This evaluation research project was conducted to describe local perspectives on creating and implementing mandated memoranda of understanding (MOU) between Medicaid managed care organizations and local health departments (LHD), to provide insights into the strengths and limitations of MOU, and to share information on their use as tools for interorganizational service coordination.
A cross-sectional, qualitative and quantitative, self-administered, mail-back survey was conducted with employees of MCOs and LHDs in California having experience in creating and/or implementing state-mandated MOU in nine service areas. Descriptive and qualitative results are reported.
The creation of MOU was facilitated by supportive leadership, previous collaborative experience, and the state mandate. The majority of respondents believed the MOU worked well to achieve coordination of services. MOU strengths and limitations were identified. The MOU were reported to have contributed to an increase in participation in four of six types of collaborative activities. Perceived quality of health care services for Medicaid participants improved with the MOU, but no impact on quality of public health services was reported. A majority (78 percent) supported a continuation of the mandated policy. Some organizations reported expanding the use of MOU beyond mandated areas.
Study respondents held generally positive perceptions of the strengths of MOU and supported continuation of the mandated MOU policy in California. Their experience demonstrates a capacity for MCOs and LHDs to work together on health care coordination issues despite the difficulties inherent in interorganizational collaboration.
Memoranda of understanding (MOU) are formal agreements created by managed care organizations to facilitate communication, coordination, and collaboration with other organizations. Increasingly, MCOs, especially those serving Medicaid populations, are being encouraged or mandated to create MOU with public health departments. A study in 1997 revealed that approximately half of all states included provisions addressing relationships with state or local public health agencies in their Medicaid managed care (MCMC) contracts or requests for proposals.1 A 1998 survey of Medicaid programs found that nine states mandated formal agreements between MCOs serving Medicaid populations and local health departments (LHDs) (California, Illinois, Kansas, Kentucky, Maryland, New York, Vermont, Texas, and Indiana), and two additional states required agreements with state health departments (Delaware, New Mexico).2 The survey also revealed that MOU between MCMC plans and LHDs were encouraged and actually existed in 22 additional states.2
Collaboration between organizations is often difficult and time consuming.3 As more MCOs embark on the challenge of creating collaborative agreements with LHDs and other organizations, it is instructive to study the experiences of others who have successfully created and implemented MOU. The study reported here was conducted three years after the initiation of a policy to mandate agreements between MCOs and LHDs in California, at a time when the majority of the required MOU were in place. This survey of MCO and LHD employees directly involved in creating and/or implementing MOU provides descriptive and qualitative information about the process of creating MOU, participant insights on MOU strengths and limitations, and information on the use of MOU as tools for interorganizational service coordination.
Setting. In 1993, California redesigned its Medicaid program and greatly expanded the number of participants enrolled in managed care plans.4 In 12 counties, the Medicaid population, eligible due to their participation in the federal Temporary Assistance to Needy Families program, was offered a choice of enrollment in one of two health plans: a locally developed "Local Initiative" plan and a commercial managed care plan.4 The state also required the MCOs that were contracting to provide these services to have MOU in 10 service areas with the LHDs in their counties.5 The 10 areas included sexually transmitted disease, human immunodeficiency virus, tuberculosis, immunization, WIC (Supplemental Nutrition Program for Women, Infants, and Children), family planning, maternal and child health, California Children's Services (CCS — California's program for children with special health care needs), Child Health and Disability Prevention (CHDP — California's Early Periodic Screening, Diagnosis and Treatment program), and population-based services. Eleven health plans (eight local initiative plans and three commercial plans) and 15 LHDs (12 county and 3 city departments) were covered under the policy. At the time of this study, both health plans were commercial in 2 of the 12 targeted counties, and only 1 (commercial) health plan was available in one county.
Detailed guidance on the expected content of the MOU was provided in a series of policy letters from the state Medicaid program, issued between 1994 and 1997. In general, the MOU addressed interorganizational communication processes, outreach and education activities, case-management services, data collection and sharing mechanisms, and special-projects participation. The requirement for an MOU in population-based services was not being enforced at the time of this study, due to the lack of policy guidance related to MOU content in this area.
