BACKGROUND
While head and neck cancers represent only 3.8% of newly diagnosed cancers (i.e., ~4950 new cases per year in Canada; 1250 in Quebec) (Public Health Agency of Canada, 2009), they are considered to have a poor prognosis, with 5-year survival rates varying from 35 (oral) to 65% (larynx), depending on tumor site and staging (Canadian Cancer Society, 2011). Including cancers of the upper aerodigestive tract (i.e., oral cavity, nasopharynx, oropharynx, hypopharynx, and larynx), the paranasal sinuses, and the salivary glands (Sobin et al., Reference Sobin, Gospodarowicz and Wittekind2010), head and neck cancers are generally diagnosed in advanced stages III or IV (50%), and in men (male:female ratio of 3:1) (Public Health Agency of Canada, 2009). The most prevalent cancers are oral (3600 new cases and 1150 deaths in Canada) and laryngeal (1150 and 490) (Canadian Cancer Society, 2011). A cross-sectional study of 8,265 cancer patients identified a significantly higher prevalence of mixed anxiety/depression symptoms in patients with head and neck, stomach, pancreatic, or lung cancer, compared with other cancer sites (Brintzenhofe-Szoc et al., Reference Brintzenhofe-Szoc, Levin and Li2009). Head and neck cancer compromises quality of life for up to 12–36 months posttreatment (Bjordal et al., Reference Bjordal, Ahlner-Elmqvist and Hammerlid2001; Hammerlid & Taft, Reference Hammerlid and Taft2001), with depression peaking at around 2–3 months postdiagnosis (Hammerlid et al., Reference Hammerlid, Ahlner-Elmqvist and Bjordal1999a; Hammerlid et al., Reference Hammerlid, Silander and Hornestam2001). Predictors of distress during and immediately following treatment include: female gender (especially involving facial surgery) (de Leeuw et al., Reference de Leeuw, de Graeff and Ros2001; Hassanein et al., Reference Hassanein, Musgrove and Bradbury2001; Katz et al., Reference Katz and Irish2003); young age (Hassanein et al., Reference Hassanein, Musgrove and Bradbury2001); advanced tumor stage (Hammerlid & Taft, Reference Hammerlid and Taft2001; Hammerlid et al., Reference Hammerlid, Ahlner-Elmqvist and Bjordal1999a; Reference Hammerlid, Silander and Hornestam2001; de Leeuw et al., Reference de Leeuw, de Graeff and Ros2001); illness-related symptoms (Hammerlid & Taft, Reference Hammerlid and Taft2001) and functional status (de Graeef et al., Reference de Graeef, de Leeuw and Ros2000; Devin et al., Reference Devin, Otto, Irish and Holland2010); surgery (Allal et al., Reference Allal, Nicoucar and Mach2003; Campbell et al., Reference Campbell, Marbella and Layde2000); combined treatment modality (surgery, radiotherapy, and/or chemotherapy) (Campbell et al., Reference Campbell, Marbella and Layde2000; de Graeef et al., Reference de Graeef, de Leeuw and Ros2000; Hassanein et al., Reference Hassanein, Musgrove and Bradbury2001); pain (de Graeef et al., Reference de Graeef, de Leeuw and Ros2000); disfigurement (Katz et al., Reference Katz and Irish2003); pretreatment distress (de Leeuw et al., Reference de Leeuw, de Graeff and Ros2001; de Graeef et al., Reference de Graeef, de Leeuw and Ros2000); dispositional optimism (de Graeef et al., Reference de Graeef, de Leeuw and Ros2000); living alone (Kugaya et al., Reference Kugaya, Akechi and Okuyama2000); poor emotional support/restricted social network (de Leeuw et al., Reference de Leeuw, de Graeff and Ros2001; Hassanein et al., Reference Hassanein, Musgrove and Bradbury2001; Katz et al., Reference Katz and Irish2003); smoking and alcohol abuse/dependency (de Graeef et al., Reference de Graeef, de Leeuw and Ros2000); and use of helpless/hopeless and avoidance coping (Campbell et al., Reference Campbell, Marbella and Layde2000).
