Document Sample

Objective: To examine sociodemographic,                         Margaret V. McDonald, MSW; Liliana E. Pezzin, PhD, JD;
clinical, and self-management characteristics                           ; Timothy R. Peng, PhD; Penny H. Feldman, PhD
of a sample of urban, African American
patients admitted to home health care with
uncontrolled hypertension and to determine
                                                   management preparedness.         (Ethn   Dis.    Examination Survey indicate that while
the extent to which these factors are associated
with disease severity.                                                                              approximately three-quarters of African
                                                   Key Words: Hypertension, African Ameri-
                                                                                                    Americans with hypertension are aware
Methods: We conducted a cross-sectional                                                             of their diagnosis, only 57% receive
                                                   can, Home Care, Diabetes, Patient Activation
study of 498 hypertensive African American                                                          hypertension treatment, and only ap-
patients newly admitted to home health care.
                                                                                                    proximately one-quarter have their
Data for this study were drawn from patient-
level clinical and functional assessment data      INTRODUCTION                                     blood pressure under control.8
derived from the uniform home health assess-                                                            The purpose of this study was to
ment system mandated by the Centers for                African Americans have dispropor-            examine sociodemographic, clinical,
Medicare and Medicaid Services and patient         tionately high rates of hypertension and         and self-management characteristics of
in-home interviews.                                poorer outcomes compared with their              a sample of urban, African American
                                                   White counterparts. An estimated 1 in 4          patients admitted to home health care
Results: Forty percent of patients had stage 1
hypertension, and 60% had the more severe          US adults has hypertension, while the            with uncontrolled hypertension and to
uncontrolled stage 2. Multivariate analyses        prevalence among non-Hispanic Blacks             determine the extent to which these
found that factors associated with stage 2 were    is more than 1 in 3.1 African Ameri-             factors are associated with disease sever-
co-morbid diabetes, poor appointment keep-         cans experience significantly higher             ity. Prospective studies have demon-
ing, low activation, and longer time since                                                          strated repeatedly that progressively
                                                   rates of hypertension-related nephropa-
diagnosis. Protective factors associated with a
lower likelihood of severe uncontrolled hyper-
                                                   thy, stroke, heart failure, type 2 diabe-        higher levels of blood pressure lead to
tension were older age and recent discharge        tes, and end-stage renal disease.2 More-         progressively increased risk of cardio-
from a hospital. More co-morbid conditions         over, the death rate per 100,000 people          vascular disease, stroke, and renal insuf-
also appeared to be protective, although the       from high blood pressure is more than            ficiency.9,10 A metaanalysis of .1
association did not reach significance.            twice as high for Blacks (40.9 for Black         million people indicated that for every
                                                   women and 51.0 for Black men in                  20 mm Hg systolic blood pressure
Conclusions: Our findings highlight the need
to address hypertension control among the          2004) as for the population as a whole           (SBP) or 10 mm Hg diastolic blood
African American, dually diagnosed diabetic        (18.1 in 2004).3                                 pressure (DBP) increase, the death rate
hypertensive population and underscore the             Data from randomized controlled              from both ischemic heart disease and
critical role of treatment adherence, widely       trials, however, indicate that African           stroke doubles.11,12 In this study, we
recognized as a key issue in managing                                                               focus on people with severe hyperten-
                                                   Americans who are prescribed and
hypertension and other chronic conditions.
                                                   adhere to appropriate medication and             sion, a group that should receive
Successful strategies will likely require more
aggressive action by home health nurses, both      dietary regimens achieve blood pressure          heightened attention to reduce the
to alert patients’ primary care providers to       control,4,5 a finding that rules out             potential for serious consequences.
