Health Educ. Res. 1998 Shediac Rizkallah 87 108

HEALTH EDUCATION RESEARCHTheory & Practice

Vol.13 no.l 1998Pages 87-108

Planning for the sustainability of community-basedhealth programs: conceptual frameworks and future

directions for research, practice and policy

Mona C. Shediac-Rizkallah and Lee R. Bone

Abstract Introduction

Attention to the sustainability of healthintervention programs both in the US andabroad is increasing, but little consensus existson the conceptual and operational definitions ofsustainability. Moreover, an empiricalknowledge base about the determinants ofsustainability is still at an early stage. Planningfor sustainability requires, first, a clearunderstanding of the concept of sustainabilityand operational indicators that may be used inmonitoring sustainability over time. Importantcategories of indicators include: (1) maintenanceof health benefits achieved through an initialprogram, (2) level of institutionaUzation of aprogram within an organization and (3)measures of capacity building in the recipientcommunity. Second, planning for sustainabilityrequires the use of programmatic approachesand strategies that favor long-term programmaintenance. We suggest that the potentialinfluences on sustainability may derive fromthree major groups of factors: (1) project designand implementation factors, (2) factors withinthe organizational setting, and (3) factors in thebroader community environment Future effortsto develop sustainable health interventionprograms in communities can build on theconcepts and strategies proposed here.

Department of Health Policy and Management, School ofHygiene and Public Health, The Johns HopkinsUniversity, 624 North Broadway, Baltimore, MD 21205,USACorrespondence to Lee R. Bone

Throughout the world, considerable resources arespent implementing community-based healthprograms that are discontinued soon after initialfunding ends (e.g. US Agency for InternationalDevelopment, 1988; Steckler and Goodman, 1989;Bamberger and Cheema, 1990; Bossert, 1990). Inrecent years, program sustainability has been anissue of growing concern, both in the US andabroad. Attention to the long-term viability ofhealth intervention programs is likely to increaseeverywhere, as policy makers and funders becomeincreasingly concerned with allocating scarceresources effectively and efficiently.

Often designed as seed-funded or demonstrationprograms, the primary focus of many community-based programs has traditionally been ondetermining program efficacy, while the long-termviability of potentially successful programs hasbeen a 'latent' concern (Goodman and Steckler,1987/88), if considered at all. Systematic researchabout the long-term maintenance of programs isat an early stage and the very notion of what ismeant by sustainability yields different approaches.The first part of this article synthesizes a diverseliterature that reveals multiple perspectives on theconcept of sustainability. Using this synthesis asan organizing framework, we examine indicatorsthat can be used to monitor sustainability over time.

Increasing efforts are being directed towardaddressing the sustainability of major community-wide health promotion programs. For example, aninternational study by Bossert (1990) involved acomparative analysis of 44 projects in CentralAmerica and 13 projects in Africa funded by the

O Oxford University Press 87

at Physical Sciences Library on June 3, 2012

http://her.oxfordjournals.org/D

ownloaded from

http://her.oxfordjournals.org/

M. C. Shediac-Rizkallah and L. R. Bone

US Agency for International Development. Inthe US, Steckler and Goodman's work on theinsti nationalization of 10 health promotionprograms funded by the Virginia State HealthDepartment (Goodman and Steckler, 1989a,b;Steckler and Goodman, 1989) has served as a basisfor recent advances in frameworks andmethodologies for conceptualizing and measuringinstitutionalization. Also notable are recent reportsfrom the now 'classic' community-basedcardiovascular disease prevention programs inPawtucket, Rhode Island (Lefebvre, 1990),Stanford (Jackson et al., 1994) and Minnesota(Weisbrod et al., 1992; Bracht et al, 1994).Initiated with federal funding in the 1980s, thesemajor demonstration programs are just now, 10and 15 years later, generating reports about theirexperiences in attempting to maintain the programsin the communities beyond the end of the grantperiod. While the above works have been describedin the literature, little attempt has been madeto consolidate what we know about factors thatinfluence sustainability across different studies.The second part of the paper presents an initial setof potential guidelines and strategies for fosteringprogram sustainability within the dynamic contextof community. The article ends with a discussionof future directions for researchers, practitionersand policy makers regarding sustainability.

Why be concerned about programsustainability?

Clearly, there are circumstances under which thediscontinuation of a program is appropriate. Glaser(1981) rightly observes that:

Not all innovations should last or endure forlong periods of time. Circumstances, people,situations, and problems change. When avalidated, more efficacious, more suitable ormore cost-effective means for meeting a givenproblem comes to light, the former modusoperandi very appropriately may be supplanted.Or the problem the given innovation wasdesigned to address may have changed ordisappeared.

While not all programs should endure, there areat least three reasons why the failure to sustainprograms over the long term^may present seriousproblems. First, program termination iscounterproductive when the disease that a programwas established to address remains or recurs. Manyexamples may be provided in public health wherecontinuing disease control, for both chronic andinfectious disease, is simply essential. A study byHolland et al. (1993) clearly illustrates the resultof discontinuing cancer screening in a community.In this study, an examination of cervical cancertrends over a 19 year period in Newark, NewJersey showed that the ratio of in situ to invasivecancer cases increased (an indication of a shift toan earlier, more curable stage of disease) with theintroduction of educational and screening programsfor the prevention of cervical cancer, only to revertto pre-intervention levels when funding for theprevention programs ended. The New Jerseyexample highlights the critical importance of themaintenance of screening for the continuing controlof invasive cervical cancer.

Second, sustainability is a concern common tomany community health programs; having incurredsignificant start-up costs in human, fiscal andtechnical resources, many programs see their fundswithdrawn before activities have reached fullfruition. In one study, sustainability was identifiedas a concern in six out of nine foundation-fundedprograms in the western US, despite their differentsettings, target groups and health goals (Altaianet al., 1991). Specifically, program staff,community coalition members and otherrepresentatives from surveyed community healthpromotion projects identified deficient funding, andthe need for a diversified and reliable long-termfunding base as obstacles to achieving currentgoals and objectives. Further, securing resources toensure self-sufficiency and integrating the programinto the community to ensure that health promotionremains when funding ends were stated as twofuture goals in eight and six communities,respectively, out of the nine surveyed. Janz et al.(1996) found that adequacy in duration of fundingand the ability to locate additional funds were

88



ownloaded from


Sustainability of community-based programs

reported as factors that impeded interventioneffectiveness by staff of almost half of 37 AIDSprevention and service projects nationwide.Surprisingly, the frequency of the reporting ofthese concerns exceeded that of other concernsthat might be expected to present particularchallenges in AIDS prevention, such as the targetpopulation's reluctance to be involved with AIDS(only 27% reported this to be highly problematic)and reaching the target group (11%).

A third reason to be concerned with programsustainability is that new programs may encounterdiminished community support and trust incommunities with a history of programs that wereabruptly or inappropriately terminated (Goodmanand Steckler, 1987/88).

Sustainability: from a 'latent' goal to aplanned approach

Sustainability was a goal of many of the projectsthe authors have collaborated on over the years inan urban African-American community. This paperis the result of an ongoing process over severalyears to organize our cumulative insights andexperiences from field work, research and therelevant literature to provide the beginnings of acoherent and systematic knowledge base aboutprogram sustainability. This paper is also intendedto guide others who are also attempting to buildsustainable programs.

Experiences in community-based projectsfor breast and cervical cancer control

Involvement by one of the authors (M. C. S.-R.)in a community-based cancer control preventionprogram over a 5 year period provided extensiveexperience in planning, implementing andevaluating a program which recruited layneighborhood residents to educate inner-citywomen in Baltimore, Maryland about cervicalcancer and encourage them to seek appropriatescreening (Mamon et al., 1991). In the final yearof the program, the success of this program,combined with the impending termination of itsfunding from the National Cancer Institute, led to

various activities to continue the program beyondthe grant period. For example, several meetingsand focus groups were held with representativesfrom local and state health departments, therecipient community, voluntary associations andfunding agencies. Lay community volunteers werevery enthusiastic about continuing their healtheducation work with the women in theircommunities, even expanding it to include otherwomen's health concerns. In addition, the localAmerican Cancer Society (ACS) demonstrated agreat deal of interest in taking over supervision ofthe network of lay community volunteers that hadbeen developed during the project; this fit wellwith the ACS's desire at the time to increase theirtargeting efforts to lower income communities.However, despite best intentions, the ending phaseof the project meant that remaining staff, time andresources were too few to permit building adequatemechanisms for a successful transition of thiscomponent of the project to another organization.Our experience in this project was consistent withGoodman and Steckler's (1987/88) observationthat sustainability is often a 'latent' concern inmany health promotion programs, i.e. variousconstituencies may well wish the program tocontinue but, in the absence of early and activeplanning, the conditions that would most enhancethe prospects for sustainability in the long termare not created and sustainability does not occur.

Experiences from the East Baltimorepartnership program

The second case example presented in this paperis a program which began in 1974 with a clinicaltrial to test the effectiveness of educational andbehavioral interventions in improving bloodpressure care and control, and subsequentmorbidity and mortality, in an urban African-American patient population in Baltimore,Maryland. The success of the trial led to acommunity-based demonstration phase in the samecommunity. This second phase was intended totest the feasibility, acceptability and impact of anoutreach approach which linked trained communityhealth workers (CHWs) with provider institutions.