While studying the predictors and correlates of distress and quality of life in head and neck cancer patients is important, it is also necessary to understand from their perspective whether their needs resulting from their cancer experience are being adequately met (e.g., physical and daily living needs, health system and information needs, patient care and support needs, psychological needs, and sexuality needs). Compared to quality-of-life evaluations or patient satisfaction surveys, needs surveys offer: (1) the ability to directly assess patients' perceived need for help; (2) quantification of the magnitude of need for help, thus allowing for prioritization of services; and (3) identification of individuals or patient subgroups with higher levels of need, thereby allowing us to target these patients and implement relevant preventive interventions (Bonevski et al., Reference Bonevski, Sanson-Fisher and Girgis2000).
Needs surveys have focused on general cancer patients (including a mixture of head and neck, breast, colorectal, prostate, lung, melanoma, and non-Hodgkin's lymphoma cancer patients) (Harrison et al., Reference Harrison, Young and Price2009), or have studied a specific cancer population such as melanoma outpatients or Chinese oral cancer patients (Chen et al., Reference Chen, Lai and Liao2009a; Reference Chen, Liao and Lin2009b; Reference Chen, Tsai and Liu2009c; Reference Chen, Lai and Cheng2011b). A systematic review of 57 studies (Harrison et al., Reference Harrison, Young and Price2009) quantifying the needs of general cancer patients identified the following needs in order of prevalence: information (6–93%), physical (7–89%), psychosocial (1–89%), psychological (12–85%), and activities of daily living (1–73%). Sexuality (33–63%), communication (2–57%), and existential/spiritual (14–51%) needs were seldom investigated. Patients in active treatment expressed the most varied and intense levels of needs, while a higher number of patients reported unmet needs posttreatment (Harrison et al., Reference Harrison, Young and Price2009). Highest levels of needs were associated with a younger age (Bonevski et al., Reference Bonevski, Sanson-Fisher and Girgis2000), female gender, cancer type (Boyes et al., Reference Boyes, Hall and Zucca2010), patients not in remission (Boyes et al., Reference Boyes, Girgis and Lecathelinais2009) or undergoing active treatment (Sanson-Fisher et al., Reference Sanson-Fisher, Boyes and Girgis2000), and a visit to an outpatient clinic during the last three months (Bonevski et al., Reference Bonevski, Sanson-Fisher and Girgis2000).
No study has systematically investigated the specific supportive care needs of general head and neck cancer patients using validated measures. Identifying these needs is a necessary first step to improving the care we provide in head and neck oncology clinics. It is recommended as the first step in intervention development in the Pan-Canadian Clinical Practice Guideline of the Canadian Partnership Against Cancer (2009) (Howell et al., Reference Howell, Currie and Mayo2009).
STUDY PURPOSE
The primary purpose of this study was to identify the met and unmet supportive care needs of head and neck cancer patients in such domains as physical and daily living needs, health system and information needs, patient care and support needs, psychological needs, and sexuality needs. Secondarily, the study aimed to: (1) explore how these needs differ according to certain demographic characteristics (i.e., gender, age, presence of medical comorbidities), disease variables (i.e., stage, trajectory), phase (pretreatment, active treatment, or posttreatment), treatment type (surgery, radiation, and/or chemotherapy), time elapsed since the cancer was diagnosed), levels of distress (i.e., on the Hospital Anxiety and Depression Scale [HADS]) (Zigmond & Snaith, Reference Zigmond and Snaith1983; Hopwood et al., Reference Hopwood, Howell and Maguire1991; Razavi et al., Reference Razavi, Delvaux and Bredart1992), and quality of life (i.e., physical, functional, emotional, and social/family well-being); and (2) on the basis of the above-mentioned primary and secondary analyses, target priorities for action and underline interventions worthy of pursuit.
METHODS
Design
We conducted a supportive care needs survey of patients seen in otolaryngology–head and neck surgery outpatient departments.
Eligibility and Exclusion Criteria
Patients were eligible if they came for an appointment at the otolaryngology–head and neck surgery outpatient clinic and were diagnosed with head and neck cancer, at least 18 years old, alert and capable of giving free and informed consent, and able to read English or French.
