Feasibility of 24-Hr Urine Collection for Measurement of Biomarkers in Community-Dwelling Older Adults

Abstract

Biologic markers are becoming a key part of gerontological research, including their measurement at multiple intervals to detect changes over time. This report examined the feasibility and quality of 24-hr urine collection to measure neuroendocrine biomarkers in a community-based sample of older caregivers and non-caregivers. At each interview, participants were instructed on the correct method to collect and store the sample. As incentives, participants selected a day for urine collection within 5 days of the interview, received a reimbursement, and study staff travelled to their home to retrieve the specimen. Between 2008 and 2013, 256 participants were enrolled; all but two participants (99%) provided a baseline urine specimen, of which 93% were considered adequate. Urine collection and quality remained high over three annual follow-up interviews and did not vary by caregiver status or perceived stress level. Our results indicate that 24-hr urine collection is feasible in active, community-dwelling older adults.

Keywords

biological samples caregiving longitudinal study

Introduction

Neuroendocrine biomarkers are important for understanding the health effects of stress (Chrousos, 2000; Lutgendorf & Costanzo, 2003; McEwen, 1998; Wurtman, 2002), particularly among adults under chronic stress, such as family caregivers (Bauer et al., 2000; Kiecolt-Glaser et al., 2003). Most endocrinologists advocate the use of 24-hr urine for the assessment of these biomarkers. A 24-hr urine sample provides an integrated measure of the biomarker over a 1-day period, accounting for diurnal variations, periods of activity and sleep, and averages across stressful and non-stressful encounters (Rosmond & Bjorntorp, 2000; Steptoe, Cropley, Griffith, & Kirschbaum, 2000; Vedhara et al., 1999). However, epidemiologic studies, including those of older adults, have generally used 12-hr urine collection (Goldman et al., 2004; Masi, Rickett, Hawkley, & Cacioppo, 2004; Reuben, Talvi, Rowe, & Seeman, 2000), spot urines (Sluiter, Frings-Dresen, van der Beek, Meijman, & Heisterkamp, 2000), and other methods such as saliva (Bauer et al., 2000), because they are assumed to be less taxing for the participants. Projected challenges to collecting 24-hr urine include acceptability, flexibility with participants’ schedules, convenience of sample retrieval, and quality of specimens. There is little information on the practicality of obtaining 24-hr specimens in studies of community-based older adults (Neuhouser et al., 2008; Schatzkin et al., 2003; Young et al., 1992). Therefore, the present study aimed to evaluate the feasibility and usability of 24-hr urine collection in a sample of older community-dwelling caregivers and non-caregivers.

Two main homeostatic systems that are disrupted by chronic stress are the hypothalamic-pituitary-adrenal (HPA) axis and the central sympathetic nervous system. Dysregulation of these systems is indicated by elevated cortisol and higher levels of the catecholamines epinephrine and norepinephrine. In studies of the effect of stress in older adults, cortisol has been assayed from 12-hr (i.e., overnight) urine (Goldman et al., 2004; Masi et al., 2004; Seeman, McEwen, Singer, Albert, & Rowe, 1997), saliva (Bauer et al., 2000; Kudielka, Buske-Kirschbaum, Hellhammer, & Kirschbaum, 2004; Vedhara et al., 1999), and plasma (Kudielka et al., 2004) samples. Twelve-hour urine collection also has been used to measure catecholamines in the MacArthur Studies of Successful Aging (Karlamangla, Singer, Greendale, & Seeman, 2005; Reuben et al., 2000). Because the excretion of catecholamines is not uniform over the course of the day (Schöfl, Becker, Prank, von Zur Mühlen, & Brabant, 1997), a 24-hr urine sample provides a more integrated measure of epinephrine and norepinephrine levels. The same is true for cortisol (Steptoe et al., 2000), whose levels also fluctuate during the day. Only a few published studies have measured catecholamines and cortisol in 24-hr urine samples from older, community-dwelling adults (Lutgendorf et al., 2002; Otte, Neylan, Pipkin, Browner, & Whooley, 2005; Vogelzangs et al., 2007; Young et al., 1992).

