Self-Care for Common Colds by Primary Care Patients: A European Multicenter Survey on the Prevalence and Patterns of Practices—The COCO Study

Background. Patients use self-care to relieve symptoms of common colds, yet little is known about the prevalence and patterns across Europe. Methods/Design. In a cross-sectional study 27 primary care practices from 14 countries distributed 120 questionnaires to consecutive patients (≥18 years, any reason for consultation). A 27-item questionnaire asked for patients' self-care for their last common cold. Results. 3,074 patients from 27 European sites participated. Their mean age was 46.7 years, and 62.5% were females. 99% of the participants used ≥1 self-care practice. In total, 527 different practices were reported; the age-standardized mean was 11.5 (±SD 6.0) per participant. The most frequent self-care categories were foodstuffs (95%), extras at home (81%), preparations for intestinal absorption (81%), and intranasal applications (53%). Patterns were similar across all sites, while the number of practices varied between and within countries. The most frequent single practices were water (43%), honey (42%), paracetamol (38%), oranges/orange juice (38%), and staying in bed (38%). Participants used 9 times more nonpharmaceutical items than pharmaceutical items. The majority (69%) combined self-care with and without proof of evidence, while ≤1% used only evidence-based items. Discussion. This first cross-national study on self-care for common colds showed a similar pattern across sites but quantitative differences.


Introduction
Common colds are the most frequently encountered human diseases worldwide [1]. Common cold is a conventional term used for mild upper respiratory illnesses, which comprises a heterogeneous group of self-limited diseases caused by numerous viruses. The frequency is age-specific with fewer episodes in adults than in younger children [2,3] and a seasonal variation with more episodes in winter and fall [1]. Beyond impairing the quality of life [4], common colds have a tremendous economic burden on societies due to work absenteeism (70 million workdays annually due to noninfluenza-related viral respiratory tract infections in the US, which corresponds to US$8 billion indirect costs) [5][6][7]. If consulted, general practitioners are asked to provide supportive advice, symptomatic treatment, and reassurance [8].
Studies on self-care have shown that common colds are the most frequent cause for self-care [9][10][11][12]. This is reasonable, as common colds have a self-limited course and resolve without treatment. Unsystematic practice observations and the few studies available indicate that patients use a mixture of evidence-based and non-evidence-based self-care practices including self-medication and traditional home remedies to reduce symptoms and improve subjective well-being (e.g., [9,[12][13][14]). Nevertheless, only few self-care practices have been studied and proven to be effective in relieving symptoms of common colds in adults, such as nonsteroidal anti-inflammatory agents [15], antihistamines [16], and oral antihistamine-decongestant-analgesic combinations [17].
Although self-care for minor illnesses is recognized as an important component of primary health care, little is known about the prevalence and patterns of self-care for common colds across Europe. Drawing on data from 27 primary care sites of 14 countries, the cross-sectional COCO study describes the variety and pattern of reported self-care practices.

Methods and Materials
2.1. Study Design. This multicenter, cross-sectional survey was performed at 27 primary sites in 14 European countries. Details of the study design have been published earlier [18]. Briefly, the cross-national survey was designed and conducted by the Working Group on Self-Care of the European General Practice Research Network (EGPRN). The data collection was completed in April, 2014.

Study Instrument and Data Collection.
A 27-item selfadministered patient questionnaire provided 94 self-care practices and free text options in 11 categories: over-thecounter medication, specific foods or drinks, herbal teas, alcoholic drinks, special self-prepared recipes, pastilles or drops, something for the nose, inhalation, throat gargle/spray, items applied externally, and extras at home. In addition, the following patient characteristics were requested: age, gender, health insurance status, number of school years, pills taken daily, and expenditures for the last common cold. To assure random samples at all sites, the questionnaires were distributed to consecutive patients independent of their reason for physician consultation. Inclusion criteria were the following: being ≥18 years, being able to understand the questionnaire, or being in attendance of someone providing assistance.

Patient Involvement.
Primary care physicians who developed and conducted the study had a dual role as professionals and as patients who experienced common colds themselves. Results of the COCO study will be disseminated in form of patient information materials.

