1
00:00:00,000 --> 00:00:15,600
Hello and welcome back to season three of What Do Scientists Do?
2
00:00:15,600 --> 00:00:19,480
This is the show where we talk to expert guests and learn about what science they do, how
3
00:00:19,480 --> 00:00:21,660
they do it, and why they do it.
4
00:00:21,660 --> 00:00:25,600
My name is Mab and today I got to talk to Colin Vibert.
5
00:00:25,600 --> 00:00:30,840
Colin teaches us all about a field called biostatistics and how we can use math to make health care
6
00:00:30,840 --> 00:00:31,840
better.
7
00:00:31,840 --> 00:00:37,280
But what are statistics and how does Chat GPT help our computer programs?
8
00:00:37,280 --> 00:00:40,760
And what does all of this have to do with why writing skills matter?
9
00:00:40,760 --> 00:00:42,360
We're doing math after all.
10
00:00:42,360 --> 00:00:45,240
Well, we answer these questions and more.
11
00:00:45,240 --> 00:00:48,200
So let's jump in and find out.
12
00:00:48,200 --> 00:00:50,840
Hello and welcome back to What Do Scientists Do?
13
00:00:50,840 --> 00:00:54,880
This is a show where we talk to wickedly smart guests and learn all about what they do and
14
00:00:54,880 --> 00:00:57,240
what science has got to do with it.
15
00:00:57,240 --> 00:01:00,140
Today we're going to talk about a field called biostatistics.
16
00:01:00,140 --> 00:01:04,880
So biostatistics as a word might seem complicated, but we can actually break it down into smaller
17
00:01:04,880 --> 00:01:07,680
parts and we get to recognizable bits.
18
00:01:07,680 --> 00:01:11,400
So you probably recognize the word bio or the study of living things.
19
00:01:11,400 --> 00:01:13,280
And then we also have statistics.
20
00:01:13,280 --> 00:01:16,800
So statistics goes hand in hand with probability usually.
21
00:01:16,800 --> 00:01:19,720
And this is something you might have learned about in school.
22
00:01:19,720 --> 00:01:24,560
So statistics is all about collecting and looking at data about certain things.
23
00:01:24,560 --> 00:01:28,560
So if we put this whole word pizza together into a whole pizza again, we actually have
24
00:01:28,560 --> 00:01:34,520
the study of data and using data to figure out how living things work.
25
00:01:34,520 --> 00:01:39,240
So that's just kind of the Mab analysis, Mab synthesis of what this all means.
26
00:01:39,240 --> 00:01:43,220
So luckily we actually have an expert here to talk to us about what that actually means.
27
00:01:43,220 --> 00:01:45,000
So here's Colin.
28
00:01:45,000 --> 00:01:47,480
So hi Colin, thank you so much for joining us today.
29
00:01:47,480 --> 00:01:50,560
Do you want to introduce yourself with your name and pronouns?
30
00:01:50,560 --> 00:01:51,560
Yeah.
31
00:01:51,560 --> 00:01:56,560
My name is Colin Vibert and my pronouns are he, him and his.
32
00:01:56,560 --> 00:01:58,400
And what kind of scientist are you?
33
00:01:58,400 --> 00:02:01,560
I am a biostatistician.
34
00:02:01,560 --> 00:02:05,940
So I think your description was very accurate.
35
00:02:05,940 --> 00:02:15,640
And I work with in the life sciences, pharmaceutical industry, and also have done some work in other
36
00:02:15,640 --> 00:02:18,000
areas in healthcare.
37
00:02:18,000 --> 00:02:23,240
Yeah, so something that I kept kind of running into when I was looking at biostatistics
38
00:02:23,240 --> 00:02:27,000
was how much it has to do with healthcare.
39
00:02:27,000 --> 00:02:29,960
So what do you do in healthcare?
40
00:02:29,960 --> 00:02:32,640
What does a biostatistician do in healthcare?
41
00:02:32,640 --> 00:02:40,000
A biostatistician in healthcare will often be working either at a hospital or a pharmaceutical
42
00:02:40,000 --> 00:02:51,920
company or an agency and they will be helping for a disease area or for a medical product
43
00:02:51,920 --> 00:02:59,200
will be looking to understand the product better or understand the disease better.
44
00:02:59,200 --> 00:03:02,240
And we do that with numbers, essentially.
45
00:03:02,240 --> 00:03:08,360
Yeah, so numbers, we get all of this data, maybe hospitals are constantly collecting
46
00:03:08,360 --> 00:03:13,680
data about how many patients there are or maybe what kind of diseases or illnesses are
47
00:03:13,680 --> 00:03:15,600
bringing them into the hospital.
48
00:03:15,600 --> 00:03:21,240
And then you take all that data and you work it into knowledge or understanding?
49
00:03:21,240 --> 00:03:22,880
Yeah, exactly.
50
00:03:22,880 --> 00:03:33,640
So in Canada and in most countries, but in Canada, we have a lot of science that gets
51
00:03:33,640 --> 00:03:41,520
published and we also have agencies that require for pharmaceutical products output to be recommended
52
00:03:41,520 --> 00:03:42,520
and recorded.
53
00:03:42,520 --> 00:03:49,320
So when a product is being developed, we need to be very open and transparent about how
54
00:03:49,320 --> 00:03:53,040
well it works and what are the possible safety risks.
55
00:03:53,040 --> 00:03:56,400
And we use data for understanding that better.
56
00:03:56,400 --> 00:04:00,160
This all kind of feeds into the scientific method as well.
57
00:04:00,160 --> 00:04:06,080
So you maybe have a hypothesis, so you have a question and then you maybe have an experiment.
58
00:04:06,080 --> 00:04:10,560
And then after your experiment is done, you have all of this data or the results.
59
00:04:10,560 --> 00:04:12,480
And then from there, you just have raw numbers.
60
00:04:12,480 --> 00:04:16,200
So you need to know what to do with the actual numbers, right?
61
00:04:16,200 --> 00:04:17,200
Absolutely.
62
00:04:17,200 --> 00:04:22,880
Yeah, with statistics, having a plan and following through with the plan is very important.
63
00:04:22,880 --> 00:04:23,920
Yeah, exactly.
64
00:04:23,920 --> 00:04:28,480
You want to be able to have numbers that you can kind of work with.
65
00:04:28,480 --> 00:04:36,440
So do you specifically work with like collecting data about pharmaceuticals or do you work
66
00:04:36,440 --> 00:04:37,440
in a hospital?
67
00:04:37,440 --> 00:04:44,480
Yeah, so I work right now for a company that is doing work with pharmaceutical companies.
68
00:04:44,480 --> 00:04:50,040
And so we're focused on the performance of different pharmaceutical products and medical
69
00:04:50,040 --> 00:04:55,320
devices on how they perform for a specific disease.
70
00:04:55,320 --> 00:04:57,040
Okay, that's really cool.
71
00:04:57,040 --> 00:05:03,200
So if you're comparing treatments, for example, maybe if you're comparing different cancer
72
00:05:03,200 --> 00:05:07,760
treatments, if one treatment is more efficient than the other.
73
00:05:07,760 --> 00:05:08,760
Exactly.
74
00:05:08,760 --> 00:05:09,760
Awesome.
75
00:05:09,760 --> 00:05:13,660
So this might be kind of a tricky question here.