ongoing, unsuccessfully treated hypertension       biological differences as an explanation
and to remediate patients’ inadequate self-        for disparate racial outcomes. This
                                                   finding implies that African Americans           METHODS
                                                   are more likely to encounter obstacles to
     From the Center for Home Care Policy          accessing appropriate hypertension care,         Study Sample
and Research, Visiting Nurse Service of            as has been found in other investigations            We evaluated a sample of consecu-
New York, New York (MVM, TRP, PHF);
                                                   of health service use in the United              tive, prospectively identified adult Afri-
Department of Medicine and Health Policy
Institute, Medical College of Wisconsin,           States.6,7 Another implication is that           can American patients with uncon-
Milwaukee, Wisconsin (LEP).                        efforts to address patient knowledge,            trolled hypertension who were newly
                                                   self-management, and barriers to hyper-          admitted to a large, Medicare-certified,
     Address correspondence and reprint            tension management can increase rates            non-profit, urban home health agency
requests to: Margaret V. McDonald, MSW;            of blood pressure control among Afri-            from February 2006 through August
Center for Home Care Policy and Research,
Visiting Nurse Service of New York; 5 Penn         can Americans, as well as the hyperten-          2007. Hypertension status was deter-
Plaza, 14th Fl; New York, NY 10001; 212-           sive population in general. Data from            mined by a primary, secondary, or
609-5761;              the National Health and Nutrition                tertiary diagnosis of hypertension (In-

148                                                   Ethnicity & Disease, Volume 19, Spring 2009
                                                                     HYPERTENSIVE HOME CARE PATIENTS - McDonald et al

ternational Classification of Diseases,     taking; each item is answered on a 4-         economic, and clinical characteristics
Ninth Revision, Clinical Modification       point Likert-type scale.13 General self-      on the likelihood of having severe,
codes 401, 402, 403, or 404) in the         management preparedness was mea-              uncontrolled (stage 2) hypertension.
home care record and SBP $140 mm            sured by using the patient activation         Sociodemographic characteristics in-
Hg or DBP $90 mm Hg ($130/80 for            measure (PAM), which assesses a per-          cluded age, sex, educational level, pri-
patients with diabetes or chronic kidney    son’s self-reported knowledge, skills,        mary source of payment for the home
disease). Trained field interviewers took   and confidence for self-management of         health stay, borough of residence, and
patients’ blood pressures in their home     his health care.14 Questions about            usual source of care (private office vs
by using a Microlife Model 3AA1-2           medication adherence barriers were            any other settings). Clinical characteris-
(Microlife USA, Inc, Dunedin, Flor-         drawn from the Brief Medication Ques-         tics included obesity (BMI .29 kg/m2),
ida), a device that uses an oscillometric   tionnaire (BMQ).15 Co-morbid condi-           co-morbid conditions, history of hospi-
algorithm validated by the British          tions were collected by using the             talization, months since hypertension
Hypertension Society. The average of        Charlson co-morbidity index, self-re-         diagnosis (and its square to control for
3 readings was used to determine study      port version, a 10-item validated scale       nonlinearities), and evidence of depres-
eligibility. We excluded patients with      that includes information on formal           sive symptoms (GDS .9). In addition
cognitive impairment, severe heart fail-    diagnosis of asthma, emphysema,               to models with this core set of variables,
ure, end-stage renal disease, life expec-   chronic bronchitis, arthritis or rheuma-      additional analyses were performed to
tancy ,6 months and those who had           tism, cancer diagnosed in the last 3          test the independent effect of self-
had a kidney transplant or were on          years, diabetes, digestive problems, heart    management and adherence behaviors
dialysis. The study was approved by the     trouble, HIV infection, kidney disease,       on the probability of severe, uncon-
appropriate institutional review boards.    liver problems, and stroke.16 Depressive      trolled hypertension. Because including
                                            symptoms were assessed by using the           these potentially endogenous variables
Data Sources and                            15-item geriatric depression scale            did not materially affect the core
Study Measures                              (GDS).17 Height and weight were               variables’ results, they were retained in
    Data were drawn from the uniform        collected to calculate body mass index        the final model.
home health assessment system mandat-       (BMI).                                            We tested the sensitivity of our
ed by the Centers for Medicare and              On the basis of blood pressure            findings by examining several variants
Medicaid Services and structured, in-       readings taken at the in-home interview       of the model. In some versions, we
home patient interviews. Information        as described above, patients were classi-     reclassified hypertension severity among
on age, sex, insurance status, health       fied according to disease severity stages     people with diabetes by using the same
status (pre-existing medical conditions),   defined in the seventh Report of the          threshold for people without diabetes.
