Health Education Research, Vol. 14, No. 2, 289-298,
April 1999
© 1999 Oxford University Press
Key informant surveys as a tool to implement and evaluate physical activity interventions in the community
Prevention Research Center, School of Public Health, St Louis University, 3663 Lindell Boulevard, St Louis, MO 63108-3342, and
1 Department of Health Research and Policy, Stanford University School of Medicine, Palo Alto, CA 94304, USA
Correspondence to: R. C. Brownson
 |
Abstract |
|---|
 Top Abstract IntroductionBackgroundMethodsResultsDiscussionSummaryReferences |
|---|
Key informant surveys are important tools for planning and evaluating community health programs. A survey was conducted to gather views on policies toward physical activity from four sets of key informants: physicians, church leaders, business leaders and civic leaders. Surveys were mailed to 797 key informants who were selected from 12 southeastern Missouri counties. For comparison, data from a telephone survey of 2106 persons in the general population were also analyzed. The majority (>85%) in all four key informant groups were very supportive of required physical education in schools, but less supportive (<69%) of 0government funding for places where community members can exercise. Physicians perceived community members as having somewhat greater access to places to exercise relative to the other key informant groups. Comparisons of the key informant surveys to the population survey indicated similar levels of support for physical activity policy. The information from this survey has been useful in identifying support for physical activity policy and gaining access to potential influences for community change. Since key informant research in the area of physical activity policy and cardiovascular disease prevention is sparse, there is a need for future studies.
|
Introduction |
|---|
|
TopAbstractIntroductionBackgroundMethodsResultsDiscussionSummaryReferences |
|---|
In the past decade, community-based health promotion programs have become an integral part of overall health promotion efforts (Shea and Basch, 1990
). These community-based programs attempt to alter social and physical environments as well as public health policies in order to influence health behaviors and reduce health risks of community members. As with any health promotion program, proper assessment and planning influences program success (Green and Kreuter, 1991). Population-based surveys are one way to assess the needs and attitudes of community residents. Another approach is the use of key informant surveys. Assessments of the community by key informants can provide information distinct yet complementary to that obtained in population surveys.
The key informant survey is a method of obtaining data from persons whose professional and/or organizational roles imply they have knowledge about specific characteristics of the population being studied as well as potential pathways and constraints for community change (Warheit et al., 1978; Von Korff et al., 1992). Involving key informants in needs assessments often leads to initial community mobilization (Marti-Costa and Serrano-Garcia, 1983). For example, during the process, contacts are made with individuals who may assist with program policy support or become involved in implementation of community health coalitions (Gregor and Galazka, 1985). The key informant method has been successfully used in several areas of study. Examples include mental health (Warheit et al., 1978; Vega et al., 1985; Morrisey et al., 1994), primary care (Gregor and Galazka, 1990; Williams et al., 1994), AIDS policy (McDonald and Natarajan, 1989) and community activation (Von Korff et al., 1992).
Theoretically, key informants generally fall into three categories. `Economic dominants' are occupants of major economic posts in the community such as area business leaders. `Prescribed influentials' are community leaders who hold positions formally designed to sanction and facilitate influence in the community such as civic leaders. `Attributed influentials' are those community members who are perceived by others as being influential in community decision making (Jennings, 1964). Examples of `attributed influentials' used in past health research include physicians (Mittelmark et al., 1988; Kofron et al., 1990; Luepker et al., 1994) and church leaders (Mollica et al., 1986; Olson et al., 1988; Thomas et al., 1994).
One area that has been little studied using the key informant approach is physical activity policy. Changes in policy and the environment to foster and maintain individual-level behavior change (such as improvements in physical activity levels) are an important aspect of chronic disease prevention (Schmid et al., 1995; King et al., 1995). For instance, efforts to promote walking as a form of physical activity may be futile if there is no access to a safe place to walk. In this case, policies that support walking trails are an example of how policies can complement individual-behavior change efforts. Gaining information on level of support for physical activity policy from key community leaders can help guide the efforts of community coalitions.