Recruitment
Following ethics approval from McGill University's institutional review board (IRB) and the involved hospitals' ethics committees, patients were recruited in the outpatient head and neck oncology waiting rooms of two McGill University teaching hospitals: the Jewish General Hospital and the McGill University Health Centre (Royal Victoria Hospital and Montreal General Hospital sites). Eligible patients were approached by doctors, nurses, secretaries, and/or unit coordinators involved in their care. They explained that the clinic was conducting a study to evaluate patients' supportive care needs and that this involved completing a brief questionnaire. They were given a pencil, a paper copy of the questionnaire with a cover letter explaining the study, and an envelope. The cover letter that accompanied the questionnaire described the study in detail and explained that completing the questionnaire implied consent to participate in the study. Patients were reminded that they were free to refuse to participate and that their decision would not affect their care. Interested participants completed the questionnaire while they were awaiting an appointment. Once they completed the questionnaire, they were instructed to place it in the envelope, seal it, and bring it back to the secretarial office. They were also reminded that they could complete the survey at home if they preferred and bring it back at their next appointment. This procedure was approved by the IRB and hospital ethics committees, given the minimal risks posed by this study. McGill University medical students (LH, MM, JWY, XL, SL) were available during clinics to answer any questions potential participants had and to offer referral for emotional support if needed.
MEASURES
Measures were selected on the basis of psychometric properties, potential to be self-administered, and utility for head and neck cancer patients. The questionnaire took approximately 20 to 30 minutes to complete.
The Supportive Care Needs Survey–Short Form (SCNS-SF34) (McElduff et al., Reference McElduff, Boyes and Zucca2004) has 34 items that assess the supportive care needs of cancer patients in five domains: psychological, health system and information, physical and daily living, patient care and support, and sexuality. The SCNS-SF34 has been translated into French using back-to-back translation and was validated in a group of 384 breast cancer patients (Bredart et al., Reference Bredart, Kop and Griesser2012). Participants respond to items on a 5-point Likert scale, including “0–Not applicable (no need),” “1—Need satisfied,” “2—Low need,” “3—Moderate need,” and “4—High need.” The overall score varies from 0 to 136, with higher scores indicating higher levels of needs. Internal consistency coefficients for the SCNS-SF34 subscales vary from 0.87 to 0.97 for the English version (McElduff et al., Reference McElduff, Boyes and Zucca2004) and from 0.80 to 0.93 for the French version (Bredart et al., Reference Bredart, Kop and Griesser2012). Internal consistency as calculated in this study using Cronbach's alpha was 0.97 for the total scale and varied from 0.89 to 0.96 for the subscales. Four relevant items were added to the questionnaire following multidisciplinary team discussions, as they were not covered in the SCNS-SF34 and were deemed relevant to head and neck cancer patients: “smoking cessation,” “alcohol cessation,” “feeling better about my appearance,” and “finding meaning and purpose in life.”
The Hospital Anxiety and Depression Scale (HADS) (Zigmond & Snaith, Reference Zigmond and Snaith1983) is a 14-item questionnaire measuring level of distress over the past week in two separate 7–item subscales. Patients respond to items on a 4-point Likert scale. Scores for each subscale range from 0 to 21; scores for the combined scale (HADS Total) range from 0 to 42. A total score of 16 on the HADS is customarily used as a threshold for clinically significant distress (Hopwood et al., Reference Hopwood, Howell and Maguire1991; Razavi et al., Reference Razavi, Delvaux and Bredart1992). Internal consistency is 0.93 for the anxiety subscale and 0.90 for the depression subscale (Zigmond & Snaith, Reference Zigmond and Snaith1983).
The Functional Assessment of Cancer Therapy–General (FACT-G ) (Cella et al., Reference Cella, Tulsky and Gray1993) has 27 items measuring quality of life in cancer patients in four domains during the past seven days: physical, social and family, emotional, and functional well-being. The FACT-Head and Neck Module (FACT-H&N) (List et al., Reference List, D'Antonio and Cella1996) has 12 additional items covering symptoms typically experienced by head and neck cancer patients. Scores on the FACT-G & FACT-H&N combined range from 0 to 156, with higher scores reflecting the highest quality of life. Internal consistency is 0.65–0.89 for the FACT-G subscales (Cella et al., Reference Cella, Tulsky and Gray1993) and 0.70 for the FACT-H&N (List et al., Reference List, D'Antonio and Cella1996). Test–retest reliability is 0.82–0.92 for the FACT-G subscales (Cella et al., Reference Cella, Tulsky and Gray1993).
