Evidence that intergenerational income mobility is the strongest predictor of drug overdose deaths in U. S. Midwest counties

Key Question Being Explored in This Insight:

What were the key findings of the study analyzing opioid overdose deaths in 1,055 U.S. Midwest counties, and how did socioeconomic factors, such as intergenerational income mobility, opioid prescription rates, and social capital, contribute to the variance in overdose death rates from 2006 to 2021?

View the page of: Gene M. Heyman

Institution: Boston College

Explore more insights in:

Cognitive Psychology Psychology Social Sciences

Media Literacy Toolkit

Faculti's Media Literacy Toolkit helps viewers critically engage with academic insights by analyzing the research context, identifying perspectives, and encouraging thoughtful evaluation.

Critical Questions to Consider ▼

What assumptions does the research make?
Are there alternative perspectives not explored?
What are the limitations of the research method?

Bias and Perspective Awareness ▼

This research comes from Boston College . Reflect on how the institution's academic focus and research partnerships may shape the questions being explored.

Further Reading ▼

Read the Study

Image courtesy of interviewee. February 26, 2025

Log-in or Sign-up to Faculti
Currently viewing this subject insight as a guest. You have insight(s) remaining for this month. Login to view 8000+ figures on the platform.

Generating Workspace...

Export as PDF

This video discusses research findings related to opioid addiction and overdose epidemic in the United States. It highlights the complexity of factors contributing to drug addiction beyond just pharmaceutical companies and physicians, such as social and economic conditions. The research indicates a correlation between income mobility and opioid overdose rates, suggesting that people may turn to drugs as a response to feeling dissatisfied with their life prospects. This video also emphasizes the importance of individual stories and recovery, with a call for greater focus on supporting individuals in overcoming addiction.

Language:

Export as PDF

Export as PDF
1. **Initial Focus on Pharmaceutical Companies and Physicians**: The video starts by highlighting the initial focus on pharmaceutical companies and physicians in relation to the opioid epidemic. There was a perception that these entities had misled the public into perceiving opioids as safe, resulting in a supply-side explanation of the epidemic.

2. **Complexity Beyond Media Portrayal**: The author delves into the complexity of motivations behind drug use and opioid addiction beyond what was portrayed in the media. They express interest in understanding the multifaceted conditions that motivate individuals to use opioids, acknowledging that available data suggests a more complex picture.

3. **Impact of Social and Economic Factors**: Through state-level analysis, the video reveals substantial differences in overdose rates across states, with social capital emerging as a significant non-prescription factor influencing overdose deaths. This measure of social capital, indicative of cooperation and mutual support in communities, was found to have a strong explanatory power in predicting overdose rates.

4. **Role of Income Mobility**: The video highlights income mobility as a significant predictor of opioid overdose rates, with intergenerational income mobility being identified as a key factor. Despite shifts in opioid prescription rates and the type of drugs involved in overdoses, income mobility remained consistently correlated with overdose deaths, suggesting its critical role in influencing susceptibility to drug use.

5. **Need for Individualized Approaches and Emphasis on Recovery**: The video suggests the necessity of more individualized analyses to understand the complex interplay of social, economic, and psychological factors contributing to drug addiction. Additionally, it promotes a shift towards emphasizing individuals' capacity for recovery and the prevalence of successful outcomes in overcoming drug dependency, calling for a focus on stories of recovery as part of addressing the opioid epidemic.

Export as PDF
Key Quotes from the Text:

1. "The focus was largely on the pharmaceutical companies and on physicians. And the claim was that the pharmaceutical companies had kind of seduced the public into thinking that these drugs were safe."

2. "I was curious what might account for people all of a sudden wanting to use opioids to such an extent, and even when they were so dangerous, in terms of like overdoses."

3. "We did a state level analysis and found... large differences from state to state in the United States in overdose rates."

4. "In America, we found that social capital... was a very important measure and the most important non-prescription measure."

5. "The rate of overdose increases until just this last year has increased at exactly the same rate they were before the decrease in opioid prescription rates."

6. "Income mobility was by far the strongest predictor... even though prescription access to opioids decreased."

7. "What was tying together all the changes in the overdose epidemic was this one factor, income mobility."

8. "Income mobility has decreased in the United States... income mobility has decreased in the United States. So in 1940, 90% of the children had incomes that were higher than their parents. Now it's about 50% in the United States."

9. "Income mobility, even though overall, 50% of the population has higher incomes than their parents varies a great deal from one geographic region to another."

10. "People, if you're using drugs every day, or a lot, it's hard to... conventional life is not providing enough rewards clearly, so that drugs become a better choice, at least in the short term."

11. "We really need a more individual analysis... in terms of applications, I think there should be more emphasis on people's ability to recover."

12. "One of the responses to the opioid epidemic should be focusing on the large numbers of people who have stopped using the drugs."

13. "Recovery often comes from people belonging to organizations like alcoholics anonymous."

X-TIMESTAMP-MAP=LOCAL:00:00:00.000,MPEGTS:90000

1
00:00:00.035 --> 00:00:04.885
Context was, at the beginning, you know,

2
00:00:04.945 --> 00:00:07.965
the, of the opioid epidemic,

3
00:00:08.395 --> 00:00:10.845
drug overdose epidemic in the United States.

4
00:00:11.585 --> 00:00:15.445
Uh, the focus was largely on

5
00:00:16.065 --> 00:00:20.205
the pharmaceutical companies and on physicians.

6
00:00:20.345 --> 00:00:23.365
And the claim was that the pharmac pharmaceutical

7
00:00:23.925 --> 00:00:28.165
companies had kind of seduced the public into thinking

8
00:00:28.165 --> 00:00:29.845
that these drugs were safe

9
00:00:30.545 --> 00:00:33.485
and that the physicians had over-prescribed them.

10
00:00:33.945 --> 00:00:38.845
So it was kind of a, a supply side, explanation.

