Psychometrics of the Lifetime Total Physical Activity Questionnaire in Adults with a History of Depression

Abstract

Background:

Depression earlier in life is a risk factor for dementia later in life. Physical activity is associated with less risk of depression and dementia. However, prevention of Alzheimer’s disease may be most effective before the onset of brain pathology, which precedes dementia symptoms by 20+ years, requiring the measure of physical activity across the life course.

Objective:

Assess the reliability and validity of the Lifetime Total Physical Activity Questionnaire (LTPAQ) in community-dwelling cognitively intact adults with and without a history of depression.

Methods:

Test-retest reliability was assessed with intraclass correlations (ICC). Evidence of construct validity was assessed with descriptive statistics using the contrasted group approach and a priori hypotheses testing with each intensity (sedentary, light, moderate, vigorous) within each domain (school/occupation, transportation, household, leisure-time) across the life course (school-age, adolescence, young adult, middle adult).

Results:

Evidence of test-retest reliability and construct validity was found during school-age and adolescence for moderate school/occupation (ICC=.562, p=.006; ICC=.605, p=.003, respectively) and household (ICC=.945, p<.001; ICC=.829, p<.001, respectively); during young adulthood for moderate school/occupation (ICC=.844, p<.001), transportation (ICC=.747, p<.001), and household (ICC=.608, p=.002); and during middle adulthood for moderate transportation (ICC=.977, p<.001). At both time points, there were zero reported school-age vigorous school/occupation and household; adolescent vigorous transportation; young adult vigorous household; and middle adult vigorous transportation and household physical activity.

Conclusions:

The LTPAQ may be used to examine physical activity across the life course and risk for Alzheimer’s disease in community-dwelling cognitively intact adults with and without a history of depression.

Background & Purpose

Depression is the leading cause of disability worldwide and a major contributor to the overall global burden of disease.¹ While depression later in life can be a prodrome of dementia, depression earlier in life is a risk factor for dementia.² For example, depression earlier in life has consistently been associated with a greater than two-fold increase in the risk of dementia later in life.³ Depression, both earlier and later in life, shares similar neurobiological changes with dementia, particularly white matter disease, potentially indicating shared risk factors.²

Physical activity is associated with less risk of depression⁴ and dementia.⁵ Physical activity is also associated with less white matter disease in healthy individuals.⁶ For example, walking or biking to or from school or work in early life was associated with less white matter hyperintensities in late life.⁷ This is important because dementia prevention, particularly Alzheimer’s disease, may be most effective before the onset of brain pathology, which may precede expected dementia symptom onset by 20+ years.⁸ Further examination of this association requires measuring physical activity across the life course with a valid and reliable tool.

There are currently three measures of physical activity across the life course. The Historical Leisure Activity Questionnaire measures one domain of physical activity, i.e., leisure-time physical activity.⁹ The Modified Historical Leisure Activity Questionnaire added household physical activity to the Historical Leisure Activity Questionnaire.¹⁰ The Lifetime Total Physical Activity Questionnaire (LTPAQ)¹¹ assesses four domains; school/occupation, transportation, household, and leisure-time physical activity.¹² The validity of the LTPAQ has been examined in healthy adults age 50+^13–14 and in adults age 50+ with a history of late-onset depression,¹⁴ as one depressive symptom is a reduced capacity to focus,¹⁵ hindering the ability to remember information.

Evidence of construct validity of the LTPAQ has been found using hypothesis testing. The intensities within each of the domains in the LTPAQ were previously assessed in adults with no history of depression and those with a history of late-onset depression.¹⁴ Based on nationally representative norms which measured physical activity with metabolic equivalent of tasks (METs), their responses, which also measured physical activity with METs, tested the following hypotheses; 1) as the majority of a typical adult’s average day is spent engaged in sedentary activities, sedentary activities account for more of the day than light intensity activities, 2) light intensity activities account for more of the day than moderate intensity activities, 3) moderate intensity activities account for more of the day than vigorous intensity activities.^16–17 Using descriptive statistics, evidence of validity was found in both adults with no history of depression and those with a history of late-onset depression for moderate and vigorous intensities as they corresponded with nationally represented norms; moderate intensity activities accounted for more of the day than vigorous intensity activities.¹⁴ However, evidence of validity was not found for sedentary or light intensities in any of the domains as they contradicted nationally represented norms; sedentary activities accounted for less of the day than light intensity activities, and light intensity activities accounted for less of the day than moderate intensity activities.¹⁴ The construct validity of the intensities of the LTPAQ in adults with a history of depression, regardless of symptom onset, is not known.

