• We are sorry, but NCBI web applications do not support your browser and may not function properly. More information
Logo of nihpaAbout Author manuscriptsSubmit a manuscriptNIH Public Access; Author Manuscript; Accepted for publication in peer reviewed journal;
Cancer. Author manuscript; available in PMC Mar 1, 2012.
Published in final edited form as:
PMCID: PMC3065823

Caregiver symptom burden: the risk of caring for an underserved patient with advanced cancer



The growing diversity of America’s population and the high burden of cancer-related symptoms reflect the need for caregiver research within underserved groups. In this longitudinal study, we assessed changes in symptom severity in caregivers and underserved minority patients diagnosed with advanced solid tumors being treated at public hospitals.


85 matched patient/caregiver dyads completed the M. D. Anderson Symptom Inventory 3 times during 20 weeks of chemotherapy. At each time point, we assessed symptom severity and interference with daily activities. Group-based trajectory modeling was used to classify caregivers into high or low symptom burden groups.


Sadness and distress were more prevalent among caregivers (P = .005). Symptom burden remained stable among caregivers in the high-symptom group (40%), whereas the low-symptom group (60%) showed a statistically significant decrease over time. Multivariate analysis found being a family-member caregiver (ADJ-OR 4.1; 95% CL 1.4, 11.6) and caring for a highly symptomatic patient (ADJ-OR 8.0; 95% CL 1.5, 41.4), rather than race, ethnicity, or sociodemographic characteristics, were significant predictors of the caregiver’s membership in the high symptom burden group.


Forty percent of the caregivers in this study were at increased risk for moderate to severe sadness and distress, which remained severe throughout the patient’s treatment course at public hospitals. To our knowledge, this study marks the first time that the concept of symptom burden has been used to measure caregiver burden, and the first time that symptom burden has been measured and documented in dyads of caregivers and underserved minority patients.

Keywords: Caregiver, cancer, vulnerable populations, ethnic groups, symptom burden, survivors


The physical and psychological health of informal (ie, unpaid) caregivers who look after patients with cancer is a growing public health issue. Many studies focused on informal cancer caregivers in general have been conducted, but little attention has been given to the health of such caregivers when the patients they assist are underserved. Patients and caregivers alike may be considered underserved if they are “members of racial and ethnic groups, those of lower socioeconomic status, and recent immigrants”.1

The growing diversity of America’s population and the high burden of cancer-related symptoms reflect the need for caregiver research within underserved groups.2,3 Knowing which factors contribute to the physical and emotional symptoms of caregivers of underserved cancer patients is critical, for several reasons. First, few studies have examined the prevalence and severity of, or the interference produced by, the multiple symptoms associated with cancer caregiving (“caregiver symptom burden”) in multiethnic or underserved settings. Second, longitudinal studies are needed to track changes in both patients’ and caregivers’ symptoms so that possible associations can be identified and strategies for addressing the symptoms can be devised.46 Third, and perhaps most important, cancer caregiver studies suggest that when the patient’s or the caregiver’s own symptoms are left untreated, the caregiver’s health outcomes are negatively affected.79

Disparities in cancer incidence, prevalence, and treatment availability contribute to poorer outcomes among patients who have financial difficulties and belong to ethnic minority groups.1012 These disparities are attributable to multiple complex factors that affect patients dealing with a cancer diagnosis. The lives of patients and their families become even more disrupted when the patient has advanced disease, severe symptoms, or uncertain survival potential.13,14 Patients who are highly symptomatic require more caregiver time than patients with less-bothersome symptoms. Similarly, persons caring for a highly symptomatic cancer patient may be at increased risk for various psychological, physical, financial, and social reactions when giving care, whether or not they themselves are undeserved; this situation is exacerbated when the caregiver also has limited resources and/or is in poor health.1517

We conducted a longitudinal descriptive study of symptom burden reported over time by patient-caregiver dyads in which the patient had advanced cancer and was categorized as underserved1. On the basis of literature showing the harmful effects of cancer caregiving on the caregiver’s psychological health,79 we expected that greater distress and sadness would be reported by caregivers than by patients. We applied group-based trajectory modeling to categorize caregivers into high symptom burden or low symptom burden groups while controlling for various patient and caregiver characteristics. We hypothesized that the predictors contributing to a caregiver’s being in the high symptom-burden group would include (1) a patient with poor performance status or severe symptoms, and (2) a caregiver that lived with or cared for the patient for more than 20 hours per week.


