Making Supervision Supportive and Sustainable: New Approaches

m a x i m i z i n g a c c e s s a n d q u a l i t y

MAQ PAPER NO. 4 . 2002

M A QMaking Supervision Supportive andSustainable: New Approaches to OldProblems

EXECUTIVE SUMMARY

Supervision provides critical support for the delivery of health services. Despiterecognition of the importance of supervision in managing human resources forhealth care, the “promise” of supervision is often not achieved in developing-country health systems. Supervision still tends to emphasize inspection and controlby external supervisors, who often believe that workers are naturally unmotivatedand require strong controls to perform adequately. At the same time, many linesupervisors lack the requisite technical and managerial skills or have limited author-ity to resolve service delivery problems.

Numerous studies and projects have attempted to inform and improve supervi-sion of primary health care and family planning programs in developing countries.While many supervision interventions have been effective on a pilot basis, sustainedimprovements have been elusive. Too often, short-term successes have fadedthrough staff turnover, end of donor support, and failure to make the systemicchanges in human resource management necessary to maintain gains in healthworker performance.

Evidence favors a different approach to supervision. Research studies andprogram evaluations suggest a different approach to make supervision more condu-cive to improvement in health worker performance: supportive supervision. Support-ive supervision expands the scope of supervision methods by incorporatingself-assessment and peer assessment, as well as community input.

Supportive supervision shifts the locus of supervisory activity from a

single official to the broader workforce. A key concept in supportive supervi-sion is that it is a process implemented by many parties, including officially desig-nated supervisors, informal supervisors, peers, and health providers themselves.Supportive supervision promotes quality outcomes by strengthening communica-tion, focusing on problem-solving, facilitating teamwork, and providing leadershipand support to empower health providers to monitor and improve their ownperformance.

Continuous implementation of supportive supervision generates sus-

tained performance improvement. Instead of occurring only when an externalsupervisor visits a facility, supportive supervision takes place continuously, asongoing performance monitoring and quality improvement become a routine part ofhealth workers’ jobs. Supportive supervision occurs in multiple places: on the job,both formally and informally; in one-on-one meetings; in peer discussions; inmeetings outside the work site; and when health workers review their own perfor-mance against standards.

MAQ

PAP

ERS

Supplement to Population Reports, Volume XXX, No. 4

2

m a x i m i z i n g a c c e s s a n d q u a l i t yM A Q PA P E R NO. 4 . 2002

MAQ

PAP

ERS

Making Supervision Supportive

and Sustainable: New Approaches

to Old Problems

A U T H O R S

Lani Marquez

University Research Co., LLC

Linda Kean

University Research Co., LLC

This MAQ paper was produced for the

MAQ Initiative by the Management and

Leadership Program of Management

Sciences for Health under USAID coop-

erative agreement HRN-A-00-00-00014-

00 and distributed by the INFO Project at

the Johns Hopkins Center for Communica-

tion Programs under USAID cooperative

agreement GPH-A-00-02-00003-00.

The development of this paper was supported

by the Bureau for Global Health of the United

States Agency for International Development.

This paper does not represent official state-

ments of policy by USAID or the position of

the organization represented by the authors.

This paper is a publication of the Maximizing Access and Quality (MAQ)

Initiative—an initiative of the United States Agency for International Develop-

ment (USAID), cooperating agencies, country partners, and other collabora-

tors to apply state-of-the-art methods to maximize access to and the quality

of family planning and other selected reproductive health services.

M A Q

MAQ PAPER NO. 4 . 2002

Supportive supervision is feasible. Recentexperiences from countries in different regions showthat supportive supervision offers a powerful alterna-tive to traditional approaches. Research findings alsoprovide compelling evidence for the effectiveness ofkey elements and tools of supportive supervision,including structured audit and feedback, self-assess-ment, and peer assessment. Supportive supervisionrequires:

■ new thinking about who does supervision and howand when it occurs;

■ motivation on the part of supervisors and staff aliketo adopt new behaviors;

■ locally appropriate and tested tools;

■ time and investment to establish and take root;

■ the commitment of top management and somedecentralized decision-making authority;

■ integration into existing human resource manage-ment systems rather than creation of a parallelsystem to “work around” problems.

More evidence of results and savings will help

advocacy to expand supportive supervision. Moreevaluation of the costs of supportive supervision andits impact on health services performance willstrengthen the case for health organizations to adoptsupportive supervision approaches, as will betterdocumentation and broader sharing of tools forsupportive supervision. These activities will also revealbest practices for (1) motivating, training, and coach-ing external supervisors to perform supportive super-vision and (2) motivating and enabling health workersto effectively conduct self-assessment, peer assess-ment, and internal supervision.

3

W H Y W E S H O U L D I N V E S T I N S U P E R V I S I O N M A Q PA P E R NO. 4 . 2002

INTRODUCTION

This paper distills lessons from recent efforts toimprove the supervision of family planning and healthprograms in developing countries and identifiesapproaches that may be more effective and sustainable.It describes supportive supervision, an approach tosupervision that emphasizes joint problem-solving,mentoring, and two-way communication betweensupervisors and those being supervised. It also ex-pands the concept of effective supervision by explor-ing how self-assessment and peer assessment, as wellas community input, can be seen as vital componentsof results-oriented, supportive supervision.

The paper’s conclusions are based on a review of theliterature of supervision; an informal, qualitativesurvey conducted among USAID reproductive healthand child survival cooperating agencies; and discus-sions with MAQ participants. The agencies includeJohn Snow International, Management Sciences forHealth, University Research Co., LLC, EngenderHealth,Pathfinder International, the International PlannedParenthood Federation, JHPIEGO, the Johns HopkinsCenter for Communication Programs, and INTRAH.The paper draws on responses summarizing theexperiences of approximately 16 field programs. Basedon review of these experiences in implementing such astyle of supervision, the paper provides a frameworkfor what supportive supervision means in practice andidentifies key lessons from recent efforts, as well asgaps in our knowledge.

WHY WE SHOULD INVEST IN SUPERVISION

There is widespread agreement among those in theinternational health community that supervision is acritical part of human resource management for thedelivery of basic health services.1 As noted in anearlier MAQ Paper (Vol. 1, No. 3, p. 8), supervision isone of the key approaches to improving the quality ofhealth care and the performance of health care provid-ers, especially given the labor-intensive nature ofhealth service delivery. This is particularly true indeveloping countries, where health program managers

recognize that supervision is “important to gettingthings done”2 and where supervision remains one ofthe most direct ways for an organization to affect whatits staff does.

At the same time, there is pervasive disappointmentthat the “promise” of supervision is frequently notrealized or sustained. USAID and its cooperatingagencies have invested significant financial and techni-cal resources in trying to strengthen supervisionsystems in developing countries through supervisortraining and development of supervisory tools andchecklists, yet interventions to strengthen supervisionsystems often have not sustained their effects past apilot phase or demonstrated results independent ofother efforts to improve service delivery.

Why invest in supervision in health and familyplanning delivery settings in developing countries? Isthere reason to expect that we can do better? The MAQSubcommittee on Management and Supervisionbelieves we can.

There are several reasons to be hopeful aboutrealizing the promise of supervision. First, there is anemerging consensus on what supervision shouldentail––what the key functions of supervision are.Second, a growing body of experience from differentsettings suggests that expanding the realm of howsupervision functions can be performed––with waysof doing supervision that involve health workersthemselves, peers, even communities––may allowsupervision to be accomplished more effectively andefficiently. Ample evidence supports the feasibility ofusing nontraditional, expanded supervision ap-proaches in developing countries. There is also limitedrecent evidence that these alternative approachesachieve better health worker performance and out-comes than traditional supervisory approaches, andsome evidence that these approaches may be moresustainable.

Growing, broad-based interest in improving qualityof care has also strengthened the argument for im-proving supervision, to use supervision as a means todirectly affect health worker performance and to buildhealth worker capacity for problem identification andresolution.3 There is evidence that providing health

4


professionals with effective supervision can helpimprove the quality of health care and even patientoutcomes.4 Moreover, a growing body of successfulexperience in developing countries shows how super-visors can be involved in leading quality and perfor-mance improvement.5

The push toward decentralization and deconcentrationof health care management functions in many countrieshighlights the importance of supervision. Whenresponsibility for oversight and technical direction istransferred from the central level to lower levels in thehealth service delivery system, these lower levels oftenhave a great need for capacity-building to implementtheir expanded supervisory role.

This paper examines the experience of USAIDcooperating agencies that are working to improvehealth and population services in developing countriesby strengthening or redesigning supervision systemsto make them more effective and sustainable. Learningfrom this experience begins with understanding whatsupervision is in the context of health service deliverysystems.

THE KEY FUNCTIONS AND TASKS OFSUPERVISION

Supervision is the process of “directing and sup-porting staff so that they may effectively perform theirduties.”6 Supervision may include periodic events,such as site visits or performance reviews, but it goesbeyond such episodes to refer to the ongoing relation-ship between health care providers and supervisors.Supervision includes oversight and implementation ofclinical and nonclinical tasks and activities that affectthe organization, management, and technical deliveryof health services, including control of work processesand systems, maintenance of facilities and infrastruc-ture, and monitoring and improvement of systemwideperformance and effectiveness. Beyond this technicalrole, there is also an important human dimension to

the supervisor-health worker relationship. In develop-ing countries, where many health workers work aloneor in small groups in remote sites, the supervisor mayrepresent the only link to the larger health system.

In the context of strengthening health and familyplanning services, numerous USAID cooperatingagencies have analyzed the major objectives andactivities of supervision in developing-countrysettings. Table 1 summarizes the main functions ofsupervision as described by several cooperatingagencies that collaborate on MAQ. While the defini-tions use different terminology, the essential func-tions and activities of supervision are similar:defining objectives, monitoring performance, provid-ing supplies, providing training, solving problems,and motivating and supporting health providers toimprove their performance.