We examined the feasibility and acceptability of obtaining a 24-hr urine specimen in the Health Pathways Study, a population-based cohort study of older caregivers and non-caregivers. The objective of this study was to investigate the relationships between caregiving-related stress, physiologic functioning, and health decline, with an emphasis on urinary and serologic biomarkers of stress and the development of metabolic syndrome. Participants were informed that they would be expected to collect a 24-hr urine sample at each of four annual interviews. We hypothesized that caregivers and individuals experiencing more stress would be less likely to provide a useable urine specimen. It has been proposed that more time caregiving is associated with less time engaging in self-care activities (Aurora & Wolf, 2014), and studies have shown that caregivers spend less time on preventive health behaviors than non-caregivers (Burton, Newsom, Schulz, Hirsch, & German, 1997; Fredman, Bertrand, Martire, Hochberg, & Harris, 2006). Thus, we anticipated that preoccupation with caregiving would interfere with completing the urine collection protocol. In the same way, participants with high-perceived stress, whether from caregiving or other circumstances, might find it too demanding to collect a 24-hr urine sample.

Method

Study Population

The Health Pathways Study is a prospective cohort study of caregivers and non-caregivers who were 60 years of age or older, lived in the Boston metropolitan area, spoke English as a primary language, and were free of cognitive impairment at baseline. Enrollment into the cohort began in May 2008 and ended in May 2013. Caregivers were defined as persons helping a relative or friend with Alzheimer’s disease (AD) or Parkinson’s disease (PD) with at least one instrumental activity of daily living (IADL) task (i.e., use the telephone, get to places out of walking distance, shop, prepare meals, manage medications, manage finances, do heavy housework) or at least one basic activity of daily living (ADL) task (i.e., walk across a room, groom, transfer from bed to chair, eat, dress, bathe, use the toilet). Non-caregivers were persons who had not helped a relative or friend with any IADL/ADL task during the year prior to the baseline interview. The study was approved by the Boston University Medical Campus Institutional Review Board.

Caregivers were recruited from the Boston University (BU) Alzheimer’s Disease Center and from the BU Parkinson’s Disease Center and newsletter. Non-caregivers were recruited from the Harvard Cooperative Program on Aging database of adults who have participated in research studies and through announcements in local newspapers, the Massachusetts Department of Elder Affairs, and other local services. The study coordinator made clear to all potential participants that 24-hr urine collection would be required at all study interviews.

Data Collection

Interview

Data collection took place at the BU General Clinical Research Unit (GCRU) in the early morning to collect fasting blood samples and to control for diurnal variation in some biological parameters. Trained study staff conducted structured face-to-face interviews and physical performance measures; GCRU nurses obtained anthropometric measures and sitting blood pressure. Participants were interviewed annually, for up to three follow-up visits, through May 2014. Participants who were unable to travel to the GCRU for a follow-up interview were offered a home interview.

24-hr urine collection

At the baseline interview, a GCRU nurse instructed the participant on the correct method to collect and store the 24-hr urine specimen and gave them a collection container and a written set of instructions. Specific collection instructions included the following elements: not to collect the first void of the day; to collect all subsequent specimens in the container provided; to keep the container in the refrigerator or on ice; to continue saving all specimens for the next 24-hr, ending at the same time as the participant started the previous day; and, when finished, to store the bottle in the refrigerator until a study staff member comes to collect it. As incentives for urine collection, the participant selected the day for the collection within 5 days of the interview and received a monetary reimbursement. In addition, study staff retrieved the urine specimen from the participant’s home and transported it on ice to the GCRU for processing and storage of aliquots at −80°C. For the follow-up interviews, participants were asked if they preferred to complete the urine collection the day before the interview at the GCRU, in which case they were mailed a collection jug and instructions as well as ice packs and a fabric cooler bag for transport to the interview. When an in-home interview was performed, a collection container and instructions were mailed to the participant. Study staff then retrieved the sample at the home visit.

To assess whether a urine specimen represented a 24-hr collection, creatinine levels were measured in a 10-ml aliquot of the specimen. A urine sample was considered inadequate if the creatinine level was below 0.4 g/24 hr for women and 0.6 g/24 hr for men (ARUP Laboratories, http://ltd.aruplab.com/tests/Pub/0020473) or the volume collected was less than 500 ml.

Perceived stress

At the baseline and follow-up interviews, perceived stress experienced in the past month was measured using the 14-item Perceived Stress Scale (PSS; Cohen, Kamarck, & Mermelstein, 1983). The PSS reflects how much respondents’ experiences are perceived as demanding beyond their ability to cope in their life as a whole or in specific situations (Cohen, 1986). Possible scores range from 0 to 56, with higher scores indicating more stress.