89



ownloaded from



Over the years, the program has changed to reflecta broader approach to health promotion and diseaseprevention, and enhanced partnership efforts withcommunity members and organizations, with aparticular focus on including the clergy. Beginningwith the clinical trial, the high blood pressurecontrol efforts in the East Baltimore communityhave been in place for over 20 years. The variousphases and components of the program have beendescribed in detail elsewhere (Bone et al., 1989;Levine et al., 1992a,b, 1994).

At the inception of the community-based phaseof the program, the approach was to mobilizeexisting community and provider resources in acoordinated effort to improve the health of thecommunity. The program recruited and trainedcommunity members as CHWs to provideeducation, outreach and follow-up, and wasdirected by a Community-Provider AdvisoryBoard, a steering committee of communityresidents, leaders and health care providers.

Initially, the program was fully supported by theNational Heart, Lung and Blood Institute (NHLBI)in both the clinical trial and the demonstrationphases. Later, support was secured through localresources and continuing support from grants. Thefinal phase in fiscal sustainability is to seek third-party reimbursement for CHW services.

Consistent with the commitment to having thecommunity directing these programs was thecommitment to provide skill building and jobopportunities for a community with a very highunemployment rate. We recognized early in theproject that if the CHW model was shown to beeffective, it would have the potential to beexpanded to include other health problems in thecommunity. Gradually, other cardiovascular riskfactors, including obesity, smoking, heavy alcoholconsumption and diabetes, were added to theCHWs' training and service provision. Continuoustraining with the introduction of added skillbuilding enhanced career development andpotential upward mobility is necessary.

In the East Baltimore project, early planning forsustainability focused on planning for the retentionof (and sources of remuneration for) CHWs. The

CHWs, initially volunteers, were later providedstipends and eventually a few workers wereemployed full-time by the program. While therehas been attrition, new community health workershave been trained each year and one workerremained with the program for 10 years. Initialplanning in the more prosperous climate of the1970s focused on ways of integrating CHWs withinexisting organizations in the community. Thisstrategy failed, partly due to the sluggish economicclimate at the time (1980s). A subsequent strategywas to integrate the program into the local healthdepartment. This strategy was also unsuccessfulbecause of the community's ambivalence towardthe health department. Eventually, the programbecame incorporated into the provider system byplacing CHWs in the local hospital emergencyroom. The program now continues with CHWsplaced in various prevention sites in the community.

The program was a partnership between thecommunity, health care providers and faculty fromthe Johns Hopkins Medical Institutions in EastBaltimore. This Advisory Board planned, directedand evaluated the program. The chairperson of theBoard was viewed as the program champion withinthe community, with support from the project staff.In her role as Assistant Director of a decentralizedmayor's office which provides health and humanservices at the local level, she was committed toimproving the health status of the community. Shewas also the primary source for recruitment ofCHWs and served as their leader and motivator.She advocated for the program and particularly forits sustainability at critical points in its history.Although her commitment to the community andthe program did not change, her influence andability to advocate for the program at the city-wide level shifted when a new mayor was elected.

The two projects described above providedmultiple lessons learned about sustainability.Unlike the cervical cancer project which was notsustained beyond its initial grant period, planningfor the sustainability of the East Baltimorepartnership program began early during the firstyear with the community demonstration phase.The motivation for this early planning, lacking

90



ownloaded from



knowledge of whether the intervention would beeffective, was concern over disrupting thecommunity • and damaging the academic center'srelationship with the community. An importantfactor contributing to the sustainability of the EastBaltimore project has been the continuity over a20 year period in principal program-relatedpositions and in some key community persons. Wealso learned that the choice of the host organizationfor the program has important implications for theway it is perceived by the community as well asfor the viability of program integration. A relatedpoint is that having multiple alternative strategiesfor sustainability is critical. Another lesson learnedwas that active community involvement indecision-making through the Community-ProviderAdvisory Board was an important factor insustaining a long-term commitment of thecommunity to the program. We also gained insightinto the value of sustainability as capacity buildingof the community through training and skillbuilding of community members.

In our projects, while we were committed to theconcept of sustainability, we were not researchingthe process or the factors influencing it; rather, ourprimary interest was continuity of our community-based fieldwork. In this paper, we are proposing aresearch-based perspective on sustainability. Informulating an empirical basis for sustainability,we have found the following definition of planningby Levey and Loomba (1973, p. 273) to bevery helpful:

Planning is the process of analyzing andunderstanding a system, formulating its goalsand objectives, assessing its capabilities,designing alternative courses of action or plansfor the purpose of achieving these goals andobjectives, evaluating the effectiveness of theseplans, choosing the preferred plan, initiatingnecessary actions for its implementation, andengaging in continuous surveillance of thesystem in order to arrive at an optimalrelationship between the plan and the system.

This definition suggests that movingsustainability from a 'latent' goal to a planned

approach will require formulating sustainabilitygoals and objectives and developing andimplementing strategies specifically to fostersustainability. A third critical component isassessment or evaluation where both objectivesand strategies are continuously monitored andrevised. This view of planning underlies theorganization of the paper. After discussingavailable definitions of sustainabUity, we reviewmeasurement issues for assessing progress towardsustainability. A subsequent section addressesstrategies to foster sustainability in appliedcommunity settings.

Sustainability defined

Little consensus exists in the literature on theconceptual and operational definitions ofsustainability. Several terms have been in use torefer to the phenomenon of program continuation.Among these are: program 'maintenance','sustainability', 'institutionalization', 'incorpora-tion', 'integration', 'routinization', local orcommunity 'ownership' and 'capacity building'.

In the literature, the following definitions havebeen offered as clarification of the variousconcepts:

Definition ISustainability is the capacity to maintain servicecoverage at a level that will provide continuingcontrol of a health problem (Claquin, 1989).Definition IIProject sustainability is defined by manyeconomists and international developmentagencies as the capacity of a project to continueto deliver its intended benefits over a longperiod of time (The World Bank's definition inBamberger and Cheema, 1990).Definition IIIA development program is sustainable when itis able to deliver an appropriate level of benefitsfor an extended period of time after majorfinancial, managerial and technical assistancefrom an external donor is terminated (USAgency for International Development, 1988).

91



ownloaded from



Definition IVThe term ' institutionalization' refers to the long-term viability and integration of a new programwithin an organization (Steckler andGoodman, 1989).Definition VOrganizational change ultimately involves theprocess by which new practices becomestandard business in a local agency. Whetherthe process is called routinization,institutionalization, incorporation, or someother term, it is central to all organizations...(Yin, 1979).Definition VI'...the optimal way to maintain heart diseaseprevention activity was to develop the healthpromotion capacity of community healtheducators. We defined health promotion capacityas the extent to which a community has localaccess to the knowledge, skills and resourcesneeded to conduct effective health promotionprograms.' (Jackson et al., 1994).

While these terms tend to be used inter-changeably, they are not synonymous. Thedefinitions of sustainability advanced by leadingdevelopment agencies (Definitions I—IH)emphasize health benefits as being at the heart ofthe sustainability process. In contrast, the emphasisin 'institutionalization' (Definitions IV and V) ison the persistence of the program itself rather thanon the benefits it delivers. 'Institutionalization'may cany the connotation of inflexibility andadoption of a program in toto. Finally, DefinitionVI characterizes sustainability as a processoccurring at the level of the community as a whole.

Interestingly, 'institutionalization' has been theterm more commonly used domestically to describeprogram continuation (e.g. Steckler and Goodman,1989; Lefebvre, 1990; Altman et al, 1991;Goodman et al., 1993a) while 'sustainability' ismore often used in the international context (USAgency for International Development, 1988;Bamberger and Cheema, 1990; Bossert, 1990;Lafond, 1995). This difference in usage may wellstem from perceived differences in institutional

strength in the two realms. The developed countriesare seen to possess well-developed andsophisticated health systems into which newprograms can be readily integrated, whereas,internationally, the third world is characterizedby weaker and more fragmented administrativestructures less capable of nesting new programs.Greater emphasis appears to have been placedby international donor agencies on infrastructuredevelopment in developing countries (Bossert,1990; Lafond, 1995) because third world countrygovernment capacity to fund even the most basichealth services is often too limited for short-termfinancial self-sufficiency to be a realistic goal.Other perceived differences include greaterpolitical stability and more affluence indeveloped nations.

Sustainability thus appears to be a multi-dimensional concept of the continuation processand the term encompasses a diversity of forms thatthis process may take. For example, an entireprogram may be continued under its original oran alternate organizational structure, parts of theprogram may be institutionalized as individualcomponents, or there may be a transfer of thewhole or parts to community ownership (Sheaet al., 1996). Moreover, continuation may occur atlevels other than the organizational level, includingthe individual and network levels. At the individuallevel, key community members assume a personalcommitment to continuing program messages,products or services. At the network level,individuals and organizations are brought togetherto create networks that reinforce program goalsand promote coordinated efforts (Lefebvre, 1990).

Sustainability is the global term we will usehereafter to refer to the general phenomenon ofprogram continuation. The choice of the term isdictated by two considerations. First, sustainabilityis a broad term that incorporates essential notionsin continuation (permanence, time) withoutlimiting its manifestations to any particular form.For example, unlike the terms 'institutionalization','incorporation' or 'integration', sustainability doesnot restrict continuation to survival within anorganizational structure. Second, sustainability

92



ownloaded from



does not imply a static program, in contrast to thenotions of institutionalization and routinizationwhich imply something that is repetitive but fixed.A dictionary definition of the word 'sustain' is 'tosupply with sustenance: nourish', suggesting aliving entity with the power to respond and change,just as a program must adjust to new needs andcircumstances if it is to continue. Indeed, Pressmanand Wildavsky (1979) have observed that changeis essential to program survival; they write that 'abasic reason programs survive is that they adaptthemselves to their environment over a long periodof time'. In sum, sustainability appears to bettercapture the dynamic process involved in programcontinuation and the broad range of its potentialforms than the notion of 'institutionalization'.