Two major objectives of the current study were to analyze the perceptions of physical activity policy among community leaders and to compare responses to physical activity policy data from a previously collected population-based survey from the same locale.
|
Background |
|---|
|
TopAbstractIntroductionBackgroundMethodsResultsDiscussionSummaryReferences |
|---|
The Prevention Research Center at St Louis University is implementing an on-going community intervention called the Ozark Heart Health Project. The goal of the project is to reduce three behavioral risk factors for cardiovascular disease (CVD)—physical inactivity, cigarette smoking and poor diet—through communityfacilitated change promoted by coalitions (Brownson et al., 1997
). The Ozark Heart Health Project targets low-income families in a 12 county region of southeast Missouri. The study area is primarily rural, medically underserved and has a higher rate of CVD than the rest of the state (Brownson et al., 1997). In a related project, the Prevention Research Center is studying the determinants of physical activity (specifically, physical activity policy) in the same population. Part of this research included a key informant survey that was sent to area civic leaders, physicians and nurse practitioners, business leaders, and church leaders. The survey was tailored to each of these four groups, but all contained a series of core physical activity policy questions and assessment of perception of access to physical activity resources. Similar questions were asked in a population-based telephone survey conducted in the Ozark region in 1995.
|
Methods |
|---|
|
TopAbstractIntroductionBackgroundMethodsResultsDiscussionSummaryReferences |
|---|
Key informant surveys
The key informant surveys were developed as a compilation of questions from various surveys in addition to several original questions into a four-page paper-and-pencil instrument. The core questions (physical activity policy and access) were taken from the Ozark Risk Factor Survey (Brownson et al., 1997
). Questions individually tailored for each group were selected from different sources. For instance, several health promotion questions on the business leader survey were taken from the 1992 Worksite Health Promotion Survey (USDHHS, 1992). The key informant survey method used in this study is unique. The broad content, including perceptions of access to physical activity resources and support for physical activity policy, is different from topics covered in other key informant research. Also, the sample of key informant groups used in this study differs from previous research. Our survey was sent to four different groups (physicians, business leaders, civic leaders and church leaders). These groups were chosen for several reasons. First, the groups appeared to have influence on community-wide behaviors. Second, we chose groups of leaders with large enough sample sizes for meaningful analysis. Third, gaining information from four separate groups of leaders may give coalition members several different avenues to approach physical activity policy in their communities. For instance, in this rural study area, churches are institutions that serve a greater community purpose than enhancing spirituality. Many of the churches in the Ozark region are also social, recreational and support entities. Church leaders who support physical activity policy can advocate such support to church members and perhaps open their church facilities for community exercise and recreation programs.
We did not include recreation leaders or exercise instructors for several reasons. First, very few exercise facilities or recreation programs exist in this rural area. Because of the small numbers of people in these positions, meaningful analysis would be difficult. Also, we wanted information on support for physical activity policy from those not directly involved with such programs.
The surveys were pre-tested (n = 46) in two counties not included in the Ozark Heart Health Project, but in the same geographic region and with similar demographics as the study area. After pre-test results were collected and analyzed, revisions to the survey were made.
Nineteen zip codes (out of 109) from the 12 counties in the Ozark Heart Health Project served as the basis of our sample. Theses zip codes were chosen based on population size (US Department of Commerce, 1992) and contained 66% of the population in the 12 counties. Because of group characteristics, the method of sampling was slightly different for each group of key informants. For physicians, the list of names of all those practicing in the area (19 zip codes) was obtained through the state medical licensing office in the Missouri Department of Health. This database contains a comprehensive list of physicians licensed and practicing in the state. Certified Nurse Practitioners (NP) were added to this sample (N = 3). Since the intervention area is medically underserved, the physician/NP sample was small (n = 93).
The civic leader sample included all mayors, county presiding commissioners, deputy directors, public administrators and sheriffs in the 19 zip codes. The list of names (n = 64) was gathered from the 1995–1996 Missouri Roster (Cook, 1995) and the Municipal Officials Directory (American Business Information, 1995b).
Unlike physicians and civic leaders, the sampling of business leaders and church leaders was more complex. A list of area businesses (for the 19 zip codes) was compiled using the American Business Information (ABI) database (American Business Information, 1995a)—an index of businesses in the US and their demographics. Businesses with over 25 employees were chosen because larger businesses are more likely to be involved with health promotion activities (USDHHS, 1992). Restaurants and fast-food establishments were omitted from the sample due to the likelihood of many part-time employees and high employee turnover. Names of personnel directors or human resource managers were obtained by calling the businesses on the list. These names were then used to personalize the survey cover letter. The total number of surveys mailed to businesses was 266.