Subject Recruitment and Enrollment

The study was approved by the institutional review boards at The University of Texas MD Anderson Cancer Center, the University of Texas Health Science Center, and the Harris County (Texas) Hospital District. All participants provided informed consent.

To ensure that our patient sample would include only underserved patients, we recruited from 2 public hospitals in the Harris County Hospital District—Lyndon B. Johnson General Hospital and Ben Taub General Hospital. Patients treated at these hospitals are required to meet strict income-based economic criteria based on federal poverty-level guidelines for means-tested programs such as Medicare and Medicaid.18 In 2009, a family of 4 would not qualify for county services if the household income was more than $22,050/year.

Patient Prescreening

During the course of this study (2006 – 2010), clinical coordinators screened medical charts of patients from the outpatient oncology clinics of the 2 public hospitals. From these records and with advice from oncology fellows and physicians as needed, we identified patients who (1) were ≥18 years old; (2) had a diagnosis of advanced solid-tumor cancer; and (3) were scheduled to begin a new chemotherapy regimen within the next 2 weeks. Information about language preference (English or Spanish) was collected from the medical record when available.

Patient Eligibility for Study Participation

Eligible patients identified from screening were approached in the oncology outpatient clinic when they came for their first chemotherapy evaluation. The language preference of the patient was confirmed at that time. Bilingual, trained clinical coordinators provided details of the study, obtained informed consent from the patients who elected to enroll, and distributed study packets. Patients chose to complete self-administered questionnaires themselves or to have the questionnaires read aloud to them in English or Spanish by the clinical coordinator.

We enrolled only black, Hispanic, or white non-Hispanic patients, because these were by far the largest racial and ethnic groups treated at the 2 public hospitals; all other racial and ethnic groups were excluded. Information on patient ethnicity and race was collected by asking respondents to self-describe their race; the response choices were black or African-American non-Hispanic, Hispanic or Latino, and white non-Hispanic; anyone unable to self-identify as one of these races was not enrolled. In addition, we categorized respondents as “minority” (Hispanic/Latino or black/African-American) or “nonminority” (white non-Hispanic).

Finally, patients were asked to identify a person who would act as their primary, unpaid caregiver and would spend 4 or more hours per week helping them manage activities related to their disease or treatment.

Caregiver Eligibility for Study Participation

To be eligible for enrollment, the patient-identified caregiver had to (1) be at least 18 years old; (2) self-identify as black or African-American non-Hispanic, Hispanic or Latino, or white non-Hispanic; (3) understand English or Spanish; and (4) intend to assist, without pay, the patient for at least 4 hours per week. Caregivers were not required to be underserved.

If the caregiver had accompanied the patient to their visit and agreed to participate in the study, the clinical coordinators followed the procedures used for patients to determine preference for language, obtain informed consent, and distribute study packets. If the caregiver had not accompanied the patient on that day, the clinical coordinators contacted the potential caregiver by telephone to explain the study and invite them to participate; if the caregiver agreed to participate, the study coordinator scheduled a time to obtain informed consent, explain the study procedures, and complete the study packet.

Data Collection and Materials

The patient study packet included forms capturing demographic, social, and clinical information. Demographic information (sex, marital status, ethnicity and race, and education) was collected from patients’ medical charts. Patients described their relationship with their informal caregiver (eg, family member, friend), and estimated the number of hours each week that the caregiver would spend assisting them. Clinical data, tumor location, tumor stage, and Eastern Cooperative Oncology Group (ECOG) performance status were abstracted from medical records.

Patients also completed the M. D. Anderson Symptom Inventory (MDASI),19 a valid, reliable questionnaire that measures the self-reported severity of cancer-related and treatment-related symptoms (13 items) and their interference with daily activities (6 items). Symptom severity is measured on a scale of 0 (not present) through 10 (as bad as can you can imagine). Symptom interference is measured using a scale of 0 (did not interfere) through 10 (interfered completely). The time period used for recall of the symptoms and their interference was “within in the past 24 hours” for patients and “within the past week”for caregivers.

The caregiver study packet included similar forms for collecting demographic and social information. In addition, caregivers were asked to state their annual household income and whether they lived with the patient. Caregivers completed an earlier version of the MDASI that had 26 items, including the 13 validated MDASI symptom severity items, the 6 interference items, and additional affective items that we believed could be important to assess in the caregiver sample.

