Global Public Interest and Seasonal Variations in Alzheimer's Disease: Evidence From Google Trends

Background: As the world's population ages, Alzheimer's disease (AD), a common neurodegenerative disease, is a major challenge to human health in the future. Understanding the information needs on AD of the global public can contribute to the prevention and control of AD. The purpose of this study was to explore global public interest and seasonal variations in AD using Google Trends (GT).Methods: GT was used to obtain relative search volume (RSV) of the keyword “Alzheimer's disease” in six English-speaking countries (Australia, New Zealand, the USA, the UK, Canada, and Ireland) and the world from January 2004 to December 2020. Cosinor analysis was applied to detect the seasonality of AD-related RSV. Time series plot was used to observe the trend of annual mean AD-related RSV. Globally, hot topics and top rising topics related to AD were also analyzed. In addition, we also explored the geographical distribution characteristics of AD-related RSV.Results: AD-related RSV declined steadily from January 2004 to December 2013 and rose steadily from January 2014 to December 2020. Search popularity of AD is low in the southern hemisphere, compared to the northern hemisphere. Cosinor analysis showed that there were significant seasonal variations in AD-related RSV in six English-speaking countries (all P < 0.05). Interestingly, regardless of the hemisphere, peaks were observed in the winter months and trough in the summer months. Topics related to the characteristics and etiology of AD, early onset AD, AD-related associations, care of AD patients, and diseases that can easily be confused with AD had received special attention.Conclusions: There is increasing global public interest for AD and a significant seasonal variation in AD. A better understanding of the seasonal variations and public interest of AD by governments, health workers and patients can contribute to the prevention, management, and treatment of AD.

Global Public Interests and Dynamic Trends in Osteoporosis From 2004 to 2019: Infodemiology Study

Background With the prolonging of human life expectancy and subsequent population aging, osteoporosis (OP) has become an important public health issue. Objective This study aimed to understand the global public search interests and dynamic trends in “osteoporosis” using the data derived from Google Trends. Methods An online search was performed using the term “osteoporosis” in Google Trends from January 1, 2004, to December 31, 2019, under the category “Health.” Cosinor analysis was used to test the seasonality of relative search volume (RSV) for “osteoporosis.” An analysis was conducted to investigate the public search topic rising in RSV for “osteoporosis.” Results There was a descending trend of global RSV for “osteoporosis” from January 2004 to December 2014, and a slowly increasing trend from January 2015 to December 2019. Cosinor analysis showed significant seasonal variations in global RSV for “osteoporosis” (P=.01), with a peak in March and a trough in September. In addition, similar decreasing trends of RSV for “osteoporosis” were found in Australia, New Zealand, Ireland, and Canada from January 2004 to December 2019. Cosinor test revealed significant seasonal variations in RSV for “osteoporosis” in Australia, New Zealand, Canada, Ireland, UK, and USA (all P<.001). Furthermore, public search rising topics related to “osteoporosis” included denosumab, fracture risk assessment tool, bone density, osteopenia, osteoarthritis, and risk factor. Conclusions Our study provided evidence about the public search interest and dynamic trends in OP using web-based data, which would be helpful for public health and policy making.

Prognostic Value of Circadian Rhythm of Brain Temperature in Traumatic Brain Injury

Hypothermia has been used in postoperative management of traumatic brain injury (TBI); however, the rhythmic variation and prognostic value of brain temperature after TBI have never been studied. This study describes diurnal brain temperature patterns in comatose patients with TBI. Mesors of brain temperature, amplitude, and acrophase were estimated from recorded temperature measurements using cosinor analysis. The association of these patterns with clinical parameters, mortality, and functional outcomes in a 12-month follow-up was examined. According to the cosinor analysis, 59.3% of patients presented with circadian rhythms of brain temperature in the first 72 h postoperatively. The rhythm-adjusted mesor of brain temperature was 37.39 ± 1.21 °C, with a diminished mean amplitude of 0.28 (±0.25) °C; a shift of temperature acrophase was also observed. Multivariate logistic regression analysis revealed that initial Glasgow coma scale score, age, elevated blood glucose level, and circadian rhythm of brain temperature seemed to be predictive and prognostic factors of patients’ functional outcomes. For the prediction of survival status, younger patients or those patients with mesor within the middle 50% of brain temperature were more likely to survive. The analysis of brain temperature rhythms in patients with moderate and severe TBI provided additional predictive information related to mortality and functional outcomes.