As recommended by the Canadian Partnership Against Cancer (2009) (see Howell et al., Reference Howell, Currie and Mayo2009), we included basic sociodemographic and medical information such as age, sex, and tumor site. In addition, we collected information such as marital status, ethnicity, highest level of education, income, employment status, presence or absence of current disease, date of initial cancer diagnosis, disease stage, trajectory phase (pretreatment, active treatment, or posttreatment), type of treatment (surgery, radiotherapy, and/or chemotherapy), and time elapsed since last treatment. Additional questions specifically designed for this study assessed patients' willingness to participate in such activities as psychotherapy, receiving psychiatric medication, meeting another head and neck cancer patient, receiving medical information about the cancer and its treatment, and smoking and alcohol cessation. Patients rated their willingness on an 8-point Likert scale from 0 (“I would definitely not participate”) to 7 (“I would definitely participate”). An 8-point scale was used to avoid having a “neutral” middle response category, instead forcing patients to commit to either wanting or not wanting to participate in the listed services (Streiner & Norman, Reference Streiner and Norman2008).
Data Analyses
The data were analyzed using SPSS version 17.0. First, independent sample t tests, ANOVAs (for categorical independent variables), and Pearson correlations (for continuous independent variables) were run to explore associations between level of overall supportive care needs (i.e., total score on the Supportive Care Needs Survey or SCNS-SF34) and the following variables: sociodemographic characteristics, including gender, age, marital status, living arrangements, and medical comorbidities; disease variables, including cancer site, stage, trajectory phase, treatment type, and time elapsed since the cancer was diagnosed; level of distress, using HADS subscale scores; and quality of life, including physical, functional, emotional, and social/family well-being, as well as head and neck cancer–related items. Significant variables were entered as independent variables in a backward elimination multiple linear regression, conducted with overall level of supportive care needs on the SCNS-SF34 as the dependent variable. Three dummy variables were created for marital status, since this categorical variable presented more than two levels (i.e., married/common-law, divorced/separated, single, and widowed). The reference group used was the category “married/common-law,” since it was hypothesized as a protective factor (Olstad et al., Reference Olstad, Sexton and Soogard2001). The presence of this level was categorized as “1” and its absence “0.”
The data were then analyzed descriptively, identifying areas of supportive care needs for head and neck cancer patients. For the purpose of these analyses, each item from the SCNS-SF34 was analyzed separately and dichotomized following SCNS-SF34 guidelines (McElduff et al., Reference McElduff, Boyes and Zucca2004), whereby a score of 2 and above indicated an area of need and a score of below 2 the absence of need. We then further explored needs categorized as being of low (i.e., score of 3), moderate (i.e., score of 4), or high (i.e., score of 5) intensity. The Likert-type scale used to evaluate service preference was next dichotomized into 0–3 (“Did not wish to participate”) and 4–7 (“Wished to participate”). We then ran chi-square analyses to ascertain the relationship between level of distress on the HADS (threshold for distress >16 [Hopwood et al., Reference Hopwood, Howell and Maguire1991; Razavi et al., Reference Razavi, Delvaux and Bredart1992]) and service preference.
RESULTS
Sample
Of 194 survey questionnaires distributed, 146 were returned, yielding a response rate of 75.3%. Of the 146 returns, 127 (87%) had complete data available for analysis purposes. Participants were mostly men (66.7%), had a mean age of 61 years (SD = 12.7), were married/common-law (63.8%), and had been initially diagnosed with cancer less than 5 years before (91.5%). The oral, larynx, pharynx, and thyroid cancer sites were the most common. Tables 1 and 2 summarize well-being indicators, and the sociodemographic and medical characteristics of our sample.
Table 1. Well-being indicators and their association with the Supportive Care Needs Survey (N = 127)
1This percentage indicates the mean of percentage for each subscale.