Kilminster and Jolly, in their systematic literaturereview of supervision in medical practice settings,7

defined a similar role for supervision in health care inthe US and the UK. They argued that supervision inthe health professions consists of three basic func-tions: management, education, and support. Theydefined supervision as “the provision of monitoring,guidance, and feedback on matters of personal,professional, and educational development in thecontext of the doctor’s care of patients” in order to“ensure patient/client safety and promote professionaldevelopment.” The emphasis on patient safety andprofessional development of health care workers isworth noting, since the role of supervision in assur-ing these two goals has often not been made explicitin developing-country health systems. Supervision indeveloping countries has often been viewed only as aninstrument through which to impose the healthsystem’s needs on health care providers rather than asa means to address health workers’ multiple needs.However, the growing concern with quality assuranceand improvement in developing countries is creating aclimate where greater attention is paid to both patientand provider safety and human resource development.

5


Table 1. The Functions of Supervision

ManagementSciences forHealth1

Set individualperformanceobjectives (theactivities)

Monitor andevaluate perfor-mance

Manage perfor-mance problemsthat arise

Motivate staffmembers andprovide feedback,solve problems,and provideguidance, assis-tance, and support

Provide training

Assist withresources andlogistics

EngenderHealth2

Address providers’needs for goodmanagement andsupervision

Address providers’needs for goodsupplies and siteinfrastructure

Address providers’needs for informa-tion, training, anddevelopment

URC/QA Project3

Communicate andverify understand-ing of standards

Monitor perfor-mance by assess-ing compliancewith standards

Provide feedbackon errors andsuggest solutions

Educate and trainproviders

Assist in theidentification ofproblems impedingquality and in thedevelopment andenabling of solu-tions

Motivate, coach,and empower staffto solve problems

INTRAH/PRIME IIProject4

Set clear jobexpectations

Review perfor-mance andprovide immediateperformancefeedback

Ensure staff haveadequate physicalenvironment andtools

Motivate andrecognize goodperformance

Ensure staff haveappropriate skillsand knowledge

Facilitate organiza-tional support

JHPIEGO5

Plan and coordinatesupervision activi-ties

Set performanceobjectives, manageconflict, deploystaff, develop workteams, improvestaff motivation

Assess skills,provide guidanceand training

Interpret, use, andshare data

Manage suppliesand equipment

Identify and solveproblems

Communicateeffectively andprovide construc-tive feedback

1. Management Sciences for Health, “Supervising and Supporting Your Staff,” The Family Planning Manager’s Handbook (West Hartford,CT: Kumarian Press, 1991), pp. 115–55.2. B. Ben Salem and K. J. Beattie, “Facilitative Supervision: A Vital Link in Quality Reproductive Health Service Delivery,” AVSC WorkingPaper no. 10 (New York: EngenderHealth, 1996).3. W. Stinson, L. Malianga, L. Marquez, and C. Madubuike, “Quality Supervision,” QA Brief vol. 7, no. 1 (1998): 4–6. Bethesda, MD:Quality Assurance Project. Y. M. Kim, P. Tavrow, L. Malianga, S. Simba, A. Phiri, and P. Gumbo, “The Quality of Supervisor-ProviderInteractions in Zimbabwe,” Operations Research Results vol. 1, no. 5 (Bethesda, MD: Quality Assurance Project, 2002).4. INTRAH, “Supportive Supervision,” PRIME Page (Chapel Hill: University of North Carolina at Chapel Hill, 2002).5. L. Chege and R. Transgrud, “Defining a Performance Improvement Intervention for Kenya Reproductive Health Supervisors: Resultsof a Performance Analysis” (1999). http://www.reproline.jhu.edu/english/6read/6pi/super/tr_super1.htm.

The Supervision Process

At whatever level it occurs, the process of supervi-sion may be thought of as consisting of four basictasks, as shown in Figure 1. These tasks encompassthe functions described in Table 1 and are consistentwith the wide variety of approaches, tools, andmethods used for supervision and quality improve-

ment interventions. To emphasize the ongoing natureof the process, the diagram shows an action loop withno starting or end points. The supervisor facilitatesthis process by communicating about, assessing, andfacilitating the work of others. The supervisor’sactivities include the following:

6


■ Set expectations: A prerequisite for effectivesupervision is the existence of clear expectations orstandards against which performance and resultscan be measured. Where these do not exist, thesupervisor helps define and implement them.

■ Monitor and assess performance: Once thestandards or guidelines are set, the task of gaugingthe extent to which they are met becomes anongoing activity that occurs at all levels of thesystem: for individual health care providers, withinand among facilities, and at the district (regional)and national (central) levels.

■ Identify problems and opportunities: Wherethere are gaps between expectations and results, thesupervisor facilitates a team process for examiningpotential causes and possible solutions. By facilitat-ing open communication and teamwork, the super-visor can help spot opportunities to improve theoverall quality of care.

■ Take action: The supervisor helps marshal theresources necessary (human, financial, material,political, institutional) and motivates and supportsproviders to implement interventions and activitiesto address performance gaps or opportunities forimprovement. The process continues as new activi-ties begin, with the establishment of expectationsfor results.

This loop describes what supervision is intended todo but does not prescribe how. Differentiating super-vision functions from how these tasks are accom-plished helps to expand the range of possibilities forimproving supervision. In practice, supervisors canuse a wide range of methods for carrying out thesetasks. For example, a positive development in recentyears in many health systems has been replacing theconcept of a single, formal supervisor with a broaderapproach that involves multidisciplinary teams, as wellas health workers themselves, in the supervisionprocess. Similarly, political decisions to increasecommunity participation have led to support in manycountries for the creation of community health com-mittees, users’ committees, or other groups of commu-nity members that oversee health care delivery; suchgroups can also serve as agents of health care providersupervision. Peer review and peer supervision––engaging health care providers themselves in provid-ing technical support to colleagues––is anotherpromising approach, particularly for its applicabilityto the private sector.

In many countries, however, the practice of supervi-sion does not effectively perform the four basic tasksshown in Figure 1. To understand how new ways ofthinking about supervision can improve its effective-ness and sustainability, the next section reviewstypical problems with supervision in health servicedelivery systems in developing countries and howthese problems act as barriers to the effective perfor-mance of supervision functions.

SUPERVISION IN HEALTH PROGRAMSHAS TRADITIONALLY EMPHASIZEDINSPECTION RATHER THAN FACILITATION

Supervision in family planning and primary healthcare programs in developing countries has tradition-ally used an inspection and control approach––that is,controlling workers’ adherence to policies and proce-dures, based on an attitude that workers are naturally

7


unmotivated and require strong external controls toperform adequately. In most health service deliverysystems, supervisors tend to focus on administrativeissues, such as inspection of facilities (hours of opera-tion, maintenance), use of resources (financial, mate-rial, human), supply logistics (quantity, maintenance ofequipment, procurement), review of records, andcommunication of information and directives fromhigher to lower levels. In a climate of inspection andcontrol, problem-solving is reactive and episodic.Supervisors often blame individuals rather than look forroot causes in deficient processes. For this reason,traditional supervision systems have not tended to“empower” staff to engage in problem-solving and totake the initiative to improve service quality and access.

The tendency to focus on inspection arises in partfrom the lack of technical and managerial skills neededto engage health workers in analyzing problems andfinding solutions. Supervisors in developing countriesare often expected to deal with every type of problem,but they may lack the skills and knowledge to carryout that ambitious role. Supervisors who are notknowledgeable about the technical tasks thatsupervisees perform can monitor and support them,but they cannot effectively train them in these tasks.

Lack of skills, especially those related to communi-cation, team-building, and facilitation, which are vitalto directing and supporting health workers, is some-times exacerbated by lack of interest on the part ofsupervisors in performing key supervisory functionsor performing them well. This lack of interest isespecially likely to affect individuals who becomesupervisors as a result of a promotion or politicalconsiderations rather than due to genuine interest inor aptitude for the supervisory role. For example, afteran intervention to improve supervision in India bydefining a clearer role and responsibilities for supervi-sors and shifting emphasis toward joint problem-solving, some supervisors reported that they did notlike the more participatory, supportive style andpreferred the status they enjoyed when supervision

was geared toward inspection and control.8 But evenwhen supervisors want to be more effective, changecan be difficult and threatening.

Another pervasive problem is the lack of authorityof line supervisors to take action, for example, to helpsolve a problem, reward good performance, or sanctionpoor performance. Supervisors who have not beengiven authority to act or make decisions based onperformance have limited credibility with supervisees.

Limited approaches to conducting supervision poseanother barrier to the effective performance of super-vision. In general, supervision at health facilitiesaround the world continues to be carried out primarilythrough site visits by an external supervisor. The sitevisit is typically short—a couple of hours duringwhich the supervisor focuses on filling out forms andchecklists and reviewing these results (rather curso-rily) with the medical officer in charge at the facility.Often such visits are isolated events, not tied to whathappened during a previous site visit or to what mayhappen during the next visit. This type of traditional,“event-oriented” external supervision focuses on onlyone part of the supervision process.

A related problem is that resources for supervisoryactivities––for example, transportation resources orhuman resources for supervision––are frequentlyunavailable, even when budgeted or mandated byorganizational policy. Resource shortages result ininfrequent, episodic visits that are narrowly focusedon only part of the supervision process.

Finally, lack of planning, failure to define priorities,nonadherence to work plans, diversion of resourcesfrom planned allocations, lack of financial stability, andlack of accountability, and the low morale amonghealth workers that often results from these condi-tions, are all systemic problems plaguing healthsystems in many developing countries. Suchoverarching problems in the health sector underminethe effectiveness of supervision at all levels of thesystem and can make attempts to improve supervisionirrelevant.

8


These issues are summarized in Table 2. The nextsection summarizes lessons from efforts to try toimprove supervision through training and otherinterventions.

