Explore
For Enterprise
Join for Free
Log In

艺术与人文
商务
计算机科学
数据科学
信息技术
健康
数学和逻辑
个人发展
物理科学与工程
社会科学
语言学习
學位
證書

瀏覽 Coursera 的全部課程

瀏覽
搜索
企業版
登錄
免費加入

Lecture 4E: Recency of Birth

Loading...

重性抑郁人群：公共卫生解决方法

约翰霍普金斯大学

4.6（151 個評分） | 13K 名學生已註冊

Public Mental Health is the application of the principles of medicine and social science to prevent the occurrence of mental and behavioral disorders and to promote mental health of the population. This course illustrates the principles of public health applied to depressive disorder, including principles of epidemiology, transcultural psychiatry, health services research, and prevention. It is predicted that by 2020 depressive disorder will be the most important cause of disease burden in the entire world! Every human being suffers from feeling depressed at some point or other, but only about one fifth of the population will experience an episode of depressive disorder over the course of their lives. This course illuminates the public health approach to disease, and the particular complexities of applying this approach to mental disorders, using depression as the exemplar.

查看授課大綱

審閱

4.6（151 個評分）

5 stars
111 ratings
4 stars
29 ratings
3 stars
7 ratings
2 stars
3 ratings
1 star
1 ratings

CM

Dec 13, 2018

Great course ! I learned information that i never knew existed.

DW

Apr 06, 2018

Very interesting course. Good structure and presentation, too.

從本節課中

The Search for Etiologic Clues

This week the focus is on possible causes for depressive disorder, including inheritance, stress, social life, work life, and the evolution of modern living.

Lecture 4A: Inheritance9:28

Lecture 4B: Stress & Diathesis7:35

Lecture 4C: Social Supports5:37

Lecture 4D: Work, Family, and Life Stage7:39

Lecture 4E: Recency of Birth20:42

教學方

William Eaton, PhD
Professor, Mental Health
Wietse A. Tol, PhD
Ali & Rose Kawi Assistant Professor, Mental Health
Ramin Mojtabai, MD
Associate Professor, Mental Health

腳本

Now we'll consider one more risk factor for depressive disorder.

And this is recency of birth.

There is an article published by Olle Hagnell in

the 1980's, called Are We Entering an Age of Melancholy?

And that is one of the things that the people will talk about a little bit.

And here is some data relevant to that.

This is data from the epidemiologic catchment area, ECA sites.

It shows the, the risk for depression, the cumulative percent

that we're depressed according to when the individual was born.

So if you look at the chartreuse

the lower, green slide, that's the oldest people.

Those people were 65 to 74 at the time of the survey.

And if they're 65 to 74 it means, they

were in the cohort of birth from 1905 to 1914.

So these are the old folks.

And you

can see, over their lifetimes, about 3% or 4%

of them ever have an episode of depressive disorder.

Pretty low percentage if we consider the teal

bar there which is the next bar to the

chartreuse bar, those people are 55 to 64 years of age at the time of the survey.

And that means they were born from 1915 to 1924,

and they have a slightly higher risk for depressive disorder.

It's still pretty low though,

over their lifetime even by the time their 64, it only gets up to about 4%.

Now let's shift and look at the younger population.

Let's look at people who were 15 to 24 at the time of the interview.

That's the dark blue bar on the very left most bar.

And, of course we can't follow those

people, until they're 34, because they aren't

that old, they're only 24 or less, at the, at the time of the interview.

But we can see that about 10% of them,

even by the time of the interview, even by the time

they're 24, 10% of them have had an episode of depressive disorder.

So what we're seeing is that births before 1935

have lower risk over their lifetime than births after 1935.

That's the cohort trend in the rate of depressive disorder.

When you collect data like this, these are population based surveys.

The data that I just

showed you was from five urban areas in

the United States, the E-, so called ECA Program.

But basically wherever you collect this data like this, and analyze

the data as I've just shown you, you get nearly identical effect.

And so we refer to this as the cohort effect.

That is younger cohorts have higher risk for depression.

That's the the, the new age of melancholy.

So here are some explanations for the cohort effect.

It could be people with depression just die off.

So when we do our survey and ask them if they've ever had an

episode of depression, the people with an

episode of depression aren't there, because they're dead.

Because they died sooner.

It's possible that, that has some effect to produce the data the way I've shown it.

But when we study the rate of depression as it's connected to mortality

we see that the connection is not strong enough to produce that effect.

It could be that older cohorts, the old folks, are less introspective.

This is a cultural shift.