Conceptual frameworks andmeasurement issues

The three categories of definitions reviewed aboveprovide three sharply different perspectives onsustainability. These are: (1) maintaining healthbenefits achieved through the initial program, (2)continuation of the program activities within anorganizational structure and (3) building thecapacity of the recipient community. A diverseliterature from a range of different fields is neededto advance our thinking about sustainability fromeach of the three perspectives. In the section below,we discuss concepts and principles from suchdiverse fields as public health, organizationaltheory, and community-level change anddevelopment to show how they can contribute toadvance our thinking about sustainability fromeach of the three perspectives.

Concepts and principles from publichealth

hi the first perspective on sustainability('sustainability as the maintenance of healthbenefits over time'), concepts and approaches fromthe field of public health can give insight intomethods of tracking health-related behaviors andhealth problems to assure the continuing controlof disease. Both practitioners and researchers agree

that many programs are prematurely terminated,resulting in recidivism in negative health outcomes.

The complexity of preventive interventions forthe control of both infectious and chronic diseasehighlights the importance of sustaining slow anddifficult changes in health behaviors. There aretwo basic reasons why intervention programs mustattend to ways to sustain behavior, not just developstrategies for initial behavior change.

First, modifications in populations' health habitsare only slowly achieved through education andsocial change, hence the need for an environmentin which change is supported and reinforced(Resnicow and Botvin, 1993; Prasad and de LCostello, 1995). To cite but one example, whilethere is support for the short-term effectiveness ofschool-based substance use prevention programs,studies have reported decaying program effects inlong-term follow-up. Resnicow and Botvin (1993)caution not to think that these prevention programsdo not work, but rather that it is necessary to usestrategies which may improve the durability ofpositive effects, such as booster sessions,enhancing implementation fidelity, utilizingmultiple intervention strategies and attending toattrition issues in evaluation studies.

Second, educational messages and otherintervention activities need to remain in place fornew generations of individuals to be exposed tothem. For example, in the area of communicablediseases of childhood, Streefland discusses 'theneed to continuously vaccinate new cohorts ofchildren with a high degree of coverage and in anappropriate manner' (Streefland, 1995). In Natal/Kwazulu, South Africa, a national mass measlesimmunization campaign led to a drastic reductionin measles admissions to a major referral hospitalin the first 12 months following the campaign;however, within 2 years of the campaign, measlesadmissions had risen steadily to above pre-campaign levels, attesting to the consequencesof the failure to maintain adequate vaccinationcoverage levels (Abdool Karim et ai, 1993).

This illustrates the need for long-term strategiesfor disease control, not short-term solutions.Disease surveillance has been defined as the

93



ownloaded from



'continued watchfulness over the distribution andtrends of incidence through the systematiccollection, consolidation, and evaluation ofmorbidity, and mortality reports and other relevantdata' and the regular dissemination to 'all whoneed to know'. Diseases that are now rare due tosuccessful control measures may be perceived asno longer important, but this should be assessedin light of their potential to re-emerge (Halperinand Baker, 1992). The resurgence of tuberculosismay be the best recent example of the unfortunateconsequences of the failure to adhere to thisimportant public health principle. The return oftuberculosis has been attributed by some to sharpreductions in funding, leading to die breakdownof the infrastructure—public health preventivesystems of case finding and supervision of infectionand disease—needed to maintain effective long-term control (Joseph, 1993; Reichman, 1993).

In recent years, concepts and approaches in thearea of surveillance have been expanded to includemonitoring methods for chronic disease, not justinfectious diseases (Halperin and Baker, 1992).Community-wide behavioral change must besustained over a long period of time before anysignificant decrease in actual morbidity or mortalitycan occur and be measured. This is welldemonstrated in the North Karelia Project wherea delay of several years is reported between healthbehavior and risk factor changes and respectivechanges in coronary heart disease rates (Puskaet al., 1985).

Theories of organizational change andinnovation

In the second perspective (sustainability as'institutionalization' of a program or programcomponents within an organization), theories oforganizational change and innovation (Yin, 1979;Rogers, 1983; Goodman and Steckler, 1989a,b;Steckler and Goodman, 1989) offer a conceptualapproach for how new programs becomeincorporated into organizations and institutions.Most of these models view program sustainabilityas die final stage of program implementation in aprocess that occurs over time.

The organizational change perspective suggeststhat a process of mutual adjustment occurs such thatboth the innovation and the organization change toadjust to each other. The innovation eventuallyloses its separate identity and becomes part of theorganization's regular activities, a process that hasbeen referred to as 'routinizing' or 'routinization'.The conditions that lead an innovation to becomeroutinized are internal conditions specific to thelocal agency or organization (Yin, 1979; Rogers,1983; Goodman et al., 1993a).

Steckler and Goodman's study of 10 healthpromotion programs in Virginia provided the basisfor the development of a framework forconceptualizing program institutionalization(Goodman and Steckler, 1989a). Building on priorwork by Yin and Katz and Kahn, these researchersconceptualized the level of institutionalization ofa program as a function of two dimensions. Thefirst dimension (y-axis) represents die extensivenessof a program's integration into the subsystems ofits host organization, of which Katz and Kahnidentified five: production, maintenance, sup-portive, adaptive and managerial. Intensiveness,or the depth of program integration into eachsubsystem, forms the second dimension (x-axis) ofinstitutionalization and has three degrees: passages,routines and niche saturation. The first two degreesare based on Yin's (1979) prior work which viewedroutinization as a combination of 'passages' and'cycles'—the key events in the life history of anorganization. In Yin's view, a 'passage' (e.g. thetransition in an innovation's financial support fromtemporary to permanent funding) is a significantchange in organizational structure or proceduresthat generally occurs once, while a 'cycle' (e.g. anannual budget cycle) is an organizational eventthat occurs repeatedly in the life of an organization.Goodman and Steckler introduced two changes toYin's original formulation. First, they replaced theterm 'cycles' with 'routines', arguing that passages(the first degree of institutionalization intensity)need to be transformed into routines (the seconddegree of intensity) in order to become permanent,but while many routines are cyclical, others areeidier intermittent or continuous. Second, they

94



ownloaded from



added the notion of 'niche saturation' (the thirddegree of intensity), defined as an institutionalizedprogram's maximum feasible expansion withina host organization. These two dimensions ofinstitutionalization provide a measure of thisconcept: the more cells in die resulting matrix thatthe health promotion program occupies, the moreinstitutionalized it can be considered to be. Findingsfrom a recent study of the construct validity of thelevel of institutionalization (Loin) scale showedinitial support for a multi-factor modelcorresponding to die various cells in theinstitutionalization matrix (Goodman et al., 1993a).A study of the inter-rater reliability of the Loinscale has also been reported (Bryant, 1993).

Models of community-level change

In the third perspective (sustainability ascommunity 'capacity building'), models ofcommunity-level change and communitydevelopment can provide the appropriateconceptual perspective from which to view theprocess of building the problem-solving abilitiesof individuals and the larger community.

The last 15 or 20 years have witnessed aparadigm change in health promotion and diseaseprevention whereby the community has becomethe new 'center of gravity' for health promotion(Green and Kreuter, 1991). The 'new' healthpromotion movement (Robertson and Minkler,1994) has shifted die locus of health promotionfrom the individual or lifestyle to the community,based, at least in part, on a growing recognitionthat lasting, widespread behavioral change is bestbrought about by changes in norms of acceptablebehavior at die level of the community as awhole. The principle of participation, central tocommunity-based approaches to healdi, posits thatchange is more likely to occur when die people itaffects are involved in die change process. Rifkinet al. (1988) define community participation as 'asocial process whereby specific groups with sharedneeds living in a defined geographic area activelypursue identification of dieir needs, take decisionsand establish mechanisms to meet these needs'.

In addition to 'participation', related constructs

increasingly found in die literature include thenotions of community 'involvement','empowerment', 'ownership', 'competence' and'capacity' (Wallerstein and Bernstein, 1994a,b).While these constructs are loosely used, it appearsthat 'participation', 'involvement' and'empowerment' all refer to the actual process ofenabling individuals and communities, inpartnership with healdi professionals, to participatein defining their health problems and shapingsolutions to diose problems (Rifkin et al., 1988;Wallerstein, 1992; Minkler, 1994; Robertson andMinkler, 1994). Intervention programs frequentlypay only Up service to participation, hence theneed to assess the actual level (depth and scope)of participation (Rifkin et al., 1988).

What is die relationship of 'participation'('involvement', 'empowerment') to 'ownership'and 'capacity' ('competence')? The literaturesuggests that community 'participation' enhancescommunity 'ownership'; in turn, 'ownership' leadsto increased 'capacity' (or 'competence') andpromotes program maintenance (Bracht andKingsbury, 1990; Flynn, 1995). Robertson andMinkler (1994, p. 308) state that 'high-levelcommunity participation...increases capacity on dieindividual and community levels'. Similarly,Wallerstein (1992, p. 201) writes that 'communitycompetence has been proposed as an importantresearch outcome of social network and communityparticipation interventions' (emphasis added).Similarly, Flynn (1995) proposes a key mediatingrole of perceived ownership as an intermediatevariable between community participation andprogram effectiveness and maintenance.

