PropertyValue
is nif:broaderContext of
nif:broaderContext
is schema:hasPart of
schema:isPartOf
nif:isString
  • All patients with a diagnosis of malignant melanoma (ICD-10: C43) who had been registered by their dermatologist (n = 112) with the Cancer Registry of Rhineland-Palatinate as part of routine documentation were eligible. Inclusion criteria were a) registration in the time period from 2000 to 2005, b) age of 14 years or older at diagnosis and signed written informed consent for study participation. We excluded those patients a) who were not informed about their registration and b) did not have adequate German language comprehension or cognitive ability to participate. For legal reasons and in order to insure confidentiality, only the Cancer Registry was allowed to decode the name of the patient. Only the dermatologist who had originally registered the patient was eligible to contact the patient, who was then provided with a written informed consent form and the questionnaire. Patients who did not react to the letter within six weeks received a reminder letter and were contacted once more. Those who still did not respond were not contacted further. Data analysis was performed with coded data without reference to any personal identification. Two thirds (67%) of the dermatologists took part. They requested questionnaires for n = 1702 patients, representing 80.5% of all eligible patients (n = 2113). Of all contacted patients (n = 1320), n = 689 questionnaires were returned. Thus, 52.2% of the contacted patients took part, representing 32.6% of registered patients. We could not reach 19.4% of the patients because the corresponding dermatologists were not reachable. The protocol was approved by the Ethics Committee of the Statutory Physician Board of the State of Rhineland Palatinate (Reference number 837.161.11, 7703). Sociodemographic data were obtained using a standardised self-reporting questionnaire. Illness-related information (UICC stage, tumour site, time since diagnosis) was extracted from the Rhineland-Palatinate Cancer Registry. Psychosocial care needs were assessed by using the 9 Item Short Form of the Hornheide Questionnaire (HQ-S) [24], including physical and mental wellbeing, tumour-related anxiety, tension, self-esteem, social support, physician support and the information needs of skin cancer patients. This reliable measure (Cronbach´s α = .87 in the MeLa data set) has been developed for skin cancer patients and has shown good validity [25, 26]. The questionnaire allows for the identification and classification of patients in need of professional psychosocial or psycho-oncologic intervention. The degree of the distress is measured on a scale from 0 "does not apply" to 5 = "applies and troubles me extremely”. Item cut-off scores for each distress dimension and a cut-off score for the scale sum (≥ 16) were developed by the author of the test and validated by clinical interviews in order to classify the need for psychosocial intervention. Depression was measured using the German version of the Patient Health Questionnaire Depression Module (PHQ-9 [27]; Cronbach´s α = .86 in the MeLa data set). Categorisation of depression is defined as a score of PHQ-9 ≥ 10, indicating moderate to severe depressive symptoms for the past two weeks [28]. Sensitivity is 92% for major depression (cut-off ≥ 10) in primary care, while specificity is somewhat lower (80%) [29]. The Generalized Anxiety Disorder Questionnaire (GAD-7) of the PHQ was used to identify probable cases of generalised anxiety disorder and to assess symptom severity. The GAD-7 is based on the most prominent diagnostic features of the DSM-IV diagnostic criteria for generalised anxiety disorder. Seven items are scored on a four-point Likert scale over the past two weeks from 0 (“not at all”) to 3 (“nearly every day”), with a total score ranging from 0 to 21. Reliability, validity and standard values were established using a representative German population sample (n = 5036; age 48 ± 18 years, 54% female [30]. Internal consistency in the MeLa data set is Cronbach´s α = .89. The EORTC Quality of Life Core Questionnaire EORTC-QLQ-C30 [31] measures cancer-related quality of life reliably and with good validity using five functional scales (physical, role, cognitive, emotional, social; in the MeLa data, Cronbach´s α = .80/.90/.68/.89/.85 in order of the previous list), three symptom scales (pain, fatigue, nausea/vomiting; in the MeLa data set Cronbach´s α = .89/.85/.33), a global health and quality of life scale, and six single items assessing additional symptoms commonly reported by patients (e.g. appetite loss, sleep disturbance) as well as the perceived financial impact of the disease and treatment. The EORTC QLQ-C30 consists of 30 items that are scored on 4-point Likert scales, ranging from 1 (“not at all”) to 4 (“very much”). Two items in the global health and quality-of-life subscale are scored on a 7-point linear analogue scale. All functional scales and individual item scores are transformed to a 0–100 scale. Fatigue was assessed using the Multidimensional Fatigue Inventory (MFI), an internationally validated, multidimensional