Data were collected 3 times during the 20-week study: before treatment (at enrollment), during treatment, and at the end of treatment. To avoid taxing the participants, we chose these data collection points to coincide with a scheduled chemotherapy session. However, if the caregiver was not going to accompany the patient to the clinic, participants could choose another location for data collection: in the outpatient clinic, at home during a visit by a study coordinator, or by telephone. Participants could also choose between a written or a face-to-face interview with a study coordinator and whether the interview would be conducted in English or Spanish.

The baseline (pretreatment) assessment consisted of all materials in the study packets and was performed on the first day of chemotherapy, immediately before the treatment session began. Both patients and caregivers repeated their respective MDASI questionnaires 2 more times: 6–9 weeks after the first assessment (during the chemotherapy course), and after completion of chemotherapy.

Statistical Analysis

Dyad Eligibility

We excluded from analysis any patient-caregiver dyad whose membership changed during the patient’s chemotherapy regimen. In addition, if data had not been collected for at least 2 time points from both members of the dyad, the case was removed from the final analysis.

Patient and Caregiver Characteristics

Descriptive analyses were conducted for each group to explore the relationships among all demographic, psychosocial, and clinical variables. Means, standard deviations (SD), and proportions were used to describe, rank order, and compare demographic characteristics of the dyads, caregiving-related characteristics, and patient clinical characteristics. We were interested in identifying the most severe symptoms and their interference for both caregivers and patients. Differences between important demographic and patient clinical data were explored using chi-square tests for categorical outcome variables and analysis of variance for continuous outcome variables. These preliminary data formed the construct of caregiver symptom burden and were used in the group-based trajectory modeling described below.

Symptom burden

Symptom Burden was defined by both the caregivers’ and patients’ ratings of symptom severity and symptom interference with physical and psychological functioning. Caregiver symptom burden, the dependent variable in these analyses, was conceptualized as the combination of the effects of a cancer patient’s symptoms on the caregiver’s physical and psychological health and the subsequent impact on their ability to function as they did before the patient commenced chemotherapy. For each time point, caregiver symptom burden was calculated using a composite score of the 4 most severe of the 13 MDASI symptoms. We were most interested in the prevalence of symptoms rated moderate to severe, which we operationally defined as symptoms rated 5 or greater on the MDASI’s 0 to 10 scale, on the basis of previous studies on pain in community-dwelling adults and patients diagnosed with cancer.20, 21 Patient symptom burden was calculated in a similar fashion, using patient-reported data.

Caregiver Symptom Burden Groupings

Group-based trajectory modeling was used to identify caregivers and patients with high or low symptom burden. This technique classifies individual trajectories into unique groups.2224 Using this approach, we were able to categorize caregivers into either high symptom burden or low symptom burden groups by using changes in the overall (caregiver plus patients) patterns in symptom severity over time. Descriptive statistics were used to describe the combined level of symptom severity in each group.

To determine whether group trajectories could have been affected by the characteristics and responses of the dyads that were lost during the study period, we used the same strategy described above to fit a 2-group model from all the dyads enrolled in the study and compared the results with the completed cases.

Potential Predictors of High or Low Caregiver Symptom Burden

A priori, we used 3 sets of variables as potential predictors of whether a caregiver would be in the high or low symptom burden group: demographic characteristics of caregivers and patients, caregiving-related characteristics, and patients’ clinical characteristics

Once the groups of high and low caregiver symptom burden were identified, we used logistic regression to conduct an exploratory univariate analysis to identify potential predictors of high symptom burden among caregivers. We determined potential predictors to be those that met a 2-tailed significance level of .05, without adjusting for multiplicity in the univariate analyses so as not to exclude potentially useful predictors in the initial screening. Because we were exploring relationships, we used a 2-stage screening for potential predictor variables that could be included in the model. These variables were then used in the multivariate analyses. The multivariate logistic approach allowed us to examine the effect of the predictor variables simultaneously.

Using the group-based trajectory analyses, we examined how the longitudinal course of symptoms, ECOG performance status, and other variables of interest over the 3 assessment times (before, during, and after treatment) might influence whether a caregiver would fall into the high or low symptom burden group.

All P values were 2-tailed with alpha = .05. All statistical procedures were conducted using SPSS version 16.0 software for Windows (SPSS Inc., 2008) and SAS for Windows version 9.1 (SAS, 2004).