IMPROVING SUPERVISION REQUIRESCHANGING THE CONTENT AND THEOBJECTIVES AND NATURE OF SUPERVISION

There have been many projects and operationsresearch studies conducted during the past twodecades to improve the supervision of primary healthcare and family planning workers in developingcountries. The most frequent attempts to improvesupervision have involved increasing the frequencyand duration of supervisory visits, changing theactivities undertaken during supervision encounters,using multidisciplinary teams to conduct supervision,introducing tools for conducting supervision visits(for example, guidelines and checklists), and providing

additional training for supervisors. The effectivenessof such interventions has typically been evaluated interms of increases in the number of activities orservices performed by health workers. It has rarelybeen linked to health outcomes or improvements inservices. An exception is the work of the Family Plan-ning Management Development II (FPMD II) Projectwith Asociación Hondureña de Planificación de Familia(ASHONPLAFA) in Honduras, where the effects of aseries of interventions to improve supervision (includ-ing the development of supervision policies, training ofsupervisors, and use of data to prioritize sites needing asupervisory visit) were measured in terms of changes inaccess to and quality and sustainability of the servicesprovided. The lack of evidence of the effectiveness ofinterventions to improve supervision stems largely fromthe considerable methodological difficulties of demon-strating associations between supervision improvementsand health outcomes, independent of other interven-tions or health system changes.

Table 2. Typical Barriers to Effective Supervision in Developing-CountryHealth Systems

Problem

Narrow focus of supervision on inspection ofcertain areas

Punitive approach

Lack of supervisory skills and knowledge

Lack of supervisor authority to reward or sanctionperformance

Infrequent or irregular supervision due to lack ofresources

Lack of direction and accountability in the overallhealth system

How It Limits the Performance of BasicSupervision Tasks

Emphasizes monitoring to the detriment of otherkey supervisory tasks, especially problem-solvingand taking action

Inhibits problem-solving and demoralizes staff

Inhibits supervisors from effectively performing anyof the basic supervision tasks and underminessupervisor credibility

Deters supervisors and staff alike from takingaction, because no consequences (positive ornegative) result

Undermines continuity and limits supervision to onlycertain tasks, such as facility assessment

Undermines the performance of all supervisiontasks and demoralizes staff and supervisors alike

9


Providing tools and guidelines to structure thesupervisory encounter. The introduction of check-lists to guide supervisory visits has gained popularityin many developing-country health programs throughdonor-supported projects, particularly because suchinstruments lend themselves to focusing the supervi-sion encounter on the services or activities of greatestinterest to funding agencies. For example, operationsresearch studies funded by the Primary Health CareOperations Research (PRICOR) Project in the 1980sfound that providing guidelines, protocols, andchecklists to supervisors could increase the level ofactivity of health workers in priority health services,at least in the short term.9 These studies also notedthat when checklists attempted to be exhaustive, theytypically became quite lengthy and actually hinderedsupervision by causing fatigue and mechanical use.Anecdotally, many of the MAQ cooperating agenciesthat completed questionnaires for this paper reportedintroducing supervision guidelines, manuals, and/orschedules to systematize supervision and make it moreeffective. Most such supervisory tools have beenintroduced without rigorous evaluation of theireffectiveness on health worker performance. Anexception is the prospective, controlled field trial byLoevinsohn et al.10 in the Philippines to test theeffectiveness of supervisory visits based on an inte-grated checklist design to facilitate follow-up. Theyfound that health worker performance, as measured byaverage combined scores on the 20 indicators measuredby the checklist, improved by 44% (from 43% to62%) in the experimental group but by only 18%(45% to 53%) in the control group (p < .05). TheMinistry of Health of Uganda recently implementednational supervision guidelines, including checklists,for all levels of the system. These guidelines weredeveloped using quality assurance principles toredesign the national supervision system; an evalua-tion of the impact of the guidelines on health workerperformance is underway.

Increasing the frequency and/or duration ofsupervisory visits. Although the goal of many effortsto improve supervision is to increase the frequency ofsupervisory visits, there is little empirical basis forarguing for a particular minimum frequency. In thestudy cited above, Loevinsohn et al. also found acorrelation between frequency of supervisory visits(ranging from monthly to every six months) andimprovements in scores in their experimental groupbut not in the control group, suggesting that increas-ing the frequency of supervision helps only if theactivities that occur during supervision are productiveand directly related to improving health workerperformance. Foreit and Foreit found in a controlledfield trial conducted in a community-based contracep-tive distribution program in Brazil that reducing thefrequency of supervision visits from monthly toquarterly reduced costs considerably without nega-tively affecting program performance (as measured bynew acceptors, revisits, and distributor turnover).11

The authors speculated that a possible explanation forthis finding was that most supervisory visits wereconcerned with collecting service data and contributedlittle to actual program performance; thus, reducingfrequency did not matter.

Training supervisors. Performance analyses ofsupervision often find that health professionalscharged with supervision responsibilities lack theknowledge and skills needed to perform these respon-sibilities effectively.12 When surveyed, supervisors indeveloping-country health systems frequently com-ment on their need for more training, and training isprobably the most common intervention used in anattempt to improve supervision practices. Areas inwhich supervisors need training that were cited in thesurvey of MAQ cooperating agencies included problemidentification, problem-solving, time management,communication, monitoring, coaching, and technical/clinical updates. Like the use of checklists, trainingprograms for supervisors are rarely subjected to

10


rigorous evaluation. Moreover, it is well accepted inthe field of human performance technology thattraining alone will not produce sustained behaviorchange.13 In her review of supervision in familyplanning programs, Simmons noted that one-timetraining should not be expected to change supervisors’behaviors and that many of the skills needs for effec-tive supervision are best learned on the job or throughcoaching and mentoring.14

Changing the focus of supervision. Anothermajor trend in efforts to improve supervision has beento shift the focus of supervision encounters away fromsimply inspecting facilities and gathering servicestatistics to concentrate on the performance of clinicaltasks and resolution of problems experienced by thehealth worker, as well as to increase feedback fromsupervisors. Such efforts have often been accompaniedby the introduction of checklists and guidelines toreorient the supervisory encounter. Survey responsesfrom MAQ cooperating agencies for this paper aboutpractices in field projects consistently highlightefforts to refocus supervision toward activities such asassessing compliance with quality standards, transfer-ring knowledge and skills, providing clear feedback tosupervisees, identifying problems, and developingaction plans. Anecdotal evidence suggests that focus-ing on the clinical performance of providers duringsupervision does increase adherence to standards, atleast in the short term. However, there are few evalua-tions of the impact of such changes independent ofother interventions to improve program effectiveness.

Sustaining Improved Supervision Practices

Most efforts to improve supervision have been pilottests to demonstrate the efficacy of a particular ap-proach and have ignored the issue of the long-termaffordability of interventions that may have been

demonstrated to be feasible and effective. In only asmall number of studies were costs clearly measuredor even estimated. In the Philippines, Loevinsohn etal.15 reported that training supervisors to use asystematic checklist to guide supervisory encounterscost only US$19.92 per health facility and only $1.85per health facility per year in recurrent costs forforms. Vernon et al.16 found that the costs of supervi-sion could be reduced with better effects on healthworker performance (measured in terms of rates ofcontraceptive distribution) by replacing one of twoannual supervision visits with either group meetingsof all district service providers with the supervisor(costing $97 per year per supervised facility) orfacilitated problem-solving following self-assessmentby the supervised health workers (costing $114 persupervised facility). Traditional supervision (twoannual visits without specific orientation to problem-solving) cost $118 per supervised facility. This limitedevidence suggests that supervision can be improvedwithin the resource constraints of many developing-country health systems. More work is needed todocument the costs and results of supervisory mecha-nisms such as peer review and self-assessment and thecosts of internal and external supervision.

Beyond the financial sustainability of improvedsupervision practices, there is even less evidence abouthow to successfully institutionalize these practices indeveloping-country health care organizations. Asnoted above, most of the successful experiences withimproving aspects of supervision have been pilotefforts that have yet to be implemented on a sustainedbasis on a national scale. The many reasons for lack ofsustainable improvements to supervision includefunding shortfalls, inadequate training, and staffturnover. Pressure from donors or politicians todemonstrate short-term results has also exacerbatedthe problem by encouraging resource-intensive

11


interventions or the creation of parallel systems thatcannot be supported without outside assistance. Ingeneral, however, the failure to translate short-termsuccesses into lasting changes reveals a failure to makesystemic reforms to the management and delivery ofhealth care to support the gains in performance andquality of care that result from more effective supervi-sion. A key lesson from successful quality and perfor-mance improvement interventions is that for change tobe sustained and institutionalized, there must be aninternal enabling environment conducive to initiating,expanding, and sustaining the change. This enablingenvironment includes policies, leadership, organiza-tional values, and adequate resources to supportimproved practices.17 Short-term improvements insupervision thus cannot be sustained or successfullyscaled up unless organizations strengthen their overallhuman resource management systems.

Making Supervision More Effective

Supervision, like the delivery of family planningand primary health care services, has evolved inresponse to the growing emphasis on quality, process,and system issues. Evidence from the past two decadespoints to the need to change not only the frequency,duration, and structure of supervisory encounters, butalso the nature and objectives of supervision, to makeit more supportive and facilitative.18

Concern about these issues has led to more emphasison data collection and analysis, problem-solving, andteamwork in supervision, and to growing interest inthe facilitative possibilities of supervision and insupervision as a solution to the problem of the gradualdecline of health workers’ technical skills after train-ing. This evolution has also been supported by thedevelopment of clearer standards of care in bothvertical and integrated programs.19

This thinking is supported by the literature onquality management, which emphasizes the importanceof teamwork and of the listening, teaching, andfacilitative roles of effective leaders.20 It is also sup-ported by a recent comprehensive review of theliterature on effective supervision in postgraduatemedical education clinical practice in the UK and theUS, which found that the single most important factorassociated with better supervisory or performanceoutcomes was the quality of the supervisory relation-ship. Effective supervisors were those who have“empathy, offer support, flexibility, instruction,knowledge, interest in supervision, [and] good track-ing of supervisees, and are interpretative, respectful,focused and practical.”21 Other factors that theseauthors cited as contributing to the effectiveness ofsupervision were clear feedback about strengths andweaknesses and recognition of the need for healthworkers to have some control over and input into thesupervisory process.

