Health Education Research, Vol. 16, No. 4, 425-441,
August 2001
© 2001 Oxford University Press
Determinants of implementing heart health promotion activities in Ontario public health units: a social ecological perspective
School of Geography and Geology, McMaster University, Hamilton, Ontario L8S 4K1 and
1 Office of the Vice-President Research, University of Victoria, Victoria, BC V8W 2Y2, Canada
 |
Abstract |
|---|
 Top Abstract IntroductionTheoretical frameworkMethodResultsDiscussionReferences |
|---|
This paper reports the results of a study undertaken to explain levels of implementation of heart health promotion activities observed in Ontario public health agencies in 1997. Organizational-level data were collected by surveying all 42 health departments in 1994, 1996 and 1997 as part of the Canadian Heart Health Initiative Ontario Project. Guided by social ecological and organizational theories, the model examines relationships between implementation and four sets of possible determinants of activity: (1) the predisposition of agencies to undertake heart health promotion activities, (2) their capacity to undertake these activities, (3) internal organizational factors and (4) external system factors. A small set of five variables explains almost half of the variance in implementation (R2 = 0.46): organizational capacity (ß = 0.40), priority given to heart health (ß = 0.36), coordination of programs (ß = 0.19), use of resource centers (ß = 0.12) and participation in networks (ß = 0.09). The results suggest that models integrating organizational and socio-ecological theories can help us understand the implementation of community-based heart health promotion activities by public health agencies. Implications for future research, policy and practice are discussed.
|
Introduction |
|---|
|
TopAbstractIntroductionTheoretical frameworkMethodResultsDiscussionReferences |
|---|
Lifestyle behaviors may explain up to 50% of preventable coronary heart disease mortality (Fries et al., 1993
; Byers et al., 1998) and hence the emphasis on behavior change in heart health promotion programmes. Widespread behavior change in populations requires a full spectrum of effective interventions, from `downstream' interventions focussed on individuals at high risk for illness or with existing symptoms, to `upstream' interventions focussed on macro-public policies (McKinlay, 1995). The `new public health' (Frank, 1995) calls for mid-stream, or population-based, interventions characterized by targeting defined populations for the purpose of changing and/or preventing health-damaging behaviors (McKinlay, 1995; Crichton, 1997). Comprehensive community heart health programs are one application of the new public health. These programs typically aim to change behaviors of defined populations as well as social and physical environments that support healthy behaviors (Shea and Basch, 1990; Advisory Board of the First International Heart Health Conference, 1992; Health and Welfare Canada, 1992; Elder et al., 1993).
For population impact, programs must be effective and have broad reach. Effectiveness of community-based heart health programs has typically been assessed using communities as the unit of intervention and analysis, and placing primary emphasis on behavioral and risk factor outcomes. The earliest projects, which began in the 1970s, report some positive outcomes [cf. (Vartiainen et al., 1994; Schooler et al., 1997)]. Subsequent projects have generally yielded modest and mixed results, with the inability to discern effects attributed, in part, to methodological challenges and secular trends [cf. (Mittelmark et al., 1993; Dobbins and Thomas, 1996; Ebrahim and Smith, 1997; Schooler et al., 1997; Sellers et al., 1997; Viswanath and Finnegan, 1997)]. While recognizing the need to expand the knowledge base on the effectiveness of (heart) health promotion, sufficient evidence supports the widespread application of community-based heart health programs (Cameron et al., 1996; Frankish et al., 1996; Nutbeam, 1996). Furthermore, widespread application requires that heart health programs be integrated into the existing system of public health programs and services (Health and Welfare Canada, 1992; Advisory Board of the Second International Heart Health Conference, 1995).
Implementation of heart health programs among public health agencies, however, is universally low (Advisory Board of the Third International Heart Health Conference, 1998). Yet very few studies address organizational uptake of health promotion activities (Johnson et al., 1996; Orlandi, 1996; Hawe et al., 1997). Those that do typically focus on single interventions (Steckler and Goodman, 1989; Orlandi et al., 1990; Parcel et al., 1990; Rogers, 1995) rather than a cluster of interventions characteristic of comprehensive, community-based health promotion. Also, few studies examine the influence of internal organizational and external system factors on agency practices (Orlandi, 1996; Richard et al., 1996), yet these factors are increasingly recognized as important determinants of organizational performance (Champagne et al., 1993). This paper examines internal organizational and external system factors influencing implementation of heart health promotion activities by public health agencies in Ontario.
The research is part of the Ontario Project of the Canadian Heart Health Initiative (CHHI), described in detail in a previous paper (Elliott et al., 1998). In brief, the CHHI is a multiphase initiative which began in 1986. A policy development phase was followed by provincial surveys of cardiovascular risk factors (MacDonald et al., 1992), and a demonstration phase in which communities within each province developed and evaluated programs for possible widespread application (Stachenko, 1996). A subsequent dissemination phase, completed in Ontario in 1998 and at various stages of development in other provinces, is aiming to increase adoption of best practices in heart health promotion within communities across Canada. The CHHI aims to integrate heart health promotion into the existing public health system.
Consistent with the philosophy and strategies of the CHHI, the dissemination phase of the Ontario Project focussed on factors influencing the dissemination of heart health promotion activities in the formal public health system. Guided mainly by Green and Kreuter (Green and Kreuter, 1991), factors of primary concern were the predisposition (motivation) and capacity (skills and resources) of health departments to implement heart health promotion activities. Data collection involved quantitative and qualitative components. A quantitative survey was administered to all health departments at three points in time (1994, 1996 and 1997), and in-depth interviews were conducted in a subset of health units in 1995 and 1997. Findings reported in this paper build on previous papers that report cross-sectional findings from the quantitative surveys (Elliott et al., 1998; Taylor et al., 1998a,b). An important extension to this work is to conduct a longitudinal analysis to understand levels of implementation of heart health activities. This paper uses path analysis to examine the factors influencing levels of implementation of heart health activities reported by public health agencies in 1997 using survey data from all three points in time from 1994 through 1997.