Table 2. Sociodemographic and medical variables, and their association with the Supportive Care Needs Survey (N=127)
Supportive Care Needs
We first looked at the relationship between overall level of needs on our primary outcome (i.e., Supportive Care Needs Survey or SCNS-SF34) and variables of interest using ANOVAs, independent sample t tests, and Pearson correlations (see Tables 1 and 2). Overall level of needs varied according to: marital status [F(5,119) = 3.36, p = 0.004], time elapsed since the cancer was last diagnosed [r = –0.211, p = 0.021], anxiety (HADS) [r = 0.531, p = 0.01], depression (HADS) [r = 0.444, p = 0.01], physical well-being (FACT-G subscale) [r = –0.497, p = 0.01], functional well-being (FACT-G subscale) [r = –0.404, p = 0.01], emotional well-being (FACT-G subscale) [r = –0.533, p = 0.01], and head and neck cancer–related quality of life (FACT-H&N) [r = –0.363, p = 0.01]. No other significant associations were found.
A backward elimination multiple linear regression indicated that patients had a higher level of overall unmet needs when they were divorced or separated [ß = 13.35, p = 0.03], had a high level of anxiety (HADS anxiety subscale) [ß = 1.40, p = 0.02], or had low physical (FACT-G physical well-being subscale) [ß = –1.80, p = 0.001] or emotional well-being (FACT-G emotional well-being subscale) [ß = –1.42, p = 0.02; r 2adj = 0.47]. Table 3 presents the results of this multiple linear regression.
Table 3. Multiple linear regression analysis of head and neck cancer patients' responses on the Supportive Care Needs Survey (SCNS-SF34) (N = 127)
Since the HADS anxiety and FACT-G emotional well-being subscales were found to be significantly associated to overall level of unmet needs on the SCNS-SF34 and could have been found as a mere artifact of the SCNS-SF34 including psychological needs, we ran exploratory Pearson correlations between scores on the HADS anxiety subscale and the FACT-G emotional well-being subscale, respectively, and scores on the SCNS total minus the psychological needs subscale. These correlations were still significant [r = 0.473, p = 0.01 and r = –0.434, p = 0.01]. For the same reason, we also ran an exploratory Pearson correlation between scores on the FACT-G physical well-being subscale and scores on the SCNS total minus the physical and daily living needs subscales. This correlation was also significant [r = –0.431, p = 0.01].
While not hypothesized a priori, an additional exploratory analysis using an independent sample t test found that French-speaking head and neck cancer patients presented a significantly higher level of unmet needs compared to English-speaking ones [t(125) = –3.08, p = 0.003]. We thus ran the previous regression analysis including language as an additional variable. The results of this analysis are presented in Table 4.
Table 4. Multiple linear regression analysis of head and neck cancer patients' responses on the Supportive Care Needs Survey (SCNS-SF34), including language as an additional variable (N = 127)
Second, we examined specific areas of unmet needs by dichotomizing each item from the Supportive Care Needs Survey (SCNS-SF34) following SCNS-SF34 guidelines (McElduff et al., Reference McElduff, Boyes and Zucca2004). Sixty-eight percent of head and neck cancer patients experienced at least one unmet need for which they wanted help. Most frequently endorsed unmet needs predominantly involved psychological ones (7 of the top 10 needs), and included fears of the cancer spreading (40.5%), uncertainty about the future (34.3%), sadness (32.6%), worries about family/friends (32%), anxiety (31.8%), feelings about death and dying (29.5%), and feeling down/depressed (29.1%). Needs around body image (24.4%) and finding a sense of meaning and purpose in life (23.6%) were also frequent. Notably, very few patients reported a need for help with smoking (5.6%) or alcohol (6.4%) cessation. Table 5 presents head and neck cancer patients' overall unmet needs in descending frequency.
Table 5. Head and neck cancer patients' unmet needs in descending frequency
1Most frequently endorsed unmet needs of only moderate or high intensity.
2Defined as a score of 3, 4, or 5 on the Supportive Care Needs Survey (SCNS-SF34).
3Defined as a score of 3 on the SCNS-SF34.
4Defined as a score of 4 or 5 on the SCNS-SF34.