Since a comprehensive list of area churches was not available, an aggregation of several lists was used. Names and addresses of churches were obtained by using the National Directory of Churches, Synagogues, and Other Houses of Worship (Melton and Krol, 1994), area telephone books and the ABI database. Once the final list was compiled, calls were made to verify the address and identify the name of the church leader to personalize the survey cover letter. After extensive calling (including calls made on Sundays), only 65% of church leaders were identified by name. Generic cover letters were sent to those churches without an identified church leader. A total of 374 surveys were mailed.
After the initial responses to the survey were received, a second copy of the survey was mailed to those who had not yet responded. After the second mailing, reminder phone calls were made and third surveys were sent. Overall response rate for the survey was 65%. Final response rates for each of the groups were as follows: civic leaders, 63%; physicians, 59% (100% for NP); business leaders, 83%; and churches, 49%.
Ozark Risk Factor Survey
The Ozark Risk Factor Survey was a telephone survey conducted with 3024 adults 18 years of age and older, residing in the 12 counties of interest. Study subjects were randomly selected (three-stage random digit dialing) from residents of the six intervention and six comparison counties. The response rate was 70% among eligible households. A detailed description of this survey and methods is published elsewhere (Brownson et al., 1997). Of the 3024 adults surveyed, only those respondents living in the same 19 zip codes as used in the key informant sample were used in analyses reported here (n = 2106). Data from the Ozark Risk Factor survey was cleaned and edited prior to this analysis. Although this was a comprehensive survey, only variables on physical activity policy, access and physician counseling were used in this study.
Analysis
Data from the key informant surveys were entered into the SPSS statistical analysis program. Frequencies and percentages were computed and analyzed. Mean and median age as well as sex of the respondents are depicted in Table I. Frequencies and confidence intervals for the variables of interest are depicted in Table II. Rate ratios (also depicted in Table II) were computed using the physicians/NP as the comparison group. Physicians/NP were chosen as the comparison group because they were more likely to show support and, presumably, of lower risk than the other groups.
|
|
|
Results |
|---|
|
TopAbstractIntroductionBackgroundMethodsResultsDiscussionSummaryReferences |
|---|
The majority of respondents in all groups were male with the exception of the business leaders. Key informant groups were asked if they `participated in any exercise, sport, or physically active hobby in the past 2 weeks'. Physicians had the highest response rate to this question. Ninety-two percent indicated that they were physically active in the past 2 weeks. The business leaders had the lowest mean age of the four groups (42), while the physician/NPs had the highest (70). The large difference between the mean and median age of the physician/NP group was due to six respondents who were 15 years or more older than the next oldest respondent.
As the introductory question on the survey, all groups were asked their level of concern with physical inactivity within their community (Table II
). Concern over physical inactivity was the highest among physicians. Ninety-three percent of the physicians/NP who responded said they were `concerned' or `very concerned'. Business leaders also demonstrated high levels of concern for physical inactivity (86%). A lower percentage of both civic leaders and church leaders expressed concern for physical inactivity in their community (55% for each).
Perception of community access to places/programs for physical activity was also analyzed (see Table II). Physicians/NP believed that the community members had somewhat greater access to walking/jogging trails, pools, school gyms, etc., than did church leaders, civic leaders or business leaders. Less than 50% of civic leaders thought that community members had access to any of the physical activity places/programs mentioned in the survey (walking/jogging trail, community pool, indoor recreation center, structured exercise programs and use of school gyms/tracks).
In comparison to the key informant surveys, one general question about community access to places where people can exercise was asked on the population-based Ozark Risk Factor Survey. Only 51% of the respondents thought community residents had access to places where they can exercise.
Two physical activity policy variables were analyzed in both the key informant survey and the population-based Ozark Risk Factor Survey. One question covered attitudes on government funding for places where people can exercise. For civic leaders and physicians, 55 and 56%, respectively, indicated that they supported government funding for places to exercise. Both church leaders and business leaders expressed higher levels of support for government funding for places to exercise (69 and 70%, respectively). When this same question was asked on the Ozark Risk Factor Survey, 71% of respondents indicated support (Brownson et al., 1998).
