Trend Pattern Analysis
Table 1 summarizes the AAPC of the terms “finasteride,” “Propecia,” “finasteride side effects,” “Propecia side effects,” and “post-finasteride syndrome” compared across the US, UK, and AUS. The mean ARSV for “finasteride” was 15% in 2004 and increased significantly to 57% in 2020 (APC: +9.25%, 95% CI 8 to 10.5, p<0.001). Join-point regression APC curves of “finasteride” and “finasteride side effects” are demonstrated in Figure 1A-B.
The “finasteride side effects” ARSV in the US presented initially a statistically significant increase in interest between 2004 and 2010 (APC: +48.1, 95% CI 30.6 to 68, p<0.001), a subsequent non-statistical decrease between 2010 and 2013 (APC -17, 95% CI 060.6 to 75.1, p=0.586), and a significant positive increase in interest between 2013 and 2020 (APC: +18.9, 95% CI 7.6 to 31.3, p=0.004). This led to an overall significant overall constant increase in trend between 2004 and 2020 with a cumulative APC of +20.7 (p<0.001). This significant increase in trend for “finasteride side effects” was consistent across the UK and AUS between 2004 and 2020, APC +25.7 (p<0.001) and APC +20.1, respectively. Specifically, in the UK, the “finasteride side effects” ARSV presented initially a steep statistically significant increase between 2004 and 2009 (APC: +78.4, 95% CI 47 to 116.6, p<0.001), and a subsequent linear increase between 2009 and 2020 (APC +7.3, p=0.023). Furthermore, the “finasteride side effects” ARSV in AUS had a statistically significant increase between 2004 and 2009 (APC +82.8, 95% CI 37.6 to 142.8, p<0.001) and a non-significant decrease in ARSV between 2009 and 2020 (APC -0.8, 95% CI -8.9 to 8.1, p=0.851). Similarly, searches for the term “post-finasteride syndrome” significantly increased overall between 2004 and 2020 in the USA (APC +29.2, 95% CI 13.4 to 47.3, p<0.001).
There was an initial non-significant increase in “post-finasteride syndrome” between 2004 and 2009 (APC +2.9, 95% CI -12.6 to 21.1, p=0.7), a steep growth between 2009 and 2012 (APC +151.7, 95% CI 21.6 to 421.3, p=0.018), and a linear growth between 2012 and 2020 (APC +16, 95% CI 7.2 to 25.6, p=0.002).
On the contrary, we observed a significant decrease in “Propecia” interest over time (APC: -9.8, 95% CI -10.8 to -9.8, p<0.001). The interest in “Propecia side effects” significantly decreased from 2004 to 2020 (APC: -4.2, 95% CI -14.1 to 6.9, p<0.001) in the USA although this was not consistent in the UK (APC: -0.5, p=1) or AUS (APC: +10.3, p<0.001). In the US, the term “Propecia side effects” demonstrated three distinct points of change in trends. Between 2004 and 2012, we observed an initial non-significant increase (APC +4.7, 95% CI -2.1 to 11.8, p=0.155), a non-significant decrease between 2012 and 2015 (APC -27.1, 95% CI -60.3 to 33.9, p=0.27), and a non-significant decrease between 2015 and 2020 (APC -2, 95% CI -14.5 to 12.3, p=0.743). Interestingly, “Propecia side effects” ARSV significantly increased between 2004 and 2020 (APC +10.3, 95% CI -7.5 to 31.6, p<0.001). Between 2004 and 2006, there was an initial steep non-significant increase in interest (APC +231.2, 95% CI -25.5 to 1372.1, p=0.106) and a significant slight decrease between 2006 and 2020 (APC -5.7, 95% CI -12.1 to 1.1, p=0.091).
Impact of FDA communications on finasteride side effects treatment trends
When trends were compared before and after FDA communications in 2011(12), the mean (SD) ASRV of “finasteride”, “finasteride side effects”, and “post-finasteride syndrome” increased from 20.4 (± 5.8) to 41.5 (±10.9) (p=0.001), from 17.6 (±2.3) to 40.2 (±19.0) (p=0.014), and from 2.0 (±5.9) to 36.1 (± 13.7) (p<0.001), respectively (Figure 2).