Use of personal care products (PCPs) is an integral part of our daily lives; therefore it is vital to ensure a high level of PCP safety. The various ingredients contained in PCPs have the potential to enter our body via multiple routes, i.e.
, through dermal absorption, ingestion, and inhalation. There is a debate in the literature as to whether once inside the body PCP ingredients can affect human health. Concerns have been raised about the possible endocrine disrupting activity of numerous PCP ingredients detectable in human matrices, including phthalates [1
], parabens [3
], organic ultraviolet (UV) filters [5
], and triclosan [10
]. Furthermore, a variety of nanoparticles, now frequently employed in a wide range of PCPs [11
] have been shown to have a cytotoxic, genotoxic [13
], and neurotoxic potential [17
]. Many common PCP ingredients may also produce cutaneous side-effects, e.g., contact or photocontact allergies [19
New ingredients are continuously introduced to the market, creating a need for comprehensive risk research and thorough exposure assessments. Furthermore, a single ingredient is nowadays often contained in different categories of PCPs that can be concurrently used by the same consumer [22
]. In order to accurately estimate human exposure and consequently assess the health risk of a PCP ingredient, exposures across the different PCP categories must be aggregated. Realistic PCP use and co-use patterns thus represent a key input to aggregate exposure assessment models.
Studies collecting use patterns of various PCP categories have been conducted previously in North American [23
] and European populations [29
]. To date in Europe, however, PCP use patterns among children are very limited and only provided for children aged 12 years and older in the German database VerbraucherAnalyse Jugend (English: Youth Consumer Analysis). Children are believed to be more susceptible to the effects of chemical exposures than adults [34
]. They have immature organ systems, higher metabolic rates, and greater body surface area-to-mass ratios. The impact of small changes caused by chemical exposures early in life may not be immediately apparent but may lead to late-life effects [35
]. Reliable PCP use data are hence urgently needed in Europe for younger children in order to assess their exposure to various PCP ingredients. Diverse socio-economic, cultural, and environmental factors across Europe may also influence the lifestyles of consumers and affect their PCP use patterns. However, at present, we do not know whether there are appreciable differences in PCP use patterns amongst European consumers since detailed and publicly available national-level data are currently unavailable for most European countries.
The present study was conducted as part of a larger research project focused on organic UV filters in PCPs. It aimed to collect the first set of individual PCP use patterns in the German-speaking population of Switzerland that can be used to calculate consumer exposure to PCPs. Eight widely used leave-on PCP categories were selected (face cream, body lotion, aftershave lotion/balm, hand cream, makeup foundation, lip care, lipstick, sunscreen) for their likelihood of containing organic UV filters. These categories can be used on a regular basis and simultaneously, which has to be taken into account when assessing aggregate exposures; therefore, we also investigated their co-use patterns. For the first time in Europe, we included children below the age of 12 into our study population, to start filling the existing information gap on PCP use in European children. In addition, we sought to compare our results with existing PCP use patterns publicly available in Europe, to provide first insights into the inter-country variation in prevalence and frequency of PCP use, and intra-country variations in PCP use over time.
2.1. PCP Use Survey: Settings and Study Design
A postal questionnaire survey (with pre-paid reply envelopes) has been selected as the most feasible and inexpensive methodology for a large-scale study to collect PCP use patterns across all ages as compared to more costly daily diary methodology. The survey comprising open and closed questions was conducted between January and March 2011 to collect PCP use data in the Swiss-German population (approx. 6.5 million people; 66% of the Swiss population [36
]). Hereafter, we will use the term Swiss-German to denote the population and the term German-speaking Switzerland if the region is mentioned. Two questionnaire versions were designed: one for children and adolescents and one for adults (English versions of the original German questionnaires can be found in Supplementary Section S1
). Throughout this work, we will use the term “children and adolescents” for participants aged 0–17 years (in the discussion further split into “children” aged between 0 and 12 years, and “adolescents” aged between 13 and 17 years), and the term “adults” for those aged ≥18 years. In both questionnaire versions, we asked the household member whose birthday was next to fill out the questionnaire. For respondents under 14 years of age, parents were asked to fill out the questionnaires. General reminder letters were sent four weeks after the initial mail-out. Second reminders, including a new copy of the questionnaire, were sent four weeks later to those who did not respond to the general reminder.