self-administered instrument [32]. Its 20 items form the following subscales (Cronbach α in the MeLa data set in parentheses): General (α = 81) physical (α = .86), and mental fatigue (α = .81), reduced motivation (α = .65), and reduced activity (α = .84). Each subscale contains four items (range from 4 to 20, with higher scores indicating increased fatigue). The German version shows moderate to good reliability Convergent validity was moderate for aspects of quality of life [33]. The 24-item German version of the Illness-specific Social Support Scale (ISSS) measures emotional, informational and practical support [34]. The two scales”positive social support” (15 items) and “detrimental interaction” (9 items) are scored on a 5-point Likert scale ranging from 0 (“never”) to 4 (“always”). The authors report a high reliability and good factorial validity. Internal consistency in the MeLa data set is Cronbach´s α = .94 for the positive social support scale and .72 for the detrimental interactions scale. Construct validity is supported by positive correlations to selected variables of social interaction. The Brief Cope (BC) is a short version of the COPE questionnaire [35, 36], which has proven useful in health-related research. It consists of 14 coping scales with a Likert scale (1 = “never” until 4 = “very often”), each with two items (e. g. “self-distraction”, “active coping”, “denial”, “substance use” and “use of emotional support”). Due to the fact that more than half of the scales have a Cronbach Alpha coefficient under .70, we decided to perform a cross validation in our sample. An exploratory factor analysis with the Brief Cope items led to three interpretable coping scales with acceptable reliability (accounting for 38% of total variance). The first dimension could be labelled as “seeking external support” (α = .75), the second “denial/self-blaming” (α = .74), and the third “active coping” (α = .76). The dimensions are in good accordance with other Brief-Cope-related analytic results [37, 38]. Since there were no standardised scales for measuring service and information needs for melanoma patients, we developed nine dichotomised single questions (“yes/no”) addressing main domains [14, 19] of support sources (psychological counselling, social counselling, pastoral care, self-help group, cancer care counselling, telephone counselling, internet counselling, internet forum with concerned persons). For measuring information needs we did not find a suitable standardised questionnaire and developed an ad hoc scale with 15 illness, therapy or psychosocial care related items covering main aspects: diagnosis, chance of recovery, course of the disease, treatment, side effects, other therapy options, course of the therapy, follow-up care, genetic risk, prevention, rehabilitation, psychosocial support, self-help groups, participation in studies, and a second opinion. Patients rated on Likert scales with five steps (1 = “not at all” to 5 = “very much”) about how adequately they felt informed about the issues. Ratings were dichotomised (0 = rather/much informed: no information deficit; 1 = not at all/little/a bit informed: information deficit) to determine the percentage of patients with an information deficit. For descriptive analysis, absolute and relative frequencies were computed for the categorical variables. Missing data were replaced according to the recommendations of the test authors. Associations between the standardised questionnaires were determined using Spearman´s correlation coefficients in order to determine the validity of the HQ-S. For the categorisation of ‘being in need’ of psychosocial intervention, the HQ-S item cut-off score and the sum score cut-off (calculated from the item scores) were used. Social class was determined using a common index [39]. In order to estimate the effects of the socio-demographic data (gender, age, marital status, education, social class, urban/rural destination), social variables (partnership/the medical issues (UICC status, time since diagnosis), comorbidity (PHQ-9, GAD-7) and psychosocial variables (scales of the MFI, Brief COPE, ISSS; EORTC-Symptoms, EORTC-Global Health) on the care needs (HQ-S), a logistic regression model was fitted. We present adjusted odds ratios, the corresponding 95% confidence intervals and p-values. Because of missing values in the potential predictors, a multiple imputation was performed [40]. We used a multiple imputation with k = 10 imputed data sets via the SAS procedure MI. For two randomly chosen single imputation data sets, we selected a set of covariates with a logistic regression model using forward and backward selection with a selection level of 5%. For each single imputation data set, we then performed logistic regression models for the previously selected covariates and combined the results via the SAS procedure MIANALYZE. This logistic regression model was used for confirmatory analyses. Statistical analysis was performed with the Statistical Package for the Social Sciences (SPSS, Chicago, IL, USA, version 21.0) and SAS 9.2 for Windows 9.2 TS Level 1M0 (SAS Institute Inc. Cary, NC, USA).
rdf:type