Contextual and Stress Process Factors Associated with Head and Neck Cancer Caregivers’ Physical and Psychological Well-being

Introduction

A cancer diagnosis may present significant symptoms and treatment-related side effects for patients, necessitating care tasks for informal caregivers such as family members or friends (Girgis, Lambert, Johnson, Waller, & Currow, 2013). In the United States, caregivers caring for patients with cancer are a fundamental source of cancer care. There are at least 2.8 million informal caregivers in the US, who provide care to patients with a primary diagnosis of cancer (National Alliance for Caregiving, 2016). The population of cancer caregivers is expected to increase, in conjunction with a projected dramatic increase in cancer survivors (DeSantis et al., 2014). The vital role that caregivers play is increasingly recognized, and advancing intervention development for cancer caregivers has been deemed a clinical priority for informal cancer caregiving (Kent et al., 2016).

While providing care for a loved one with cancer has been associated with positive outcomes such as personal growth and an enhanced relationship with the care recipient (Li & Loke, 2013), cancer caregivers face a myriad of challenges and are at risk for negative psychological and physical health outcomes (Balfe, O’Brien, et al., 2016; Northouse, Williams, Given, & McCorkle, 2012). For example, cancer caregivers have high unmet needs across the cancer trajectory including psychosocial, medical, daily activity, healthcare service and information needs, and financial needs (Kim, Kashy, Spillers, & Evans, 2010; Longacre, Galloway, Parvanta, & Fang, 2015; Lund, Ross, Petersen, & Groenvold, 2015; Sklenarova et al., 2015), and often enter a caregiving role with little preparation or training (Hashemi-Ghasemabadi, Taleghani, Yousefy, & Kohan, 2016). Anxiety and depression are common in cancer caregivers (40% and 39%, respectively) (Braun, Mikulincer, Rydall, Walsh, & Rodin, 2007; Janda et al., 2007), and studies have shown that cancer caregivers have worse mental and physical well-being when compared to population norms (Butow et al., 2014).

Fletcher et al. (2012) proposed a conceptual model of cancer caregivers’ health and well-being, adapted from a previous model proposed in 2000 by Weitzner and colleagues. The updated model has three main interacting components that ultimately contribute to caregiver health and well-being, including contextual factors, the stress process, and the cancer trajectory (Fletcher et al., 2012). Contextual factors include the caregiver’s personal, sociocultural, economic, and health care characteristics. The stress process component includes primary (e.g., patient illness factors and caregiving characteristics) and secondary (e.g., spillover effects of caregiving such as employment) stressors, appraisal (e.g., perceived burden), and cognitive-behavioral responses (e.g., self-care and caregiving behaviors) of the caregiver (Fletcher et al., 2012). The cancer trajectory encompasses all stages of a cancer illness beginning with cancer diagnosis and concluding with bereavement or survivorship and acknowledges the different care experiences that may be encountered at different points in the cancer experience (Fletcher et al., 2012). Although the updated model is grounded in a decade of cancer caregiving research (Fletcher et al., 2012), few studies have explored the impact of contextual factors and stress process variables on health and well-being in vulnerable caregiver populations of patients with complicated care regimens (Kent et al., 2016), such as caregivers of patients with head and neck cancer (HNC).

Head and neck cancer includes cancers of the upper aerodigestive tract (larynx, pharynx, nasopharynx, and oral cavity), treated with a combination of surgery, radiation, and chemotherapy (Ward & van As-Brooks, 2014). Treatment for HNC may cause or contribute to disfigurement, impaired speech, difficulty swallowing, and pain (Penner, 2009). Patients commonly experience high levels of psychological distress and report high care demands (Badr, Gupta, Sikora, & Posner, 2014; Chen et al., 2009). Caregivers engage in a variety of tasks related to patient oral care, tube feeding, skin care, speech and voice, tracheostomy care, and more (Bond et al., 2016). Although studies have highlighted vulnerable characteristics among HNC caregivers (Balfe, Butow, et al., 2016; Longacre, Ridge, Burtness, Galloway, & Fang, 2012; Nightingale et al., 2016), there is limited knowledge regarding contextual factors and stress process variables contributing to HNC caregivers’ physical and psychological well-being. Further, little research has focused on caregivers of HNC patients receiving major surgery, and this population of caregivers may have distinct experiences from caregivers of patients receiving other treatment modalities. Increased knowledge regarding factors associated with negative psychological and physical outcomes in HNC caregivers will aid in identifying caregivers most at risk for poor outcomes; further, this knowledge may provide insight regarding potential targets for interventions aimed at improving psychological or physical well-being in HNC caregivers. Thus, the current study examines the following exploratory research question: What are the contextual factors and stress process variables that are associated with physical and psychological well-being in caregivers of patients treated with major surgery for HNC? Figure 1 presents the conceptual model of our research, which is based on Fletcher et al.’s model.

