must be able to
52
do without difficulty.11,12 Stages represent both the severity and types of limitations experienced
53
and specify clinically meaningful patterns of increasing difficulty with self-care skills. By
54
distinguishing activities older people are still able to do without difficulty from those that they
55
find difficult, stages enhance opportunities for discourse about specific strategies for reducing
56
disparities. No research to date has examined perceptions of care according to ADL and IADL
57
activity limitation stages.
EP
AC C
58
TE D
50
Older adults with disabilities have difficulty obtaining healthcare services despite the
59
availability of efficacious treatment. As individuals advance in age, their functional capabilities
60
decline increasing the complexity of their medical care and influencing satisfaction.13 Our goal
61
was to examine patient satisfaction and perceived quality of medical care among community-
5
ACCEPTED MANUSCRIPT
dwelling persons 65 years of age and older with differing ADL and IADL activity limitation
63
stages using data from the Medicare Current Beneficiary Survey (MCBS), a systematic,
64
representative sample. The objectives of the present study were 1) to describe how older
65
Medicare beneficiaries assess satisfaction with care, access, and physician quality categorized
66
under 5 patient satisfaction and perceived quality dimensions: care coordination and quality,
67
access barriers, technical skills of primary care physicians, interpersonal skills of primary care
68
physicians, and quality of information provided by primary care physicians; and 2) to assess
69
whether satisfaction with care, access, and perceived physician quality were related to stages of
70
activity limitation. We hypothesized that persons 65 years of age and older at higher stages of
71
activity limitation would report less satisfaction with care coordination and quality, greater
72
access barriers, and lower perceived technical skills, interpersonal skills and quality of
73
information provided by primary care physicians compared to persons with no activity
74
limitations.
SC
M AN U
TE D EP AC C
75
RI PT
62
6
ACCEPTED MANUSCRIPT
76
METHODS
77
79
Study Sample from the Medicare Current Beneficiary Survey (MCBS) Our study is cross-sectional using MCBS questions. The study sample is derived from the
RI PT
78
MCBS, using data from calendar years 2001-2011 (n=42584). The MCBS is conducted by the
81
Centers for Medicare and Medicaid Services through in-person beneficiary or proxy
82
interviews.14-16 The majority of the sample responded for themselves, but 11.6% used proxies.
83
Reasons for proxy use included hospitalization, institutionalization (temporarily), language
84
problems, lack of mental or physical capability, absence of medical records, preference for the
85
proxy to answer, or unavailability. Respondents are sampled from the Medicare enrollment file
86
and the oldest old (80 and over) are oversampled to permit more detailed analyses of this sub-
87
population.17 The sample was weighted to be representative of the fee-for-service and Health
88
Maintenance Organization Medicare beneficiaries 65 years of age and older who were living in
89
the community. The MCBS applies a rotating panel survey design whereby a panel is followed
90
for 12 interviews. We used the Access to Care files at the time of their first interview. This study
91
was approved by the Institutional Review Board at the University of Pennsylvania.