11
00:00:39.345 --> 00:00:44.245
Uh, and the individual users were seen as, as you know, as,

12
00:00:44.465 --> 00:00:49.205
as, as victims, who, have felt prey to these,

13
00:00:49.385 --> 00:00:52.405
predatory, companies

14
00:00:52.745 --> 00:00:56.645
and these, physicians who were just careless.

15
00:00:57.425 --> 00:01:00.925
And I've been working on, drug use

16
00:01:00.945 --> 00:01:02.645
and addiction for a number of years,

17
00:01:03.385 --> 00:01:05.605
and it's motivated behavior.

18
00:01:05.865 --> 00:01:09.925
You know, people, have motivations to take drugs there.

19
00:01:09.925 --> 00:01:12.885
They, there are reasons why people take drugs and or,

20
00:01:13.265 --> 00:01:14.285
and, you know,

21
00:01:14.285 --> 00:01:16.525
and there's lots of evidence to su support

22
00:01:16.555 --> 00:01:19.965
that there was some, it was a much more complicated picture

23
00:01:20.715 --> 00:01:23.245
than the media,

24
00:01:23.625 --> 00:01:27.405
and many people in the field were presenting.

25
00:01:27.905 --> 00:01:30.365
Um, for example,

26
00:01:31.645 --> 00:01:35.445
millions upon millions of people are prescribed opioids

27
00:01:35.545 --> 00:01:39.845
for pain, for surgeries and injuries.

28
00:01:40.225 --> 00:01:41.725
And they don't become addicted.

29
00:01:41.985 --> 00:01:43.565
You know, they don't become,

30
00:01:43.635 --> 00:01:45.485
it's really only a small percentage.

31
00:01:46.145 --> 00:01:48.165
Um, so there was much more to it.

32
00:01:48.465 --> 00:01:52.285
Uh, and so I was interested in what were the conditions

33
00:01:53.195 --> 00:01:57.445
that were motivating this besides the availability of drugs.

34
00:01:57.545 --> 00:02:01.085
And I don't wanna, I don't want to,

35
00:02:02.485 --> 00:02:05.165
undercut how important availability is,

36
00:02:05.705 --> 00:02:07.645
but it isn't the only factor.

37
00:02:08.345 --> 00:02:10.085
And so that was my reasons.

38
00:02:10.365 --> 00:02:15.285
I was curious what might account for people all of a sudden

39
00:02:15.915 --> 00:02:19.245
wanting to use opioids to such an extent,

40
00:02:19.385 --> 00:02:24.165
and even when they were so, dangerous,

41
00:02:24.625 --> 00:02:25.965
in terms of like overdoses.

42
00:02:25.965 --> 00:02:28.085
So that was the motivation, is what else?

43
00:02:28.465 --> 00:02:31.925
You know, it had to be a more complex picture than the media

44
00:02:32.185 --> 00:02:33.925
and many experts were presenting.

45
00:02:34.345 --> 00:02:37.005
And so I wanted to know what were some of the conditions.

46
00:02:38.625 --> 00:02:39.865
I, I wasn't able

47
00:02:40.045 --> 00:02:44.305
to find out about individual psychological factors so much,

48
00:02:44.305 --> 00:02:45.625
which are clearly important.

49
00:02:46.665 --> 00:02:49.105
I mean, terribly important, but I could,

50
00:02:49.205 --> 00:02:51.985
but the data available for finding out about social

51
00:02:52.205 --> 00:02:54.385
and economic conditions were available

52
00:02:55.005 --> 00:02:57.305
was just the more complic clearly had

53
00:02:57.305 --> 00:02:58.745
to be a more complicated picture.

54
00:02:59.345 --> 00:03:01.685
And so we looked at what we could find,

55
00:03:02.185 --> 00:03:05.245
and what we could find was in the social economic world,

56
00:03:05.245 --> 00:03:06.325
there were data available.

57
00:03:06.785 --> 00:03:08.445
We did a state level analysis

58
00:03:08.625 --> 00:03:11.845
and found, you know, that there are large differences

59
00:03:12.595 --> 00:03:15.765
from state to state in the United States in overdose rates.

60
00:03:16.675 --> 00:03:17.955
I mean, like 20 to one.

61
00:03:18.035 --> 00:03:20.475
I mean, just really remarkably large differences.

62
00:03:21.215 --> 00:03:25.235
And again, there were data available at the state level

63
00:03:25.695 --> 00:03:28.835
to find out what could be driving this.

64
00:03:29.295 --> 00:03:30.595
And so we did a state level

65
00:03:30.615 --> 00:03:32.195
and we did the state level analysis.

66
00:03:32.195 --> 00:03:36.235
We also, I always included drug availability,

67
00:03:36.645 --> 00:03:40.735
which would be, in that, in the case

68
00:03:40.875 --> 00:03:44.415
of opioid prescription rates and opioid.

69
00:03:44.635 --> 00:03:47.455
And we were able to translate those, so

70
00:03:47.455 --> 00:03:50.495
that we had them in terms of, commensurable units

71
00:03:50.595 --> 00:03:52.415
of morphine milligrams.

72
00:03:52.415 --> 00:03:54.215
So we knew about prescription rates,

73
00:03:54.545 --> 00:03:56.975
which was everybody talked about was the driver.

74
00:03:57.875 --> 00:03:59.575
Um, so we had that data,

75
00:03:59.675 --> 00:04:03.255
but then we also could get, data on economic

76
00:04:03.395 --> 00:04:04.495
and social conditions.

77
00:04:04.715 --> 00:04:06.815
And it turned out at the state level,

78
00:04:07.155 --> 00:04:10.375
and looking at all the 48, continental states

79
00:04:10.995 --> 00:04:15.615
in America, north America, we, found that social,

80
00:04:16.035 --> 00:04:19.375
something called social capital, which is a measure

81
00:04:20.005 --> 00:04:23.855
that is of interest to sociologists and economists,

82
00:04:24.235 --> 00:04:27.415
and it's apo, was this was a very important measure.