In a study of the influence of managers’ behavior onnurses’ job satisfaction, McNeese-Smith22 identifiedseveral factors that affected nurses’ job satisfactionand, indirectly, performance. Positive behaviors bysupervisors included giving recognition, exhibitingleadership skills, being supportive of the team, facili-tating nurses doing their job, setting standards, andcreating open communication. Behaviors that nega-tively affected nurses were not providing recognition,lack of follow-up on agreements, criticizing in a crisis,and not communicating effectively.

Based on research and field experience about what hasand has not been effective in supervision, the next sectionof this paper describes how new ways of thinking aboutsupervision can lead to better health worker performanceand, ultimately, improved health outcomes.

12


MAKING SUPERVISION SUPPORTIVE CANLEAD TO BETTER OUTCOMES

Research studies and program evaluations recom-mend facilitating on-the-job learning, quality improve-ment, and problem-solving to make supervision moreconducive to improvement in health workers’ perfor-mance. These recommendations have coalesced, allow-ing discussion of a new approach to supervision,which the MAQ Subcommittee on Management andSupervision calls “supportive supervision.” Thissection describes what supportive supervision is andhow it is implemented. The next section summarizesevidence from field programs about the effectivenessof supportive supervision.

What Is Supportive Supervision?

Supportive supervision is a process that promotesquality at all levels of the health system by strength-ening relationships within the system, focusing on theidentification and resolution of problems, and helpingto optimize the allocation of resources. Supportivesupervision promotes continuous improvements in thequality of care by providing the necessary leadershipand support for quality improvement processes and bypromoting high standards, teamwork, and better two-way communication.

Supportive supervision is facilitative, fosteringrelationships that help improve individuals’ skills andperformance. Supervisors are intermediaries whoimplement institutional goals, solve problems at lowerlevels, and serve as a link to higher levels of authorityto resolve lingering problems.23 The supportivesupervisor brings people and resources together topursue clear objectives, assess results, and identifyand solve problems, and develops relationships basedon trust and responsiveness.

Supportive supervision focuses on the results ofprocesses as well as individual performance. It encour-ages open communication and building team ap-proaches that facilitate problem-solving. It focuses on

monitoring performance against expectations andusing data for decision-making. Measuring perfor-mance also allows supervisors to be held accountablefor results and helps foster continuous improvementsin quality at all levels of the health system

In this paper, supportive supervision is synonymouswith “facilitative supervision.” EngenderHealth(formerly AVSC) has defined facilitative supervision as“an approach to supervision that emphasizesmentoring, joint problem-solving, and two-waycommunication between the supervisor and thosebeing supervised.”24 Although EngenderHealth definesfacilitative supervision in the context of a broaderquality improvement approach that includes the use ofits client-oriented, provider-efficient (COPE) tech-nique, the principles and methods of supportivesupervision are the same. Supportive supervision alsoembraces the concepts articulated in ManagementSciences for Health’s work on team supervision, inwhich joint problem-solving is the focus of thesupervisory interaction, and supervisors act as on-the-job teachers who support their staff rather than asinspectors.25

Another key aspect of supportive supervision is anexplicit concern for meeting the needs of clients, bothexternal and internal (staff members). Experience withusing COPE for supportive supervision shows thatemphasizing the needs of the client helps focus theentire facility on solving problems as they arise. Thisfocus, in turn, improves staff performance and thequality of care.26 Supportive supervision also ad-dresses the human needs and aspirations of healthworkers themselves, recognizing that they need clearexpectations, feedback, skills, and materials to effec-tively perform their jobs, and a safe working environ-ment, recognition, and opportunities for professionaldevelopment and advancement.

Supportive supervision makes continuous improve-ments in the quality of care possible by providing thenecessary leadership and support for quality improve-ment processes, in the form of high standards, team-

13


work, and good communication. Indeed, many of thecharacteristics of supportive supervision parallel theemphasis of the quality movement, including a focuson the customer, team approaches, problem-solving,and strengthened systems and processes.27

Finally, a key concept in supportive supervision isthat it is implemented by multiple parties, includingofficially designated supervisors, informal supervisors,peers, and health care providers themselves. Whereastraditional supervision locates supervisory activityentirely in the person of the official supervisor,supportive supervision recognizes that supervisoryactivities to identify and solve problems, meet clientneeds, and motivate and support health providers indoing their jobs may be carried out by many actors,including health workers themselves.

Expanding the notion of who supervises has directimplications for where and when supervision occurs.While in traditional supervision approaches, supervi-sion most often takes place only when an externalsupervisor shows up at a facility, supportive supervi-sion occurs continuously, as ongoing performancemonitoring and quality improvement become a routinepart of health workers’ jobs. Supportive supervisiontakes place on the job, both formally and informally, inone-on-one meetings, in peer discussions, in meetingsoutside the work site, and when health workersreview their own performance against standards.

In sum, there are several key characteristics ofsupportive supervision:

■ The focus of supervision is on problem-solving toassure quality and meet client needs.

■ The entire team (including the external supervisor)is responsible for quality, so attention shifts fromindividuals to teams and processes.

■ Health providers are empowered to monitor andimprove their own performance.

■ The external supervisor acts as facilitator, trainer,and coach.

■ Health workers participate in supervising them-selves and each other.

■ Decision-making is participatory.

The hypothesis is that if providers are empoweredand motivated to provide high-quality services andtheir needs satisfied, and if clients’ needs are satisfied,then client satisfaction and outcomes will be enhanced.Box 1 summarizes these expected outcomes of support-ive supervision.

Box 1. Expected Results of ImplementingSupportive Supervision

Service delivery sites provide access to qualityservices that clients want or need.

Service providers and institutions continuouslyseek ways to improve the quality of their ser-vices.

Service providers and institutions are responsiveto client needs.

Service providers and supervisors are continu-ously improving their own performance, haveopportunities for increased job satisfaction, andsee their work as part of a larger picture.

Supervisors provide encouragement and supportto providers in continuously improving the qualityof services.

Supervisors are able to help sites translateinstitutional goals into services that clients wantand need.

Supervisors are able to provide managementwith information about the quality of servicesbeing provided and help identify constraints toimproving that quality.

There is a reduction in the costs of poor quality.

Source: B. Ben Salem and K. J. Beattie, “Facilitative Supervision:A Vital Link in Quality Reproductive Health Service Delivery,”AVSC Working Paper no. 10 (New York: EngenderHealth, 1996).

14


Implementing Supportive Supervision:Rethinking Who, How, and When

Implementing and institutionalizing supportivesupervision present an enormous challenge for healthsystems, especially those dominated by longstandinghierarchies as reflected in traditional, inspection-and-control supervision.28 Introducing a new approach tosupervision represent fundamental organizationalchange. Implementing supportive supervision requiresrethinking who does supervision, and how and whenit occurs, and it represents a major change fromtraditional approaches to supervision (see Table 3). Yet,as will be discussed in the next section, it can be done.

Operationalizing this concept of expanding thelocus of supervision to include peers and healthworkers themselves can be understood by viewingsupportive supervision as happening at three levels:external supervision (from outside the health facility),internal supervision (from within the health facility),and self- and peer supervision (by health workers, of

themselves and each other). As depicted in Figure 2,these three mechanisms of supervision are simulta-neous, complementary, and overlapping.

■ External supervision is the process used tooversee the operations and performance of individu-als and facilities within a larger system, such as adistrict, regional, or national health system. Exter-nal supervisors make site visits; set and implementclear program goals and standards; jointly defineperformance expectations with supervisees; monitorperformance against those expectations; allocateresources within the system; facilitate supervision atlower levels of the system; and follow up to solveproblems that require intervention from higherlevels of the health system.

■ Internal supervision is the process in a particularfacility or department to oversee the performance ofindividuals and the quality of service delivery.Internal supervisors set and monitor standards;

Table 3. Comparison of Traditional and Supportive Supervision

Traditional Supervision

External supervisors designated bythe service delivery organization

During periodic visits by externalsupervisors

Inspection of facility, review ofrecords and supplies, supervisormakes most of the decisions,reactive problem-solving by super-visor, little feedback or discussionof supervisor observations

No or irregular follow-up

Supportive Supervision

External supervisors designated by the servicedelivery organization, staff from other facilities,colleagues from the same facility (internalsupervision), community health committees,staff themselves through self-assessment

Continuously: during routine work, team meet-ings, and visits by external supervisors

Observation of performance and comparison tostandards, provision of corrective and support-ive feedback on performance, discussion withclients, provision of technical updates or guide-lines, on-site training, use of data and clientinput to identify opportunities for improvement,joint problem-solving, follow-up on previouslyidentified problems

Actions and decisions recorded, ongoing moni-toring of weak areas and improvements, follow-up on prior visits and problems

Who performssupervision

When supervi-sion happens

What happensduring supervi-sion encounters

What happensafter supervi-sion encounters

15


support and motivate providers with materials,training, and recognition; build teams and promoteteam-based approaches to problem-solving; fostertrust and open communication; and collect and usedata for decision-making.

■ Self- and/or peer supervision is the process bywhich individuals monitor and improve their ownskills and performance or that of their colleagues.The process encompasses setting clear performanceexpectations (including professional standards);assessing skills and measuring performance; elicit-ing customer feedback; and monitoring healthoutcomes, among others.

Together, these three mechanisms of supervisionhelp guarantee that the overall supervisory processprovides adequate support and succeeds in accom-plishing the core tasks of supervision: set expecta-tions, monitor and assess performance, identifyproblems and opportunities, and take action. Theshaded area of Figure 2 represents the full performanceof the basic tasks of supervision through the contribu-tions of each mechanism of supportive supervision.