Participant Accrual and Retention

A total of 129 patient-caregiver dyads met the eligibility criteria; of these, 22 caregivers or patients (17%) refused to join the study. A 17% refusal rate is considered low for the underserved minority population from which the sample was drawn. The remaining 107 dyads were enrolled into the study. At the end of the study, data from the 107 dyads were screened to determine if each dyad had 2 time points for which concurrent data were collected and that the same caregivers remained in the study across time. Twenty-two dyads did not meet these criteria and were excluded from the analysis. Thus, data from 85 dyads were included in this analysis.

Participant Characteristics

Overall, caregivers were significantly younger than patients (45.7 years, SD = 15.0 vs 55.5 years, SD = 10.3; P < .01). Both caregivers and patients were predominantly female, married, and Hispanic, and did not have a high school degree (Table 1). Three quarters of participants self-identified as being Hispanic/Latino or black/African-American non-Hispanic (ie, minority). We found no significant differences in the primary demographic and clinical characteristics between participants (85 matched dyads used in the analysis) and nonparticipants (22 unmatched dyads not used in the analysis).

Table 1
Characteristics of Caregivers and Patients before Treatment (N = 85 Dyads)

The patient-caregiver dyad responded to the MDASI on the same day 70% of the time; the median time difference between patients and caregivers in reporting of symptoms and their severity was 0 days, both overall and by time point.

Clinical Characteristics of Patients

According to the patients’ medical charts, lung and breast cancers constituted more than half of the cases of solid tumors (Table 1). Gastrointestinal cancers made up almost a quarter of the cases and gynecological and head and neck cancers each accounted for less than 10% of the tumors. Almost three quarters of the patients had poor ECOG performance status (grades 2–4).

Caregiving-Related Characteristics

Eighty-one per cent of caregivers were family members (Table 1). According to the caregivers, almost half of were a spouse or partner; other relationships included child, parent, or sibling. Most lived with the patient and provided help for >20 hours/week. Most the caregivers reported an annual household income of $25,000 or less.

Symptom Burden

Symptom burden was conceptualized as the 4 most severe symptoms reported by caregivers or patients (Fig. 1, top panel). Among caregivers, the 4 most severe symptoms overall were sadness and distress (psychological symptoms) and fatigue and disturbed sleep (physical symptoms). These symptoms differed notably across time (pretreatment, during treatment, and posttreatment). The relative severity of other of the caregivers’ symptoms varied across time and usually, but not always, decreased over time.

Figure 1
Mean Symptom Burden in Caregiver and Patient Groups, compared by phase (N = 85 dyads)

For the patient sample, the most severe symptoms overall (fatigue, pain, drowsiness, and disturbed sleep) differed across time. Before treatment, the top 4 symptoms reported by patients included fatigue and dry mouth, followed equally by disturbed sleep, distress, and drowsiness. During and after treatment, pain replaced dry mouth as a top symptom. Fatigue was the most severe symptom across all treatment phases for patients, with small variations in severity scores noted before, during, and after treatment.

The mean interference in day-to-day activities reported by caregivers and patients is also shown in Fig. 1 (bottom panel). Among caregivers, symptoms interfered most with mood, albeit only mildly. Among patients, symptom interference with the ability to work was rated the highest across all time points.

Overall, among caregivers, the most prevalent moderate to severe symptoms (rated ≥5 on the MDASI’s 0–10 scale; 13 validated MDASI symptom items only) were sadness, distress, fatigue, and disturbed sleep. Patients reported more severe symptoms over time compared with their caregivers, except for sadness and distress.

Trajectory of Caregiver Symptom Burden Groups

We next explored 2-group trajectories of caregivers in the high symptom burden group and low symptom burden groups. We found that a 2-group model was optimal on the basis of empirical model fit using the Akaike information criterion24 and on the interpretability of resulting groups. Caregiver characteristics for both high and low symptom burden groups reflected the characteristics of the entire caregiver group: they were likely to be Hispanic/Latino, female, married, living with and related to the patient, and to spend more than 20 hours a week taking care of the patient. The 40% of caregivers in the high symptom burden group had stable symptom severity over time (pretreatment, during treatment, and posttreatment), whereas among the 60% of caregivers in the low symptom burden group the severity decreased significantly, by 1.0 point on the 0 to 10 scale (P = .008) (Fig. 2).