Cottrel was die first to develop the concept of'community competence', characterizing acompetent community along eight dimensions.Capacity building has also been defined as 'dienurturing of and building upon die strengdis,resources and problem-solving abilities alreadypresent in individuals and communities' (Robertsonand Minkler, 1994, p. 303). More recendy, anexpert panel convened by the Centers for DiseaseControl and Prevention identified 10 dimensions

95



ownloaded from



of community capacity (Goodman et al., inpreparation).

In thinking about community competence orcapacity, community is the appropriate unit ofanalysis. However, just as with the related constructof empowerment, capacity may also be measuredas an individual and organizational phenomenonto the extent that capacity building also involvesthe individuals and organizations in the community.Measurement at all three levels acknowledges thatchanges at one level may be related to changes atother levels (Wallerstein, 1992; Israel et al., 1994).

In summary, the three perspectives onsustainability reviewed above suggest different setsof frameworks as approaches to conceptualizingsustainability. The implications of the differentconceptual approaches for the development ofmeasurable objectives for sustainability arediscussed next.

Measuring sustainabilityIn order to lay the groundwork for research in thearea of sustainability, there is a clear need todevelop operational definitions of the term.Sustainability is probably a matter of degree ratherthan an all-or-none phenomenon. 'Gold standards'for sustainability may not be appropriate as theremay be wide variability in what can or should besustained by project type, setting or resources.Nevertheless, indicators are needed for planningwhat is to be sustained, how or by whom, howmuch and by when. These indicators can serve assustainability objectives to be monitored duringand after the project period. We discuss below thedevelopment of objectives appropriate to each ofthe three perspectives on sustainability and presentsome examples from the literature. This is alsoshown in Table I.

Maintenance of health benefits from programs

Two examples may serve to illustrate the usefulnessof measurable objectives for tracking andmaintaining the health benefits achieved through aprogram. First, the unique achievement of theworld-wide eradication of smallpox presents ahistorical example of how program activities were

adapted to changing needs and objectives. Overtime, as cases of disease were diminishing, theprogram's emphasis (the what in sustainability)shifted from routine vaccination to investigationof reports of suspected residual smallpox cases.This required a continued post-eradicationsmallpox surveillance system (the how) to monitorpossible recurrence. This surveillance systemremained in place 10 years after the last case ofsmallpox was reported (the when) (Henderson,1987; Jezek et al., 1987).

We can also refer back to the tuberculosisexample discussed earlier. It has been suggestedthat (the differences between the two diseasesnotwithstanding), as was required with theeradication of smallpox, interest and commitmentto the control and eventual elimination oftuberculosis should not have been allowed todecline, even when cases of disease diminishedover time. Public health activities, such as casefinding and contact tracing, should have beencontinued and even intensified, if the disease wasto be kept under control or eliminated (Reichman,1993). In order to face the resurgence oftuberculosis, a critical task facing public health nowis the revitalization of tuberculosis control systems.

National periodic surveys of the distribution andtrends of health-related behaviors and diseasessuch as the National Health Interview Survey(conducted annually by the National Center forHealth Statistics) or the Behavioral Risk FactorSurveillance System (conducted by the Centers forDisease Control and Prevention) provide additionalmethods to track health-related behaviors anddiseases of the American population over time.Surveillance findings can help determine whetherpublic health interventions are meeting theirobjectives, both in the short and longterm (Halperinand Baker, 1992).

Institutionalization objectives

The development of institutionalization objectivescorresponds to the second view of sustainabilityas organizational integration. The level ofinstitutionalization scale (Goodman and Steckler,1989; Goodman et al., 1993a) provides a beginning

96



ownloaded from



Tfcble L Conceptual approaches to sustainability planning, corresponding operational measures and examples of planning objectives

Concept Operational measure Example of a planning objective

Maintenance of healthbenefits

Institutiooalization

Capacity building

Continued monitoring for control of thehealth problem

Integration of the program within anorganization

Local community's access to knowledge,skills and resources

Maintenance of a public health system tomonitor possible recurrence of disease

Incorporating intervention programs incommunity organizations after thewithdrawal of funding

Training community members to serve asa source of information and expertisefor health promotion for thecommunity

basis for developing quantitative measures ofinstitutionalization which can be repeated atdifferent time periods to monitor changes in thedegree of institutionalization of a program withinan organization.

Other examples from the literature include theMinnesota Heart Health Program's tracking of the'incorporation' status of intervention programs instudy communities through annual surveys. Usingthis measure of incorporation, survey findingsshowed initial rates of 67% (18 out of 27 programs),61% (16 out of 26 programs) and 84% (21 out of 25programs) in each of the three study communities in1989, the year of federal funding withdrawal. Threeyears later, the incorporation rate had droppedto 56, 58 and 68%, in each of the respectivecommunities (Bracht et al., 1994).

Capacity-building objectives

Developing capacity-building objectives corres-ponds to the third view of sustainability ascommunity capacity building. For example, in theStanford Five-City Project (FCP), the capacity-building partnership between FCP staff and theMonterey County Health Department (MCHD)included three long-term objectives: (1) self-sufficiency of the MCHD in the use of acquiredresources and knowledge to design, implement andevaluate heart disease prevention programs in thecommunity, (2) the MCHD would serve as a sourceof health promotion information and expertise forhealth educators in the community, and (3) MCHDand FCP staff would collaborate on new projects

where collaboration would be viewed as mutuallybeneficial (Jackson et al., 1994).

Several studies have described attempts atdeveloping instruments to track changes incommunity capacity (and intervening variablessuch as ownership) over time as a way ofmonitoring progress in achieving capacity-buildingobjectives. For example, Flynn (1995) tested theCommunity Ownership Scale in three programsutilizing a similar university-communitypartnership model. In this study, community leaderswere asked to rate the amount of control theyperceived for themselves, the university and thelocal program staff, on 14 key program functions.Based on initial evidence of the scale's validity,the author suggests its application at differentstages in the life of a program to monitor the degreeof perceived community ownership over time.

Goeppinger and Baglioni (1985) developed aquestionnaire for use by trained community healthworkers to assess community competence.Telephone interviews of 433 community residentsin five rural Virginia communities demonstratedthat none of the communities seemed to surpassany of the others in overall competence, rathereach community demonstrated its specific areas ofstrength and weakness. Fourteen of the 22 surveyitems discriminated between the five communities;these 14 items also successfully captured six ofCottrell's eight dimensions of communitycompetence.

Eng and Parker (1994) also measured yearlychanges in community competence after

97



ownloaded from



implementation of a health promotionempowerment program in three communities.Community representatives generated 23 traits ofa community that 'could get it together'. Evaluatorsand program staff then constructed a questionnaireto measure community competence (itemsmeasured CottreLTs original dimensions and a newdimension—social support—which was con-tributed by the community members). Changes incommunity competence over time were monitoredby interviewing 15 key informants in the threeproject communities at baseline and once a yearafter implementation. Results showed that levelsof competence became more similar acrosscommunities after year 1 of the intervention; in allthree communities, the pattern of competenciesalso changed, e.g. from internal social interactionsto a more external focus on mediating with outsideinstitutions and officials.

Influences on program sustainability

The first part of this paper discussed conceptualdistinctions in the various constructs used inreferring to the general process of sustainability.Planning for sustainability may be very differentdepending on what one envisions to sustain, withdifferent implications for measuring andmonitoring sustainability. This second part of thepaper reviews what is known about how to facilitateprocesses that will lead to sustainability.

Understanding the conditions under whichprograms are most likely to continue is requiredto move from a 'latent' or passive approach tosustainability towards active attempts to modifyconditions to maximize the potential for long-termsustainability. A framework for conceptualizingprogram sustainability is presented in Figure 1,listing three major groups of factors as potentialinfluences on sustainability, derived from ourreview of the available sustainability literature:

(1) Project design and implementation factors.(2) Factors within the organizational setting.(3) Factors in the broader community environ-

ment

The factors grouped under these three mainheadings were not derived using a quantitativesummary of research findings (e.g. meta-analysis)nor are they an exhaustive list of all factors thatmay be found to influence sustainability. Rather,these factors were distilled through a review of theliterature using Medline searches of publications inthe last 15 years, manual searches of major healtheducation and health promotion journals, andsearches of references cited in published papers.These factors are presented here as a starting pointfor summarizing similar findings across multiplestudies regarding potential influences onsustainability. These influencing factors, presentedin Table II as potential guidelines for sustainabilityplanning, are reviewed below. Table II can also beused as a checklist for researchers, practitionersand policy makers interested in maximizing thepotential sustainability of their programs.

Project design and implementation factorsThe first group of factors relates broadly to theresources available to the project, including staff,financial resources and time for project activitiesto reach fruition, and the implementation activitiesdetermining the use of these resources.

Project negotiation process

Projects imposed by a funding agency may be lesslikely to be sustained than those which are theresult of a 'mutually respectful negotiating process'between funders and host governments (Bossert,1990). A similar perspective is offered by Bermejoand Bekui (1993) who write that, compared tothe 'participatory' approach in disease controlprograms where goals, targets and time framesemerge from the interaction of local people andservice providers, and are adapted as the projectevolves, the 'project' approach with pre-specifiedobjectives and time frames is much less conduciveto community participation.