83
00:04:27.955 --> 00:04:32.215
Um, and it was the most important non-prescription measure.

84
00:04:32.835 --> 00:04:34.815
And it account, and it, it, it

85
00:04:35.335 --> 00:04:38.855
provided the, its explanatory power was, is quite large,

86
00:04:38.855 --> 00:04:40.615
almost as large as prescriptions.

87
00:04:40.755 --> 00:04:43.935
And then I gradually realized that states are pretty

88
00:04:44.825 --> 00:04:46.695
gross units of analysis.

89
00:04:46.795 --> 00:04:49.135
You know, there are millions of people in each state.

90
00:04:49.955 --> 00:04:54.495
Uh, and so I tried to, next I looked for a more molecular

91
00:04:55.395 --> 00:04:57.615
county, but I mean, a county is gross stoop,

92
00:04:58.155 --> 00:04:59.895
but not as gross as a state.

93
00:05:00.635 --> 00:05:04.295
Uh, and so we did the, the state first

94
00:05:04.875 --> 00:05:08.055
and the, there, it, you know,

95
00:05:08.375 --> 00:05:11.415
I think everybody would share the intuition that it has

96
00:05:11.415 --> 00:05:13.895
to be more complicated than it was being presented.

97
00:05:14.395 --> 00:05:16.575
And it turned out that this social capital,

98
00:05:16.945 --> 00:05:21.305
which is supposedly measures the degree to which people

99
00:05:22.585 --> 00:05:26.205
in a geographic area are willing to cooperate

100
00:05:26.435 --> 00:05:30.525
with each other for mutual ends,

101
00:05:30.585 --> 00:05:32.645
for mutual needs, mutual ends.

102
00:05:32.945 --> 00:05:35.645
And so they use measures like the, the number

103
00:05:35.665 --> 00:05:38.645
of voluntary organizations, you know,

104
00:05:38.645 --> 00:05:42.605
nonprofit organization as in geographic area, willingness,

105
00:05:42.905 --> 00:05:47.085
the degree to which people vote in elections, kind of,

106
00:05:47.105 --> 00:05:50.765
measures of social cohesion, social interest,

107
00:05:50.765 --> 00:05:51.885
social motivation.

108
00:05:52.345 --> 00:05:54.805
And that turned out to be a very strong predictor.

109
00:05:55.665 --> 00:06:00.005
States where there was greater social capital,

110
00:06:00.145 --> 00:06:04.565
had much lower rates of overdose deaths than states with,

111
00:06:04.665 --> 00:06:06.445
low social capital.

112
00:06:06.595 --> 00:06:09.365
Initially. The literature was,

113
00:06:09.905 --> 00:06:12.165
and this is what the media picked up on the literature,

114
00:06:12.165 --> 00:06:15.445
was emphasizing prescription rates and physicians.

115
00:06:16.425 --> 00:06:20.325
And there are lots of articles about, pharmaceutical

116
00:06:20.925 --> 00:06:24.005
companies, driving this,

117
00:06:24.345 --> 00:06:27.125
and, drug mills, you know, physicians

118
00:06:27.185 --> 00:06:28.845
who make their living by cell

119
00:06:28.945 --> 00:06:31.245
by prescribe prescribing opioids.

120
00:06:32.025 --> 00:06:33.805
Um, there were some, there,

121
00:06:33.805 --> 00:06:37.405
there was always some data on the social economic drivers.

122
00:06:38.065 --> 00:06:42.975
Um, again, people choose those in part

123
00:06:42.975 --> 00:06:44.135
because they're available.

124
00:06:44.515 --> 00:06:48.575
So the literature, is now, what people have come

125
00:06:48.575 --> 00:06:51.695
to realize and what's in the most recent paper is that

126
00:06:52.635 --> 00:06:55.655
opioid prescription rates have greatly decreased.

127
00:06:58.065 --> 00:07:01.805
But what we document for the Midwest counties is that

128
00:07:02.505 --> 00:07:06.365
the rate of overdose increases until just this last year

129
00:07:06.985 --> 00:07:11.205
has increased at exactly the same rate they were before.

130
00:07:11.505 --> 00:07:13.965
The decrease in opioid prescription rates,

131
00:07:14.395 --> 00:07:18.205
that is drug overdose rates did not decline.

132
00:07:18.265 --> 00:07:20.805
And this is a very interesting observation.

133
00:07:21.465 --> 00:07:25.285
Did not decline even though opioid prescription rates have

134
00:07:25.325 --> 00:07:26.405
markedly declined.

135
00:07:26.945 --> 00:07:31.085
Opioid overdose, drug overdose deaths,

136
00:07:31.515 --> 00:07:34.045
kept up at exactly the same rate.

137
00:07:34.705 --> 00:07:38.805
Uh, they, in 2013 from 20,

138
00:07:38.805 --> 00:07:42.045
the data we've tracked is from 1999

139
00:07:42.145 --> 00:07:44.725
or 20, 2001,

140
00:07:44.745 --> 00:07:47.365
and the most recent paper 2006

141
00:07:47.825 --> 00:07:49.925
to 2021.

142
00:07:50.865 --> 00:07:54.805
Um, over that stretch of time, there was a

143
00:07:55.325 --> 00:08:00.125
slightly exponential increase in overdose deaths in Midwest

144
00:08:00.405 --> 00:08:04.085
counties at the same exact rate over for year

145
00:08:04.085 --> 00:08:05.245
after year after year.

146
00:08:05.985 --> 00:08:07.165
In contrast,

147
00:08:07.985 --> 00:08:12.845
opioid prescription rates increased from 2006 to 2012,

148
00:08:13.385 --> 00:08:16.725
but then in 2013, they started decline.

149
00:08:16.785 --> 00:08:20.285
And by 2018, they were lower than they were in 2006.

150
00:08:21.455 --> 00:08:24.835
So, the two rates, opioid prescription rates

151
00:08:25.015 --> 00:08:29.235
and overdose deaths have shifted in the opposite direction,

152
00:08:30.015 --> 00:08:32.475
for about the last 12, you know, last,

153
00:08:33.575 --> 00:08:35.155
12 years.