In the self-administered questionnaire, participants were asked to recall their use of the eight investigated PCP categories, namely, face cream, body lotion, aftershave lotion/balm (hereafter referred to as aftershave), hand cream, makeup foundation, lip care, lipstick, and sunscreen, over the past year. We were aware that some of the investigated PCP categories, e.g., lipstick, have traditionally only been used from a certain age on and only by one sex (females). Nevertheless, we asked everyone about the use of all PCP categories to avoid gender/age bias and thus reflect more accurately the actual use patterns in modern societies.
The core survey questions focused on the prevalence and frequencies of PCP application. For sunscreens, participants were additionally requested to report the product application frequency to different areas of the body, and season of application. The questionnaire also included questions about the specific products used, including the Sun Protection Factor (SPF) values for sunscreens. Participants were asked to provide basic demographic information such as gender, age, and educational attainment (adults only). Data on physical characteristics (body weight, body height, skin type [37
]) were also collected.
2.2. Study Population
The questionnaire for children and adolescents was mailed to 1,000 eligible families with children and adolescents aged 0–17 years recruited using a commercially available address database (Schober Addresses-Shop [38
]). The adult questionnaire was sent to 2,500 household addresses randomly selected from the Swiss telephone directory. We obtained a response rate of 48.8% for children and adolescents and 36.8% for adults. The response rates were calculated by dividing the number of returned questionnaires by the total number of questionnaires, excluding those sent out to invalid addresses, i.e.
, for example to respondents who died, moved away or could not answer in German. Our response rates are consistent with previous paper-based postal surveys with two reminders and even slightly higher for children and adolescents [39
]. After exclusion of ineligible, incomplete, and contradictory responses, the final dataset included 397 children and adolescents and 799 adults (exclusion criteria are detailed in the Supplementary Section S2
). If a participant chose not to answer specific questions, the responses were coded as missing values.
2.3. Data Analysis
Basic descriptive statistics were used to summarize the demographic and physical characteristics of the study subjects. Prevalence and frequency of PCP use were depicted graphically. Variations in the prevalence of PCP use were analyzed by gender (female, male), age group (four age groups among children and adolescents (0–4, 5–8, 9–12, 13–17); four age groups among adults (18–42, 43–52, 53–65, 66+)), level of education (adults only; primary, secondary school/upper secondary school, university), and skin type (very fair/fair/light brown, medium brown/dark brown/black) using the Chi-squared or Fisher test statistic. The Mann-Whitney U test or the Kruskal–Wallis test were used to assess variations in the frequency of PCP use using the same predictor variables as above. Spearman’s correlation coefficient (R) was used to measure the strength of correlations between frequencies of use of the different PCP categories. Correlations were considered very weak if 0 < R < 0.19, weak if 0.20 < R < 0.39, moderate if 0.40 < R < 0.59, strong if 0.60 < R < 0.79, and very strong if 0.80 < R < 1.00. A p value of <0.05 indicated statistical significance unless noted otherwise.
Data were analyzed using the statistical software package SPSS 19.0 (SPSS Inc., an IBM company, Chicago, IL, USA, 2010). PCP co-use patterns were generated for the eight PCP categories using an in-house generated Python (version 2.5) script. The complete set of data at an individual level and the Python script details are accessible from the authors upon request.
2.4. Inter- and Intra-Country Comparisons of PCP Use Patterns: Data Sources and Treatment
In order to compare the PCP use patterns of consumers in three countries in Western Europe, namely, Switzerland (Swiss-German population), Germany, and the Netherlands, we used survey data on PCP use patterns from two other sources: the German VerbraucherAnalyse (VA) database (more than 31,000 respondents; the second database version for 2011 was used to more closely match the time frame of the other two datasets), and a Dutch dataset (516 respondents) from Biesterbos et al.