cognitive status, and function (limita-     Joint National Committee on Preven-           In others, we included interaction terms
tions in activities of daily living) was    tion, Detection, Evaluation, and Treat-       between certain characteristics of inter-
derived from the nurse’s routine assess-    ment of High Blood Pressure, with             est (eg, age and a diagnosis of diabetes,
ment conducted during the initial visit.    additional adjustment for lower blood         age and time since hypertension diag-
Most other information came from the        pressure recommendations for people           nosis) or considered additional regres-
patient interviews conducted by trained     with diabetes. Specifically, patients were    sors, such as income and specific
research staff.                             categorized as stage 1 if their average in-   barriers to adherence (eg, ability to
    The patient interview elicited de-      home blood pressure reading was 140-          pay, medication side effects, difficulty
tailed information on patient’s blood       159/90-99 mm Hg (130-149/80-89                remembering medication doses). Our
pressure, hypertension knowledge and        mm Hg for people with diabetes or             findings were robust to these alternative
self-management preparedness, and ad-       chronic kidney disease) and stage 2, or       specifications, none of which yielded an
herence to hypertension care, including     severe, uncontrolled hypertension, if         improved model fit.
medications and dietary restrictions.       their average in-home blood pressure
Whenever possible, we employed stan-        reading was $160/100 mm Hg ($150/
dardized concepts and measures. The         90 mm Hg for people with diabetes or          RESULTS
Hill-Bone compliance to high blood          chronic kidney disease).
pressure therapy scale was used to                                                            In total, 498 adult African American
measure hypertension-related health be-     Statistical Analysis                          patients with uncontrolled hypertension
haviors. This 14-item scale poses ques-         Multivariate logistic regression mod-     were admitted to home health care
tions in 3 domains: sodium intake,          els were used to examine the indepen-         during the study period and participated
appointment keeping, and medication         dent effects of sociodemographic,             in the interviews (Table 1). Two-thirds

                                            Ethnicity & Disease, Volume 19, Spring 2009                                         149

                                                                                                                   istics according to self-management be-
Table 1. Sociodemographic and clinical characteristics of 498 African American
patients admitted to home health care with uncontrolled hypertension from February                                 haviors and self-management readiness
2006 through August 2007                                                                                           are described in Table 2.
                                                                                                                        After controlling for potential con-
                        Characteristic                                               Mean (SD) or n (%)*
                                                                                                                   founders, the probability of having severe,
Female, n (%)                                                                        342 (68.7)                    uncontrolled stage 2 hypertension was
Mean age (SD), years                                                                 64.6 (10.8)
                                                                                                                   significantly higher among persons with
Age category, n (%)
  ,45 years                                                                            21   (4.2)                  diabetes, those with poor appointment
  45–64 years                                                                         201   (40.4)                 keeping, and those in the lowest patient
  65–74 years                                                                         167   (33.5)                 activation category. Longer time since
  75–79 years                                                                         109   (22.0)
Educational level, n (%)
                                                                                                                   diagnosis also increased the likelihood of
  Less than high school graduate                                                      202 (40.8)                   severe, uncontrolled hypertension. How-
  High school graduate or higher                                                      296 (59.2)                   ever, the relationship between time since
Income, n (%)*                                                                                                     diagnosis and severe, uncontrolled hyper-
  ,$10,000                                                                            223 (52.4)
  $10,000–$29,999                                                                     146 (34.3)
                                                                                                                   tension exhibited a convex shape (ie, the
  $$30,000                                                                             57 (13.4)                   probability of severe, uncontrolled hyper-
Primary home health payer, n (%)                                                                                   tension increased with time since diagno-
  Medicaid only or dually eligible                                                    213   (42.8)                 sis up to an inflection point and decreased
  Medicare only                                                                       179   (35.9)
  Private insurance                                                                    59   (11.8)                 thereafter). Older patients and, to a lesser
  Other                                                                                47   (9.4)                  extent, those with more education and
Usual source of care, n (%)                                                                                        those admitted from a hospital were less
  Clinic                                                                              241 (48.4)
                                                                                                                   likely to have severe, uncontrolled hyper-
  Private doctor’s office                                                             229 (46.0)
  Other/no usual source of care                                                        28 (5.6)                    tension. Finally, patients with more co-
Blood pressure, mean (SD) SBP/mean (SD) DBP                                                                        morbid conditions also were less likely to
  Full sample                                                                       157.9   (21.4)/87.7 (14.9)     experience severe, uncontrolled hyperten-
  Stage 1 subsample                                                                 140.5   (8.6)/79.6 (10.0)
  Stage 1, n (%)                                                                     199    (40)
                                                                                                                   sion, although the coefficient failed to
  Stage 2 subsample3                                                                169.4   (15.2)/93.1 (15.2)     reach significance. (Table 3)
  Stage 2, n (%)                                                                     299    (60)
Mean (SD) months since hypertension diagnosis                                       169.1   (149.9)
Mean (SD) number of additional co-morbidities                                         2.1   (1.3)
Diabetes, n (%)                                                                      300    (60.2)                 DISCUSSION