154
00:08:35.935 --> 00:08:40.475
Um, and, so opioid prescription rates,

155
00:08:40.775 --> 00:08:43.555
may be even less important in our analyses.

156
00:08:43.735 --> 00:08:45.355
For the Midwest, they're less important.

157
00:08:45.355 --> 00:08:47.435
They're like the third most important factor.

158
00:08:48.015 --> 00:08:51.955
Uh, we use something called multiple regression, which tries

159
00:08:51.975 --> 00:08:55.885
to measure, the amount

160
00:08:56.785 --> 00:09:00.165
of variation between counties, which can be described

161
00:09:00.225 --> 00:09:01.405
by a particular factor.

162
00:09:02.145 --> 00:09:06.065
And what we found was this measure,

163
00:09:06.445 --> 00:09:09.305
intergenerational income mobility was

164
00:09:09.405 --> 00:09:11.345
by far the strongest predictor.

165
00:09:11.855 --> 00:09:15.025
Even, even though prescription a a access

166
00:09:15.025 --> 00:09:16.465
to opioids decreased.

167
00:09:16.525 --> 00:09:20.505
And even though the particular drug, the particular type

168
00:09:20.505 --> 00:09:23.025
of drug that was accounting for most

169
00:09:23.705 --> 00:09:27.225
overdoses changed from over, there were three kind of,

170
00:09:27.435 --> 00:09:30.665
first it was opioid prescriptions, and it was heroin,

171
00:09:30.885 --> 00:09:34.025
and it was fentanyl, even though the drugs were changing,

172
00:09:34.215 --> 00:09:38.265
even though the access to legal opioids was changing, the

173
00:09:38.975 --> 00:09:41.545
correlation between overdose deaths

174
00:09:41.805 --> 00:09:45.665
and, income mobility remained the same.

175
00:09:46.815 --> 00:09:50.955
And so that was a, you know, very powerful evidence that

176
00:09:51.025 --> 00:09:55.015
what was tying together all the, you know, the,

177
00:09:55.155 --> 00:09:59.215
the changes in the overdose epidemic was this one factor,

178
00:09:59.715 --> 00:10:01.015
income mobility.

179
00:10:01.675 --> 00:10:04.895
Now, what, what income mobility is actually measuring is

180
00:10:04.895 --> 00:10:09.055
kind of mis, you know, mysterious, you know, it, it has

181
00:10:09.055 --> 00:10:12.855
to have a psychological component, which we we're not,

182
00:10:13.235 --> 00:10:14.455
we don't have access to.

183
00:10:14.625 --> 00:10:17.815
We've begun to think about what it is measuring,

184
00:10:17.835 --> 00:10:19.295
what's interesting about it.

185
00:10:20.435 --> 00:10:22.615
One of the things interesting about it, I just started,

186
00:10:24.185 --> 00:10:28.085
we looked, the measure we used in our research

187
00:10:29.105 --> 00:10:33.565
was income mobility for those individuals whose parents

188
00:10:34.275 --> 00:10:36.245
were in the 25th percentile.

189
00:10:37.155 --> 00:10:39.805
That is, they were in the middle of the bottom half

190
00:10:39.865 --> 00:10:41.365
of the income distribution.

191
00:10:41.825 --> 00:10:42.925
So that was our focus.

192
00:10:44.055 --> 00:10:45.475
We have access to data

193
00:10:45.655 --> 00:10:49.355
for PA individuals whose parents are in the 50th,

194
00:10:50.155 --> 00:10:54.895
in the first, in the 75th, and in the 99th,

195
00:10:55.035 --> 00:10:57.535
or, and I've started to look at that.

196
00:10:57.755 --> 00:11:02.075
So everybody's income mobility

197
00:11:02.135 --> 00:11:03.315
is different than their parents.

198
00:11:04.255 --> 00:11:08.195
For the people in the 75th, their income mobil,

199
00:11:08.695 --> 00:11:11.755
you look at the children's,

200
00:11:12.355 --> 00:11:15.075
parents' place in the income distribution

201
00:11:15.095 --> 00:11:18.475
of the United States when they're 16 years old,

202
00:11:20.375 --> 00:11:21.795
and then,

203
00:11:22.095 --> 00:11:24.675
you measure it again when they're in their mid thirties,

204
00:11:24.925 --> 00:11:28.475
their income re their income,

205
00:11:28.475 --> 00:11:31.955
their place in the income distribution when they themselves

206
00:11:32.695 --> 00:11:33.795
are in their mid thirties.

207
00:11:34.535 --> 00:11:35.675
So we're comparing it

208
00:11:35.695 --> 00:11:38.875
to their families when they're 16 about,

209
00:11:40.435 --> 00:11:42.335
and then 20 years later when they

210
00:11:42.335 --> 00:11:43.615
themselves are in their thirties.

211
00:11:45.075 --> 00:11:50.055
And it turns out that people from the lower half,

212
00:11:50.105 --> 00:11:52.455
their income rank is typically

213
00:11:52.675 --> 00:11:55.685
Higher than their parents and people,

214
00:11:55.905 --> 00:11:56.965
and it sort of makes sense.

215
00:11:57.185 --> 00:11:58.725
And people in the upper half,

216
00:11:58.735 --> 00:12:02.575
their income rank is lower than their parents.

217
00:12:02.835 --> 00:12:07.445
If you look at people's, how people change relative

218
00:12:07.445 --> 00:12:09.565
to their parents,

219
00:12:09.905 --> 00:12:12.645
and you compare it, his look at it historically,

220
00:12:13.055 --> 00:12:14.725
since people began measuring this,

221
00:12:14.725 --> 00:12:17.925
which was in 1940 in the United States to the present

222
00:12:19.145 --> 00:12:23.045
income mobility has decreased in the United States.