Figure 1.

Study Conceptual Model

Methods

Participants and procedures

This cross-sectional study was performed in two academic medical centers in North and South Carolina from 2011–2012. The study sample consisted of HNC patients and their nominated caregivers. Patients with HNC were identified using clinic administrative databases for a parent study (Hatcher et al., 2016). Eligible HNC patients for the parent study were: (1) 18 years of age or older, (2) diagnosed with stage I-IV squamous cell carcinoma of the upper aerodigestive tract, (3) scheduled for a major head and neck surgery (i.e., tumor resection, neck dissection, plus or minus reconstruction), and (3) able to read/understand English. A total of 89 HNC patients were participated in the parent study and 43 of them nominated their caregivers to complete a one-time caregiver survey by mail or telephone (Hatcher et al., 2016; Nightingale et al., 2016). Inclusion criteria for the informal caregiver included: (1) nomination by an eligible patient, (2) age18 years or older, (3) provision of uncompensated care for the nominating patient, and (4) able to read/understand English. Ethics approval for this study was obtained from the institutional review boards of both Wake Forest University School of Medicine and the Hollings Cancer Center. All participants provided informed consent.

Measures

Contextual Factors

Caregiver age, gender, race/ethnicity, education, marital status, and number of comorbidities were collected as contextual factors. Caregiver comorbid conditions were defined as a count of self-reported physician diagnoses of heart disease, stroke, hypertension, diabetes, cancer, chronic obstructive pulmonary disease, asthma, or arthritis (categorized </≥ 2 comorbidities). Patient data collected as contextual factors included age, gender, and race/ethnicity.

Stress Process Variables

Primary and Secondary Stressors

Caregiver characteristics including days per week of care provided (</≥ 4 days/week), hours of daily care provided (</≥ 5 hours/day), and living with patient (yes/no), were collected as primary stressors. Patient clinical characteristics (i.e., cancer stage, month since diagnosis) were abstracted from patient medical charts. Patient head and neck cancer-related symptoms and dysfunction were assessed using the European Organization for Research and Treatment of Cancer (EORTC) QLQ-H&N35 (Sherman et al., 2000). This instrument is a reliable and empirically validated scale assessing pain, swallowing, senses, speech, social eating, and social contact with greater scores indicating more problems with each symptom (Sherman et al., 2000). Patient’s comorbidity was defined as < 2 or ≥ 2 chronic conditions, based on clinical data from the chart-based medical records (i.e., hypertension, coronary artery disease or congestive heart failure, cerebrovascular disease or peripheral vascular disease, chronic obstructive pulmonary disease, asthma, diabetes, or chronic renal insufficiency). Patient’s fatalism about health was measured by five items from a scale reported from a study of cancer (e.g., I’ve given up trying to get better) with higher scores indicating increased fatalism (Schnoll et al., 2002). Secondary stressors included caregivers’ full-time employment status (yes/no), providing child care (yes/no), and short nighttime sleep duration (</≥ 7 hours/night).

Appraisal and Cognitive-Behavioral Responses

Caregiver burden was evaluated using a brief screening version of the Zarit Burden Interview (ZBI) (Bédard et al., 2001). The screening version of the ZBI has scores ranging from 0 to 16 and higher scores reflect more burden. The ZBI has demonstrated excellent internal consistency (Majerovitz, 1995; Zarit, Anthony, & Boutselis, 1987). Caregivers’ physical activity (</≥ 3 days/week) was collected as a cognitive-behavioral response variable and measured using an item from the Cancer Care Outcomes Research and Surveillance (CanCORS) caregiver study (Ayanian et al., 2004).