Evidence of construct validity of an online version of the LTPAQ has also been found using the contrasted group approach; 1) males have higher participation in occupational activity than females; 2) females have higher participation in household activity than males; and 3) males have higher participation in sports/recreational activity than females.¹³ The contrasted group approach has not been assessed in face-to-face interviews conducted with the LTPAQ.

When determining the reliability of the LTPAQ administered to the same participants on two separate occasions, the test-retest reliability of all the intensities, domains, and developmental stages combined was 0.74 with Pearson correlation¹¹ and 0.88–0.89 with Intraclass Correlation Coefficient.¹³ Occupation combined with transportation had the highest correlation with 0.77 to 0.87, followed by household 0.67 to 0.77, and leisure-time with 0.72 to 0.89.^11,13 The reliability of transportation physical activity separate from occupation physical activity is not known.

The LTPAQ also measures physical activity across developmental stages such as school-age/childhood, adolescence, young/early adulthood, middle adulthood, and older adulthood.¹¹ However, the construct validity and reliability of the developmental stages in the LTPAQ have not been assessed despite their extensive use.^7,18–23 The purpose of this study was to assess the construct validity and test-retest reliability of the developmental stages in the LTPAQ in community-dwelling cognitively intact adults with no history of depression and those with a history of depression regardless of symptom onset.

Methods and Procedures

Participants

Community-dwelling cognitively intact adults were recruited from the Wisconsin Registry for Alzheimer’s Prevention and the Wisconsin Alzheimer’s Disease Research Center, where they underwent magnetic resonance imaging (MRI) to examine biomarkers associated with dementia.²⁴ Exclusion criteria comprised 1) MRI scanner incompatibility, 2) history of major psychiatric disease except for depression (i.e., schizophrenia, substance abuse), or 3) major medical conditions (history of neurologic disorders including prior head trauma with loss of consciousness, cancer requiring chemotherapy or radiation, insulin-dependent diabetes, and untreated hypertension), 4) abnormal cognitive testing (more than 1.5 standard deviation below normal on Rey Auditory Verbal Learning total learning or delayed free recall, or 2 standard deviations below IQ), or 5) abnormal structural MRI indicating disease or a condition that would affect normalization procedures (tumor or hydrocephalus). Prior to recruitment, approval was obtained from the Institutional Review Board at the University of Wisconsin-Madison (2011–0907-CR002). Written informed consent was obtained. All procedures were done in accord with the Helsinki Declaration of 1975.

Participant Characteristics

Participant characteristics were obtained at baseline. Age was measured in years. Gender was measured by self-report. Race was measured by self-report according to 2010 Census categories. Education was measured as completion of high school, some college (no degree), Associate degree (2-year college), Bachelor’s degree (4-year college), Master’s degree, or professional degree (i.e., JD, MD, PhD).

Depression

All participants were assessed for a history of current and past major depressive episodes, major depressive disorder, current and past manic and hypomanic episodes, and dysthymic disorder by the interviewer using the Structured Clinical Interview for Diagnostic and Statistical Manual of Mental Disorders 4^th edition (SCID-IV) Text Revision Axis I Disorders Research version Non-patient Edition.²⁵ The SCID-IV is a semi-structured interview guide for making DSM-IV diagnoses. The interviewer was a registered nurse trained by a psychiatrist in the administration of the SCID-IV.

Physical activity

Conduction of face-to-face interviews with the LTPAQ¹¹ and measurement of school/occupation, transportation, household, and leisure-time domains and intensities were completed as previously described.¹⁴ Whereas the original LTPAQ combined school/occupation and transportation physical activity,¹¹ the current study examined school/occupation and transportation physical activity separately,¹² as done in previous studies.^7,26 All interviews were completed by the same interviewer trained as a registered nurse in the lab of the creator of the LTPAQ to use cognitive interviewing skills while administering the LTPAQ.

Each activity was converted into energy expended by multiplying the hours spent by the metabolic equivalent of task (MET) of that activity abstracted from the 2011 Compendium of Physical Activities, which characterizes sedentary behavior as 1.0–1.5 METs, light-intensity as 1.6–2.9 METs, moderate-intensity as 3.0–5.9 METs, and vigorous-intensity as ≥6 METs.²⁷ The average MET hours/week/year were calculated as previously described.¹⁴ The interviewer was trained in the lab of the creator of the LTPAQ to calculate the average MET hours/week/year. Values were subsequently stratified by developmental stages based on Erikson’s Theory of Psychosocial Development;²⁸ school-age (6–11), adolescence (12–18), young (19–39), middle (40–64), and older adulthood (≥65).