5Responses of all participants were included for the purpose of capturing all collected data.
A different set of unmet needs appeared when an exploratory analysis of the care needs applied an alternative categorization, whereby the patients' unmet needs were graded by intensity from low through medium to high. Under these circumstances, the prevalent unmet needs included two physical and daily living needs, four psychological needs, and four health system and information needs (see items related to note 1 in Table 5).
Service Preference
Responses to the service preference questions were dichotomized to reflect willingness or not to participate in various supportive/interventional services. The percentage of patients willing to participate in the various services were as follows: receiving medical information about the illness, treatments, and side-effects (53.2%); finding solutions to daily living problems (e.g., speech, eating, drinking) (37.9%); receiving psychotherapy to help them deal with the cancer (32%); meeting with another patient who has the same cancer for support (31.5%); and being helped with changes in appearance (23.6%). To a much lesser degree, they felt willing to participate in interventions addressing smoking (10.3%) or alcohol cessation (7.1%).
Patients experiencing a high level of psychological distress on the HADS (score ≥16) were twice as likely to report wanting to undergo psychotherapy (53.3%), to be helped to find solutions to daily living problems (56.7%), and to be helped with changes in appearance (78.6%) than those experiencing a low level of psychological distress [26.1, 33.0, and 46.4%; χ 2(122) = 7.62, p = 0.006; χ 2(121) = 5.34, p = 0.021; and χ 2(123) = 9.05, p = 0.003]. Patients' level of psychological distress on the HADS (dichotomized) did not influence their willingness to receive medical information about the illness, treatments, and side-effects [63.3 vs. 50.5%; χ 2(123) = 1.49, p = 0.23], to meet with another patient for support [40 vs. 30%; χ 2(123) = 1.54, p = 0.21], and to participate in interventions addressing smoking [10.3 vs. 9.8; χ 2(121) = 0.008, p = 0.93] or alcohol cessation [6.7 vs. 6.5; χ 2(122) = 0.001, p = 0.98].
DISCUSSION
The results of our study highlight the importance of psychological needs in head and neck cancer patients. They also indicate an openness to receiving help, supporting the creation of intervention programs specifically tailored to this population and targeting concerns highly endorsed by patients (e.g., fear of recurrence or progression, dealing with uncertainty and with the possibility of a shortened life, mourning losses related to the illness, including changes in appearance and function, and reclaiming control over one's life). It is noteworthy that among the top ten unmet needs we found a lack of energy/tiredness (second in frequency), not being able to carry on usual activities (third in frequency), and being informed about cancer that is under control or diminishing (i.e., in remission). These would also need to be attended to. The distinctive nature of head and neck cancer may account for patients' higher level of unmet psychological needs when compared with that of the general cancer population studied to date, for whom informational and service-related needs are most prevalent (up to 93 vs. 18.4% in our study) (Harrison et al., Reference Harrison, Young and Price2009). Furthermore, the highest levels of needs were associated in other studies with a younger age (Bonevski et al., Reference Bonevski, Sanson-Fisher and Girgis2000), female gender, cancer type (Boyes et al., Reference Boyes, Hall and Zucca2010), patients not in remission (Boyes et al., Reference Boyes, Girgis and Lecathelinais2009) or undergoing active treatment (Sanson-Fisher et al., Reference Sanson-Fisher, Boyes and Girgis2000), and a visit to an outpatient clinic within the last three months (Bonevski et al., Reference Bonevski, Sanson-Fisher and Girgis2000).
In contrast, we found that head and neck cancer patients experienced a higher level of unmet needs when they were divorced or separated, had a low physical quality of life, or manifested either high levels of anxiety or low emotional well-being. French-speaking patients also reported increased unmet needs compared to their English counterparts. Head and neck oncology clinics may be mindful of these markers and include them when screening for patients at increased need for help.