Health and Well-being Outcomes

The primary outcomes, physical and psychological well-being, were assessed through reliable and valid self-report instruments. Caregivers’ physical well-being was assessed using the Veterans RAND 12-item Health Survey (VR-12) physical component score (PCS) (Ware, Kosinski, & Keller, 1996). The VR-12 has been demonstrated to be valid and reliable (Ware et al., 1996). Possible scores range from 0 to 100, with higher scores indicating better perceived physical health functioning.

We used several instruments to assess caregivers’ psychological well-being, including mental health functioning, depressive symptoms, and anxiety. Mental health functioning was measured using the VR-12 mental component score (MCS). Possible MCS scores range from 0 to 100 and higher scores indicate better mental health functioning. Depressive symptoms were assessed using the 10-item Center for Epidemiologic Studies Depression Scale (CESD-10) (Radloff, 1977). The CESD-10 is a widely used depressive symptoms measure and has been shown to be both reliable and valid (Andresen, Malmgren, Carter, & Patrick, 1994; Radloff, 1977). Scores on the CESD-10 range 0 to 30, with higher scores indicating greater depressive symptoms. A cutoff score of ≥ 10 on the CESD-10 has been previously established to denote significant depressive symptoms (Andresen et al., 1994). Caregivers’ anxiety was assessed using the 7-item National Institutes of Health (NIH) Patient-Reported Outcomes Measurement Information System (PROMIS) Emotional Distress short form - Anxiety 7a (Pilkonis et al., 2011). The reliability and validity of PROMIS-Anxiety has been well documented (Cella et al., 2010). Raw scores were converted to t-scores (0-100 scale) with a mean of 50 and SD of 10 (Cella et al., 2010). Higher t-scores indicate greater anxiety.

Data Analysis

Descriptive statistics were conducted to describe sample characteristics. General linear models (GLM) were utilized to test for associations between predictor variables and physical and psychological well-being. Least-squares means (LS means) were estimated for categorical predictor variables. Each outcome variable was analyzed individually. Statistical analyses were conducted with SAS (v 9.4, SAS Institute, Cary, NC). All statistical tests were two-sided and statistical significance was at a 0.05 alpha level.

Results

Sample characteristics

A total of 43 nominated HNC caregivers were invited to participate in this study. The survey response rate was 79% (n = 34) for caregivers. The caregiver survey data were linked with patient data. Of the 34 pairs of HNC patients and their caregivers, 33 dyads were included in this study; 1 dyad was ineligible for analyses due to missing data. Socio-demographic, clinical, and caregiving characteristics are shown in Table 1. A majority of caregiver respondents were non-Hispanic White (85.0%) and female (81.8%). Most caregivers were a spouse or partner (72.7%) and lived with the patient (87.9%), were not employed full-time (81.8%), and had less than a college education (66.7%). Most caregivers provided care for ≥ 4 days/week (81.8%) and spent < 5 hours/day caring for the patient (75.0%). Over half of the caregivers had ≥ 7 hours of sleep/night (57.6%), 66.7% engaged in physical activity on ≥ 3 days/week, and 45.5 % had ≥ 2 comorbidities. The HNC patient group was predominantly male (84.9%) and had a mean age of 60.1 years. Seventeen patients (51.5%) were diagnosed with AJCC stage IV disease and 16 patients (48.5%) had ≥ 2 non-cancer comorbidities. The mean scores of the PCS and MCS for caregivers were 45.8 (SD = 13.6) and 46.8 (SD = 11.7), respectively, which are both slightly lower compared to the general population (M = 50, SD = 10) (Selim et al., 2009). The mean of caregivers’ CESD-10 score was 9.7 (SD = 7.7). The mean t-score of anxiety was 54.1 (SD = 10.2), which is somewhat higher than that of the general population (Cella et al., 2010).

Table 1.