Statistical Analysis

Baseline characteristics were compared between caregivers and non-caregivers using chi-square or Fisher’s exact statistic for categorical variables and t test for continuous variables applied to evaluate the associations of caregiver status and PSS with urine sample usability, respectively.

Results

During study enrollment, 330 individuals (148 caregivers, 182 non-caregivers) were screened and found eligible for the study (Figure 1). Seventy-four (22.4%) of those who were eligible chose not to participate; 54 refused participation at the initial telephone screening, and 20 never attended their scheduled baseline interview. Only seven (2.1%) individuals declined to participate because collection of a urine and/or blood sample was required for the study. The nonparticipation rate was higher for caregivers (26%) than for non-caregivers (20%), although the difference was not statistically significant (p = .20). Caregivers who did not participate were more likely than non-caregivers who did not participate to be male (26% vs. 15%) and younger (M age, 72.8 vs. 79.0 years).

Figure 1.

Study enrollment and follow-up.

In total, 256 participants (110 caregivers, 146 non-caregivers) completed a baseline interview and were enrolled in the study. At baseline, the mean age of participants was 73.4 (±7.9) years, 69% were female, 84% were White, non-Hispanic, and 64% had attended college (Table 1). Caregivers were more likely than non-caregivers (p < .01) to be younger, of White, non-Hispanic race, and married and to have an annual household income of at least US$30,000, as well as to have attended college (p = .04); they did not differ with respect to gender. Caregivers had higher perceived stress scores than did non-caregivers (p < .01).

Table 1.

Baseline Characteristics of Participants in the Health Pathways Study, by Caregiver Status.

	Total	Caregivers	Non-caregivers
Characteristic	(n = 256)	(n = 110)	(n = 146)	p value^a
Age, years
M ± SD (range)	73.4 ± 7.9 (60.0-97.8)	71.3 ± 7.3 (60.0-89.3)	75.2 ± 7.8 (60.0-97.8)	<.0001
Gender (%)
Male	79 (30.9%)	35 (31.8%)	44 (30.1%)	.77
Female	177 (69.1%)	75 (68.2%)	102 (69.9%)
Race (%)
White, non-Hispanic	215 (84.0%)	102 (92.7%)	113 (77.4%)	<.0001
White, Hispanic	2 (0.8%)	1 (0.9%)	1 (0.7%)
Black, non-Hispanic	37 (14.5%)	6 (5.5%)	31 (21.2%)
Black, Hispanic	1 (0.4%)	1 (0.9%)	0
Refused	1 (0.4%)	0	1 (0.7%)
Marital status (%)
Married	143 (55.9%)	97 (88.2%)	46 (31.5%)	<.0001
Widowed	34 (13.3%)	2 (1.8%)	32 (21.9%)
Divorced/separated	57 (22.3%)	8 (7.3%)	49 (33.6%)
Never married	22 (8.6%)	3 (2.7%)	19 (13.0%)
Education level (%)
Less than high school	11 (4.3%)	2 (1.8%)	9 (6.2%)	<.0001
High school	29 (11.3%)	7 (6.4%)	22 (15.1%)
Vocational school/some collage	59 (23.1%)	25 (22.7%)	34 (23.3%)
College	63 (24.6%)	27 (24.6%)	36 (24.7%)
Graduate school	94 (36.7%)	49 (44.6%)	45 (30.8%)
Household income (%)
<US$30,000	72 (28.1%)	10 (9.1%)	62 (42.5%)	<.0001
US$30,000-US$69,999	102 (39.8%)	53 (48.2%)	49 (33.6%)
≥US$70,000	75 (29.3%)	45 (40.9%)	30 (20.6%)
Refused/do not know	7 (2.7%)	2 (1.8%)	5 (3.4%)
Perceived stress score
M ± SD (range)	16.9 ± 8.1 (2.0-42.0)	19.6 ± 7.6 (2.0-41.0)	14.8 ± 7.8 (2.0-42.0)	<.0001

p value for comparison of caregivers with non-caregivers; chi-square or Fisher’s exact test for categorical variables and t test for continuous variables.

At the baseline interview, almost all participants (N = 254, 99%) collected a 24-hr urine sample, and 97% provided a blood sample (Table 2). Only two non-caregivers (1.4%) did not provide a urine specimen. The mean time between the interview and urine collection was approximately 2 days, with a range from 0 to 28 days; the four specimens that were obtained more than 12 days after the interview occurred over the university’s holiday intersession period during which the GCRU was closed and could not process urine specimens. Based on measured creatinine levels and sample volume, urine samples from six caregivers (5.5%) and 11 non-caregivers (7.6%) were considered inadequate.