94
M AN U
TE D
EP
93
Patient Satisfaction and Perceived Quality
AC C
92
SC
80
In the MCBS, a series of questions was used to measure satisfaction with care, access,
95
and perceived physician quality. These questions were categorized under the 5 patient
96
satisfaction and perceived quality dimensions listed below. Each dimension is comprised of
97
patient or proxy answers to several questions rated either as: “very satisfied, satisfied,
98
dissatisfied, or very dissatisfied” or as: “strongly agree, agree, disagree, or strongly disagree”
7
ACCEPTED MANUSCRIPT
99
during the patient’s baseline interview. Persons reporting “no experience” with an item were counted as a missing value for only that particular item. The patient satisfaction and perceived
101
quality dimensions and component questions address: 1: Care Coordination and Quality: 5
102
questions on overall quality of care, information given, follow-up care, concern for overall
103
health, and needs met at same location. 2: Access Barriers: 3 questions on availability of
104
night/weekend care, ease/convenience, and out-of-pocket costs. 3: Technical Skills of Primary
105
Care Physician (PCP): 3 questions about checking “everything”, understanding medical
106
history, and understanding what is wrong. 4: Interpersonal Skills of PCP: 4 questions about
107
whether or not the PCP is in a hurry, does not explain, does not discuss, and acts like doing a
108
favor. 5: Quality of Information provided by PCP: 4 questions about whether the PCP “tells
109
all I want to know,” “answers all my questions,” “has my confidence,” and dependence on
110
doctor(s) to “feel better both physically and emotionally.” Categorizing these questions under the
111
5 patient satisfaction and perceived quality dimensions is supported by factor analysis.18 Patient
112
satisfaction and perceived quality dimensions 1 and 2 were available for all beneficiaries, but
113
PCP patient satisfaction and perceived quality dimensions were available only for 95.1% of
114
beneficiaries who have regular PCPs. The responses to the three to five questions under each
115
patient satisfaction and perceived quality dimension, values ranging from 1 to 4, were used to
116
calculate an average summated score. The average summated score corresponding to the upper
117
quartile was considered highly satisfied for patient satisfaction dimensions and highly favorable
118
for perceived quality dimensions. For each patient satisfaction and perceived quality dimension,
119
the highest quartile was compared to the 3 lower quartiles similar to prior work.19 Figure 1
120
contains the questions about patient satisfaction with care, access, and perceived physician
121
quality categorized by the 5 patient satisfaction and perceived quality dimensions.
AC C
EP
TE D
M AN U
SC
RI PT
100
8
ACCEPTED MANUSCRIPT
122 123
Stages of Activity Limitation Stages of activity limitation have been described in detail elsewhere.20 Briefly, stages of
125
activity limitation were built to reflect the International Classification of Functioning, Disability
126
and Health. These patient-centered outcomes were derived from patient- or proxy-reported
127
answers to simple questions about difficulties performing basic activities. The ADL and IADL
128
stages are ordered based on the probability that people with various levels of difficulty (i.e.,
129
none, mild, moderate, severe, or complete) describe difficulty performing each activity.21 The
130
ADLs were eating, toileting, dressing, bathing/showering, getting in or out of bed/chairs, and
131
walking. The IADLs were using the telephone, managing money, preparing meals, doing light
132
housework, shopping for personal items, and doing heavy housework. For each of the ADLs and
133
IADLs, respondents are asked, “Because of a health condition, do you (or does the person you
134
are answering for) have difficulty with…?”22,23
135
137
Covariates
Sociodemographic characteristics included age (65-74, 75-84, and 85 and over24); gender;
EP
136
TE D
M AN U
SC
RI PT
124
race/ethnicity (non-Hispanic African American or Black, Hispanic, non-Hispanic White, and
139
other25); education (high school diploma or greater/less than high school diploma26); income
140
(≤$25,000/>$25,00027); supplemental insurance type (Medicare only, Medicare and Medicaid-
141
dual enrollee, private-supplemental, other (i.e., Champus, VA)); living arrangement (lives alone,
142
with spouse, with children, with other relatives or non-relatives, or in a retirement community);
143
and the presence or absence of accessibility features in the home. Health status was captured by
144
self-reported chronic conditions and impairments.
AC C
138
9
ACCEPTED MANUSCRIPT
145 146 147
Analysis Our analysis proceeded in two phases. First, a descriptive analysis was applied to estimate the distribution of baseline ADL and IADL stages, sociodemographic characteristics,
149
and self-reported health conditions and impairments. Second, logistic regression models were
150
run to determine whether the five patient satisfaction and perceived quality dimensions differed
151
by ADL and IADL stages. ADL stages (Stage 0 as reference) and IADL stages (Stage 0 as
152
reference) were analyzed separately. For each patient satisfaction and perceived quality
153
dimension, we performed a logistic regression using the specific dimension as the dependent
154
variable and sociodemographic characteristics (age, gender, race/ethnicity, education, and annual
155
income), supplemental insurance, living arrangement, home accessibility features, proxy status,
156
self-reported health conditions and impairments as explanatory variables. Our measures of
157
association were adjusted odd ratios with 95% confidence intervals. We tested collinearity
158
between the variables in the final models applying criteria established by Belsley et al.28 and
159
illustrated by Mason.29 According to this criteria, harmful collinearity is found when a condition
160
index is larger than 20. P-values were two-sided, with statistical significance at p
