AMCHP 2005 ANNUAL CONFERENCE
DELIVERING RESULTS, IMPROVING PREGNANCY & BIRTH
February 19-23, 2005
PAMELA XAVERIUS: Good afternoon. I want to talk specifically about a project that we are working on in Missouri geared toward training providers of how to talk to smokers about not smoking. And I thought I would begin with giving you an idea of why we were one of the dark blue states that Janis showed and show you some of that data first. I'm a data geek and I probably have too many grafts in here so just bear with me.
My previous speakers have already spoken to these issues, but, you know, smoking during pregnancy has a rate that's unacceptably high, and we need to work on getting women to stop smoking for all the reasons just stated.
And in Missouri, you can see that it's clear that the blue line here is just overall smoking rates during pregnancy and it's been fairly high overtime from 1991 to the present, even in 2003 it was almost one in five pregnant women were smoking in Missouri. This is perhaps the only sort of maternal health outcome disparity one where you see the African American rate is lower than the white women's smoke rate instead of every other one, it's flip-flopped. But still, the rate is much higher than Healthy People 2010 suggest that no more than 1 percent of women smoke and it's almost 20 percent in Missouri
The story gets bleaker when you look at younger pregnant women they are smoking at higher rates than really overall. So sort of starting to target toward, you know, here is a graph that shows women between the ages of 18 and 19, and another one that is showing it between 20 and 24. And you can see those rates are almost one in four. 25 percent of women are smoke between the ages of 20 and 24 while pregnant in Missouri. Unacceptable.
That's looking at vital static data. Sometimes you can look a variety of different ways at the information. We also have a periodic survey that goes out regarding perinatal substance abuse. And when we look at that data and look at age under 20 by race we see '93, '97 and 2001, those rates of smoking are just going straight up and even at higher rates. So either way we cut the pie, whether we look at vital statistics data or surveys, women are smoking at a much higher rate than we should accept in Missouri.
Also, there's evidence out there that talks to how to go about getting women to not smoke. There is this -- the Community Guide, are you all familiar with the Community Guide? Its just a great resource where they go ahead and do the scientific research and you can go and use what they recommendations because they followed some really stringent methodology to make some recommendations, but you can go to the Community Guide and they have done the research on what works on helping people not smoke. And there's a variety of things that are pretty effective of helping people not smoke. And this little media campaign, when you look at the list here, if you see two dots it means really effective, if you see one dot it means sufficient. And then the question marks are, we are not sure. But media campaigns are pretty effective, increasing the price of tobacco obviously pretty effect testify. Media campaigns with intervention. Provider reminder systems which is just telling a provider you need to remember to tell women not no smoke. Provider reminder systems with education and that's showing them how to talk to people about not smoking.
And my experiences having traveled the country and trained providers on how to talk to people about not smoking, they really don't know how to do it. Oftentimes, the first thing they'll say to somebody is, do you smoke? They ask. They say yes. And they say, you do know it can kill you. They start with the real ineffective psychologically anyway in terms of convincing, motivating someone to not smoke. So there is room for education on how to do that and do it effectively. Reducing the cost for treatment and having support lines are all pretty effective.
This is just essentially saying all the things that are effective in terms of trying to control tobacco use in communities, whether it's through community intervention or healthcare systems, what our intervention in Missouri, at least this evaluation that I'm going to talk to you briefly about focused on is the healthcare system, a provider reminder system with education. And I just described what the provider reminder system is with education and really, brief counseling is different than intensive counseling. Brief counseling is really a five to 15 minute conversation with somebody about tobacco use. Intensive treatment would be like a 16 week program somebody would go through to really go through weekly and talk about quitting and how to do that.
When they did the review of the literature and looked at hundreds of hundreds of articles, they found overall that when intervention included patient education, provider education and provider reminders, advice increased 22 percent and quit rates increased five percent. One of the things I say when I look at when I look at that is five percent, wow that's not really a lot. But when you look at it from a public health perspective, that's quite a few people that are not smoking. So you are really getting a lot of bang for your buck. So it's pretty effective. And when you just have the provider reminder and education without necessarily the patient education, quit rates are still close to five percent. So it's really an effective way from a public health perspective if you can get people trained on how to do that to get smoking rates to go down.