The second physical activity policy question focused on the general requirement of physical education in schools. Support for this variable was overwhelmingly high. Ninety percent of church leaders indicated support, as did 94% of both business leaders and physicians/NP. The group with the lowest support of the four groups was civic leaders (85%). In the Ozark Risk Factor Survey, 93% of respondents thought that schools should require physical education.
The third part of our analysis of the key informant survey focused specifically on physicians. Physicians/NP were asked `about how often do you counsel sedentary patients on physical activity?'. Sixty-four percent indicated `always.' A comparison to the Ozark Risk Factor Survey indicated a discrepancy. Only 18% of sedentary respondents (defined as not having participated in any physical activity or exercise in the past 2 weeks) reported being counseled by their physicians in the past year to become more physically active.
|
Discussion |
|---|
|
TopAbstractIntroductionBackgroundMethodsResultsDiscussionSummaryReferences |
|---|
This key informant survey provided valuable information that can be used for both planning and evaluation of the Ozark Heart Health Project and related interventions. The groups indicated low access to places to exercise. This perception is a reality for the majority of the communities. However, with the aid of the heart health coalitions, several improvements are being made. For example, walking trails are being built in many of the communities. Additionally, gymnasiums and tracks in several schools are open to community residents to use for exercise. Churches are also being targeted as sights for sports/recreation programs and exercise classes. Also, area businesses are given information on ways to incorporate physical activity at the worksite and other worksite health promotion topics.
Another important finding from our analysis was that policy questions had the majority support from all four key informant groups as well as from the population-based Ozark Risk Factor Survey. Over half of all group respondents support government funding for places for people in the community to exercise. Interestingly, all four groups as well as the general population were very supportive of physical education as a requirement in schools. In the Ozark region, only 50 min per week of physical education is required in grades 1–5, 50 min twice a week in grades 6–9 and only one semester of physical education is required for graduation in high school. Perhaps level of support would have been different for both the government funding and required physical education questions if support for a tax levy was included as part of the questions.
Clearly, civic leaders showed the lowest concern over physical inactivity in their community and lowest support for physical activity policy of the four groups surveyed. This has implications for intervention implementation. Civic leaders (those most responsible for policy making) appeared to have the most negative views. Although many of these `prescribed influentials' claim to participate in exercise themselves, perhaps they have priorities other than physical activity when it comes to allocating their energies and spending the community budget. Coalitions must work with these leaders to emphasize the importance of physical activity policy.
Physicians/NP appeared to have high levels of concern over physical inactivity in the community and perceived somewhat greater access to physical activity resources than the other groups. They also had the highest rates of physical activity when asked if they participated in any `exercise, sport or physically active hobby in the past 2 weeks'. Perhaps their own habits coupled with the perception of the resources already being available are reasons why they are less likely (compared to business and church leaders) to endorse government funding for places to exercise.
Research has demonstrated physician counseling may increase patient behavior such as physical activity (USDHHS, 1996; Lewis, 1988). Sixty-four percent of the physician/NP respondents reported `always' counseling sedentary patients on physical activity. However, results from the Ozark population-based survey indicates that only 18% of sedentary respondents reported being advised by their physicians in the past year on physical activity or exercise. One possible explanation for this is the broad definition of `counseling'. While physicians may view mentioning a suggested increase in physical activity as `counseling`, patients may view `counseling' as more of a detailed plan of action. Another possible explanation for this discrepancy is that many health care providers may not believe that physical activity is an important topic to discuss with their patients given the time allotted for care and may lack effective counseling skills (USDHHS, 1996) but social desirability of the survey question warranted a positive response (Lewis, 1988; Hoppe et al., 1990).
An additional benefit of the survey was that half of the respondents (50%) expressed interest in such programs by requesting more information on the project as well as a summary of survey results, thus demonstrating the reactive nature of key informant surveys. Those who expressed interest in the project may be able to serve as potential coalition members and physical activity policy advocates.