Figure 2
Trajectory of the Observed Mean Caregiver Symptom Severity across Time for the High and Low Symptom Burden Groups (N = 85)

Sensitivity analysis indicated that the 2-group trajectory was still the best fitting model when all the cases (N = 85 dyads) were analyzed. In the high-symptom group the symptom severity remained stable over time, whereas in the low symptom burden group it decreased.

Predictors of High or Low Caregiver Symptom Burden

Univariate Analysis

The characteristics of caregivers and patients were screened to identify potential predictors for inclusion in the multivariate models (Table 2). Caregiver membership in the high symptom burden group was significantly more likely when the patient had high symptom burden, minority status, or poor ECOG performance status, or when the caregiver had minority status or was a family member (P ≤ .05 for all variables).

Table 2
Potential Predictors of High Caregiver Symptom Burden Included in the Multivariate Model (N = 85 Dyads)

Multivariate Analysis

The potential predictors identified in the univariate analysis (patient high symptom burden, minority status, poor ECOG performance, and being a family member) were subjected to logistic regression. The dependent variable was caregiver symptom burden (high vs. low). Only 2 predictors remained statistically significant in the multivariate model (Table 2). A caregiver’s membership in the high-symptom burden group was significantly more likely when the patient’s symptom burden was high (ADJ OR, 4.1; 95% CL, 1.4, 11.6; P = .01) and when the caregiver was a family member (OR, 8.0; 95% CL, 1.6, 41.4; P = .01).


This study paired patients and their caregivers to examine the effects of taking care of an underserved person with advanced cancer on the health and well-being of the caregiver. To identify factors that contribute to a high symptom burden among caregivers, we used the MDASI questionnaire to evaluate the prevalence, severity, and interference of multiple symptoms in a sample of caregivers and their patients. These patients had been diagnosed with advanced solid tumors and were beginning a chemotherapy regimen at a public hospital. Our hypothesis was that a caregiver would belong to a higher symptom burden group if (1) the caregiver’s patient had poor performance status or severe symptoms, or (2) the caregiver lived with the patient or provided care for more than 20 hours per week. To our knowledge, this study marks the first time that the concept of symptom burden has been used to measure a component of caregiver burden, and the first time that symptom burden has been measured and documented in dyads of caregivers and underserved minority patients.

As we had expected, across time the caregivers reported greater prevalence and severity of affective symptoms, such as sadness and distress, than the patients did. In contrast, patients reported more severe physical symptoms, such as fatigue, pain, and disturbed sleep. There were differences in the co-occurrence of multiple symptoms between the patient and caregiver groups. These differences in type and severity of symptoms support the need for interventions tailored to each group.

We divided the caregivers into low symptom burden and high symptom burden groups. Most of the caregivers in this study had low symptom burden. In this group the burden decreased significantly over the study period. This finding could indicate that the caregivers adjusted to their roles and responsibilities over time or that those patients found that their symptom burden lessened after treatment ended and thus required less assistance from their caregiver. In contrast, the caregivers who initially had a high symptom burden continued to have a high burden, even when the patient had completed treatment. This group of caregivers would be an ideal population to receive tailored interventions that could ameliorate the constantly high symptom burden during their experience. Any intervention addressing the caregivers of underserved patients with advanced cancer would need to consider the strengths and limitations of developing such a program in outpatient oncology clinics of public hospitals.

In partial support of our hypothesis, the 2 factors that predicted whether a caregiver would have a high symptom burden included caring for a patient with high symptom severity and being a relative of the patient. The first risk factor is not surprising, for as the patient’s cancer progresses and symptoms increase in prevalence, severity, and interference, the caregiver is likely exposed to greater stress, disturbed sleep, and fatigue. Our findings are consistent with other cancer caregiver study results indicating a relationship between increasingly severe and aggressive symptoms in a patient and an increasingly affected mental and physical health for the caregiver.5, 14, 25 In a meta-analysis, Pinquart and colleagues26 found that among caregivers of older adults, being a family member (e.g., spouse or adult child) increased a caregiver’s risk for negative health effects. Patients with advanced disease and receiving chemotherapy require more care. Thus, our findings quantify the intuitive explanation that caring for a relative (especially a spouse) with cancer, poor functioning, and high symptom severity increases the caregiver’s symptom burden.