The five A's, here they are. I raised my hand when she asked if we knew what they were. Now one of the -- the deal with the study that we did was how do we know that people are trained? They might go to his training and say oh, yes, I'm very satisfied. But do they actually have any -- is there quality assurance. Are we sure these people have the skills, the information, the knowledge to be able to counsel folks. So that was really the question we were asking ourselves when we were starting to train providers in Missouri how to counsel people not to smoke.
Really, Arizona and Massachusetts, at least initially, have led the way in sort of quality assurance in terms of if you train these folks, how do you make sure you train them in an effective way? Arizona has a certification program that they published about that they are really actively involved with where they teach the counseling skills, the basic skills to somebody. It's a four hour training session and it gives them the skills. Then they have cessation specialist who are sort of like the train the trainer model when those folks are taught how to train people on how to quit smoking. And they have, they give sort of pretest, post test, follow-up test to those folks to make sure they are doing what they've taught to do.
And Massachusetts has sort of a one level certification program, but again it's sort of empirically based and they've been trying to make sure people have the skills that they were intending to get.
So we thought we would try to do that in Missouri also. And we really used the training that was described just a moment ago in terms of following our guidelines and if you're going to counsel pregnant women who smoke that, you know, you need to be careful of the pharmacotherapy or nicotine replacement issues and follow all the guidelines. I won't repeat what was just said a moment ago.
So, what we did in Missouri was there were training sessions that were offered to service providers to teach them how to counsel women of child bearing age on not smoking. And we sort of thought of this as our pilot. So initially we had three training sessions in three different parts of Missouri. One in Springfield which is southern Missouri, one in Jefferson City and one in Kirksville. So we covered the lower, middle and upper portions of Missouri. We had 93 people attend those sessions. And at each session, when somebody walked in the door, we were vigilant is the nice word. We were vigilant about making sure everybody filled out a survey initially. So they walked through the door and we didn't hunt them down, but we kindly asked them, would you fill this out before the training. So we had these pretests and we asked them all sorts of question and we asked them, what do you know about the five A's, what do you know about the intensive treatment, what do you know about blah, blah, blah. And then they had this four hour training session and then we gave them the same exact survey again at the end, and we said okay, what do you know about these things now. So we had 88 people fill out the pretest and we had 86 folks fill out the post test. So we only missed two. So we really captured almost everybody who attended these training session in terms of has your knowledge changed at all on these issues. Then after the training was over, three months later, we sent surveys out to all of those folks, and we said now that you have this knowledge, have you been using it? Have you been talking to folks about tobacco use. And we had 47 folks return those surveys. And we are in the process right now of analyzing hat data. And one of the things we've been able to do is match folks. So we -- they were anonymous, they didn't have names on there, but we could tell by age groups or location or whether they used a pencil or pen and pull those together and we were able to match 76 of the post tests and 34 of all three. So we are in the process and what I'm going to do is present to you some of the information that we've gleaned so far.
Essentially the independent variable here was the training section. We are in the process of modifying that now, but right now the training session was pretty much just, I'm going to talk to you and you're going to absorb, not necessarily a practice. We're modifying that now to include a practice component to it because we think that's pretty important. But right now, that's pretty much what the independent variable was. And like I said, the measures, or the dependent variables, were these tests, the pretest, post test and the three month follow ups.
And essentially, through all the groups, it was primarily women that attended the training like 93 percent of women attended the training. The age ranges really sort of covered the gamut from 15 to 24 to 65 and older with the bulk of folks between 25 and 64 on all tests. So they were pretty comparable. Each group of tests was pretty comparable in terms of the demographic mix of those folks. Primarily our attendees were white. Although there was a couple of gender sort of mix, other racial ethnic mixture, but there was white folks.
They really sort of -- the blue box here suggested they work in urban settings and the purplish color is rural settings. If you don't know Missouri, it's fairly rural. There's Kansas City on one side and St. Louis on the other and then a whole bunch of rural stuff in the middle. Lots of these folks are servicing folks in rural Missouri. And these are the three locations that I talked about, Springfield, Jefferson City and Kirksville.
There were some questions we asked just purely about satisfaction. Did you like the training? Did you think it was worthwhile? And overall, people could give it a score of 1 to 5. One was hated it, five was loved it. And essentially, everybody said they loved it. They gave it an average of 4.77, in terms of they thought the quality of the instruction was great, they thought the material was useful, they thought the content was relevant and they really thought overall it was valuable. But being a person with a bachelors, Master's and Ph.D. In behavioral psychology, what people say is one thing, but what people do is often really different. And so while this is nice to know this is standardized, what I really want to know is did they do anything with it.
