Health promotion evaluation
Notes
http://homepage.ntlworld.com/gary.sturt/health/Health%
20promotion.htm
PowerPoint
http://homepage.ntlworld.com/gary.sturt1/health/Health
%20Promotion.ppt
Meyerowitz and Chaiken (1987) the effectiveness of Gain and Loss messages. Gain
message worked best. Therefore campaigns should use a positive message.
Meyerowitz and Chaiken (1987) positive message better than negative.
Secondary prevention - taboos - embarrassment.
Barriers to health promotion

Not sure what causes poor health.

Not sure how behaviours develop.

Much behaviour is learnt in the home.

For young people illness and death are a long way away.

Unrealistic optimism.
Rise (1992) used the TRA as a theoretical framework in order to study 'condom
behaviour' defined as 'a decision based upon consideration of the expected
consequences of using or not using condoms' (p. 185). A postal questionnaire about
condom use was completed by 1,172 Norwegian adolescents aged 17 to 19 years and all
non-virgins. The following variables were measured:
The intention to use condoms at the next intercourse (behavioural intention). .
Beliefs about condom use, e.g. 'condoms protect me against sexually transmitted
diseases' (behavioural beliefs).
Evaluation of behavioural beliefs and outcomes, e.g. 'How much do you fear STD?'
(values).
Significant others' evaluation of the respondent's condom use (normative beliefs).
,
Importance of significant others' evaluation (motivation to comply).
Previous/habitual condom use (prior behaviour).
Items (2) and (3) together provided a measure of attitudes towards condom use; (4) and
(5) combined assessed subjective norms. With the exception of prior behaviour, all
variables were from the TRA.
.
Rise (1992) observed that past behaviour was by far the strongest predictor of
intention to use condoms at the next intercourse. Next came subjective norm followed
by attitude. All relationships were statistically significant. Behavioural beliefs related to
pleasure and sensation (e.g. 'Condom use reduces my physical pleasure') discriminated
best between intenders and non-intenders whereas traditional risk appraisal beliefs (e.g.
'Condom use protects me against STD') did not discriminate. Among normative beliefs,
sexual partners' expectations had the best discriminatory power.
Postal questionnaire in Rise study good for sensitive information.
Good sample should generalise to western world but may not generalise to older age
groups.
Results useful in planning promotion campaigns.
Sick role behaviour could be determined in the family. A study of female college students
accessed whether they had been encouraged during adolescence to adopt the sick role for
menstruation or had seen their mothers get upset over menstruation. These women
reported more menstrual symptoms, disability and attended clinic more often in
adulthood compared with other students (Whitehead et al, 1986).
Whitehead et al 1986) ignores genetic factors - nature/nurture debate.
Learn all about the Yale model of communication.
Janis and Feshbach (1953)
Minimal fear appeal - 36% conformity (evidence based on self-report).
Strong fear appeal - 8% conformity
Illustrated lectures (15 mins) on dangers of tooth decay and need for oral hygiene.
Fear appeals must not be too frightening.
The Janis and Feshbach 1953 study could suffer because it was a one-sided message
aimed at intelligent people - see the Yale model.
More of all this will come as part of health promotion in the workplace, community and
schools.
Health Promotion in Schools
An experiment in 22 elementary schools introduced a carefully designed curriculum with
emphasis on nutrition and physical fitness (Walter et al., 1985). The schools were
randomly assigned so that their students either participated in the program or served as a
control group. The researchers compared the two groups after a year. Relative to the
control subjects, the children who participated in the program showed improvements in
their blood pressure and cholesterol levels.
Walter et al (1985) participants are randomly assigned which avoids bias.
Good sample in
Edwards and Hartwell (2002) investigated whether children, aged 8-11
Edwards and
years could correctly identify commonly available fruit and vegetables;
to assess the acceptability of these; and to gain a broad understanding
Hartwell (2002).
of children's perceptions of 'healthy eating'. Fruit and vegetables used
were those readily available in retail outlets in the UK. Data were
collected from 221 children using a questionnaire supported by
semistructured interviews and discussions. Overall, fruit was more
popular than vegetables and recognition of fruit better; melons being
the least well identified. Recognition of vegetables increased with age;
the least well identified being cabbage which was confused with lettuce
by 32, 16 and 17% of pupils in their respective age groups. Most
children (75%) were familiar with the term healthy eating, citing school
(46%) as the most common source of information. Pupils showed an
awareness and understanding of current recommendations for a
balanced diet, although the message has become confused. If fresh
fruit and vegetables are to form part of a balanced diet, the 'health
message' needs to be clear. Fruit is well liked; vegetables are less
acceptable with many being poorly recognized, factors which need to
be addressed.
Many schools do
not provide health
education at all
because of
underfunding, the
courses being
poorly designed