In the US, the funding of intervention programsis often driven by priorities for categorical healthproblems. To legitimize health programs andincrease their acceptance among communityleaders and groups, services should be provided

98



ownloaded from



/

community environment

Project desiftnan

implementa

4

i

\

\

ition factors

Factors within theorf«nizational setting

\Program sustainability

1. Maintenance of healthbenefits from a program

2. Institutionalization ofa program within anorganization

3. Capacity building in therecipient community

Fig. 1. A framework for conceptualizing program sustainability.

Tfcble IL Guidelines for sustainability planning

Project design and implementation factors1. Project negotiation process. Are project approaches and goals discussed with recipient community members, as equal

partners? Are the needs of the community driving the program or those of external donor agencies and technical experts? Is anegotiation or consensus-building process in place to reach a compromise for addressing everyone's (including donors,community, technical experts) needs?

2. Project effectiveness. Is the project (perceived as) effective? Is it visible? What are the (desirable and undesirable) secondaryeffects of the program?

3. Project duration. What is the project's grant period (number of years in operation)? Is it a new project or is it an existingprogram that is acquiring additional funds?

4. Project financing. What are the sources of funds for the program (internal, external, a mixture)? What are the community'slocal resources? Can the community afford the program (e.g. is it able to pay maintenance and recurrent costs)? How muchare community members willing/able to pay for services? What strategies are in place to facilitate gradual financial self-sufficiency?

5. Project type. What type of project is it (e.g. preventive versus curative)?6. Training. Does the project have a training component (professional or para-professional)?

Factors within the organizational setting7. Institutional strength. What organization will be implementing the program? How mature (developed, stable, resourceful) is

this organization? Is it likely to provide a strong organizational base for the program?8. Integration with existing programs/services. Is the program vertical (categorical) or is it a horizontal (comprehensive or

integrated) program? Are goals, objectives and approaches pre-specified or are they adapted to the local population andsetting and over time? Is the program integrated into the standard operating practices of its host organization? Is the missionof the program compatible with the mission and activities of its host organization? Is the implementing organization therecipient of program funds or is there an intermediary organization?

9. Program champion/leadership. Is there a program champion? What are his/her attributes? If not, can one be identified/nurtured so that he/she may serve as an advocate for the continuation of the program? Is the program endorsed from the top?How well is it supported?

Factors in the broader community environment10. Socweconomic and political considerations. How favorable is the general socioeconomic and political environment for the

sustainability of the program to be a realistic goal?11. Community participation. What is the level of community participation? What is the depth (amount) of involvement? What is

the range of involvement (types of activities)?

Planning for sustainability along these various guidelines must begin early in the program and assumes a minimal level ofpolitical and economic stability.

99



ownloaded from



that are desired by the community (Shea et al.,1992). This requires a negotiation or consensus-building process about how to best serve thecommunity.

Project effectiveness

It has been noted that although successful programimplementation does not guarantee sustainability(Goodman and Steckler, 1989b), a program is onlyworth institutionalizing if it has been shown to beeffective (Steckler and Goodman, 1989). In reality,community-based programs are difficult (andexpensive) to evaluate and programs may besustained with no real evidence of impact orregardless of the direction of the evidence. Forexample, in Bossert's study (1990, p. 1019), 'it wasthe reputation for effectiveness and not objectiveevidence that was important for sustainability'.Others have similarly maintained that highvisibility in the community, through thedissemination of information on project activitiesand early evaluation results, is essential to programcontinuation (Bracht and Kingsbury, 1990).

Project duration

In general, the short-term horizon of governmentsand funding agencies, due to a crisis mode ofoperation, short budget cycles and internal politicalpressures, has negatively affected the process ofsustainability (Bamberger and Cheema, 1990). Ina cross-case analysis of the development of healthsystems in several low-income countries, Lafond(1995) finds that two features of the traditional aidsystem exert a detrimental effect on sustainability:its inward focus and its short-term investmentcycles. Donor agencies are accountable toinstitutions which demand swift and visibleevidence of their investments; these requirementsconflict with the long-term needs of the recipientcommunities.

Available research shows that short grant periodsfor establishing new programs impedeinstitutionalization. Steckler and Goodman (1989)found that a grant period of 3 years was tooshort to achieve institutionalization of new healthpromotion programs. They suggest that funding

agencies may want to consider supportingworthy programs up to 5 years to enhanceinstitutionalization prospects. Similarly, Scheirer's(1990) study of the Fluoride Mouth Rinse Programin schools shows that number of years in operationwas strongly related to the likelihood that theprogram continued: on average, continuing siteshad used the program for 6.6 years compared toonly 3.7 years for discontinued sites, a statisticallysignificant difference.

Project financing

Financing is probably the most prominent factorin sustainability. In international aid programs, thefinancial sustainability of a health project beyondexternal donor support is typically dependent onone of two sources of national funding: hostcountry government support or beneficiary supportthrough cost-recovery mechanisms (Bossert, 1990).In the Bossert study, the availability of nationalfunds after the end of external funding was relatedto efforts at gaining alternative sources of financialsupport during the life of the project and gradualindependence from external support (e.g. throughprogressive absorption of recurrent costs into thegovernmental budget).

There has been increased reliance on communityfinancing as a funding source for health programsin the last two decades, as a result of declininggovernment resources and the global recession ofthe 1980s (Abel-Smith and Dua, 1988; Gertler andvan der Gaag, 1990). Haws et al. (1992) documentthe experience of 20 family planning clinics inMexico, the Dominican Republic and Brazil inimplementing strategies for financial sustainabilityfollowing the withdrawal of external funds: themost common strategy was an increase in clientfees. In most clinics, the effect of introducing userfees was a reduction in the volume of clients anda change in client profile with a fewer number ofthe poorer clients served. However, a few clinicswhich sought various additional cost-recoverystrategies—including donations from localcommunity groups, use of sliding-scale fees forservices, diversification of services and contractswith public- and private-sector companies—did

100



ownloaded from



not experience a change in size or type of caseload.Based on the study findings, the authors made thefollowing recommendations to assist programs inachieving sustainability: (1) the need for carefulplanning by donors and grantees for eventualcutbacks in funding, (2) the ability to identify costsand set realistic fees, (3) adopting anentrepreneurial spirit in seeking alternative sourcesof funding, and (4) diversification of services.

An aspect of financial sustainability that is oftenoverlooked is that there is a supply side and ademand side. The availability of external resourcesillustrates the familiar supply side of sustainability.Focusing on the demand side of sustainabilityshifts the attention away from the donor to therecipient's behavior. Beneficiaries' willingness andability to pay for services is a central issue for thedemand side of sustainability (Jensen, 1991). Thedemand side also underscores the need for servicesto achieve a high level of quality: good qualityleads to increased demand; in turn, demand forservices attracts monetary resources (Ashford andHaws, 1992). Paradoxically, excessive outsidefunds (the supply side) can inhibit sustainability,as often happens when a program requires recurrentfunds for continuation that exceed local resources(the demand side) (Mburu and Boerma, 1989).

Project type

Lesser resources tend to be allocated to preventivethan to curative care, perhaps because the Americanhealth care system has generally been lessprevention oriented (Abel-Smith and Dua, 1988).Under this modus operandi, it follows thatpreventive programs may be more difficult tosustain than curative programs. Bossert's (1990)research supports this hypothesis; in his work,health service projects (including health workertraining, clinic construction and other infrastructuredevelopment) were most likely to be sustained withnational funds, while preventive health projects(family planning, malaria control and nutritionplanning) were least likely to be sustained.However, significant variation was found withinproject type.

Training

Projects with training (professional and para-professional) components are more likely to besustained than those without: those trained cancontinue to provide benefits, train others and forma constituency in support of the program(Bossert, 1990).

The experience of the Stanford FCP providesfurther support for the inclusion of training as asustainability-enhancing strategy. A keycomponent of the Stanford FCP's capacity-buildingapproach to intervention maintenance involvedtraining a cadre of local health educators tocontinue the work in heart disease education, begunby the university-community partnership program(Jackson et al., 1994). A 'training of trainers'model was employed where a core group of 16health educators from the county health departmentreceived intensive training enabling them tosubsequently transmit knowledge and skills toother health educators in the community, therebybenefiting lay members of the community at large.

Factors within the organizational settingFactors considered in this category includeorganizational and managerial structures andprocesses. These encompass factors related to theorganizational location of the program, itsadministrative structure as well as politicalprocesses within the organization that inhibit orsupport program continuation.

Institutional strength

The 'strength' of the institution or organizationthat is implementing the program was identified inthe Bossert (1990, p. 1018) study as a variablepositively related to program sustainability; in thisstudy, institutional strength referred to 'institutionswhich were well integrated, had goal structuresthat were consistent with the project goals, and hadstrong leadership and relatively high skill levels'.

In the US, Steckler and Goodman (1989)similarly found that implementing organizationswhich were stable and mature were more likely topromote program institutionalization by providinga strong organizational base for new programs. In

101



ownloaded from



this study, a health promotion program which wasimplemented in a school district fared better interms of achieving institutionalization than anotherprogram which was implemented in a free-standingcommunity wellness center. The researchers' casestudy analysis suggests that, compared to the schooldistrict, with its multiple divisions offering theprogram alternative supportive structures intowhich it could be integrated, the community centerwas less well developed organizationally and lessresourceful.