Divorced or separated patients may be living alone or lacking the support of a partner or spouse, thereby predisposing them to require more help (Olstad et al., Reference Olstad, Sexton and Soogard2001). This can especially be the case in head and neck cancer, a disease causing high levels of functional impairments and reliance on family caregivers for support (Ledeboer et al., Reference Ledeboer, van der Verden and de Boer2005). Married or common-law patients may have more needs met by their spouse, while those lacking such support can need either an intervention to help mobilize their supports or more hospital- or community-based help. This help may be needed for a longer period of time, since the disease is often fraught with visible and permanent sequelae (e.g., disfigurement, speech impairments, difficulties eating and breathing, pain) (Brintzenhofe-Szoc et al., Reference Brintzenhofe-Szoc, Levin and Li2009; Bjordal et al., Reference Bjordal, Ahlner-Elmqvist and Hammerlid2001; Hammerlid & Taft, Reference Hammerlid and Taft2001; Hammerlid et al., Reference Hammerlid, Ahlner-Elmqvist and Bjordal1999a). This circumstance, coupled with the reality of a poor prognosis and the possibility of cancer recurrence, may heighten the adaptation toll and bring psychological and existential angst to the forefront. It may also explain the new finding of a significant positive relationship between level of anxiety and level of need for help (Chen et al., Reference Chen, Yu and Chu2010), as anxiety has been shown to have a cumulative effect (i.e., enhancing levels of physical symptoms such as pain [Zaza & Baine, Reference Zaza and Baine2002] and fatigue [Tavio et al., Reference Tavio, Milan and Tirelli2002]). As for language, it remains unclear whether French-speaking patients' higher level of needs reflects an increased openness in acknowledging a need for help (Miyamoto & Ryff, Reference Miyamoto and Ryff2011), a communication barrier leading them to be underserved when treated in a traditionally English-speaking hospital environment (Health Canada, 2001), or that the French-speaking population is more geographically dispersed (i.e., in rural areas) outside the island of Montreal, thus limiting their access to care (Quebec Government, 2002).
Despite the overwhelming presence of psychological needs in head and neck cancer patients, very few psychosocial interventions have been developed for this specific cancer population, with most providing psychoeducation (i.e., coping [NUCARE; Allison et al., Reference Allison, Edgar and Nicolau2004; Vilela et al., Reference Vilela, Nicolau and Mahmud2006], bibliotherapy [Semple et al., Reference Semple, Dunwoody and Kernohan2009], and telehealth interventions [Head et al., Reference Head, Studts and Bumpous2009)]); group psychotherapy (Hammerlid et al., Reference Hammerlid, Persson and Sullivan1999b; Ostroff et al., Reference Ostroff, Ross and Steinglass2004); smoking cessation therapy (Duffy et al., Reference Duffy, Ronis and Valenstein2006); and physical rehabilitation (i.e., nutritional [Bertrand et al., Reference Bertrand, Piquet and Bordier2002; Isenring et al., Reference Isenring, Capra and Bauer2004a; Reference Isenring, Capra and Bauer2004b; Falciglia et al., Reference Falciglia, Whittle and Levin2005], speech [Perry et al., Reference Perry, Shaw and Cotton2003], and exercise interventions [McNeely, Reference McNeely2008]). Posttreatment head and neck cancer patients admitted to needing help with fears about the cancer spreading, which is the single most frequently expressed need. Patients with head and neck cancer may be especially prone to experience fears of recurrence, as the consequences of their cancer and treatments are often very visible (e.g., disfigurement and functional changes in eating, speech, and breathing), serving as a constant reminder of their plight (Humphris & Ozakinci, Reference Humphris and Ozakinci2008). Thus, more scholarship is required to better understand this construct in head and neck cancer and how fears of recurrence interface with disfigurement, as well as functional and sensory impairments. This comprehension could guide the implementation of relevant interventions such as the AFTER intervention, designed specifically to help head and neck cancer patients adjust to the fear, threat, or expectation of recurrence. This latter intervention was developed based on Leventhal's self-regulation model and employs cognitive-behavioral techniques and health-psychology principles to target recurrence fears, cancer-related beliefs, and inappropriate checking behaviors. While the AFTER intervention seems relevant to head and neck cancer patients and was reported by them to be acceptable in previous studies (Humphris & Ozakinci, Reference Humphris and Ozakinci2008), it has just begun to be empirically tested in randomized controlled trials (Humphris & Rogers, Reference Humphris and Rogers2012; Van Der Meulen et al., Reference Van Der Meulen, May and Ros2013).