Statistics

Results were stratified by history of depression vs. no history of depression. Chi-square was performed to determine if there was a statistically significant difference in gender or education between the two groups. Fisher’s exact test was performed to determine if there was a statistically significant difference in race between the two groups. A T-test was performed to determine if the two groups had a statistically significant difference in age.

Using descriptive statistics, evidence of construct validity of each of the intensities within each of the domains within each of the developmental stages in the LTPAQ was assessed with hypothesis testing. Based on nationally representative norms, the following hypotheses were tested; 1) sedentary activities account for more of the day than light intensity activities, 2) light intensity activities account for more of the day than moderate intensity activities, 3) moderate intensity activities account for more of the day than vigorous intensity activities.^16–17

Using descriptive statistics, evidence of construct validity was also assessed using the contrasted group approach; 1) males will have higher participation in school/occupational activity than females; 2) females will have higher participation in household activity than males; and 3) males will have higher participation in leisure-time activity than females.¹³

A subset of participants agreed to be interviewed with the LTPAQ two years after the initial interview. The test-retest reliability of the intensities and domains across developmental stages of the two time periods in the LTPAQ was assessed with the Intraclass Correlation Coefficient (ICC) using a 2-way mixed-effects with absolute agreement and a single measurement.²⁹ ICC values were interpreted as follows: ICC < .5 is indicative of poor reliability; .5 < ICC < .75 as moderate reliability, .75 < ICC < .9 as good reliability, and ICC > .9 as excellent reliability.²⁹ All analyses were performed using SPSS (version 29).

Results

Ninety-nine individuals consented to the interview. One participant did not complete the study due to a family emergency during the LTPAQ administration and was excluded from the analysis, resulting in a final sample size of 98. Sixty-six percent of the sample was female, 93% were White with an average age of 62, and 65% had a bachelor’s degree or higher (see Table 1). Thirty-two percent of the sample had a history of depression. With regards to the age of first depression onset, 10% was during childhood, 10% during adolescence, 32% during young adulthood, 45% during middle-adulthood, and 3% unknown. Of those with a history of depression, 13% had current depression. Of those with current depression, 75% were being treated with medication, 25% were seeing a therapist. None had a current or past manic or hypomanic episode, or dysthymic disorder. There was no significant difference in gender (Χ² = 3.3, p = .07), race (p = .20), education (Χ² = 6.0, p = .31) or age (t = −.27, p = .79) between those with vs. without a history of depression. As only 39% of the sample were age 65+, the older adult developmental stage was omitted from further analyses.

Table 1.

