Ecological Momentary Assessment of Physical Activity

Genevieve F. Dunton, PhD, MPHGenevieve F. Dunton, PhD, MPH

Department of Preventive MedicineUniversity of Southern California

Recall Instruments

• Memory errors and biases

• Not completed in the environment in which

Limitations of Recall and Observational Methods

the behavior occurs

Observational Methods

• Often limited to a single setting

• Do not measure mood or subjective perceptions

Accelerometer

• Difficulty measuring activity type, load bearing, incline, bicycling, swimming

Limitations of Objective Methods

GPS

• Difficulty differentiating some modes of travel (scooter chair, wheel chair, stroller)

• Do not measure mood or subjective perceptions

• Real-time self-report responses in naturalistic settings

• Can simultaneously measure:

Ecological Momentary Assessment (EMA)

• Can simultaneously measure:

1) Activity type (e.g., soccer, watching TV)

2) Where (e.g., playground, trail, sidewalk)

3) With whom (e.g., alone, friends, siblings)

4) Perceived characteristics (e.g., safety, traffic)

5) Mood (e.g., positive affect, negative affect, stress)

Electronic EMA Equipment/Technology

• Mobile phone • Personal Digital Assistant (PDA)• Internet (Laptop, Desktop, iPad)• Internet (Laptop, Desktop, iPad)

• Event-contingent - information recorded during or after a pre-determined behavior

• Interval-contingent- information recorded according to specific pre-set time frames (e.g., at 8am and 12noon everyday)

EMA Sampling Schedule

(e.g., at 8am and 12noon everyday)

• Signal-contingent – information recorded when prompted, often at random times throughout the day

• Context-contingent – information recorded when a context or environment is sensed (GPS, heart rate, etc)

Example EMA Sampling Schedule

(Interval-Signal Contingent Hybrid)

• Monitoring occurred across 4 days (Fri-Mon) for

each wave.

• No prompts during school hours on Friday or • No prompts during school hours on Friday or

Monday.

Eco log ical Momen tary Assessment Prompting Schedule Day 8:30 -

10am 10am-12pm

12-2pm 2-4pm 4-6pm 6-8pm 8-8:30pm

Friday X X X Satu rday X X X X X X X

Sunday X X X X X X X Monday X X X

Note: Question sequences were prompted at a random time within each interval.

EMA Items

Compliance (children ages 9-13 years)

• Children answered 80% (range 7% – 100%) of

EMA surveys

• Unanswered surveys were more common among

African-American (21%), Asian (22%), Mixed/BiracialAfrican-American (21%), Asian (22%), Mixed/Biracial

(22%), and Other (25%) as compared with

White/Caucasian children (11%)

• No differences by day of the week, time of

day, sex, age, income or weight status

Dunton GF, Liao Y, Intille S, Spruijt-Metz D, and Pentz M. (2011). Investigating children’s physical activity and sedentary behavior using Ecological Momentary Assessment with mobile phones. Obesity. 1205-1212.

Physical Activity Levels During Answered and Unanswered Prompts

(Steps in the 30 minutes before the EMA prompt)

.

200

250

300

350

Adj. Wald F = 3.09, df = 1, p = .08)

Adjusted for day of the week, time of day, sex, age, race/ethnicity, income, weight status.

0

50

100

150

200

Answered Prompts Unanswered Prompts

Ste

ps

Extent to Which EMA Surveys

Disrupted Activity

(15 min. before/after the EMA prompt)

.

70

80

90

100

Sedentary Behavior

250

300

Physical Activity

Adj. Wald F = 2.29, df = 1, p = .13 Adj. Wald F = 0.15, df = 1, p = .70

Adjusted for day of the week, time of day, sex, age, race/ethnicity, income, weight status.

0

10

20

30

40

50

60

70

Before After

Ste

ps

0

50

100

150

200

Before After

Ste

ps

Validity of EMA Activity Responses

(30 minutes before the EMA prompt)

.

300

400

500

600

700

Ste

ps

Underweight/At Risk for Underweight/Normal WeightAt Risk for Overweight/ Overweight

Dunton GF, Liao Y, Intille S, Spruijt-Metz D, and Pentz M. (2011). Investigating children’s physical activity and sedentary behavior using Ecological Momentary Assessment with mobile phones. Obesity. 1205-1212.

0

100

200

• Where?

• With whom?

• Do these patterns differ according to?

Using EMA Describe the Contexts of PA and Sedentary Behavior

• Do these patterns differ according to? - Demographic factors (sex, age, ethnic,income)

- Temporal factors (time of day, day of the week, seasons)

Age and Income Differences in Physical Activity Contexts

30%

40%

50%

60%

Perc

en

t o

f P

hysic

al A

cti

vit

y

Rep

ort

ed

Ou

tdo

ors

30%

40%

50%

60%

70%

80%

Perc

en

t o

f P

hysic

al A

cti

vit

y

Rep

ort

ed

Ou

tdo

ors

Dunton, G. F., Kawabata, K., Intille, S., Wolch, J., & Pentz, M. (2012). Assessing the social and physical contexts of children’s leisure-time physical activity: An Ecological Momentary Assessment study. American Journal of Health Promotion. 26, 135-142.

0%

10%

20%

9-10 years 11-13 years

Perc

en

t o

f P

hysic

al A

cti

vit

y

Rep

ort

ed

Ou

tdo

ors

0%

10%

20%

30%

Perc

en

t o

f P

hysic

al A

cti

vit

y

Rep

ort

ed

Ou

tdo

ors

• How do physical activity levels

(e.g., intensity, duration) differ across

physical and/or social contexts?

