The use of social networks in scientific research with questionnaires

Aim: The aim of this study was to determine the response rate of the SF-36 quality of life questionnaire sent and received by mail and over a social network to caregivers of individuals with epidermolysis bullosa (EB) in Brazil. Methods: All volunteers were first-degree relatives of patients with EB that directly spent time helping them with their basic activities of daily living. A maximum of two caregivers per patient could answer the questionnaire. Volunteers were divided into two groups: for group 1, questionnaires were sent to 53 members of a support association for the disease by mail, and for group 2, 798 members of the FacebookTM page of friends and relatives of patients with EB were invited to participate using an access link to Google DriveTM to gain access to the questionnaire. The data from both groups were analyzed 150 days after the start the study. Descriptive analysis was performed by EpiInfo8 and the return of questionnaires was evaluated according to age, sex, and time of return using the Chi-squared and Fisher’s exact test. Results: After 150 days, 30 questionnaires were returned, 17 (56.7%) of which from group 1 and 13 (43.3%) from group 2. Approximately 12 questionnaires were returned by mail and 11 over the social network. Conclusions: The data collection of the SF-36 quality of life questionnaire from caregivers of individuals with EB over a social network seems to be efficient mainly when it is necessary to collect results within a short timeframe, highlighting the importance of social networks as a means for conducting this type of research. However, in our study, the most efficient method was distributing the questionnaires by mail.


Introduction
Questionnaires are an important tool in clinical research for outlining a situation or the way of life of a population [1][2][3][4][5][6] .Usually, the forms are distributed to the public through the mail or via censuses 3,5 .However, with the rise of the internet and the advent of electronic social networks, such as Facebook™ in 2004, distances to the target population were shortened, especially in Brazil 1,5,[7][8] .Studies have assessed the acceptability and representability of social networks in information science in Brazil 1 , which led researchers to apply electronic surveys more often 3,5,8 .In addition, the SciELO organization (online library) held a seminar in 2012 on the use of social networks in scientific communication, emphasizing the importance of this communication vehicle 1,2,[7][8] .
The distribution of questionnaires through social networks might be an effective approach to study epidermolysis bullosa (EB), which is a serious and rare dermatological genetic disorder with a prevalence of 5:100,000 live births and no racial or gender difference 9 .EB is a group of inherited connective tissue disorders characterized by the absence of a cohesion protein that results in the defective connection between epidermis and dermis, making the skin fragile [10][11] .Based on the skin layer that is affected, EB can be classified into four main types: EB simplex, junctional EB, dystrophic EB, and the recently established Kindler syndrome 9,12 .Depending on the severity of the disease, which can vary from benign to life threatening, symptoms can include skin fragility, blistering of the skin following mild friction or trauma, and blistering of mucous membranes or internal organs [11][12][13] .Currently, the disease has no cure and clinical management is focused on relieving symptoms 11,13 .Topical agents and dressings are typically used for the treatment of skin lesions, and appropriate follow-up is essential to monitor the patient and his caregivers for psychosocial problems and psychiatric symptoms [11][12][13][14] .
In addition, the occurrence of problems related to mobility, self-care, and usual activities, pain, discomfort, anxiety, and depression also impact the quality of life of patients and caregivers 12,13 .Non-professional caregivers, who are often relatives, provide informal care spending time helping the individuals with EB with their basic activities of daily living 15 .
Patients with EB have direct medical costs with hospital admissions, emergency visits, and outpatient care (rehabilitation, medical tests and examinations, visits to health professionals and home medical care) and non-medical costs with transportation, social care services (formal care), and caregiver's time.The time dedicated to the patient and the high informal care costs account for more than half of EB's socio-economic burden.The hidden social costs of EB are relevant and have consistent impact on the quality of life of patients and their caregivers, especially when considering the impact on family income levels 15 .
Thus, the aim of this study was to determine the response rate of the SF-36 quality of life questionnaire delivered by mail and through a social network to first-degree relatives of individuals with EB in Brazil.We discuss the use of surveys and questionnaires in EB caregivers, emphasizing the acceptability of social networks as an important tool for scientific research.

Sample characteristics
This study was approved by the Human Research Ethics Committee from the Federal University of Minas Gerais (protocol number ETIC 285 204).Each participant was informed about the research, and answered questions on their relationship with the individual with EB.All agreed by signing the informed consent that was sent with the SF-36 questionnaire.
The inclusion criteria for the participants were first-degree relatives that spent time directly helping patients with their basic activities of daily living.The maximum of two caregivers could answer the questionnaire and they were divided into two groups.Group 1 was selected from 83 files from the Minas Gerais Association of Relatives, Friends and Patients with Epidermolysis Bullosa (AMPAPEB), and the questionnaire was sent to 53 members by mail.Group 2 was selected from the Facebook™ page of the association, and the questionnaire was administered over this social network to 798 members, including parents, health professionals, family members, and friends of EB patients.
After being accepted as a friend in the association's Facebook™ page, the researcher invited all first-degree relatives from group 2 members to participate in the study, to a maximum of two caregivers per individual with EB.The inform consent form and the questionnaire were sent through Google Drive™ software, which were filled out by an access link.Previously, a pilot study with a virtual questionnaire was performed among people not involved in this research.The pilot test was done with 5 persons (10% of the expected total sample) and included a professor and students of Dentistry.