and taught by
disinterested
teachers.
Could be a dated
finding though.
In Perry et al’s
(1989) study,
younger children (ages 8—9 years) participated in either a treatment or control schoolbased program designed to increase healthy eating habits. The intervention program
included modelling through stories and role-playing, self-monitoring of behaviours,
behavioural contracting, and material rewards. Treatment participants showed significant
reductions in the use of salt (it was not stated whether minorities were included).
Together, these studies reviewed above provide evidence that incorporating directly
observable behavioural objectives—such as setting written goals, modelling behaviours,
and providing feedback—can successfully result in long-term dietary change.
Perry et al etc - setting written goals, modelling and feedback works.
Must involve the support of staff and canteen staff where food is involved.
French et al (2001) examined the effects of pricing and promotion strategies on
purchases of low-fat snacks from vending machines. Low-fat snacks were added
to 55 vending machines in a convenience sample of 12 secondary schools and
12 worksites. Four pricing levels (equal price, 10% reduction, 25% reduction,
50% reduction) and 3 promotional conditions (none, low-fat label, low-fat label
plus promotional sign) were crossed in a Latin square design. Sales of low-fat
vending snacks were measured continuously for the 12-mo intervention. Results
show that price reductions of 10%, 25%, and 50% on low-fat snacks were
associated with significant increases in low-fat snack sales; percentages of lowfat snack sales increased by 9%, 39%, and 93%, respectively. Promotional
signage was independently but weakly associated with increases in low-fat snack
sales. Average profits per machine were not affected by the vending
interventions. It is concluded that reducing relative prices on low-fat snacks was
effective in promoting lower-fat snack purchases from vending machines used by
both adult and adolescent populations.
If healthy foods are reduced in price then sales soar and profits remain unaffected (see
French)
Ewart, Loftus and Hagberg (1995) evaluated the efficacy of school-based aerobic
exercise program for lowering blood pressure in a high-risk urban sample of ninth-grade
African American girls. Girls in the intervention group received a one-term aerobics
class of fitness instruction and training designed to be enjoyable and engaging for highrisk girls. Eighteen 50-min class periods involved lecture and discussion and 60 class
periods were spent performing aerobic exercise. Girls assigned randomly to the control
group just received the regular PE curriculum. After completing the course 81% wished
to continue for another term, demonstrating their enjoyment and a developing
commitment to regular exercise.
Ethics - some students get treatment others do not (Ewart et al 1995).
Peer based programmes are effective.
Health Promotion at worksites
Advantages
Saves companies money
Regular attendance at work so regular attendance at intervention sessions
Social support from co-workers
Facilities provided
Positive reinforcement from bosses
Less expensive
Convenient
Disadvantage
Those who need it the most refuse to take part.
An attempt to encourage people to quit smoking was carried out at five worksites. All the sites
received a six-week programme in cognitive behaviour therapy which focused on the skills of
giving up. The workers who enrolled in the programmes in four of the sites were put into
competing teams, with the workers at the fifth site acting as a control. At the end of the
programme 31 per cent of the people in the programme at the control site and 22 per cent the
competition sites had stopped smoking. A follow-up study after six months found that 18 per cent
of the control group and 14 per cent of the competition groups had stayed off the cigarettes. This
appears to suggest that the control group were doing better than the competition groups, but this
was not the case. At the competition sites 88 per cent of the smokers joined the programme, but
only 54 per cent did so at the control site, suggesting that the incentive of competition encouraged
more people to attempt to give up. When the data was compared for the total number of smokers
at each site to give up, there was an overall reduction of 16 per cent at the competition sites and
only 7 per cent at the control site (Klesger et al. 1986).
Smoking reduction - the results look as though no competition is better than competition,
but the competitive health intervention was more effective in attracting more smokers and
was more successful in the number of smokers who quitted.
Gomel et al 1993 - previously covered in preventing and quitting substance abuse can be
used here to make the point that banning smoking at work can lead to compensatory
smoking.
Problem with health promotion at work is the lack of consultation with workers.
Control Data's "StayWell" Program
Each StayWell participant completes a health screening, receives a resulting confidential
health risk profile, and attends a workshop that focuses on interpreting the profile. The
person can then join courses taught by professionals that provide information about
lifestyle and health and teach the skills needed to change unhealthful behaviors. There are
courses in physical fitness, nutrition, weight control, stopping smoking, and stress
management. The individual can also join action teams that focus on two things:


(1) making the work environment more healthful,
(2) forming support groups whereby members help one another in changing their
behaviour.
Evaluation of the StayWell program uses two approaches.
1. Some sites did not offer the Staywell program, and therefore could be used as
controls.
2. Employess exhibited varying degrees of participation in the Staywell program so
comparisons could be made.
Often there is no change in the work environment or works practices, but see Control data
who use action teams to tackle work related health issues.
Employers ignore research and do what they think is right instead.
Other criticisms
Mental health often ignored and the unemployed or homeless are not reached.
Health Promotion in the Community
There are too many uncontrollable variables to properly establish cause and effect.
Coronary heart Disease and mass media appeals
It is difficult to evaluate the effect of mass media appeals. In the case of product advertising the
effect can be measured in sales. In the case of health behaviour it is difficult to come up with
appropriate measures since there are so many influences on us every day. One of the most famous
studies on the effectiveness of mass media messages was the Stanford Heart Disease Prevention
Programme (see, for example Farquhar et al., 1977). This study looked at three similar small
towns in the US. Two of the towns received a massive media campaign concerning smoking, diet
and exercise over a two-year period. This campaign used television, radio, newspapers, posters
and mailshots. The third town had no campaign and so acted as a control.
The researchers interviewed several hundred people in the three towns between the ages of 35
and 60. They were interviewed before the campaign began, after one year, and again after two
years when the campaign ended. The interviews included questions about health behaviours,
knowledge about the risks of heart disease, and physical measures such as blood pressure and
cholesterol levels. In one of the two campaign towns, the researchers used the interview data to
identify over one hundred people who were at high risk of heart disease and offered them one-toone counselling.
The people in the control town showed a slight increase in risk factors for heart disease, and
the people in the campaign towns showed a moderate decrease. The campaign produced increased
awareness of the dangers of heart disease but produced relatively little change in behaviour. The
exception to this was the people who had been offered one-to-one counselling this group showed
significant changes in behaviour. This study suggests that mass media campaigns by themselves
produce only small changes in behaviour, but they can act as a cue to positive action if further
encouragement is offered.
—
The Stanford Heart Disease three towns project is a mixture of a natural and a field
experiment.
Small ethical point about one town not getting the appeal.
Usual strengths and weaknesses of self-reports and physiological measures.
A control town needed for counselling alone as the good results from the counselling
town could be due to counselling alone or being followed up and not the campaign!
Link to following up by health professionals from your adherence notes as well as to
targeting from the Yale model of communication.
Reducing skin cancer risk
Over the past twenty years there has been a large growth in the incidence of skin cancers, which
might be due to a combination of changes in the environment and changes in lifestyles. There are
a number of health promotion campaigns to encourage safe behaviours in the sun. A study on the
effectiveness of these programmes was carried out by McClendon and Prentice (2001). White
students who chose to tan were given a health promotion intervention based on protection
motivation theory (PMT). The intervention was made up of brief lectures, an essay, short
discussions and a video about a young man who died of melanoma (a particularly dangerous form
of skin cancer). There were two sessions, each just over one hour long and taking place two days
apart.
The researchers used psychometric tests to estimate responses to a range of variables
including:
•
vulnerability
•
severity of the threat
•
self-efficacy
•
costs and rewards
•
intentions.
With the exception of self-efficacy, these variables all showed some significant change after the
intervention and remained effective one month later. However, the issue is not whether people
intend to change their behaviour, but whether they actually do change their behaviour. This is
always more difficult to measure. In this study, however, they took photographs of the
participants at the start of the study and again after one month. These pairs of photographs were
then judged by four blind-raters (judges who did not know whether the pictures were before or
after) to see whether the students’ skin had tanned further or become lighter. The students were
not aware that this judgement would take place. Of the 32 individuals photographed, 23 (72 per
cent) were judged to have lighter skin tone after one month, 4 (12.5 per cent) were rated as having
no change and 5 (16 per cent) were judged to have darker skin.
The Skin cancer study had a very short follow-up period. Compare this snapshot study to
the previously mentioned longitudinal study.
It used students who were eager to please the experimenter (social desirability effects).
Again the effect could be due to the students knowing they were to be followed up.
Writing an essay lacks ecological validity as a promotion technique. No control group.
However, the use of blind raters reduced experimenter bias.
Targeting the homeless is difficult even though they are perhaps the most at need. There
is lack of collaboration between health workers. Lack of housing initiatives. They are
alienated from messages and suffer from low self-esteem.
Notes
http://homepage.ntlworld.com/gary.sturt/health/promoting%20health%20of%20a%20spe
cific%20problem.htm
Powerpoint http://homepage.ntlworld.com/gary.sturt1/health/Aids.ppt
AIDS - Health Promotion of a specific problem.
Targeting - Yale Model of communication. - Young people, Gay people, Intravenous
Drug users, Sex workers.
Need to tackle prejudice as well.
Should select partners carefully - but symptoms take a long time to develop.
People who believe they are not in the high risk group tend not to take precautions.
Unrealistic Optimism. Oxford Undergraduates felt they were not at risk.
Gay men are at risk because of promiscuity and risky sexual practices.
Having another STD also increases the risk.
Intravenous drug users are adopting safer practices with regard to injecting but are not
using condoms. Costs and rewards.
Policing policy can make matters worse. The Edinburgh police arrested people for
carrying injecting equipment, whereas the Glasgow police did not. This led to large
anonymous shooting galleries that caused the infection rate to increase to 50%! This
study is backed up by the fact that in New York where there are shooting galleries AIDS
is a bigger problem than in San Francisco where shooting galleries are not prevalent.
Harm minimisation.
This controversial program includes needle exchanges. An advantage is that
this would bring addicts into contact with health workers. There is a hierarchy
of behavioural changes for addicts to reduce the risk of HIV infection:




Do not use drugs;
If you use drugs, do not inject;
If you must inject, do not share injecting equipment;
If you must share, sterilise the injecting equipment before each injection.
Harm minimisation is controversial as it seems to be encouraging drug use at the
taxpayers expense. In fact, Harm minimisation is accepting the reality of the situation
and attempts to reduce the spread of the HIV virus by whatever means are likely to be
accepted by drug users.
The gay community in San Francisco has done much to change it’s members sexual
practices and has thus produced a reduction in the spread of HIV.
Outreach workers are effective.
Backlash - deliberate unsafe sex as a form of empowerment.
Kelly et al. (1994), for example, compared a five-session skills-based programme with an
education-only control group in a group of socially disadvantaged sexually active women
considered to be at high risk for HIV infection. The intervention included risk education,
training in condom use, practising sexual assertiveness, problem-solving and risk trigger
self-management. At the three-month follow-up, women from both conditions reported
having a similar number of sexual partners. However, what they did with those partners
differed; women in the intervention group reported that more of their partners used
condoms, and on more occasions.
Kelly et al (1994) target vulnerable women and educate them about condom use. Good
use of outreach workers.
Kolata (1987) reported a study involving gay men, in which participants discussed issues
relating to safer sex following either provision of written materials or watching an erotic
safer sex video. Those in the latter group evidenced increased use of safer sex practices in
comparison with those who received only written materials.
Video is moderately successful at providing modelling (Kolata 1987).
In a more radical study, Robert and Rosser (1990) compared the effectiveness of four
interventions aiming to reduce unsafe sex practices in a sample of gay men. Each was
randomly assigned to one of four conditions. The first involved watching a 15-minute
video on AIDS, which provided information on and modelled a number of safer sex
behaviours, including placing a condom on an erect penis and refusing to participate in
unsafe sexual behaviour. The second condition involved 20 to 30 minutes of individual
counselling in which major sexual concerns were discussed and standardized information
on safer sex given. A third condition involved attendance at a workshop, which explored
how to eroticise safer sex practices. The final intervention involved attending a workshop
in which the social impact of HIV/AIDS and safer sex guidelines were discussed. Trend
analysis revealed that individual counselling was most effective in increasing condom
use, while the erotic safer sex group was the only condition to evidence a significant
reduction in the frequency of anal intercourse. Of interest was the finding that the
conditions which engaged individuals in active consideration of how to change their
behaviour evidenced the greatest behavioural change, and that these changes reflected the
topics discussed. That is, each intervention may have produced effects specific to the
issues addressed within it.
Robert and Rosser (1990) used video, counselling, fun workshop and serious workshop
interventions. Counselling was best followed by fun workshop.
Teenagers need a chance to practice decision making not just information. Practical work
has been done on getting teenage boys over the embarrassment of purchasing condoms.
In a study of 234 male and 91 female teenagers living in the San Francisco
area, Kegeles et al (1988) found that the large majority of subjects agreed that
using a method, which prevents both pregnancy and sexually transmitted
diseases, was of great importance. In spite of this only 2.1 percent of females
and 8.2 percent of males reported using condoms. The sample might have seen
themselves as at low risk of HIV infection and saw condom use as having costs
that outweighed the perceived benefits.
Kegeles et al (1988) demonstrates that although people see AIDS prevention as being of
great importance, they still do not use condoms.
Link Health Belief Model to studies wherever possible - seriousness, susceptibility, costs
and rewards.
Also link Yale model of communication particularly targeting.