Integration with existing programs/services'Vertical' (i.e. stand alone or self-contained)programs are less likely to be sustained thanprograms that are well integrated with existingsystems (Bossert, 1990) or into the standardoperating practices of their host organizations(Glaser, 1981). 'Vertical' programs are privilegedbecause they can focus resources and activitieson well-defined goals and with little pressure tocompromise; but they also tend to createinstitutional jealousy and are less likely to attractnational sources of funding, making themvulnerable to demise when external funding ends(Bossert, 1990). Similarly, Lafond (1995) foundthat vertical programs 'limit the potential spin-offsof investment by insulating external interventionsfrom the health bureaucracy. Parallel managementstructures created for donor-funded interventionsaimed to ensure strict control of funds. However,they can also fragment the health system'. Thus avertical approach seems to help initialimplementation but not long-term sustainability.

The likelihood of integration may be influencedby the fit or compatibility of the program withorganizational mission and activities. In Stecklerand Goodman's (1989) study, a factor favoringintegration was direct funding to the organizationimplementing the program rather than to anintermediary organization. Funding implementingorganizations directly presumably fosters thenecessary local adaptations in the organization thatultimately would institutionalize the program.

A study of a vaccination program in India

provides further support for the positiverelationship between program continuity and itsincorporation in the regular health system. Inaddition, a second key aspect of integration inthis study was adaptation of the program to thesociocultural environment of the host community(Streefland, 1989). Adjustment of formal programrules to local conditions contributes to continuityas the program and the environment adapt to eachother. This type of adaptation is facilitated wheninformation is transferred across the differentorganizational levels. For example, setting targetlevels for vaccination coverage should be done notby upper-level administrators acting alone but asa joint task with input from field workers at lowerlevels (Streefland, 1995).

Program champions/leadership

It has been argued that while programimplementation is primarily focused on immediateprogrammatic concerns, the process of programinstitutionalization is politically oriented and islargely one of generating goodwill for thecontinuation of a program (Goodman and Steckler,1987/88). Such goodwill seems to be mosteffectively garnered by influential individualswithin the implementing organization acting asprogram advocates or 'champions'. 'Linkingagents', external to the organization, may seek outor substitute for the role of the internal championin diffusing a program to multiple adoptingorganizations (Orlandi, 1986; Monahan andScheirer, 1988). Attributes of program championsinclude: mid- to upper-level managerial positionwithin the organization; a sense for thecompromises necessary to build support for theprogram; and negotiating skills. Programs with achampion possessing all three attributes exhibitedthe highest levels of institutionalization in a studyof 10 Virginia programs (Steckler and Goodman,1989). Results of a study of discontinuationprocesses surrounding the Fluoride Mouth RinseProgram in public school districts also points tothe key role of an internal champion in bothinitially adopting the program and preventing itfrom being discontinued later (Scheirer, 1990).

102



ownloaded from



Glaser (1981) also found that endorsement andsupport of the program from the top of the hostorganization is an important factor in its survival.

Factors in the broader communityenvironmentA program does not operate in a vacuum. Therelationship of the program with the larger'environment' (political, economic, social) mustbe considered. The political and economicenvironment, and the depth and range ofinvolvement of target community members willinfluence program impact and endurance.

Socioeconomic and political considerations

Bossert's comparative analysis of health projectsdemonstrates that projects in Africa weresignificantly less likely to be sustained than thosein Central America; this was attributed to agenerally less favorable environment forsustainability in Africa, due to greater economicdeterioration and weaker governmental institutionscompared to Central America. Cross-countryanalyses by both Bossert (1990) and Lafond (1995)suggest that minimal levels of economic resourcesmay be necessary for sustainability to be attainable.

In the US, Shea et al. (1992, 1996) identifyseveral challenges to the dissemination of thecommunity model in disadvantaged communities.They identify competing problems (of poverty,unemployment, crime, housing and homelessness,overcrowded schools, and drug abuse) as one ofthe potential barriers to adoption of a programmodeled after earlier cardiovascular diseaseprevention programs in North Karelia, Stanford,Minnesota and Pawtucket, which were allimplemented in more affluent and stablecommunities. Moreover, they question theassumption that the community can continue theprogram on its own beyond an initial fundingperiod, given that it has few local resources todraw upon. They conclude that programs in poor,disadvantaged communities may require anongoing commitment of resources from externalagencies in order to be viable in the long run.

Others have argued, however, that even the

poorest of communities have resources andfocusing on communities' strengths rather thandeficits is a more fruitful avenue. For example,McKnight (1987) advocates a vision of'community as the basic context for enablingpeople to contribute their gifts' and 'communityassociations as contexts to create and locate jobs,provide opportunities for recreation and multiplefriendships, and to become the political defender'of the rights of people. McKnight suggests thatsocial policy that excludes community and itsunique capacities will ultimately fail.

Community participation

As previously noted, the involvement andparticipation of the beneficiary community indesigning and implementing health programs isreceiving increasing attention in the literature.The literature overwhelmingly shows a positiverelationship between community participation andsustainability in both domestic (e.g. Bracht andKingsbury, 1990; Flynn, 1995) and internationalsettings (e.g. Rifkin, 1986) although a few studies,notably Bossert's (1990), have indicated norelationship except with respect to communityfinancing. This may be related to the differentmeanings attached to community participation: theconcept lacks a widely accepted definition becauseof the multiple meanings of each of the terms'community' and 'participation' (Rifkin et al.,1988) and the varying approaches and levels ofinvolvement of community members in healthprograms (Rifkin, 1986).

There appear to be many avenues by whichthe process of community involvement enhancesprogram sustainability. First, a major premise of acommunity approach to health behavior change isthat lasting widespread change is more likely tooccur if a broad range of health professionals,health institutions, community groups and privatecitizens are involved in a 'collective attack onhealth risk behaviors and the conditions thatproduce and support them' (Kinne et al., 1989).Other avenues by which community participationinfluences program sustainability is through theintermediate process of promoting a sense of

103



ownloaded from



ownership of the program (Bracht and Kingsbury,1990; Flynn, 1995). Also, community participationenhances overall community competence andcapacity (Wallerstein, 1992; Robertson andMinkler, 1994).

Future directions for research,practice and policy

Informed planning for program sustainability willrequire more research into the sustainabilityprocess, more training to teach programs toincorporate strategies specifically directed atfostering sustainability and changes in policymaking.

Directions for research

As in all scientific endeavors, the process ofbuilding upon the available knowledge base forsustainability will involve a cumulative process ofresearch across different locations, populations andhealth problems. Different sustainability objectivesmay be appropriate in different contexts, but futureresearch must clearly identify the approach usedto address sustainability in that particular contextand specify appropriate operational measures. Forexample, in the past, the specification of anacceptable time frame for assessing projectsustainability has been limited to indefinitesuggestions such as 'over a long period of time'(see Definition II above) or 'for an extendedperiod of time' (see Definition III above), usuallytranslating, in practice, to 3 and 5 years post-projectperiod as cut-off points. While the appropriate timeframe for sustainability may be dependent onproject type and how soon the benefits from theproject activities can be expected (Bamberger andCheema, 1990), the critical point is that the timeframe issue should not be arbitrary and should bemonitored.

Few appear to disagree about the value ofsustainability as a general goal. However, there isless consensus about what is to be sustained. Forexample, Green (1989) has questioned whetherlong-term program continuation, or 'institutiona-lization', is the proper goal of grant-funded

programs, arguing that long-term program effectsmay best be seen 'as investments in people ratherthan investments in programs.' In Green's view,grants should seek to develop problem-solvingskills and community leadership and confidencerather than to seek institutionalization of programsthat may become sterile bureaucracies. Green'sperspective on sustainability has two importantimplications for what should be sustained. Thefirst implication is that the merit of sustainingprograms should probably be gauged, first andforemost, against the intended benefits of aprogram, not its activities. Bossert (1990) alsostresses the need to evaluate (and not assume) thatcontinued activities actually produce continuedbenefits (e.g. when latrines are built, are theyused?). Thus an evaluation of the health benefitsachieved through a program, not just the programactivities, is essential to an overall assessment ofsustainability.

Secondly, the notion of investing in peopleimplies that capacity building is a legitimateprogram benefit. Others (Rifkin, 1986) havesimilarly maintained that since the community-based approach to health is a communitydevelopment model rather than a purely medicalor health services model, expected program benefitsinclude improvements both in health status andcommunity capacity.

Further research comparing the process ofsustaining programs in different contexts andsettings, e.g. public versus private organizations,and profit versus not-for-profit organizations, isneeded to advance the available empiricalknowledge base. We would expect that theframework to consider in planning forsustainability would be similar in the differentcontexts, but different priority may be given todifferent factors in diverse settings. For example,cost considerations may be most prominent in for-profit agencies and less important in not for-profitorganizations.

Some instruments for measuring programinstitutionalization and capacity building alreadyexist. For example, we have discussed the Loinscale (Goodman et al., 1993a), the community

104



ownloaded from



ownership scale developed by Flynn (1995), andcommunity competence questionnaires developedby Goeppinger and Baglioni (1985) and Eng andParker (1994). These instruments need furthertesting and validation before they can be broadlyapplied in different settings. Moreover, instrumentsfor measuring sustainability should tap its differentlevels, including the personal, organizational andcommunity levels (Lefebvre, 1990; Wallerstein,1992; Israel et al., 1994).