Sample Size Calculation.
To obtain a representative sample for a practice of up to 10,000 patients, a total of 94 patients had to be studied (CI 95%; SE 0.05; accuracy 0.1). This estimate was based on the assumption that the sample of patients surveyed is random for the respective practice. To account for nonresponders, oversampling by 25% was planned, which led to a sample size of 117.5 (94 + 23.5 patients), rounded to 120 patients per primary care setting. This sample size also yielded representative results for items with a prevalence in the range of 40% to 60% in the general population. For details see Weltermann et al. [18].
2.5. Statistical Analysis. Self-care items were classified by 2 researchers in terms of their mode of application: "intestinal absorption," "intranasal application," "local oral effects," "inhalation," "topical use in throat," "external use," "foodstuffs," and "extras at home." In a subclassification, single items were further subclassified using the ATC (WHO Anatomical Therapeutic Chemical) classification system for pharmacological substances and-if not listed-plausible groups and subcategories. Items were listed individually if they had a utilization rate of at least 1% ( ≥ 27) prior to age standardization; otherwise, they were summarized as "other" within each mode.
Descriptive analyses of all self-care practices were performed for the total study population and stratified by country. To account for differences in age distributions between European countries and practice samples, results were agestandardized using the 2013 European Standard Population [19]. Data were analyzed on the basis of five aspects: (1) "Number of self-care practices used": (a) mean item use; (b) the average number of items per 100 participants stratified by country; and (c) prevalence of using any self-care were considered.
(2) "Mode of application": (a) distribution of all single items reported per mode of application; (b) % of participants using at least one item per mode; and (c) mean item use per mode (subclassification) were considered.
(3) "Single items used" (subclassification): (a) the most frequently used single items; (b) the sum of the five most frequently used items per country; (c) frequently used items (≥5% utilization rate); and (d) rarely used items (1% to 5% utilization rate) were considered (selected items used by <1% of the participants were reported).
(5) "Analysis by the countries' purchasing power standard (PPS)": based on the PPS, countries were classified into three groups. The PPS is an artificial value which is based on each nation's Gross Domestic Product and eliminates price level differences among countries [32]: PPS I included Austria, Sweden, Germany, and Finland; PPS II included France, Israel, Italy, Spain, and Slovenia; PPS III included Poland, Turkey, and Macedonia. Analyses included (a) the mean item use, (b) the use of evidence-based self-care (at least 1, mean), (c) the use of pharmaceutical products (at least 1, mean), and (d) the mean expenditure in the 3 PPS groups.
The analysis included all returned questionnaires. Statistical analyses were performed using IBM5 SPSS5 Statistics, version 20 and SAS version 9.4 statistical package (SAS Institute, Inc., Cary, NC, USA).

Characteristics of the Participants.
A total of 27 sites from 14 European countries participated in the survey. 3,074 questionnaires were returned. After excluding data from sites  0 20 40 60 80 100 Sweden (1 site) Figure 1: Pattern of self-care by mode of application stratified by country: percentage (%) of using at least 1 item within each mode (weighted by age).

Number of Practices Used.
In total, 527 items were reported. Of these, 483 items were reported as free text options. The age-standardized total mean item utilization was 11.5 (±SD 6.0, range: 0-53). In total, 99% of participants used any kind of self-care practices. Comparative analyses for open and closed answers showed mean item utilization of 1.6 (±SD 1.0) and 11.0 (±SD 6.6, range: 0-53), respectively.
Comparisons between countries showed a variation by factor 6.4 in the spectrum size for the stated number of different items per 100 participants (35/100 in Macedonia compared to 224/100 in Turkey). We largely abstained from comparisons between countries, because within-country comparisons showed differences of up to 38% in modes of applications and 4.9% in item utilization (see Additional File 1 in Supplementary Material).