Hospitalized within 14 days of admission to home care, n (%)                         330    (66.3)
Mean (SD) BMI, kg/m2                                                                 30.9   (8.3)                      Our findings highlight the need to
  Underweight, n (%)                                                                  10    (2.1)
  Healthy weight, n (%)                                                              112    (23.8)
                                                                                                                   address hypertension control among
  Overweight, n (%)                                                                  135    (28.7)                 patients who also have diabetes. People
  Obese, n (%)                                                                       213    (45.3)                 with diabetes, who made up 60% of our
Depressive symptoms, n (%)                                                                                         overall sample of African Americans with
  Normal                                                                              298 (59.8)
  Mild depression                                                                     162 (32.5)
                                                                                                                   uncontrolled hypertension and 71% of
  Moderate/severe depression                                                           38 (7.6)                    those with severe hypertension, were 3
  SD 5 standard deviation, SBP 5 systolic blood pressure, DBP 5 diastolic blood pressure, BMI 5 body mass index.   times more likely than those without
  * Information missing for 14.5%; denominator used to calculate percentages was 426.                              diabetes to have severe, uncontrolled
  3 Stage 2 hypertension is defined as blood pressure $160/100 mm Hg ($150/90 mm            Hg for people with
                                                                                                                   hypertension. Recent evidence from clin-
diabetes or kidney failure).
                                                                                                                   ical trials suggests that greater reductions
                                                                                                                   in morbidity and mortality may result
of the patients were hospitalized in the 14                conditions, most commonly diabetes                      from intensive control of blood pressure
days before home care admission: 9.4%                      (60.2%), arthritis or rheumatism                        in older people with type 2 diabetes than
were referred from another institutional                   (50.6%), or heart disease (27.0%). One                  may result from tight glycemic control.18
setting (eg, nursing home, rehabilitation                  out of 5 patients had a history of stroke.              In home care as in the primary care
center) and 24.3% were referred by a                       On average, patients were taking almost 6               setting, however, preoccupation with
community physician. On average, these                     prescribed medications, including 2 med-                managing diabetes-specific treatment
patients had been diagnosed with hyper-                    ications for hypertension. Nearly half                  may overwhelm both clinicians and
tension for 14 years. In addition to                       were obese, and an additional 28.7%                     patients and divert attention from the
hypertension, they had 2 other chronic                     were overweight. The sample’s character-                need to address other serious conditions.19

150                                                            Ethnicity & Disease, Volume 19, Spring 2009
                                                                                     HYPERTENSIVE HOME CARE PATIENTS - McDonald et al

                                                                                                               Our findings also underscore the
Table 2. Self-management behaviors and barriers to adherence among 498 African
American patients admitted to home health care with uncontrolled hypertension                             critical role of treatment adherence, which
from February 2006 through August 2007                                                                    has been widely recognized as a key issue
                                                                                                          in managing hypertension and other
                                Variable                                            Mean (SD) or n (%)
                                                                                                          chronic conditions.20,21 People with these
Mean (SD) Hill-Bone compliance to high blood pressure therapy scale                     18.4 (3.3)        conditions often must follow complex
                                                                                                          medication and diet regimens, monitor
   Medication adherence subscale (range 8–24)                                            9.4   (2.0)
   Sodium intake subscale (range 3–12)                                                   4.5   (1.4)      their own health status, alter their life-
   Appointment keeping subscale (range 3–12)                                             4.5   (1.5)      styles, and effectively interact with various
Currently smokes cigarettes, n (%)                                                      100    (20.1)     healthcare providers.22 Multiple factors
Mean (SD) patient activation measure                                                    58.5   (14.9)
   Stage 1 (least activated), n (%)                                                     107    (21.5)
                                                                                                          influence treatment adherence, including
   Stage 2, n (%)                                                                        72    (14.5)     demographic and clinical factors, medi-
   Stage 3, n (%)                                                                       202    (40.6)     cation complexity and side effects, health
   Stage 4 (most activated), n (%)                                                      117    (23.5)     beliefs, and health system issues.23–27 Our
Selected Brief Medication Questionnaire (BMQ) barriers
Medication causes side effects, n (%)
                                                                                                          study provides information on a less well-
   Not at all                                                                           329 (69.4)        studied phenomenon by using PAM,
   Somewhat/a lot                                                                       145 (30.6)        which measures general self-management
It is hard to pay for meds, n (%)
                                                                                                          preparedness.14 Our multivariate results
   Not at all                                                                           330 (68.9)
   Somewhat/a lot                                                                       149 (31.1)        show that patients at the lowest PAM level
It is hard to remember doses, n (%)                                                                       (stage 1) were 80% more likely to have
   Not at all                                                                           394 (82.3)        severe, uncontrolled hypertension than
   Somewhat/a lot                                                                        85 (17.7)
                                                                                                          were patients who were at the 3 higher
  SD 5 standard deviation.                                                                                levels of awareness, knowledge, and
 * Higher scores indicate poorer adherence.