Sample Characteristics

Baseline Characteristics	Total (n = 98)	History of depression (n = 31)	No History of depression (n = 67)
	No. (%)
Gender
Women	65 (66.3)	25 (80.6)	40 (59.7)
Men	33 (33.7)	6 (19.4)	27 (40.3)
Race
White	91 (92.9)	27 (87.1)	64 (95.5)
Black	7 (7.1)	4 (12.9)	3 (4.5)
Education
High school	12 (12.2)	2 (6.5)	10 (14.9)
Some college (no degree)	6 (6.2)	1 (3.2)	5 (7.5)
Associate degree (2-year college)	16 (16.3)	4 (12.9)	12 (17.9)
Bachelor’s degree (4-year college)	30 (30.6)	14 (45.2)	16 (23.8)
Master’s degree	28 (28.6)	9 (29.0)	19 (28.4)
Professional degree (i.e., JD, MD, PhD)	6 (6.1)	1 (3.2)	5 (7.5)
	Mean (SD)
Age	62.1 (5.9)	62.4 (7.0)	62.0 (5.4)
Type of Activity measured in MET hours/week/year
SCHOOL-AGE
School/Occupation activity
Sedentary behavior	0.0 (0.2)	0.0 (0.1)	0.0 (0.2)
Light	5.2 (1.1)	5.2 (1.5)	5.2 (0.9)
Moderate	1.1 (1.6)	1.2 (1.4)	1.1 (1.7)
Vigorous	0.0 (0.4)	0.2 (0.7)	0
Transportation physical activity
Light	0.0 (0.1)	0.0 (0.1)	0.1 (0.2)
Moderate	0.5 (0.6)	0.7 (0.8)	0.4 (0.4)
Vigorous	0.0 (0.1)	0	0.0 (0.1)
Household physical activity
Light	0.3 (0.8)	0.4 (0.8)	0.3 (0.8)
Moderate	1.4 (3.1)	1.4 (2.5)	1.4 (3.4)
Vigorous	0	0	0
Leisure-time physical activity
Light	0.0 (0.1)	0.0 (0.0)	0.0 (0.2)
Moderate	6.4 (6.4)	4.6 (4.5)	7.2 (7.0)
Vigorous	0.9 (1.9)	1.0 (2.2)	0.8 (1.7)
ADOLESCENT
School/Occupation activity
Sedentary behavior	0.3 (0.6)	0.2 (0.4)	0.4 (0.7)
Light	4.5 (1.0)	4.3 (1.0)	4.5 (1.0)
Moderate	7.2 (3.3)	7.3 (3.0)	7.1 (3.5)
Vigorous	0.6 (1.9)	0.9 (2.7)	0.4 (1.4)
Transportation physical activity
Light	0.1 (0.1)	0.0 (0.1)	0.0 (0.1)
Moderate	0.5 (0.5)	0.5 (0.5)	0.5 (0.4)
Vigorous	0.0 (0.1)	0	0.0 (0.2)
Household physical activity
Light	0.5 (1.0)	0.5 (1.0)	0.5 (1.0)
Moderate	1.8 (3.6)	1.8 (2.8)	1.8 (3.9)
Vigorous	0	0	0
Leisure-time physical activity
Light	0.1 (0.4)	0.0 (0.2)	0.1 (0.5)
Moderate	2.9 (3.8)	2.6 (3.6)	3.1 (3.9)
Vigorous	1.8 (2.8)	2.3 (3.6)	1.6 (2.3)
YOUNG ADULT
School/Occupation activity
Sedentary behavior	9.0 (7.7)	9.1 (9.0)	8.9 (7.2)
Light	3.3 (3.9)	3.0 (2.2)	3.4 (4.5)
Moderate	22.9 (22.2)	20.5 (16.8)	24.0 (24.4)
Vigorous	2.5 (8.1)	0.1 (0.5)	3.6 (9.6)
Transportation physical activity
Light	0.0 (0.1)	0.0 (0.1)	0.0 (0.1)
Moderate	0.9 (2.3)	1.2 (3.6)	0.8 (1.2)
Vigorous	0.1 (0.9)	0.0 (0.2)	0.2 (1.1)
Household physical activity
Light	6.2 (13.6)	4.6 (10.9)	7.0 (14.7)
Moderate	15.0 (24.7)	14.8 (22.5)	15.1 (25.8)
Vigorous	0.0 (0.1)	0	0.0 (0.1)
Leisure-time physical activity
Light	0.3 (1.2)	0.1 (0.3)	0.4 (1.4)
Moderate	1.4 (2.0)	1.5 (1.6)	1.3 (2.1)
Vigorous	2.8 (5.1)	3.5 (5.8)	2.5 (4.7)
MIDDLE ADULT
School/Occupation activity
Sedentary	11.3 (9.2)	9.1 (7.6)	12.4 (9.7)
Light	1.4 (3.4)	1.1 (2.5)	1.5 (3.8)
Moderate	18.6 (20.6)	21.7 (19.4)	17.2 (21.2)
Vigorous	2.8 (11.7)	0.9 (4.3)	3.7 (13.7)
Transportation physical activity
Light	0.0 (0.0)	0.0 (0.1)	0.0 (0.0)
Moderate	0.3 (1.2)	0.3 (1.3)	0.4 (1.2)
Vigorous	0.2 (1.1)	0.2 (1.0)	0.2 (1.1)
Household physical activity
Light	4.9 (8.8)	3.0 (6.5)	5.8 (9.6)
Moderate	10.9 (14.6)	9.9 (11.9)	11.3 (15.8)
Vigorous	0.0 (0.0)	0	0.0 (0.0)
Leisure-time physical activity
Light	0.3 (0.9)	0.2 (0.4)	0.4 (1.1)
Moderate	2.8 (3.8)	3.3 (4.2)	2.6 (3.6)
Vigorous	3.8 (8.4)	5.6 (12.2)	3.0 (5.8)