Using EMA to Examine Differences inPhysical Activity Across Contexts

physical and/or social contexts?

• How do physical activity experiences

(e.g., enjoyment, positive and negative

affect) differ across physical and/or

social contexts?

Physical Activity Level by Social Context(30-min. before EMA prompt)

200

250

300

350

400

Ste

ps

Dunton, G. F., Liao, Y., Intille, S., Wolch, J., & Pentz, M. (2011). Social and physical contextual influences on children’s leisure-time physical activity: An Ecological Momentary Assessment study. Journal of Physical Activity and Health, 8(Suppl 1), S103-S108.

.

0

50

100

150

200

Family and friends

Friends only Family only Alone

Ste

ps

Mood During Physical Activity by

Physical Context

1.5

2

2.5

3

Avera

ge M

oo

d R

ati

ng

Positive Affect

1.5

2

2.5

3

Avera

ge M

oo

d R

ati

ng

Enjoyment

Dunton, G. F., Liao, Y., Intille, S., Wolch, J., & Pentz, M. (2011). Social and physical contextual influences on children’s leisure-time physical activity: An Ecological Momentary Assessment study. Journal of Physical Activity and Health, 8(Suppl 1), S103-S108.

0

0.5

1

1.5

Outdoors Yard Other Home Someone else's house

Avera

ge M

oo

d R

ati

ng

0

0.5

1

1.5

Outdoors Yard Other Home Someone else's house

Avera

ge M

oo

d R

ati

ng

• Is the likelihood of participating in a physical activity bout related to prior or current mood, stress, pain, fatigue, etc?

Using EMA to Examine Antecedents,

Concomitants, Consequences of PA Episodes

mood, stress, pain, fatigue, etc?

• Does participating in a physical activity bout influence subsequent predict subsequent mood, stress, pain, fatigue, etc?

Conceptual Model of

Temporal Relationships

7:45am 11:45am 3:45pm 7:45pm

MVPA MVPA MVPA MVPA

SE SE SE SE

PA PA PA PA

NA NA NA NANA NA NA NA

Energy Energy Energy Energy

Fatigue Fatigue Fatigue Fatigue

Demand Demand Demand Demand

Control Control Control Control

Lagged

Effects

Variable Group Avg.

Coeff. (SE)

p

Self-EfficacyT-1 0.08 (0.02) <.001

Positive AffectT-1 0.06 (0.02) .003

Negative AffectT-1 -0.11 (0.03) <.001

EnergyT-1 0.04 (0.02) .066

Associations with MVPAT

FatigueT-1 -0.02 (0.01) .135

ControlT-1 0.05 (0.02) .004

DemandT-1 -0.01 (0.01) .978

Pos. Soc. Inter. T 0.17 (0.05) .001

Prob. Soc. Inter. T 0.02 (0.06) .692

Stressful EventT -0.01 (0.06) .875

Dunton, G. F., Atienza, A., Castro, C. M., & King, A. C. (2009). Using ecological momentary assessment to examine antecedents and correlates of physical activity bouts in adults age 50+ years: A pilot study. Annals of Behavioral Medicine, 38, 249-255.

• Do some people show stable patterns of physical activity self-efficacy across the day or from day to day whereas other people’s patterns are variable?

Using EMA to Examine Intraindividual Variability

patterns are variable?

• Are between-person differences in degree of intraindividual variability in self-efficacy related to physical activity levels?

0

1

2

3

4

5

6

Se

lf-E

ffic

ac

y

Time

Low Variability Case

0

1

2

3

4

5

6

7

8

9

Se

lf-E

ffic

ac

y

Time

High Variability Case

Intraindividual Variability in

Self-Efficacy and Physical Activity

Intraindividual variability in self-efficacy was lower during weeks when

brisk walking was higher (τ = -0.01, SE = 0.002, p < .01)

Dunton, G. F., Atienza, A. A., Huh, J., Castro, C., Hedeker, D., King, A. C. (under review). Examining within-person variability in physical activity self-efficacy: An Ecological Momentary Assessment study.

Future Areas of Research

Context-Sensitive EMA with Environmental, Biological and Behavioral Triggers

-Integration of data from internal/external

sensors (GPS, accelerometers, salivary

amylase, air pollution monitors, instrumented

asthma inhalers)

Challenges and Limitations

• Missing data

• Reactance

Data

• Participant burden

• Costs

Acknowledgments• Collaborators: Donna Spruijt-Metz, Ph.D (USC)

Mary Ann Pentz, Ph.D (USC) Jennifer Wolch, Ph.D (UCB)

Audie Atienza, Ph.D (NCI/DHHS)

Abby King, Ph.D (Stanford)Abby King, Ph.D (Stanford)

Cynthia Castro, Ph.D (Stanford)

Stephen Intille, Ph.D (Northeastern)

Alex Rothman, Ph.D (Univ. of MN)

Don Hedeker, Ph.D (Univ. of IL, Chicago)

• Programming: Jennifer Beaudin (MIT)

• Project Manager: Keito Kawabata, B.A.

• Ph.D Student: Yue Liao, MPH

Acknowledgments (Cont.)

• Robert Wood Johnson Foundation #65837 (Dunton, PI)

• American Cancer Society #118283-MRSGT-10-012-01-

CPPB (Dunton, PI)CPPB (Dunton, PI)

• National Cancer Institute #R01-CA-123243 (Pentz, PI).

• National Institute of Aging #R01-AG-12358 (King, PI)

Thank You