The setting for the research is the formal public health system in Ontario (Elliott et al., 1998). At the time of data collection, Ontario had 42 local health units, each administered by a local board of health, and regulated by provincial legislation and program guidelines. In 1989, public health in Ontario experienced a strategic shift in programming direction by re-focussing on non-communicable disease prevention, with a particular emphasis on cardiovascular disease (CVD). In addition to existing responsibilities, health units were required to provide extensive programming in tobacco use prevention, nutrition promotion and physical activity promotion (Ontario Ministry of Health, 1989). By 1997, approximately 10% of public health resources in Ontario were targeted to these program areas (Ontario Ministry of Health, 1998). Health units were also required to work collaboratively with a wide variety of local agencies and groups to achieve public health goals (Schabas, 1996).
The change in public health mandate stimulated other structural changes. Local health departments hired staff with a wider range of health promotion skills (e.g. community development, program evaluation, social marketing) and re-organized into multidisciplinary teams. Various networks and coalitions (mostly consisting of agency representatives) were formed at local and provincial levels for heart health and individual issue areas (e.g. tobacco, active living). In addition, a provincial resource system was established to support health promotion activities of public health and other community agencies. The system consisted of over 20 resource organizations, which provided technical assistance in general health promotion skills (e.g. planning, evaluation) and for issue-specific programming (e.g. tobacco, nutrition). The resource system and other provincial level developments related to heart health promotion in Ontario are described in more detail elsewhere (Riley, in preparation).
|
Theoretical framework |
|---|
|
TopAbstractIntroductionTheoretical frameworkMethodResultsDiscussionReferences |
|---|
Previous work on the dissemination of health promotion programs draws primarily on diffusion theory, organizational theory and individual behavior change theories applied to organizations (Parcel et al., 1990
; Orlandi et al., 1990; Orlandi, 1996; Nutbeam and Harris, 1998). A consistent conclusion from this work is the need to view organizations from a systems or ecological perspective, whereby the interactions among organizations and their environment are a central focus. This study consolidates a diverse literature, but draws most heavily on a socio-ecological perspective (Green et al., 1996), recognizing the importance of the context in which agencies undertake health promotion activities.
The theoretical framework for this study is illustrated in Figure 1. Although the goal is improved health of the population, the outcome of interest in this study is implementation of comprehensive, community-based programs to prevent CVD and promote heart health. While recognizing a wide range of factors determining cardiovascular health (Evans and Stoddart, 1990; Lomas, 1998), community-based programs typically focus on changing health behaviors and social/physical environments to support healthy behaviors (Ontario Ministry of Health, 1993). A comprehensive approach would address multiple behaviors (notably, tobacco use, physical inactivity, unhealthy diet), target populations in several community locations (e.g. schools, workplaces, health care settings) and use a variety of population-based approaches (e.g. community-wide education, environmental and policy initiatives) (Burns, 1991; Elder et al., 1993; Nutbeam, 1996).
|
A diverse literature suggests implementation by organizations is influenced by aspects of motivation, characteristics of the organization (e.g. skills, resources, structures, processes), and the environment in which organizations function. According to the provisional framework for this study, implementation is most directly influenced by (1) organizational predisposition and (2) organizational capacity.
Following Green and Kreuter (Green and Kreuter, 1991), predisposition refers to the motivation to undertake heart health promotion activities. Even though heart health promotion activities are part of the legislated public health mandate, health departments are locally autonomous units and can choose to delay implementation or move at a slower (or faster) rate. The importance of a shared commitment among staff to organizational directions is increasingly recognized as an important precondition for effective organizational action (Rogers, 1995; Goodman et al., 1998; Senge, 1999).
In this study, capacity refers to the skills and resources of public health agencies to undertake heart health promotion activities (Green and Kreuter, 1991; Clark and McLeroy, 1995). Our view of capacity was informed by literature on efforts to strengthen the public health system in the US (Institute of Medicine, 1988; Roper et al., 1992), and capacity building for community-based CVD and other prevention programs (Kreuter, 1992; Schwartz et al., 1993; Steckler et al., 1997). In this literature, there is general agreement that the organization must be able to effectively assess, plan, prioritize, organize, implement, evaluate, adjust and maintain organizational initiatives. Accordingly, our notion of capacity refers to how well public health agencies conduct a set of organizational practices related to assessing, planning, organizing resources to support implementation and evaluating heart health promotion activities (see Method). The most recent literature, which post-dates the definition and measurement of constructs for our research, defines capacity as a more global construct, comprised of aspects of motivation, organizational structures and processes, and the environment (Hawe et al., 1997; Goodman et al., 1998). The framework (Figure 1
) includes all of these dimensions, but as separate constructs rather than as dimensions of a global concept of capacity.
Guided by a social ecological view, we propose that organizational predisposition and capacity are influenced by a variety of factors related to the internal organization as well as the external system. With respect to the former, appropriate financial and human resources are key (Hoover and Schwartz, 1992; Ornstein et al., 1992; Schwartz et al., 1993; Hawe et al., 1997). Also, organizational structures and processes must encourage a focus on heart health promotion, and facilitate multidisciplinary activities, collaborative planning with community agencies and coordination of individual programs related to heart health (e.g. tobacco, nutrition, physical activity) (Kaluzny and Hernandez, 1988; Green and Kreuter, 1991; Goodman et al., 1998). A final dimension of internal organizational factors is leadership, with the type and strength of leadership provided by medical officers being particularly relevant (Becker, 1970; Schwartz et al., 1993). However, opinion leadership and champions for heart health promotion can emerge from any level within the organization and can strongly influence organizational performance (Rogers, 1995).
