Single-centre prospective follow-up study of colorectal cancer patients diagnosed at the Complexo Hospitalario Universitario A Coruña (A Coruña, Spain) during the period 2011–2012. Incident cases with anatomopathological confirmation of colorectal cancer according to the International Classification of Diseases (ICD) 9th revision (codes 153 and 154) are included. Prevalent or recurrent cases are excluded, together with cases of multiple cancers, cases that are only dealt with in private hospitals, cases detected through colorectal cancer screening, and cases diagnosed in another hospital but which are referred for treatment to the hospital where the study is conducted.
Eligible patients are identified from the Department of Pathology within one month of diagnosis. At an initial interview with a trained nurse, patients are informed about the aims and characteristics of the study. After signing the Informed Consent Form, information is obtained from patients regarding their sociodemographic characteristics, family history of cancer, first symptoms attributable to cancer perceived by the patient, symptom perception and reaction to early symptoms. At the same time, patients are asked to complete the questionnaires concerning their health-related quality of life and satisfaction with hospital care. The patients’ functional impairment is also evaluated using Karnofsky Performance Status.
In the case of patients qualifying for curative surgical treatment, the initial interview is performed in the month following discharge after surgery. Patients subjected only to palliative procedures are interviewed within three months of the diagnosis. All of them will be followed-up for a maximum of 2 years, until December 2013. For survivors, health-related quality of life and functional status assessments will be re-evaluated by means of a personal interview with a trained nurse at one and two years after the diagnosis during the study period.
Clinical records are also reviewed to collect information about comorbidity at diagnosis (Charlson comorbidity index), tumour characteristics at diagnosis (site according to ICD-9th, tumour size, histological grade, TNM stage, location of metastases and infiltration of adjacent organs), treatment received (surgery, chemotherapy and/or radiation therapy), hospital consultations related to the colorectal cancer and exploratory procedures in the follow-up. Occurrence of the following incidents in the follow-up is also registered: local recurrence, development of metastases, appearance of a new tumour, and mortality.
Symptoms-to-diagnosis interval is defined as the time elapsed from the date the patient perceived the first symptoms until the cytohistological confirmation of the diagnosis of cancer (date of biopsy or direct surgery). Treatment delay is defined as the interval from date of diagnosis to date of treatment (i.e., date of surgery, date of preoperative or postoperative radiotherapy or chemotherapy or date of palliative treatment). For those patients who do not receive any treatment, the date of the visit closest to the decision not to treat the patient will be recorded.
Surveillance strategies will be classified based on the frequency of follow-up and the investigations performed, taking into account: a) history and physical examinations, b) serum carcinoembryonic antigen (CEA) determinations, c) imaging examinations (ecographies, positron emission tomographies, computed tomographies, magnetic resonances), and d) endoscopy explorations (colonoscopies or rectoscopies).
Health-related quality of life
Health-related quality of life, the main outcome of this study, is assessed with the Quality of Life Questionnaire Core 30 (QLQ-C30) (version 3.0)  and the colorectal cancer-specific quality of life questionnaire module (QLQ- CR29) , developed by the European Organisation for Research and Treatment of Cancer (EORTC). These questionnaires have been translated and validated for use in Spain [36, 37].
The EORTC QLQ-C30 is a generic cancer health-related quality of life questionnaire, which has to be supplemented by disease-specific modules for each type of cancer or tumour location. The EORTC QLQ-C30 consists of 30 items and includes a scale measuring the global health status/health-related quality of life, 5 functioning scales (physical, role, emotional, cognitive and social), 3 symptom scales (fatigue, nausea/vomiting and pain), and 6 single-item scales (dyspnea, insomnia, appetite loss, constipation, diarrhoea and financial impact).
The QLQ-CR29 is a colorectal cancer disease–specific module consisting of 29 items. These are grouped into 4 functioning scales (body image, anxiety, weight and sexual interest) and 18 symptom scales (urinary symptoms, stool characteristics, gastrointestinal symptoms, pain and other symptomatology) . Items related to gastrointestinal symptoms are collected differently in stoma and non-stoma patients. Problems associated with sexual activity are evaluated differently for men and women, with a separate item for each gender.
The items on both questionnaires will be scaled and scored using the recommended EORTC procedures . After estimating the average of the items that contribute to each scale, raw scores will be linearly transformed to a scale ranging from 0 to 100, with a higher score representing a higher (better) level of functioning or a higher (worse) level of symptoms. According to EORTC scoring guidelines, this procedure is only applicable to those patients with at least half of the items answered for each of the scales. In that case, any items with missing values will be ignored when making the calculations.
Satisfaction with hospital-based care
Patient satisfaction with care is assessed by means of the validated Spanish version of the EORTC IN-PATSAT32 questionnaire [30, 31]. This questionnaire was developed by the EORTC Quality of Life Group for assessing cancer patients’ perception of the quality of hospital-based care.
The EORTC IN-PATSAT32 is composed of 32 items organised into eleven multi-item scales and three single items assessing cancer patients’ perception of the quality of hospital doctors and nurses, as well as selected aspects of the care organisation and hospital environment. Each question is administered using a 5-category Likert response scale (“poor”, “fair”, “good”, “very good” or “excellent”).
