Patient Engagement Following Acute Myocardial

Infarction and Its Influence on Outcomes

Anthony E. Peters, MD, and Ellen C. Keeley, MD, MS*

The Patient Activation Measure (PAM) is a validated assessment tool that evaluates how
engaged patients are in their own health care. The more engaged or “activated” patients
are, the higher the score and the more likely they are to adhere to medical therapy and
make healthy lifestyle choices. Little is known regarding patient activation in patients after
an acute myocardial infarction. From March 2016 to December 2016, we administered PAM
surveys to patients after myocardial infarction at the time of a clinic visit scheduled within
10 days of hospital discharge. Demographic and outcome data were collected. The primary
outcome was defined as a composite end point of major medication errors, emergency de-
partment visits, and/or unplanned readmission. The secondary outcome was continued tobacco
use after discharge. A total of 93 patients were enrolled and 39 (42%) were positive for the
primary outcome. PAM scores ranged from 40.9 to 100 (median 62.6, interquartile range
56.0 to 72.1). In multivariable analysis, adjusting for age, gender, and burden of co-
morbidities, patients with lower PAM scores were more likely to have the primary outcome
(odds ratio 1.063, 95% confidence interval 1.020 to 1.109, p = 0.0041). Patients with lower
PAM scores also were more likely to continue to use tobacco after discharge (odds ratio
1.060, 95% confidence interval 1.005 to 1.118, p = 0.0325). In conclusion, we found an as-
sociation between lower PAM scores and subsequent adverse clinical outcomes, including
unplanned readmissions. Further investigation into the potential effect of education and
coaching interventions in patients with low PAM scores after acute myocardial infarction
is warranted. © 2017 Elsevier Inc. All rights reserved. (Am J Cardiol 2017;120:1467–1471)

The principle of patient activation reflects patients’ un- Methods

derstanding of what their role is in the overall care plan; how
Patients were enrolled from March 2016 to December 2016
engaged they are in their own health care. Patient activation
in the University of Virginia post-MI clinic, a multidisci-
incorporates a combination of knowledge regarding their
plinary clinic where patients with AMI (ST elevation and non-
illness, skill, and self-confidence in the management of their
ST elevation) are seen within 10 days of hospital discharge.
medical conditions.1 It has been shown to be associated with
Investigational review board approval was obtained to ad-
health behaviors, metrics of chronic disease, morbidity, and
minister a 10-question survey (PAM-10) after obtaining written
hospitalizations.2–6 Lower patient activation scores have been
consent. Answers were converted to numerical scores through
associated with higher health-care costs.7 The Patient Acti-
a validated algorithm. PAM scores range from 1 to 100 and
vation Measure (PAM) is an effective survey tool that quantifies
are separated into the following levels: level 1 (≤47.0) re-
a patient’s activation level. It has been validated in diabetes,8
flects patients who are passive, lack confidence, and are
chronic obstructive pulmonary disease,9 multiple sclerosis,10
nonadherent; level 2 (47.1 to 55.1) reflects patients who have
and congestive heart failure.11–13 Several studies have found
gained some knowledge of their condition and can set simple
an association between PAM scores and cancer screening,
goals, but they still believe that their health is largely out of
smoking, obesity, emergency room visits, and readmissions.2–5
their control; level 3 (55.2 to 67.0) reflects patients who are
The principle of patient activation has not been rigorously
taking action and building self-management skills; and level
studied in the population after acute myocardial infarction
4 (≥67.1) reflects patients who have adopted healthier be-
(AMI) to date. The goal of this study was to evaluate the
haviors and are trying to maintain a healthy lifestyle.14
impact of PAM on clinical outcomes including unplanned re-
Baseline demographics, length of hospital stay, medical
admissions in patients after AMI.
co-morbidities, and adherence to discharge medication regimen
was collected during the clinic visit and through retrospec-
tive chart review. Overall burden of co-morbidities was defined
as the total number of co-morbidities per patient. During the
clinic visit, patients were assessed for symptoms, medica-
Department of Medicine, University of Virginia, Charlottesville, Virginia. tion adherence, current tobacco use, and emergency room visits
Manuscript received May 1, 2017; revised manuscript received and ac-
and readmissions since hospital discharge. The medication
cepted July 21, 2017.
Funding: This work was supported by the American Heart Association
regimen of each patient was reviewed, and discrepancies from
(13IRG14560018 and 16IRG27180006 to ECK). discharge regimen were noted. Information of emergency de-
See page 1470 for disclosure information. partment visits and unplanned readmissions over the
*Corresponding author: Tel: (352) 273-9065; fax: (352) 392-3606. subsequent 3 months were collected through chart review using
E-mail address: Ellen.Keeley@medicine.ufl.edu (E.C. Keeley). an electronic medical record by Epic Systems Corporation