So we talk about the psychologization of society.

That is where the young people are all the time

thinking about themselves and or they're willing to talk about themselves.

But old people feel a little more reticent, and

it's just not the thing to do to talk

about yourself.

And so when you are asked a question about depressive disorder, you're more

likely to say no, because it's just not proper, really, to talk about that.

It could be that as we've developed more and more

treatment options for depressive disorder, people become more aware of depression.

They are perhaps more likely to get treatment.

They visit a doctor.

They say what's wrong and he says you have depressive disorder.

Or they for one reason or other they, the

change in the system in mental health care might lead

to better recall from 19 births in the early part

of the century to births later part of the century.

And then the final thing is that and actually the most convincing

possibility is older subjects forget episodes that occur earlier in their life.

So you have a 65 year old guy, and you're asking over his entire life has he had

an episode of depressive disorder.

Well, it could be that he just doesn't remember

the episodes that occur when he was 25 or 35.

If he were 40, those episodes which would be a lot

closer in time and he might be more likely to remember them.

So Lou Giuffre has done a, a simulation of how that cohort effect might be produced.

And what he's done is he said look, let's make

a model in which people forget every year they

forget 1% of the episodes of depression that they've had.

And then if we have that model let's see what happens and

we'll make a cohort be born in 1900 and 1930 and 1960.

And you can see, if you are forgetting 1% of your episodes per year.

You tend to get a lower lifetime history

of depressive disorder, if you're born earlier than

if you're born later.

And let's say that you forget 3% of episodes forgotten each year.

That's a large percentage of episodes to be forgotten.

On the other hand you see those curves.

They look like the curves that I just showed you.

It looks like the age of melancholy curve.

And here is yet another example of

depression forgetting increasing with each age cohort.

A 10% and 5% and 1%, they just look just

like the cohort effect. So we do have this repeated finding that

depression is more likely to happen in cohorts born later in the century.

But we still have this important possibility that

it's an effective recall differences between the age cohort.

So we can study actually the prevalence of depressive disorder in the United States,

in different studies that have been conducted at different points in time.

So we can look at different data on prevalence

of disorder from service conducted at different points in time.

Over the recent several decades.

And we might look at the

National Longitudinal Alcohol Epidemiologic Survey conducted

in 1991 and 1992, which was a survey of the U.S. population.

It had 42,862 people, a huge survey.

As it

happens, it was not longitudinal, but they hoped it would be longitudinal.

But nevertheless, it's a good cross

sectional survey and gives us prevalence data.

And we can compare that data to the

National Epidemiology Survey on Alcohol and Related Conditions.

Sometimes called the NESAR data.

That was about ten years later, and again had about 40,000

or so individuals in the same population, the United States population.

You can see here, I've shown you the prevalence by age.

And what you can see is, actually that it

seems that the prevalence of depression has gone up.

Between 1991, 92 and 2001.

And in what, in ten years, so for 18 to 29 year old men, that's gone 4.8 to 7.3.

For 18 to 29 year old women, it's gone from 7.1 to 12.6.

So those are not trivial differences.

Those are large differences, in only a ten year period.

So that is lending support to the idea that

we are entering this somehow strange age of melancholy.

Now let's consider why that might be happening in contemporary society.

And I'm going to show you several pieces of data

from a wonderful book by Putnam called Bowling Alone.

It was done in 2000, and explaining the title

is important.

what Putnam studied was the decline of social connections.

From about, actually, this slide is from 75 to 2000, but

he studied it over about 50 years, from 1950 to 2000.

And, in 1950 and in 2000 there were at both points in time there

were bowling alleys, and at both points in time they were full of people.

But, what he discovered is, in 1950 the bowlers were all in leagues and groups and

clubs and networks. But, in 2000 there were people just

bowling by themselves, so that's the sort of sad note in his title, Bowling Alone.

And he tabulates many, many different sources of data.

They all had the same quality to them.

For example, how many times did you entertain at home last year?

You can see there are data from 1975 through 2000.

That's the blue squares there.

And it goes from about 14 times last

year you entertained at home down to about eight.

And how many times did you play cards last year?

My parents used to play cards all the time with their friends.

I hardly ever do, that's the red line going down from

about 16 times to about 8 times, that's the left axis.

And then how many people agree our whole family eats

dinner together.

And that's the green line about 50%, and this

is the right axis now for the, the left hand

most part of the green line there and it

goes down to about 30% on the right hand side.

And likewise, I went to the home of friends last

week has dropped from about 40% to less than 20%.

So this is suggestive.

Let me just show you a few more examples of this.