The MAQ Subcommittee on Management andSupervision sees the interplay among these threemechanisms of supervisory activity as dynamic andfluid, with the relative importance and effectiveness ofthe three types of activities changing over time. The

ability of the system to self-adjust to address supervi-sory weaknesses or build on strengths is a strength ofsupportive supervision.

Ideally, a supportive supervision system involves allthree; but in practice, many programs that attempt toimplement supportive supervision may find that onesupervision mechanism is initially emphasized over theother two. In the long term, the greatest impact isexpected when all three types of supervision takeplace. Nevertheless, supervision can still be supportiveeven if all three types of supervision do not occur oroccur to different degrees.

For example, when self- and/or peer supervision ishighly effective, internal and external supervisors havemore time to focus on improving other aspects of theoverall quality of care. Conversely, if there are gaps inself/peer supervision, then internal and/or externalsupervisors will need to devote more time and re-sources to monitoring the performance of individuals.

The relationship between internal and externalsupervision is similar: where internal supervision isstrong and effective, external supervisors can focus onother activities to improve the overall quality of thesystem (such as sharing effective approaches amongfacilities and addressing systemwide needs). But whereinternal supervision is deficient, the external supervi-sion process must be refocused to address problemsand fill gaps.

At present, there is more emphasis and dependenceon external supervision in most developing-countryhealth care systems. But external supervision isgenerally the most resource-intensive and logisticallydifficult mechanism. External supervisors generallyoccupy higher positions in the organizational hierar-chy; they tend to have more skills and experience andusually receive higher compensation. Furthermore,external supervision is resource intensive: it usuallyrequires travel and involves numerous people at a site.

In contrast, internal and self/peer supervision aregenerally done on site, with no travel, which means thatstaff members spend less time away from the job. Fewerpeople are generally involved in these processes, decreas-

16


ing the human resource costs associated with them.While introducing supportive supervision requires

resources initially to develop staff capacity to imple-ment it, in the long term, organizations may end upspending less on supervision once external supervisorsengage in more facilitative, team-based activities andhealth providers regularly perform self- and peersupervision.29

EVIDENCE POINTS TO THE FEASIBILITYAND EFFECTIVENESS OF SUPPORTIVESUPERVISION

Evidence from Field Experiences inImplementing Supportive Supervision

Responses to the survey of MAQ cooperatingagencies commissioned by the Subcommittee on

Management and Supervision revealed that while manyCA-supported projects are attempting to implementaspects of supportive supervision, there are few well-developed, ongoing examples of supportive supervi-sion. Most of these experiences have been described inproject reports but not rigorously evaluated, have beenimplemented only experimentally, or offer only partialevidence to substantiate the effectiveness of supportivesupervision. Nonetheless, these experiences takentogether, in different regions and in health systems atdifferent levels of development, demonstrate thefeasibility of supportive supervision and suggest thatit offers a potentially powerful alternative to tradi-tional supervision.

Table 4 provides a summary of how supportivesupervision has been implemented in some of the “bestevidence” programs and with what results. These sixprograms include four receiving technical support fromEngenderHealth (in Bangladesh, Kenya, Nepal, andTanzania); one supported by JHPIEGO, ManagementSciences for Health, and the Center for CommunicationPrograms (Brazil); and one supported by ManagementSciences for Health (Honduras). Four are implementedby private-sector family planning organizations andtwo by government health agencies (national and state).All six experiences incorporate supportive supervisionas part of a larger quality-focused approach: four aspart of quality improvement including COPE and two aspart of performance improvement.

Of the six experiences described, one (Brazil) is ademonstration project whose results are only nowbeing scaled up. The Honduras, Kenya, and Nepal casesdemonstrate the application of supportive supervisionthroughout an organization, although within NGOsproviding a narrow set of services. The Nepal andTanzania experiences are an interesting combination of

The Supervisor’s View of SupportiveSupervision

“My role has changed. I have more autonomy thanbefore and that has given me more confidence tosupport the clinics rather than to check up onthem. I used to only go to the clinics to find faults,but now I go to support them. My job is to makesure that the clinics have enough supplies, to helpthem maintain equipment, to troubleshoot, to helpwith record keeping, and to train staff on the jobwhere I can. I feel that they like me more in mynew role and that makes my job more rewarding.”

—Njagi Muchiri, FPAJ Area Manager for Nairobi,Kenya, cited in J. Bradley, “Using COPE to

Improve Quality of Care: The Experience of theNational Family Planning Association of Kenya”

17


Table 4. Well-Developed Field Experiences in Implementing Supportive Supervision

Interventions to AchieveSupportive Supervision

Performance Improvement model(including setting standards forquality performance; training andtechnical assistance to increaseprovider and client capacity toperform at minimum standards;logistics and facility improvements;and accreditation system to moti-vate and reinforce continuouslearning and improvement), self-assessments, external accreditationsurveys, monthly progress reviewmeetings; supervisors lead qualityimprovement teams.

Results

At end of 18-month demonstra-tion project, fourof the five partici-pating clinicsreceived accredi-tation; servicequality improvedin all clinics.

Country/Context

Brazil/PROQUALI,Secretariats ofHealth of Bahia andCeará

JHPIEGO, MSH,CCP1

Lessons

The use of self-assessment andfocus on quality improvementmotivated and empowered bothsupervisors and health careproviders. Performance im-provements were greatestwhere baseline performancewas moderate to high, wherethe project champion had for-mal position of power, andwhere staff resistance waslowest.The successful demon-stration is now being scaled upto include approximately 25more clinics in Bahia and Cearástates.

Bangladesh/ Ministryof Health, NationalIntegrated Populationand Health ProgramQuality ImprovementPartnership

EngenderHealth2

Clinical supervision teams that su-pervise gov’t and NGO facilities.The gov’t-managed teams providecontraceptive updates and orienta-tion, support use of COPE, conductannual quality assurance site visitsto assess essential health services.The Quality Improvement Partner-ship (QIP) has developed standards,guidelines, job aids, and curricula.QIP also supports 45 urban andrural NGOs with over 300 staticclinics and 9,000 satellite clinics.

Data from annualvisits show gen-eral improvementof quality of es-sential servicesat NGO sites.Composite indica-tors for clinicfacilities, counsel-ing, family plan-ning, sexuallytransmitted infec-tions, and vitaminA services allshowed improve-ments in bothurban and ruralNGO sites.

A key lesson learned in design-ing a quality monitoring andsupervision system for NGOclinics in the National Inte-grated Population and HealthProgram has been the impor-tance of involving multiplestakeholders in the process,including urban and rural NGOmanagers, clinic managers, andnational and international tech-nical experts and donors.

18


Table 4. Well-Developed Field Experiences in Implementing Supportive Supervision (cont.)


ResultsCountry/Context

Lessons

Honduras/ASHONPLAFA

Management Sci-ences for Health3

Participatory workshops withsenior management and se-lected supervisors from alllevels of the organization todevelop vision and strategiesfor improved supervision;new supervision manual,procedures, and tools devel-oped to guide supervisoryvisits. Supervision now in-cludes joint performanceplanning by supervisors andsupervisees, with evaluationand determination of perfor-mance goals every twomonths; ongoing long-dis-tance learning and annualsupervisor workshops todevelop and reinforce super-visor skills.

Increased open dialogue be-tween supervisors andsupervisees about perfor-mance planning and evalua-tion has led to high achieve-ment of performance goals.Standards for supervisory ex-cellence have been inte-grated into organization’soverall quality assurance sys-tem. The proportion ofASHONPLAFA’s operatingbudget obtained from localsources has continued to rise,from 51% before the super-vision interventions to 63%20 months after the launchof the new system. Clientsatisfaction has remainedhigh (97%), and client ac-cess, as measured in coupleyears of protection (CYP),has likewise increased, from185,808 in 1999, to237,105 in 2000. At the be-ginning of June 2001, CYPwas 105,434, indicating acontinued upward trend.

Before the supervision sys-tem interventions, the oper-ating units of the organiza-tion were under strong pres-sure to reduce operatingcosts to meet sustainability(self-financing) goals. Bud-gets, including funding for su-pervisory field visits, werecut, hindering the organiza-tion’s ability to achieve itsgoal of improving servicequality. However, supportfrom senior management andemployees throughout the or-ganization for creative alter-native approaches tostrengthen supervision re-sulted in more effective su-pervision practices withoutincreasing costs.

19




COPE, improved over time;whole-site training; support-ive supervision

Results

Staff at all 14 FPAK repro-ductive health clinics carryout facility self-assess-ment and problem-solvingsessions every threemonths.

Country/Context

Kenya/Family Plan-ning Association ofKenya (FPAK)

EngenderHealth4

Lessons

FPAK concluded that qualityimproved after introducingCOPE and supportive supervi-sion, whereas it had not im-proved after training,introduction of guidelines andsupervisory checklists, and“legislating quality” from thecentral level; greater partici-pation and empowerment inquality improvement, delega-tion of authority to improvequality, and encouragement todo self-assessment and act onit were needed to remove theblaming and punitive natureof the management system—to “trust employees to makedecisions for themselves.”Early attempts to use COPEresults to take punitive actionand to use peer assessmentrather than self-assessmentfailed—too threatening. COPE+ supportive supervision ismore labor intensive and re-quires motivation on the partof supervisors.

Nepal/Nepal FertilityCare Center

EngenderHealth5

“Quality of Care ManagementCenter” (QOCMC) staffed byNGO team but based in theMOH. Four field officers pro-vide monthly 2–3-day whole-site monitoring-supervisionvisits to 24 FP clinics in 21districts. Quick response tologistics and equipment prob-lems. Coordination of trainingneeds. Continuity betweenvisits. Monthly monitoring ofuniform set of quality of careindicators. Strong “customerorientation” on the part ofQOCMC staff.

Increases in aggregatescores (combining all 24service sites) for all qualityof care indicators overtwo years.

The client-oriented approachof QOCMC staff, due to theirbeing NGO employees, was adistinct departure from thetraditional attitude of thegov’t bureaucracy. The focuson performance of the wholesite, not individuals, fosteredteamwork and shared respon-sibility.