]. Several design features of the three datasets required adaptations to ensure a satisfactory level of data comparability. First, the German VA consists of individuals above 12 (VA Jugend, English: Youth) or 14 (VA Klassik, English: Classic) years of age and the Dutch survey consists of individuals above 18 years of age. The VA is not flexible in terms of restricting the dataset to make a population sample of individuals above 18 years; therefore we restricted our Swiss-German population sample to the age range of respondents in the selected version of the VA for the desired PCP category (some PCP categories are not available in the VA Jugend). Also, the Dutch population is restricted by its narrower age-range of respondents (18–70 years). Second, the VA conforms to assumed gender stereotypes (e.g., lipstick use by females only). Therefore, for both the Swiss-German and the Dutch population, we used a reduced dataset that matches the gender of respondents in the VA for a particular PCP category. Third, the VA contains use data for a wide range of facial care product categories for females, e.g., day cream, night cream etc.
As a result, our “face cream” category for females was excluded from the comparison, as we were unable to match it with a single representative PCP category in the VA. Likewise, “face cream” does not have a matching PCP category for males in the Dutch database. Hence for “face cream”, we only provide a comparison of use between the Swiss-German and the German male populations. Finally, the PCP use frequency categories differ between the three datasets, but we easily merged them into common frequency categories.
Adequate data needed to assess intra-country variations in use of the investigated PCP categories over time are currently lacking for Switzerland. A single study conducted by Berret and colleagues in 2000/2001 is available on the use of sunscreen in families living in Switzerland (N = 1,239; age 0–72 years) [42
]. Berret et al.
] focused on sunscreen use during summer holidays, therefore, we compare their findings to our results for sunscreen use (whole-body application) in the summer/autumn season, as we do not have data for summer holidays alone. The focus of the comparison was given to: overall prevalence of sunscreen use, number of daily sunscreen applications, and SPF values of the sunscreen products used.
We provide the first assessment of use patterns for eight widely used leave-on PCP categories across all age groups in the Swiss-German population. Our findings build upon and extend the body of current knowledge on PCP use patterns worldwide. Furthermore, data we collected on PCP use amongst young children are an essential prerequisite for exposure assessments of substances contained in PCPs, thus far lacking in the published literature for Europe.
In our female population, 8.8% of female children and adolescents and 15.3% of adult females applied aftershave products in the past year. Although typically perceived as PCPs for males, aftershave products for females exist, in particular those designed for the sensitive bikini area. Our results indicate that aftershave products have become increasingly popular amongst females. Such a finding is not surprising since removal of unwanted body hair is normative in contemporary Western cultures [44
]. Furthermore, amongst adolescent boys aged 13–17 years, 8.4% used lipstick in the past year. Nowadays, also men use decorative cosmetics, traditionally used by women, to improve their looks. Biesterbos et al.
have recently reported that 0.5% of adult Dutch males used make-up products (lipstick or lip gloss, makeup foundation, make-up remover, and powder or rouge) [29
]. Hence, gender-based stereotyping concerning PCP use becomes less pronounced. Similarly, age-based stereotyping in PCP use tends to diminish. In our study in fact, 56.1% of female adolescent participants aged 13–17 years indicated that they used makeup foundation, previously worn mainly by adult women. Moreover, 7.2% of the youngest children aged 0–4 years used lipstick (see Figure 2
). In the US, Wu et al.
showed that a high percentage of females (45%) age 5 or less used nail polish (not investigated in our study) [28
]. On the other hand, our female participants in the oldest age group aged 66 years and older had the highest prevalence of lipstick use (72.6% respondents). As the population ages, older adults also want to retain good appearance in the youth-oriented society and therefore continue using decorative cosmetics until very old age. In order to obtain a realistic representation of people’s actual PCP use patterns that mirrors societal change, it is desirable that future PCP use surveys avoid gender and age product stereotyping.
Knowledge of PCP co-use and the associations of the frequency of use between the individual PCP categories are vital pieces of information for evaluating aggregate exposures of PCP ingredients. Concurrent use of the eight leave-on PCP categories examined in this study was very common, albeit varied between children and adolescents and adults as well as between genders. The overall prevalence rate of concurrent PCP use was higher in children and adolescents than in adults. About 80% of the children and adolescent population (79.8% females; 78.4% males) used at least two PCP categories concurrently on a regular basis in the past year. However, despite the lower prevalence of concurrent use amongst adults (64.8% females; 71.2% males), their co-use patterns were much more complex with a considerably higher number of different PCP combinations (particularly among males) compared to children and adolescents. The substantially greater inter-individual variability in PCP co-use means that it is more difficult to provide an accurate characterization of the aggregate ingredient intake. This finding emphasizes the need for more research focusing on detailed co-use of multiple PCP categories, especially amongst adults (stratified by age and gender and ideally other socio-demographic factors), in order to help to identify the potentially highest exposed subpopulations of consumers. The use frequencies of face cream and body lotion were moderately correlated, the only case that occurred in all age/gender subgroups. The correlation of use frequencies of skincare PCP categories appears to have a similar pattern worldwide: Wu et al.
found moderately positive correlations between the use frequencies of face moisturizer with body lotion in the US [28
]. Similarly, Biesterbos et al.
found moderately positive correlations between the use frequencies of day/night cream with body lotion in the Netherlands [29
]. As highlighted by Wu et al.
ingredients in topical skin care preparations, which are intended to stay on the skin for prolonged periods of time, can be similar, which may be reflected in individual’s aggregate exposure [28
Among the three age groups of children, i.e.
, 0–4 years, 5–8 years, and 9–12 years, the prevalence of use of face cream and body lotion for both genders was highest (>79% of respondents) in the youngest group of children who may be more vulnerable to the potential adverse effects of some PCP ingredients. The specific PCPs that were used on children aged 0–4 years were almost exclusively products labeled for use by infants and children. Parents may assume that infant/children PCPs are rather safe to use; nevertheless, it has been shown previously that even these PCPs may contain potentially harmful ingredients, e.g., fragrances and preservatives [46
], known to cause allergies and/or are suspected of causing endocrine disruption. Future research should include thorough exploration of use patterns of other PCP categories frequently used by young consumers, including rinse-off products, to accurately assess their aggregate exposure to PCP ingredients.
PCP use patterns might be influenced by regional variations in environmental and socio-economic conditions and also cultural behaviors. Ours is the first comparative analysis of PCP use patterns among three Western European countries: German-speaking Switzerland, Germany, and the Netherlands. Overall, the prevalence and the frequency of PCP use were especially high in German-speaking Switzerland. One possible explanation of the higher use of skin care PCPs, particularly the considerably higher use of sunscreen and lip care, is that many Swiss spend a significant part of their leisure-time outdoors all year round. Switzerland is a largely mountainous country and recreational activities such as hiking, skiing, and rock climbing, are very popular. UV radiation levels are high in mountain areas and reflection from snow further increases exposure to UV radiation. Thus sun protection is highly needed and sunglasses and sunscreen use is encouraged [48
]. Additionally, it has been shown previously that in high-altitude ski areas individuals who follow sunscreen advice carefully are also more likely to practice other sun-protection practices, such as the use of UV filter containing lip balm [49
]. Also, according to 2010 L’Oreal’s annual report, Switzerland has one of the highest per capita cosmetics consumption in the world, which they assume reflects the high income levels of the country—once basic human needs are satisfied, people are able to spend on luxuries [50
]. The higher income levels in Switzerland, compared to Germany and the Netherlands, are likely to enable the Swiss to spend more money on manufactured goods, which in turn may increase their PCP consumption. We acknowledge that our inter-country comparison of PCP use is partially compromised by the differences in survey design of the three studies compared (i.e.
, different categories of frequency of PCP use, target populations in terms of age and size, and data collection methodologies). Despite these limitations, we are confident that this comparison is meaningful, as it demonstrates, for the first time, that there are differences in PCP use between countries within Europe and that data available from other countries must be used with caution.
Besides inter-country differences, intra-country variations in the extent and patterns of PCP use might also take place over time within a country. Due to lack of available datasets on PCP use in Switzerland, we were only able to carry out a limited comparison of sunscreen use based on data from Berret et al.
]. It is important to mention here that the annual all-sky sunshine duration (SD) for 2000 was comparable to that of 2010 and in both years the annual all-sky SD was slightly lower than the annual all-sky SD mean [51
]. Overall, it seems that the prevalence of sunscreen use in 2000 was very similar to that of 2010. However, compared to a decade ago, our respondents in 2010 used sunscreen products with higher SPF values for more efficient UV protection. At the same time, they appear to assume that fewer applications are adequate for all day exposure. Our results support those of Autier et al.
, who showed that use of products with high SPF values gives consumers a false sense of security and often leads, counterproductively, to prolonged sun exposure [52
]. We have to mention, however, that the decrease in the number of daily applications in our study might have been somewhat influenced by our longer study period (summer and autumn compared to summer only in Berret et al.
]). In the autumn, whole-body sunscreen application is likely to be lower than in the summer. Furthermore, Berret et al.
also included respondents from the French- and Italian-speaking parts of Switzerland [42
There are several limitations of the present study that warrant consideration. For children under 14 years of age we relied on one parent/guardian to report the PCP patterns of their child, therefore we might have missed PCP usage unknown to that person. Furthermore, respondents were asked retrospective questions between January and March. Therefore, we assume some level of recall error, especially, in the case of the sunscreen frequency of use during the past year. Also, unlike for PCPs used regularly throughout the year, for the sunscreen frequency of use an open-ended questioning format was used to account for the important change in patterns of sunscreen use between seasons. However, open-ended questions were difficult for respondents to answer and we indeed had to exclude 29 respondents who reported an unrealistically high number of days during which they had used sunscreen (details in the Supplementary Section S2
). Hence we further assume some level of error in consumers’ responses because of the questioning format used. On the other hand, we are aware that people who have spent their summer or winter holidays abroad in sunnier regions than Switzerland might have applied sunscreen very frequently. It should also be mentioned that we used open-ended questions to gather information on specific PCPs used by the survey respondents in order to obtain as much details about the products as possible. However, the gathered dataset was extremely complex as the degree of detail given by respondents varied widely. Responses included, e.g., brand names without any further description (e.g., Labello), exact product names (e.g., Labello hydro care), different names describing the same product (e.g., Labello pearl & shine; Labello Perlglanz), and descriptive labels that could be aligned with several products of the same brand (e.g., Labello pure & natural could refer to Labello pure & natural-milk & honey or olive & lemon etc.
) or even with different brands (e.g., Aloe Vera). Labello, in particular, is also often used as a synonym for lip care in Switzerland and other German-speaking countries. Identification of the individual products to determine the percentages of respondents using the same product is very subjective; therefore we chose not to discuss the PCP brand names that we derived from the questionnaire answers. One also has to bear in mind that the amount of PCP applied on the skin per application influences individual’s exposure levels to ingredients contained in PCPs. However, we have not measured the amount applied for our population sample. In addition, leave-on PCPs investigated can be washed off and the subsequent exposure depends on the time before the next washing. Finally, we cannot rule out that the respondents of our questionnaire might be people who use fewer PCPs/use PCPs less frequently than the average German-Swiss population, as they might be more health-conscious consumers willing to spend the time filling out the questionnaire, unlike their less health-conscious counterparts using more PCPs/PCPs more frequently.