Limitations
They are several limitations to this study. First, there is the possibility that selection bias may be an issue due to the sampling of only the major zip codes in the 12 counties. The zip codes with the largest population in each county were chosen. Perhaps the more rural/isolated key informants would have responded to the survey differently. While sample sizes for churches and businesses were large, the sample sizes for physicians/NP and civic leaders were small, thus limiting our ability for more complex multivariate analyses. Additionally, by sending the business leader survey to the attention of the human resource manager, we may have received a higher percentage of women respondents than the other groups and an over-representation of respondents who do not have direct influence on business decisions. Also, the results from this rural, southeastern Missouri region may not be generalizable to other areas and similar research in other settings is warranted.
|
Summary |
|---|
|
TopAbstractIntroductionBackgroundMethodsResultsDiscussionSummaryReferences |
|---|
Key informant surveys of community leaders are important to the community intervention process for several reasons. First, by surveying multiple key informant groups, information from the surveys can be used to tailor interventions to each group. Methods to promote physical activity policy advocacy in civic leaders will likely be different from those used with physicians. For example, civic leaders that indicated support for government funding of places physical activity can be contacted by coalitions to advocate building and maintenance of walking trails in their communities. An example of a different approach used for physicians who expressed interest in the project is development and distribution of a `prescription pad for physical activity' that can be readily provided to sedentary patients. This `prescription' will include basic guidelines for exercise and local telephone numbers for area resources, and will be signed by the physician.
Second, the key informant method is consistent with the social planning approach to community organization. Social planning is a top-down approach that primarily involves expert planners (Rothman and Tropman, 1987). In our study, these `expert planners' are the key informants. One of the main strengths of a key informant approach is reliance on experienced community organizers rather than inexperienced community members (Rothman and Tropman, 1987). Established networks and prestige in the community are two examples of key informants qualities that may foster health promotion interventions.
Third, the key informant approach used in conjunction with other population-based data is a good example of triangulation of data. Triangulation is the use of multiple methods in the study of a phenomenon and can help overcome bias inherent in any one method (e.g. bias of surveying only those who have telephones in a telephone survey) and to increase validity because different methods highlight different aspects of the phenomenon (Denzin, 1978). When comparing the results of the key informant survey to the population-based Ozark Risk Factor Survey, it was interesting to identify similar levels of overall support from both key informants as well as the population data sets (e.g. requiring physical education in schools). The discrepancy in physician counseling responses suggests a difference between patient and physician reporting of behavior change recommendations. This is consistent with the literature on physical activity counseling (Roter and Russell, 1994) as well as with smoking cessation counseling (Ward and Sanson-Fisher, 1996).
The results of this key informant survey have already been put to use. For example:
- We have compiled lists of those key informants who indicated interest in the project and provided the list to the project coordinator in the Ozark region.
- For civic leaders, intervention is focused on policy issues surrounding access to places for physical activity. Heart Health coalitions are working with civic leaders to secure funds for the building of walking trails.
- Sparked by interest in both the survey and the physical activity policy project, six of the mayors in the intervention area have signed proclamations promoting physical activity in their communities.
- A booklet on promoting physical activity in the church community has been developed, and will be distributed and evaluated among church leaders who expressed interest in increasing physical activity participation of their congregation.
- Physicians who responded to the survey are being invited to join the heart health coalitions in the area.
- The community coordinator will distribute `physical activity prescription pads' to each clinic in order to make counseling sedentary patients more convenient for the health care providers.
- Business leaders who expressed interest will be contacted by the community coordinator and invited to attend a coalition-sponsored workshop on workplace health promotion.
In summary, our key informant survey of physicians, business leaders, civic leaders and church leaders has provided us with valuable information that can be used for intervention development and evaluation. Such an approach is relatively inexpensive ($3.00 was the approximate cost per survey, which included copying, postage, administrative time, analysis and supplies), and may provide important health promotion planning and evaluation tool for other community-based projects.
|
Acknowledgments |
|---|
This project was funded through the Centers for Disease Control and Prevention contract U48/CCU710806 (Center for Research and Demonstration of Health Promotion and Disease Prevention).
|
References |
|---|
|
TopAbstractIntroductionBackgroundMethodsResultsDiscussionSummaryReferences |
|---|
American Business Information, Inc. (1995a) American Business Disc. American Business Information, Inc., Omaha, NE.
American Business Information (1995b) Missori Municipal Officals: 1995–1996. American Business Information, Inc., Jefferson City, MO.