With respect to the external system, interorganizational relationships (partnerships) are especially relevant. There is widespread recognition of the need for public health agencies to work effectively with other service providers (Bracht and Kingsbury, 1990; Butterfoss et al., 1993; McLeroy et al., 1994; Steckler et al., 1997; Goodman et al., 1998), citizens (Goodman et al., 1998) and organizations at other levels (e.g. federal and provincial) (Green et al., 1996; Steckler et al., 1997) to plan and carry out health promotion activities.
The external system also includes activities to support community (heart) health promotion by public health agencies (Florin et al., 1993; Schwartz et al., 1993; Jackson et al., 1994). The primary purpose of these activities is to enhance the knowledge, skills and resources for local organizations and groups to conduct effective health promotion. Major support functions include consultation and training, rewards and incentives, and feedback on performance. Feedback on performance, consistent with Green and Kreuter's (Green and Kreuter, 1991) reinforcing factors, is especially important for sustainability.
Many other characteristics at different levels (e.g. local/regional, provincial, federal) may also influence heart health promotion of public health agencies. These are referred to as contextual factors in the framework, and include social/physical characteristics of communities, community priorities, and trends in the health and social policy environment (Green et al., 1996; Robinson and Elliott, 1999).
|
Method |
|---|
|
TopAbstractIntroductionTheoretical frameworkMethodResultsDiscussionReferences |
|---|
Data collection
Quantitative surveys were conducted in all 42 public health units in Ontario in order to measure organizational level predisposition, capacity and implementation of community-based heart health promotion activities in 1994, 1996 and 1997. In December 1994, a two-stage Survey of Capacities, Activities and Needs (SCAN) of Ontario public health units was administered, and is described in detail in a previous paper (Elliott et al., 1998
). The first stage of the SCAN measured levels of implementation of community-based heart health activities over the previous year in the geographic areas served by the public health units. An organizational response was completed jointly by the local Medical Officer of Health (MOH) and staff most involved in heart health promotion. The response rate was 100%. The second-stage survey in 1994 measured organizational capacity and predisposition for heart health promotion as well as internal organizational factors (e.g. human and financial resources, leadership for heart health) and external system factors (e.g. community interest, partnerships). Surveys were completed by the MOH and approximately six additional unit staff (representing between 1–14% of public health staff) nominated by the MOH as those most familiar with managing and/or delivering heart health activities. The response rate was 90% (n = 262). In December 1996 and January 1998, the surveys were repeated to determine heart health predisposition, capacity and implementation for the preceding 12 months, as well as factors related to the external system and the internal organization of the health unit. Thus, comparable data were collected for all 42 health units over three time periods. The main difference between the first and subsequent surveys was that the corporate and individual surveys administered in 1994 were combined into a single instrument which was completed corporately (i.e. one survey from each unit) by the MOH and staff most involved with heart health promotion (six staff per unit on average). Consistent with guidelines for collecting organizational data (Steckler et al., 1997), the comparability of the data is increased by the overlap of the respondents (59% from 1994 to 1996, 68% from 1996 to 1997 and 46% from 1994 to 1997), the similar distribution of positions within units over time (medical officers, directors/managers, program staff) and verification of data by respondents.
Definitions of variables and measures
Predisposition
Predisposition refers to the motivation to undertake heart health promotion activities. It was operationally defined as a collective belief among staff in the importance of the organization conducting a set of public health activities to support community-wide implementation of heart health promotion activities. The primary indicator of predisposition was importance ratings of 18 organizational practices supportive of heart health, categorized into four areas: assessment, planning, activities to support implementation and evaluation (Table I). The selection of organizational practices was informed by: (1) expert consultation with public health professionals and researchers within and outside Ontario, (2) information on the process to develop performance indicators for public health in the US [cf. (Turnock et al., 1994)], and (3) literature on community organization processes [cf. (Bracht and Kingsbury, 1990)].
|
Predisposition was calculated as the mean of 18 organizational practices, each rated on a four-point scale from `not at all important' to `very important'. In 1994, average scores from individuals within units were used as corporate scores after confirming strong correlations between individual scores and within-unit means using the procedures described by James (James, 1982
) and Shrout and Fleiss (Shrout and Fleiss, 1979). The mean score on four subscales was used to construct a multi-item scale. Levels of predisposition were consistently high from 1994 to 1997 (rated as `very important') (Table II). Variability across units was low at all three data collection times. The multi-item scale yielded good internal consistency with 
coefficients ranging from 0.61 to 0.87 from 1994 to 1997. Construct validity was established by: (1) expert review, (2) positive feedback from respondents, and (3) fairly consistent relationships between predisposition and capacity (Table II) and predisposition and implementation (Table IV).
|
|
Capacity
Capacity refers to skills and resources required to implement community-based heart health activities. It was operationally defined as effectiveness in performing organizational practices to support heart health promotion activities and was measured by effectiveness ratings on a five-point scale from `not aware activity was conducted' to `activity was conducted and was very effective'. Item and scale construction were the same as for predisposition (Table I
).