0002-9149/$ - see front matter © 2017 Elsevier Inc. All rights reserved. www.ajconline.org
https://doi.org/10.1016/j.amjcard.2017.07.037
1468 The American Journal of Cardiology (www.ajconline.org)

(EPIC) and the Care Everywhere platform, allowing us to
obtain data on patients who were evaluated and/or readmit-
ted to outside hospitals. Smoking status at follow-up was
documented by the pharmacist who provided smoking ces-
sation coaching.
Patient demographics, co-morbidities, and discharge medi-
cation regimens were compared with the Wilcoxon rank sum
test or Kruskal-Wallis test (continuous variables) and chi-
square test (categorical variables). The Fisher’s exact test
was used for categorical variables with small (<5) expected
values. The primary outcome was defined as major medica-
tion errors identified at clinic visit after MI, emergency
department visits, and/or unscheduled readmission. Major
medication error was defined as a medication error that threat-
ened the efficacy of standard of care for patients with AMI
and, if left unresolved, would lead to adverse outcomes. The
secondary outcome was continued tobacco use after dis-
charge. Univariable logistic regression was used to analyze
continued tobacco use after discharge. Multivariable logis-
tic regression was used to compare scores with regard to
outcomes, adjusted for age, gender, and burden of co-
morbidities. Receiver operating characteristic curve analysis
was performed to evaluate the performance of the model as
a predictor of the primary outcome. Performance was com-
pared with established predictive scores: the Thrombolysis
in Myocardial Infarction score,15,16 the Global Registry of
Acute Coronary Events score,17 and the Primary Angioplasty
in Myocardial Infarction score.18 Analyses were performed
with Statistical Analysis System software, version 9.4 (SAS
Institute, Cary, NC). A p value of <0.05 was considered sta-
tistically significant.
Results
A total of 93 patients who attended the clinic appoint-
ment were enrolled, with PAM scores ranging from 40.9 to
100 (median 62.6, interquartile range 56.0 to 72.1). Base-
line patient demographics and co-morbidities were not
significantly different in patients with different PAM levels
(Table 1). Patients who experienced the primary outcome were
more likely to be female and had a higher burden of co-
morbidities (Table 2). All patients completed the 3-month
follow-up. A total of 39 (42%) patients had the primary end
point, including 19 major medication errors, 17 emergency
department visits, and 18 unplanned readmissions for a total
of 47 events; several of these events occurred in the same pa-
tients and were counted as a single “positive” outcome. Median
number of days from discharge to readmission was 17
(interquartile range 6 to 45) for the entire cohort. Major medi-
cation errors included duplicate or incorrect dosing of
antiplatelet agents (n = 4), and errors in hypertension medi-
cations (n = 8), hyperlipidemia medications (n = 3), diabetes
medications (n = 3), and anticoagulation (n = 1).
In a multivariable model adjusting for age, gender, and
burden of co-morbidities, lower PAM scores were signifi-
cantly associated with the primary outcome (odds ratio [OR]
1.063, 95% confidence interval [CI] 1.020 to 1.109, p = 0.0041).
Other predictors of the primary outcome were younger age
(0.953, 95% CI 0.914 to 0.994, p = 0.0244), female gender
(OR 13.676 CI 3.211 to 58.251, p = 0.0004), and increased
burden of co-morbidities (OR 2.738 CI 1.675 to 4.475,
p <0.0001) (Table 3). For comparison with established scores,
continuous variables were categorized based on the strength
of their association with the primary outcome as follows: age
<65, 1 point; female sex, 3 points; co-morbidity burden, 0 to
6 points; and inverse of PAM level, 0 to 3 points. This score
compared favorably with the Thrombolysis in Myocardial In-
farction and Global Registry of Acute Coronary Events scores
(Figure 1). The Primary Angioplasty in Myocardial Infarc-
tion score produced a c-statistic of 0.5227 for MI patients with
ST-elevation (n = 29). The primary end point was driven largely
by unplanned readmissions and major medication errors,
whereas emergency department visits alone played a less sig-
nificant role (Figure 2). PAM scores also were significantly
associated with continued tobacco use at post-MI clinic follow-
up (OR 1.060, 95% CI 1.005 to 1.118, p = 0.0325).