Note: Participant gender, race, education, or age did not differ between those with and without a history of depression

Construct Validity with Hypothesis Testing

For the total sample, during school-age there was less reported school/occupation sedentary (0) than light intensities (5.2), contradicting norms, but more reported light than moderate (1.1) and moderate than vigorous school/occupation intensities (0), coinciding with norms. There was less reported light (0–0.3) than moderate intensities (0.5–6.4), contradicting norms, but more reported moderate than vigorous transportation, household, and leisure-time physical activity (0–0.9), coinciding with norms. Results did not differ by group (see Table 1). Thus, evidence of construct validity was found for light school/occupation physical activity, as well as moderate and vigorous intensities in all the domains during school-age in those with and without a history of depression since these results coincide with nationally representative norms.

For the total sample, during adolescence, there was less reported school/occupation sedentary (0.3) than light activities (4.5) and less light than moderate intensity school/occupation (4.5 vs. 7.2), transportation (0.1 vs.0.5), household (0.5 vs. 1.8), and leisure-time physical activity (0.5 vs. 1.8) respectively, contradicting norms. There was more reported moderate than vigorous intensity school/occupation (7.2 vs. 0.6), transportation (0.5 vs.0.0), household (1.8 vs. 0), and leisure-time physical activity (2.9 vs. 1.8) respectively, coinciding with norms. Results did not differ by group (see Table 1). Thus, evidence of construct validity was found for moderate and vigorous intensities in all domains during adolescence in those with and without a history of depression since these results coincide with nationally representative norms.

For the total sample, during young and middle adulthood, even though there was more reported school/occupational sedentary (9–11.3) than light intensities (1.4–3.3), coinciding with norms, there was less reported sedentary than moderate intensities (18.6–22.9), contradicting norms. There was less reported light than moderate intensity school/occupation (1.4–3.3 vs. 18.6–22.9), transportation (0.0 vs.0.3–0.9), household (4.9–6.2 vs. 10.9–15.0), and leisure-time physical activity (0.3 vs. 1.4–2.8) respectively, contradicting norms. There was more reported moderate than vigorous intensity school/occupational (18.6–22.9 vs. 2.5–2.8), transportation (0.3–0.9 vs. 0.1–0.2) and household physical activity (10.9–15.0 vs. 0.0) respectively, coinciding with norms, but less reported moderate than vigorous intensity leisure-time physical activity (1.4–2.8 vs. 2.8–3.8) respectively, contradicting norms. Results did not differ by group (see Table 1). Thus, evidence of construct validity was found for moderate and vigorous-intensity school/occupational, transportation and household during young and middle adulthood in those with and without a history of depression since these results coincide with nationally representative norms.

Construct Validity with Contrasted Group Approach

Since no group differences were found with hypothesis testing, subsequent analyses combined the two groups. Since no evidence of construct validity with hypothesis testing was found for sedentary intensities across any of the domains during any of the developmental stages, sedentary intensities were omitted from the contrasted group approach. Since evidence of construct validity with hypothesis testing for light intensities was only found for school/occupation during school-age, light intensities were omitted from the contrasted group approach. Consistent with our a priori hypothesis, males reported higher participation in school/occupational activity than females during school-age (1.2 vs. 1.1, respectively), adolescence (7.9 vs. 7.7, respectively), and young adulthood (33.2 vs. 21.5, respectively), but not middle adulthood (20.1 vs. 22.2, respectively). Consistent with our a priori hypothesis, females reported higher participation in household activity than males during each developmental stage (see Table 2). Consistent with our a priori hypothesis, males reported higher participation in leisure-time activity than females during each developmental stage. Thus, evidence of construct validity was found for school/occupation physical activity during school-age, adolescence, and young adulthood; and household and leisure-time physical activity during all developmental stages.

Table 2.