Levels of capacity were low to moderate between `somewhat' and `fairly' effective, and increased over time (Table II). Internal consistency was high with coefficients ranging from 0.84 to 0.92 from 1994 to 1997. Construct validity was established using several methods: expert review of measures; positive reviews by respondents; a high correlation between a global rating of capacity in 1997 (on a five-point scale from low to high) and the multi-item score (r = 0.45, P = 0.003); and consistent and strong correlations between capacity and predisposition (Table II) as well as capacity and implementation (Table IV).
Implementation
Implementation, in this study, refers to the performance of community-based heart health activities. Respondents rated levels of implementation for 75 activities, organized by risk factor and setting, on a five-point scale from `not aware of any organized activity being planned or implemented' to a `high level of implementation' (Table III). Ratings were made relative to `full implementation', defined as `ideal implementation in your community, not just to the extent that resources allow'.
|
The inventory of heart health activities incorporated risk factors, settings and approaches and was adapted from the US Public Health Service comprehensive approach to tobacco control (Burns, 1991
). Risk factors are behavioral targets for change, and included tobacco use reduction, nutrition promotion, physical activity promotion and promotion of heart health in general (i.e. general heart health or two or more of the other risk factors). Settings are locations for program activities, and included schools, workplaces, health care settings and the community at large. Approaches are the strategies to produce change, and included education, environmental support and policy initiatives.
Average scores were calculated for overall implementation. Implementation increased from 1994 to 1997, with the average health unit at a low level of implementation by 1997 (Table IV). Internal consistency for overall implementation was high, with coefficients ranging from 0.75 to 0.94 for the three measurement times. Evidence of construct validity was from: (1) expert review, (2) a strong correlation between a global rating of implementation in 1997 (on a five-point scale) and the multi-item scale (r = 0.61, P = 0.000), (3) positive assessments from respondents, (4) a high correlation between implementation in 1996 and 1997, and (5) consistent relationships between predisposition and implementation as well as capacity and implementation (Table IV).
Internal organizational factors
Internal organizational factors refer to resources, structures, processes and leadership within health departments. Facilitators and barriers address all dimensions of internal organizational factors shown in Figure 1 and specific items are listed in Table V. Additional indicators were developed for all dimensions except leadership. Indicators of financial resources included: whether or not the health department had a budget line for heart health and budget per capita. Health units were unable to estimate resource allocation for heart health activities since these activities are part of several and variably defined program areas (e.g. healthy growth and development, healthy lifestyles). Indicators of human resources included: staff time spent on heart health and working with volunteers. Indicators of organizational structure included: coordination of programs within the health unit and priority of heart health in the organization. Table V shows how the indicators were measured, scoring procedures for the path analysis and the range of scores.
|
External system factors
External system factors refer to conditions and other factors beyond the direct control of the health departments, and include partnerships, support from resource organizations and contextual factors. Facilitators and barriers address all dimensions of external system factors in Figure 1
. An additional indicator of partnerships was participation in networks. Support from resource organizations was measured by use and usefulness of resource centers and helpfulness of supports offered by the Canadian Heart Health Initiative Ontario Project (CHHIOP). Indicators of contextual factors included: population served by each health unit, duration of a heart health program in the community and priority of heart health in the community. Table V shows how the indicators are measured, scoring procedures for the path analysis and the range of scores.
Path analysis procedures
Path analysis is a statistical method that builds on multiple regression techniques and is often used with exploratory models. It was used to estimate the direct and indirect effects of external system factors, internal organizational factors, predisposition and capacity on 1997 levels of implementation. A central assumption was that scores on explanatory variables over time (rather than scores at one point in time) would provide the most theoretically plausible explanation for 1997 levels of implementation. This assumption is based on the process of change characteristic of health promotion programs undertaken using a community development approach (Mittelmark et al., 1993; Frankish and Green, 1994; Nutbeam and Harris, 1998). That is, the development and implementation process is often extended over a period of years and requires sustained activity from a number of agencies. Consequently, scores on factors, such as the amount of partnering with community agencies, that reflect the time period from 1994 through 1997 provide a stronger basis for explanation of 1997 levels of implementation rather than single scores at any one point in time. Candidate variables to help explain implementation, therefore, were composite scores using survey data from all three time periods (Table V). The limited degrees of freedom due to the small number of observations (n = 42) also influenced how variables were constructed. Composite measures were created to optimize the use of data and reduce the number of variables. Such composite measures, however, may mask embedded relationships. A case in point is the measures of facilitators and barriers. Four aggregate scores were computed: internal and external classes for each of facilitators and barriers. These aggregate measures take into account all items (15 facilitators, 14 barriers) and reflect overall scores on a wide range of factors helping or impeding progress. Correlations with individual items were also explored to better understand factors that contribute most to the aggregate measures.
A staged modeling approach was used whereby separate regression models were estimated for predisposition, capacity and implementation. At each stage, correlation analyses were performed to identify candidate variables for inclusion in the regression model (using P < 0.10 as the inclusion criterion to prevent premature elimination of variables). Correlation screening was used because of a high number of candidate variables and relatively few degrees of freedom. The bivariate correlations between all variables in the model are shown in Table VI.
|
For each dependent variable (i.e. predisposition, capacity, implementation), models were estimated for external system factors and internal organizational factors separately and then in combination. The final model for capacity also included predisposition and the final model for implementation included both predisposition and capacity. This cumulative approach was used in order to provide insight as to how the variables behave individually as well as in combination with respect to the outcome of interest.
|
Results |
|---|
|
TopAbstractIntroductionTheoretical frameworkMethodResultsDiscussionReferences |
|---|
Modeling predisposition to undertake heart health promotion activities
The dependent variable was the mean importance rating on 18 organizational practices related to heart health throughout 1996 and 1997. The explanatory variables were those with statistically significant correlations with the dependent variable (P < 0.10). In the separate model estimated with external system factors, usefulness of resource centers (X9) was retained in the model (ß = 0.30; t[41] = 1.98, P = 0.06; R2 = 0.09). Priority of heart health in the organization (X15) was the only variable retained for internal organizational factors and in the combined model, with the same statistical result (ß = 0.34; t[41] = 2.28, P = 0.03; R2 = 0.12) in both models.