The dimensions measured include doctors’ and nurses’ technical skills, interpersonal skills, information provision, and availability; satisfaction with other hospital staff interpersonal skills and information provision; exchange of information with the care team; waiting-time; hospital access; hospital comfort; and overall satisfaction with care. Following the standard scoring procedure for the EORTC IN-PATSAT32, all scores will be linearly transformed to a 0–100 scale, with higher scores reflecting a higher level of satisfaction.
Functional status of colorectal cancer patients is assessed by means of the Karnofsky Performance Status Scale (KPS) . The KPS is an 11-point rating scale which ranges from normal functioning (100) to dead (0), used to assess patients’ physical functional level related to cancer and its treatment.
Sample size is limited by both the duration of the study and the number of incident colorectal cancer patients diagnosed per year. During the period 2011–2012 approximately n = 500 patients will be diagnosed of colorectal cancer. Assuming a non-response rate of 25%, a sample size of n = 375 patients is expected.
Based on their 0–100 scoring system, and using the range rule of thumb, a standard deviation of 25 points (one quarter of the range score) is assumed for the final punctuations of the EORTC QLQ-C30, QLQ-CR29 and IN-PATSAT32 questionnaires. Therefore, a sample size of n = 375 patients will allow us to estimate mean values with a precision of ±2.6 for a 95% confidence interval. Working with a power of 80% and an alpha of 0.05, score differences of 8 points between groups of patients will be detected as statistically significant, assuming an exposure of 50% to the variable of interest. This sample size will also allow detecting as statistically significant (p ≤ 0.05) correlation coefficients ≥0.15 among the studied variables and the questionnaires scores.
Descriptive analyses will be performed for all variables. Continuous variables will be reported using means ± standard deviations (SD) or median (interquartile range). For dichotomous/categorical variables, absolute numbers and percentages will be computed, together with their 95% confidence intervals.
Health-related quality of life, satisfaction with care and functional status scores will be compared according to patients’ characteristics and disease variables. The comparison of means will be carried out using Student’s t test, Mann–Whitney test, analysis of variance or the Kruskall-Wallis test as appropriate. The association of qualitative variables will be carried out using Chi-square statistics. The correlation among quantitative variables will be assessed using Spearman’s Rho correlation coefficient, due to the expected non-normal distribution of the questionnaires scores and in order to detect nonlinear relatioships.
In the multivariate analysis, multiple linear and logistic regression models will be used to identify those variables independently associated with patients’ functional status after the diagnosis, quality of life and satisfaction with care scores. Separate regressions will be conducted for each of the three outcomes. Box-Cox normalizing transformations will be used when necessary to ensure the normality assumption in the linear regression model.
Finally, evolution of quality of life at one and two years from diagnosis will be analysed based on the change from baseline scores for each time point. Significance of the changes will be assessed using the Wilcoxon signed ranks test. Clinical relevance will be analyzed by means of the minimal important difference (MID), effect sized, and standard error of measurement (SEM). Information provided by the developers of the EORTC QLQ-C30 about what could be considered as a MID will be used. Therefore, a mean change score of 5 to 10 will be a “little” change, about 10 to 20 a “moderate” change, and more than 20 a large change. Cohen’s guidelines for interpreting small (effect size 0.2-0.49), moderate (effect size 0.5-0.79), and large (effect size ≥0.5) will also be considered. Finally, a value greater than 1 SEM will be considered as clinically significant.
Since measurements of QOL will be made repeatedly on the same patients, we will be in a repeated measures context. Longitudinal QOL will be analysed with a linear mixed-effects regression model. More specifically, the relationship of QOL with time will be determined using a random coefficients model, which generalizes linear regression techniques to allow for repeated observations. These models take into account the correlation within observations on the same subject and allow for inclusion of data on subjects who have only partial follow-up without imputing missing data.
Therefore, a linear random coefficients regression model will be performed, with QOL scores over time as the dependent variable. The relationship between QOL and time will be modeled by including a quantitative time effect (months since diagnosis) as a covariate in the model, fitting patient effect and patient*time interaction as random effects. In addition to the effect of time on the QOL outcome measures, we will incorporate other factors into the regression models, in order to adjust for those variables of interest and potential confounding factors such as: sociodemographic variables, diagnosis/treatment delay, follow-up strategies, comorbidy, TNM stage at diagnosis, or presence of colostomy.
Two-sided tests will be used, and p-values < 0.05 will be considered as statistically significant. Statistical analyses will be performed using SPSS for Windows (version 19.0, SPSS Inc., Chicago, Illinois) and R (version 2.12.2). Apart from performing an analysis of all of the patients, the patients with colon cancer and patients with rectal cancer will be analysed independently.
The study will be carried out according to the Good Clinical Practice guidelines of the Helsinki declaration. Informed consent is obtained from each patient to take part in the study and to review their clinical records. This project was approved by the corresponding ethics review board (Clinical Research Ethical Committee of Galicia, decision 2010/304).