Characteristics of Head and Neck Cancer Caregivers and Patients (N=33)

Variables		N (%)	Mean (SD)
Caregiver characteristics
Age			60.0 (11.2)
Gender	Female Male	27 (81.8) 6 (18.2)
Race/ethnicity	Non-Hispanic White Other	28 (84.9) 5 (15.1)
Education	< College ≥ College graduate	22 (66.7) 11 (33.3)
Marital status	Married/partnered Other	29 (87.9) 4 (12.1)
Full-time work status	Yes No	6 (18.2) 27 (81.8)
Relationship to patient	Spouse/partner Other	24 (72.7) 9 (27.3)
≥ 7 hours of sleep/night	Yes No	19 (57.6) 14 (42.4)
≥ 3 days physical activity/week	Yes No	22 (66.7) 11 (33.3)
≥ 2 comorbidities	Yes No	15 (45.5) 18 (54.6)
Living with the patient	Yes No	29 (87.9) 4 (12.2)
Days per week of care provided	1-3 days/week ≥ 4 days/week	6 (18.2) 27 (81.8)
Hours of daily care provided	< 5 hours/day ≥ 5 hours/day	24 (75.0) 8 (25.0)
Providing child care	Yes No	5 (15.2) 28 (84.9)
Caregiving burden			5.4 (3.9)
Patient characteristics
Age			60.1 (10.7)
Gender	Female Male	5 (15.2) 28 (84.9)
AJCC Cancer stage	I-III IV	16 (48.5) 17 (51.5)
Month since diagnosis			28.8 (34.6)
≥ 2 comorbidities	Yes No	16 (48.5) 17 (51.5)

Six factors were significantly associated with caregiver physical well-being in linear models (Table 2). Caregivers’ age was positively associated with physical well-being, indicating that older age was associated with better physical well-being (β = .45, p < .05). Caregivers who had < 2 comorbidities had better physical well-being than those with ≥ 2 comorbidities (LS mean = 51.4 vs. 39.1, p < .01). Positive associations were observed between patients’ difficulties with swallowing and speech and caregivers’ physical well-being. Providing care for a patient with more swallowing and speech difficulties were each associated with better physical well-being in caregivers (both β = .15, p < .05). The LS mean score on the PCS was significantly higher in caregivers who reported that they slept ≥ 7 hours/night than those who slept < 7 hours/night (50.3 vs. 39.8, p < .05) and who engaged in ≥ 3 days of physical activity/week had better physical well-being than those who did not (49.6 vs. 38.2, p < .05).

Table 2.

Least-squares means and beta coefficients for physical and psychological well-being among head and neck cancer caregivers (N=33) by selected demographic, caregiving, and clinical variables