Table 2.

Specimen Collection in the Health Pathways Study, by Caregiver Status.

	N	Caregivers	Non-caregivers	p value^a
Baseline specimen collection
n interviewed	256	110	146
Urine samples
n obtained (%)	254 (99.2)	110 (100)	144 (98.6)	.32
n inadequate (%)	17 (6.7)	6 (5.5)	11 (7.6)	.16
Blood samples
n obtained (%)	248 (96.9)	106 (96.4)	142 (97.3)	.26
Follow-up specimen collection
n interviewed	508	205	303
Home interviews	30	15	15
Phone interviews	5	1	4
Urine samples
n obtained (%)	489 (96.3)	195 (95.1)	294 (97.0)	.34
n inadequate (%)	7 (1.4)	1 (0.51)	6 (2.0)	.25
Blood samples
n obtained (%)	482 (95.4)	193 (94.1)	289 (95.4)	.55

p value for comparison of caregivers with non-caregivers, based on chi-square or Fisher’s exact test.

A total of 508 follow-up interviews were conducted (205 among caregivers, 303 among non-caregivers); 130 participants completed three follow-up visits (48 among caregivers and 82 among non-caregivers, respectively), 43 completed two (25 and 18, respectively), and 32 completed one (11 and 21, respectively). Thirty of these interviews were performed at the participant’s home, and five by telephone. One non-caregiver and no caregivers terminated the study because of the urine collection; the nonparticipation rate across the follow-up interviews was low for both caregivers (8%) and non-caregivers (5%; Figure 1). Loss-to-follow-up was the same for the two groups (7.5%).

Twenty-four-hour urine samples were collected for 96% of the follow-up interviews, with no difference by caregiver status (Table 2). Five of the urine samples could not be collected because the interview was conducted by phone; however, for only two of the 30 home interviews, a urine specimen was not obtained. The interval from a follow-up interview to the urine collection was similar to that for the baseline time period. Seven (1.4%) urine specimens (one from a caregiver, six from non-caregivers) were judged to be inadequate. At the follow-up interviews, blood samples were drawn from 96% of the participants; four blood samples could not be collected during a home visit.

We compared caregiver status and perceived stress level between participants who provided a useable urine sample and those who did not, at each study interview (Table 3). Although the numbers were small, there were no differences according to caregiver status or median PSS scores. When caregivers and non-caregivers were considered separately, median PSS score did not vary significantly by whether the participant provided an adequate urine specimen (data not shown).

Table 3.

Comparison of Caregiver Status and Perceived Stress Level Between Participants With and Without a Usable Urine Sample at Each Study Interview.

Interview	Participants with a useable sample	Participants without a useable sample	p value^a
Baseline	n = 237	n = 19
Caregivers, n (%)	104 (43.9)	6 (31.6)	.30
PSS, median (range)	16 (2-42)	17 (6-39)	.47
First follow-up	n = 196	n = 9
Caregivers, n (%)	79 (40.3)	5 (55.6)	.49
PSS, median (range)	17 (2-39)	22.5 (9-29)	.15
Second follow-up	n = 165	n = 8
Caregivers, n (%)	70 (42.4)	3 (37.5)	1.0
PSS, median (range)	16 (2-46)	17.5 (9-21)	.90
Third follow-up	n = 121	n = 9
Caregivers, n (%)	45 (37.2)	3 (33.3)	.91
PSS, median (range)	16 (2-41)	18 (4-28)	.96

Note. PSS = perceived stress score.

p value for comparing participants with and without a usable sample, using Fisher’s exact test for caregiver status and Wilcoxon rank sum test for PSS.

Discussion

In this prospective study of community-dwelling older adults, nearly all participants, 99%, provided a 24-hr urine specimen at baseline. A follow-up sample was obtained from 70% of the individuals initially enrolled in the study; of those who continued their participation in the study, 96% provided a follow-up sample. Compliance with the protocol for collection of the urine sample also was very high, with less than 10% of the specimens considered inadequate. Contrary to our hypothesis, neither caregivers nor those with higher perceived stress levels were less likely to provide a useable urine sample. This study demonstrates the feasibility of collecting 24-hr urine specimens longitudinally in active, community-dwelling older adults. With sufficient instruction and logistical support, the participants found the procedures for collecting a urine specimen to be acceptable and contributed a sample adequate for measurement of neuroendocrine biomarkers.