223
00:12:24.105 --> 00:12:27.565
So in 1940, 90% of the children,

224
00:12:27.675 --> 00:12:31.005
born in 1940 had incomes

225
00:12:31.005 --> 00:12:33.635
that were higher than their parents.

226
00:12:34.655 --> 00:12:38.125
Now it's about 50% in the United States.

227
00:12:38.265 --> 00:12:40.765
So it's still, you know, there's still income mobility,

228
00:12:41.655 --> 00:12:43.155
but it's relatively higher.

229
00:12:43.535 --> 00:12:46.435
And what, turns out that one

230
00:12:46.435 --> 00:12:48.675
of the things about it in the United States,

231
00:12:48.815 --> 00:12:53.115
and I'm sure this is true elsewhere to varying degrees, is

232
00:12:53.115 --> 00:12:57.475
that income mobility, even though overall,

233
00:12:58.255 --> 00:12:59.355
50%

234
00:12:59.355 --> 00:13:03.555
of the population has higher incomes than their parents,

235
00:13:03.895 --> 00:13:05.115
in terms of the distribution.

236
00:13:05.115 --> 00:13:06.755
And so it's a relative measure.

237
00:13:08.285 --> 00:13:10.985
It varies a great deal,

238
00:13:11.175 --> 00:13:13.705
from one geographic region to another.

239
00:13:14.125 --> 00:13:16.025
So there are counties where lots

240
00:13:16.025 --> 00:13:19.345
and lots of people have higher incomes than their parents

241
00:13:19.485 --> 00:13:21.705
and their counties, where lots

242
00:13:21.705 --> 00:13:25.185
and lots of people have lower incomes than their parents.

243
00:13:25.815 --> 00:13:30.065
What we find is this very strong correlation, such

244
00:13:30.065 --> 00:13:33.105
that counties where it's higher, they're low overdose rates,

245
00:13:33.485 --> 00:13:36.585
you know, in, in the same states,

246
00:13:37.125 --> 00:13:39.265
and the, and vice versa.

247
00:13:39.645 --> 00:13:41.705
And so, and it's true.

248
00:13:41.965 --> 00:13:45.625
So, and if we use this measure, the,

249
00:13:45.685 --> 00:13:49.025
the predictions are strongest for, for families

250
00:13:49.205 --> 00:13:51.985
or for individuals whose parents were in the 25th.

251
00:13:52.325 --> 00:13:55.385
But if we get people in the 75th whose parents are in the

252
00:13:55.385 --> 00:13:59.835
75th and they live, they grew up in a county

253
00:14:00.565 --> 00:14:03.555
where even though their parents were in the 75th,

254
00:14:04.415 --> 00:14:05.995
but they grew up in a county

255
00:14:06.975 --> 00:14:09.555
in which there's lower income mobility,

256
00:14:09.865 --> 00:14:13.195
that county has higher, they have higher, rates

257
00:14:13.535 --> 00:14:14.995
of, of overdoses.

258
00:14:15.615 --> 00:14:19.475
So, it, it, it's a measure in which it looks like

259
00:14:19.475 --> 00:14:21.795
what people are doing is comparing themselves

260
00:14:22.095 --> 00:14:24.035
to their co, to their cohort.

261
00:14:25.055 --> 00:14:29.835
Uh, and those who, do more poorly

262
00:14:30.475 --> 00:14:34.595
relative to their cohort are those individuals who seem

263
00:14:34.595 --> 00:14:38.435
to have, be more and more susceptible to,

264
00:14:39.095 --> 00:14:43.355
the allure of what opioid drugs have to offer

265
00:14:43.735 --> 00:14:45.795
or what drugs have to offer, you know,

266
00:14:45.795 --> 00:14:48.355
changes in consciousness and so on.

267
00:14:48.715 --> 00:14:52.445
I think it, the, the natural intuition to have is that

268
00:14:53.625 --> 00:14:55.765
people looking at their life

269
00:14:56.505 --> 00:14:59.085
and looking at their prospects, so you're, you know,

270
00:14:59.085 --> 00:15:02.325
you're in your late twenties, early thirties,

271
00:15:02.705 --> 00:15:03.885
and you look around you

272
00:15:04.065 --> 00:15:05.405
and people, there are a lot of people

273
00:15:05.465 --> 00:15:09.085
who are doing better than you are and have better prospects.

274
00:15:09.225 --> 00:15:13.005
So my my intuition is that it has a lot to do

275
00:15:13.195 --> 00:15:16.565
with people thinking about, their present

276
00:15:16.745 --> 00:15:19.165
and their future and what the chances are

277
00:15:19.545 --> 00:15:21.605
of having a good life.

278
00:15:22.265 --> 00:15:25.885
Um, and, I mean, that's pretty simple,

279
00:15:26.265 --> 00:15:29.965
simple minded, but, that's what the data suggests,

280
00:15:29.965 --> 00:15:32.165
that it's something as sort of fundamental

281
00:15:32.745 --> 00:15:34.485
and basic as that.

282
00:15:35.545 --> 00:15:40.485
Now, again, if we had individual data, we could look at how

283
00:15:40.515 --> 00:15:42.445
that correlates with measures, let's say,

284
00:15:42.445 --> 00:15:44.205
of anxiety or depression.

285
00:15:44.925 --> 00:15:47.805
I mean, 'cause obviously not everybody, in fact,

286
00:15:47.835 --> 00:15:50.605
it's a small number of people from each of these counties

287
00:15:50.835 --> 00:15:52.645
that end up overdosing.

288
00:15:53.105 --> 00:15:54.805
Um, it, it really is.

289
00:15:55.045 --> 00:15:58.245
I mean, when you add it all up, it's, it's, tens

290
00:15:58.245 --> 00:16:00.325
of thousands of people, you know, maybe,

291
00:16:00.705 --> 00:16:03.205
but the United States has 300 million people,

292
00:16:03.385 --> 00:16:06.325
and so maybe last year there was something like a hundred,

293
00:16:06.325 --> 00:16:08.165
maybe a hundred thousand overdosed,

294
00:16:08.465 --> 00:16:12.205
but that's really still a relatively small percentage.