Variables	PCSa	MCSa	Depressive symptoms (CESD-10)b	PROMIS Anxietyb
Contextual Factors
Age c	.45* (.20)	.12 (.19)	−.30* (.11)	−.30ǂ (.15)
Gender  Female  Male	44.7 (2.61) 51.0 (5.53)	45.7 (2.25) 51.7 (4.76)	10.7 (1.45) 5.2 (3.08)	55.3 (1.92) 48.4 (4.07)
Race/ethnicity  Non-Hispanic White  Other	47.5ǂ (2.48) 36.1ǂ (5.86)	46.6 (2.25) 48.1 (5.32)	9.5 (1.48) 10.6 (3.51)	53.6 (1.94) 56.8 (4.60)
Education  < College  ≥ College graduate	42.8ǂ (2.78) 51.9ǂ (3.93)	46.1 (2.53) 48.3 (3.58)	10.9 (1.63) 7.3 (2.31)	56.1 (2.11) 50.0 (2.99)
Marital status  Married/partnered  Other	45.8 (2.56) 45.6 (6.89)	47.9 (2.13) 38.7 (5.75)	8.9 (1.40) 15.3 (3.78)	53.1 (1.86) 61.0 (5.00)
≥ 2 comorbidities  Yes  No	39.1** (3.16) 51.4** (2.88)	47.2 (3.07) 46.5 (2.81)	11.5 (1.98) 8.2 (1.81)	56.2 (2.62) 52.3 (2.39)
Primary Stressors
Days per week of care provided  1–3 days/week  ≥ 4 days/week	47.2 (5.62) 45.5 (2.65)	48.2 (4.85) 46.5 (2.29)	7.2 (3.17) 10.3 (1.49)	51.7 (4.20) 54.6 (1.98)
Hours of daily care provided  < 5 hours/day  ≥ 5 hours/day	47.7 (2.67) 43.2 (4.62)	47.6 (2.45) 44.9 (4.25)	9.0 (1.61) 11.8 (2.79)	54.0 (2.12) 53.3 (3.68)
Living with the patient  Yes  No	46.0 (2.56) 44.2 (6.88)	47.3 (2.20) 43.1 (5.91)	9.7 (1.46) 9.8 (3.93)	53.8 (1.92) 55.9 (5.17)
Patient’s Cancer stage  I-III  IV	49.8ǂ (3.29) 42.0ǂ (3.20)	48.4 (2.95) 45.3 (2.86)	7.8 (1.91) 11.5 (1.85)	50.8ǂ (2.46) 57.1ǂ (2.38)
Patient’s symptoms/dysfunction c  Pain  Swallowing  Senses  Speech  Social eating  Social contact	−.06 (.09) .15* (.07) .00 (.09) .15* (.07) .07 (.06) .08 (.08)	−.08 (.08) −.05 (.06) −.10 (.07) −.10 (.06) −.03 (.06) −.15* (.07)	.10*(.05) −.01 (.04) .04 (.05) .02 (.04) −.01 (.04) .06 (.05)	.10 (.06) −.02 (.06) .11 (.07) −.02 (.06) −.01 (.05) .04 (.06)
Patient’s month since diagnosisc	.03 (.07)	−.01 (.06)	−.00 (.04)	.00 (.05)
Patient’s ≥ 2 comorbidities  Yes  No	44.3 (3.42) 47.2 (3.32)	43.7 (2.87) 49.8 (2.78)	11.7 (1.90) 7.8 (1.84)	56.8 (2.50) 51.5 (2.42)
Patient’s fatalism	−.78 (3.70)	−4.01 (3.12)	3.32 (2.03)	2.55 (2.75)
Secondary Stressors
Full-time employment status  Yes  No	43.6 (5.61) 46.3 (2.64)	49.3 (4.84) 46.3 (2.28)	8.0 (3.19) 10.1 (1.50)	52.8 (4.22) 54.4 (1.99)
Providing child care  Yes  No	51.2 (6.07) 44.9 (2.57)	50.4 (5.28) 46.2 (2.23)	8.6 (3.51) 9.9 (1.48)	51.3 (4.60) 54.6 (1.94)
≥ 7 hours of sleep/night  Yes  No	50.3* (2.91) 39.8* (3.39)	49.9ǂ (2.59) 42.6ǂ (3.02)	6.1*** (1.51) 14.6*** (1.75)	49.7** (2.05) 60.0** (2.38)
Appraisal Caregiving burden c	−.24 (.62)	−1.72*** (.44)	1.15*** (.29)	1.30** (.41)
Cognitive-Behavioral Response
≥ 3 days physical activity/week  Yes  No	49.6* (2.69) 38.2* (3.80)	49.2 (2.43) 42.1 (3.44)	7.3** (1.50) 14.5** (2.11)	50.4** (1.90) 61.3** (2.68)

Note. All variables were examined in independent models. Numbers in parentheses indicate standard errors.

^aHigher score indicates better physical or mental health.

^bHigher score indicates more depressive symptoms or greater anxiety level.

^cβ.

^***p < .001,

^**p < .01,

^*p < .05

^ǂp < .10

PCS = VR-12 Physical Component Score; MCS = VR-12 Mental Component Score

Only two factors were significantly associated with caregiver mental health functioning, patient’s difficulties with social contact and caregiving burden. Providing care for a patient with more social contact problems and having higher caregiving burden were both associated with a poorer mental health functioning in caregivers (β = −.15, p < .05; β = −1.72, p < .001, respectively). Older age in caregivers was significantly associated with lower depressive symptoms (β = −.30, p < .05), whereas providing care for patients with higher levels of pain was associated with more depressive symptoms among caregivers (β = .10, p < .05). Caregivers with short nighttime sleep duration (< 7 hours) had significantly greater depressive symptoms and anxiety than those with ≥ 7 hours sleep/night (LS mean for CESD-10 = 14.6 vs. 6.1, p < .001; LS mean for PROMIS anxiety = 60.0 vs. 49.7, p < .01, respectively). Higher caregiving burden was associated with greater depressive symptoms and anxiety (β = 1.15, p < .001; β = 1.30, p < .001, respectively). Caregivers who engaged in ≥ 3 days of physical activity/week reported lower depressive symptoms and anxiety than those who did not (LS mean for CESD-10 = 7.3 vs. 14.5, p < .01; LS mean for PROMIS anxiety = 50.4 vs. 61.3, p < .01, respectively).