Table 1
Patient characteristics stratified by Patient Activation Measure level
Variable Level 1–2 (n = 23) Level 3 (n = 48) Level 4 (n = 22) p-Value
Age 69 (58–76) 59.5 (48.5–73) 64.5 (46–75) 0.1979
Male 19 (83%) 37 (77%) 14 (64%) 0.3089
Race
Caucasian 19 (83%) 41 (85%) 15 (68%) 0.1965
African-American 3 (13%) 5 (11%) 7 (32%)
Other 1 (4%) 2 (4%) 0 (0%)
Current smoker 7 (30%) 16 (33%) 6 (27%) 0.8754
Hypertension 18 (78%) 37 (77%) 14 (64%) 0.4298
Hyperlipidemia 17 (74%) 28 (58%) 15 (68%) 0.4030
Diabetes 14 (61%) 21 (44%) 9 (41%) 0.3163
Chronic kidney disease 5 (22%) 5 (10%) 4 (18%) 0.4025
Chronic obstructive pulmonary disease 2 (9%) 7 (15%) 3 (14%) 0.8474
Known coronary artery disease 10 (43%) 14 (29%) 7 (32%) 0.4812
Comorbidity burden 3 (2–4) 2 (1–3.5) 2.5 (1–4) 0.3575

Values are median (interquartile range) or count (percentage). Hypertension is defined by a previous documentation of blood pressure >140/90 mm Hg or
current use of antihypertensive medication. Hyperlipidemia is defined by a previous documentation of total cholesterol >200 mg/dl or low-density lipopro-
tein >130 mg/dl or high-density lipoprotein <40 mg/dl or current use of lipid-lowering agent. Co-morbidity burden was defined as the sum of co-morbidities
(hypertension, hyperlipidemia, diabetes, chronic kidney disease, chronic obstructive pulmonary disease, and known coronary artery disease) for each patient.
 Coronary Artery Disease/Patient Engagement and Outcomes Post-MI 1469

Table 2
Patient characteristics according to primary outcome
Variable With primary outcome Without primary outcome p-Value
(n = 39) (n = 54)

