MATERIALS AND METHODS
Design and sample
This design was a descriptive, observational, case–control study that was performed following the Strengthening the Reporting of Observational Studies in Epidemiology statement and checklist.13Between January 2016 and May 2017, 200 patients were recruited from podiatric medicine and surgery clinics that provided treatment for foot diseases and disorders at University of Extremadura (Plasencia, Spain) and the Hospital Infanta Cristina (Badajoz, Spain). A consecutive sampling method was used to select the 200 participants. All patients provided consent and were enrolled in the study.
Inclusion criteria for the case group were women >18 years of age with breast cancer diagnosed by an oncologist and those who were undergoing chemotherapy treatment. Regarding the case group, patients with breast cancer had been undergoing chemotherapy treatment and were recruited during the time from the first session to the first year of treatment with chemotherapy. The control group consisted of healthy women without cancer.
Exclusion criteria for both groups were women <18 years of age, presence of other types of cancer, previous radiotherapy treatments, existence of foot deformities prior to undergoing chemotherapy treatment, neurological alterations, any systemic or chronic diseases, being non- or semi-independent in daily activities, and/or unable to understand instructions.
Sample size calculation
Sample size calculation was performed by means of the difference between two independent groups using the G*Power 188.8.131.52 software and based on the foot function domain scores (mean ± standard deviation [SD]) from the Foot Health Status Questionnaire (FHSQ) from a pilot study (n=40) with the two groups.14,15 A case group consisting of 20 women with breast cancer undergoing chemotherapy treatment (FHSQ foot function domain score =67.95±27.53) and a control group consisting of 20 healthy women (FHSQ foot function domain score =54.37±29.80) were used for the data analysis. Indeed, a one-tailed hypothesis, an effect size of 0.47, an α error probability of 0.01, a power (1-β error probability) of 0.80, and an allocation ratio (N2/N1) of 1 were used for the sample size calculations. Therefore, a total sample size of 182 women, 91 in each group, was estimated.
A single trained examiner, who measured height and weight, examined each participant. Furthermore, body mass index (BMI) was calculated with height (meters) and weight (kilograms) using Quetelet’s equation (BMI = weight/height2).16 All subjects were asked to complete the FHSQ.15 The investigator remained throughout the process of completing the questionnaire to help resolve any misunderstandings. Furthermore, the podiatric examiner performed a thorough examination in order to observe the foot pathologies that appeared after chemotherapy treatment.
Section 1 of the questionnaire consists of 13 questions reflecting foot health-related domains such as foot pain, foot function, footwear, and general foot health. This questionnaire has demonstrated a high degree of content, criterion, construct validity (Cronbach α =0.89–0.95), and high retest reliability (intraclass correlation coefficient =0.74–0.92).15 Indeed, this assessment tool was shown to be the most appropriate measure of foot health-related QoL. This questionnaire presents 13 questions that assess four health domains of the feet: 1) pain; 2) function; 3) general health; and 4) footwear. Each domain has a specific number of questions (four questions with regard to pain, four with regard to function, three with regard to footwear, and two regarding general foot health). Indeed, pain and function assessments are based on physical phenomena. Footwear evaluation utilized practical aspects related to availability and shoe comfort, and the perception of general foot health was based on patients’ self-assessment of the state of their feet. Several answers were permitted for each question permitted, and a Likert-type ordinal scale (words or phrases corresponding to a numeric scale) was applied to score the answers. The descriptors for this scale varied for each domain, and only one response was selected to be the most adequate and appropriate. The participant then circled this option on the Likert scale. The questionnaire did not provide a global score but rather generated an index for each domain. In order to obtain these scores, the responses were fed into a computer program that processes the data (The FHSQ, Version 1.03). After processing the data, a score ranging from 0 to 100 was possible. A score of 0 represented the worst state of foot health, while a score of 100 score was considered the best possible health status. Furthermore, the software provided graphical images to represent the outcomes. Section 2 evaluated questions that reflect four general health-related domains such as general health, physical activity, social capacity, and vigor. Questions were largely adapted from the Medical Outcomes Study 36-Item Short-Form Health,15 which has been demonstrated to be a valid measurement tool.17 The third section contained sociodemographic data such as participants’ age, their medical characteristics, educational level, profession, civil status, and medical records. Furthermore, the Spanish FHSQ version showed adequate psychometric properties.18
Finally, the Spanish EuroQol Health Questionnaire 5 Dimensions, which is a widely used non-disease-specific questionnaire to evaluate health-related QoL, negatively correlated with the FHSQ for pain (r= –0.445) and positively correlated with general foot health and function (r=0.261, –0.579), confirming criterion-related validity.18
The Bioethics and Biosafety Committee at the University of Extremadura approved this study (Spain, registry number: 28/2017). All participants provided written informed consent before inclusion in the study. The ethical standards for human experimentation given in the Declaration of Helsinki (World Medical Association) and the Council of Europe Convention on Human Rights and Biomedicine in addition the Universal Declaration of the UNESCO about the Human Genome and Human Rights and other appropriate national or institutional organizations’ ethical requirements were followed.
All variables were examined for normality of distribution using the Kolmogorov–Smirnov test, and data were considered to have a normal distribution if P>0.01. Demographic characteristics (such as age, height, weight, and BMI) and FHSQ domains (foot pain, foot function, footwear, general foot health, general health, physical activity, social capacity, and vigor) were described. Mean ± SD for parametric data, median ± interquartile range for nonparametric data, and maximum and minimum (range) values were used for the total sample: women with breast cancer who were undergoing chemotherapy (case group) and the healthy women (control group).