Title: Internetbased secondary prevention of alcohol problems
1Internet-based secondary prevention of alcohol
problems
2Interest in different forms of self-help in a
general population sample of drinkers(Koski-Jänne
s, A. Cunningham, J. 2001)
3Use of Internet
- Fast expansion all over the world ever since the
middle of the nineties. - Wide general interest in health topics
- 80 of US Internet users have searched online
information about health topics of those 8 have
searched for information about problems with
alcohol and drugs. - In 2006 four out of five Finns between 15 and 79
have used Internet 85 of them use it weekly. - 92-99 of people under the age of 35 are current
users. - ?
- Internet has turned to a viable channel for
secondary prevention of heavy drinking.
4Merits of Internet-based services
- Highly accessible in time and space
- Affordable
- Discreet
- Appealing to people who want to avoid being
labeled as problem drinkers or alcoholics - Could be used either on ones own or as an
adjunct to other services - Easily modifiable can be kept up-to-date
5Previous research
- A number of pilot studies and feasibility trials
have reported on participants' initial
evaluations of Internet sites providing self-help
materials for problem drinkers - Cunningham, Humphries Koski-Jännes (2000a)
- Cloud and Peacock (2001)
- Lieberman (2003)
- Westrup et al. (2003)
- Linke, Brown and Wallace (2004)
- Saitz et al. (2004)
- Squires and Hester (2002 2004)
- RCTs
- Hester Delaney (1997) Kypri et al. (2004)
- participants used the Internet in the
researchers office or a health-care centre. - ? Comparability to the use of Internet at home?
6Finnish pilot
- An Internet site where people could receive
normative feedback on their drinking was launched
in the web pages of the Addiction link in June
2003. - Juomatapatesti Drinking Habit Test
- http//www.paihdelinkki.fi/testaa/juomatapatesti
- - Comprises of a brief assessment and produces
a personalized feedback summary designed to
motivate change in drinkers. - Modeled after the one created in Toronto (see
Cunningham, Humphreys, Koski-Jännes 2000) with
minor changes in its set-up.
7Drinking habit test
- 10 AUDIT questions
- 6 psychosocial consequence questions
- of drinks consumed on each day of a typical
week, or if irregular drinking pattern drinking
diary over the past month - of drinks in the last drinking occasion
- Demographic data (age, sex, weight)
- Anonymous responses were saved with an ID number
in a safe dataset by the A Clinic Foundation - Time needed for responding 5 minutes.
8Personalized normative feedback
- Comparison of the participants drinking with
other Finns (males and females) - Total of drinking days in a year
- T of drinks in a year
- Money spent on alcohol in a year
- Caloric intake in a typical drinking day
- Drinks consumed and BAC
- Probability of accidents at different levels of
BAC - A link for estimating hours to eliminate daily
intake - of hours under the influence in a year
- Personal AUDIT score and level of risk
- Upper limits of sensible drinking
9(No Transcript)
10Goals of the study
- Comparison of the volunteer study participants
with the anonymous users of this Internet service - Survey on participant reactions to the
personalized feedback. -
- Comparison of alcohol consumption and problems at
the time of screening and three months later
among the volunteer study participants.
11Subjects
- During the recruitment period (March October,
2004) - 22 536 self-assessments were completed,
- 7 of whom responded to the survey.
- The survey respondents (n1598) were asked if
they wanted to participate in the follow-up
study. - 22 (n351) agreed by providing their e-mail
address - 3 persons who responded for somebody else and 5
persons with incomplete data were excluded. - Study sample (n 343)
- Comparison sample A random sample (n 581) of
the service users
12Table 1. Background data on the study subjects
compared to a random sample of service users
during the same period __________________________
____________________________________________ S
tudy sample Service usersVariable (n
343) (n 581) p____________________________
__________________________________ women
60.6 49.2 .001 Mean (SD) age 32.6
(11.9) 30.4 (12.5) .007 Mean (SD) AUDIT 17.0
(8.7) 14.1 (8.1) .000
Mean (SD SqrtMean)
drinks in the last occasion 8.5 (6.2
2.6) 7.8 (5.7 2.7) ns.
Mean
(SD SqrtMean) drinks in a typical week
19.0 (18.5 2.7) 12.2 (14.4 3.8) .000
Mean (SD)
consequences of drinking 2.7(2.1)
1.9 (1.9) .000
_________________________________________________
_____________ SqrtMean is the mean square root
transformed value employed in the comparison for
this variable because of positive skew .
Mann-Whitney U-test.