76
00:05:13,660 --> 00:05:19,880
But if you had to summarize what you do to like, if Colin had a billboard across Canada,
77
00:05:19,880 --> 00:05:23,480
what would you put on your billboard to summarize what you do?
78
00:05:23,480 --> 00:05:25,040
Well, that's a good question.
79
00:05:25,040 --> 00:05:28,720
I would put I would say planning.
80
00:05:28,720 --> 00:05:36,920
I plan a lot and I do computer coding a lot.
81
00:05:36,920 --> 00:05:47,280
And I am thinking and reading and doing research for understanding data methods and data analysis
82
00:05:47,280 --> 00:05:48,560
better.
83
00:05:48,560 --> 00:05:53,400
So when you talk about planning, what kind of things do you have to plan for?
84
00:05:53,400 --> 00:05:58,720
Yeah, so whenever you want to do just like you said with the scientific method, whenever
85
00:05:58,720 --> 00:06:05,880
you want to do a study before you do the study, you have to have a good understanding of what
86
00:06:05,880 --> 00:06:09,400
you want to do and what data you want to collect.
87
00:06:09,400 --> 00:06:15,080
That means you need to have a plan written out of what you're going to actually end up
88
00:06:15,080 --> 00:06:18,960
doing before you do it.
89
00:06:18,960 --> 00:06:27,760
And so what that means is, let's say you want to study how well a medical device is performing
90
00:06:27,760 --> 00:06:34,040
for a disease, you need to first specify, well, what are we going to measure when we
91
00:06:34,040 --> 00:06:37,620
say that we want to see if the device is performing well?
92
00:06:37,620 --> 00:06:39,680
What score do we use to measure that?
93
00:06:39,680 --> 00:06:46,080
So we would plan by thinking about what score is important, what scores are important, additional
94
00:06:46,080 --> 00:06:52,760
data, demographic data, so data about like the age of the patients, the race of the patients,
95
00:06:52,760 --> 00:06:58,080
the location, all of that is also important to figure out how we're going to characterize
96
00:06:58,080 --> 00:07:00,020
our population.
97
00:07:00,020 --> 00:07:07,640
Because with health studies, it's very important that you are collecting accurate populations
98
00:07:07,640 --> 00:07:12,580
that represent the patients and the disease correctly.
99
00:07:12,580 --> 00:07:16,560
If you don't, then the conclusions you arrive at will be biased.
100
00:07:16,560 --> 00:07:21,160
So you want to make sure that you're collecting enough data that represent what's actually
101
00:07:21,160 --> 00:07:22,720
going on.
102
00:07:22,720 --> 00:07:29,000
So one of the things about statistics, Colin, you could correct me if I'm wrong, but I'm,
103
00:07:29,000 --> 00:07:32,600
I don't do this for a living and it's been a while since I took a stats course.
104
00:07:32,600 --> 00:07:39,040
But one thing about stats is that stats aims to collect small representations, comparatively
105
00:07:39,040 --> 00:07:45,120
small representations about a wider population because it's just not possible to ask every
106
00:07:45,120 --> 00:07:47,840
single person every single question.
107
00:07:47,840 --> 00:07:54,280
So what statistics tries to do is ask a couple people, a diverse and representative group
108
00:07:54,280 --> 00:08:01,800
of people, a couple things to then get an idea or try to predict a larger population.
109
00:08:01,800 --> 00:08:06,840
So it sounds like in health sciences as well, you want to make sure that you're predicting
110
00:08:06,840 --> 00:08:10,280
accurately because these are people's health and wellness on the line.
111
00:08:10,280 --> 00:08:14,480
You want to make sure that you're doing a good job to represent everybody, which I think
112
00:08:14,480 --> 00:08:18,320
is really cool and something that I never really thought about, I guess.
113
00:08:18,320 --> 00:08:21,920
Yeah, that's, that's exactly right on.
114
00:08:21,920 --> 00:08:28,920
You want, you know, if you have, let's say, let's say the medical device could be given
115
00:08:28,920 --> 00:08:34,020
to 10,000 people in North America.
116
00:08:34,020 --> 00:08:41,220
You would, you would not really be able to go out and collect from all 10,000 people.
117
00:08:41,220 --> 00:08:51,320
So you use statistics to try to understand how this device performs with a much smaller
118
00:08:51,320 --> 00:08:53,480
sample, say 100.
119
00:08:53,480 --> 00:08:58,000
And I guess another thing that's kind of sticking out to me is that you work in health care.
120
00:08:58,000 --> 00:09:02,040
And at least when I was younger, I thought working in health care meant being a doctor
121
00:09:02,040 --> 00:09:05,080
or being a nurse.
122
00:09:05,080 --> 00:09:10,320
And it's cool to see, I mean, that your job is just one example of working in health care
123
00:09:10,320 --> 00:09:15,520
and the health care industry that has an impact on people's health and wellness without being,
124
00:09:15,520 --> 00:09:19,280
you know, a doctor or a nurse, but still being kind of health oriented.
125
00:09:19,280 --> 00:09:23,120
There's a lot of people working behind the scenes on health, which is something I think
126
00:09:23,120 --> 00:09:25,640
I always forget about and maybe other people do.
127
00:09:25,640 --> 00:09:26,640
Absolutely.
128
00:09:26,640 --> 00:09:32,800
I would say from my experience so far, I don't have experience working as a health care worker,
129
00:09:32,800 --> 00:09:39,880
as say a frontline health care worker, like a nurse, a doctor, a physician's assistant,
130
00:09:39,880 --> 00:09:41,880
but it is very important.
131
00:09:41,880 --> 00:09:50,160
I found and I believe this, that monitoring disease and monitoring the performance of
132
00:09:50,160 --> 00:09:57,120
the products that are used to treat disease are very, very frequently and continuously done
133
00:09:57,120 --> 00:10:03,880
so that you have a good understanding of how well the treatments are and how well and how
134
00:10:03,880 --> 00:10:05,000
much better you can be.
135
00:10:05,000 --> 00:10:09,080
I think that's really an important part of health care is that we're always thinking
136
00:10:09,080 --> 00:10:13,240
of developing better treatments to help patients more.
137
00:10:13,240 --> 00:10:17,320
And so that's what the research avenue brings.
138
00:10:17,320 --> 00:10:22,880
Whereas if you're a frontline worker, you're doing your very best to treat with the means
139
00:10:22,880 --> 00:10:24,240
you have available.
140
00:10:24,240 --> 00:10:31,240
So your job is kind of looking in the future and trying to come up with ways to make health
141
00:10:31,240 --> 00:10:33,560
care in the future better as well.
142
00:10:33,560 --> 00:10:34,560
Exactly.
143
00:10:34,560 --> 00:10:35,560
Yeah.
144
00:10:35,560 --> 00:10:36,560
Cool.
145
00:10:36,560 --> 00:10:40,640
So another question is you mentioned computer programming and I was just curious about how
146
00:10:40,640 --> 00:10:46,520
much your job is sitting in front of a database and doing the nitty gritty computer programming
147
00:10:46,520 --> 00:10:50,240
and like what computer programming and coding has to do with your job?
148
00:10:50,240 --> 00:10:59,680
Yeah, I would say so I as a biostatistician, I do a lot of work in the R programming language.
149
00:10:59,680 --> 00:11:09,320
There are also other languages like SAS and SPSS that can do computer programming to some
150
00:11:09,320 --> 00:11:13,200
extent, but R is a very popular one right now.