                                                                                                          confidence in their ability to manage their
                                                                                                          chronic conditions. Independent of PAM
                                                                                                          stage, our multivariate analysis also found
                                                                                                          that poorer adherence to appointment
                                                                                                          keeping, as measured by higher scores in
                                                                                                          the relevant Hill-Bone subscale, was
Table 3. Factors associated with severe uncontrolled hypertension among 498                               significantly associated with severe, un-
African American patients admitted to home health care with uncontrolled                                  controlled hypertension.
hypertension from February 2006 through August 2007*
                                                                                                               Our analysis also identified several
                                                                     Odds Ratio         95% Confidence    protective factors associated with a
                    Characteristic                                    (P Value)            Interval       lower likelihood of severe, uncontrolled
  Female sex                                                          .84   (.46)              .52–1.35   hypertension. Consistent with other
  Age, years (mean, sd)                                               .97   (.04)              .94–1.00   studies,1,28 higher education was associ-
  Educational Level: High school graduate or more                     .68   (.09)              .43–1.05
  Insurance status: Medicaid or dually eligible                      1.16   (.51)              .74–1.82
                                                                                                          ated with better blood pressure control.
  Usual source of care: MD office                                    1.18   (.45)              .77–1.83   We also found older age to be protec-
Clinical Characteristics                                                                                  tive, in contrast to most national
  Months since hypertension diagnosis                                1.01   (.02)          1.01–101       studies, which have found younger
  Months since diagnosis squared                                     1.00   (.01)           .99–1.01      rather than older age to exert a
  Number of additional co-morbidities                                 .94   (.11)           .86–1.02
  Diabetes                                                           3.19   (,.01)         2.02–5.03
                                                                                                          protective influence.29,30 Our finding
  Hospitalized within 14 days of admission to home care               .68   (.10)           .43–1.07      may reflect in part the fact that we
  BMI: Obese:                                                         .81   (.34)           .52–1.25      controlled for patient activation level,
  Moderate/severe Depressive Symptoms                                1.19   (.45)           .75–1.88      which declines with age,14 while to our
Barriers to Adherence                                                                                     knowledge no other published hyper-
  Patient Activation Measure (PAM): Stage 1                          1.80 (.04)            1.01–3.19
                                                                                                          tension studies have introduced a com-
  Hill Bone Compliance Scale
    Medication Adherence                                              .94 (.48)             .80–1.11      parable control for self-management
    Sodium Intake                                                     .92 (.26)             .79–1.06      preparedness. Our findings suggest that
    Appointment Keeping                                              1.15 (.02)            1.01–1.29      once the effects of patient activation,
 * All analyses also control for county of residence.                                                     combined with other clinical and be-
                                                                                                          havioral factors, are separated from age,

                                                        Ethnicity & Disease, Volume 19, Spring 2009                                                151

other circumstances of older age may           female, elderly, diabetic, and having                Yet our results highlight the challeng-
lead to better blood pressure control.         multiple co-morbid conditions. These             es ahead for home care agencies seeking
     Finally, admission to home care from      risk factors may overpower obesity and           more aggressive, effective strategies for
the hospital (compared with referral by a      smoking as predictors of hypertension            treating African American patients with
community physician) yielded a 30%             severity in this population. Another             uncontrolled hypertension. Our sample
lower probability of severe, uncontrolled      potential consideration is that risky            was marked by low education, low
hypertension, while each additional co-        behaviors such as obesity and smoking            income, heavy reliance on clinics for
morbidity also had a protective influence.     may be more powerful predictors of any           primary care, high likelihood of obesity
The prior hospitalization result suggests      hypertension than of severe hypertension         and multiple comorbidities, taking al-
that hypertension may receive more             among a population such as ours, all of          most six prescribed medications – all
concentrated attention in the hospital         whom have uncontrolled blood pressure.           factors complicating medical and behav-
setting than in the community, where                This study is not without limita-           ioral regimens, treatment adherence and
physicians frequently have been cited for      tions. It relies on self-reported measures       interactions with the health care system.
