Alcohol Cessation Among Head and Neck Cancer Survivors

Alcohol Cessation Among Head and Neck Cancer Survivors: A Pilot RCT of a Tailored Text Message Based Intervention

Head-and-neck cancers (HNC) account for 4 percent of cancer diagnoses in the United States and for more than 66,000 annual cancer diagnoses. The prevalence rate of HNC among Veterans is 150% higher than the rate in the general population. Together with smoking, alcohol drinking is a major risk factor for HNC, responsible for approximately one-third of the cases worldwide. Overwhelming evidence from population-based studies show that alcohol drinking significantly increases the risk of recurrence of the primary HNC and of second primary malignancies, as well as negatively impacts HNC survivors' psychosocial health. Hence, several organizations (i.e., American Cancer Society, American Society of Clinical Oncology, and the World Cancer Research Fund) have issued guidelines recommending that individuals with HNC reduce or avoid alcohol altogether. Despite these recommendations, a substantial proportion of HNC survivors continue to use alcohol. The overall goal of the proposed research is to: Adapt an existing evidence-based text message alcohol cessation intervention for HNC survivors in both civilian and VA settings (i.e., at two sites, Northwell Health and the Brooklyn VA Medical Center); and Preliminarily evaluate, in a two-arm pilot RCT, the acceptability and preliminary efficacy of the intervention, as well as feasibility of conduct a future RCT. The investigators hypothesize that: H1: The tailored text-message intervention will be 1) feasible to evaluate in a large-scale RCT, defined as achieving an enrollment rate of ≥70% in this pilot; and 2) acceptable to participants, defined as a score ≥4 on a 5-point Likert scale ranging from "not at all" to "extremely" acceptable. H2: Compared to the control condition of alcohol assessment and feedback (AF), the tailored text messages will result in a 30% increase in cessation among survivors (assuming also a 20% increase in cessation in the AF arm).

HNC survivors (baseline N=138) will be enrolled from Northwell Health and the Brooklyn VA Medical Center, at the 3-month post treatment completion follow-up visit, when alcohol resumption is most likely to occur according to pilot data. The investigators will randomize survivors to either the control condition (AF; n=69) or the text message arm, consisting of: AF plus HNC tailored texts for civilians (n=34) and AF plus HNC tailored texts for Veterans (n=35), for a total of n = 69. The investigators expect 15% attrition over 6 months for a final sample size of N=120. The study's primary outcome is alcohol cessation at 6 months post-baseline. Secondary outcomes are: feasibility of enrolling patients (consenting 70% of eligible approached patients), acceptability of the text messages (a score ≥4 on a 5-point Likert scale ranging from "not at all" to "extremely" acceptable), and improved psychosocial functioning (improved QOL, and decreased depression, social isolation, loneliness). If patients are interested, the HNC survivor will meet with a member of the study team to give informed consent, record baseline data, and assess drinking patterns with the AUDIT questionnaire and the Time-Line Follow Back (TLFB) alcohol use assessment procedure. The study staff will access provide alcohol use feedback and randomize the participant to either the active comparator (usual care) or the intervention text messaging arm. Block randomization will occur to ensure that an even number of civilian and Veteran HNC survivors will be allocated to both conditions, and that even numbers per condition across sites are maintained overall. Usual care at both study sites is to provide general information about smoking and alcohol. For the present study, the investigators will enhance this usual care by assessing alcohol use systematically with the AUDIT questionnaire and the TLFB to obtain an accurate assessment of drinking. Regardless of drinking level, the patient will receive feedback about the reported alcohol consumption. The feedback will emphasize that, according to leading medical associations, any consumption of alcohol is not advisable for cancer survivors, as continued alcohol use is associated with a higher risk of cancer recurrence and reduced quality of life. A sample feedback statement is "one or two drinks might not seem much, but it can increase the participant's chances of cancer recurrence". Non-drinkers will be told to continue to abstain from consuming alcohol. Smoking cessation information is also part of usual care. Patients will be urged to stop smoking and referred to a smoking cessation program if so desired. The investigators will further remind patients to complete two follow-up questionnaires, one at 3 and one at 6 months post baseline, during scheduled follow-up visits. No further effort will be required, but patients randomized to the control condition will be offered the HNC text messages after completing all study procedures. Survivors randomized into the intervention condition will receive usual care (described above) plus HNC tailored alcohol cessation text messages tailored for the general public or specifically designed for veterans, depending on recruitment site (Northwell vs. VA, respectively). Survivors will receive messages three times a week, usually during the late afternoon/early evening period, when the risk of alcohol drinking is greatest. Messages will be delivered for 3 months. Demographic information, personal and family health history, and treatment type will be assessed. Patient will report any alcohol and smoking cessation advice they might have received from physicians. Overall health will be assessed with the SF-12 questionnaire. In addition, the AUDIT and TLFB will be used to assess frequency of alcohol use during the previous three months. Feasibility of a future RCT will be assessed through the number of survivors recruited vs. approached (enrollment rate threshold of 70%), Acceptability of the intervention will be determined through a positive evaluation of intervention (mean value of 4 on a 5-point scale, with higher values = higher acceptability). The primary outcome will be assessed using the TFLB interview, which will be converted into a binary drinking/no-drinking variable. Secondary outcomes include quality of life, depression, social isolation, and loneliness. The primary goal of the proposed work is to assess the acceptability of the proposed intervention, to estimate effect sizes for the two arms, as well as to determine feasibility of a future RCT. For the estimation of effect sizes for the primary measure of alcohol cessation, the needed sample size is 138 survivors, equally divided among AF and the intervention arm (n = 69 each) across the two sites. The sample size is not powered to detect a statistically significant difference between the intervention and AF arms. It will provide investigators, however, with effect size estimates to power a future RCT. The current sample size is based on well-established literature that pilot studies should accrue at least 10% of the patients needed for a fully powered trial. Based on the prior literature, the investigators assume that the rate of abstinence in the AF arm will be 20%, and among the text arm 30%. Thus, the investigators estimate an average effect size of 10% between AF and the text conditions. These effect sizes are considered small (h=0.2) and would be clinically meaningful.