151
00:11:13,200 --> 00:11:20,120
And I would say probably 50 to 60 percent of my day is spent on a given day doing some
152
00:11:20,120 --> 00:11:23,400
coding in some form.
153
00:11:23,400 --> 00:11:29,320
But I am someone who is because I have a math and statistics background from university,
154
00:11:29,320 --> 00:11:33,360
I'm kind of known within my company as being someone who is a math and stats and coding
155
00:11:33,360 --> 00:11:34,360
person.
156
00:11:34,360 --> 00:11:36,020
So they go to me.
157
00:11:36,020 --> 00:11:41,980
Other biostatisticians are a little bit less number less coding focused and they can still
158
00:11:41,980 --> 00:11:49,760
do really, really great work and bring a lot to the table and contribute more on the planning
159
00:11:49,760 --> 00:11:51,480
and methods phase.
160
00:11:51,480 --> 00:11:59,040
But in terms of in terms of a division of what I do, I do probably 50 ish percent coding
161
00:11:59,040 --> 00:12:07,840
and then the other the other amount of what I do is done in terms of planning or writing.
162
00:12:07,840 --> 00:12:10,060
Report writing is also important.
163
00:12:10,060 --> 00:12:13,760
And I think this is something that sometimes people will think students will think.
164
00:12:13,760 --> 00:12:17,160
And I remember I thought this when I was in high school that, you know, there's there's
165
00:12:17,160 --> 00:12:20,200
an art stream and then there's a science stream.
166
00:12:20,200 --> 00:12:24,820
And you don't if you're you know, if you're artsy, you you you can write.
167
00:12:24,820 --> 00:12:28,400
And if you if you do science, you can't write.
168
00:12:28,400 --> 00:12:29,600
And that changes.
169
00:12:29,600 --> 00:12:32,040
You have to develop your writing skill.
170
00:12:32,040 --> 00:12:35,740
The further along in science you go, there's a lot of writing in science and it's it's
171
00:12:35,740 --> 00:12:39,800
a different kind of writing than creative writing.
172
00:12:39,800 --> 00:12:44,520
But but yeah, I'd say that's the other the other part of what I do for for work.
173
00:12:44,520 --> 00:12:47,080
But not to derail the coding part.
174
00:12:47,080 --> 00:12:54,840
A lot of a lot of what I do is is looking at our double checking in either books or
175
00:12:54,840 --> 00:12:58,520
online to make my code better.
176
00:12:58,520 --> 00:13:04,600
Or recently, I don't know if I should say this, Chat GPT is helpful to we're very at
177
00:13:04,600 --> 00:13:05,600
my company.
178
00:13:05,600 --> 00:13:06,600
We're very open with it.
179
00:13:06,600 --> 00:13:12,440
And in school, it's it's a little bit more of a controversial topic, but it's it's been
180
00:13:12,440 --> 00:13:13,960
it's been helpful there.
181
00:13:13,960 --> 00:13:19,920
But yeah, yeah, that's a good amount of coding, I'd say it's important.
182
00:13:19,920 --> 00:13:25,240
I love that you brought up writing and report writing because so often I know I didn't think
183
00:13:25,240 --> 00:13:28,980
I was like, OK, science will have like a science report here and there.
184
00:13:28,980 --> 00:13:31,240
But a lot of science is a lot of writing.
185
00:13:31,240 --> 00:13:36,560
And in the last couple episodes of this podcast, actually, we've talked a lot about how creative
186
00:13:36,560 --> 00:13:39,080
creativity and science kind of go hand in hand.
187
00:13:39,080 --> 00:13:40,800
So I think it's I don't know.
188
00:13:40,800 --> 00:13:45,880
Did you want to touch on that a little bit more about if creativity has anything to do
189
00:13:45,880 --> 00:13:47,280
with your day to day?
190
00:13:47,280 --> 00:13:51,120
Yeah, I'd say creativity is very important.
191
00:13:51,120 --> 00:13:57,320
Usually it's it's easy to just kind of think that with statistics that you just understand
192
00:13:57,320 --> 00:14:01,680
everything at some point and that there's nothing more to do and that and there's definitely
193
00:14:01,680 --> 00:14:06,160
a bit of a temptation to think that, oh, the computer once I have once I have enough computer
194
00:14:06,160 --> 00:14:10,800
code written, it'll just be automatic and I won't have to do anything.
195
00:14:10,800 --> 00:14:15,360
And there's conveniences that come with having good coding skills and things set up that
196
00:14:15,360 --> 00:14:16,880
you can repeat.
197
00:14:16,880 --> 00:14:21,400
But I think that part is what takes away from some of the creativity.
198
00:14:21,400 --> 00:14:27,480
And what's important in this field is always thinking of ways to solve problems.
199
00:14:27,480 --> 00:14:29,640
And you need creativity for that.
200
00:14:29,640 --> 00:14:34,680
So you need to be thinking of what what are some other ways that I could approach this
201
00:14:34,680 --> 00:14:41,520
problem and what are some other ways that I could analyze this data that would be valid
202
00:14:41,520 --> 00:14:47,560
and that would help to reduce the bias in the data and in the analysis?
203
00:14:47,560 --> 00:14:53,800
Because bias is always an issue in health research and it's always a very big concern
204
00:14:53,800 --> 00:15:00,800
because we need to be as unbiased as possible and as focused on getting the most representative
205
00:15:00,800 --> 00:15:04,400
sample that we can of our population.
206
00:15:04,400 --> 00:15:09,600
So I think for those for those major things, creativity is very important because it keeps
207
00:15:09,600 --> 00:15:15,040
you it keeps you active, it keeps you engaged, it keeps you reading and thinking about how
208
00:15:15,040 --> 00:15:16,920
you can solve problems.
209
00:15:16,920 --> 00:15:23,440
And I guess this might also come back to planning where you want to have a strong plan so that
210
00:15:23,440 --> 00:15:29,800
you can kind of problem solve before the problems turn into unsolvable problems or before your
211
00:15:29,800 --> 00:15:35,320
entire database is skewed or biased or anything like that.
212
00:15:35,320 --> 00:15:40,640
Exactly and I think I think it's planning and creativity can go hand in hand.
213
00:15:40,640 --> 00:15:44,600
You can plan for creative time.
214
00:15:44,600 --> 00:15:51,560
So you can you can set part of your day to be a time to do research and a time to think
215
00:15:51,560 --> 00:15:55,340
and a time to think about how to how to tackle the problem more.
216
00:15:55,340 --> 00:16:01,800
So it can really yeah it can go it doesn't it doesn't have to be one or the other.
217
00:16:01,800 --> 00:16:05,240
So it's kind of switch gears a little bit here.
218
00:16:05,240 --> 00:16:10,980
Do you have a favorite science or industry kind of thing that you've done before?
219
00:16:10,980 --> 00:16:18,120
I would say I really like the work that we do at my company right now is really interesting
220
00:16:18,120 --> 00:16:22,580
because we do we do something called indirect comparisons.
221
00:16:22,580 --> 00:16:32,040
And indirect comparisons are basically trying to compare two different studies that didn't
222
00:16:32,040 --> 00:16:34,620
didn't compare something within the studies.
223
00:16:34,620 --> 00:16:39,680
So it's taking data from one study taking data from another study and putting them together
224
00:16:39,680 --> 00:16:41,520
and doing it in the right way.