Seasonality of Back Pain in Italy: An Infodemiology Study

Background: E-health tools have been used to assess the temporal variations of different health problems. The aim of our infodemiology study was to investigate the seasonal pattern of search volumes for back pain in Italy. Methods: In Italian, back pain is indicated by the medical word “lombalgia”. Using Google Trends, we selected the three search terms related to “lombalgia” with higher relative search volumes (RSV), (namely, “mal di schiena”, “dolore alla schiena” and “dolore lombare”), representing the semantic preferences of users when performing web queries for back pain in Italy. Wikipedia page view statistics were used to identify the number of visits to the page “lombalgia”. Strength and direction of secular trends were assessed using the Mann–Kendall test. Cosinor analysis was used to evaluate the potential seasonality of back pain-related RSV. Results: We found a significant upward secular trend from 2005 to 2020 for search terms “mal di schiena” (τ = 0.734, p < 0.0001), “dolore alla schiena” (τ = 0.713, p < 0.0001) and “dolore lombare” (τ = 0.628, p < 0.0001). Cosinor analysis on Google Trends RSV showed a significant seasonality for the terms “mal di schiena” (pcos < 0.001), “dolore alla schiena” (pcos < 0.0001), “dolore lombare” (pcos < 0.0001) and “lombalgia” (pcos = 0.017). Cosinor analysis performed on views for the page “lombalgia” in Wikipedia confirmed a significant seasonality (pcos < 0.0001). Both analyses demonstrated a peak of interest in winter months and decrease in spring/summer. Conclusions: Our infodemiology approach revealed significant seasonal fluctuations in search queries for back pain in Italy, with peaking volumes during the coldest months of the year.

A SAS Macro for Cosinor Analysis of Automated Heart Rate Data in the ICU

Cosinor analysis, developed by Franz Hallberg and colleagues in the 1960s, allows for the fitting of a cosine curve to data of known period and is frequently used in the analysis of circadian data. While software packages exist which will perform cosinor analyses, SAS has historically steered users to a reference (Ferrarra, 1993) that contains a flaw that results in miscalculation of the acrophase (peak) and nadir (minimum) values about half the time. The R package ‘cosinor’ includes the same error as the SAS reference. The macro we have written 1.) outputs a dataset with the correct values of acrophase, mesor, amplitude, nadir p-value for rhythmicity and R-squared for a curve fit to an individual’s data and 2.) plots the resulting curve and underlying data. While the macro does not allow for the simultaneous analysis of data from multiple individuals it will output the data from multiple curves so that these parameters can be used in downstream data analysis. We demonstrate the use of this macro on a dataset of automated heart rate data collected on patients in the intensive care unit. Given the growing use of automated data collection, it is vitally important that software that provides accurate statistical analysis of circadian phenomena be made widely available to clinical investigators.

Global Public Interests and Dynamic Trends in Osteoporosis From 2004 to 2019: Infodemiology Study (Preprint)

BACKGROUND With the prolonging of human life expectancy and subsequent population aging, osteoporosis (OP) has become an important public health issue. OBJECTIVE This study aimed to understand the global public search interests and dynamic trends in “osteoporosis” using the data derived from Google Trends. METHODS An online search was performed using the term “osteoporosis” in Google Trends from January 1, 2004, to December 31, 2019, under the category “Health.” Cosinor analysis was used to test the seasonality of relative search volume (RSV) for “osteoporosis.” An analysis was conducted to investigate the public search topic rising in RSV for “osteoporosis.” RESULTS There was a descending trend of global RSV for “osteoporosis” from January 2004 to December 2014, and a slowly increasing trend from January 2015 to December 2019. Cosinor analysis showed significant seasonal variations in global RSV for “osteoporosis” (<i>P</i>=.01), with a peak in March and a trough in September. In addition, similar decreasing trends of RSV for “osteoporosis” were found in Australia, New Zealand, Ireland, and Canada from January 2004 to December 2019. Cosinor test revealed significant seasonal variations in RSV for “osteoporosis” in Australia, New Zealand, Canada, Ireland, UK, and USA (all <i>P</i>&lt;.001). Furthermore, public search rising topics related to “osteoporosis” included denosumab, fracture risk assessment tool, bone density, osteopenia, osteoarthritis, and risk factor. CONCLUSIONS Our study provided evidence about the public search interest and dynamic trends in OP using web-based data, which would be helpful for public health and policy making.