There is limited published information on the collection of 24-hr urine samples in population-based studies in the United States. Among men aged 43 to 85 years in the Normative Aging Study, 81% provided a 24-hr urine sample (Young et al., 1992); of the collected samples, 3% were low volume or had other collection errors. In the Observing Protein and Energy Nutrition (OPEN) Study of men and women between 40 and 69 years of age, two 24-hr specimens were obtained over an 11- to 14-day period (Subar et al., 2003). Of the possible specimens, 3% were either not collected or had inadequate volumes; 24% of analyzable urine samples were considered incomplete based on measurement of para-aminobenzoic acid (PABA) and were excluded from additional analysis. Participants in the Women’s Health Initiative Nutritional Biomarkers Study collected a 24-hr urine specimen, which they brought to their second study visit (Neuhouser et al., 2008). Missed or spilled voids occurred in only 1% of these 50- to 79-year old women in Seattle, Washington; based on PABA measurement, 5% of tested samples were considered incomplete. We observed similar, or better, rates of compliance and adequacy of collection in the present sample of older individuals recruited for a study of stress and health.

By contrast, several studies of community-dwelling older adults have collected 12-hr, overnight urine. In the MacArthur Studies, 75% of the elderly (70-75 years of age) cohort members provided an overnight urine specimen of sufficient volume and acceptable creatinine level (Reuben et al., 2000); in this study, higher refusal rates and incomplete collection occurred with 24-hr urine collection. More than 95% of the 50- to 67-year-old men and women in the population-based Chicago Health, Aging, and Social Relations Study provided a usable overnight urine sample (Masi et al., 2004). Although a 12-hr overnight urine sample has been considered more feasible to obtain from older adults than a 24-hr specimen collection, compliance in our study was comparable with these studies

Outside of the United States, a few population-based studies of older individuals have included the collection of a 24-hr urine sample, with mixed success. Similar to the present study, compliance was high in the InChianti Study in Italy; more than 95% provided a urine sample, of which about 5% were incomplete (Vogelzangs et al., 2007). A lower proportion of participants provided 24-hr urine (83%) in a cross-sectional study in Australia, although only 1% of the samples were excluded because they were considered to be inaccurately collected (Huggins et al., 2011). In the Uppsala Longitudinal Study of Adult Men in Sweden, a urine specimen was obtained from just three quarters of the older men (Helmersson-Karlqvist, Ärnlöv, Carlsson, Härmä, & Larsson, 2014).

We posited that, given the demands on their time and energy, caregivers would be less likely than non-caregivers to provide a usable urine specimen. It is known that caregivers are less likely to engage in healthy activities (Burton et al., 1997; Fredman et al., 2006) and are more prone to experience stress (Bertrand, Fredman, & Saczynski, 2006; Fredman, Cauley, Hochberg, Ensrud, & Doros, 2010; Hooker et al., 2002; Pinquart & Sörensen, 2003). We also expected to observe higher perceived stress levels among those without than those with an adequate urine sample. Neither of these hypotheses was supported by our data. Despite having higher perceived stress, caregivers were not less likely to comply with the urine collection protocol for the study. It is possible that certain aspects of caregiving serve to moderate the stress effects of providing care (Pearlin, Mullan, Semple, & Skaff, 1990). Development of coping mechanisms and access to social supports may enable caregivers to sufficiently manage their commitment to the care recipient such that they can engage in other purposeful endeavors (Corcoran, 2011). Caregivers who participate in research studies also may comply with burdensome study procedures, such as collecting 24-hr urine, because they want to contribute to research that will help future caregivers. An alternate explanation for comparable results in caregivers and non-caregivers may be the higher socioeconomic status of the caregivers in our sample. Their higher education level and household income may have mitigated against the constraints imposed by caregiving, with respect to their participation with all aspects of the study. Nonetheless, compliance with the urine protocol was very high for both caregivers and non-caregivers.

Several strategies were used to enhance compliance with providing a 24-hr urine sample by the study participants. Emphasis was placed on arranging the study interview and subsequent urine collection to best accommodate the participant’s schedule. At the follow-up interviews, participants were given the option to have the collection container and instructions mailed to them prior to the interview and to bring the urine specimen with them. Allowing participants to be interviewed in their homes, rather than coming in for a visit at the GCRU, minimized potential loss-to-follow-up, particularly as the study population aged or experienced stress. Of note, participants who chose to be interviewed at home had higher perceived stress levels then did those who were interviewed at the clinic.