225
00:16:41,520 --> 00:16:49,980
And I find there's a lot there's a lot of nuance and there's a lot of interesting pitfalls
226
00:16:49,980 --> 00:16:53,480
that you can fall into if you are not careful.
227
00:16:53,480 --> 00:16:59,120
So it's it's it's it's something that you need to really be thinking about and be working
228
00:16:59,120 --> 00:17:00,920
as a team on.
229
00:17:00,920 --> 00:17:10,520
So I find I find that is is very rewarding when we get when we complete a project in
230
00:17:10,520 --> 00:17:16,600
in that area and are able to deliver something really helpful to our clients because they
231
00:17:16,600 --> 00:17:23,200
gain valuable information from from seeing the comparisons that we have to make that
232
00:17:23,200 --> 00:17:26,920
they wouldn't otherwise be able to see if they just read what's published.
233
00:17:26,920 --> 00:17:33,720
I think one of my favorite things coming from a research background was when I could actually
234
00:17:33,720 --> 00:17:38,000
connect the dots as well and have something here and have something here and finally make
235
00:17:38,000 --> 00:17:43,400
a connection and have it not confirmed because science is always like evolving.
236
00:17:43,400 --> 00:17:47,000
And that's a whole other discussion for a podcast that's much longer than 30 minutes
237
00:17:47,000 --> 00:17:50,040
about whether or not things can be confirmed for real.
238
00:17:50,040 --> 00:17:55,560
But having things like OK, because we did this, this happened.
239
00:17:55,560 --> 00:17:58,200
That was one of the coolest and most rewarding things.
240
00:17:58,200 --> 00:18:02,760
And it sounds like to be able to do that on such a large scale with different like health
241
00:18:02,760 --> 00:18:08,160
studies, I can see how that would be really rewarding and just really wicked and cool.
242
00:18:08,160 --> 00:18:12,040
Yeah, making making connections is really that's the best part.
243
00:18:12,040 --> 00:18:15,000
I really I enjoy it.
244
00:18:15,000 --> 00:18:20,360
It makes you feel like you have have some expertise in an area.
245
00:18:20,360 --> 00:18:26,000
And that's that's something that is very rewarding and it's very it's very helpful and clients
246
00:18:26,000 --> 00:18:29,800
appreciate it too when they when they know that you're a reliable expert.
247
00:18:29,800 --> 00:18:35,480
So kind of the last question here, this is kind of similar to the billboard question.
248
00:18:35,480 --> 00:18:39,920
What is one thing you think everyone should know about your topic?
249
00:18:39,920 --> 00:18:45,760
You will probably need to do statistics at some point, but you don't need to do too much.
250
00:18:45,760 --> 00:18:51,880
And and it's something that you can approach it, take some time, think about it and come
251
00:18:51,880 --> 00:18:52,880
back to it.
252
00:18:52,880 --> 00:18:55,200
You don't have to force it.
253
00:18:55,200 --> 00:18:59,960
And I think that's really something that that dissuades people from doing going down math
254
00:18:59,960 --> 00:19:05,560
and statistics in general, is they will run into a wall and get frustrated.
255
00:19:05,560 --> 00:19:10,160
And you will get frustrated, probably, and you will make mistakes, but just just give
256
00:19:10,160 --> 00:19:11,160
it some time.
257
00:19:11,160 --> 00:19:18,300
Take some time and and and see see what turns out and and and how important it is to talk
258
00:19:18,300 --> 00:19:25,180
to people and be social about the problems you're you're having and and how what they
259
00:19:25,180 --> 00:19:26,800
think would be a good solution.
260
00:19:26,800 --> 00:19:31,440
So yeah, I think patience and being being kind of talkative at a good team player.
261
00:19:31,440 --> 00:19:36,460
So important, and I think being a good team player is important for every job, but maybe
262
00:19:36,460 --> 00:19:42,480
sometimes not something you'd think about when you're thinking about really hard math.
263
00:19:42,480 --> 00:19:45,480
But I mean, it's important everywhere just to kind of wrap up.
264
00:19:45,480 --> 00:19:49,720
Is there any kind of question you'd like to go back to or anything else you'd kind of
265
00:19:49,720 --> 00:19:51,840
like to say or expand on?
266
00:19:51,840 --> 00:19:59,720
I guess if if statistics, if biostatistics doesn't sound too appealing to you, there's
267
00:19:59,720 --> 00:20:05,400
also just statistics in general, which is something that would be used across the board
268
00:20:05,400 --> 00:20:07,120
for any for any science.
269
00:20:07,120 --> 00:20:09,200
And there's a lot of overlap there.
270
00:20:09,200 --> 00:20:14,700
There's different levels of precision needed in different different areas of interest,
271
00:20:14,700 --> 00:20:19,320
but you can really add statistics and pair it to any other science industry and you'll
272
00:20:19,320 --> 00:20:25,600
be able to work with researchers and work with labs easier.
273
00:20:25,600 --> 00:20:31,280
100% if you're going into science or thinking about going into science, statistics are going
274
00:20:31,280 --> 00:20:32,280
to be there.
275
00:20:32,280 --> 00:20:37,280
So you might as well try to have a good relationship with statistics and try to make them your
276
00:20:37,280 --> 00:20:38,280
friend.
277
00:20:38,280 --> 00:20:39,280
Exactly.
278
00:20:39,280 --> 00:20:42,200
Well, thank you so much, Colin.
279
00:20:42,200 --> 00:20:44,480
I learned so much today.
280
00:20:44,480 --> 00:20:46,560
It was so to talk to you.
281
00:20:46,560 --> 00:20:47,560
Yeah, thanks.
282
00:20:47,560 --> 00:20:48,560
Thanks a lot for having me, Mab.
283
00:20:48,560 --> 00:20:51,840
It was it was great to be a part of this.
284
00:20:51,840 --> 00:20:56,600
A humongous thank you to Colin for joining us today on What Do Scientists Do?.
285
00:20:56,600 --> 00:21:01,920
I had a great time learning about biostatistics and as always a ginormous thank you to everybody
286
00:21:01,920 --> 00:21:03,080
listening out there.
287
00:21:03,080 --> 00:21:07,320
If you're commuting, if you're in the back of your parent and guardian's car, if you're
288
00:21:07,320 --> 00:21:13,660
biking home safely with a helmet on and obeying all traffic signals, I hope you're having
289
00:21:13,660 --> 00:21:15,620
a great time listening as well.
290
00:21:15,620 --> 00:21:20,960
For more science fun, you can find us on Twitter or Instagram at SuperNOVAatDal.
291
00:21:20,960 --> 00:21:24,240
Do you have a question that you'd like answered by one of our experts?
292
00:21:24,240 --> 00:21:30,480
Send us an email or a voice recording at whatdoscientistsdo@superstaff.ca and we might
293
00:21:30,480 --> 00:21:32,560
just answer your question on the phone.
294
00:21:32,560 --> 00:21:34,620
Show no phone.
295
00:21:34,620 --> 00:21:43,400
Thank you so much for listening and we will see you but not hear you next episode.
296
00:21:43,400 --> 00:21:48,080
Bye bye for now.
297
00:21:48,080 --> 00:21:52,840
This show was made by SuperNOVA at Dalhousie University, a network member of Actua.
298
00:21:52,840 --> 00:22:21,760
For more information on our workshops, summer camps, and more, visit supernova.dal.ca.