Head and Neck Cancer, Veterans Health, Substance Abuse, Quality of life, Depression, Social Isolation, Loneliness, Alcohol Cessation, Alcohol Drinking

The study model is a randomized control pilot trial consisting of two arms, control and intervention group. Control condition will consist of a provider-delivered alcohol cessation intervention with usual-care-advice. The intervention condition will consist of a text based intervention designed to promote alcohol cessation in addition to usual-care-advice.

Intervention condition will consist of a text message program designed to promote alcohol cessation tailored to HNC patients and targeted to VA or Civilian populations in addition to usual-care-advice.

Control condition will consist of a provider-delivered alcohol cessation intervention with usual-care-advice.

Survivors will self-report the amount of alcohol they are consuming at the 6 month follow-up assessment using the Time-Line Follow Back (TFLB) interview, a semi-structured interview designed to collect information about survivors' daily substance use. The TLFB was validated in 1996 and has demonstrated good test-retest reliability, convergent validity, and agreement with collateral reports of alcohol abuse. Participants will be asked to report their substance use over the prior 4 weeks via phone or electronic survey. Responses will be converted to a binary outcome of drinking reported or no drinking reported. Fewer participants reporting alcohol use would be considered success.

Survivors will self-report the amount of alcohol they are consuming at the 3 month follow-up assessment using the Time-Line Follow Back (TFLB interview), a semi-structured interview designed to collect information about survivors' daily substance use. The TLFB was validated in 1996 and has demonstrated good test-retest reliability, convergent validity, and agreement with collateral reports of alcohol abuse. Participants will be asked to report their substance use over the prior 4 weeks via phone or electronic survey. Responses will be converted to a binary outcome of drinking reported or no drinking reported. Fewer participants reporting alcohol use would be considered success.

Quality of life will be assessed with the Head and Neck Quality of Life instrument, a 20-item Likert scale measure that evaluates quality of life related to eating, communication, pain, and emotion in the last 4 weeks. This measure has been validated with both civilian and VA populations. This survey will be administered via phone or electronic survey during the baseline assessment. Univariate and descriptive analyses will be performed and normalizing and/or variance stabilizing transformations will be conducted. The investigators will examine the data for outliers and use descriptive and inferential statistical techniques for data characterization. Lower scores indicate less impairment on quality of life.

Quality of life will be assessed with the Head and Neck Quality of Life instrument, a 20-item Likert scale measure that evaluates quality of life related to eating, communication, pain, and emotion in the last 4 weeks. This measure has been validated with both civilian and VA populations. This survey will be administered via phone or electronic survey during the 3 month follow-up assessment. Univariate and descriptive analyses will be performed and normalizing and/or variance stabilizing transformations will be conducted. The investigators will examine the data for outliers and use descriptive and inferential statistical techniques for data characterization. Lower scores indicate less impairment on quality of life.

Quality of life will be assessed with the Head and Neck Quality of Life instrument, a 20-item Likert scale measure that evaluates quality of life related to eating, communication, pain, and emotion in the last 4 weeks. This measure has been validated with both civilian and VA populations. This survey will be administered via phone or electronic survey during the 6 month follow-up assessment. Univariate and descriptive analyses will be performed and normalizing and/or variance stabilizing transformations will be conducted. The investigators will examine the data for outliers and use descriptive and inferential statistical techniques for data characterization. Lower scores indicate less impairment on quality of life.

Depression will be evaluated with the Patient Health Questionnaire (PHQ) 2-item version, a brief and accurate measurement of the core symptoms/signs of depression and will be used to help assess patients' quality of life. This survey will be administered via phone or electronic survey during the baseline assessment. Univariate and descriptive analyses will be performed and normalizing and/or variance stabilizing transformations will be conducted.The investigators will examine the data for outliers and use descriptive and inferential statistical techniques for data characterization. Higher scores indicate increased symptoms of depression.