Brownson, R. C., Mayer, J. P., Dusseault, P., Dabney, S., Wright, K. S., Jackson-Thompson, J., Malone, B. R., Goodman, R. M. and Thurman, L. (1997) Developing and evaluating a cardiovascular risk reduction project. American Journal of Health Behavior, 21, 333–344.[Web of Science]
Brownson, R. C., Schmid, T. L., King, A. C., Eyler, A. A., Pratt, M., Murayi, T., Mayer, J. P. and Brown, D. R. (1998) Support for policy intervention to increase physical activity in rural Missouri. American Journal of Health Promotion, 12, 263–266.[Web of Science][Medline]
Cook, R. M. (1995) Missouri Roster 1995–1996: A Directory of State, District, and County Officials. Jefferson City, MO.
Denzin, N. K. (1978) Sociological Methods: A Source Book, 2nd edn. McGraw Hill, New York.
Green, L. W and Kreuter, M. W. (1991) Health Promotion Planning: An Educational and Environmental Approach, 2nd edn. Mayfield, Mountain View, CA, p. 45.
Gregor, S. and Galazka, S. S. (1990) The key use of key informant networks in assessment of community health. Family Medicine, 22, 118–121.[Medline]
Hoppe, R. B., Farquhar, L. J., Henry, R. and Stoffelmayr, B. (1990) Residents' attitudes toward and skills in counseling: using undetected standardized patients. Journal of General Internal Medicine, 5, 415–420.[Web of Science][Medline]
King, A. C. Haskell, W. L., Young, D. R., Oka, R. K. and Stafanic, M. L. (1995) Long-term effects of varying intensities and formats of physical activity on participation rates, fitness, and lipoproteins in men and women aged 50–65 years. Circulation, 91, 2596–2604.
Kofron, P. M., Rastam, L., Pirie, P. L., Holder, W. and Leupker, R. V. (1991) Physician practice for cardiovascular disease risk-factor reduction in six upper Midwestern communities. Journal of Family Practice, 32, 49–55.
Leupker, R. V, Murray, D. M, Jacobs, D. R, Jr, Mittlemark, M. B., Bracht, N., Carlaw, R., Crow, R., Elmer, P., Finnegan, J. and Folsom, A.R. (1994) Community education for cardiovascular disease prevention: risk factor changes in Minnesota Heart Health Program. American Journal of Public Health, 84, 1383–1393.
Lewis, C. E. (1988) Disease prevention and health promotion practices of primary care physicians in the United States. American Journal of Preventative Medicine, 4 (Suppl.), 9–16.
Marti-Costa, S. and Seranno-Garcia, I. (1983) Needs assessments and community development: an ideological perspective. Prevention in Human Services, 2, 75–88.[Medline]
McDonald, J. R, and Natarajan, T. (1989) Community care needs of people with AIDS—the key informant study: a research method for policy development, service planning, and achieving consensus. Journal of Palliative Care, 5, 16–19.
Melton, J. G., Krol, J. (eds) (1994) National Directory of Churches, Synagogues, and other Houses of Worship: Midwestern States. Gale Research, Farmington, MI, vol. 2.
Mittelmark, M. B., Leupker, R. V., Grimm, R., Jr, Kottke, T. E. and Blackburn, H. (1988) The role of physicians in a community-wide program for prevention of cardiovascular disease: the Minnesota Heart Health Program. Public Health Reports, 103, 360–365.[Web of Science][Medline]
Mollica, R. F., Streets, F. J., Boscarino, J. and Redlich, F. C. (1986) A community study of formal pastoral counseling activities among clergy. American Journal of Psychiatry, 143, 323–328.
Morrissey, J. P., Ridgely, M. S., Goldman, H. H. and Bartkow, T. (1994) Assessment of community mental health support systems: a key informant approach. Community Mental Health Journal, 30, 565–579.[Medline]
Olson, L. M., Reis, J., Murphy, L. and Gehm, J. H. (1988) The religious community as a partner in health care. Journal of Community Health, 13, 249–254.[Medline]
Roter, D. L. and Russell, N. K. (1994) Validity of physician self-report in tracking patient education objectives. Health Education Quarterly, 21, 27–38.[Web of Science][Medline]
Rothman, J. and Tropman, J. E. (1987) Models of community organization and macro practice perspectives: their mixing and phasing. In Cox, F. M., Erlich, J. L., Rothman, J. and Tropman, J. E. (eds), Strategies of Community Organization: Macro Practice. Peacock, Itasca, IL, pp. 3–25.