20




Training of project-hired areateams in facilitative supervi-sion; area teams providedsupervision to 120 gov’t andprivate clinics. In 1997, areateams began training gov’tDistrict Health ManagementTeams (DHMTs) in facilitativesupervision tools and ap-proaches; package of DHMTreproductive and child healthtraining and supervision toolsdeveloped in 1998. Supervi-sion approach includes in-creased involvement of gov’tsupervisors in problem-solving at both public andprivate sites; annual qualityself-assessment (Quality Mea-suring Tool) by facility staff.

Results

Increase in use of self-assessment (26% ofsites at baseline to 77%in 1999); increased par-ticipation of supervisorsin COPE; increased feed-back and support by su-pervisors; better coverageand scores on QMT self-assessment.

Country/Context

Tanzania/MOH, Fam-ily Planning Associa-tion of Tanzania,Marie Stopes

EngenderHealth6

Lessons

Strengthening supervisionwas only one part of a largerquality improvement effortthat also emphasized COPEand whole-site training.

1. ReproLine, “Lessons from the Field: Performance Improvement Case Studies: Brazil Case Study,” http://www.reproline.jhu.edu/english/6read/6pi/case/case_brz.htm, cited 8/30/01.

2. B. Ben Salem and K. J. Beattie, “Facilitative Supervision: A Vital Link in Quality Reproductive Health Service Delivery,” AVSC WorkingPaper no. 10 (New York: EngenderHealth, 1996). EngenderHealth, “EngenderHealth Annual Report to USAID, July 1, 2000–June 30,2001” (New York: EngenderHealth, 2001), submitted to the Office of Population, USAID, Washington, DC.

3. Performance Improvement Consultative Group, “Performance Improvement Case Study–ASHONPLAFA,” http://www.pihealthcare.org/lessons.htm, cited 7/15/02.

4. J. Bradley, “Using COPE to Improve Quality of Care: The Experience of the National Family Planning Association of Kenya,” Quality/Calidad/Qualité no. 9. Available at http://www.popcouncil.org/publications/qcq/qcq09.pdf, cited 10/29/02.

5. H. Stanley, D. R. Shrestha, M. Barone, and M. Linehan, “The Quality of Care Management Center in Nepal: Improving Services withLimited Resources,” AVSC Working Paper no. 13 (New York: EngenderHealth, 2001).

6. J. Bradley, E. Mielke, G. Wambwa, J. Mashafi, and M. Nasania, “Family Planning Services in Tanzania: Results from a Project to ImproveQuality, 1996-1999” (New York: EngenderHealth, 2000).

21


public and private collaboration in that supervisoryteams are employees of a private-sector organizationbut are providing support to public-sector staff andfacilities. The fact that the supervision teams in Nepalare NGO staff rather than MOH employees was notedas possibly contributing to the successful implementa-tion of supportive supervision. EngenderHealth hasnoted elsewhere that COPE often has greater success innongovernmental institutions than in public-sectorfacilities and has speculated that this difference isattributable to the relatively greater control that NGOstaff have over financial and human resources.30 TheBrazil case offers perhaps the best potential evidencefor the success of scale-up of supportive supervision inpublic-sector health systems, although the expansionof PROQUALI throughout the two state health systemsin Bahia and Ceará is still in an early stage.

Evidence for the Effectiveness of SpecificElements of Supportive Supervision

Another source of evidence for the effectiveness ofsupportive supervision is the findings of research onspecific elements of supportive supervision or toolsused. While there are limits to how far evidence forthe effectiveness of individual elements of supportivesupervision can be generalized to the approach as awhole, several studies do provide evidence for theeffectiveness of key elements of supportive supervision.

Structured audit and feedback. In Nepal, arandomized controlled trial of the effects of audit andfeedback by district health officers using a structuredchecklist that focused on prescribing practices inprimary health care facilities resulted in statisticallysignificant differences in adherence to standardtreatment schedules (22.2% in the control versus40.5% in the intervention districts).31 Zeitz et al.found in an uncontrolled trial that supervisors’ use ofa checklist for diarrhea case management duringmonthly visits to rural health facilities in Nigeriaresulted in improvements in history-taking, physical

examination, disease classification, treatment, andcounseling.32 In Mexico, Kim et al. found that struc-tured observation and focused performance feedbackby supervisors, accompanied by joint identification ofopportunities for improvement, increased facilitativecommunication and provision of information to clientsby rural doctors.33 The supervisor feedback, two-waydiscussion of how the provider could improve, andidentification of skills to develop through assignmentsto be recorded in a homework log served to add aneducational dimension to supervision that resulted inmeasurable improvements in provider performance.

However, an operations research study carried outby the Quality Assurance Project in Niger found thatchanging the content of the supervision encounter toexamine health worker compliance with IntegratedManagement of Childhood Illness (IMCI) standardsand provide structured feedback on key areas of IMCIperformance resulted in significant improvements inperformance only when the health workers alsoreceived on-site IMCI training.34 Feedback alone,without training of health workers in IMCI standards,did not improve health worker performance. At thesame time, among health workers who received IMCItraining after feedback on their clinical performancefrom supervisors, improvement was found only inthose areas that were targeted in supervisor feedback;performance of clinical tasks that were not the focus offeedback deteriorated. This finding suggests thatsupervisory feedback may have an impact on perfor-mance only when providers have the necessary knowl-edge and skills to perform in accordance withstandards.

Self-assessment and peer review. There is ampleevidence in the health field that self-assessment is auseful method for self-instruction and some evidencethat self-assessment can be effective in causing desir-able behavior change.35 The study by Kim et al. of theeffectiveness of self-assessment and peer review asinterventions to reinforce training in interpersonal

22


skills in Indonesia provides convincing evidence of theeffectiveness of these two mechanisms of supervisionin maintaining and, in some cases, increasing improve-ments in health worker performance after training.36

Interestingly, in the Indonesia study, self-assessmentand peer review had a greater impact among providerswith more than 10 years of experience, suggesting thatexperienced providers were better able to use andapply lessons from self-assessment and peer discus-sions than were their less experienced colleagues.37 Intheir Mexico work, Kim et al. found that askingdoctors to audiotape their own consultations and listento the tapes to critique their own communicationperformance increased the power of self-assessmentand proved to be a strong source of motivation toimprove. They concluded that self-criticism (from theself-assessment exercises, including the audiotaping)was a far more compelling motivator than outsidecriticism (from the supervisors).38 However, Kilminsterand Jolly concluded in their review that self-supervi-sion alone was not effective and that some input froman external supervisor or colleague was needed forimprovements to be achieved in performance.39

In Kenya, Stanback et al. conducted a rigorousevaluation of the impact of a one-time supportivesupervision visit that focused on reinforcing messagesabout adherence to new guidelines for reproductivehealth service delivery, after health workers partici-pated in training workshops to introduce the guide-lines.40 They found that adding the supportivesupervision visit, and a package of materials designedto help the trainee transfer information on the newguidelines to colleagues, resulted in statisticallysignificant gains of approximately nine percentagepoints in knowledge and practice scores beyond thegains achieved through training alone.

In an operations research study on the impact ofself-assessment and peer feedback on health worker

compliance with standards for fever care in children inMali, Kelley et al. found that combining self-assess-ment with peer feedback, when done regularly, canhave a significant effect on compliance with stan-dards.41 However, it is clear that self-assessmentrequires resources. All the individuals interviewedfrom the intervention pool characterized the extrawork required by the intervention as burdensome.The researchers reported that the intervention costabout $6 per provider.

Challenges in Implementing SupportiveSupervision

It is also important to consider the risks of support-ive supervision. Having too many actors with diffuseresponsibilities can make supervision a non-systemwith no one in charge––no one ultimately responsiblefor ensuring that supervisory tasks take place. Over-reliance on self-directed teams can place unreasonableor unrealistic performance expectations on the workforce.

Part of the challenge of the transition to a morefacilitative approach to supervision is that change canbe threatening and daunting. It is more labor inten-sive. It requires motivation and behavior change onthe part of supervisors. Some may believe that it is notachievable in developing-country settings. The experi-ences described above suggest otherwise. Supportivesupervision, especially when implemented in a largercontext of quality improvement, has been shown toincrease quality of care indicators, provider satisfac-tion with the support provided by the health system,and problem-solving at the facility level.

Even when the hurdle of effecting change at various(and sometimes all) levels of an organization can becleared, sustaining such improvements over time canbe an even greater challenge. Most field experiences inimplementing supportive supervision are still fairlyrecent and only now addressing scale-up issues.

23


LESSONS IN IMPLEMENTING ANDSUSTAINING SUPPORTIVE SUPERVISION

What lessons can be drawn from these experiences?

Supportive supervision requires motivation onthe part of supervisors and staff alike, to adoptnew behaviors. Supportive supervision requireschanges in human behavior. Supervisors are the criticalcatalysts for change and must themselves be convincedof the need for and value of the supportive supervisionapproach. Supervisors need to model appropriatebehaviors, and supervisors’ commitment is needed tosustain the approach. Such change is not easy toobtain and will require multifaceted interventionstrategies based on overcoming individuals’ resistanceto change, building on readiness to change, developingnew skills needed to effect the change, and adaptingthe work environment to reinforce the change. BenSalem and Beattie cited several personal characteristicsof supervisors as contributing to effective supportivesupervision, including the ability to delegate andcomplete work through others, the desire to achieve athigh levels, confidence in one’s own ability and theability of staff, and the ability to instill a sense ofvalue about the organization’s goals in others.42

The attitudes of supervisors and providers towardservice delivery must also change, to be integratedrather than vertical, and client focused rather thanprovider focused. Supervisors need to adopt ap-proaches that facilitate work, convey support, andbuild self-confidence among staff.