Discussion

This study examined factors associated with HNC caregivers’ physical and psychological well-being, guided by a comprehensive conceptual model of family cancer caregiving proposed by Fletcher et al. (2012). Better physical well-being among HNC caregivers was associated with older age and having < 2 comorbid conditions (contextual factors), more patient problems with swallowing and speech (primary stressors), longer caregiver nighttime sleep duration (secondary stressor), and more physical activity (cognitive-behavioral response). Better mental health functioning among HNC caregivers was associated with fewer difficulties with social contact (primary stressor) as well as less caregiving burden (appraisal). Both caregiver anxiety and depression were significantly associated with short sleep duration (secondary stressor), caregiving burden (appraisal), and less physical activity (cognitive-behavioral response). Younger caregiver age (contextual) and more patient pain (primary stressor) were additionally related to depressive symptoms.

Previous literature is inconsistent with regard to the relationship between caregivers’ age and physical health. Although some caregiver studies have found that older age was related to poorer physical health (Kim & Schulz, 2008; Pinquart & Sörensen, 2007), others have found no significant relationship between age and physical health (Kurtz, Kurtz, Given, & Given, 2004). In this study, caregiver’s older age was related to better physical well-being. A potential explanation is that compared to younger caregivers, older caregivers may have less outside competing demands that typically require additional physical effort (e.g., child care) or may have additional help providing care for the patient, resulting in less physical strain. Another potential explanation is that older caregivers may engage in downward social comparison, a process by which the caregiver compares their health to the patient’s health, resulting in a more favorable perception of their own health (Taylor & Lobel, 1989). Downward social comparison is more salient in older age (Heckhausen & Krueger, 1993), thus older caregivers may be more likely to engage in this process compared to younger caregivers.

As suggested in other caregiver research (Vitaliano, Katon, & Unützer, 2005; Vitaliano, Zhang, & Scanlan, 2003), HNC caregivers in this study with comorbidities had worse physical well-being. This study also revealed that HNC patient’s symptoms, such as difficulties with swallowing and speech, were associated with better physical well-being in caregivers. This finding is somewhat unexpected, because it is generally believed that increased severity in patient symptoms may adversely affect caregivers’ health outcomes (Girgis et al., 2013; Nijboer, Triemstra, Tempelaar, Sanderman, & van den Bos, 1999). This result may reflect the possibility that potential gaps exist between severity of HNC patients’ perceived symptoms and corresponding levels of patient’s dependency on the caregiver for assistance. Similar to findings in other caregiver studies (Burton, Newsom, Schulz, Hirsch, & German, 1997; King & Brassington, 1997), our results showed that shorter nighttime sleep duration and low physical activity were significantly associated with poorer physical well-being. These findings suggest limited sleep and physical activity may results in diminished physical well-being for HNC caregivers, which has previously been suggested for cancer survivors and other types of cancer caregivers (Rock et al., 2012; Speck, Courneya, Mâsse, Duval, & Schmitz, 2010).

In the present study, we observed a significant relationship between a primary stressor (e.g., patient’s difficulties with social contact) and worse mental health functioning in caregivers. This finding extends previous studies with HNC patients which have demonstrated that difficulty with social contact is a risk factor for poor mental health outcomes in patients (Psoter, Aguilar, Levy, Baek, & Morse, 2012; Wu et al., 2016). Taken together, findings in this study and others (Psoter et al., 2012; Wu et al., 2016) suggest that patients’ difficulties with social contact may contribute to both patient and caregiver mental health. In accordance with other caregiver studies (Chang, Chiou, & Chen, 2010; Grunfeld et al., 2004; Rhee et al., 2008), caregiving burden (included in the appraisal domain) was significantly related to caregiver’s mental health functioning, depression, and anxiety. These findings provide preliminary evidence that altering the negative appraisal of the caregiving situation may influence all aspects of psychological well-being among HNC caregivers. Similar to Fletcher et al.’s model (2012), patient and caregiver factors together contributed to HNC caregiver depressive symptoms. Our findings are consistent with prior caregiver research suggesting patient’s pain (Kozachik et al., 2001; Sherwood et al., 2006; Weitzner, McMillan, & Jacobsen, 1999), caregiver’s younger age (Braun et al., 2007), short nighttime sleep duration in caregivers (Carter, 2002; Carter & Chang, 2000), and caregiver physical inactivity (Castro, Wilcox, O’Sullivan, Baumann, & King, 2002; King & Brassington, 1997; McCurry, Logsdon, Teri, & Vitiello, 2007) as risk factors for caregiver depressive symptoms.