Several caregiver studies have focused on sociodemographic factors that contribute to caregiver burden. Factors generally associated with an increased risk of caregiver burden, disturbed sleep, and depressive symptoms include being younger, being female, belonging to an ethnic or racial minority, having a lower income, and being a spousal caregiver.2729 In our study, we did not find that age, sex, or race or ethnicity contributed significantly to a high symptom burden in caregivers, but that lower socioeconomic status and spousal relationship to the patient did.

There are some limitations in this study worth noting. First, the generalizability of the findings is limited to caregivers of patients treated at public hospitals for advanced solid tumors. Second, the patients in the high symptom burden group tended to have very advanced disease and poor ECOG performance status, which may have contributed to an overestimation of symptom severity in both groups. Third, the sample size in our study was relatively small for examining differences by characteristics such as age or ethnicity and race. Finally, our sample of underserved caregivers and patients was heterogeneous in several ways, such as living arrangements between caregiver and patient or type of solid tumor, which may have affected the significance of our findings.

Our findings indicate that one third of cancer caregivers are at risk for high symptom burden. When family members care for highly symptomatic patients, clinicians should screen caregivers early for negative physical and psychological symptoms. The development and validation of brief, cross-cultural screening tools would help identify caregivers at high risk.

Our results also support the need for further research to identify factors that should be addressed when interventions are being developed to target underserved, multiethnic caregiver-patient dyads. Certain characteristics are associated with underserved cancer populations, including being noninsured or underinsured, having limited health literacy, belonging to an ethnic or racial group, presenting with advanced disease, and having a poorer survival rate. There is a critical need to consider these characteristics when designing culturally appropriate caregiver interventions that will lead to better caregiver outcomes and allow them to provide optimal care to the cancer patient.

In conclusion, we found that symptom burden in caregivers of patients with advanced cancer is a complex clinical and social problem that is influenced by both patient and caregiver characteristics. Additional research is warranted on the ethnic and racial differences in cancer caregiver burden to identify positive or protective factors that will enrich, or at least ease, the experience for underserved caregivers.


The authors would like to acknowledge the following individuals: Katherine Ramsey, BA and Lucy Balderas, BA for their recruitment and retention efforts; Gary Mobley, MA for consulting on data analysis; and Jeanie F. Woodruff, ELS for editorial support.

Funding: This project was supported by Award Number K07 CA102482 from the National Cancer Institute. The content is solely the responsibility of the authors and does not necessarily represent the official views of the National Cancer Institute or the National Institutes of Health.