Age 65 (52–76) 61 (47–74) 0.5055

Male 24 (62%) 46 (85%) 0.0091
Race
Caucasian 30 (77%) 45 (83%) 0.6715
African-American 7 (18%) 8 (15%)
Other 2 (5%) 1 (2%)
PAM score 59.3 (52.9–64.2) 65.8 (56.8–75.5) 0.0216
Current smoker 9 (23%) 20 (37%) 0.1516
Hypertension 37 (95%) 32 (59%) 0.0001
Hyperlipidemia 33 (85%) 27 (50%) 0.0006
Diabetes 25 (64%) 19 (35%) 0.0058
Chronic kidney disease 10 (26%) 4 (7%) 0.0152
Chronic obstructive pulmonary disease 5 (13%) 7 (13%) 0.9839
Known coronary artery disease 14 (36%) 17 (31%) 0.6558
Comorbidity burden 3 (2–4) 2 (1–3) 0.0001
Aspirin
P2Y12 inhibitor 38 (97%) 54 (100%) 0.4194
Beta-blocker 35 (90%) 49 (91%) 1.0000
Angiotensin converting enzyme inhibitor 36 (92%) 45 (83%) 0.2308
Calcium channel blocker 28 (72%) 29 (54%) 0.0772
Statin
8 (21%) 2 (4%) 0.0154
36 (92%) 52 (96%) 0.6463
Left ventricular ejection fraction < 40% 7 (17.95%) 9 (17%) 0.8716
Length of stay (days) 3 (2–3) 2.5 (2–3) 0.7591

Values are median (interquartile range) or count (percentage). Hypertension is defined by a previous documentation of blood pressure >140/90 mm Hg or
current use of antihypertensive medication. Hyperlipidemia is defined by a previous documentation of total cholesterol >200 mg/dl or low-density lipopro-
tein >130 mg/dl or high-density lipoprotein <40 mg/dl or current use of lipid-lowering agent.

Table 3
Multivariable logistic regression results for primary outcome
Variable β coefficient Standard OR (95% CI) p-Value
error
PAM score 0.0614 0.0214 1.063 (1.020–1.109) 0.0041
Age −0.0478 0.0212 0.953 (0.914–0.994) 0.0244
Female sex 2.6156 0.7394 13.676 (3.211–58.251) 0.0004
Comorbidity 1.0073 0.2507 2.738 (1.675–4.475) <0.0001
burden

PAM = Patient Activation Measure.

Co-morbidity burden was defined as the sum of co-morbidities (hyper-
tension, hyperlipidemia, diabetes, chronic kidney disease, chronic obstructive
pulmonary disease, and known coronary artery disease) for each patient.