We also observed that a secondary stressor in caregivers (short nighttime sleep duration) and a cognitive-behavioral response (low physical activity) were risk factors for caregiver’s anxiety, consistent with results from previous studies (Flaskerud, Carter, & Lee, 2000; Ströhle, 2008). These results taken together suggest that caregiver’s nighttime sleep and physical activity are important modifiable factors that may potentially contribute to reductions in both depressive symptoms and anxiety among HNC caregivers. These results suggest that clinicians and healthcare professionals should consider lack of sleep and physical inactivity as critical risk factors. Behavioral-educational intervention programs targeted at promoting nighttime sleep and encouraging physical activity may be particularly important for decreasing depressive symptoms and anxiety in HNC caregivers.

Among the general caregiver population, empirical evidence suggests the critical importance of various stress process factors in understanding caregiver health outcomes (Haley, LaMonde, Han, Burton, & Schonwetter, 2003; Matthews, Baker, & Spillers, 2003; Pearlin, Mullan, Semple, & Skaff, 1990). Fletcher et al.’s (2012) model was extended to include a comprehensive set of risk and protective factors (e.g., stressors, appraisal, cognitive-behavioral responses, contextual, and cancer trajectory factors) that may contribute to physiological and psychological outcomes of cancer caregivers specifically. The current findings align closely with Fletcher et al.’s model and suggest that the model indeed applies to an understudied HNC caregiver populations. In addition, our findings describe in detail factors affecting each dimension of well-being separately. Given the small sample size, our research is exploratory in nature. However, the current results provide an addition to the relatively scarce research literature in the area of HNC caregivers.

Study Strengths and Limitations

Our study has several notable strengths. First, the current study was carried out exclusively in an understudied population of HNC caregivers. This study, therefore, makes an important contribution towards understanding potential risk factors for poor mental and physical health outcomes among HNC caregivers. Second, this analysis was conducted based on a conceptual model, which is grounded in family caregiving research and provides a scientific basis for understanding family cancer caregivers’ health and well-being. The inclusion of a comprehensive group of study variables based on a conceptual framework and the assessment of broader outcomes allowed us to extend our knowledge regarding potential risk and protective factors in HNC caregivers’ physical and psychological well-being. Third, the inclusion of both HNC caregivers and patients allowed us to investigate the contribution of both caregiver and patient factors associated with caregivers’ physical and psychological well-being.

The current study has some limitations. Although this analysis was guided by a stress process model of caregiving, the cross-sectional design of this study limits causal inferences regarding the stress process and temporal relationships between the study variables and outcomes, and limits the ability to examine the illness trajectory domain of the model. An important limitation of this study is that because of the small sample and large number of study variables, the results should be considered with caution. In addition, the sample may not be representative of the larger population of HNC caregivers. We also relied on a convenience sample of caregivers of HNC patients receiving major surgery who were recruited from two institutions in North and South Carolina. The majority of study participants were female and non-Hispanic White. Further, patients were heterogeneous in terms of their clinical and disease characteristics such as tumor persistence or recurrence, patient’s surgery status, primary tumor site (oral cavity, oropharynx, hypopharynx, larynx, and other), and length of time since diagnosis. Consequently, these findings should be interpreted with caution as caregiving experiences are likely to differ based on these clinical factors.

Implications

The findings from this study emphasize the importance of a comprehensive understanding of factors affecting HNC caregivers’ physical and psychological well-being. Health care professionals working in areas that serve HNC patients can help caregivers whose lives are more likely to be impacted negatively after caregiving begins. Health care professionals and practitioners should be aware of the risk factors associated with poorer physical and psychological outcomes in HNC caregivers. A better understanding of factors affecting HNC caregivers’ outcomes would prove valuable in providing optimal support to HNC caregivers through educational and health promotion interventions, such as encouraging participation in physical activity. This study adds important information to a relatively limited literature base on HNC caregivers and may inform future interventions targeting HNC caregivers. Future studies should include larger and more diverse samples that allow generalization, as well as longitudinal research designs that can support evaluation of the stress process.