Directions for practiceAssistance to community groups in maintainingprograms should focus on the three groups offactors proposed as influences on programsustainability: project design and implementationfactors, factors within the organizational setting,and factors in the broader community environment.At the present time, some of these guidelines arewell accepted, while others have received lessempirical support.

Some of these factors are more amenable tocontrol by program staff than others. For example,strategies that are within the control of programstaff include the extent of community involvement;training community members to promote programmaintenance after the withdrawal of externalassistance; the choice of an organizational basefor the program or program components; and'cultivating' and nurturing program championswho can advocate for program continuation.

The community model of health promotion anddisease prevention often requires collaborativeapproaches leading to interorganizationalcollaboration and coordination, such as communitycoalitions (Goodman et al., 1993b), university-community partnership programs (Levine et al.,1992a,b, 1994; Bracht et al., 1994; Jackson et al.,1994), and strategic alliances among health serviceproviders (Kaluzny, 1991). Adequate start-up timeis needed for a new program to develop linkageswith existing programs and organizations in thecommunity that will enhance sustainability; policymakers and hinders must adjust to likely delays indemonstrating program impact.

While process and intermediate measures of

program effectiveness may not be predictive ofeventual impact on health status, they can be usedto disseminate project activities and enhance thevisibility of the program, and commitment of thestaff and leadership. These early results may bethe stimulus for earlier sustainability planning.

Directions for policyFactors, such as financial resources, programduration and the process of project negotiation,are ultimately a matter of policy. Under currentpractices, program funding is often driven byfunding agencies' own time frames, budget cyclesand internal political pressures (Lafond, 1995).The process of sustainability is unlikely to besignificantly facilitated until hinders and policymakers alter their funding practices. The followingare some suggested criteria which hinders shouldconsider to enhance sustainability prospects.

• Community health programs must be driven bythe needs of communities, not those of externaldonor agencies or technical experts. There isbeginning awareness of the need for change inthis area (Lafond, 1995). In the US, university-community partnership models for community-based public health must not allow academics'own needs for professional advancement tosupersede or interfere with community-chosengoals and activities (Breslow and Tai-Seale,1996).

• Sound planning for sustainability dictates thatprograms be designed with local capability inmind. A program is more likely to be sustainedwhen its host community can afford it.Considerations of affordability must include, notonly financial aspects, but also other costs suchas time (Yacoob and Walker, 1991) and technicalresources.

• While excess resources may not be justifiable,too little is not satisfactory either. For example,Schwartz et al. (1993) find that community-based cardiovascular disease programs havesuffered from inadequate financial support tostate health agencies. They suggest thatinadequate resources lead to a 'poverty cycle'where a poor resource base leads to a poorly

105



ownloaded from



designed intervention; in turn, the interventionyields only a modest impact, a situation whichonly stifles the ability to obtain appropriatefunding in the future. Unless enough resourcesare allocated to yield initial success, long-termsustainability is unlikely.

• Allocating resources to cover the maintenanceand recurrent costs of existing programs orservices with a proven track record rather thanmaking investment decisions that are biasedtoward spending on new programs (Steckler andGoodman, 1989; Lafond, 1995).

Concluding comments

This paper has presented an organizing frameworkfor conceptualizing and measuring sustainabilityand tentative guidelines to facilitate sustainabilityin community programs. Throughout the paper, wehave emphasized that sustainability is a dynamicprocess, and that goals and strategies for achievingit must continuously adapt to changingenvironmental conditions. Future efforts to developsustainable health intervention programs incommunities can build on the concepts andstrategies proposed here.

The research reported throughout this papersuggests that many of the same factors are foundto influence program sustainability in the US andin developing countries. There are clearly sharedproblems and common lessons to be learned. Forexample, major system flaws exist in some areasof the developed world, as demonstrated by thetypes of issues that surfaced during recent attemptsat health care reform in the US. There are alsopockets of social and political instability indeveloped nations that reflect underlyingsocioeconomic conditions similar to those seenin third world poverty areas. Finally, adequateresources for health care is a critical issue in allprograms. These parallels should not be ignored,lest important lessons go unheeded.

A final point relates to placing the discussionon program sustainability in its broader contextThe examination of the sustainability of a. particularhealth program—which was a major focus of this

article—is merely the narrowest perspective onsustainability. A broader perspective onsustainability arises when one considers the totalhealth and well-being of communities, not just thebenefits associated with improvements on a singlehealth problem or program viewed in isolation(King, 1990a,b). The third and broadest perspectiveon sustainability comes from viewing the healthof communities in the context of ecologicalsustainability (the health of the planet). Sustainabledevelopment is an increasingly debated topicamong environmentalists and other developmentspecialists. Defined as 'development that meets theneeds of the present without compromising theability of future generations to meet their ownneeds', it is seen as requiring the use of aninterdisciplinary approach (Ruttan, 1994).

Acknowledgements

The authors are grateful to Drs Charlyn Cassady,David M. Levine and Mary Ann Scheirer fortheir very helpful comments on earlier drafts ofthis paper.

References

Abdool Karim, S. S., Abdool Karim, Q., Dilraj, A. andChamane, M. (1993) Unsustainability of a measlesimmunisation campaign—rise in measles incidence within 2years of the campaign. South African Medical Journal, 83,322-323.

Abel-Smith, B. and Dua, A. (1988) Community-financing indeveloping countries: the potential for the health sector.Health Policy and Planning, 3, 95-108.

Altaian, D. G., Endres, J., Linzer, J., Long, K., Howard-Pimey,B. and Rogers, T. (1991) Obstacles to and future goals often comprehensive community health promotion projects.Journal of Community Health, 16, 299-314.

Ashford, L. S. and Haws, J. M. (1992) Family planning programsustainability: threat or opportunity? Studies in FamilyPlanning, 23, 63-65.

Bamberger, M. and Cheema, S. (1990) Case Studies of ProjectSustainability: Implications for Policy and Operations fromAsian Experience. The World Bank, Washington, DC.

Bermejo, A. and Bekui, A. (1993) Community participation indisease control. Social Science and Medicine, 36, 1145—1150.

Bone, L. R., Mamon, J., Levine, D. M., Walrath, J., Nanda, J.,Gurley, H. T., Noji, E. K. and Ward, E. (1989) Emergencydepartment detection and follow-up of high blood pressure:use and effectiveness of community health workers. AmericanJournal of Emergency Medicine, 7, 16-20.

106



ownloaded from



Bossert, T. J. (1990) Can they get along without us?Sustainability of donor-supported health projects in CentralAmerica and Africa. Social Science and Medicine, 30,1015-1023.

Bracht, N. and Kingsbury, L. (1990) Community organizationprinciples in health promotion: a five-stage model. In Bracht,N. (ed.), Health Promotion at the Community Level. Sage,Newbury Park, CA.

Bracht, N., Fmnegan, J. R., Rissel, C, Weisbrod, R., Gleason,J., Corbett, J. and Veblen-Mortenson S. (1994) Communityownership and program continuation following a healthdemonstration project Health Education Research, 9, 243-255.

Breslow, L. and Tai-Seale, T. (1996) An experience with healthpromotion in the inner city. American Journal of HealthPromotion, 10, 185-188.

Bryant, C. (1993) Interrater Reliability of the Level ofInstitutionalization (Loin) Scale. Department of Public HealthEducation, University of North Carolina at Greensboro.

Claquin, P. (1989) Sustainability of EPl: Utopia or Sine QuaNon Condition of Child Survival. REACH, Arlington, VA.

Eng, E. and Parker, E. (1994) Measuring communitycompetence in the Mississippi Delta: the interface betweenprogram evaluation and empowerment. Health EducationQuarterly, 21, 199-220.

Flynn, B. S. (1995) Measuring community leaders' perceivedownership of health education programs: initial tests ofreliability and validity. Health Education Research, 10,27-36.

Gertler, P. and van der Gaag, J. (1990) 77K; Willingness to Payfor Medical Care: Evidence from Two Developing Countries.The Johns Hopkins University Press, Baltimore, MD.

Glaser, E. M. (1981) Durability of innovations in human serviceorganizations: a case study analysis. Knowledge: Creation,Diffusion, Utilization, 3, 167-185.

Goeppinger, J. and Baglioni, A. J., Jr (1985) Communitycompetence: a positive approach to needs assessmentAmerican Journal of Community Psychology, 13, 507-523.

Goodman, R. M. and Steckler, A. B. (1987/88) The life anddeath of a health promotion program: an institutionalizationcase study. International Quarterly of Community HealthEducation, 8, 5-21.

Goodman, R. M. and Steckler, A. B. (1989a) A framework forassessing program institutionalization. Knowledge in Society:The International Journal of Knowledge Transfer, 2, 57-71.

Goodman, R. M. and Steckler, A. B. (1989b) A model for theinstitutionalization of health promotion programs. Familyand Community Health, 11, 63—78.

Goodman, R. M., McLeroy, K. R., Steckler, A. B. and Hoyle,R. H. (1993a) Development of level of institutionalizationscales for health promotion programs. Health EducationQuarterly, 20, 161-178.

Goodman, R. M., Burdine, J. N., Meehan, E. and McLeroy, K.R. (eds) (1993b) Special issue: Community coalitions forhealth promotion. Health Education Research, 8(3).

Green, L. W. (1989) Is institutionalization the proper goal ofgrant-making? American Journal of Health Promotion, 3,44.