At the same time, the reports on the experiences ofcooperating agencies with supportive supervisionrepeatedly note that health workers often derivesatisfaction from self-assessment, joint problem-solving, and other participatory processes that pro-mote involvement, ownership, and commitment to

improvement. Supervisors also need support fromtheir superiors and positive reinforcement to sustainsupportive supervision. Such support may come fromboth within the health system or organization andoutside, such as from clients, community organiza-tions, NGOs, or professional associations.

Supportive supervision offers a wide range ofmechanisms to accomplish supervisory tasks. It isimportant to explore the alternatives for implementingsupportive supervision to determine what is workablein a given situation. Sometimes there are no internalsupervisors, or external supervisors may have inad-equate time and resources to provide the supportrequired for good performance. In such situations,peer support may be a feasible and effective alternativeto external supervision. For example, in a recentsurvey of private nurse-midwives in Kenya who havebeen trained by PRIME in the area of postabortioncare, about 50% of the providers surveyed stated thatthey are members of a group of nurse-midwivesorganized to support each other’s work in reproductivehealth, and most of these had contacted, or beencontacted by, a peer for assistance.43

Supportive supervision requires locally appro-priate and tested tools. The tools and techniques ofsupportive supervision, such as structured guidelinesfor external supervision and self-assessment instru-ments, have shown to be readily usable by healthworkers and supervisors in a wide range of settings.44

Responses from the survey emphasize the need forsimple tools, not long, involved ones that end updominating the supervisory encounter. Tools to helpsupervisors understand what they are to do and howto do it and defined standards of clinical practicefacilitate supportive supervision, giving providers andsupervisors alike clear objectives to which they candirect their efforts. Self-assessment tools must be easyto use and be able to be interpreted by the staff using

24


them. There is also a critical need for tools to addressboth client and community needs and incorporatethese into supervision to improve quality and perfor-mance.45

Supportive supervision takes time and invest-ment to establish. Because supportive supervisioninvolves behavior change, it is not a “quick fix” thatcan be implemented through one manual or trainingcourse. It requires relatively more of frontline supervi-sors than traditional supervision. EngenderHealth hasarticulated a three-stage process for introducingsupportive supervision in a developing-countryinstitution, taking approximately two years. Cooperat-ing agencies implementing elements of supportivesupervision uniformly underscored the need forinvestment in:

■ developing in supervisors the skills needed forsupportive supervision (such as facilitation, inter-personal, problem-solving, and analytical skills);

■ orienting staff to its tools and methods (peerreview, performance assessment tools);

■ upgrading the technical or clinical skills of thoseassigned to provide external supervision.

Teaching these skills poses a challenge to health careprograms, which generally have scarce resources andmust meet numerous critical learning needs.

Top management must be committed to sup-portive supervision. Support from senior officials isvital to introduce and sustain supportive supervisionand quality improvement processes in general. Seniormanagers must become involved in and visibly supportnew initiatives to improve supervision. Investing insupportive supervision should be linked to account-ability for results, and rewards or recognition formaking supervision more supportive. Moreover, thistop management support must be nurtured, in light ofthe high potential for turnover at this level, especiallyin the public sector. This lesson also suggests that

where top management does not fully embrace sup-portive supervision, it is not appropriate to try toinstigate the approach from the bottom up.

Supportive supervision should be integratedinto the existing human resource managementsystem, rather than introduced as an isolatedintervention or parallel system. The key to im-proving sustainability is to build capacity for im-proved supervision into existing systems andprocesses, rather than imposing entirely new systemsfrom the outside, since changes and improvements that“work around” current systems and processes aregenerally less sustainable.46

Build on success. One way to improve the sustain-ability of change is to focus on results and build onsuccess. In practice, this strategy means identifyingsome desirable and achievable results and then imple-menting the most promising approaches for attainingthose results. Each success builds support and momen-tum for further change. Therefore, when supportivesupervision is introduced, efforts must be made tohelp external supervisors build confidence in the newway of doing things and to motivate staff to invest inpeer support and self-assessment. Several respondentsto the survey commissioned for this paper noted thatan important factor in sustaining quality improvementprocesses was for staff to achieve an early “win”––inother words, to motivate teams to continue withquality improvement activities, it was important forthese teams to achieve tangible results early in theprocess. In the absence of quick results, teams oftenget discouraged and abandon efforts.

Some decision-making authority must bedecentralized for supportive supervision to work.Frontline supervisors and health care providers mustbe empowered to make some decisions to solve prob-lems and implement changes in the management anddesign of services. If such authority is not delegated,efforts to institute supportive supervision will gener-ate frustration and ultimately fail.

25


Long-term sustainability depends on public-sector institutions embracing supportive supervi-sion. For supportive supervision to take hold in acountry, it requires the legitimacy that acceptance bythe Ministry of Health and major donors affords. Mostof the experiences described in the literature onimplementing aspects of supportive supervision havetaken place in private reproductive health organiza-tions. Ongoing experiences in Brazil (PROQUALI),Kenya (the Family Planning Association of Kenya’scollaboration with the Ministry of Health), and Uganda(the Quality Assurance Department of the Ministry ofHealth) need to be followed and analyzed to under-stand how to successfully institutionalize supportivesupervision in public-sector health systems.

RECOMMENDATIONS FOR MAQ

This review of the promise and practice of support-ive supervision leads to several recommendations forconsideration by MAQ.

Document experiences and best practices.Improved tools to guide supportive supervision andself-assessment are needed. For example, many check-lists are being created in different countries, sometaking considerable resources to develop (the Ugandanational supervision guidelines, for example). Greatersharing of such tools for supportive supervision and ofevaluation results from using them would help ad-vance the state of the art.47 Best practices need to beidentified from successful approaches to motivating,training, and coaching external supervisors to performsupportive supervision and to motivating and enablinghealth workers to effectively conduct self-assessment,peer assessment, and internal supervision. Identifyingbest practices should include defining the costs of thethree mechanisms of supportive supervision (external,

internal, and self/peer supervision) in different set-tings. Also needed is further study of supportivesupervision over time, to better understand whatfactors help to sustain it.

Study scale-up, including the startup andrecurrent costs of supportive supervision. There isvery little information about the costs and cost effec-tiveness of supportive supervision in developing-country health systems. More documentation isneeded of the startup and recurrent costs of thevarious mechanisms of supportive supervision indifferent country and organizational contexts.EngenderHealth has suggested that supportive super-vision requires considerable investment in staffdevelopment and capacity-building initially, but that,in the long run, programs may end up spending lesson external supervision and training after supportivesupervision has been established.48

Evaluate the impact of supportive supervisionon the performance of health services. The reportsof how various organizations are implementing aspectsof supportive supervision are mainly descriptive, and,with some exceptions, generally do not provideobjective evidence of supportive supervision’s impacton health worker performance, client satisfaction, orother outcomes. While it is difficult to isolate theeffects of an intervention in field settings, moreattempts to evaluate the impact of supportive supervi-sion would help to expand its adoption. Long-termfollow-up on the sustained effects of supportivesupervision in sites where it has been in place forseveral years would greatly enrich our understandingof institutionalization issues.

Advocate supportive supervision. Better documen-tation of the impact of supportive supervision and itscosts will help to provide a stronger case for advocatingfor the expansion of supportive supervision.

26


ACKNOWLEDGMENTS

The authors would like to acknowledge Lynn Bakamjian ofEngenderHealth, Thada Bornstein and Diana Silimperi ofUniversity Research Co., LLC, and Wayne Stinson of Manage-ment Sciences for Health for technical direction. We alsothank the reviewers: Maj-Britt Dohlie of PRIME II/EngenderHealth, Alison Ellis of Management Sciences forHealth, Young-Mi Kim and Adrienne Kols of the JohnsHopkins University Center for Communication Programs, PamLynam of JHPIEGO, Erin Mielke of EngenderHealth, and JimShelton of USAID.

NOTES

1. R. Simmons, “Supervision: The Management of FrontlinePerformance,” in R. J. Lapham and G. B. Simmons, eds.,Organizing for Effective Family Planning (Washington, DC:National Academy Press, 1987), pp. 233–40. W. Stinson,L. Malianga, L. Marquez, and C. Madubuike, “QualitySupervision,” QA Brief vol. 7, no. 1 (1998): 4–6 (Bethesda,MD: Quality Assurance Project).

2. Management Sciences for Health, “Supervising andSupporting Your Staff,” The Family Planning Manager’sHandbook (West Hartford, CT: Kumarian Press, 1991), pp.115–55. Also available on the Electronic Resource Centerat http://erc.msh.org/fpmh_english.

3. J. Heiby, “Quality Assurance and Supervision Systems”(editorial), QA Brief vol. 7, no. 1 (1998) (Bethesda, MD:Quality Assurance Project).

4. S. M. Kilminster and B. C. Jolly, “Effective Supervision inClinical Practice Settings: A Literature Review,” MedicalEducation vol. 34 (2000): 827–40.

5. B. Ben Salem and K. J. Beattie, “Facilitative Supervision:A Vital Link in Quality Reproductive Health ServiceDelivery,” AVSC Working Paper no. 10 (New York:EngenderHealth, 1996). J. Bradley, E. Mielke, G.Wambwa, J. Mashafi, and M. Nasania, “Family PlanningServices in Tanzania: Results from a Project to ImproveQuality, 1996–1999” (New York, EngenderHealth, 2000).M. Dohlie, E. Mielke, F. K. Mumba,G. E. Wambwa, A. Rukonge, and W. Mongo, “UsingPractical Quality Improvement Approaches and Tools inReproductive Health Services in East Africa,” Journal onQuality Improvement vol. 25, no. 11 (1999): 574–87. Y. M.Kim, P. Tavrow, L. Malianga, S. Simba, A. Phiri, and P.Gumbo, “The Quality of Supervisor-Provider Interactionsin Zimbabwe,” Operations Research Results vol. 1, no. 5(2000) (Bethesda, MD: Quality Assurance Project). Stinsonet al., “Quality Supervision.”