A cross-sectional descriptive survey of employees of the health plans and LHDs involved in creation and/or implementation of the MOU was conducted as part of a larger study.6 Participants were recruited in a two-stage process. First, names of potential participants were gleaned from archival review of state records related to the MOU. Second, directors of the LHDs and health plans were sent a letter requesting that they name additional individuals in their organizations who were familiar with the MOU. A survey instrument was designed and pilot tested. Each survey queried for responses about one set of LHD/health plan MOU. Individual participants received multiple surveys when they were expected to be familiar with multiple LHD/health plan MOU. The questionnaire included 110 closed-ended and 27 open-ended questions. The survey(s) were sent via first-class mail, self-administered by hand, and returned postage-paid to the researcher. Participation was encouraged with postcard and telephone reminders as well as a nominal incentive program. Data collection took place between October 1999 and February 2000.
Descriptive statistics for the closed-ended responses are presented here. Only one questionnaire per individual respondent (n=123) was used for the closed-ended responses to avoid statistical data dependence. When multiple surveys were returned by an individual, the choice of which survey to include was made randomly. Qualitative methods for open-ended responses included verbatim recording of responses, identification of themes among the responses to each question, and consolidation into the major themes for each question. Data from all returned surveys (n=193) were included in the qualitative analysis to capture as much of the qualitative experiences of these respondents as possible.
Survey response rates were greater than 70 percent in all categories: 78 percent of unduplicated individuals (n=123) returned 72 percent (n=193) of all surveys. The response rate from LHD respondents was 77 percent (n=80), and 80 percent (n=43) from health plan respondents. The response rates from commercial plans and from local initiative plans were 77 percent (n=17) and 81 percent (n=26) respectively. All organizations (n=26) were represented among the respondents with a range of 1 to 12 respondents per organization (mean = 4.7).
MOU creation. The survey participants were asked about factors that promoted or facilitated the creation of the MOU. As shown in Table 1, high percentages of LHD and health plan respondents agreed that supportive leadership, previous experience with collaboration, the state mandate for MOU development, and sharing information between organizations were helpful or very helpful in creating the MOU.
|TABLE 1 Factors that facilitated MOU creation: percent helpful/|
|Type of respondent|
|Factor||n||%||LHD %||HP %|
|Support of LHD leadership||100||94||92||97|
|Previous experience with collaboration||101||94||97||89|
|Support of HP leadership||98||93||90||97|
|Training about managed care||85||88||86||93|
|Previous experience with this organization||57||86||81||92|
|Sharing information with other HP or LHD||84||82||83||81|
|Direction provided by policy letters||90||80||78||83|
|State-mandated MOU policy||103||79||84||69|
|Technical assistance by state PH||83||64||71||48|
|Technical assistance by MMCD||81||48||55||37|
|Note: Ratings were made on a four point Likert-type scale (not at all helpful, not helpful, helpful, very helpful, don't know). Missing observations in this table were "don't know" responses.|
Abbreviations: LHD=Local health department; HP= Health plan; PH= State public health programs; MMCD= State Medicaid program
Technical assistance from the state public health and Medicaid programs was seen as helpful by a smaller percentage of respondents. When respondents were asked to indicate the most important factor in promoting the ability of their organization to create the MOU, the most frequent response themes were as follows:
Participants were asked to share lessons they had learned in the process of creating the MOU. The most frequent themes that emerged from the answers to this question were:
MOU strengths. Majorities of respondents reported that the MOU were working well or extremely well in achieving coordination between their organizations in the nine service areas covered by the MOU (see Table 2).
|TABLE 2 Achievement of coordination of MOU by type, respondent type, and health plan type|
|Type of respondent||Type of health plan|
|MOU type||n||All (%)||LHD (%)||HP (%)||Commercial (%)||Local initiative (%)|
|Note: Ratings were made on a four point Likert-type scale (not working at all, not well, well, extremely well, and don't know). Missing observations were "don't know" responses. Overall n=123.|
The percentage of respondents who thought the MOU worked well or extremely well was higher for health plan respondents compared to LHD respondents in all categories.