Modeling capacity to undertake heart health promotion activities
The dependent variable was the mean effectiveness rating (over 2 years) of 18 organizational practices related to heart health and the explanatory variables were those significantly correlated with the primary indicator of capacity. In the first model two of four external system factors were retained: participation in networks (X5) (ß = 0.37; t[41] = 2.67, P = 0.01) and use of resource centers (X8) (ß = 0.35; t[41] = 2.54, P = 0.02) with an R2 of 0.36. In the next model, coordination of programs (X12) (ß = 0.43; t[41] = 2.57, P = 0.01) and priority of heart health in the organization (X15) (ß = 0.30; t[41] = 1.77, P = 0.08) were significant (R2 = 0.45). The combined model yielded an R2 of 0.57, and maintained participation in networks (X5) (ß = 23; t[41] = 1.92, P = 0.06), use of resource centers (X8) (ß = 0.29; t[41] = 2.45, P = 0.02) and coordination of programs (X12) (ß = 0.48; t[41] = 4.23, P = 0.00).
Modeling implementation of heart health promotion activities
The dependent variable was the mean implementation score for 75 community-based heart health activities. The explanatory variables were those with statistically significant correlations with mean implementation; variables carried forward from the previous models for predisposition (priority of heart health in the organization) and capacity (participation in networks, use of resource centers, and coordination of programs); and mean importance and mean effectiveness ratings as indicators of predisposition and capacity, respectively. In the model estimated using external system factors, priority of heart health in the community (X7) (ß = 0.34; t[41] = 2.27, P = 0.03) and usefulness of resource centers (X9) (ß = 0.33; t[41] = 2.27, P = 0.03) were retained (R2 = 0.18). In the model estimated using internal organizational factors, both coordination of programs (X12) (ß = 0.34; t[41] = 2.02, P = 0.05) and priority of heart health in the organization (X15) (ß = 0.36; t[41] = 2.11, P = 0.04) were retained (R2 = 0.42). The combined model yielded an R2 of 0.46 and the significant predictors of overall implementation in 1997 were capacity (X19) (ß = 0.40; t[41] = 2.76, P = 0.01) and priority of heart health in the organization (X15) (ß = 0.36; t[41] = 2.49, P = 0.04). The direct and indirect effects on 1997 levels of implementation are displayed in Figure 2. Indirect effects were calculated by multiplying the ß weights for direct effects of explanatory variables on capacity (0.23, 0.29 and 0.48 for X5, X8 and X12, respectively) and the ß weight for the direct effect of capacity on implementation (0.40). Analyses were repeated using a weighted implementation score based on the role of the health unit (i.e. lead/support/no role) in undertaking community-based heart health promotion activities. The strength of associations decreased with the weighted scores.
|
|
Discussion |
|---|
|
TopAbstractIntroductionTheoretical frameworkMethodResultsDiscussionReferences |
|---|
This paper developed a path model to explain 1997 levels of implementation of heart health promotion activities in Ontario's 42 health unit jurisdictions. Organizational-level data were collected by surveying all health departments in 1994, 1996 and 1997. Informed by ecological and organizational theory, 19 explanatory variables were used to estimate path models. One primary indicator was used for each of predisposition and capacity, and other variables were grouped into factors related to the external system in which public health agencies operate and factors related to the internal organization of public health agencies. The final model includes five variables that explain almost half of the variance (i.e. 46%) in 1997 levels of implementation of heart health promotion activities in Ontario public health units. These are strong results, especially given the exploratory nature of the work, statistical limitations and the complexity of the public health system (Champagne et al., 1993
; Crichton, 1997).
The path model supports a number of relationships hypothesized in our preliminary framework. It supports a strong and direct relationship between capacity and implementation. This finding is consistent with Champagne et al. (Champagne et al., 1993), who examined the influence of organizational and environmental factors on performance of public health agencies in Quebec, and found a strong relationship between organizational practices (referred to as capacity in our model) and organizational performance (implementation in our model). Our result is also consistent with the presumed link between capacity and implementation in the health promotion literature; however, our measure of capacity was limited to organizational practices and did not include the multiple dimensions recently proposed by others (Hawe et al., 1997; Goodman et al., 1998). In our research, other dimensions of capacity, such as motivation, organizational structure and contextual factors, were defined and measured as separate constructs.
The path model also indicates that external system and internal organizational factors impact on implementation primarily by influencing organizational practices to support heart health promotion (i.e. capacity). Of the external system factors, partnerships with other local agencies (measured by participation in networks) and support from resource organizations were most strongly related to the effectiveness of organizational practices. The central importance of partnerships in health promotion is now well-recognized (Bracht and Kingsbury, 1990; Advisory Board of the First International Heart Health Conference, 1992; Schwartz et al., 1993; Nutbeam and Harris, 1998); however, little empirical work has confirmed relationships between partnerships and organizational practices. Similarly, although the literature on resources to support community-based health promotion is expanding (Florin et al., 1993; Schwartz et al., 1993; Jackson et al., 1994), few studies demonstrate an empirical link between such resources and performance of local agencies. Of the internal organizational factors, organizational structure, measured by coordination of programs within public health units, was shown to have the strongest relationship to capacity. This finding may support new organizational models in public health agencies. That is, recent shifts away from traditional disciplinary activities towards more integrated and multidisciplinary programming that targets specific problems or goals may facilitate implementation, assuming these shifts enhance program coordination. Other indicators of organizational structure are needed to strengthen this conclusion.