Discussion
We found that the first 3 months following discharge
for AMI is a high-risk period for patients, with nearly 20%
requiring readmission and an additional 12% requiring
evaluation in the emergency department or resolution of a
major medication error. Patient activation (a measure of
how engaged patients are in their own health care) was
strongly correlated with adverse clinical outcomes in the
post-AMI period, independent of age, gender, and burden Figure 1. Receiver operating characteristic curves for the model using the
of co-morbidities. PAM score (blue), the TIMI score (red), and the GRACE score (green).
Efforts to minimize adverse events after discharge have PAM = Patient Activation Measure; TIMI = Thrombolysis in Myocardial In-
focused on effective communication through discharge farction; GRACE = Global Registry of Acute Coronary Events.
1470 The American Journal of Cardiology (www.ajconline.org)
Figure 2. Clinical events according to Patient Activation Measure (PAM) level.
planning, early postdischarge follow-up, and risk factor op-
timization. Although previous studies have shown some relation
between these principles and adverse events after dis-
charge, the data have been mixed with few reliable and
reproducible methods.19–26 Our findings suggest that patient
activation should be considered as an additional risk factor
that can be easily assessed, tracked, and potentially modified.
Targeting and achieving improvement in patient activa-
tion will require a multidisciplinary effort, as evidenced by
other postdischarge programs.27–29 In our post-MI clinic, pa-
tients are seen by a physician, pharmacist, exercise
physiologist, dietitian, and social worker (when indicated).
This clinic model uncovered medication errors and pro-
vided a platform to assess patient activation. Thus far,
exploratory efforts to coach several patients low scoring in
the PAM through bimonthly interactions with social workers
on a longitudinal basis have resulted in substantial in-
creases in PAM scores, indicating activation may be a highly
modifiable risk factor. This is important because there is evi-
dence from a previous study that systematic coaching and
communication interventions can reduce readmissions in pa-
tients with complex care needs.30
Our study was limited by the typical constraints of a ret-
rospective, observational study. In addition, there may have
been selection bias in that completion of the PAM question-
naire was optional, and despite the wide range of scores in
our study group, it is possible that patients with certain char-
acteristics may have been less likely to participate in the study.
Lastly, we are just beginning to assess whether coaching will
improve PAM scores in this population and therefore do not
have long-term follow-up data at this time.
Disclosures
The authors have no conflicts of interest to disclose.
1. Hibbard JH, Greene J. What the evidence shows about patient activa-
tion: better health outcomes and care experiences; fewer data on costs.
Health Aff 2013;32:207–214.
2. Greene J, Hibbard JH. Why does patient activation matter? An exami-
nation of the relationships between patient activation and health-
related outcomes. J Gen Intern Med 2012;27:520–526.
3. Mosen DM, Schmittdiel J, Hibbard J, Sobel D, Remmers C, Bellows
J. Is patient activation associated with outcomes of care for adults with
chronic conditions? J Ambul Care Manag 2007;30:21–29.
4. Skolasky RL, Green AF, Scharfstein D, Boult C, Reider L, Wegener
ST. Psychometric properties of the patient activation measure among
multimorbid older adults. Health Serv Res 2011;46:457–478.
5. Salyers MP, Matthias MS, Spann CL, Lydick JM, Rollins AL, Frankel
RM. The role of patient activation in psychiatric visits. Psychiatr Serv
2009;60:1535–1539.
6. Schmittdiel J, Mosen DM, Glasgow RE, Hibbard J, Remmers C, Bellows
J. Patient Assessment of Chronic Illness Care (PACIC) and improved
patient-centered outcomes for chronic conditions. J Gen Intern Med
2008;23:77–80.
7. Hibbard JH, Greene J, Overton V. Patients with lower activation asso-
ciated with higher costs; delivery systems should know their patients’
“scores.” Health Aff (Millwood) 2013;32:216–222.
8. Sacks RM, Greene J, Hibbard J, Overton V, Parrotta CD. Does patient
activation predict the course of type 2 diabetes? A longitudinal study.
Patient Educ Couns 2017;100:1268–1275.
9. Müllerová H, Landis SH, Aisanov Z, Davis KJ, Ichinose M, Mannino
DM, Maskell J, Menezes AM, Molen T, van der Oh YM, Tabberer M,
Han MLK. Health behaviors and their correlates among participants in
the continuing to confront COPD international patient survey. Int J COPD