PTA members per 100 families for people with a

kid under 18 joining the PTA is a possibility.

But you can see, less people are doing it in 2000 than in 1960.

And you can also look at the union membership.

As a percent of the non-farm labor force union membership.

Again this is a community association. It's gone down.

So now you're suppose to be thinking about

social supports, as we think about this trend.

What, what's the connection between social supports and depressive disorder.

Here's another one.

A divorce rate.

Divorce rate is a little bit about social support.

It's also about life stress, and the divorce rate has risen from 1960 to

2000 from 10 per 1,000 to 20. It's practically doubled.

So check out this one, this is social connectedness and a happiness index.

Here's the happiness index questions, I wish I could leave

my present life and do something different, that's an unhappy person.

Or a happy person is I'm very satisfied with

the way things are going in my life these days.

Or the bottom, I am much happier now than I was ever before.

And this is in Bowling Alone, again.

You can see for people who volunteered, who attended club

meetings, who entertained at home or attended church, the more they

do those things, the more times per year

they do those community things, those social things,

the less likely they are to be unhappy, and the more likely they are to be happy.

Now, happiness is not the same as depression, but it's still suggestive.

And here's another community association, Decline in

Weekly Religious Attendance in America from 1960.

We all know this has happened.

And we can think about in I suppose theological

terms, but it has to do with social connectedness as well.

And it's in decline.

So, all these things are in decline.

And we wonder if our social life is changing so much.

Does that affect our risk for depressive disorder?

We know that stress is related to depressive disorder.

Social support is related to depressive disorder.

Our social life is changing.

Maybe that's raising our risk for depressive disorder.

Now we also have to consider other changes

in our social life that may be related to risk for depressive disorder.

And maybe substituting or, you know, counteracting these changes.

And so we have text messaging and the Internet, and the web.

We can form Facebook friends and so forth. And so I made this up.

You know, we have two people talking on text messages.

Wuzzup?

Where are you?

the mall. Did you see Jenny with K?

K, oh my god. Why?

Jenny is with R, not K.

Oh. See you later, okay.

So here is a commentary on social life, Jenny and K and R, you know.

But these people are being friends.

They aren't visiting in the home, they aren't playing cards in the home.

But they are making social contact.

So we have to wonder if the new social technologies,

to some extent, are replacing old forms of social life.

We just can't notice it, it's less face to face interaction.

More digital interaction I suppose you'd say.

And we actually don't know the extent to which

those social technologies are replacing our face to face communication.

And how they affect the risk for depressive disorder,

but it is still a possibility that they are.

So now, I think I showed

you earlier that when we want to study

the force of morbidity we need an incidence study.

And to do an incidence study, we have to have a

perspective study where we interview people at one point in time.

And we determine that there are some people who are at

risk for the first onset of a given disease or disorder.

And, of those risk set people, what percentage of them, or

what rate of onset is there.

And so to do an incidence study, and study incidence over time, is very difficult.

In fact there are only three studies in the world that I can

find that are based in the population that actually let us do this.

This chart has them placed there, so we have the study in Lundby, Sweden.

you can see the, the left hand most black diamond there which is about

2.3 on the left axis there.

The annual incidence is about 2.3 per 1,000 per year.

And that is a study conducted between 1947 when the

risks that was established in a small town Sweden in 1957.

And the follow up when the incidence rate

was established, so that's the left most black diamond.

And then from 1957 they established another

risk set in that same population, and

followed up in 1975 that's the second life time.

And so actually to make a trend in depressive disorder, you need

three points in time, each incidence rate requires two points in time.

You can see.

And this is the study in Lundby is the one led by Ula Hagnel who wrote the

article after looking at this data, these black

diamonds, saying are we entering an age of melancholy?

Because it seemed to show that

the rate of incidence of depressive disorder went from 2.3 to 4.5.

And so we only have two other studies that can do this.

One of these studies is in Atlantic, Canada.

And again this is a three wave study.

They start in 1952 and then followed up through 1970.

And looked and established all the new cases of depressive disorder, divided by

the personal years of observation, and then you've got a rate of 4.5

per 1,000.

Then they did this again from 1970 to 1992, 22 years.

And their rate was more like 3.8 per 1,000.

Finally the third study is the Baltimore ECA project which I was leading, and there

was a study in 1981 to 1993, and the incidence rate was three per 1,000.

You can see these incidence rates are not too far off,

even though they're in very different areas.

The incidence of depression is kind of close.

And when we follow that group through 2004, the new cases

of depression produced an incidence rate of about two per 1,000.