295
00:16:12.305 --> 00:16:15.245
So clearly there are individual factors,

296
00:16:15.465 --> 00:16:17.925
but, there's also a social milieu

297
00:16:17.995 --> 00:16:19.405
that is playing a big role.

298
00:16:20.025 --> 00:16:24.325
And, if you trans, if you kind of just sort of guess at

299
00:16:24.325 --> 00:16:26.445
what that social milieu is doing, is

300
00:16:26.445 --> 00:16:29.285
that people looking at their future prospects

301
00:16:29.785 --> 00:16:31.285
and their present prospects

302
00:16:31.585 --> 00:16:35.005
and not being enthusiastic about it.

303
00:16:35.125 --> 00:16:39.295
I mean, it takes, it takes a lot of, you know,

304
00:16:39.295 --> 00:16:42.855
I mean, using drugs is, is competing

305
00:16:43.525 --> 00:16:45.375
with other aspects of life.

306
00:16:45.515 --> 00:16:48.975
You know, like having a career, raising children,

307
00:16:49.515 --> 00:16:53.695
you know, it interferes with all of those, with those,

308
00:16:53.955 --> 00:16:56.375
you know, those sort of everyday it interferes

309
00:16:56.375 --> 00:16:57.935
with conventional everyday life.

310
00:16:58.005 --> 00:16:59.335
It's hard to carry, you know,

311
00:16:59.335 --> 00:17:03.375
if you're using drugs every day, are a lot, it's hard to,

312
00:17:03.635 --> 00:17:06.655
carry on a conventional life.

313
00:17:07.075 --> 00:17:11.615
Um, and so their conventional life is not providing,

314
00:17:11.635 --> 00:17:15.215
enough rewards clearly, so

315
00:17:15.215 --> 00:17:18.495
that drugs become a, a better, a better choice,

316
00:17:18.675 --> 00:17:20.135
at least in the short term.

317
00:17:20.555 --> 00:17:21.695
Uh, at least at the moment.

318
00:17:22.035 --> 00:17:24.615
My que answers have suggested is

319
00:17:24.615 --> 00:17:29.415
that we really need a more individual, analyses.

320
00:17:29.755 --> 00:17:33.415
And, and that's a very much more difficult to do than,

321
00:17:33.415 --> 00:17:34.575
that,

322
00:17:34.715 --> 00:17:37.255
but that's what's needed in terms of just the research.

323
00:17:38.235 --> 00:17:41.335
Um, and in terms of,

324
00:17:43.225 --> 00:17:48.205
of applications, which, I think there,

325
00:17:49.585 --> 00:17:52.805
and this is completely separate from my research

326
00:17:52.875 --> 00:17:54.045
that I've described here,

327
00:17:54.625 --> 00:17:58.205
but in terms of applications, I think I and,

328
00:17:58.225 --> 00:18:02.045
and, think that there should be more emphasis

329
00:18:02.945 --> 00:18:05.485
on people's ability to

330
00:18:06.705 --> 00:18:08.815
kinda restore their lives.

331
00:18:09.555 --> 00:18:12.615
Uh, and instead of, that there should be, this is,

332
00:18:12.615 --> 00:18:15.655
there's been lots of focus on restricting opioids.

333
00:18:16.105 --> 00:18:19.895
There should be focus on people on the capacity and of,

334
00:18:20.275 --> 00:18:24.415
and sort of advertising of the fact that most people

335
00:18:26.115 --> 00:18:30.095
who are having a hard time with drugs actually stop

336
00:18:30.625 --> 00:18:31.775
using the drugs,

337
00:18:32.195 --> 00:18:35.935
and that individuals are not, they have the capacity

338
00:18:36.395 --> 00:18:40.415
to restore their lives, even if they've been using drugs

339
00:18:40.675 --> 00:18:41.815
for a very long time.

340
00:18:42.635 --> 00:18:44.735
Now, that's completely, that doesn't come out

341
00:18:44.895 --> 00:18:46.375
of this research, but it comes out

342
00:18:46.375 --> 00:18:47.775
of other things that I've been doing.

343
00:18:48.195 --> 00:18:50.415
But I think that one of the responses

344
00:18:50.415 --> 00:18:53.695
to the opioid epidemic should be focusing on the large

345
00:18:53.695 --> 00:18:56.975
numbers of people who have stopped using the drugs.

346
00:18:57.315 --> 00:18:59.495
And then that should be part of the story.

347
00:18:59.835 --> 00:19:04.175
And in fact, it should be perhaps maybe the leading,

348
00:19:04.655 --> 00:19:06.415
headline, in this story.

349
00:19:06.955 --> 00:19:09.095
Um, and that comes,

350
00:19:09.285 --> 00:19:11.415
that comes from other work that I've been doing.

351
00:19:11.635 --> 00:19:14.695
But in terms of just the research finding out, and,

352
00:19:14.695 --> 00:19:17.495
and actually maybe the two go together now that I'm thinking

353
00:19:17.495 --> 00:19:18.975
of talking about it, is

354
00:19:18.975 --> 00:19:22.495
that finding out about the individual stories,

355
00:19:22.675 --> 00:19:26.255
we would also learn about recovery, as well.

356
00:19:26.915 --> 00:19:30.855
And that has been lost in this, in this discussion.

357
00:19:31.315 --> 00:19:36.175
Um, and the recovery often comes from people, belonging

358
00:19:36.195 --> 00:19:39.695
to organizations and with alcohol, like alcohol's anonymous.

359
00:19:40.355 --> 00:19:43.735
Um, but, there's a lot of these self-help groups

360
00:19:43.805 --> 00:19:47.335
that I think have been very, effective, in,

361
00:19:47.435 --> 00:19:49.375
in the drug story overall.

362
00:19:50.075 --> 00:19:53.895
So, so a, a greater focus on individuals.