A direct influence of internal organizational factors on implementation was also supported. Specifically, priority given to heart health promotion within the public health organization had a direct and strong relationship with implementation. This finding suggests that a shared commitment to organizational priorities impacts directly on implementation. Literature on organizational performance, including relatively recent literature on learning organizations, supports this finding (Senge, 1999). Nevertheless, practical implications have not been considered in depth. For example, how do public health agencies most effectively develop priorities and a shared commitment to them? How many priorities can be addressed with finite resources? The results of the path model raise the importance of these practical issues for public health professionals.
Predisposition, measured by importance ratings of public health practices to support heart health promotion, was not retained in the final model. The most plausible reasons are its high scores and low variability across units. Predisposition, therefore, may be important even though the path modeling procedures were unable to demonstrate hypothesized relationships between predisposition, other explanatory variables and implementation. In addition, it may be that predisposition is more important at earlier or later stages in the dissemination process (e.g. adoption) and less relevant during the implementation stage. Further study is warranted on the role of predisposition at various stages in the dissemination process.
The path analysis undertaken in this paper represents an advance over previous, related research in its attempt to specify structural relationships between various explanatory variables and implementation, and to distinguish between their direct and indirect effects. No claim is made to identify causal relationships, but rather to clarify the links among multiple factors influencing levels of implementation. The results suggest that a model integrating organizational and socio-ecological theories can help us understand movement within the implementation stage of disseminating mandated, community-based, heart health promotion activities among public health agencies. Conceptually, priority given to heart health in the organization and organizational capacity (i.e. effectiveness of organizational practices to support heart health) exert a direct influence on implementation. Consistent with socio-ecological approaches to health promotion, internal organizational factors and external system factors influence implementation indirectly through organizational capacity.
The structural relationships among variables suggest that a useful research strategy is to continue to `unbundle' capacity, i.e. to examine relationships among the multiple dimensions of a global concept of capacity recently proposed by others (Hawe et al., 1997; Goodman et al., 1998). Further study is needed to examine relationships not supported in the path model, notably the role of financial and human resources, leadership, and contextual factors. Qualitative findings suggest that these factors exert a strong influence on implementation of heart health activities; however, the interplay of these factors needs further study. In order to substantially increase the application of findings, other useful directions are to examine the extent to which similar variables (1) influence other areas of health promotion practice (e.g. injury prevention) and (2) operate in other jurisdictions.
Results also have implications for policy and practice. Based on the results of the path model, the practitioner who wants to increase implementation of heart health promotion activities would make heart health an organizational priority, and strengthen organizational practices to assess, plan, mobilize resources for implementation and evaluate heart health promotion activities. Primary strategies to improve these practices would be to participate in networks, access support from the resource system and coordinate individual programs related to heart health (e.g. tobacco, nutrition, physical activity) within the health unit. Provincial public health authorities with an interest in enhancing dissemination of heart health promotion activities would ensure supports are available to strengthen the priority given to heart health by public health agencies and organizational practices supporting heart health activities. Policy makers would also encourage the integration of program delivery within health units.
|
Acknowledgments |
|---|
The authors wish to thank Susan Watt and Josie d'Avernas for their helpful comments on an earlier version of this paper.
|
References |
|---|
|
TopAbstractIntroductionTheoretical frameworkMethodResultsDiscussionReferences |
|---|
Advisory Board of the First International Heart Health Conference (1992) Bridging the Gap: Science and Policy in Action. Declaration of the First International Heart Health Conference, Victoria, BC.
Advisory Board of the Second International Heart Health Conference (1995) The Catalonia Declaration: Investing in heart health. Declaration of the Second International Heart Health Conference, Barcelona.
Advisory Board of the Third International Heart Health Conference (1998) The Singapore Declaration: Forging the Will for Heart Health in the Next Millennium. Declaration of the Third International Heart Health Conference, Singapore.
Becker, M. H. (1970) Factors affecting diffusion of innovations among health professionals. American Journal of Public Health, 60, 294–304.
Bracht, N. and Kingsbury, L. (1990) Community organization principles in health promotion: a five-stage model. In N. Bracht (ed.), Health Promotion at the Community Level. Sage, Newbury Park, CA, pp. 66–88.
Burns, D. M. (1991) The scientific rationale for comprehensive community-based smoking control strategies. In Strategies to Control Tobacco Use in the United States: A Blueprint for Public Health Action in the 1990s. Smoking Control Monograph 1. US DHHS, Public Health Service, Rockville, MD.
Butterfoss, F. D., Goodman, R. M. and Wandersman, A. (1993) Community coalitions for prevention and health promotion. Health Education Research, 8, 315–330.
Byers, T., Anda, R., McQueen, D., Williamson, D., Mokdad, A., Casper, M., Ford, E. and Marks, J. (1998) The correspondence between coronary heart disease mortality and risk factor prevalence among states in the United States, 1991–1992. Preventive Medicine, 27, 311–316.[ISI][Medline]
Cameron, R. J., Brown, K. S. and Best, J. A. (1996) The dissemination of chronic disease prevention programs: linking science and practice. Canadian Journal of Public Health, 87, S50–S53.
Champagne, F., Leduc, N., Denis, J.-L. and Pineault, R. (1993) Organizational and environmental determinants of the performance of public health units. Social Science and Medicine, 37, 85–95.
