The best magazine
Methods
We conducted a cross-sectional questionnaire study of patients visiting a GP. Data collection took place between March and June 2013 in waiting rooms of GP practices. Data was collected using a convenience sample in three different regions in Germany: Lower Saxony (urban and rural GP practices), Thuringia (urban and rural GP practices), and Baden-Wurttemberg (urban and rural GP practices). Participation in the survey was entirely voluntary, and by filling in the questionnaire, patients agreed with participation in the study. The following ethics committees stated that ethics approval was not necessary for this anonymous questionnaire study: the Medical Ethics Committee of the University Hospital of Heidelberg, and the Ethics Committee of the Medical Association of Lower Saxony and the Ethics Committee of the Medical Association of Thuringia (personal communication).
The Questionnaire
No validated questionnaires in this context were available. Consequently, a standardised questionnaire was developed based on available literature and available questionnaires about knowledge and use of complementary and alternative medicine, for example, a study among elder patients in rehabilitation centres and a study about patients using both complementary and conventional medicine and their information about their practitioners' qualification.
The developed questionnaire consisted of four parts. The first part contains a list of 49 home remedies adapted from the textbook "Home remedies in modern medicine". The predefined list of home remedies contains five groups: teas, foodstuffs, wet packs, baths and skin applications. Survey participants indicated whether each of these home remedies was known or not known, or already used successfully or not successfully.
The second part of the questionnaire includes items about the use of home remedies in general. For example, whether home remedies are used regularly, why they are used, and in which cases, and reporting of "subjective" effects. Also, the questionnaire asks how patients act when concerned about a common health complaint and from where their knowledge about home remedies comes. All items in this section were measured on 5-point-Likert scales from "never/not at all" to "very often/frequently".
The third part of the questionnaire contains open questions about frequent health problems as defined by the German federal government's health monitoring statistics. Patients were requested to write down which home remedy they would use to treat symptoms of a cold, diarrhoea, constipation, back pain, headache, sleeping disorder/nervousness, wound healing and cystitis.
The last part of the questionnaire contains socio-demographic questions. The questionnaire was initially piloted with 10 volunteers. Subsequently, in a focus group with 12 doctoral students, the questionnaire was also piloted and then immediately afterwards discussed in a cognitive debriefing process; after which modifications to improve understanding and clarity were made. The revised version was piloted with a group of 10 patients recruited in the waiting room of a GP practice. No modifications were found to be necessary and the revised version was implemented for this study.
Participants and Recruitment
Patients had to be 18 years or older to be included in the survey. GP practices from three regions in Germany were targeted. Region selection was convenience based with a location in the north, the south and the east chosen: the cities of Heidelberg (in southern Germany), Erfurt (in eastern Germany) and Hannover (in northern Germany) and surrounding semi-rural areas. Within these regions, GP practices were randomly selected by their addresses. In all regions, patients from GP practices in both urban and rural areas were recruited with a final sample size of 400 patients in 40 practices (10 patients per practice). Addresses were obtained from the websites of the regional associations of statutory health insurance physicians (kassenärztliche Vereinigung). 100 GPs in each region were contacted by mail; this was followed by a reminder telephone call if there was no reply. In cases of a positive answer, dates for data collection were arranged. Information sheets and questionnaires were made available to patients in the waiting room of participating GP practices, and they were asked if they were willing to participate in the study by the first author during morning and afternoon consulting hours. On rare occasions, when a patient was not able to read or write, they were assisted in filling in the questionnaire. Completed questionnaires were collected in a clearly marked box left in the waiting room.
Data Analysis
Questionnaires were reviewed for completeness. When more than a third of the questions were not answered, including the socio-demographic data, questionnaires were excluded from analysis.
A descriptive analysis was performed with all other questionnaires, even if some data was incomplete. On top of variables related to the survey questions, an additional variable was created for analysing, per respondent, the entire number of named home remedies that were mentioned in the questionnaire. For this purpose, we tallied up home remedies from our list provided in the questionnaire as well as those named by participants in answers to open-ended questions. Home remedies were only counted once. The overall number of home remedies is reported as mean and standard deviation. A rating list was created for the 15 most frequently used home remedies from the predefined list in the questionnaire. Participant answers from open-ended questions were also classed into our pre-defined major categories i.e. teas, wet packs, baths, foodstuffs, skin applications. The use of home remedies by participants, based on our provided list of home remedies in the first part of the questionnaire, is presented as frequencies. Single items of the questionnaire regarding use of home remedies are presented as frequencies and percentage in case of pre-defined major categories. ANOVA was used to detect differences in the amount of home remedies used between sex, age, regions and rural or urban area. All answers to items measured in 5-point-Likert scales are presented in tables.
For statistical calculations, SPSS 21.0 and Excel 7 were used.
Source: ...