‘‘inertia’’ with respect to treatment of       of adherence and on cross-sectional data         Subjects had on average a 14-year history
high blood pressure.31,32 The co-morbid-       that suggest, but cannot prove, causal-          of grappling with hypertension, while six
ity result is consistent with several recent   ity. Further, our study focuses on               in ten patients had severe, uncontrolled
studies that found multi-morbidity to be       African American patients served by a            Stage 2 hypertension. Low education,
associated with better quality of care.33,34   single urban home health organization            long duration of hypertension, and
This relationship, in turn, seems partly       that provides care to a low-socioeco-            diabetes were especially pronounced
due to greater frequency of physician          nomic status population that is more             among the latter group, indicating their
visits and higher rates of specialty care      likely to be dually eligible for Medicare        heightened socio-economic and clinical
among people with multiple chronic             and Medicaid than is the typical home            vulnerability. Further, individuals in this
conditions.33                                  care population. Because we report on a          group were handicapped with a very low
     Surprisingly, patients in our study       patient sample specifically selected to          level of patient activation, indicating lack
with blood pressure at differing stages of     target high blood pressure among                 of awareness, knowledge, skills or confi-
severity did not differ with respect to a      African Americans, we cannot report              dence in their ability to deal with this
number of factors that conventional            comparable data on nonhypertensive               long-term life-threatening but silent con-
wisdom would tell us matter in deter-          African Americans or members of other            dition.
mining the likelihood of better-con-           racial/ethnic groups with or without                 Successful strategies will require more
trolled hypertension. Contrary to find-        uncontrolled hypertension. However,              aggressive action on the part of home
ings in another study that urgent and          we believe that a strength of our study          health nurses to alert both patients and
emergent hypertension, among a sample          is its narrow focus on an undertreated,          patients’ primary care physicians (PCPs)
of inner-city, minority emergency de-          understudied population among whom               to the presence of ongoing, unsuccessfully
partment patients, were more common            high blood pressure has proven to be             treated hypertension – an action compli-
among patients who had no primary              particularly intractable and damaging.           cated by physicians’ inertia and the fact
care physician,23 we found that having a            To our knowledge, this is the first         that so many patients rely on clinics,
physician’s office (as opposed to clinic       study to target home healthcare pa-              including hospital outpatient clinics with
or other setting) as a usual source of care    tients, an especially complex, high-risk         a string of rotating residents. Linking
did not reduce the likelihood of severe,       group characterized by prior hospital-           ‘‘unconnected’’ patients to a ‘‘medical
uncontrolled hypertension. Our finding         ization or referral by a community               home’’ – a usual source of care where the
may reflect the fact that we controlled        physician for skilled nursing care. Home         PCP gets to know the needs of the
for appointment keeping. Clinical and          health care represents a segment of the          hypertensive patient and assures that the
behavioral risk factors such as obesity        healthcare system where a substantial            patient is on the most effective hyperten-
and smoking also were not significantly        number of high-risk hypertension pa-             sion medication regimen over time – may
associated with severe uncontrolled            tients are served, where hypertension            be essential to long-term BP control but
hypertension in our sample, even when          management likely requires improve-              will require broader systems reform than
other potentially collinear variables,         ment, and where nursing personnel are            can be effected by most home care
such as adherence behaviors and PAM            uniquely positioned to mobilize care             providers during the typical Medicare/
score, were eliminated from the regres-        management and patient-education                 Medicaid home care stay. In contrast,
sion. Our study sample is more homog-          skills to increase the proportion of             developing education and behavioral
enous than most—entirely African               treated hypertension patients who                strategies designed to address varying
American and disproportionately urban,         achieve adequate blood pressure control.         levels of patient preparedness for self-care

152                                               Ethnicity & Disease, Volume 19, Spring 2009
                                                                                       HYPERTENSIVE HOME CARE PATIENTS - McDonald et al

management should be more doable                            8. Ashaye MO, Giles WH. Hypertension in                21. DiMatteo M, Giordani PJ, Lepper HS,
                                                               Blacks: a literature review. Ethn Dis.                  Croghan TW. Patient adherence and medical
within the current scope of home health
                                                               2003;13(4):456–462.                                     treatment outcomes: a meta-analysis. Med
care practice. With proper training and                     9. MacMahon S, Peto R, Cutler J, et al. Blood              Care. 2002;40:794–811.