So, we have Hagnel's work in Lundby, Sweden, with a sharp rise from 47 to 75,

and then the study in Atlantic Canada by Jane Murphy in Collaborators

showing a decline, and then the Baltimore, Maryland study showing decline.

So we conclude, the data does not suggest a rise

in the incidence of depressive disorder with the passage of time.

But now let's consider the prevalence of depressive disorder as compared to the

incidence of depressive disorder, and this is data from the Baltimore ECA cohort.

I'm going to show you this data

categorized by age.

So in 1981, for example, the prevalence was 2.1.

And between 1981 and 1993 the incidence of people

who were aged 18 to 29 in 81 was 1.5.

Now we can't study the prevalence in 1993 because there isn't anybody aged

18 to 29 in our cohort any longer right, they've all aged out of that cohort and

entered the 30 to 44 cohort. What you can see is if you compare the

incidence apparently declined in the age cohorts from 3.8 to 1.5, for example.

And this is for males.

But the prevalence has, you might say, it has risen.

You would have trouble. It seems to have declined a little bit.

And then risen again.

We don't see evidence of a strong trend in these data for males.

Now let's look at the females.

This is the same type of chart except it's for the females.

You can see that the incidence of depression, take the 30

to 44 year old females, those, these are females in 1981.

Who are at risk for nuance and a

depressive disorder, and were followed from 1981 to 1993.

That incidence was 6.5, and they were followed again from 1993 to 2004.

And since its 4.5 so

incidence is declining if anything, but now look at the prevalence data.

And this is especially from 1993 to 2004, we see

a relatively strong increase in the prevalence of depressive disorder.

So, we're talking in 1993 for middle aged females 30

to 44, prevalence in 1993 of 5.1 is more than doubled to 11.0, and we have

3.3 increasing to 4.3. So now, what this seems

to indicate is, people who are becoming depressed are staying depressed longer.

That's what makes the prevalence rate go up in this case.

That's what this data seems to suggest.

If we're entering an age of melancholy, it's not

that we're having more new cases of depressive disorder.

Is that the cases that get started, they stayed depressed longer.

This may be connected to the decline in social

connectedness that we saw in Putnam's work Bowling Alone.

So here's the idea, cohorts born, this is

our original slide on trends, and births after 1935.

What we're proposing, and again, this is a, I would say not proof.

It's a suggestion that women born in later cohorts,

after 1935, are likely to have more chronicity of depressive

disorder if they become depressed. So let's review this.

This is getting a little more complicated, I think

you can see, that's what makes psychiatric epidemiology so interesting.

Doesn't look like we're entering an age of melancholy.

And we do seem to see large trends and higher

prevalence in a later born cohorts, but at least partly

that is the the way that people recall when they respond to a survey.

It doesn't seem there's been a rise in

incidence in the past half century, but there

has been a rise in prevalence for women and cohorts born from 1935 to about 1954.

Here's what may be interesting, women born between

1935 and 1954, they entered work and married

life between 1955 and 1974.

They have greater chronicity of depressive disorder, we think.

It's also true that this is what we term a cohort effect.

They entered work and married life in a very special time,

that is the era of title nine and the women's liberation movement.

And the possibility that women had a wider

range of choices for their life trajectories than earlier.

They could enter the workforce with a lot

more possibility of getting a good job than before.

And this may have complicated their life, and it

may have complicated the way they formed social supports.

And may have raised the possibility that

they will stay depressed if they get depressed.

So, that's the final conclusion.

Trends in social supports and divorce may be connected

to trends in the chronicity of depression for women.

So,

that's the end of this lecture.

This is a complicated lecture on risk factors for depressive disorder.

And, of course, there are many other risk factors, but I've

reviewed the ones that seem to be the most potent and interesting.

And the next lecture doctor Moshtabi will discuss with you the system of

treatment for depression and the degree to which people actually get that treatment.

探索我們的目錄

免費加入並獲得個性化推薦、更新和優惠。

Coursera 致力於普及全世界最好的教育，它與全球一流大學和機構合作提供在線課程。

© 2019 Coursera Inc. 保留所有權利。

通過 App Store 下載

通過 Google Play 獲取

Coursera
關於
管理團隊
工作機會
目錄
證書
MasterTrack™ 證書
學位
企業版
政府版

社區
Learners
Partners
Developers
Beta Testers
Translators

連接
博客
Facebook
領英
Twitter
Instagram
YouTube
技術博客

更多
條款
隱私
幫助
內容訪問
媒體
聯繫我們
目錄
附屬公司