Depression will be evaluated with the Patient Health Questionnaire (PHQ) 2-item version, a brief and accurate measurement of the core symptoms/signs of depression and will be used to help assess patients' quality of life. This survey will be administered via phone or electronic survey during the 3 month follow-up assessment. Univariate and descriptive analyses will be performed and normalizing and/or variance stabilizing transformations will be conducted.The investigators will examine the data for outliers and use descriptive and inferential statistical techniques for data characterization. Higher scores indicate increased symptoms of depression.

Depression will be evaluated with the Patient Health Questionnaire (PHQ) 2-item version, a brief and accurate measurement of the core symptoms/signs of depression and will be used to help assess patients' quality of life. This survey will be administered via phone or electronic survey during the 6 month follow-up assessment. Univariate and descriptive analyses will be performed and normalizing and/or variance stabilizing transformations will be conducted.The investigators will examine the data for outliers and use descriptive and inferential statistical techniques for data characterization. Higher scores indicate increased symptoms of depression.

Social isolation will be evaluated with the 6-item Lubben Social Network Scale, a validated instrument designed to gauge social isolation in older samples by measuring the number and frequency of social contacts with friends and family members. This survey will be administered via phone or electronic survey during the baseline assessment. Univariate and descriptive analyses will be performed and normalizing and/or variance stabilizing transformations will be conducted. The investigators will examine the data for outliers and use descriptive and inferential statistical techniques for data characterization. Higher scores indicate a larger social network.

Social isolation will be evaluated with the 6-item Lubben Social Network Scale, a validated instrument designed to gauge social isolation in older samples by measuring the number and frequency of social contacts with friends and family members. This survey will be administered via phone or electronic survey during the 3 month assessment. Univariate and descriptive analyses will be performed and normalizing and/or variance stabilizing transformations will be conducted. The investigators will examine the data for outliers and use descriptive and inferential statistical techniques for data characterization. Higher scores indicate a larger social network.

Social isolation will be evaluated with the 6-item Lubben Social Network Scale, a validated instrument designed to gauge social isolation in older samples by measuring the number and frequency of social contacts with friends and family members. This survey will be administered via phone or electronic survey during the 6 month assessment. Univariate and descriptive analyses will be performed and normalizing and/or variance stabilizing transformations will be conducted. The investigators will examine the data for outliers and use descriptive and inferential statistical techniques for data characterization. Higher scores indicate a larger social network.

Loneliness will be evaluated with the 6-item De Jong Gierveld Loneliness Scale, a reliable and valid continuous measure for overall, emotional and social loneliness. This survey will be administered via phone or electronic survey during the baseline assessment. Univariate and descriptive analyses will be performed and normalizing and/or variance stabilizing transformations will be conducted. The investigators will examine the data for outliers and use descriptive and inferential statistical techniques for data characterization. Higher scores indicate increased loneliness.

Loneliness will be evaluated with the 6-item De Jong Gierveld Loneliness Scale, a reliable and valid continuous measure for overall, emotional and social loneliness. This survey will be administered via phone or electronic survey during the 3 month assessment. Univariate and descriptive analyses will be performed and normalizing and/or variance stabilizing transformations will be conducted. The investigators will examine the data for outliers and use descriptive and inferential statistical techniques for data characterization. Higher scores indicate increased loneliness.

Loneliness will be evaluated with the 6-item De Jong Gierveld Loneliness Scale, a reliable and valid continuous measure for overall, emotional and social loneliness. This survey will be administered via phone or electronic survey during the 6 month assessment. Univariate and descriptive analyses will be performed and normalizing and/or variance stabilizing transformations will be conducted. The investigators will examine the data for outliers and use descriptive and inferential statistical techniques for data characterization. Higher scores indicate increased loneliness.

The investigators will evaluate acceptability of the intervention, defined as a positive evaluation of the intervention as indicated by a mean value of 4 on a 5-point scale, with higher values indicating higher acceptability.

The investigators will evaluate acceptability of the intervention, defined as a positive evaluation of the intervention as indicated by a mean value of 4 on a 5-point scale, with higher values indicating higher acceptability.

The investigators will evaluate feasibility of conducting a future, large-scale efficacy RCT, defined as a recruitment rate of 70% (number of survivors recruited over approached) in this pilot RCT.

The investigators will evaluate feasibility of conducting a future, large-scale efficacy RCT, defined as a recruitment rate of 70% (number of survivors recruited over approached) in this pilot RCT.

Inclusion Criteria: Adult male or female patients 18 years or older Diagnosed with primary cancer of: pharynx (nasopharynx, oropharynx, hypopharynx), larynx (all subsites), oral cavity (all subsites) Having completed surgical, radiation, and/or chemotherapy treatment; Post-treatment completion status of at least 3 months; Able to communicate and read in English; Possess a telephone with text messaging capability; and Confirmation of alcohol consumption. Exclusion Criteria: Non-English speaker Non-drinker Previously enrolled in adaption phase focus-groups to determine intervention content Clinically significant depression, as indicated by an assessment of ≥ 16 on the CES-D 11-item Iowa short form scale