6. Stinson et al., “Quality Supervision.”

7. Kilminster and Jolly, “Effective Supervision in ClinicalPractice Settings.”

8. Population Council, “Evaluation of the SupportiveSupervision Intervention,” Update vol. 11 (1998): 1–8.

9. Center for Human Services, “Solving Operational Prob-lems in Primary Health Care, 1981–87: Final Report of thePRICOR Project” (Chevy Chase, MD: Primary Health CareOperations Research Project, 1987). Center for HumanServices, Examining the Link between Supervision andHealth Worker Performance: Malaria Case Management inSenegal (Bethesda, MD: The Primary Health Care Opera-tions Research Project, 1990).

10. B. P. Loevinsohn, E. T. Guerrero, and S. P. Gregorio,“Improving Primary Health Care through SystematicSupervision: A Controlled Field Trial,” Health Policy andPlanning vol. 10, no. 2 (1995):144–53.

11. J. R. Foreit and K. G. Foreit, “Quarterly versus MonthlySupervision of CBD Family Planning Programs: AnExperimental Study in Northeast Brazil,” Studies inFamily Planning vol. 15, no. 3 (1984): 112–20.

12. JHPIEGO, Executive Summary of “Defining a PerformanceImprovement Intervention for Kenya Reproductive HealthSupervisors: Results of a Performance Analysis,”JHPIEGO Technical Report JHP-07 (November 1999). Alsoavailable at http://www.jhpiego.org/pubs/TR/tr907sum.htm. Kim et al., “The Quality of Supervisor-Provider Interactions in Zimbabwe.”

13. H. D. Stolovitch and E. J. Keeps, eds., Handbook of HumanPerformance Technology: A Comprehensive Guide forAnalyzing and Solving Performance Problems in Organiza-tions (San Francisco: Jossey-Bass Publishers, 1992).

14. Simmons, “Supervision: The Management of FrontlinePerformance.”

15. Loevinsohn et al., “Improving Primary Health Carethrough Systematic Supervision.”

16. R. Vernon, A., Staunton, M. Garcia, J. J. Arroyo, and R.Rosenberg, “Test of Alternative Supervision Strategies forFamily Planning Services in Guatemala,” Studies inFamily Planning vol. 25, no. 4 (1994): 232–38.

17. L. M. Franco, D. R. Silimperi, T. Veldhuyzen van Zanten,C. MacAulay, K. Askov, and B. Bouchet, Sustaining QualityHealthcare: Institutionalization of Quality Assurance(Bethesda, MD: Quality Assurance Project, 2002).

18. W. Stinson, L. Bakamjian, S. C. Huber, and D. Silimperi,“Managing Programs to Maximize Access and Quality:Lessons Learned from the Field,” MAQ Paper vol. 1, no. 3(2000) (Washington, DC: Maximizing Access and Quality[MAQ] Initiative, USAID).

19. Heiby, “Quality Assurance and Supervision Systems.”

20. P. E. Atkinson, Creating Cultural Change: The Key toSuccessful Total Quality Management (Toronto: Pfeiffer &Company, 1990). J. A. Barker, Leadershift: Five Lessons forLeaders in the 21st Century (West Des Moines, IA: AmericanMedia, Inc., 1999).

27



22. D. K. McNeese-Smith, “The Influence of ManagerBehavior on Nurses’ Job Satisfaction, Productivity, andCommitment,” Journal of Nursing Administration vol. 27,no. 9 (1997): 47–55.

23. Ben Salem and Beattie, “Facilitative Supervision.”


25. Management Sciences for Health, “Improving Supervi-sion: A Team Approach,” The Manager vol. 2, no. 5(Boston: Management Sciences for Health, 1993). Alsoavailable on the Electronic Resource Center at http://erc.msh.org/mainpage.cfm?file=2.2.3.htm&module=hr&language=English.


27. A. J. Kols and J. E. Sherman, “Family Planning Programs:Improving Quality,” Population Reports vol. 26, no. 3(Baltimore, MD: Population Information Program, Centerfor Communication Programs, Johns Hopkins School ofPublic Health, Nov. 1998).


29. M. Dohlie et al., “Using Practical Quality ImprovementApproaches and Tools in Reproductive Health Services inEast Africa.”

30. P. Lynam, L. M. Rabinovitz, and M. Shobowale, “The Useof Self-Assessment in Improving the Quality of FamilyPlanning Clinic Operations: The Experience of COPE inAfrica,” AVSC Working Paper no. 2 (New York:EngenderHealth, 1992).

31. K. K. Kafle, Y. M. S. Pradhan, A. D. Shrestha, S. B. Karkee,P. L. Das, N. Shrestha, R. R. Prasad, “Better PrimaryHealth Care Delivery through Strengthening the ExistingSupervision/Monitoring” (poster exhibited at the FirstInternational Conference on Improving Use of Medicines,Chiang Mai, Thailand, April 1–4, 1997), http://dcc2.bumc.bu.edu/prdu/ICIUM_Posters/2e3_txtf.htm,cited 8/30/01.

32. P. S. Zeitz, C. G. Salami, G. Burnham, S. A. Goings, K.Tijani, and R. H. Morrow, “Quality Assurance Manage-ment Methods Applied to a Local-Level PrimaryHealthcare System in Rural Nigeria,” International Journalof Health Planning and Management vol. 8, no. 3 (1993):235–44.

33. Y. M. Kim, M. E. Figueroa, A. Martin, R. Silva, S. F.Acosta, M. Hurtado, P. Richardson, and A. Kols, “Partici-patory Supervision with Provider Self-AssessmentImproves Doctor-Patient Communication in RuralMexico,” Operations Research Results vol. 2, no. 12 (2002)(Bethesda, MD: Quality Assurance Project).

34. E. Kelley, C. Geslin, S. Djibrina, and M. Boucar, “TheImpact of QA Methods on Compliance with the IntegratedManagement of Childhood Illness Algorithm in Niger,”

Operations Research Report vol. 1, no. 2 (2000) (Bethesda,MD: Quality Assurance Project).

35. S. Bose, E. Oliveras, and W. N. Edson, “How Can Self-Assessment Improve the Quality of Healthcare?,” QA IssuePaper vol. 2, no. 4 (2001) (Bethesda, MD: Quality Assur-ance Project).

36. Y. M. Kim, F. Putjuk, E. Basuki, and A. Kols, “ImprovingProvider-Client Communication: Reinforcing IPC/CTraining in Indonesia with Self-Assessment and PeerReview,” Operations Research Results vol. 1, no. 6 (2000)(Bethesda, MD: Quality Assurance Project).

37. This finding contrasts with the observation by Kilminsterand Jolly in “Effective Supervision in Clinical PracticeSettings” and others that external supervision has moreeffect on less experienced health providers.

38. Kim et al., “Participatory Supervision with Provider Self-Assessment in Mexico.”


40. J. Stanback, S. J. Griffey Brechin, P. Lynam, C. Toroitich-Ruto, T. Smith, and the Kenya Guidelines Update Evalua-tion Study Group, “The Effectiveness of NationalDissemination of Updated Reproductive Health/FamilyPlanning Guidelines in Kenya,” final report (ResearchTriangle Park, NC: Family Health International, 2001).

41. Kelley et al., “The Impact of Self-Assessment with PeerFeedback on Health Provider Performance in Mali.”


43. M. Dohlie et al., “Kenya: Survey on Peer Support amongPrivate Nurse-Midwives,” PRIME II Technical Report(Chapel Hill, NC: PRIME II/INTRAH, forthcoming).

44. Lynam et al., “The Use of Self-Assessment in Improvingthe Quality of Family Planning Clinic Operations: TheExperience of COPE in Africa.”

45. M. Dohlie, E. Mielke, T. Bwire, D. Adriance, and F.Mumba, “COPE, a Model for Building CommunityPartnerships That Improve Care in East Africa,” Journalfor Healthcare Quality vol. 22, no. 5 (2000): 34–39.

46. V. Doyle, Enabling Factors for Ensuring Long-Term Sustain-ability of QA Systems (London: Health Sector ReformResearch Work Programme, n.d.).

47. For example, JHPIEGO has made available on CD-ROM thesupervision learning resource package it developed withthe Ministry of Health of Kenya. The CD-ROM, “Supervis-ing Health Services: Improving the Performance ofPeople,” by Nancy Caiola, Kama Garrison, Rick Sullivan,and Pamela Lynam, may be obtained from JHPIEGOMaterials Management (Tel.: 410-538-1825; Fax: 410-537-1472; e-mail: [email protected]).

48. Dohlie et al., “Using Practical Quality ImprovementApproaches and Tools in Reproductive Health Services inEast Africa.”

Copies of this MAQ paper may

be obtained by contacting the

MAQ fellow at:

Office of Population and

Reproductive Health/Service

Delivery Improvement Division

USAID

Washington, DC 20523-3601

Phone: 202-712-4548

Fax: 202-216-3046

MAQ PAPERS

MAQ Papers are published by the Maximizing Ac-

cess and Quality (MAQ) Initiative—an initiative of

the United States Agency for International Develop-

ment (USAID), cooperating agencies, country part-

ners, and other collaborators to apply state-of-the-

art methods to maximize access to and the quality

of family planning and other selected reproductive

health services.

The participating organizations are:

■ Abt Associates, Inc.■ CARE■ Centre for Population and Development Activities■ Deloitte Touche Tohmatsu■ EngenderHealth■ Family Health International■ Futures Group■ Innovative Technologies for Health Care Delivery

■ Institute for Reproductive Health,

Georgetown University■ JHPIEGO■ John Snow, Inc.■ Johns Hopkins Bloomberg School of Public

Health, Center for Communication Programs■ Management Sciences for Health

■ ORC Macro■ Pathfinder International■ Plan International■ Population Council■ Population Services International■ Program for Appropriate Technology in Health■ Public Health Institute■ Save the Children USA■ University Research Co., LLC, Center for

Human Services

Documents

Making Supervision Supportive and Sustainable: New Approaches