Data collection
SF-36 quality of life questionnaire data was stored in a database.The feedback for both groups was expected to be received no more than 150 days from the start date of the study.After 30 days, the same material (letter, 2 consent forms, and 2 questionnaires) was resent to non-responders in the same way as the first delivery: by mail for group 1 and through Facebook™ and Google Drive for group 2. Additionally, the researchers had access to the date and time of the response through the Google Drive™ software in group 2.

Statistical analysis
Descriptive statistical analysis was performed with EPI INFO 8™ (Center for Disease Control and Prevention-CDC, USA).The return of questionnaires was evaluated for groups 1 and 2 according to age, sex, and time of return.Data were analyzed using the Chi-squared and Fisher's exact test.The level of significance of the analyses was set at 5% 3 .

RESULTS
After 30 days, five envelopes with seven questionnaires were returned from group 1 and eight were filled out by group 2. After 150 days, 10 more questionnaires were returned by mail, and five filled out over the social network.
Thus, after 150 days from the study start date, 30 questionnaires were returned, 17 (60.0%)from group 1 (from 53 members, 32.1% return rate) and 12 (40.0%)from group 2 (from 798 members, 1.50% return rate).During the first 30 days, group 2 had a rapid return, although it did not persist until the end of the evaluation (Figure 1).

Discussion
The Internet presents an opportunity for innovative recruitment modalities [1][2][7][8]17 . Social etworks were developed to bring people together and increase the sharing of information.Today, Facebook™ is one of the most used social networks and has been considered a viable recruitment option for health behavior assessment in young adults 18 .The use of a social network in human research has been reported; however, dental research has not yet been conducted using this recruitment method 1,3,5 .
A systematic review found that using mailed questionnaires for health research can improve the quality of the data 19 .According to Leece et al. 20 and Kongsved et al. 21, the prevalence of returned questionnaires via internet is significantly lower than via mail.However, the study of Frandsen et al. 22 and Skonnord et al. 8 demonstrated the opposite.The study conducted by Mazzon et al. 16 showed a return of 15.1% of questionnaires by mail, while our study observed a return of 32.1%, a high rate compared to the return by social network of only 1.50%.It is important to emphasize, however, that  many persons in the social network were friends, aunts, uncles or health professionals that did not meet the inclusion criteria.Thus, we considered that the evaluation of the method using return rate of data collected through a social network is questionable.
The social network as a means of data collection allows easy access to the caregivers of EB individuals because most Brazilians have internet access.However, the researcher needs to constantly monitor data collection since many people might not be familiar with this type of approach.Thus, new strategies should be employed in future studies to increase response rates and quality of research such as reminder system by telephone and softwares 2,8 .
The literature demonstrates that most of the questionnaires sent by mail are returned within the first two weeks of research 2,3,[19][20][21][22][23][24][25] , which is corroborated by the current study.Moreover, a higher prevalence of return was observed through the social network in the first 30 days; after 150 days, however, the mailed questionnaires had a higher return rate, resulting in an overall higher rate (18 -60.0%).This finding suggests that social networks are important tools for questionnaire-based research when data can be collected within a short timeframe [23][24][25] ; for research in general, the most efficient way to send questionnaire and receive responses is by mail.
Women were more prevalent (76.7%) in both groups, and more men in the mail group returned the questionnaire (71.4%).These findings may be because women use social networks more frequently.
The average age of participants who returned the questionnaire was 39.2 years; those who responded over the social network were younger (43.75 years) than those who responded by mail (52.85 years).These findings show that older people tend to be more interested in participating in independent research through forms sent by mail.
Research using questionnaires is necessary and practiced widely [2][3][4][5][6][7][8][23][24][25] . It is orth emphasizing that the internet, in addition to its technological benefits, has been asserted as a powerful research mechanism 2,22 .Therefore, it is worthwhile to consider the expansion of social networks and their rapidly increasing number of users as a research method through sending and receiving questionnaires via the internet.These approaches may also be understood as important tools for conducting questionnaire-based-research, especially when a quick data collection is needed, to assess large samples, investigate rare cases, or decrease problems related to long distances from research centers and target population [22][23][24][25] .According to Fenner et al. 17 , the success of this method has worldwide implications for future research.It is worth emphasizing that our goal was to discuss the use of a social network to determine the response rate of a quality of life questionnaire sent to caregivers of individuals with EB; in this respect, social networks have become an important means of gaining access to these individuals.
The recruitment of young people for health research by traditional methods has become more expensive and challenging over recent decades 17 and social networks have the potential to overcome this disadvantage, especially in terms of having a fast turnover, lower operating costs, and broad coverage.However, over a long term, mailed material still shows greater effectiveness of return.
Although we found that data collection by mail was more effective, we believe that social networks can help researchers to recruit more people and become an important tool for conducting research using questionnaires, mainly when data need to be collected within a short timeframe, people living far from the research need to be reached, and especially for investigations involving rare diseases.

Figure 1 .
Figure 1.Return rate of questionnaires during the initial 30 days (A) and after 150 days of the start of the research (B).

Figure 2 .
Figure 2. Relative distribution of questionnaires according to sex and age during the initial 30 days and after 150 days from the start date of the research.