Contrasted Group Analysis of Gender and Physical Activity Domains across Developmental Stages

Baseline Characteristics	Males (n = 33)	Females (n = 65)
Type of Activitya	Mean (SD)
SCHOOL-AGE
School/Occupation activity	1.2 (1.4)	1.1 (1.8)
Household physical activity	1.1 (3.4)	1.5 (2.9)
Leisure-time physical activity	7.9 (6.4)	6.9 (6.9)
ADOLESCENT
School/Occupation activity	7.9 (3.7)	7.7 (4.2)
Household physical activity	1.2 (2.8)	2.1 (3.9)
Leisure-time physical activity	6.1 (5.1)	4.1 (4.1)
YOUNG ADULT
School/Occupation activity	33.2 (33.7)	21.5 (17.0)
Household physical activity	2.3 (4.6)	21.5 (28.0)
Leisure-time physical activity	6.4 (6.6)	3.0 (4.9)
MIDDLE ADULT
School/Occupation activity	20.1 (26.5)	22.2 (23.4)
Household physical activity	6.3 (10.4)	13.2 (15.9)
Leisure-time physical activity	8.9 (9.3)	5.5 (9.4)

^aPhysical activity measured in MET hours/week/year

Test-retest Reliability

Nineteen individuals consented to be interviewed two years after the initial interview (see Table 3). During school-age there was moderate reliability for moderate intensity school/occupation (ICC = .562, p = .006) and excellent reliability for moderate intensity household (ICC = .945, p < .001). No vigorous intensity school/occupation and household physical activity were reported at either time points (see Tables 3–4).

Table 3.

Physical activity characteristics across two interviews

	Average hours per week of activity
	First measurement			Second measurement
Type of Activitya	Mean	SD	Range	Mean	SD	Range
School-Age
School/Occupation Sedentary	0	0	0	0.0	0.1	0.3
School/Occupation Light	4.9	0.9	3.2	5.0	1.0	3.7
School/Occupation Moderate	1.0	1.4	4.7	1.0	1.4	5.7
School/Occupation Vigorous	0	0	0	0	0	0
Transportation Light	0.1	0.2	0.6	0.0	0.2	0.8
Transportation Moderate	0.4	0.5	1.6	1.0	2.0	8.8
Transportation Vigorous	0	0	0	0.0	0.1	0.4
Household Light	0.1	0.5	2.1	0.2	0.7	2.9
Household Moderate	0.9	2.4	10.1	0.6	2.5	10.9
Household Vigorous	0	0	0	0	0	0
Leisure-time Light	0	0	0	0.0	0.1	0.5
Leisure-time Moderate	4.7	5.0	18.0	7.3	7.0	25.8
Leisure-time Vigorous	1.0	1.9	6.2	0.2	0.4	1.2
Adolescence
School/Occupation Sedentary	0.2	0.4	1.3	0.1	0.2	0.7
School/Occupation Light	4.2	0.6	2.0	4.7	0.7	3.1
School/Occupation Moderate	6.0	2.2	7.7	6.4	2.7	11.9
School/Occupation Vigorous	0.4	1.4	6.1	0	0	0
Transportation Light	0.0	0.1	0.6	0.0	0.1	0.3
Transportation Moderate	0.6	0.5	1.5	1.6	3.9	17.5
Transportation Vigorous	0	0	0	0	0	0
Household Light	0.5	0.8	2.4	0.1	0.3	0.9
Household Moderate	1.0	2.9	12.4	0.9	2.0	7.8
Household Vigorous	0	0	0	0.0	0.1	0.4
Leisure-time Light	0.0	0.1	0.2	0.0	0.1	0.3
Leisure-time Moderate	2.3	1.9	7.5	2.9	3.5	16.3
Leisure-time Vigorous	2.1	2.5	9.3	1.2	1.6	7.0
Young Adult
School/Occupation Sedentary	8.8	6.1	21.8	8.4	6.1	17.0
School/Occupation Light	3.4	3.2	14.2	4.3	3.9	14.1
School/Occupation Moderate	20.4	15.4	45.0	23.0	18.8	61.4
School/Occupation Vigorous	2.6	7.3	27.4	0.1	0.5	1.6
Transportation Light	0.0	0.1	0.3	0	0	0
Transportation Moderate	1.0	1.4	5.0	0.6	0.8	3.4
Transportation Vigorous	0	0	0	0.0	0.0	0.2
Household Light	9.2	15.5	52.8	4.0	10.0	31.1
Household Moderate	8.8	17.8	72.0	13.0	22.5	59.0
Household Vigorous	0	0	0	0	0	0
Leisure-time Light	0.8	2.3	10.0	0.1	0.5	2.0
Leisure-time Moderate	1.0	1.2	4.5	1.0	1.2	3.8
Leisure-time Vigorous	2.4	4.1	15.3	1.9	4.2	16.0
Middle Adult
School/Occupation Sedentary	13.6	9.7	30.8	10.3	9.9	28.7
School/Occupation Light	0.2	0.8	3.2	0.7	2.5	10.9
School/Occupation Moderate	18.9	22.7	71.0	21.7	22.8	68.1
School/Occupation Vigorous	1.4	4.1	14.5	0.3	1.4	6.2
Transportation Light	0	0	0	0	0	0
Transportation Moderate	0.6	1.7	7.1	0.6	1.8	7.8
Transportation Vigorous	0	0	0	0	0	0
Household Light	8.0	10.8	34.0	1.7	4.3	18.3
Household Moderate	9.2	15.0	40.7	6.6	10.2	32.2
Household Vigorous	0	0	0	0	0	0
Leisure-time Light	0.3	0.9	4.0	0.2	0.3	1.1
Leisure-time Moderate	4.3	5.9	19.5	3.8	5.2	20.0
Leisure-time Vigorous	6.3	14.2	57.0	3.2	5.3	18.2