00:00:00,000 --> 00:00:15,600
Hello and welcome back to season three of What Do Scientists Do?
2
00:00:15,600 --> 00:00:19,480
This is the show where we talk to expert guests and learn about what science they do, how
3
00:00:19,480 --> 00:00:21,660
they do it, and why they do it.
4
00:00:21,660 --> 00:00:25,600
My name is Mab and today I got to talk to Colin Vibert.
5
00:00:25,600 --> 00:00:30,840
Colin teaches us all about a field called biostatistics and how we can use math to make health care
6
00:00:30,840 --> 00:00:31,840
better.
7
00:00:31,840 --> 00:00:37,280
But what are statistics and how does Chat GPT help our computer programs?
8
00:00:37,280 --> 00:00:40,760
And what does all of this have to do with why writing skills matter?
9
00:00:40,760 --> 00:00:42,360
We're doing math after all.
10
00:00:42,360 --> 00:00:45,240
Well, we answer these questions and more.
11
00:00:45,240 --> 00:00:48,200
So let's jump in and find out.
12
00:00:48,200 --> 00:00:50,840
Hello and welcome back to What Do Scientists Do?
13
00:00:50,840 --> 00:00:54,880
This is a show where we talk to wickedly smart guests and learn all about what they do and
14
00:00:54,880 --> 00:00:57,240
what science has got to do with it.
15
00:00:57,240 --> 00:01:00,140
Today we're going to talk about a field called biostatistics.
16
00:01:00,140 --> 00:01:04,880
So biostatistics as a word might seem complicated, but we can actually break it down into smaller
17
00:01:04,880 --> 00:01:07,680
parts and we get to recognizable bits.
18
00:01:07,680 --> 00:01:11,400
So you probably recognize the word bio or the study of living things.
19
00:01:11,400 --> 00:01:13,280
And then we also have statistics.
20
00:01:13,280 --> 00:01:16,800
So statistics goes hand in hand with probability usually.
21
00:01:16,800 --> 00:01:19,720
And this is something you might have learned about in school.
22
00:01:19,720 --> 00:01:24,560
So statistics is all about collecting and looking at data about certain things.
23
00:01:24,560 --> 00:01:28,560
So if we put this whole word pizza together into a whole pizza again, we actually have
24
00:01:28,560 --> 00:01:34,520
the study of data and using data to figure out how living things work.
25
00:01:34,520 --> 00:01:39,240
So that's just kind of the Mab analysis, Mab synthesis of what this all means.
26
00:01:39,240 --> 00:01:43,220
So luckily we actually have an expert here to talk to us about what that actually means.
27
00:01:43,220 --> 00:01:45,000
So here's Colin.
28
00:01:45,000 --> 00:01:47,480
So hi Colin, thank you so much for joining us today.
29
00:01:47,480 --> 00:01:50,560
Do you want to introduce yourself with your name and pronouns?
30
00:01:50,560 --> 00:01:51,560
Yeah.
31
00:01:51,560 --> 00:01:56,560
My name is Colin Vibert and my pronouns are he, him and his.
32
00:01:56,560 --> 00:01:58,400
And what kind of scientist are you?
33
00:01:58,400 --> 00:02:01,560
I am a biostatistician.
34
00:02:01,560 --> 00:02:05,940
So I think your description was very accurate.
35
00:02:05,940 --> 00:02:15,640
And I work with in the life sciences, pharmaceutical industry, and also have done some work in other
36
00:02:15,640 --> 00:02:18,000
areas in healthcare.
37
00:02:18,000 --> 00:02:23,240
Yeah, so something that I kept kind of running into when I was looking at biostatistics
38
00:02:23,240 --> 00:02:27,000
was how much it has to do with healthcare.
39
00:02:27,000 --> 00:02:29,960
So what do you do in healthcare?
40
00:02:29,960 --> 00:02:32,640
What does a biostatistician do in healthcare?
41
00:02:32,640 --> 00:02:40,000
A biostatistician in healthcare will often be working either at a hospital or a pharmaceutical
42
00:02:40,000 --> 00:02:51,920
company or an agency and they will be helping for a disease area or for a medical product
43
00:02:51,920 --> 00:02:59,200
will be looking to understand the product better or understand the disease better.
44
00:02:59,200 --> 00:03:02,240
And we do that with numbers, essentially.
45
00:03:02,240 --> 00:03:08,360
Yeah, so numbers, we get all of this data, maybe hospitals are constantly collecting
46
00:03:08,360 --> 00:03:13,680
data about how many patients there are or maybe what kind of diseases or illnesses are
47
00:03:13,680 --> 00:03:15,600
bringing them into the hospital.
48
00:03:15,600 --> 00:03:21,240
And then you take all that data and you work it into knowledge or understanding?
49
00:03:21,240 --> 00:03:22,880
Yeah, exactly.
50
00:03:22,880 --> 00:03:33,640
So in Canada and in most countries, but in Canada, we have a lot of science that gets
51
00:03:33,640 --> 00:03:41,520
published and we also have agencies that require for pharmaceutical products output to be recommended
52
00:03:41,520 --> 00:03:42,520
and recorded.
53
00:03:42,520 --> 00:03:49,320
So when a product is being developed, we need to be very open and transparent about how
54
00:03:49,320 --> 00:03:53,040
well it works and what are the possible safety risks.
55
00:03:53,040 --> 00:03:56,400
And we use data for understanding that better.
56
00:03:56,400 --> 00:04:00,160
This all kind of feeds into the scientific method as well.
57
00:04:00,160 --> 00:04:06,080
So you maybe have a hypothesis, so you have a question and then you maybe have an experiment.
58
00:04:06,080 --> 00:04:10,560
And then after your experiment is done, you have all of this data or the results.
59
00:04:10,560 --> 00:04:12,480
And then from there, you just have raw numbers.
60
00:04:12,480 --> 00:04:16,200
So you need to know what to do with the actual numbers, right?
61
00:04:16,200 --> 00:04:17,200
Absolutely.
62
00:04:17,200 --> 00:04:22,880
Yeah, with statistics, having a plan and following through with the plan is very important.
63
00:04:22,880 --> 00:04:23,920
Yeah, exactly.
64
00:04:23,920 --> 00:04:28,480
You want to be able to have numbers that you can kind of work with.
65
00:04:28,480 --> 00:04:36,440
So do you specifically work with like collecting data about pharmaceuticals or do you work
66
00:04:36,440 --> 00:04:37,440
in a hospital?
67
00:04:37,440 --> 00:04:44,480
Yeah, so I work right now for a company that is doing work with pharmaceutical companies.
68
00:04:44,480 --> 00:04:50,040
And so we're focused on the performance of different pharmaceutical products and medical
69
00:04:50,040 --> 00:04:55,320
devices on how they perform for a specific disease.
70
00:04:55,320 --> 00:04:57,040
Okay, that's really cool.
71
00:04:57,040 --> 00:05:03,200
So if you're comparing treatments, for example, maybe if you're comparing different cancer
72
00:05:03,200 --> 00:05:07,760
treatments, if one treatment is more efficient than the other.
73
00:05:07,760 --> 00:05:08,760
Exactly.
74
00:05:08,760 --> 00:05:09,760
Awesome.
75
00:05:09,760 --> 00:05:13,660
So this might be kind of a tricky question here.