Clark, N. M. and McLeroy, K. R. (1995) Creating capacity through health education: what we know, and what we don't. Health Education Quarterly, 22, 273–289.[ISI][Medline]
Crichton, A. (1997) Health Care: A Community Concern? Developments in the Organization of Canadian Health Services. University of Calgary Press, Calgary, Alberta.
Dobbins, M. and Thomas, H. (1996) The Effectiveness of Community-Based Heart Health Projects: A Systematic Overview (Working Paper Series 96-1). Quality of Nursing Worklife Research Unit, McMaster University and University of Toronto, Hamilton, Ontario.
Ebrahim, S. and Smith, G. D. (1997) Systematic review of randomised controlled trials of multiple risk factor interventions for preventing coronary heart disease. British Medical Journal, 314, 1666–1674.
Elder, J. P., Schmid, T. L., Dower, P. L. and Hedlund, S. (1993) Community heart health programs: components, rationale and strategies for effective interventions. Journal of Public Health Policy, 14, 463–479.[Medline]
Elliott, S. J., Taylor, S. M., Cameron, R. J. and Schabas, R. (1998) Assessing public health capacity to support community-based heart health promotion: the Canadian Heart Health Initiative, Ontario Project (CHHIOP). Health Education Research, 13, 607–622.
Evans, R. G. and Stoddart, G. L. (1990) Producing health, consuming health care. Social Science and Medicine, 31, 1347–1363.
Florin, P., Mitchell, R. and Stevenson, J. (1993) Identifying training and technical assistance needs in community coalitions: a developmental approach. Health Education Research, 8, 417–432.
Frank, J. W. (1995) Why `population health'? Canadian Journal of Public Health, 86, 162–164.[ISI][Medline]
Frankish, C. J. and Green, L. W. (1994) Organizational and community change as the social scientific basis for disease prevention and health promotion policy. Advances in Medical Sociology, 4, 209–233.
Frankish, J., Green, L. W., MacLean, D. and Stachenko, S. (1996) Canadian Conference on Dissemination Research. Proceedings from the Canadian Conference on Dissemination Research, Vancouver, BC, 1995.
Fries, J. F., Koop, C. E., Beadle, C. E., Cooper, P. P., England, M. J., Greaves, R. F., Sokolov, J. J. and Wright, D. (1993) Reducing health care costs by reducing the need and demand for medical services. The Health Project Consortium. New England Journal of Medicine, 329, 321–325.
Goodman, R. M., Speers, M. A., McLeroy, K., Fawcett, S., Kegler, M., Parker, E., Smith, S. R., Sterling, T. D. and Wallerstein, N. (1998) Identifying and defining the dimensions of community capacity to provide a basis for measurement. Health Education and Behavior, 25, 258–278.[Abstract]
Green, L. W. and Kreuter, M. W. (1991) Health Promotion Planning: An Educational and Environmental Approach, 2nd edn. Mayfield, Mountain View, CA.
Green, L. W., Richard, L. and Potvin, L. (1996) Ecological foundations of health promotion. American Journal of Health Promotion, 10, 270–281.[ISI][Medline]
Hawe, P., Noort, M., King, L. and Jordens, C. (1997) Multiplying health gains: the critical role of capacity building within health promotion programs. Health Policy, 39, 29–42.[ISI][Medline]
Health and Welfare Canada (1992) The Canadian Heart Health Initiative: a policy in action. Health Promotion, 30, 2–19.
Hoover, S. and Schwartz, R. (1992) Diffusing PATCH through interagency collaboration. Journal of Health Education, 23, 160–162.
Institute of Medicine (1988) The Future of Public Health. National Academy Press, Washington, DC.
Jackson, C., Fortmann, S. P., Flora, J. A., Melton, R. J., Snider, J. P. and Littlefield, D. (1994) The capacity-building approach to intervention maintenance implemented by the Stanford Five-City Project. Health Education Research, 9, 385–396.
James, L. R. (1982) Aggregation bias in estimates of perceptual agreement. Journal of Applied Psychology, 67, 219–229.
Johnson, J. L., Green, L. W., Frankish, C. J., MacLean, D. R. and Stachenko, S. (1996) A dissemination research agenda to strengthen health promotion and disease prevention. Canadian Journal of Public Health, 87, S5–S10.
Kaluzny, A. D. and Hernandez, S. R. (1988) Organizational change and innovation. In Shortell, S. M. and Kaluzny, A. D. (eds), Health Care Management: A Text in Organizational Theory and Behavior, 2nd edn. Wiley, New York, pp. 379–417.
Kreuter, M. W. (1992) PATCH: its origin, basic concepts, and links to contemporary public health policy. Journal of Health Education, 23, 135–139.
Lomas, J. (1998) Social capital and health: implications for public health epidemiology. Social Science and Medicine, 47, 1181–1188.
MacDonald, S., Joffres, M., Stachenko, S., Horlick, L. and Fodor, G. (1992) Multiple cardiovascular disease risk factors in Canadian adults. Canadian Medical Association Journal, 146, 2021–2029.[Abstract]
McKinlay, J. B. (1995) The new public health approach to improving physical activity and autonomy in older populations. In Heikkinen, E., Kuusinen, J. and Ruoppila, I.. (eds), Preparation for Aging. Plenum Press, New York.
McLeroy, K. R., Kegler, M., Steckler, A., Burdine, J. M. and Wisotzky, M. (1994) Community coalitions for health promotion: summary and further reflections (Editorial). Health Education Research, 9, 1–11.
Mittelmark, M. B., Hunt, M. K., Heath, G. W. and Schmid, T. L. (1993) Realistic outcomes: lessons from community-based research and demonstration programs for the prevention of cardiovascular disease. Journal of Public Health Policy, 14, 437–462.[Medline]
Nutbeam, D. (1996) Achieving best practice in health promotion: improving the fit between research and practice. Health Education Research, 11, 317–326.