Note: Two repeat administrations of questionnaire were completed, on average, two years apart. Physical activity for the second interview was determined through the end of the first interview, so both interviews covered the same time period.

^aPhysical activity measured in MET hours/week/year

Table 4.

Intraclass correlations between physical activity domains and intensities across developmental stages between the two interviews

Physical Activity Domain	Intensity	School-age (6–11)	Adolescence (12–18)	Young Adult (19–39)	Middle Adult (40–64)
School/Occupation	Sedentary 95% CI	.00 [−.44, .44]	.36 [−.10, .70]	.78** [.52, .91]	.80** [.48, .92]
	Light 95% CI	.31 [−.18, .66]	.31 [−.08, .65]	.86** [.65, .95]	.18 [−.29, .58]
	Moderate 95% CI	.56* [.15, .81]	.61* [.22, .83]	.84** [.65, .94]	.83** [.62, .93]
	Vigorous 95% CI	_____	.00 [−.44, .44]	−.01 [−.42, .42]	−.05 [−.48, .40]
Transportation	Light 95% CI	−.10 [−.56, .38]	−.05 [−.50, .40]	.00 [−.43, .44]	_____
	Moderate 95% CI	−.02 [−.44, .42]	.04 [−.39, .47]	.75** [.45, .89]	.98** [.94, .99]
	Vigorous 95% CI	.00 [−.44, .44]	_____	.00 [−.44, .44]	_____
Household	Light 95% CI	.03 [−.45, .48]	.43 [.01, .73]	.32 [−.11, .66]	.32 [−.07, .65]
	Moderate 95% CI	.95** [.87, .98]	.83** [.61, .93]	.61* [.23, .83]	−.17 [−.60, .31]
	Vigorous 95% CI	_____	.00 [−.44, .44]	_____	_____
Leisure-time	Light 95% CI	.00 [−.45, .45]	.42 [−.01, .73]	.34 [−.09, .68]	−.03 [−.48, .43]
	Moderate 95% CI	.21 [−.22, .59]	.02 [−.45, .47]	.88** [.71, .95]	.68** [.33, .86]
	Vigorous 95% CI	−.07 [−.44, .36]	.12 [−.32, .52]	.80** [.56, .92]	.07 [−.39, .50]

Note: Test-retest reliability assessed with intraclass correlation coefficient of MET hours/week/year on two repeat administrations of questionnaire completed, on average, two years apart. Physical activity for the second interview was determined through the end of the first interview, so both interviews covered the same time period. CI = confidence interval; empty cells indicate both time points had no activity reported, therefore ICC could not be determined.

^*p < .05.

^**p < .001.

During adolescence, there was moderate reliability for moderate intensity school/occupation (ICC = .605, p = .003) and good reliability for moderate intensity household (ICC = .829, p < .001). No vigorous intensity transportation physical activity was reported at either time points.

During young adult there was moderate reliability for moderate intensity household (ICC = .608, p = .002) and transportation (ICC = .747, p < .001); good reliability for sedentary (ICC = .784, p < .001), light (ICC = .862, p < .001), and moderate intensity school/occupation (ICC = .844, p < .001); and good reliability for moderate (ICC = .879, p < .001) and vigorous intensity leisure-time (ICC = .803, p < .001). No vigorous intensity household physical activity was reported at either time points.

During middle adult, there was good reliability for sedentary (ICC = .795, p < .001) and moderate intensity school/occupation (ICC = .830, p < .001); excellent reliability for moderate intensity transportation (ICC = .977, p < .001); and moderate reliability for moderate intensity leisure-time (ICC = .676, p = .001). No light and vigorous intensity transportation, and vigorous intensity household physical activity were reported at either time points.