The participants also received a small, monetary remuneration as compensation for collecting the urine specimen. Other studies have used travelling to the participants’ homes to collect the 24-hr urine sample (Neuhouser et al., 2008; Otte et al., 2005) or a combination of travelling to participants’ homes and monetary incentives (Subar et al., 2003). The use of oral and written collection instructions combined likely contributed to a low frequency of specimens found to be inadequate in the present study and others (Ferrucci et al., 2000; Huggins et al., 2011; Neuhouser et al., 2008). In particular, the written instructions may have enhanced compliance by providing concrete information on collection and storage of urine specimens and clearly stating that the participant save all urine specimens over the 24-hr period. Although we did not specifically ascertain from participants what factors contributed to their compliance (or lack of compliance) with the urine collection procedures nor did we ask them to keep a log to determine whether they missed any voids, we feel that the strategies used in our study helped to facilitate our obtaining of specimens of appropriate quality.

Some other limitations of our study should be noted. Most of the non-caregiver participants were recruited from the Harvard Cooperative Program on Aging. Given their prior research experience, these individuals might be expected to more readily comply with study protocols, including the collection of urine samples. As a result, our findings may be less representative of other community-based, older non-caregiver populations. In addition, it is possible that some individuals chose to terminate from the study or were lost to follow-up because of the inconvenience of providing a 24-hr urine specimen.

Conclusion

Biologic markers increasingly are becoming a key part of research on community-dwelling older adults, including their measurement at multiple intervals to detect changes over time. As shown in the present study, 24-hr urine samples can be obtained longitudinally in this population. Retrieval of the samples by study staff greatly facilitated collection and reduced the burden of participation. Neither caregiver status nor perceived stress level appeared to affect the ability of older adults to provide a useable specimen. The feasibility and acceptability of collecting biologic samples has important implications for the inclusion of biomarkers in epidemiologic studies of aging and health.

Footnotes

Declaration of Conflicting Interests

The authors declared no potential conflicts of interest with respect to the research, authorship, and/or publication of this article.

Funding

The authors disclosed receipt of the following financial support for the research, authorship, and/or publication of this article: This study was supported by the National Institutes of Health under the following grant numbers: R01 AG028144, R01 AG18037, and UL1 RR025771.

Author Biographies

Sherri O. Stuver is clinical professor in epidemiology at Boston University School of Public Health. With a background in cancer epidemiology, she has extensive experience in the collection of biologic specimens and measurement of biomarkers in epidemiologic studies. Her current work includes studies of stress in older adults and assessment of care at the end of life in cancer patients.

Jennifer Lyons is a doctoral student in epidemiology at Boston University School of Public Health. Her research interests focus on physical function and performance measures in older adults, as well as older caregiver psychosocial and health outcomes. She participated in the collection of data and biologic specimens for the Health Pathways Study.

Andrea Coviello currently is on the faculty of Duke University School of Medicine in the Division of Endocrinology, Diabetes, and Metabolism; she previously served as a faculty member at Boston University School of Medicine. Her research spans clinical epidemiology and clinical trials related to hormonal influences on metabolic and cardiovascular disease with a focus on gender differences in obesity, diabetes, and cardiovascular disease.

Lisa Fredman is professor in epidemiology at Boston University school of Public Health. Her research focuses on the health effects of older caregivers, particularly the association between caregiving with physiological and physical health outcomes. She is the principal investigator of the Health Pathways Study, which is the focus of this manuscript.

References

Aurora

Wolf

D. A.

(2014). Is there a trade-off between parent care and self-care? Demography, 51, 1251-1270.

Bauer

M. E.

Vedhara

Perks

Wilcock

G. K.

Lightman

S. L.

Shanks

(2000). Chronic stress in caregivers of dementia patients is associated with reduced lymphocyte sensitivity to glucocorticoids. Journal of Neuroimmunology, 103, 84-92.

Bertrand

R. M.

Fredman

Saczynski

(2006). Are all caregivers created equal? Stress in caregivers to adults with and without dementia. Journal of Aging and Health, 18, 534-551.

Burton

Newsom

Schulz

Hirsch

German

(1997). Preventive health behaviors among spousal caregivers. Preventive Medicine, 26, 162-169.