Shea, S. and Basch, C. E. (1990) A review of five major community-based cardiovascular disease prevention programs, part 1: rationale, design, and theoretical framework. American Journal of Health Promotion, 4, 203–218.[Medline]
Schmid, T. L., Pratt, M. and Howze, E. (1995) Policy as intervention: environmental and policy approaches to the prevention of cardiovascular disease. American Journal of Public Health, 85, 1207–1211.
Thomas, N. J., Key, J. D. and Ector, W. L. (1994) Bicycle-related head injuries in South Carolina children: what can we do to prevent them? Journal of the South Carolina Medical Association, 90, 355–359.
Thomas, S. B., Quinn, S. C., Billingsley, A. and Caldwell, C. (1994) The characteristics of northern black churches with community health outreach programs. American Journal of Public Health, 84, 575–579.
US Department of Commerce. (1992) 1990 Census of Population and Housing, Short Form. US Department of Commerce, Bureau of Census, Washington, DC.
US Department of Health and Human Services (1992) National Survey of Worksite Health Promotion Activities. US Government Printing Office, Washington, DC.
US Department of Health and Human Services (1996) Physical Activity and Health: A Report of the Surgeon General. US Department of Health and Human Services, Centers for Disease Control and Prevention, National Center for Chronic Disease Prevention and Health Promotion, Atlanta, GA.
Vega, W. A., Scutchfield, F. D., Karno, M. and Meinhardt, K. (1985) The mental health needs of Mexican American agricultural workers. American Journal of Preventative Medicine, 1, 47–55.
Von Korff, M., Wickizer, T., Maeser, J., O'Leary, P., Pearson, D. and Beery, W. (1992) Community Activation and Health Promotion: Identification of key organizations. American Journal of Health Promotion, 7, 110–117.[Medline]
Ward, J. and Sanson-Fisher, R. (1996) Accuracy of patient recall of opportunistic smoking cessation advice in general practice. Tobacco Control, 5, 110–113.[Abstract]
Warheit, G. J., Bulh, J. M. and Bell, R. A. (1978) A critique of social indicators, analysis and key informants surveys as needs assessment methods. Education and Program Planning, 1, 239–247.
Williams, R. L, Snider, R, and Ryan, M. J. (1994) A key informant `tree' as a tool for community oriented primary care. Family Practice Research Journal, 14, 273–280.[Medline]
CiteULike 
Connotea 
Del.icio.us What's this?
This article has been cited by other articles:
|
|
 |
|
  I. M. Munoz-Baell, C. Alvarez-Dardet, M. T. Ruiz, E. Ferreiro-Lago, and E. Aroca-Fernandez Setting the stage for school health-promoting programmes for Deaf children in Spain Health Promot. Int., December 1, 2008; 23(4): 311 - 327. [Abstract] [Full Text] [PDF]
|
|
|
|
 |
|
 M. A Muhit, S. P Shah, C. E Gilbert, S. D Hartley, and A. Foster The key informant method: a novel means of ascertaining blind children in Bangladesh Br J Ophthalmol, August 1, 2007; 91(8): 995 - 999. [Abstract] [Full Text] [PDF]
|
|
|
|
 |
|
 B. M. Reininger, M. Vincent, S. F. Griffin, R. F. Valois, D. Taylor, D. Parra-Medina, A. Evans, and M. Rousseau Evaluation of Statewide Teen Pregnancy Prevention Initiatives: Challenges, Methods, and Lessons Learned Health Promot Pract, July 1, 2003; 4(3): 323 - 335. [Abstract] [PDF]
|
|
|
|
|
|
L. M. Wen, M. Thomas, H. Jones, N. Orr, R. Moreton, L. King, P. Hawe, J. Bindon, J. Humphries, K. Schicht, et al. Promoting physical activity in women: evaluation of a 2-year community-based intervention in Sydney, Australia Health Promot. Int., June 1, 2002; 17(2): 127 - 137. [Abstract] [Full Text] [PDF]
|
|
| ||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||