R and Python
There are times when we not only look at the descriptive analysis but also want to make future predictions based on the past trends. We will look at techniques that we can use to predict the number of studies submitted or registered in future years.
We will see how we can use some of the libraries like pandas, statsmodels and matplotlib in python.
the python code is available on my github repository here:
https://github.com/kalehdoo/clintrials/blob/master/ctrials_1.py
I will also try to explain the steps and procedures to perform the analysis later.
Here is the final outcome:
Model Summary:
Studies submitted predicted:
For 2019 : 31,022
For 2020 : 32,479
The data used for the regression is from 2005-2018.
There is still some time left for 2019 to complete so I will come back next year to compare the 2019 actuals with the predicted numbers here. The actuals for 2019 is 19,990 (data until Aug 2019) which was posted in one of the previous posts here.
I have also shown how to do regression using R programming, and also how to interpret the results. The link below has complete code and the analytics:
http://rpubs.com/kalehdoo/sponsor_analytics
Source data extracted from: https://aact.ctti-clinicaltrials.org
Study Activity Dashboard
In this post, we will try to understand the clinical study activities across the globe. We will gather some inputs like population, GDP and health spending as % of GDP. Then we compare different countries by their involvement in clinical studies. The clinical study activity is based on the clinical site in that country for a particular study. It is important to keep in mind that those clinical sites or facilities may or may not have enrolled any participants. Also, the demographics data is for the year 2017 and we are considering all the clinical studies registered in the USA clinicaltrials.gov as of Aug 2019. Keeping all that in mind, we will try to get a sense of overall study activity and compare them for different countries. We will also look at the study activity from region level. So, just sit back and relax.
 |
| Figure 1.1 |
Figure 1.1 shows study activity per 100 K population of a country. Denmark tops the list with the highest number of studies per 100 K population. This is not the complete list and I have tried to display maximum I could fit in a picture. You would notice that there are countries with very small population and hence they have got a high activity per capita. Also, there are few countries with very large populations and have got a low ratio.
In figure 1.2 above, the countries are categorized under geographic region. It also shows the percentage share of the population and number of studies. The dashboard allows to drill down on a region and see the details by country.
We will look at the study activity based on the GDP and health spending as % of GDP of countries in next post.
Keep thinking till next time.
Sponsor Trends Dashboard
In the last post here, we gained some insights into top performing sponsors and overall trend. In this post, we will look further deep into how the sponsor participation has changed in last few years. The Sponsor Trends Dashboard (figure 6.1) can answer some interesting questions and see what's going on in the clinical trials industry.
 |
| Figure 6.1 |
Here, we will try to analyze the insights gained from Sponsor Trends Dashboard in Figure 6.1 above.
1. What type of study Sponsors have the highest or lowest share in study registration, and how has that changed over the last few years?
The chart in Quadrant 2 tells us about the share of studies by different types of sponsors. Clearly, Universities have the highest share and has maintained a steep rise in the last few years. The share has increased from 40% in 2005 to 50% in last 2 years. These Universities could be public or private, funded or not-funded but we do not have that information available as of now.
Hospitals have also performed well in registering the studies. The share of studies registered by Hospitals have almost doubled since 2005. On the other hand, the share of Industry sponsored studies has declined consistently and it has reduced to just 17% now which was 37% in 2008. The share does not indicate if the segment has really grown or declined. The share of one segment, Universities for example, could rise because other segments have declined. To see the trend, read the next question below.
2. Which type of sponsors have shown growth in study activity?
The chart in Quadrant 1 shows how the registered studies by different categories or types of sponsors have grown over the period.
The studies registered by Universities have grown at a continuous and rapid pace. Until 2008, the Industry segment and Universities were overlapping but after that the Universities completely outpaced the Industry Sponsors. The growth in studies registered by Industry sponsors has remained almost flat but Hospitals maintained a steady growth until 2016 after which the growth is negligible.
3. How has the total numbers of sponsors changed over the past few years?
In the above 2 charts in Quadrants 1 and 2, we looked at the trend of the studies registered by sponsors but now in this chart on quadrant 3, we will look at the trend of actual number of sponsors.
The overall trend shows that the total number of primary sponsors who registered their studies has grown consistently. All 3 important categories, Universities, Industry Sponsors and Hospitals have grown consistently. All 3 types of sponsors have increased almost 2 times of 2008 levels.
4. Has the participation of sponsors from industry has increased?
The chart in quadrant 4 shows the participation trend over the period. Industry participation means that either the study is sponsored by a sponsor from industry or at least one of the collaborators is an industry sponsor.
The chart shows that the industry participation has declined and reduced to just 22% from upper 40s in 2008. Between 2005 and 2008, the share of studies having industry participation remained in upper 40s but started declining consistently after 2008. However, it is not clear from this chart if the studies with industry participation has decreased or the studies without industry participation has increased. To figure that out, let's take a look at the trend chart that shows the number of studies with or without industry participation over the period of time.
 |
| Figure 6.2 |
The trend in chart (figure 6.2) shows that the number of studies where there is industry participation has remained nearly constant but the studies without any industry participation has increased from 6K levels in 2005 to 24K in 2018, which is four times growth.
As a food for thought, how do you think clinical research industry can increase the collaboration for the larger benefit to the whole community?
Do post your questions and comments.
Keep thinking!
Source data extracted from: https://aact.ctti-clinicaltrials.org