Chrousos

G. P.

(2000). The role of stress and the hypothalamic-pituitary-adrenal axis in the pathogenesis of the metabolic syndrome: Neuro-endocrine and target tissue-related causes. International Journal of Obesity & Related Metabolic Disorders, 24(Suppl. 2), S50-S55.

Cohen

(1986). Contrasting the Hassles Scale and the Perceived Stress Scale: Who’s really measuring appraised stress? American Psychologist, 41, 716-718.

Cohen

Kamarck

Mermelstein

(1983). A global measure of perceived stress. Journal of Health and Social Behavior, 24, 385-396.

Corcoran

M. A.

(2011). Caregiving styles: A cognitive and behavioral typology associated with dementia family caregiving. The Gerontologist, 51, 463-472.

Ferrucci

Bandinelli

Benvenuti

Di Iorio

Macchi

Harris

T. B.

, . . . for the InCHIANTI Group. (2000). Subsystems contributing to the decline in ability to walk: Bridging the gap between epidemiology and geriatric practice in the InCHIANTI Study. Journal of the American Geriatric Society, 48, 1618-1625.

10.

Fredman

Bertrand

Martire

L. M.

Hochberg

Harris

E. L.

(2006). Leisure-time exercise and overall physical activity in older women caregivers and non-caregivers from the Caregiver-SOF Study. Preventive Medicine, 43, 226-229.

11.

Fredman

Cauley

J. A.

Hochberg

Ensrud

K. E.

Doros

(2010). Mortality associated with caregiving, general stress, and caregiving-related stress in elderly women: Results of caregiver-study of osteoporotic fractures. Journal of the American Geriatrics Society, 58, 937-943.

12.

Goldman

Weinstein

Cornman

Singer

Seeman

Chang

M.-C.

(2004). Sex differentials in biological risk factors for chronic disease: Estimates from population-based surveys. Journal of Women’s Health, 13, 393-403.

13.

Helmersson-Karlqvist

Ärnlöv

Carlsson

A. C.

Härmä

Larsson

(2014). Increased urinary cystatin C indicated higher risk of cardiovascular death in a community cohort. Atherosclerosis, 234, 108-113.

14.

Hooker

Bowman

S. R.

Coehlo

D. P.

Lim

S. R.

Kaye

Guariglia

(2002). Behavioral change in persons with dementia: Relationships with mental and physical health of caregivers. The Journals of Gerontology, Series B: Psychological Sciences & Social Sciences, 57, P453-P460.

15.

Huggins

O’Reilly

Brinkman

Hodge

Giles

G. G.

English

D. R.

Nowson

C. A.

(2011). Relationship of urinary sodium and sodium-to-potassium ratio to blood pressure in older adults in Australia. Medical Journal of Australia, 195, 128-132.

16.

Karlamangla

A. S.

Singer

B. H.

Greendale

G. A.

Seeman

T. E.

(2005). Increase in epinephrine excretion is associated with cognitive decline in elderly men: MacArthur studies of successful aging. Psychoneuroendocrinology, 30, 453-460.

17.

Kiecolt-Glaser

J. K.

Preacher

K. J.

MacCallum

R. C.

Atkinson

Malarkey

W. B.

Glaser

(2003). Chronic stress and age-related increases in the pro-inflammatory cytokine IL-6. Proceedings of the National Academy of Sciences of the United States of America, 100, 9090-9095.

18.

Kudielka

B. M.

Buske-Kirschbaum

Hellhammer

D. H.

Kirschbaum

(2004). HPA axis responses to laboratory psychosocial stress in healthy elderly adults, younger adults, and children: Impact of age and gender. Psychoneuroendocrinology, 29, 83-98.

19.

Lutgendorf

S. K.

Costanzo

E. S.

(2003). Psychoneuroimmunology and health psychology: An integrative model. Brain, Behavior, and Immunity, 17, 225-232.

20.

Lutgendorf

S. K.

Kreder

K. J.

Rothrock

N. E.

Hoffman

Kirschbaum

Sternberg

E. M.

. . . Ratliff

T. L.

(2002). Diurnal cortisol variations and symptoms in patients with interstitial cystitis. Journal of Urology, 167, 1338-1343.

21.

Masi

C. M.

Rickett

E. M.

Hawkley

L. C.

Cacioppo

J. T.

(2004). Gender and ethnic differences in urinary stress hormones: The population-based Chicago Health, Aging, and Social Relations Study. Journal of Applied Physiology, 97, 941-947.

22.

McEwen

B. S.

(1998). Protective and damaging effects of stress mediators. New England Journal of Medicine, 338, 171-179.