Rigorous scientific studies are needed to evaluate psychological interventions in head and neck cancer patients, as these can have an important impact on patients' quality of life and on reducing healthcare costs (Tromp et al., Reference Tromp, Brouha and Hordijk2005). Distress and depression in cancer patients have been associated in other studies with immune function impairment, prolonged hospital stays, reduced self-care ability, noncompliance with treatment (Lydiatt et al., Reference Lydiatt, Moran and Burke2009), suicide (Yousaf et al., Reference Yousaf, Christensen and Engholm2005; Hem et al., Reference Hem, Loge and Haldorsen2004), and persistent at-risk behaviors (smoking and alcohol) (Zabora et al., Reference Zabora, Brintzenhofe-Szoc and Curbow2001; Duffy et al., Reference Duffy, Ronis and Valenstein2007), factors which may independently predict morbidity, survival, and/or disease recurrence (Hammerlid et al., Reference Hammerlid, Ahlner-Elmqvist and Bjordal1999a).
Interestingly, a very low proportion of patients in our study reported a need for help with smoking or alcohol cessation (~6–10%), while these behaviors are clear etiological factors in the development of head and neck cancers (Mehanna et al., Reference Mehanna, Paleri and West2010). Patients may have already stopped smoking or drinking alcohol at the time of questionnaire completion or, on the contrary, are unmotivated to change these behaviors. Underlying reasons for this finding should be explored in future studies, guiding the suitability of interventions such as motivational interviewing (Miller & Rollnick, Reference Miller and Rollnick1991), a technique particularly effective in motivating patients reluctant or ambivalent about changing at-risk behaviors. This is particularly important, as smoking and alcohol use significantly increase the risk of a second neoplasm (odds ratio [OR] smoking = 2.9; OR alcohol use = 5.2) (Leon et al., Reference Leon, Del Prado Venegas and Orus2009). Moreover, smokers have less chance of attaining a five-year survival (23 vs. 55% in nonsmokers) (Chen et al., Reference Chen, Chen and Vaughan2011a), a higher risk of treatment complications (poor wound healing, infections, more severe treatment side-effects) (Arcavi & Benowitz, Reference Arcavi and Benowitz2004; Goodwin et al., Reference Goodwin, McCarthy and Pusic2005), lower QOL, and less chance of disease-free survival (smokers 58 vs. non-smokers 69%) (Gritz et al., Reference Gritz, Schacherer and Koehly1999; Jensen et al., Reference Jensen, Jensen and Grau2007).
This study makes a significant contribution to our understanding of supportive care needs and related variables in head and neck cancer patients. However, several limitations are noteworthy. First, there was variability in time elapsed since the cancer was diagnosed. While this variance was important in ascertaining the impact of the disease trajectory on levels of needs, a future study could separately address patients' supportive care needs longitudinally, including the pretreatment phase, active treatment, and posttreatment, as the nature of these needs may evolve over time. Second, we included thyroid cancer patients in our analyses. Though included because they are frequently seen in head and neck oncology departments and did not differ significantly in levels of needs in our survey, they are often thought of as representing a different population (e.g., they are mostly women, with a better prognosis and younger age upon diagnosis). In this respect, their needs may be worthy of further independent inquiry. Third, some patients may not have responded to the questionnaire because their physical state precluded participation (perhaps there were very few patients in active treatment). Hence, our results may not reflect the needs of patients with a poor performance status or those with a high level of physical symptoms. Finally, because the questionnaire was only available in French and English, allophones may have been underrepresented.
Despite these limitations, this study is the first to systematically investigate the supportive care needs of general head and neck cancer patients using validated psychometric instruments. Identifying supportive care needs is a first step toward implementing meaningful clinical changes leading to improved quality of life among this underserved population. These changes are essential, considering that 68% of head and neck cancer patients are asking for help.
ACKNOWLEDGMENTS
We would like to acknowledge the valuable statistical input of Christina Holcroft, Sc.D., from the Statistical Consultation Service of the Jewish General Hospital's Centre for Clinical Epidemiology. We do not have a conflict of interest relative to this study. We have full control of all primary data and agree to allow the journal to review these data if requested.