And I apologize for this graph, this is what they teach you in Power Point 101 not to do and to even compound it and make it worse, I don't have copies of my handout. So anybody who would be interested in getting copies of my handout, if you want to give me a card or write your name down on a sheet of paper, I promise you when I get back to Missouri later this week, I'll email you a copy of it.
Essentially what we did, though, because we could not get our hands on a survey that was valid and reliable, and that means it's reliable if somebody takes this test and they take it a second time, would they answer the question the same way. And valid is for those of you that have been away from Stats 101 for quite a long time, valid is that it really is measuring what it claims to be measuring. Usually if you do research, you want to use an instrument that's already out there that's valid and reliable so you can then not have to spend a lot of time validating and making it reliable. Unfortunately, we couldn't get our hands on the Arizona or Massachusetts instrument, so we had to make up our own. So we spent quite a bit of time doing some preliminary analysis. And we are finding that it's very reliable and that it's very valid, yeah, and perhaps it will be a paper that we publish in the future.
But for now, essentially, we just asked 16 questions about, do you know what asking is, do you know what assessing is, do you know what advising is. Do you know the difference between brief and intensive counseling. So on the left column here, you see the pretest scores and on the right column you see the post test scores. And these are for the matched ones, so that means that this data has a lot of power because we know these are the same folks and we can look at it from groups or we can look at it from individuals. But overall, every single question, there was a statistically significant increase in knowledge. So we felt pretty good about that. We felt not only is the instrument valid and reliable, but people are showing us that they actually have more knowledge. But, again, behavioral psychologist, just because I said they know more does it mean they will do more. So that was the idea behind doing the three month follow-up. Unfortunately, if you remember the first slide I showed you, were only able to match 34. So we are trying to figure out exactly what we are going to do with that because we did not quite get the response rate that we had wanted.
But essentially, for those that we did get information back from, those 34 three month follow ups -- well, 34 were matched. Overall, there were 47. 100 percent of those people said the training has assisted them in delivering brief counseling. In fact, 85 percent of them said they have performed at least one brief intervention since the training within three months. 75 percent have said within the last 30 days they have performed a brief intervention.
So the study is telling us through the survey they are actually changing their behaviors, they are actually talking to women about smoking. And as we move through some of the questions, we get down to the bottom here that 97 percent of the said at least one client has reduced her smoking, and almost 70 percent have said at least one of her clients have stopped smoking.
So we pretty encouraged that this is getting us toward the way we want to go. So we think the preliminary analysis is really telling us we need to do more of this and role it out more, because the first graphs I showed you is that women in Missouri are smoking at a rate that is unacceptably high. So we are kind of excited about that and we are excited in our panel here that we could go from the broad to the very specific in presenting the information and getting down to a state specific example.
So this is all that data, it's reliable. Essentially, it's very reliable. We would want our combat alpha coefficients to be over a .70, and it's close to a .90. So it's very reliable that we could predict that if these people took the test again, they would score the same way.
We did validity. There's all different kinds of validity that you can take a look at. Post validity, content validity, context validity, and we are getting some data on all that. So there is evidence that on effectiveness that, you know, that's one way to make decisions. So if you're going to decide what you're going to do with folks, base it upon evidence first. It's a no-brainer. And then effective intervention for using tobacco use. We know that. Evaluation of provider trainings are the building blocks of comprehensive tobacco control programs. We feel like we need to contribute to that literature because it's pretty dearth right now in terms of who's showing in the literature that there are effective ways to train folks on how to counsel smokers not to smoke.
And in Missouri, like I said, we are in the process of updating the training program for wider dissemination and to include the actual -- you know, we'll break folks into dyads and have them practice and give them feedback on how to counsel somebody who is not ready to quit or ready to quit and various different scenarios and that theoretical model of change. But ultimately what we want, I showed you the graph at first where the smoking is fast and furiously study. We want to start making that go down with my made up data at the end. And hopefully with some of these trainings that we are starting to disseminate in Missouri we'll be able to do that so we can have healthy babies like my little Gracey right here. Thank you.