Discipline: Disparities Research


1. Dohan D, Schrag D. Using navigators to improve care of underserved patients: current practices and approaches. Cancer. 2005 Aug 15;104(4):848–855. [PubMed]
2. Wells JN, Cagle CS, Bradley P, Barnes DM. Voices of Mexican American caregivers for family members with cancer: on becoming stronger. J Transcult Nurs. 2008 Jul;19(3):223–233. [PubMed]
3. Siefert ML, Williams AL, Dowd MF, Chappel-Aiken L, McCorkle R. The caregiving experience in a racially diverse sample of cancer family caregivers. Cancer Nurs. 2008 Sep-Oct;31(5):399–407. [PMC free article] [PubMed]
4. Northouse LL, Rosset T, Phillips L, Mood D, Schafenacker A, Kershaw T. Research with families facing cancer: the challenges of accrual and retention. Res Nurs Health. 2006 Jun;29(3):199–211. [PubMed]
5. McMillan SC, Weitzner MA. Methodologic issues in collecting data from debilitated patients with cancer near the end of life. Oncol Nurs Forum. 2003 Jan-Feb;30(1):123–129. [PubMed]
6. Given BA, Keilman L. Cancer in the elderly population: research issues. Oncol Nurs Forum. 1990 Jan-Feb;17(1):121–123. [PubMed]
7. Given B, Wyatt G, Given C, et al. Burden and depression among caregivers of patients with cancer at the end of life. Oncol Nurs Forum. 2004 Nov;31(6):1105–1117. [PMC free article] [PubMed]
8. Northouse LL, Mood DW, Schafenacker A, et al. Randomized clinical trial of a family intervention for prostate cancer patients and their spouses. Cancer. 2007 Dec 15;110(12):2809–2818. [PubMed]
9. Golant M, Haskins NV. "Other cancer survivors": the impact on family and caregivers. Cancer J. 2008 Nov-Dec;14(6):420–424. [PubMed]
10. Medicine Io. Unequal Treatment: Confronting Racial and Ethnic Disparities in Health Care. Washington, D. C.: The Naitonal Academies Press; 2003.
11. Freeman HP. Poverty, culture, and social injustice: determinants of cancer disparities. CA Cancer J Clin. 2004 Mar-Apr;54(2):72–77. [PubMed]
12. Brawley OW, Berger MZ. Cancer and disparities in health: Perspectives on health statistics and research questions. Cancer. 2008 Oct 1;113(7):1744–1754. [PubMed]
13. Honea NJ, Brintnall R, Given B, et al. Putting evidence into Practice (R): Nursing assessment and interventions to reduce family Caregiver strain and burden. Clinical Journal of Oncology Nursing. 2008 Jun;12(3):507–516. [PubMed]
14. Schulz R, Sherwood PR. Physical and Mental Health Effects of Family Caregiving. American Journal of Nursing. 2008 Sep;108(9):23–27. [PMC free article] [PubMed]
15. Zarit SH, Reever KE, Bach-Peterson J. Relatives of the impaired elderly: correlates of feelings of burden. Gerontologist. 1980 Dec;20(6):649–655. [PubMed]
16. Poulshock SW, Deimling GT. Families caring for elders in residence: issues in the measurement of burden. J Gerontol. 1984 Mar;39(2):230–239. [PubMed]
17. Given CW, Given B, Stommel M, Collins C, King S, Franklin S. The caregiver reaction assessment (CRA) for caregivers to persons with chronic physical and mental impairments. Res Nurs Health. 1992 Aug;15(4):271–283. [PubMed]
18. U.S. Department and Human Services ASfPaE. Federal Register74. Washington, D.C.: 2009. pp. 4199–4201.
19. Cleeland CS, Mendoza TR, Wang XS, et al. Assessing symptom distress in cancer patients: the M.D. Anderson Symptom Inventory. Cancer. 2000 Oct 1;89(7):1634–1646. [PubMed]
20. Palos GR, Cantor SB, Aday LA, Mendoza TR, Mobley GM, Cleeland CS. An urban community's preferences for hypothetical outcomes of analgesic pain treatment. J Pain Symptom Manage. 2006 Nov;32(5):470–477. [PubMed]
21. Serlin RC, Mendoza TR, Nakamura Y, Edwards KR, Cleeland CS. When is cancer pain mild, moderate or severe? Grading pain severity by its interference with function. Pain. 1995 May;61(2):277–284. [PubMed]
22. Nagin D. Group-based modeling of development. Cambridge, Mass: Harvard University Press; 2005.
23. Nagin D. Analyzing developmental trajectories: A semiparametric group-based approach. Psychological Methods. 1999;4:139–157.
24. Jones BLN DS, Roeder K. A SAS procedure based on mixture models for estimating developmental trajectories. Sociological Methods and Research. 2001;29:374–393.
25. Given B, Given CW, Sikorskii A, Jeon S, Sherwood P, Rahbar M. The impact of providing symptom management assistance on caregiver reaction: results of a randomized trial. J Pain Symptom Manage. 2006 Nov;32(5):433–443. [PubMed]
26. Pinquart M, Sorensen S. Correlates of physical health of informal caregivers: a meta-analysis. Journals of Gerontology Series B: Psychological Sciences & Social Sciences. 2007;62B(2):126–137. [PubMed]
27. Carter PA. Caregivers' descriptions of sleep changes and depressive symptoms. Oncol Nurs Forum. 2002 Oct;29(9):1277–1283. [PubMed]
28. Pinquart M, Sorensen S. Gender differences in caregiver stressors, social resources, and health: an updated meta-analysis. Journals of Gerontology Series B: Psychological Sciences & Social Sciences. 2006;61B(1):33–45. [PubMed]
29. Sherwood PR, Given BA, Donovan H, et al. Guiding research in family care: a new approach to oncology caregiving. Psychooncology. 2008 Oct;17(10):986–996. [PubMed]
PubReader format: click here to try


Related citations in PubMed

See reviews...See all...

Cited by other articles in PMC

See all...


  • MedGen
    Related information in MedGen
  • PubMed
    PubMed citations for these articles

Recent Activity

Your browsing activity is empty.

Activity recording is turned off.

Turn recording back on

See more...