Discussion

The construct validity of the developmental stages in the LTPAQ assessed with hypothesis testing did not find any differences between adults with and without a history of depression, coinciding with a previous similar study of adults with a history of late-onset depression.¹⁴ Only 13% had current depression, all of whom were being treated with medication or therapy. It is important to account for a history of depression because one depressive symptom is a reduced capacity to focus,¹⁵ hindering the ability to remember information. The current study adds to the literature by examining the construct validity of the LTPAQ in adults with a history of depression regardless of symptom onset.

Evidence of construct validity (through hypothesis testing and contrasted group) and test-retest reliability was found during school-age for moderate and vigorous intensity school/occupation and household physical activity; during adolescence for moderate intensity school/occupation and household, and vigorous intensity transportation physical activity; during young adulthood for moderate intensity school/occupation, transportation and household, and vigorous intensity household physical activity; and during middle adulthood for moderate and vigorous intensity transportation, and vigorous intensity household physical activity. These results coincide with previous studies.^11,13–14 Whereas previous studies combined school/occupation and transportation physical activity,^11,13 the currently study examined school/occupation separate from transportation.

Evidence of construct validity (through hypothesis testing and contrasted group) and test-retest reliability was not found during school-age for any intensity transportation and leisure-time physical activity; during adolescence for vigorous school/occupation and household, moderate intensity transportation, or any leisure-time physical activity; during young adulthood for vigorous school/occupation and transportation, or any intensity leisure-time physical activity; and during middle adulthood for moderate household, or any intensity school/occupation or leisure-time physical activity. These results contradict a previous study that did not assess intensities and domains across developmental stages.¹⁴

In the current study, no evidence of construct validity (through hypothesis testing and contrasted group) and test-retest reliability was found for sedentary and light intensities across any of the domains during any of the developmental stages in the LTPAQ. These results coincide with a previous study that found no evidence of construct validity for sedentary and light intensities in the LTPAQ, although that study did not assess intensities across developmental stages.¹⁴ These results are not surprising, given that the capacity to correctly describe sedentary and light intensities is often weak.^30–31

Limitations and Strengths

Findings are limited by self-report measures as there were no objective measures of physical activity across the life course in this sample. All self-report physical activity measures risk recall bias such as over-reporting.³² Participants completed a recall calendar in advance for use during the interview. In addition, in advanced age long-term memory may be better preserved than recent recollections.³³ Since our study conducted face-to-face interviews, our results may not be generalizable to self-administered or online surveys. Results are specific to the target group for the instrument,³⁴ participants in a neuroimaging project examining physical activity across the life course and areas of the brain related to Alzheimer’s disease and related disorders. The nationally representative norms were based on self-report physical activity in young, middle and older adults at the time of the study,^16–17 not their recollection of physical activity during school-age or adolescence. However, objective measures in a nationally representative sample of children and adolescents found similar results. Younger children were more active than older children and adolescents, with more time spent in moderate than vigorous physical activity.³⁵ Finally, future studies should be conducted with larger and more ethnically diverse samples.

Despite limitations, to our knowledge, this is the first study to assess the validity and reliability of the face-to-face interview administered LTPAQ in adults with a history of depression regardless of symptoms onset. Our study examined occupation and transportation physical activity separately. This study is also one of the few studies to assess the validity and reliability of the developmental stages in the LTPAQ despite its extensive use.^7,18–23 This is important because there is increasing recognition of the need to study benefits of physical activity from a life course perspective,³⁶ particularly Alzheimer’s disease where interventions may be most effective before the onset of brain pathology, which may precede dementia symptom onset by 20+ years.⁸

Conclusion

Evidence of construct validity (through hypothesis testing and contrasted group) and test-retest reliability was found during school-age for moderate and vigorous intensity school/occupation and household physical activity; during adolescence for moderate intensity school/occupation and household, and vigorous intensity transportation physical activity; during young adulthood for moderate intensity school/occupation, transportation and household, and vigorous intensity household physical activity; and during middle adulthood for moderate and vigorous intensity transportation, and vigorous intensity household physical activity. The LTPAQ may be used to examine physical activity earlier in life in community-dwelling cognitively intact adults with and without a history of depression before the onset of brain pathology and risk for Alzheimer’s disease later in life.

Data availability

The data supporting the findings of this study are available on request from https://www.adrc.wisc.edu/apply-resources