363
00:19:54.235 --> 00:19:56.375
Uh, and I think that if, if he did that,

364
00:19:56.725 --> 00:20:00.095
then the recovery stories would become more prominent

365
00:20:00.315 --> 00:20:04.055
and people could see that there's hope for themselves

366
00:20:04.315 --> 00:20:05.975
and that, and,

367
00:20:05.995 --> 00:20:08.495
and I think that would be the most useful thing to come out

368
00:20:08.495 --> 00:20:09.055
of all of this.

Quick Answers

The Opioid Epidemic: Complex Motivations and Misleading Narratives

00:00:00 - Context was, at the beginning, you know, the, of the opioid epidemic, drug overdose epidemic in the United States. the focus was largely on the pharmaceutical companies and on physicians. And the claim was that the pharmac pharmaceutical companies had kind of seduced the public into thinking that these drugs were safe

00:00:33 - So it was kind of a, a supply side, explanation. and the individual users were seen as, as you know, as, as, as victims, who, have felt prey to these, predatory, companies and these, physicians who were just careless. And I've been working on, drug use

00:01:03 - and it's motivated behavior. You know, people, have motivations to take drugs there. They, there are reasons why people take drugs and or, and, you know, and there's lots of evidence to su support that there was some, it was a much more complicated picture than the media, and many people in the field were presenting. for example,

Analyzing Motivations for Opioid Use

00:01:35 - for pain, for surgeries and injuries. And they don't become addicted. You know, they don't become, it's really only a small percentage. so there was much more to it. and so I was interested in what were the conditions that were motivating this besides the availability of drugs. And I don't wanna, I don't want to,

00:02:05 - but it isn't the only factor. And so that was my reasons. I was curious what might account for people all of a sudden wanting to use opioids to such an extent, and even when they were so, dangerous, in terms of like overdoses. So that was the motivation, is what else? You know, it had to be a more complex picture than the media and many experts were presenting.

00:02:38 - I, I wasn't able to find out about individual psychological factors so much, which are clearly important. I mean, terribly important, but I could, but the data available for finding out about social and economic conditions were available was just the more complic clearly had to be a more complicated picture. And so we looked at what we could find, and what we could find was in the social economic world,

State-Level Analysis: Overdose Disparity Factors

00:03:06 - We did a state level analysis and found, you know, that there are large differences from state to state in the United States in overdose rates. I mean, like 20 to one. I mean, just really remarkably large differences. And again, there were data available at the state level to find out what could be driving this. And so we did a state level and we did the state level analysis. We also, I always included drug availability,

00:03:40 - of opioid prescription rates and opioid. And we were able to translate those, so that we had them in terms of, commensurable units of morphine milligrams. So we knew about prescription rates, which was everybody talked about was the driver. so we had that data, but then we also could get, data on economic and social conditions. And it turned out at the state level,

00:04:10 - in America, north America, we, found that social, something called social capital, which is a measure that is of interest to sociologists and economists, and it's apo, was this was a very important measure. and it was the most important non-prescription measure. And it account, and it, it, it provided the, its explanatory power was, is quite large,

State vs. County Analysis: Social Capital Impact

00:04:40 - And then I gradually realized that states are pretty gross units of analysis. You know, there are millions of people in each state. and so I tried to, next I looked for a more molecular county, but I mean, a county is gross stoop, but not as gross as a state. and so we did the, the state first and the, there, it, you know, I think everybody would share the intuition that it has

00:05:14 - And it turned out that this social capital, which is supposedly measures the degree to which people in a geographic area are willing to cooperate with each other for mutual ends, for mutual needs, mutual ends. And so they use measures like the, the number of voluntary organizations, you know, nonprofit organization as in geographic area, willingness,

00:05:47 - measures of social cohesion, social interest, social motivation. And that turned out to be a very strong predictor. States where there was greater social capital, had much lower rates of overdose deaths than states with, low social capital. Initially. The literature was, and this is what the media picked up on the literature, was emphasizing prescription rates and physicians.

The Paradox of Opioid Prescription Rates and Overdose Incidents

00:06:20 - companies, driving this, and, drug mills, you know, physicians who make their living by cell by prescribe prescribing opioids. there were some, there, there was always some data on the social economic drivers. again, people choose those in part because they're available. So the literature, is now, what people have come

00:06:52 - opioid prescription rates have greatly decreased. But what we document for the Midwest counties is that the rate of overdose increases until just this last year has increased at exactly the same rate they were before. The decrease in opioid prescription rates, that is drug overdose rates did not decline. And this is a very interesting observation.

00:07:25 - markedly declined. Opioid overdose, drug overdose deaths, kept up at exactly the same rate. they, in 2013 from 20, the data we've tracked is from 1999 or 20, 2001, and the most recent paper 2006 to 2021. over that stretch of time, there was a

Changing Trends in Opioid Prescription Rates and Overdose Deaths

00:08:00 - counties at the same exact rate over for year after year after year. In contrast, opioid prescription rates increased from 2006 to 2012, but then in 2013, they started decline. And by 2018, they were lower than they were in 2006. So, the two rates, opioid prescription rates and overdose deaths have shifted in the opposite direction,

00:08:33 - 12 years. and, so opioid prescription rates, may be even less important in our analyses. For the Midwest, they're less important. They're like the third most important factor. we use something called multiple regression, which tries to measure, the amount of variation between counties, which can be described

00:09:02 - And what we found was this measure, intergenerational income mobility was by far the strongest predictor. Even, even though prescription a a access to opioids decreased. And even though the particular drug, the particular type of drug that was accounting for most overdoses changed from over, there were three kind of, first it was opioid prescriptions, and it was heroin,

Income Mobility and Overdose Epidemic

00:09:34 - even though the access to legal opioids was changing, the correlation between overdose deaths and, income mobility remained the same. And so that was a, you know, very powerful evidence that what was tying together all the, you know, the, the changes in the overdose epidemic was this one factor, income mobility.