Mode of Application Level.
Of all 527 items used, "foodstuffs" was reported most frequently (52%). With 276 items, this is also the mode with the largest number of reported items. Much fewer items were reported for "intestinal absorption" (9%, 48 items), "inhalation" (8%, 42 items), "extras at home" (7%, 37 items), "intranasal application" (7%, 36 items), "local oral effects" (6%, 32 items), and "topical use in throat" and "external use" (both 5%, 28 items). Details on the percentages for the utilization of at least one item per mode of application and the mean item utilization for each mode stratified by country are displayed in Figures 1 and 2.

Self-Care Practices Used by at Least 1% of Participants.
After grouping single items based on (a) content (related practices) and (b) utilization rate (≥1%), the spectrum of self-care comprised 97 items. After age-standardization, the most frequently reported single items were water (43%), honey (42%), paracetamol (38%), oranges/orange juice (38%), staying in bed (38%), bath/shower (35%), rest at home (34%), lemons/lemon juice (31%), and chicken soup (30%). Water and/or honey were among the three most frequently used items in 9 countries, oranges/orange juice, staying in bed, and/or pain medication in 5 countries, and rest at home and/or chicken soup in 4 countries. In the total sample, ≥2 of the five most frequent items (water, honey, paracetamol, oranges/orange juice, and staying in bed) were used by 63% of the participants, and 34% used at least 3 of these items. The five most frequently used self-care practices of each country added up to 19 different self-care practices across all countries: bath/shower, rest at home, staying in bed, chicken soup, lemons/lemon juice, oranges/orange juice, mandarins, other fruits incl. "more fruit," lime blossom tea, mixed herbal tea, honey, hot milk with honey, water, ibuprofen, paracetamol, vitamin C, nasal decongestants, saline for nose, and pastilles for cough.
An overview on the frequently used self-care practices, that is, items with a utilization rate of at least 5%, is provided in Table 2; items rarely used (1% to 5%) are provided in Table 3. When taking all items into account, fluids of any type were used by 92% of participants, including any tea by 74% and any alcohol-containing liquids by 12%. Fruit and/or juices were used by 70% of participants and vegetables by 24%. Soup was consumed by 31%. With regard to "extras at home," 56% of participants reported preferring to "stay in bed," "sleep," or "rest at home" and 57% used something related to "warmth." For "intestinal absorption" it is noteworthy that 60% of participants used at least one pain medication. Moreover, although the questionnaire item vitamin C was chosen by  28%, when also considering fruit-based vitamin C (citrus fruits and juices) it is 68%. At least one nose drop/spray was reported by 36% of participants, at least one flavored lozenge by 33%, at least one gargle by 17%, and at least one rub by 10%.

Evidence-Based Products Proven to Be Somewhat Effective.
In total, 21 self-care items were grouped as evidence-based. See Table 4 for an overview. The mean of evidence-based products was 1.1 items (±SD 0.97) and ranged between 0 and 5 items. Of the 2,691 participants using self-care, solely evidence-based self-care was used by 0.3% of participants, solely non-evidence-based self-care was used by 31.2%, and a combination of both was used by 68.5%. In participants who combine both, participants used 8.0 times more nonevidence-based items than evidence-based items. An additional 35 items were grouped as non-evidencebased pharmacological products which were used by 4.6% ( = 124) of the participants. Any pharmaceutical productsirrespective of their evidence-based status-were used by 62% of the participants. Participants who combined pharmaceutical and nonpharmaceutical products used 9.0 times more nonpharmaceutical products than pharmaceutical products.

Analysis by the Countries' Purchasing Power Standard.
The number of items differed between PPS groups: with a mean of 12.4 (±SD 7.2) it was highest in the lowest PPS group (III) and equal-although lower-in PPS II (10.6, ±SD 5.8) and PPS III (10.5, ±SD 6.2). Conversely, the use of evidence-based products decreased with increasing PPS (PPS I: 77%, PPS II: 71%, and PPS III: 58%), as did the use of pharmaceutical products (PPS I: 69%, PPS II: 61%, and PPS III: 54%). The mean items used for evidence-based products and pharmaceutical products differed marginally (with increasing PPS: evidence-based products: 1.2, 1.1, and 0.9; pharmaceutical products: 1.0, 0.8, and 0.8).