2016;11:881–890.
10. Packer TL, Kephart G, Ghahari S, Audulv Å, Versnel J, Warner G. The
patient activation measure: a validation study in a neurological popu-
lation. Qual Life Res 2015;24:1587–1596.
11. Siddharthan T, Rabin T, Canavan ME, Nassali F, Kirchhoff P,
Kalyesubula R, Coca S, Rastegar A, Knauf F. Implementation of patient-
centered education for chronic-disease management in Uganda: an
effectiveness study. PLoS ONE 2016;11:1–12.
12. Ngooi BX, Packer TL, Kephart G, Warner G, Koh KWL, Wong RCC,
Lim SP. Validation of the patient activation measure (PAM-13) among
adults with cardiac conditions in Singapore. Qual Life Res 2017;26:1071–
1080.
13. Shively MJ, Gardetto NJ, Kodiath MF, Kelly A, Smith TL, Stepnowsky
C, Maynard C, Larson CB. Effect of patient activation on
 Coronary Artery Disease/Patient Engagement and Outcomes Post-MI
self-management in patients with heart failure. J Cardiovasc Nurs
2013;28:20–34.
14. Insignia Health. Patient activation measure. 2015. Available at: http://
www.insigniahealth.com/products/pam-survey.
15. Antman EM, Cohen M, Bernink PJLM, McCabe CH, Horacek T,
Papuchis G, Mautner B, Corbalan R, Radley D, Braunwald E. The TIMI
risk score for unstable angina/non-ST elevation MI: a method for prog-
nostication and therapeutic decision making. J Am Med Assoc
2000;284:835–842.
16. Morrow DA, Antman EM, Charlesworth A, Cairns R, Murphy SA, Lemos
JA, de Giugliano RP, McCabe CH, Braunwald E. Clinical investiga-
tion and reports TIMI risk score for ST-elevation myocardial infarction:
a convenient, bedside, clinical score for risk assessment at presenta-
tion. Circulation 2000;102:2031–2037.
17. Granger CB, Goldberg RJ, Dabbous O, Pieper KS, Eagle KA, Cannon
CP, Werf F, Van De Avezum A, Goodman SG, Flather MD, Fox KA,
Global Registry of Acute Coronary Events Investigators. Predictors of
hospital mortality in the global registry of acute coronary events. Arch
Intern Med 2003;163:2345–2353.
18. Addala S, Grines CL, Dixon SR, Stone GW, Boura JA, Ochoa AB,
Pellizzon G, O’Neill WW, Kahn JK. Predicting mortality in patients
with ST-elevation percutaneous coronary intervention (PAMI Risk Score).
Am J Cardiol 2004;93:629–632.
19. Hansen LO, Young RS, Hinami K, Leung A, Williams MV. Interven-
tions to reduce 30-Day rehospitalization: a systematic review. Ann Intern
Med 2011;155:520–528.
20. Jha AK, Orav EJ, Epstein AM. Public reporting of discharge
planning and rates of readmissions. N Engl J Med 2009;361:2637–
2645.
21. Hoyer EH, Odonkor CA, Bhatia SN, Leung C, Deutschendorf A, Brotman
DJ. Association between days to complete inpatient discharge summa-
ries with all-payer hospital readmissions in Maryland. J Hosp Med
2016;11:393–400.
1471
22. Hess CN, Shah BR, Peng SA, Thomas L, Roe MT, Peterson ED. As-
sociation of early physician follow-up and 30-day readmission after non-
ST-segment-elevation myocardial infarction among older patients.
Circulation 2013;128:1206–1213.
23. Hernandez AF, Greiner MA, Fonarow GC, Hammill BG, Heidenreich
PA, Yancy CW, Peterson ED, Curtis LH. Relationship between early
physician follow-up and 30-day readmission among Medicare beneficiaries
hospitalized for heart failure. J Am Med Assoc 2010;303:1716–1722.
24. Misky GJ, Wald HL, Coleman EA. Post-hospitalization transitions: ex-
amining the effects of timing of primary care provider follow-up. J Hosp
Med 2010;5:392–397.
25. Grafft CA, McDonald FS, Ruud KL, Liesinger JT, Johnson MG, Naessens
JM. Effect of hospital follow-up appointment on clinical event out-
comes and mortality. Arch Intern Med 2010;170:955–960.
26. Field TS, Ogarek J, Garber L, Reed G, Gurwitz JH. Association of early
post-discharge follow-up by a primary care physician and 30-day re-
hospitalization among older adults. J Gen Intern Med 2015;30:565–
571.
27. Jack BW, Chetty VK, Anthony D, Greenwald JL, Sanchez GM, Johnson
AE, Forsythe SR, O’Donnell JK, Passche-Orlow MK, Manasseh C,
Martin S, Culpepper L. A reengineered hospital discharge program to
decrease rehospitalization: a randomized trial. Ann Intern Med
2009;150:178–187.
28. Schnipper JL, Kirwin JL, Cotugno MC, Wahlstrom SA, Brown BA,
Tarvin E, Kachalia A, Horng M, Roy CL, McKean SC, Bates DW. Role
of pharmacist counseling in preventing adverse drug events after hos-
pitalization. Arch Intern Med 2006;166:565–571.
29. Boutwell AE, Johnson MB, Watkins R. Analysis of a social work-
based model of transitional care to reduce hospital readmissions:
preliminary data. J Am Geriatr Soc 2016;64:1104–1107.
30. Coleman EA, Parry C, Chalmers S, Min S. The care transitions inter-
vention: results of a randomized controlled trial. Arch Intern Med
2006;166:1822–1828.