Nutbeam, D. and Harris, E. (1998) Theory in a Nutshell: A Practitioner's Guide to Commonly Used Theories and Models in Health Promotion. University of Sydney, Sydney.
Ontario Ministry of Health (1989) Mandatory Health Programs and Services Guidelines for Local Boards of Health. Queen's Printer, Toronto, Ontario.
Ontario Ministry of Health (1993) Opportunities for Health: Promoting Heart Health. Report of the Chief Medical Officer of Health. Queen's Printer, Toronto, Ontario.
Ontario Ministry of Health (1997) Mandatory Health Programs and Services Guidelines for Local Boards of Health. Queen's Printer, Toronto, Ontario.
Orenstein, D., Nelson, C., Speers, M., Brownstein, J. N. and Ramsey, D. C. (1992) Synthesis of the four PATCH evaluations. Journal of Health Education, 23, 187–192.
Orlandi, M. A. (1996) Health promotion technology transfer: organizational perspectives. Canadian Journal of Public Health, 87 (Suppl. 2), S28–S33.
Orlandi, M. A., Landers, C., Weston, R. and Haley, N. (1990) Diffusion of health promotion innovations. In Glanz, K., Frances, M. L. and Rimer, B. K. (eds), Health Behavior and Health Education. Jossey-Bass, San Francisco, CA, pp. 288–313.
Parcel, G. S., Perry, C. L. and Taylor, W. C. (1990) Beyond demonstration: diffusion of health promotion innovations. In Bracht, N. (ed.), Health Promotion at the Community Level. Sage, Newbury Park, CA, pp. 229–251.
Richard, L., Potvin, L., Kishchuk, N., Prlic, H. and Green, L. W. (1996) Assessment of the integration of the ecological approach in health promotion programs. American Journal of Health Promotion, 10, 318–328.[ISI][Medline]
Robinson, K. and Elliott, S. J. (1999) Community development approaches to health promotion: a geographical perspective. Professional Geographer, 51, 283–295.
Rogers, E. M. (1995) Diffusion of Innovations, 4th edn. The Free Press, New York.
Roper, W. L., Baker, E. L., Dyal, W. W. and Nicola, R. M. (1992) Strengthening the public heath system. Public Health Reports, 107, 609–615.[ISI][Medline]
Schabas, R. (1996) Promoting heart health promotion. Canadian Journal of Public Health, 87 (Suppl. 2), S54–S56.
Schooler, C., Farquhar, J. W., Fortmann, S. P. and Flora, J. (1997) Synthesis of findings and issues from community prevention trials. Annals of Epidemiology, 7 (Suppl.), S54–S68.
Schwartz, R., Smith, C., Speers, M. A., Dusenbury, L. J., Bright, R., Hedlund, S., Wheeler, F. and Schmidt, T. L. (1993) Capacity building and resource needs of state health agencies to implement community-based cardiovascular disease programs. Journal of Public Health Policy, 14, 480–494.[Medline]
Sellers, D. E., Crawford, S. L., Bullock, K. and McKinlay, J. B. (1997) Understanding the variability in the effectiveness of community heart health programs: a meta-analysis. Social Science and Medicine, 44, 1325–1339.
Senge, P. (1999) The Dance of Change: The Challenges of Sustaining Momentum in Learning Organizations. Doubleday, New York.
Shea, S. and Basch, C. E. (1990) A review of five major community-based cardiovascular disease prevention programs. Part I: rationale, design, and theoretical framework. American Journal of Health Promotion, 4, 203–213.[Medline]
Shrout, P. E. and Fleiss, J. L. (1979) Intra class correlations: uses in assessing rater reliability. Psychological Bulletin, 86, 420–428.[ISI]
Stachenko, S. (1996) The Canadian Heart Health Initiative: dissemination perspectives. Canadian Journal of Public Health, 87(S2), S57–S59.
Steckler, A. and Goodman, R. M. (1989) How to institutionalize health promotion programs. American Journal of Health Promotion, 3, 34–44.
Steckler, A., Goodman, R. M. and Alciati, M. H. (1997) Collecting and analysing organizational level data for health behavior research (Editorial). Health Education Research, 12, i–iii.
Taylor, S. M., Elliott, S. J. and Riley, B. L. (1998a) Heart health promotion: predisposition, capacity and implementation in Ontario public health units, 1994–96. Canadian Journal of Public Health, 89, 410–414.[ISI][Medline]
Taylor, S. M., Elliott, S. J., Robinson, K. and Taylor, S. J. (1998b) Community-based heart health promotion: facilitators and barriers. Canadian Journal of Public Health, 89, 406–409.[ISI][Medline]
Turnock, B. J., Handler, A., Dyal, W. W., Christenson, G., Vaughn, E. H., Rowitz, L., Munson, J. W., Balderson, T. and Richards, T. B. (1994) Implementing and assessing organizational practices in local health departments. Public Health Reports, 109, 478–484.[ISI][Medline]
Vartiainen, E., Puska, P., Jousilahti, P., Korhonen, H., Tuomilehto, H. and Nissinen, A. (1994) Twenty-year trends in coronary risk factors in North Karelia and in other areas of Finland. International Journal of Epidemiology, 23, 495–504.
Viswanath, K. and Finnegan, J. (1997) Reflections on community health campaigns: secular trends and the capacity to effect change. Paper presented at the Health Communication Conference, Annenberg School for Communication, University of Pennsylvania.
Received on March 6, 2000; accepted on October 10, 2000