It's important to recognize, as we mentioned earlier,

that models are a slice of reality, a particular perspective.

Different researchers have developed models based on their own experiences.

Some have worked with specific health programs,

such as those that developed the health belief model.

Some have worked more in social psychology

in the field looking at learning generally.

And from these different perspectives that you can see, that each model, although

has a similar dependent or outcome variable in terms of human behavior.

Each model emphasize a different combination

of intermediate variables that may influence these behaviors.

Health belief model is the only one that really focuses on the variable or

concept of threat.

Social learning theory has a particular strength in emphasizing the issue

of self-efficacy, perceived ability, or confidence,

in the context of observational learning, watching other important people.

Theory of reasoned action introduces the issue of intentions and shows that

one needs to analyze the balance between social norms, perceptions of what other

people think ne should do, as well as the attitudes to the particular behavior.

And it was observed that this issue of attitude toward the behavior,

perceived benefits and constraints, outcome and value expectancies,

are basically the same variable and appear in all of the models.

All the models also, in one form or another,

take into account the broader social environment.

This is specifically spelled out in social learning theory,

not just showing that the individual is contained in, or part of an environment,

but the individual interacts actively and dynamically with the environment.

The environment is more of a background variable in theory of reason to action and

health belief model.

It's noted as external in theory of reason to action or modify.

But again, the three models do acknowledge the role of the different

environmental factors in determining individual behavior and

distinguishing the behavior of different subgroups of populations.

The final slides in this lecture provide a comparison of the models

based on the examples we've given so far to see which models provide

the best interpretation for understanding particular behaviors.

When we looked at the behavior of filtering for guinea worm control,

the health belief model shows us the contrast between

seeing that a disease is highly threatening, people perceive it's serious.

They know they're susceptible because of local beliefs.

At the same time, we can see the balance of forces happening,

such that people doubt the efficacy of the cloth in preventing the disease.

Here again, the importance of modifying factors that draw on local explanatory

models show us that people's knowledge or

believes about the disease are different than those of the scientific community.

And this is why a highly threatening disease results in inaction,

because the technologies we offer the community to solve their problem

are not perceived to work.

Again, this kind of study should be done in all health programs to give health

planners feedback about their efforts and why people will or will not accept them,

so that these technologies can be modified or substituted as the case may be.

Social learning theory showed us that with filtering behavior,

observational learning played a role.

People feel confident to filter, there's no doubts in terms of self-efficacy.

But again, this issue of the outcome expectancies, people not really

believing that the filter will work, is an over-riding concern.

Theory of reasoned action predicts, again,

that there would be low intention to use, primarily based on the attitudinal issue.

If we were to look at these three models,

it's likely that health belief model is the most useful because it provides

the stark contrast between people's perceptions of the disease seriousness,

contrasting with their doubts about the interventions that we offer.

These are the kinds of comparisons and

conclusions that you will need to make in your laboratory activities and exercises

when you try the different models on the case studies that are offered.

Or on a case study on your own that you may want to present.

The next slide compares the two models that were used to look at exclusive

breast feeding.

The theory of reasoned action, in this case, showed that both the attitude

toward the activity was negative, and the perceived social norms,

although not discouraging women from breastfeeding per se.

But did encourage her to supplement water, or herbal teas, or

other things to satisfy the child.

So, we can see that intention was low to practice this.

In fact, we are doing studies presently among nurses themselves, and

find that they are not completely committed to the idea.

So, in terms of nurses as a reference group that may encourage mothers,

their own beliefs will affect how they educate mothers on this.

Social learning theory, again,

showed the importance of outcome and value expectations.

It showed in the particular case that self-efficacy in performing

the behavior of breastfeeding was not really in question.

In this particular case, theory of reasoned action appears to offer more

explanation, more understanding of the behavior.

Finally, we look at the example of use of bed nets.

The three models had been applied there.

Health belief model showed us that people in

many rural communities in Africa do not perceive malaria as a threat.

It's a minor indisposition.

In terms, just like with guinea worm,

there may not be a connection seen between malaria as a disease or illness and

bed nets as an efficacious technology or solution to that illness.

But on the other hand, people do see benefits of using that,

that's intrinsic to the nets themselves.

Again, the only constraint would be cost and as we saw that was modified

by different personal and family economic and occupational characteristics.

So, we saw that bed nets could be adopted, but

not because people feel threatened by the disease.

In theory of reason to action, we can see that, again, the attitude toward

the behavior is similar to this cost benefit analysis and health belief model.

There is really no strong social pressure for or against having a bed net.

Although one could possibly construe the perception that

bed nets beautify the home and enhance social status as a positive value.

Social learning theory shows, again, that people may value the use of nets.

There is not a strong efficacy component involved in hanging up the net, pulling

the net down to sleep under it, re-washing or soaking the net in the insecticide.

So here, possibly, again, as with guinea worm, the contrast between people's

beliefs about the illness itself in contrast to people's perceptions about

the recommended technology, health belief seems to have some use in helping

us understand where people are coming from in terms of their bed net use behavior.

Social learning theory though shows the importance of observing other people,

and looking at bed net use in the context of the broader environment.

So, in conclusion, each model offers a different perspective, and

we should use multiple models to help us look into health issues.

Hopefully during your MPH career you'll have an opportunity to take

other courses in this school that look at some of the different research methods.

Quantitative survey methods.

Qualitative methods, such as in depth interviews, key informant interviews,

focus group interviews so that you can then use the ideas and variables

from the different theoretical models to frame your investigative questions and

learn more about people's health behavior prior to organizing a program.

And also using this information to help evaluate the acceptance of new behaviors,

to evaluate the outcome of public health programs.

The slides that conclude this lecture look at some examples

of how some of our MPH students in Abad and Nigeria have actually

done studies to test, or operationalize these different variables.