Green, L. W. and Kreuter, M. W. (1991) Health PromotionPlanning: An Educational and Environmental Approach, 2ndedn. Mayfield, Mountain View, CA.

Halperin, W. and Baker, E. L., Jr (eds) (1992) Public HealthSurveillance. Van Nostrand Reinhold, New York.

Haws, J., Bakamjian, L, Williams, T. and Lassner, K. J. (1992)Impact of sustainability policies on sterilization services inLatin America. Studies in Family Planning, 23, 85-96.

Henderson, D. A. (1987) Principles and lessons from thesmallpox eradication programme. Bulletin of the WorldHealth Organization, 65, 535-546.

Holland, B. K, Foster, J. D. and Louria, D. B. (1993) Cervicalcancer and health care resources in Newark, New Jercey,1970 to 1988. American Journal of Public Health, 83,45-48.

Israel, B. A., Cbeckoway, B., Schulz, A. and Zimmerman, M.(1994) Health Education and community empowerment'conceptualizing and measuring perceptions of individual,organizational, and community control. Health EducationQuarterly, 21, 149-170.

Jackson, C, Fortmann, S. P., Flora, J. A., Melton, R. J., Snider,J. P. and littlefield, D. (1994) The capacity-building approachto intervention maintenance implemented by the StanfordFive-City Project Health Education Research, 9, 385-396.

Janz, N. K., Zimmerman, M. A., Wren, P. A., Israel, B. A.,Freudenberg, N. and Carter, R. J. (1996) Evaluation of 37AIDS prevention projects: successful approaches and barriersto program effectiveness. Health Education Quarterly, 23,80-97.

Jensen, E. R. (1991) Cost-effectivenesi and financialsustainability in family planning operations research. InSeidman, M. and Horn, M. (eds), Operations Research:Helping Family Planning Programs Work Better. Wiley-Iiss,New York.

Jezek, Z,, Khodakevich, L. N. and Wickett, J. F. (1987)Smallpox and its post-eradication surveillance. Bulletin ofthe World Health Organization, 65, 425-^34.

Joseph, S. (1993) Editorial: Tuberculosis, again. AmericanJournal of Public Health, 83, 647-648.

Kaluzny, A. D. (1991) Community cancer programs as strategicalliances: challenges and guidelines for action. CanadianMedical Association Journal, 144, 1427-1432.

King, M. (1990a) Health is a sustainable state. Lancet, 336,664-667.

King, M. (1990b) Public health and the ethics of sustainability.Tropical and Geographical Medicine, 42, 197-206.

Kinne, S., Thompson, B., Chrisman, N. J. and Hanley, J. R.(1989) Community organization to enhance the delivery ofpreventive health services. American Journal of PreventiveMedicine, 5, 225-229.

Lafond, A. K. (1995) Improving the quality of investment inhealth: lessons on sustainability. Health Policy and Planning,10 (Strppl.), 63-76.

Lefebvre, R. C. (1990) Strategies to maintain and institutionalizesuccessful programs: a marketing framework. In Bracht, N.(ed.), Health Promotion at the Community Level. Sage,Newbury Park, CA.

Levey, S. andLoomba, N. P. (1973) Health Care Administration:A Managerial Perspective. Lippincott, Philadelphia, PA.

Levine, D. M., Becker, D. M., Bone, L. R., Stilhnan, F. A.,Tuggle, M. B., n, Prentice, M., Carter, J. and Fitippeli, J.(1992a) A partnership with minority populations: acommunity model of effectiveness research. Ethnicity andDisease, 2, 296-305.

Levine, D. M., Becker, D. M. and Bone, L. R. (1992b)Narrowing the gap in health status of minority populations:a community-academic medical center partnership. AmericanJournal of Preventive Medicine, 8, 319-323.

107



ownloaded from



Levine, D. M., Becker, D. M., Bone, L. R., Hill, M. N., Tuggle,M. B., II, and Zeger, S. L. (1994) Community-academichealth center partnerships for underserved minoritypopulations: one solution to a national crisis. Journal of theAmerican Medical Association, Til, 309-311.

Mamon, J. A., Shediac, M. C , Crosby, C. B., Celentano, D.D., Sanders, B. and Matanoski, G. M. (1991) Developmentand implementation of an intervention to increase cervicalcancer screening in inner-city women. InternationalQuarterly of Community Health Education, 12, 21-34.

Mburu, F. M. and Boerma, J. T. (1989) Community-basedhealth care 10 years post Alma Ata. Social Science andMedicine, 28, 1005-1006.

McKnight, J. L. (1987) Regenerating community. Social Policy,Winter, 54-58.

Minkler, M. (1994) Challenges for health promotion in the1990s: social inequities, empowerment, negativeconsequences, and the common good. American Journal ofHealth Promotion, 8, 403-413.

Monahan, J. L. and Scheirer, M. A. (1988) The role of Unkingagents in the diffusion of health promotion programs. HealthEducation Quarterly, 15, 417^*33.

Orlandi, M. A. (1986) The diffusion and adoption of worksitehealth promotion innovations: an analysis of barriers.Preventive Medicine, 15, 522-536.

Prasad, B. and de L Costello, A. M. (1995) Impact andsustainability of a 'baby friendly' health educationintervention at a district hospital in Bihar, India. BritishMedical Journal, 310, 621-623.

Pressman, J. L. and Wildavsky, A. (1979) Implementation, 2ndedn. University of California Press, Berkeley, CA.

Puska, P., Salonen, J. T, Koskela, K., McAlister, A., Kottke,T. E., Maccoby, N. and Farquhar, J. W. (1985) Thecommunity-based strategy to prevent coronary heart disease:conclusions from the ten years of the North Karelia ProjectAnnual Review of Public Health, 6, 147-193.

Reichman, L. B. (1993) Fear, embarrassment, and relief: thetuberculosis epidemic and public health. American Journalof Public Health, 83, 639-641.

Resnicow, K. and Botvin, G. (1993) School-based substanceuse prevention programs: why do effects decay? PreventiveMedicine, 22, 484^*90.

Rifkin, S. B. (1986) Lessons from community participation inhealth programmes. Health Policy and Planning, 1, 240-249.

Rifkin, S. B., Muller, F. and Bichmann, W. (1988) Primaryhealth care: on measuring participation. Social Science andMedicine, 26, 931-940.

Robertson, A. and Minkler, M. (1994) New Health PromotionMovement a critical examination. Health EducationQuarterly, 21, 295-312.

Rogers, E. M. (1983) Diffusion of Innovations. The Free Press,New York.

Ruttan, V. W. (Editor) (1994) Agriculture, Environment, andHealth: Sustainable Development in the 21st Century.University of Minnesota Press, Minneapolis, MN.

Scheirer, M. A. (1990) The life cycle of an innovation: adoptionversus discontinuation of the Fluoride Mouth Rinse Programin schools. Journal of Health and Social Behavior, 31,203-215.

Schwartz, R., Smith, C , Speers, M. A., Dusenbury, L. J.,Bright, E, Hedlund, S., Wheeler, F. and Schmid, T. L. (1993)Capacity-building and resource needs of state health agenciesto implement community-based cardiovascular diseaseprograms. Journal of Public Health Policy, Winter, 480-494.

Shea, S., Basch, C. E., Lantigua, R. and Wechsler, H. (1992)The Washington Heights-Inwood Healthy Heart Program: athird generation community-based cardiovascular diseaseprevention program in a disadvantages! urban setting.Preventive Medicine, 21, 203-217.

Shea, S., Basch, C. E., Wechsler, H. and Lantigua, R. (1996)The Washington Heights-Inwood Healthy Heart Program: a6-year report from a disadvantaged urban setting. AmericanJournal of Public Health, 86, 166-171.

Steckler, A. and Goodman, R. M. (1989) How to institutionalizehealth promotion programs. American Journal of HealthPromotion, 3, 34-^4.

Streefland, P. H. (1989) The continuity of vaccinationprogrammes: reflections and a case from Gujarat, India.Social Science and Medicine, 29, 1091-1098.

Streefland, P. H. (1995) Enhancing coverage and sustainabilityof vaccination programs: an explanatory framework withspecial reference to India. Social Science and Medicine, 41,647-656.

US Agency for International Development (1988) Sustainabilityof Development Programs: A Compendium of DonorExperience. USAID, Washington, DC.

Wallerstein, N. (1992) Powerlessness, empowerment, andhealth: implications for health promotion programs. AmericanJournal of Health Promotion, 6, 197-205.

Wallerstein, N. and Bernstein, E. (1994a) Special issue:Community empowerment, participatory education, andhealth—Part I. Health Education Quarterly, 21 (2).

Wallerstein, N. and Bernstein, E. (1994b), Special issue:Community empowerment, participatory education, andhealth—Part II. Health Education Quarterly, 21 (3).

Weisbrod, R. R., Pine, P. L. and Bracht, N. F. (1992) Impactof a community health promotion program on existingorganizations: the Minnesota Heart Health Program. SocialScience and Medicine, 34, 639-648.

Yacoob, M. and Walker, J. (1991) Community management inwater supply and sanitation projects: costs and implications.Aqua, 40, 30-34.

Yin, R. K. (1979) Changing Urban Bureaucracies: How NewPractices Become Routinized. D. C. Heath, Lexington, MA.

108



ownloaded from


Documents

Health Educ. Res. 1998 Shediac Rizkallah 87 108