Elsevier

Sleep Medicine

Volume 37, September 2017, Pages 119-123
Sleep Medicine

Brief Communication
Summertime blues? A re-examination of the seasonality of web searches for restless legs and leg cramps

https://doi.org/10.1016/j.sleep.2017.06.017Get rights and content

Highlights

  • There is a summer peak in Internet searches for restless legs and leg cramps.

  • Google Trends search data for these terms in the United Kingdom and Australia were examined.

  • Results were broken into trend, seasonal and noise components using an econometric software.

  • There was a significant stable seasonal effect for all searches.

  • A combined seasonality test was positive only for leg cramps in both countries.

Abstract

Objective

Studies that examined Internet searches for restless legs and leg cramps have found a strong seasonal effect with peaks in summer and troughs in winter months. The present study used an econometric approach to examine the seasonality of such searches in greater detail.

Methods

Monthly relative search volumes for ‘restless legs’ and ‘leg cramps’ from 2004 to March 2017 in the United Kingdom (UK) and Australia were obtained from Google Trends. Average percentage change from winter to summer months was examined. The TRAMO-SEATS procedure from the DEMETRA statistical software was used to decompose the data into trend, seasonal and noise components and to determine whether a combined seasonality test was positive.

Results

There were substantial percentage increases in Google Trends searches between winter and summer months regarding restless legs in the UK (median increase 46%) and Australia (33%) and regarding leg cramps in the UK (95%) and Australia (50%). However, the combined seasonality test was positive only for leg cramps and not for restless legs in both countries: although there was significant stable seasonality in restless legs searches, this was outweighed by substantial moving seasonality and noise components.

Conclusions

Examination of average percentage increase in search volume from winter to summer exaggerates the degree of seasonality. Seasonal effects for restless legs searches are non-significant when the trend and noise components of the data are considered, although this does not exclude a clinical significance for the identified stable seasonality. Significant seasonality, with a summer peak, is present for leg cramps searches and suggests an increase in the incidence or severity of leg cramps in summer.

Introduction

A substantial proportion of all Internet searches is for health-related information. Google's search engine (http://www.google.com) is the most popular worldwide, and Google Trends (www.google.com/trends), a freely accessible online tool, has been used increasingly in recent years in epidemiological research to study the effects of location and time, including seasonality, on search patterns as a potential guide to the occurrence or severity of a disease [1], [2], [3].

Although it is well recognised that some patients with Willis–Ekbom disease (restless legs syndrome [RLS]) find that heat, such as the summer months or a warm bed environment, exacerbates their symptoms [4], [5], [6], worsening of symptoms in winter has also been reported [7]. Recently, the first systematic investigation of this topic used Google Trends to examine the seasonal effect on searches for ‘restless legs’ [8]. It was found that relative search volume (RSV) increased by 24–40% during summer relative to winter months in both the Northern hemisphere (primarily the United States) and in Australia in the Southern hemisphere (where the seasons are reversed).

The effect of temperature on the occurrence of nocturnal leg cramps has also been uncertain [9], [10]. A recent study of Google searches for ‘leg cramps’ in the United States and Australia again reported a substantial peak in RSVs in the summer months and peak-to-trough differences in Internet search volume ‘were equivalent to 73.1% of the mean cramp-related search volume’ [11].

The magnitude of the seasonal effects reported for restless legs and leg cramps searches in these studies, which potentially reflect variations in the severity or occurrence of these conditions, is somewhat surprising, as marked seasonal effects on these conditions had not been previously reported. Although the studies examining search trends for restless legs [8] and leg cramps [11] examined seasonality differently, the magnitude of the effect was mainly based on comparing observed peak-to-trough differences in search volumes. This may not fully consider the effects of random variation and longer-term trends.

In this study, I used an econometric statistical analysis to examine in greater detail the seasonal effect on Google searches for ‘restless legs’ and ‘leg cramps’ in the United Kingdom (UK) and Australia.

Section snippets

Google Trends search

Monthly search results for the exact terms ‘restless legs’ and ‘leg cramps’ between January 2004 and March 2017 were examined on 27 April 2017. Search results in Google Trends are proportionate to the total searches at a particular location and time [3]. The resulting numbers are then ‘normalised’: individual values over time are calculated by dividing each time-point value by the highest value and multiplying by 100 to provide an RSV of entered queries on a scale of 0–100. Absolute search

Results

There were substantial increases in Google Trends searches regarding restless legs and leg cramps between winter and summer months in both the UK and Australia, albeit with large annual variations and some reversals (Table 1). In the UK, there were positive correlations between RSVs for restless legs (Pearson correlation coefficient 0.19) and leg cramps (0.25) and mean monthly British temperatures.

Analysis of the seasonal component after decomposition of the total search volumes confirmed that

Discussion

Time series analysis of economic data is commonly used to break down the series into trend, seasonal and noise components to identify significant seasonality [17]. Recently, studies have applied this approach for examining the seasonality of health problems [18], [19], and in the present study, it was used to examine the seasonal effect on Google searches for ‘restless legs’ and ‘leg cramps’. These precise terms were used as they allow a more direct comparison of the present results with

References (20)

There are more references available in the full text version of this article.

Cited by (11)

  • Seasonality of restless legs syndrome: symptom variability in winter and summer times

    2020, Sleep Medicine
    Citation Excerpt :

    Only one previous study tried to examine this phenomenon by analyzing Internet search query data about RLS symptomatology, and documented a seasonal trend of the search queries, with a peak in the summer months [12]. However, the previously observed seasonality of web searches for RLS symptoms and leg cramps has been re-examined later on, and that author did not find the same seasonal effect for RLS searches in Internet data, although he did not exclude a clinical significance of the previously identified seasonality of RLS symptoms [15]. In the present study, we confirmed the evidence of seasonality in RLS, showing that patients affected by RLS presented higher IRLS scores during summer compared to winter.

  • Misbeliefs and Biases in Health-Related Searches

    2021, International Conference on Information and Knowledge Management, Proceedings
  • Seasonality of back pain in italy: An infodemiology study

    2021, International Journal of Environmental Research and Public Health
View all citing articles on Scopus
View full text