support for home care professionals and                        pressure, stroke, and coronary heart disease,       22. Holman H, Lorig K. Patients as partners in
paraprofessionals, home care has the                           part 1: prolonged differences in blood pressure:        managing chronic disease. Partnership is a
potential to achieve long-term impact by                       prospective observational studies corrected for         prerequisite for effective and efficient health
developing ‘‘activation-appropriate’’                          the regression dilution bias. Lancet. 1990;335:         care. BMJ. 2000;320:526–527.
                                                               765–774.                                            23. Shea S, Misra D, Ehrlich MH, Field L, Francis
strategies tailored to individual patients’
                                                           10. He J, Whelton PK. Elevated systolic blood               CK. Correlates of nonadherence to hyperten-
knowledge and skills and designed to                           pressure and risk of cardiovascular and renal           sion treatment in an inner-city minority
‘‘break the cycle of failure’’ experienced by                  disease: an overview of evidence from obser-            population. Am J Public Health. 1992;82:
patients who are either unaware of or                          vational epidemiologic studies and randomized           1607–1612.
unable to adhere to the behavioral changes                     controlled trials. Am Heart J. 1999;138:            24. Hekler EB, Lambert J, Leventhal E, Leventhal
                                                               211–219.                                                H, Jahn E, Contrada RJ. Commonsense illness
required to achieve effective control of
                                                           11. Lewington S, Clarke R, Qizilbash N, Peto R,             beliefs, adherence behaviors, and hypertension
their chronic conditions.35                                    Collins R. Prospective Studies Collaboration            control among African Americans. J Behav
                                                               Age-specific relevance of usual blood pressure to       Med. 2008;31:391–400.
                                                               vascular mortality: a meta-analysis of individual   25. Ingersoll KS, Cohen J. The impact of
ACKNOWLEDGMENTS                                                data for one million adults in 61 prospective           medication regimen factors on adherence to
This work was supported by grant R01                           studies. Lancet. 2002;60:1903–1913.                     chronic treatment: a review of literature.
HL078585 from the National Heart, Lung                     12. Pogue VA, Ellis C, Michel J, Francis CK. New            J Behav Med. 2008;31:213–224.
and Blood Institute to Penny H. Feldman                        staging system of the fifth Joint National          26. Morisky DE, Ang A, Krousel-Wood M, Ward
(principal investigator). We acknowledge                       Committee report on the detection, evalua-              HJ. Predictive validity of a medication adher-
                                                               tion, and treatment of high blood pressure              ence measure in an outpatient setting. J Clin
Jennifer Mongoven and Ernesto Henriquez
                                                               (JNC-V) alters assessment of the severity and           Hypertens. 2008;10:348–354.
for overseeing the data collection effort.
                                                               treatment of hypertension. Hypertension.            27. Piette JD, Heisler M, Wagner TH. Cost-
                                                               1996;28:713–715.                                        related medication underuse. Do patients with
REFERENCES                                                 13. Kim MT, Hill MN, Bone LR, Levine DM.                    chronic illnesses tell their doctors? Arch Intern
 1. Hertz RP, Unger AN, Cornell JA, Saunders E.                Development and testing of the Hill-Bone                Med. 2004;164:1749–1755.
    Racial disparities in hypertension prevalence,             compliance to high blood pressure therapy           28. Grotto I, Huerta M, Sharabi Y. Hypertension
    awareness and management. Arch Intern Med.                 scale. Prog Cardiovasc Nurs. 2000;15:90–96.             and socioeconomic status. Curr Opin Cardiol.