76
00:05:13,660 --> 00:05:19,880
But if you had to summarize what you do to like, if Colin had a billboard across Canada,
77
00:05:19,880 --> 00:05:23,480
what would you put on your billboard to summarize what you do?
78
00:05:23,480 --> 00:05:25,040
Well, that's a good question.
79
00:05:25,040 --> 00:05:28,720
I would put I would say planning.
80
00:05:28,720 --> 00:05:36,920
I plan a lot and I do computer coding a lot.
81
00:05:36,920 --> 00:05:47,280
And I am thinking and reading and doing research for understanding data methods and data analysis
82
00:05:47,280 --> 00:05:48,560
better.
83
00:05:48,560 --> 00:05:53,400
So when you talk about planning, what kind of things do you have to plan for?
84
00:05:53,400 --> 00:05:58,720
Yeah, so whenever you want to do just like you said with the scientific method, whenever
85
00:05:58,720 --> 00:06:05,880
you want to do a study before you do the study, you have to have a good understanding of what
86
00:06:05,880 --> 00:06:09,400
you want to do and what data you want to collect.
87
00:06:09,400 --> 00:06:15,080
That means you need to have a plan written out of what you're going to actually end up
88
00:06:15,080 --> 00:06:18,960
doing before you do it.
89
00:06:18,960 --> 00:06:27,760
And so what that means is, let's say you want to study how well a medical device is performing
90
00:06:27,760 --> 00:06:34,040
for a disease, you need to first specify, well, what are we going to measure when we
91
00:06:34,040 --> 00:06:37,620
say that we want to see if the device is performing well?
92
00:06:37,620 --> 00:06:39,680
What score do we use to measure that?
93
00:06:39,680 --> 00:06:46,080
So we would plan by thinking about what score is important, what scores are important, additional
94
00:06:46,080 --> 00:06:52,760
data, demographic data, so data about like the age of the patients, the race of the patients,
95
00:06:52,760 --> 00:06:58,080
the location, all of that is also important to figure out how we're going to characterize
96
00:06:58,080 --> 00:07:00,020
our population.
97
00:07:00,020 --> 00:07:07,640
Because with health studies, it's very important that you are collecting accurate populations
98
00:07:07,640 --> 00:07:12,580
that represent the patients and the disease correctly.
99
00:07:12,580 --> 00:07:16,560
If you don't, then the conclusions you arrive at will be biased.
100
00:07:16,560 --> 00:07:21,160
So you want to make sure that you're collecting enough data that represent what's actually
101
00:07:21,160 --> 00:07:22,720
going on.
102
00:07:22,720 --> 00:07:29,000
So one of the things about statistics, Colin, you could correct me if I'm wrong, but I'm,
103
00:07:29,000 --> 00:07:32,600
I don't do this for a living and it's been a while since I took a stats course.
104
00:07:32,600 --> 00:07:39,040
But one thing about stats is that stats aims to collect small representations, comparatively
105
00:07:39,040 --> 00:07:45,120
small representations about a wider population because it's just not possible to ask every
106
00:07:45,120 --> 00:07:47,840
single person every single question.
107
00:07:47,840 --> 00:07:54,280
So what statistics tries to do is ask a couple people, a diverse and representative group
108
00:07:54,280 --> 00:08:01,800
of people, a couple things to then get an idea or try to predict a larger population.
109
00:08:01,800 --> 00:08:06,840
So it sounds like in health sciences as well, you want to make sure that you're predicting
110
00:08:06,840 --> 00:08:10,280
accurately because these are people's health and wellness on the line.
111
00:08:10,280 --> 00:08:14,480
You want to make sure that you're doing a good job to represent everybody, which I think
112
00:08:14,480 --> 00:08:18,320
is really cool and something that I never really thought about, I guess.
113
00:08:18,320 --> 00:08:21,920
Yeah, that's, that's exactly right on.
114
00:08:21,920 --> 00:08:28,920
You want, you know, if you have, let's say, let's say the medical device could be given
115
00:08:28,920 --> 00:08:34,020
to 10,000 people in North America.
116
00:08:34,020 --> 00:08:41,220
You would, you would not really be able to go out and collect from all 10,000 people.
117
00:08:41,220 --> 00:08:51,320
So you use statistics to try to understand how this device performs with a much smaller
118
00:08:51,320 --> 00:08:53,480
sample, say 100.
119
00:08:53,480 --> 00:08:58,000
And I guess another thing that's kind of sticking out to me is that you work in health care.
120
00:08:58,000 --> 00:09:02,040
And at least when I was younger, I thought working in health care meant being a doctor
121
00:09:02,040 --> 00:09:05,080
or being a nurse.
122
00:09:05,080 --> 00:09:10,320
And it's cool to see, I mean, that your job is just one example of working in health care
123
00:09:10,320 --> 00:09:15,520
and the health care industry that has an impact on people's health and wellness without being,
124
00:09:15,520 --> 00:09:19,280
you know, a doctor or a nurse, but still being kind of health oriented.
125
00:09:19,280 --> 00:09:23,120
There's a lot of people working behind the scenes on health, which is something I think
126
00:09:23,120 --> 00:09:25,640
I always forget about and maybe other people do.
127
00:09:25,640 --> 00:09:26,640
Absolutely.
128
00:09:26,640 --> 00:09:32,800
I would say from my experience so far, I don't have experience working as a health care worker,
129
00:09:32,800 --> 00:09:39,880
as say a frontline health care worker, like a nurse, a doctor, a physician's assistant,
130
00:09:39,880 --> 00:09:41,880
but it is very important.
131
00:09:41,880 --> 00:09:50,160
I found and I believe this, that monitoring disease and monitoring the performance of
132
00:09:50,160 --> 00:09:57,120
the products that are used to treat disease are very, very frequently and continuously done
133
00:09:57,120 --> 00:10:03,880
so that you have a good understanding of how well the treatments are and how well and how
134
00:10:03,880 --> 00:10:05,000
much better you can be.
135
00:10:05,000 --> 00:10:09,080
I think that's really an important part of health care is that we're always thinking
136
00:10:09,080 --> 00:10:13,240
of developing better treatments to help patients more.
137
00:10:13,240 --> 00:10:17,320
And so that's what the research avenue brings.
138
00:10:17,320 --> 00:10:22,880
Whereas if you're a frontline worker, you're doing your very best to treat with the means
139
00:10:22,880 --> 00:10:24,240
you have available.
140
00:10:24,240 --> 00:10:31,240
So your job is kind of looking in the future and trying to come up with ways to make health
141
00:10:31,240 --> 00:10:33,560
care in the future better as well.
142
00:10:33,560 --> 00:10:34,560
Exactly.
143
00:10:34,560 --> 00:10:35,560
Yeah.
144
00:10:35,560 --> 00:10:36,560
Cool.
145
00:10:36,560 --> 00:10:40,640
So another question is you mentioned computer programming and I was just curious about how
146
00:10:40,640 --> 00:10:46,520
much your job is sitting in front of a database and doing the nitty gritty computer programming
147
00:10:46,520 --> 00:10:50,240
and like what computer programming and coding has to do with your job?
148
00:10:50,240 --> 00:10:59,680
Yeah, I would say so I as a biostatistician, I do a lot of work in the R programming language.
149
00:10:59,680 --> 00:11:09,320
There are also other languages like SAS and SPSS that can do computer programming to some
150
00:11:09,320 --> 00:11:13,200
extent, but R is a very popular one right now.