The survey asked respondents to indicate the extent to which they agreed with a series of statements about hypothetical strengths of the MOU. A majority of participants (61 percent to 71 percent) agreed or strongly agreed that the MOU: 1) made contacting the right people at the organization easier; 2) resulted in meaningful collaboration between the organizations; and 3) provided something to fall back on when problems arose in working with the partner organization (see Table 3). Improvements in public health services in the county, improvements in health care services to all enrollees of the health plan, and improved management of reimbursement mechanisms were seen as strengths by a minority of respondents, although health plan respondents ranked the statements about the MOU leading to improved health care services and public health services more highly than did LHD respondents. There were no differences in responses by health plan type (see Table 3).
|TABLE 3 Strengths and limitations of MOU by respondent type (percent agree/strongly agree)|
|Respondent type||HP type|
|n||All (%)||LHD (%)||HP (%)||CP (%)||LI (%)|
|Contact identification easier||118||71||65||83||77||67|
|Led to meaningful collaboration||117||61||57||68||60||61|
|Improved services for all in HP||113||34||26||49||30||36|
|Back up for problem solving||117||71||80||66||75||68|
|Reimbursement works better||113||23||20||29||17||28|
|Led to improved public health||117||33||19||45||31||26|
|Do not address important PH issues||117||23||32||7||23||23|
|Take too much time||117||27||26||29||27||28|
|Describe only what we do already||118||35||34||22||31||29|
|Are not being followed||116||19||21||8||13||19|
|Note: Ratings were made on a five point Likert-type scale anchored by "strongly disagree" and "strongly agree." Missing observations were missing data.|
In an open-ended question, the respondents were asked to list additional strengths of the MOU. The two most common themes arising from these responses were as follows:
Survey respondents were asked to indicate what impact the MOU had on participation of the LHD or health plan in six types of collaborative activities (see Table 4). A majority of respondents reported the participation of their organizations had increased for health promotion coalitions (74 percent), provider or staff training (72 percent), community health promotion or education interventions (67 percent), and development of health promotion guidelines or protocols (58 percent). A minority believed that the MOU had contributed to increases in participation in data collection (44 percent) and community health assessment (40 percent) activities.
|TABLE 4 Perceived MOU impact on collaborative activities and health care/public health service quality|
|n||Increase (%)||Decrease (%)||No impact (%)|
|Health promotion coalitions||85||74||0||26|
|Community health education||82||67||0||33|
|Development of guidelines||85||58||0||42|
|Community health assessment||70||40||1||59|
|Quality of services|
|Health care services||78||56||3||40|
|Public health services||73||27||3||70|
Note: Missing observations include "not applicable" (no MOU in place), "no participation in collaboration," "don't know," and missing responses.
Survey participants were asked for their perceptions as to whether the MOUs contributed to a change in the quality of health care services for Medicaid participants (see Table 4). An increase in the quality of health care services was reported by 56 percent of respondents. Examples of ways in which the MOU have contributed to improved quality included: improved access to services (n=19); improved case management (n=16); and improved collaboration (n=15). When participants were asked whether the MOU had contributed to a change in quality of public health services in their county, 27 percent of respondents reported an increase in quality. When asked for examples of how the MOU have contributed to a change in quality, the most frequent themes in the responses were increased collaboration (n=14) and improved data collection (n=6).
MOU limitations. A minority of respondents supported hypothetical MOU limitations statements (see Table 3). There were no differences in responses by type of respondent for most statements; the exception was "Don't address important public health issues," which was supported by a statistically significantly higher percentage (though still a minority) of LHD respondents compared to health plan respondents. When asked an open-ended question about additional limitations of MOU, the most common themes in the responses were as follows:
Use of MOU. A majority (78 percent) of respondents supported continuation of the state's policy of mandated MOU in the nine service areas. A larger percentage of LHD respondents supported continuation (84 percent) compared to health plan respondents (67 percent). A slight majority of respondents (59 percent) recommended the use of MOU between LHDs and MCOs for non-Medicaid populations. A minority (36 percent) believed that the state should promote MOU in additional service areas between public health and Medicaid managed care.
The respondents were asked to report on how their organizations were using MOU as a tool in management or collaboration outside the mandated service areas. The creation of additional MOU between the MCO and the LHD in areas not mandated by the state was reported by 21 percent of respondents. Examples of these MOU included refugee health services, childhood lead poisoning, health education and promotion, nutrition, mental health, and emergency-room use. Expansion of the use of MOU as a way of increasing collaboration with other health organizations was reported by 37 percent of respondents.