23.

Neuhouser

M. L.

Tinker

Shaw

P. A.

Schoeller

Bingham

S. A.

Horn

L. V.

. . . Prentice

R. L.

(2008). Use of recovery biomarkers to calibrate nutrient consumption self-reports in the Women’s Health Initiative. American Journal of Epidemiology, 167, 1247-1259.

24.

Otte

Neylan

T. C.

Pipkin

S. S.

Browner

W. S.

Whooley

M. A.

(2005). Depressive symptoms and 24-hour urinary norepinephrine excretion levels in patients with coronary disease: Findings from the Heart and Soul Study. American Journal of Psychiatry, 162, 2139-2145.

25.

Pearlin

L. I.

Mullan

J. T.

Semple

S. J.

Skaff

M. M.

(1990). Caregiving and the stress process: An overview of concepts and their measures. The Gerontologist, 30, 583-594.

26.

Pinquart

Sörensen

(2003). Differences between caregivers and noncaregivers in psychological health and physical health: A meta-analysis. Psychology and Aging, 18, 250-267.

27.

Reuben

D. B.

Talvi

S. L.

Rowe

J. W.

Seeman

T. E.

(2000). High urinary catecholamine excretion predicts mortality and functional decline in high-functioning, community-dwelling older persons: MacArthur studies of successful aging. The Journals of Gerontology, Series A: Biological Sciences & Medical Sciences, 55, M618-M624.

28.

Rosmond

Bjorntorp

(2000). Occupational status, cortisol secretory pattern, and visceral obesity in middle-aged men. Obesity Research, 8, 445-450.

29.

Schatzkin

Kipnis

Carroll

R. J.

Midthune

Subar

A. F.

Bingham

. . . Freedman

L. S.

(2003). A comparison of a Food Frequency Questionnaire with a 24-hour recall for use in an epidemiological cohort study: Results from the biomarker-based Observing Protein and Energy Nutrition (OPEN) Study. International Journal of Epidemiology, 32, 1054-1062.

30.

Schöfl

Becker

Prank

von Zur Mühlen

Brabant

(1997). Twenty-four-hour rhythms of plasma catecholamines and their relation to cardiovascular parameters in healthy young men. European Journal of Endocrinology, 137, 675-683.

31.

Seeman

T. E.

McEwen

B. S.

Singer

B. H.

Albert

M. S.

Rowe

J. S.

(1997). Increase in urinary cortisol excretion and memory declines: MacArthur studies of successful aging. Journal of Clinical Endocrinology & Metabolism, 82, 2458-2465.

32.

Sluiter

J. K.

Frings-Dresen

M. H.

van der Beek

A. J.

Meijman

T. F.

Heisterkamp

S. H.

(2000). Neuroendocrine reactivity and recovery from work with different physical and mental demands. Scandinavian Journal of Work Environment & Health, 26, 306-316.

33.

Steptoe

Cropley

Griffith

Kirschbaum

(2000). Job strain and anger expression predict early morning elevations in salivary cortisol. Psychosomatic Medicine, 62, 286-292.

34.

Subar

A. F.

Kipnia

Trolano

R. P.

Midthune

Schoeller

D. A.

Bingham

. . . Schatzkin

(2003). Using intake biomarkers to evaluate the extent of dietary misreporting in a large sample of adults: The OPEN Study. American Journal of Epidemiology, 158, 1-13.

35.

Vedhara

Cox

N. K.

Wilcock

G. K.

Perks

Hunt

Anderson

. . . Shanks

N. M.

(1999). Chronic stress in elderly carers of dementia patients and antibody response to influenza vaccination. The Lancet, 353, 627-631.

36.

Vogelzangs

Suthers

Ferrucci

Simonsick

E. M.

Ble

Schrager

. . . Penninx

B. W.

(2007). Hypercortisolemic depression is associated with the metabolic syndrome in late-life. Psychoneuroendocrinology, 32, 151-159.

37.

Wurtman

R. J.

(2002). Stress and the adrenocortical control of epinephrine synthesis. Metabolism: Clinical and Experimental, 51(Suppl. 1), 11-14.

38.

Young

J. B.

Troisi

R. J.

Weiss

S. T.

Parker

D. R.

Sparrow

Landsberg

(1992). Relationship of catecholamine excretion to body size, obesity, and nutrient intake middle-aged and elderly men. American Journal of Clinical Nutrition, 56, 827-834.