00:10:04 - kind of mis, you know, mysterious, you know, it, it has to have a psychological component, which we we're not, we don't have access to. We've begun to think about what it is measuring, what's interesting about it. One of the things interesting about it, I just started, we looked, the measure we used in our research was income mobility for those individuals whose parents

00:10:37 - That is, they were in the middle of the bottom half of the income distribution. So that was our focus. We have access to data for PA individuals whose parents are in the 50th, in the first, in the 75th, and in the 99th, or, and I've started to look at that. So everybody's income mobility is different than their parents. For the people in the 75th, their income mobil,

'Income Mobility Trends Over Generations'

00:11:12 - parents' place in the income distribution of the United States when they're 16 years old, and then, you measure it again when they're in their mid thirties, their income re their income, their place in the income distribution when they themselves are in their mid thirties. So we're comparing it to their families when they're 16 about,

00:11:42 - themselves are in their thirties. And it turns out that people from the lower half, their income rank is typically Higher than their parents and people, and it sort of makes sense. And people in the upper half, their income rank is lower than their parents. If you look at people's, how people change relative to their parents, and you compare it, his look at it historically,

00:12:14 - which was in 1940 in the United States to the present income mobility has decreased in the United States. So in 1940, 90% of the children, born in 1940 had incomes that were higher than their parents. Now it's about 50% in the United States. So it's still, you know, there's still income mobility, but it's relatively higher.

Income Mobility and Overdose Rates: A Geographic Correlation

00:12:46 - of the things about it in the United States, and I'm sure this is true elsewhere to varying degrees, is that income mobility, even though overall, 50% of the population has higher incomes than their parents, in terms of the distribution. And so it's a relative measure. It varies a great deal, from one geographic region to another.

00:13:16 - and lots of people have higher incomes than their parents and their counties, where lots and lots of people have lower incomes than their parents. What we find is this very strong correlation, such that counties where it's higher, they're low overdose rates, you know, in, in the same states, and the, and vice versa. And so, and it's true. So, and if we use this measure, the,

00:13:49 - or for individuals whose parents were in the 25th. But if we get people in the 75th whose parents are in the 75th and they live, they grew up in a county where even though their parents were in the 75th, but they grew up in a county in which there's lower income mobility, that county has higher, they have higher, rates of, of overdoses. So, it, it, it's a measure in which it looks like

'Drugs, Desires, and Destinies: The Peril of Peer Comparisons'

00:14:22 - to their co, to their cohort. and those who, do more poorly relative to their cohort are those individuals who seem to have, be more and more susceptible to, the allure of what opioid drugs have to offer or what drugs have to offer, you know, changes in consciousness and so on. I think it, the, the natural intuition to have is that

00:14:56 - and looking at their prospects, so you're, you know, you're in your late twenties, early thirties, and you look around you and people, there are a lot of people who are doing better than you are and have better prospects. So my my intuition is that it has a lot to do with people thinking about, their present and their future and what the chances are of having a good life. and, I mean, that's pretty simple,

00:15:29 - that it's something as sort of fundamental and basic as that. Now, again, if we had individual data, we could look at how that correlates with measures, let's say, of anxiety or depression. I mean, 'cause obviously not everybody, in fact, it's a small number of people from each of these counties that end up overdosing. it, it really is. I mean, when you add it all up, it's, it's, tens

Social Milieu and Drug Use Dynamics

00:16:00 - but the United States has 300 million people, and so maybe last year there was something like a hundred, maybe a hundred thousand overdosed, but that's really still a relatively small percentage. So clearly there are individual factors, but, there's also a social milieu that is playing a big role. And, if you trans, if you kind of just sort of guess at what that social milieu is doing, is that people looking at their future prospects

00:16:31 - and not being enthusiastic about it. I mean, it takes, it takes a lot of, you know, I mean, using drugs is, is competing with other aspects of life. You know, like having a career, raising children, you know, it interferes with all of those, with those, you know, those sort of everyday it interferes with conventional everyday life. It's hard to carry, you know, if you're using drugs every day, are a lot, it's hard to,

00:17:07 - and so their conventional life is not providing, enough rewards clearly, so that drugs become a, a better, a better choice, at least in the short term. at least at the moment. My que answers have suggested is that we really need a more individual, analyses. And, and that's a very much more difficult to do than, that,

Title: Emphasizing Recovery in the Opioid Epidemic

00:17:38 - and in terms of, of applications, which, I think there, and this is completely separate from my research that I've described here, but in terms of applications, I think I and, and, think that there should be more emphasis on people's ability to

00:18:09 - and instead of, that there should be, this is, there's been lots of focus on restricting opioids. There should be focus on people on the capacity and of, and sort of advertising of the fact that most people who are having a hard time with drugs actually stop using the drugs, and that individuals are not, they have the capacity to restore their lives, even if they've been using drugs

00:18:42 - Now, that's completely, that doesn't come out of this research, but it comes out of other things that I've been doing. But I think that one of the responses to the opioid epidemic should be focusing on the large numbers of people who have stopped using the drugs. And then that should be part of the story. And in fact, it should be perhaps maybe the leading, headline, in this story. and that comes, that comes from other work that I've been doing.

Emphasizing Recovery through Personal Stories

00:19:14 - and actually maybe the two go together now that I'm thinking of talking about it, is that finding out about the individual stories, we would also learn about recovery, as well. And that has been lost in this, in this discussion. and the recovery often comes from people, belonging to organizations and with alcohol, like alcohol's anonymous. but, there's a lot of these self-help groups

00:20:08 - in the drug story overall. So, so a, a greater focus on individuals. and I think that if, if he did that, then the recovery stories would become more prominent and people could see that there's hope for themselves and that, and, and I think that would be the most useful thing to come out of all of this.

Faculti AI can make mistakes

Research

Policy

Analysis

Data

Originals

Explainers

Podcasts

Watch the latest episode of The Exchange – the show where influential policymakers and leading researchers engage in insightful discussions shaping the future.

Episode: Brazil's Economic Resilience