Discussion
This is the first study on self-care for common cold with data from several European countries. According to the COCO data, European primary care patients can consistently be considered as high users of self-care for common colds. Prevalence of application modes showed similar patterns across countries, while the number of items on country level varied markedly: the modes "foodstuffs," "intestinal absorption," and "extras at home" were the most frequently used modes in 10 countries, with "foodstuffs" representing the most frequently used one across all countries. Of the large spectrum of 527 practices reported, 19 self-care practices rank among the top five across countries.
Evidence-Based Complementary and Alternative Medicine 7

Strengths and Limitations of the Study.
This is the first study to analyze prevalence and patterns of self-care for common colds on a larger European scale. Contrary to previous studies it used a rather explorative approach and did not restrict self-care to the use of OTC or home remedies. Age standardization was used to account for differences in ages between sites and countries. This study was not without limitations. Firstly, due to the sampling strategy, results are representative only for primary care practices/sites. Therefore, we largely abstained from country comparisons. Secondly, we did not ask for the participants' reason for using certain self-care practices (e.g., reduction of duration, symptomatic relief, and targeted symptoms) and how they appraise the evidence. Thirdly, we did not ask whether patients' current physician consultation was due to an acute cold. Fourthly, coding unspecific answers might have led to an underrepresentation of pharmaceutical products. Women were overrepresented in our sample, likely due to generally more physician visits by women and the feminization of the elderly population. In addition, participants with more school years are overrepresented as compared to the EU population.

Comparison with Other
Studies. The observed prevalence of self-care utilization close to 100% is higher than reported in other studies. For instance, the prevalence was only 84% in the British general population [14] and 69% in Estonian pharmacy customers [33]. In addition, we showed that patients engage in a variety of self-care practices, as the spectrum of reported self-care items among the sites in the 12 countries comprised more than 500 items, on average 12 items per person. Few studies are available for comparison. Irrespective  [9]. Specifically regarding colds, the Estonian survey showed average utilization of 4.1 medicinal plants alone [33], and in an American survey among parents ( = 153) with a focus on children 3.8 (European Americans) and 3.2 (African Americans) items were reported on average [34]. Turning to self-medication, 62% used pharmaceutical products, which is higher than the 52% reported in the British survey ( = 4,327) [13] and the 44% reported in Estonian pharmacy customers ( = 300) who used a combination of herbal products/medicinal plants and self-medication [33]. Based on a review of the literature, we identified 21 self-care items in the COCO data that are discussed as possibly being beneficial for cold symptoms in adults; the majority of items lack an assessment for effectiveness or were shown to be ineffective. Of the five most frequently reported items (utilization of about 40% each), that is, "water," "honey," "paracetamol," "oranges/orange juice," and "staying in bed," only paracetamol is evidence-based. Overall with two thirds, the majority of participants engaged in self-care that includes both evidence-based items and items without proof of evidence.
The most frequently reported evidence-based items were nasal decongestants followed by the various pain medications (NSAIDs and paracetamol). Examples for non-evidencebased self-care are that one third of respondents still use vitamin C and another 15% use garlic, which can both be considered treatment myths [22,26]. Similarly, the use of inhalations by about one third is not supported by evidence [28]; the same applies for chest rubs [31]. With 12%, also the non-evidence-based use of alcohol-either drunk alone or added to other liquids-is surprisingly high.

Conclusions
This study adds to the knowledge on the prevalence and patterns of cold self-care and clearly demonstrates that selfcare for cold is frequent, even among those with relatively low utilization. As such, the data are representative on the level of the participating primary care practices. Overall, the documented commitment for self-care is in line with European wide attempts to promote self-care for minor illnesses in order to reduce health care system expenditures and the workload of general practitioners. Nevertheless, given the poor evidence base, the high engagement in self-care highlights the need for patient-directed information. Such information should address the evidence base and safety of self-care practices. Furthermore, studies that evaluate the effectiveness of the frequently used, poorly studied items for common colds are needed.

Additional Points
Trial Status. Participants' recruitment was finished in April 2014.