The results of this survey revealed positive perceptions among the respondents about the MOU and the mandated MOU policy. A substantial majority of respondents supported continuation of the current policy, though support was stronger among LHD respondents compared to health plan respondents. The reasons for the difference in support by type of respondent may relate to concerns about coordination, use of public health services, and loss of revenue to public health departments raised by the transition of Medicaid from a fee-for-service model to managed care.7 California's MOU policy was in part designed to address such issues. The mandate thus forced health plans to consider LHD concerns as well as roles with Medicaid populations, whereas they might not have done so without the mandate.
The MOU were perceived to have had a positive impact on the coordination of services to Medicaid participants. In addition, the MOU were perceived to have increased participation in some types of collaborative activities and to have improved the quality of health care services to Medicaid participants. These findings support previous case studies that revealed MOU to be good tools for initiating collaboration between managed care and public health agencies.8, 9 The state of California recognized the importance of coordination between these entities in its strategic plan to expand Medicaid managed care, and implemented the mandate to assure that coordination occurred.4 The policy of mandated MOU seems to have contributed to coordination of health care services for Medicaid participants. The survey indicated some organizations are expanding their use of MOU in areas not addressed by the mandate and with other health organizations. These findings support the finding by Blake and colleagues that MOU are flexible instruments that can be used more broadly than just with Medicaid managed care.8
The results of the survey provide evidence about the limitations of the MOU. These agreements were not perceived to have improved public health services in the affected counties to the same extent as they improved health care services to participants. In addition, the MOU were not viewed as having a positive impact on participation in collaborative activities that are traditionally associated with public health (health and risk-status data collection and community health assessments). These findings add support to previous literature that reports MOU between MCOs and LHDs to be focused on health care service delivery rather than population-based public health activities.810
The creation of MOU between LHDs and MCOs in California was not easy. These respondents noted the importance of organizational leadership and commitment in the successful creation of their agreements. These factors have been reported repeatedly in the collaboration literature to be important to the initiation and continuation of collaborative efforts.1113 In addition, previous experience working together was a factor that helped facilitate the MOU creation process. This finding is supported by literature on inter-organizational collaboration that has shown that prior ties between organizations help collaboration develop more quickly and efficiently.12, 14 The state mandate to create the MOU was seen as a contributor to the organization's ability to create the agreements. Mandates have been reported in the literature to be powerful promoters of collaboration between organizations.1519
It was interesting to note that the technical assistance provided by the state agencies to the MCOs and LHD was not strongly supported as a facilitative factor in the creation of the organization's agreements. This may be due to the inexperience of the agencies in dealing with issues of collaboration between managed care and public health. The emphasis on the use of MOU for this purpose is relatively recent, and the actual use of mandates for MOU between these entities remains relatively uncommon. More research about collaboration between managed care and public health organizations as well as broader dissemination of information about the experience of organizations in creating and implementing these types of agreements is needed.
Strong leadership, careful planning, involvement of key personnel, and commitment of resources are necessary to make collaboration between public health and managed care a reality.6 Since previous experience with collaboration helps organizations collaborate better, these early experiences in collaboration between MCOs and LHDs should lead to better efforts in the future. Nevertheless, the impact of any collaborative agreement is likely to be limited by its content and focus. MOU structured around health care coordination for Medicaid participants are unlikely to have heightened effects on public health unless activities with a broader population orientation are incorporated into the agreements. If increases in specific types of collaborative activities or participation in population-based activities are desired, then reference to those activities should be included in the MOU or other types of collaborative agreements. Additionally, states should provide clear policy guidance to MCOs and LHDs regarding the content of these types of agreements while allowing flexibility for addressing unique local needs. The experience of the organizations studied demonstrates a capacity for managed care and public health to work together on important health care coordination issues, despite the difficulties inherent in interorganizational collaboration.
The author would like to express appreciation to the University of California, Berkeley, Annette Blann Fellowship and Dissertation Research Grants, and the California Department of Health Services, Office of Clinical Preventive Medicine, for their financial support of this project.
Susan J. Zahner, DrPH, RN
University of WisconsinMadison
H6/252 Clinical Science Center
600 Highland Avenue
Madison, WI 53792-2455