151
00:11:13,200 --> 00:11:20,120
And I would say probably 50 to 60 percent of my day is spent on a given day doing some
152
00:11:20,120 --> 00:11:23,400
coding in some form.
153
00:11:23,400 --> 00:11:29,320
But I am someone who is because I have a math and statistics background from university,
154
00:11:29,320 --> 00:11:33,360
I'm kind of known within my company as being someone who is a math and stats and coding
155
00:11:33,360 --> 00:11:34,360
person.
156
00:11:34,360 --> 00:11:36,020
So they go to me.
157
00:11:36,020 --> 00:11:41,980
Other biostatisticians are a little bit less number less coding focused and they can still
158
00:11:41,980 --> 00:11:49,760
do really, really great work and bring a lot to the table and contribute more on the planning
159
00:11:49,760 --> 00:11:51,480
and methods phase.
160
00:11:51,480 --> 00:11:59,040
But in terms of in terms of a division of what I do, I do probably 50 ish percent coding
161
00:11:59,040 --> 00:12:07,840
and then the other the other amount of what I do is done in terms of planning or writing.
162
00:12:07,840 --> 00:12:10,060
Report writing is also important.
163
00:12:10,060 --> 00:12:13,760
And I think this is something that sometimes people will think students will think.
164
00:12:13,760 --> 00:12:17,160
And I remember I thought this when I was in high school that, you know, there's there's
165
00:12:17,160 --> 00:12:20,200
an art stream and then there's a science stream.
166
00:12:20,200 --> 00:12:24,820
And you don't if you're you know, if you're artsy, you you you can write.
167
00:12:24,820 --> 00:12:28,400
And if you if you do science, you can't write.
168
00:12:28,400 --> 00:12:29,600
And that changes.
169
00:12:29,600 --> 00:12:32,040
You have to develop your writing skill.
170
00:12:32,040 --> 00:12:35,740
The further along in science you go, there's a lot of writing in science and it's it's
171
00:12:35,740 --> 00:12:39,800
a different kind of writing than creative writing.
172
00:12:39,800 --> 00:12:44,520
But but yeah, I'd say that's the other the other part of what I do for for work.
173
00:12:44,520 --> 00:12:47,080
But not to derail the coding part.
174
00:12:47,080 --> 00:12:54,840
A lot of a lot of what I do is is looking at our double checking in either books or
175
00:12:54,840 --> 00:12:58,520
online to make my code better.
176
00:12:58,520 --> 00:13:04,600
Or recently, I don't know if I should say this, Chat GPT is helpful to we're very at
177
00:13:04,600 --> 00:13:05,600
my company.
178
00:13:05,600 --> 00:13:06,600
We're very open with it.
179
00:13:06,600 --> 00:13:12,440
And in school, it's it's a little bit more of a controversial topic, but it's it's been
180
00:13:12,440 --> 00:13:13,960
it's been helpful there.
181
00:13:13,960 --> 00:13:19,920
But yeah, yeah, that's a good amount of coding, I'd say it's important.
182
00:13:19,920 --> 00:13:25,240
I love that you brought up writing and report writing because so often I know I didn't think
183
00:13:25,240 --> 00:13:28,980
I was like, OK, science will have like a science report here and there.
184
00:13:28,980 --> 00:13:31,240
But a lot of science is a lot of writing.
185
00:13:31,240 --> 00:13:36,560
And in the last couple episodes of this podcast, actually, we've talked a lot about how creative
186
00:13:36,560 --> 00:13:39,080
creativity and science kind of go hand in hand.
187
00:13:39,080 --> 00:13:40,800
So I think it's I don't know.
188
00:13:40,800 --> 00:13:45,880
Did you want to touch on that a little bit more about if creativity has anything to do
189
00:13:45,880 --> 00:13:47,280
with your day to day?
190
00:13:47,280 --> 00:13:51,120
Yeah, I'd say creativity is very important.
191
00:13:51,120 --> 00:13:57,320
Usually it's it's easy to just kind of think that with statistics that you just understand
192
00:13:57,320 --> 00:14:01,680
everything at some point and that there's nothing more to do and that and there's definitely
193
00:14:01,680 --> 00:14:06,160
a bit of a temptation to think that, oh, the computer once I have once I have enough computer
194
00:14:06,160 --> 00:14:10,800
code written, it'll just be automatic and I won't have to do anything.
195
00:14:10,800 --> 00:14:15,360
And there's conveniences that come with having good coding skills and things set up that
196
00:14:15,360 --> 00:14:16,880
you can repeat.
197
00:14:16,880 --> 00:14:21,400
But I think that part is what takes away from some of the creativity.
198
00:14:21,400 --> 00:14:27,480
And what's important in this field is always thinking of ways to solve problems.
199
00:14:27,480 --> 00:14:29,640
And you need creativity for that.
200
00:14:29,640 --> 00:14:34,680
So you need to be thinking of what what are some other ways that I could approach this
201
00:14:34,680 --> 00:14:41,520
problem and what are some other ways that I could analyze this data that would be valid
202
00:14:41,520 --> 00:14:47,560
and that would help to reduce the bias in the data and in the analysis?
203
00:14:47,560 --> 00:14:53,800
Because bias is always an issue in health research and it's always a very big concern
204
00:14:53,800 --> 00:15:00,800
because we need to be as unbiased as possible and as focused on getting the most representative
205
00:15:00,800 --> 00:15:04,400
sample that we can of our population.
206
00:15:04,400 --> 00:15:09,600
So I think for those for those major things, creativity is very important because it keeps
207
00:15:09,600 --> 00:15:15,040
you it keeps you active, it keeps you engaged, it keeps you reading and thinking about how
208
00:15:15,040 --> 00:15:16,920
you can solve problems.
209
00:15:16,920 --> 00:15:23,440
And I guess this might also come back to planning where you want to have a strong plan so that
210
00:15:23,440 --> 00:15:29,800
you can kind of problem solve before the problems turn into unsolvable problems or before your
211
00:15:29,800 --> 00:15:35,320
entire database is skewed or biased or anything like that.
212
00:15:35,320 --> 00:15:40,640
Exactly and I think I think it's planning and creativity can go hand in hand.
213
00:15:40,640 --> 00:15:44,600
You can plan for creative time.
214
00:15:44,600 --> 00:15:51,560
So you can you can set part of your day to be a time to do research and a time to think
215
00:15:51,560 --> 00:15:55,340
and a time to think about how to how to tackle the problem more.
216
00:15:55,340 --> 00:16:01,800
So it can really yeah it can go it doesn't it doesn't have to be one or the other.
217
00:16:01,800 --> 00:16:05,240
So it's kind of switch gears a little bit here.
218
00:16:05,240 --> 00:16:10,980
Do you have a favorite science or industry kind of thing that you've done before?
219
00:16:10,980 --> 00:16:18,120
I would say I really like the work that we do at my company right now is really interesting
220
00:16:18,120 --> 00:16:22,580
because we do we do something called indirect comparisons.
221
00:16:22,580 --> 00:16:32,040
And indirect comparisons are basically trying to compare two different studies that didn't
222
00:16:32,040 --> 00:16:34,620
didn't compare something within the studies.
223
00:16:34,620 --> 00:16:39,680
So it's taking data from one study taking data from another study and putting them together
224
00:16:39,680 --> 00:16:41,520
and doing it in the right way.