For example,

one of the students looked at over 400 mothers who had recently given birth.

And their children had passed four months of age,

when they would have had a chance to practice exclusive breastfeeding or not.

And so, a number of questions were asked to find out if they had done it, and

then would they intend to do it with their next baby?

Questions to operationalize this for

the attitude toward exclusive breastfeeding were things like,

what do you think would be the effect of exclusive breastfeeding on yourself?

Is it a good effect, bad effect, uncertain?

And then, of course, we could get narrative information for

them to explain why.

And then also, another component of the attitude was asking them questions about

what they think would happen to the baby.

Of course, we’ve talked it before, some of the negative effects for

themselves is that the baby would suck them dry, they would be malnourished

themselves because poor economic conditions they won't get enough to eat,

the child would not be satisfied.

So, these were how we operationalized this variable into

some specific questions on a survey instrument.

We operationalized intention.

Do you intend to feed the baby with your next child?

We, again assume that they would be having another child.

We then operationalized the questions on the perceived social support or

social norms, and mentioned a number of different people.

Would your husband support the idea?

Would your own mother?

Would your mother in law?

Would your sisters, would the other women in the household?

And we ask them, do they think these people would agree that this is

a good idea, a bad idea or they would be uncertain?

And so we could develop a composite score of the response of the perception of

whether each of these six groups of people would support or not support.

Then, of course, the next thing when we're talking about models,

again statistics provide us with regression models.

And in this case, we put in the different variables of exclusive

breastfeeding that we've talked about.

The attitude toward the behavior,

the perceived support, the perceived norms, and the intention.

And with this particular model we found that

both of those factors were associated with intention.

So, if they had a positive attitude toward exclusive breastfeeding,

that it would make their baby healthy, strong.

If they felt that other people would think that exclusive breastfeeding

is a good idea, that most of those thought so, then their intention

was higher than those people who did not have a positive attitude,

did not have a perception, a positive perception of support from other people.

So, we can actually test these, and

then we can use this information to design programs, knowing that, in fact,

these are important variables, that if we want women to practice this,

we have to explain the benefits, and show them that it's not dangerous.

But more than that, we also have to do community education with husbands and

older women in the household,

reaching them through their social groups in the market to let them know that this

is a behavior that they should encourage among their daughters.

Another example, when we looked at the question of drinking

behavior and motorcycle taxi drivers, we've tried to test some of the variables.

We found that 75 out of 266 motorcycle

taxi drivers in this town where much of our rural research is done,

actually stopped during their workday and

had a drink of beer, palm wine, maybe even schnapps.

Interestingly enough, 65% of those who reported this drinking behavior

had also reported that some time during the past couple of years,

they had fallen off their bike.

We couldn't use the word accident because that is something that is very sensitive.

People have certain beliefs about a being caused by witches or by people with

evil intent, and so if you mention the word accident it's frightening.

So we said, we framed it in other language to say,

did you happen to fall of your bike?

And 65% of those who drank reported falling off their bike,

compared to 46% of those who did not.

And this was significant.

So again, just from a simple epidemiological point of view,

we found that drinking, reported drinking behavior was associated with accidents.

But then the next question, as we've always said from the earlier lectures,

is not just to, there's not just a need in public health to find out

what behaviors contribute to health outcomes.

But we want to know who and why.

Okay, we see in the next slide, again, the idea of developing a regression

model to try to find out who are those people who drink and drive?

And we framed self-efficacy questions.

Do you feel confident that you could avoid drinking while you were driving?

We asked a number of attitude questions about safety.

The importance of stopping at intersections,

the importance of wearing helmets, a number of safety attitude questions.

We looked at personal characteristic of the number of years that they drive and

also another personal characteristic that they own the motorcycle that they were

using as a taxi, or were they hired by someone else to drive it?

And when we looked at those, again, the first two could be some of our

cognitive variables, the second two could be their external or

modifying factors, according to the models we might be using.

Found that those who had accidents felt less confident in refusing to drink.

They also had a lower attitude about taking these safety precautions.

Interestingly enough, they were more experienced, had been driving longer.

But on the other hand,

if they owned the motorcycle, they were less likely to drink.

Of course, we can talk to them and find out more, but we would assume that those

who owned the motorcycle would be more protective of their investment.

Whereas if you were driving for someone else, if the motorcycle crashed,

you could find a job driving for someone else.

So, these were some of the issues that we look at.

And then if we were going to do a training program in safety education for

the motorcycle drivers, we could say okay,

we need to help them examine their attitude toward the safety regulations.

Why do they not feel these are important?

What are the factors that make them think they can't refuse a drink?

And help them enhance their self-efficacy for refusing.

The next slide looks at an article from Health Education Research,

a journal that you all might want to look at that's online, and has a number of

studies that use the different models that we've been talking about.

And the authors there tried to find out whether women would have

the intention to eat a high-folate diet during pregnancy.

And they found in their study that three of the various kinds of cognitive

variables that we've been talking about were actually associated with

eating a diet that was high in folate.

They thought the diet had benefits for them.

Okay if they actually perceive fewer barriers to carrying this out, in terms of

the getting the food and cooking it etc., and on a positive note, again, if they had

a high level of self-efficacy that they could buy, prepare and eat these foods.

So again, these models gives us an idea of the factors that may influence behavior.

We can do various kinds of surveys, operationalize the variables and

do surveys to see which ones really apply.

Do they apply to the whole population?

Some parts of the population?

And then using the results, where we find factors that are actually associated with

the behavior, to design interventions that will help ameliorate those factors and

hopefully result in some positive changes in health behavior.

Comparisons

Diagnosing Health Behaviors for Global Health Programs

Meet the Instructors