13Table 2. Responses to the survey (n343)
- __________________________________________________
_______________ - Questions
- __________________________________________________
_______________ - How useful did you Very useful 41
- find the feedback? Somewhat 52
- Not at all 7
- Was there something Yes 40
- surprising in it? No 60
- Has your drinking been Unproblematic 38
- problematic in your Partly problematic 40
- own opinion? Clearly problematic 22
- Audit scores no problem (lt 8) 16
- at risk drinking (8-13) 20
- high risk drinking (gt13) 64
- Did the recommended Too tight 34
14Typical causes of surprise
- That I drink so much more than people on
average. - How slowly alcohol disappears from ones body.
- The number of days spent under the influence of
alcohol. - That my drinking was clearly harmful.
- The calories.
153-month follow-up
- The subjects received an e-mail asking them to
click a link to the follow-up questions. - Their ID number was attached to this link and it
was automatically saved with their responses in a
separate data set. - Questions similar to the baseline questions
(concerned the last 3 months) - Additional questions (on past 3 months)
- Have you received any treatment or participated
in self help groups? - Have you changed your drinking in any way?
- Have you recommended this site to anybody else?
- Has the reduction of prices on alcohol had any
effect on your drinking? - 78 response rate.
- No significant differences between the responders
and non-responders.
16Table 3. Data on drinking by those who responded
to the baseline and 3-month follow-up (n
269) _____________________________________________
____________________ Variable Baseline 3-month
follow-up p _____________________________________
____________________________ Mean (SD) AUDIT
score 16.6 (8.6) 13.8 (8.0) .000 Mean (SD)
AUDIT-C score 7.4 (2.5) 6.8 (2.6) .000 Mean
(SD, sqrtMean) drinks in a typical
week 19.1 (18.5, 3.8) 13.8 (14.5,
3.1) .000 Mean (SD, sqrtMean) drinks
in the in the last occasion 8.4 (6.0, 2.7)
7.1 (6.4, 2.4) .000 Mean (SD) of drinking
consequences 2.7 (2.1) 2.1
(2.0) .000 ____________________________________
_____________________________ SqrtMean is the
mean square root transformed value employed in
the comparison for this variable because of
positive skew. Wilcoxon Signed Ranks test.
17Additional findings at 3 months
- Only 4 participated in treatment/self-help
groups during past 3 months - 27 had recommended the test to others
- 34 of women and 18 of men (p .006)
- 41 had done the test again on their own.
- Perceived effects of price reductions 13
increased, 70 no change, 18 decreased. - The largest drop in AUDIT-C scores occurred in
subjects who regarded their drinking as at least
somewhat problematic in the beginning (p lt
.05).
18Summary of main results
- 93 found the feedback useful.
- Highly significant reductions in all the outcome
variables. - Women were
- more eager to participate than men (61 vs. 40 )
- more surprised by the feedback (47 vs. 30)
- recommended the test more often to others (34
vs. 18). - No sex differences in the main outcome variables.
- People who regarded their drinking as problematic
at the baseline were more successful reducers
(plt.05) than those who did not.
19Discussion
- Were the reductions caused by the personalized
feedback? -
- AUDIT scores ? in the study sample, while they ?
among the service users in general from the first
to the last quarter of the study period (13.8
?14.2) reflecting the general increase of
drinking in Finland after the price reductions of
April 2004. - The changes in drinking were similar to those in
Cunningham et al. (2005) with a smaller sample
(ns.)
20Other possible explanations
- Attempts to please the researchers?
- not probable due to subject anonymity ( nothing
to gain by lying) - Seasonal variation?
- not probable because the follow-up data were
gathered over the heavier drinking summer and
fall months. - Regression to the mean?
- Possible in populations where heavier users
self-select themselves to the study. - Controlling for this possibility requires a
randomized study.
21Research continues...
- Cunningham, Humphries, Koski-Jännes Cordingley
(2005) - After receiving web-based normative feedback the
subjects could click a link to the consent form
and sign up for the study. - Random assignment to
- web-based self-assessment or web-based
self-assessment a mailed self-help book - Combined approach gt the self-assessment alone.
- Drinking was reduced also in the self-assessment
alone about the same amount as in Finland, but
due to small sample size the effect was ns. - ?Limitation Lack of no treatment control group
22Conclusion
- Internet-based self-assessment services provide a
highly promising avenue for primary and secondary
prevention of excessive drinking. - Huge public health potential for various
underserved populations young people, those
living in rural areas, people with limitations in
transportation, etc. - Heavy drinkers seem to profit more than lighter
drinkers - Especially appealing for women.
- Some initial research has also been made with
other drug use - Internet services for substance abusers should be
further developed and studied.