225
00:16:41,520 --> 00:16:49,980
And I find there's a lot there's a lot of nuance and there's a lot of interesting pitfalls
226
00:16:49,980 --> 00:16:53,480
that you can fall into if you are not careful.
227
00:16:53,480 --> 00:16:59,120
So it's it's it's it's something that you need to really be thinking about and be working
228
00:16:59,120 --> 00:17:00,920
as a team on.
229
00:17:00,920 --> 00:17:10,520
So I find I find that is is very rewarding when we get when we complete a project in
230
00:17:10,520 --> 00:17:16,600
in that area and are able to deliver something really helpful to our clients because they
231
00:17:16,600 --> 00:17:23,200
gain valuable information from from seeing the comparisons that we have to make that
232
00:17:23,200 --> 00:17:26,920
they wouldn't otherwise be able to see if they just read what's published.
233
00:17:26,920 --> 00:17:33,720
I think one of my favorite things coming from a research background was when I could actually
234
00:17:33,720 --> 00:17:38,000
connect the dots as well and have something here and have something here and finally make
235
00:17:38,000 --> 00:17:43,400
a connection and have it not confirmed because science is always like evolving.
236
00:17:43,400 --> 00:17:47,000
And that's a whole other discussion for a podcast that's much longer than 30 minutes
237
00:17:47,000 --> 00:17:50,040
about whether or not things can be confirmed for real.
238
00:17:50,040 --> 00:17:55,560
But having things like OK, because we did this, this happened.
239
00:17:55,560 --> 00:17:58,200
That was one of the coolest and most rewarding things.
240
00:17:58,200 --> 00:18:02,760
And it sounds like to be able to do that on such a large scale with different like health
241
00:18:02,760 --> 00:18:08,160
studies, I can see how that would be really rewarding and just really wicked and cool.
242
00:18:08,160 --> 00:18:12,040
Yeah, making making connections is really that's the best part.
243
00:18:12,040 --> 00:18:15,000
I really I enjoy it.
244
00:18:15,000 --> 00:18:20,360
It makes you feel like you have have some expertise in an area.
245
00:18:20,360 --> 00:18:26,000
And that's that's something that is very rewarding and it's very it's very helpful and clients
246
00:18:26,000 --> 00:18:29,800
appreciate it too when they when they know that you're a reliable expert.
247
00:18:29,800 --> 00:18:35,480
So kind of the last question here, this is kind of similar to the billboard question.
248
00:18:35,480 --> 00:18:39,920
What is one thing you think everyone should know about your topic?
249
00:18:39,920 --> 00:18:45,760
You will probably need to do statistics at some point, but you don't need to do too much.
250
00:18:45,760 --> 00:18:51,880
And and it's something that you can approach it, take some time, think about it and come
251
00:18:51,880 --> 00:18:52,880
back to it.
252
00:18:52,880 --> 00:18:55,200
You don't have to force it.
253
00:18:55,200 --> 00:18:59,960
And I think that's really something that that dissuades people from doing going down math
254
00:18:59,960 --> 00:19:05,560
and statistics in general, is they will run into a wall and get frustrated.
255
00:19:05,560 --> 00:19:10,160
And you will get frustrated, probably, and you will make mistakes, but just just give
256
00:19:10,160 --> 00:19:11,160
it some time.
257
00:19:11,160 --> 00:19:18,300
Take some time and and and see see what turns out and and and how important it is to talk
258
00:19:18,300 --> 00:19:25,180
to people and be social about the problems you're you're having and and how what they
259
00:19:25,180 --> 00:19:26,800
think would be a good solution.
260
00:19:26,800 --> 00:19:31,440
So yeah, I think patience and being being kind of talkative at a good team player.
261
00:19:31,440 --> 00:19:36,460
So important, and I think being a good team player is important for every job, but maybe
262
00:19:36,460 --> 00:19:42,480
sometimes not something you'd think about when you're thinking about really hard math.
263
00:19:42,480 --> 00:19:45,480
But I mean, it's important everywhere just to kind of wrap up.
264
00:19:45,480 --> 00:19:49,720
Is there any kind of question you'd like to go back to or anything else you'd kind of
265
00:19:49,720 --> 00:19:51,840
like to say or expand on?
266
00:19:51,840 --> 00:19:59,720
I guess if if statistics, if biostatistics doesn't sound too appealing to you, there's
267
00:19:59,720 --> 00:20:05,400
also just statistics in general, which is something that would be used across the board
268
00:20:05,400 --> 00:20:07,120
for any for any science.
269
00:20:07,120 --> 00:20:09,200
And there's a lot of overlap there.
270
00:20:09,200 --> 00:20:14,700
There's different levels of precision needed in different different areas of interest,
271
00:20:14,700 --> 00:20:19,320
but you can really add statistics and pair it to any other science industry and you'll
272
00:20:19,320 --> 00:20:25,600
be able to work with researchers and work with labs easier.
273
00:20:25,600 --> 00:20:31,280
100% if you're going into science or thinking about going into science, statistics are going
274
00:20:31,280 --> 00:20:32,280
to be there.
275
00:20:32,280 --> 00:20:37,280
So you might as well try to have a good relationship with statistics and try to make them your
276
00:20:37,280 --> 00:20:38,280
friend.
277
00:20:38,280 --> 00:20:39,280
Exactly.
278
00:20:39,280 --> 00:20:42,200
Well, thank you so much, Colin.
279
00:20:42,200 --> 00:20:44,480
I learned so much today.
280
00:20:44,480 --> 00:20:46,560
It was so to talk to you.
281
00:20:46,560 --> 00:20:47,560
Yeah, thanks.
282
00:20:47,560 --> 00:20:48,560
Thanks a lot for having me, Mab.
283
00:20:48,560 --> 00:20:51,840
It was it was great to be a part of this.
284
00:20:51,840 --> 00:20:56,600
A humongous thank you to Colin for joining us today on What Do Scientists Do?.
285
00:20:56,600 --> 00:21:01,920
I had a great time learning about biostatistics and as always a ginormous thank you to everybody
286
00:21:01,920 --> 00:21:03,080
listening out there.
287
00:21:03,080 --> 00:21:07,320
If you're commuting, if you're in the back of your parent and guardian's car, if you're
288
00:21:07,320 --> 00:21:13,660
biking home safely with a helmet on and obeying all traffic signals, I hope you're having
289
00:21:13,660 --> 00:21:15,620
a great time listening as well.
290
00:21:15,620 --> 00:21:20,960
For more science fun, you can find us on Twitter or Instagram at SuperNOVAatDal.
291
00:21:20,960 --> 00:21:24,240
Do you have a question that you'd like answered by one of our experts?
292
00:21:24,240 --> 00:21:30,480
Send us an email or a voice recording at whatdoscientistsdo@superstaff.ca and we might
293
00:21:30,480 --> 00:21:32,560
just answer your question on the phone.
294
00:21:32,560 --> 00:21:34,620
Show no phone.
295
00:21:34,620 --> 00:21:43,400
Thank you so much for listening and we will see you but not hear you next episode.
296
00:21:43,400 --> 00:21:48,080
Bye bye for now.
297
00:21:48,080 --> 00:21:52,840
This show was made by SuperNOVA at Dalhousie University, a network member of Actua.
298
00:21:52,840 --> 00:22:21,760
For more information on our workshops, summer camps, and more, visit supernova.dal.ca.