    2005;165:2098–2104.                                    14. Hibbard JH, Mahoney ER, Stockard J, Tusler              2008;23:335–339.
 2. Douglas JG, Bakris GL, Epstein M, et al.                   M. Development and testing of a short form of       29. Ostchega Y, Dillon CF, Hughes JP, Carroll
    Management of high blood pressure in African               the patient activation measure. Health Serv             M, Yoon S. Trends in hypertension preva-
    Americans: consensus statement of the Hyper-               Res., 2005;40:1918–1930.                                lence, awareness, treatment, and control in
    tension in African Americans Working Group             15. Svarstad BL, Chewning BA, Sleath BL,                    older US adults: data from the National
    of the International Society on Hypertension in            Claesson C. The Brief Medication Question-              Health and Nutrition Examination Survey
    Blacks. Arch Intern Med. 2003;163:525–541.                 naire: a tool for screening patient adherence           1988 to 2004. J Am Geriatr Soc. 2007;55:
 3. Rosamond W, Flegal K, Furie K, et al. Heart                and barriers to adherence. Patient Educ Couns.          1056–1065.
    disease and stoke statistics 2007—update: a                1999;37:113–124.                                    30. Duggirala MK, Cuddihy RM, Cuddihy MT,
    report from the American Heart Association             16. Chaudhry A, Jin L, Meltzer D. Use of a self-            et al. Predictors of blood pressure control in
    Statistics Committee and Stroke Statistics Sub-            report generated Charlson comorbidity index             patients with diabetes and hypertension seen in
    committee. Circulation. 2008;117:e25–e146.                 for predicting mortality. Med Care. 2005;43:            primary care clinics. Am J Hypertens.
 4. Cushman WC, Ford CE, Cutler JA, et al.                     607–615.                                                2005;18:833–838.
    Success and predictors of blood pressure               17. Sheikh JI, Yesavage JA. Geriatric depression        31. Phillips LS, Branch WT, Cook CB, et al.
    control in diverse North American settings:                scale (GDS): recent evidence and development            Clinical inertia. Ann Intern Med. 2001;135:
    The Antihypertensive and Lipid-Lowering                    of a shorter version. In: Clinical Gerontology: A       825–834.
    Treatment to Prevent Heart Attack Trial                    Guide to Assessment and Intervention. New York,     32. Ogedegbe G. Barriers to optimal hypertension
    (ALLHAT). J Clin Hypertens. 2002;4:393–404.                NY: The Haworth Press. 1986:165–173.                    control. J Clin Hypertens. 2008;10:644–646.
 5. Appel LJ, Moore TJ, Obarzanek E, et al. A              18. Schrier RW, Estacio RO, Esler A, Mehler P.          33. Higashi T, Wenger NS, Adams JL, et al.
    clinical trial of the effects of dietary patterns on       Effects of aggressive blood pressure control in         Relationship between number of medical
    blood pressure: DASH Collaborative Research                normotensive type 2 diabetic patients on                conditions and quality of care. N Engl J Med.
    Group. N Engl J Med. 1997;336:1117–1124.                   albuminuria, retinopathy and strokes. Kidney            2007 356:2496–2504.
 6. Gornick ME. Vulnerable Populations and                     Int. 2002;61:1086–1097.                             34. Min LC, Wenger NS, Fung C, et al. Multi-
    Medicare Services: Why Do Disparities Exist?.          19. Piette JD, Kerr EA. The impact of comorbid              morbidity is associated with better quality of
    New York, NY: The Century Foundation                       chronic conditions on diabetes care. Diabetes           care among vulnerable elders. Med Care.
    Press; 2000.                                               Care. 2006;29:725–731.                                  2007;45:480–488.
 7. Hargraves JL, Cunningham PJ, Hughes RG.                20. Von Korff M, Gruman J, Schaefer J, Curry SJ,        35. Hibbard JH, Tusler M. Assessing activation
    Racial and ethnic differences in access to                 Wagner EH. Collaborative management of                  stage and employing a ‘‘next steps’’ approach
    medical care in managed care plans. Health                 chronic illness. Ann Intern Med. 1997;127:              to supporting patient self-management. J
    Serv Res. 2001;36:853–868.                                 1097–1102.                                              Ambul Care Manage. 2007;30:2–8.

                                                           Ethnicity & Disease, Volume 19, Spring 2009                                                             153

Shared By: