Dry skin

Web searches for 'dry skin' reflect regional changes in weather, demographics, and housing properties

Thanks to data available from Google Trends, we are able to show the relative importance of Internet searches for the term 'dry skin' across all of the 50 states. The map below ranks the relative importance of 'dry skin' searches. For example, the state of New York is often the top ranked state for these searches because it has the highest proportion of Internet searches devoted to the search term 'dry skin'.

The key factors influencing the seasonal and geographic changes in people's interest in searching for 'dry skin' on the Internet include

  • Changes in weather

  • Age distribution of a state's population, and

  • Building properties.


In a previous blog we showed the Internet search activity for 'dry skin' across the United States for the past few years. The variations in search activity mainly represented seasonal changes in weather patterns, with the winter months exhibiting the highest searches because the levels of water vapor in outdoor air (and indoor air) are typically the lowest at this time of the year. Nevertheless, the aggregation of search activity across all states masks variations between the states. 

Age of Population

For example, as people age, their susceptibility to dry skin formation increases, and thus states with the highest proportion of residents over the age of 65 years would be expected to have higher search activity regarding 'dry skin' than states with younger populations--other factors being equal. The 5 states with the highest percentages of citizens over 65 years are

  • Florida (19%),

  • Maine (18%),

  • West Virginia (18%),

  • Vermont (17%), and

  • Pennsylvania (17%)

and the bottom 5 are

  • Alaska (9%),

  • Utah (10%),

  • Texas (11%),

  • Georgia (12%), and

  • Colorado (13%).

Even though Florida's population has the largest fraction of seniors, search activity for 'dry skin' is rarely elevated because of the state's humid climate, which reduces the magnitude of indoor drying stresses on skin. In contrast, Colorado has a very dry atmosphere, but its population is among the youngest in the US and thus the search activity for 'dry skin' is depressed.

Housing Properties

As the average age of housing in a state increases, the air exchange rate with outdoor air also tends to increase (due to air leakage across the building shell), which means that indoor humidities are more sensitive to changes in outdoor water vapor levels. The states with the largest fractions of housing built prior to 1970 are

  • New York (69%),

  • Rhode Island (63%),

  • Massachusetts (62%),

  • Pennsylvania (59%), and

  • Connecticut (58%).

The state of New York not only has old housing stock, but also has a large inventory of apartment buildings in New York City that tend to have elevated drying stresses on skin in the winter months due to low humidity levels. Pennsylvania is notable because it is in the top five category for both the age of its residents and housing. These combined factors place the residents of this state at increased risk of exhibiting dry skin and hence a greater tendency to search the Internet for this condition. 

Internet search activity for "dry skin" during the month of October, 2017. Results are normalized to the highest search popularity.

Internet search activity for "dry skin" during the month of October, 2017. Results are normalized to the highest search popularity.

Wrap Ups

  • Seasonal trends in 'dry skin' Internet searches vary across the US as function of weather conditions, age distributions among the states, and housing properties.

  • If you do a lot of business or recreational travel, be aware that where you start and end your travel will impact the drying stresses on your skin!

  • Spending the winter in Florida will result in lower drying stresses on your skin--compared to staying in the colder parts of the Midwest and Northeast.

The Water Vapor Content of Outdoor Air Really Matters to Your Skin!

We know that water vapor in outdoor air continually enters a residence and changes the level of indoor relative humidity.

Consequently, there should be a direct connection between local weather conditions, for example how dry or humid the air is, and people’s tendency to have dry skin caused by exposures to indoor humidity and temperature stressors. Moreover, it’s probably safe to assume that as dry skin forms due to such drying stresses, people will be more inclined to conduct Internet searches to learn about their dry skin and how to treat it. To examine the relationship between outdoor water vapor levels and Internet searches referring to dry skin (as a proxy for actual dry skin formation), I collected concurrent data on “dry skin” web searches (from Google Trends) and ambient water vapor levels for two major cities that reflect different demographics, building properties, and climates:  Chicago, USA, and Tokyo, Japan.

“Dry Skin” Internet Searches versus Water Vapor Levels in Outdoor Air

Let’s begin by looking at the relevant data for Tokyo, as shown on the adjacent chart. The y-axis represents the frequency of web searches for the term “dry skin” (translated to Kanji characters). Search results are normalized so that the highest response rate is assigned a value of 100. The x-axis shows the monthly values of absolute humidity (i.e., the concentration of water vapor in air, expressed in g/m3). One interesting feature of the plot is the divergent responses that occur at an absolute humidity of around 5 g/m3. To simulate this response pattern, I created a statistical model that categorized water vapor data into either [1] a special cold season, defined by the months of November through January, or [2] all other months of the year. The resulting two-season model predicts “dry skin” searches (depicted by green squares) from absolute humidity and captures the Internet search responses occurring at low absolute humidities.

Predicted and observed changes in Internet searches for “dry skin” among residents of Tokyo, Japan .

Predicted and observed changes in Internet searches for “dry skin” among residents of Tokyo, Japan.

I used the same analytical approach with the Chicago data and the results for the observed and predicted search term activity for “dry skin” also showed that a two-season model can be used to represent Internet search patterns for “dry skin”.

Predicted and observed changes in Internet searches for “dry skin” among residents of Chicago, USA.

Predicted and observed changes in Internet searches for “dry skin” among residents of Chicago, USA.

What the results mean

The elevated web search results during the “cold season” correspond to the heating seasons in both Chicago and Tokyo, which means that indoor temperatures are reduced compared to the rest of the year. In addition, the absolute humidities of outdoor air during the winter months are typically the lowest of the year in temperate climates because the colder air holds less water vapor. Cooler indoor temperatures along with lower humidities in turn produce higher values of the Dry Skin Index since it increases as relative humidity and temperature decrease. Therefore, we would indeed expect elevated Internet search activity regarding “dry skin” as a result of the higher drying stresses during those months.

Internet search activity for the term “dry skin” among the residents of both cities gradually diminishes as outdoor water vapor levels rise—even to levels as high as 20 g/m3. This suggests that dry skin care for these populations extends year round, but is less urgent during the summer months when humidities are highest.

Despite the similarities between the Chicago and Tokyo data, the unique combinations of climate, building properties, and demographics for these cities make it difficult to generalize to other locations. For example, wood-frame residences absorb/desorb water vapor differently than buildings of masonry construction and older residential populations are more susceptible to drying stressors on skin. Consequently, variations in these and other parameters will directly influence the nature and magnitude of drying stressors on skin. The best solution to the challenge of monitoring drying stresses unique to your location is to track the Dry Skin Index with a sensor that can be placed in your home or carried with you in a purse/backpack or simply placed next to your moisturizing lotions.

Wrap Up

  • Internet search activity on Google Trends for the term “dry skin” among residents of Chicago and Tokyo vary in response to changes in outdoor water vapor levels

  • The cold season months of November through January produce the highest search activity due in part to low indoor humidities and temperatures that exert higher drying stresses on skin.

  • Dry skin care in both cities is essentially a year-round activity, but is less important during the warmer, more humid months of the year.