In 2007, after 14 years of civil war, Liberia began the process of rebuilding its health delivery system through a decentralized health care sector. As part of the rebuilding, the MoHSW developed a Basic Package of Health Services (BPHS) guaranteed to all Liberians accessing services through the public sector. Through the BPHS, health facility standards were refined and introduced at each level of care, catapulting Liberia into an environment of health facility reform. In order to meet these standards, the MoHSW focused on enhancing the management capacity of health centers, clinics, and hospitals throughout the country. In addition to these higher standards, County Health Teams (CHTs) - consisting of district health workers often with limited background or experience in management - became responsible for decision-making and priority-setting at local levels.
During this time of rebuilding and decentralization, Liberia was and is transitioning from a period of relief to one of development. With this transition have come changes in partner and donor support and an adjustment from crisis management to strategic thinking and systems development. Together, these contextual issues magnify the need for strong management skills at the CHT and health care organizational levels.
Within this setting of decentralization and health sector reform, a classroom-based Health Systems Management Course for health facility and CHT managers was developed and taught by Yale University, Mother Patern College, and CHAI. Follow-up and mentoring for course participants was provided by Mother Patern faculty, on-site Yale-Clinton Foundation Fellows, and CHAI staff who assisted participants in managing projects and in reinforcing course concepts. The Health Systems Management Course is a competency-based training course focused on the core skills of health care management, including scientific problem solving and strategic thinking, human resource management, financial management, and leadership development. The content of the course was designed to teach participants how to think strategically in several key domains and then how to use the more conceptual tools to solve concrete problems, including the implementation of the Basic Package of Health Services and related policies. A core component of the training was to link classroom didactics to field-based applications, where teams were expected to use the management skills developed and apply them to specific expectations for MoHSW policy implementation. For example, problem statements were generated around expected deliverables outlined in the six focus areas of the Basic Package of Health Services (e.g. maternal and infant health, child health, reproductive and adolescent health, communicative disease control, mental health and emergency care) such as establishing regular supportive supervision in primary health care units, ensuring effective management of malaria, and establishing an effective county referral system.
Course participants included two to four managerial representatives from CHTs, the MoHSW, government hospitals, and Liberians working with international non-governmental organizations (INGOs) in the oversight of multiple health facilities. Participants were nominated to apply and selected based on successful completion of a course application, which included holding a supervisory or management position within a CHT, government institution, INGO or faith based organization; answering four short questions in a clear manner; and coming recommended by the head of one's organization or CHT.
Consistent with the goal of building capacity at Mother Patern College and transitioning the course from Yale to Liberian faculty, the course was conducted in a stepped approach with increasing responsibility transferred to Mother Patern instructors with each training cycle. Each cycle comprised three classroom periods lasting approximately 10 days each, which took place over a period of five months. In the first cycle, Yale University instructors developed and taught 100% of the material. In the second cycle, after substantial train-the-trainer sessions, instructors from Mother Patern College conducted 50% of the training with Yale facilitators conducting the remaining 50% in addition to providing supervision and feedback. In the third cycle, Mother Patern College took on 100% of course instruction, with Yale providing only back-up support. By the fourth cycle, Mother Patern instructors will teach and manage 100% of the material without Yale or CHAI support. The full transfer (the first three cycles or cohorts where Yale and CHAI were present) required approximately 2 years to complete.
In addition to classroom training, the course required substantial field-based work with supervision. Between the 10-day classroom periods, participants returned to their counties and/or health facilities to apply course skills and tools in their work settings, and to complete field-based project assignments including the use of root cause analysis balanced scorecards and comparative analysis in order to support efficient problem solving approaches. During this time, participants received on-the-ground mentoring in management skills and quality improvement approaches, as they applied classroom learning to their real-world work situations. Mentoring was provided by Mother Patern faculty, CHAI, the MoHSW, and five Yale-CHAI Fellows placed in two Monrovia hospitals and on a rotational basis in CHTs throughout Liberia.
Since the completion of the program, staff originally trained through the train-the-trainer mechanism have identified and trained additional individuals. The original cohorts of trainees are now conducting trainings themselves targeting middle-level staff in counties including clinical officers-in-charge, supervisors and coordinators. The original management course material has been used to prepare these subsequent trainings. In the 6 months following the final cohort reported in this paper, graduates of the original program have trained 27 additional staff from two counties, with plans to expand to three more counties in the next 6 months. In addition, the Liberian academic partner, Mother Patern College, is currently working from the original materials to mentor the MoHSW in developing its own management and leadership program. Through interviews and an on-the-ground assessment, Mother Patern College management staff members are conducting a program evaluation in order to measure the course's impact and identify areas for improvement, as needs and content continually evolve in the Liberian context. These efforts reflect the transfer of the program beyond the first generation of trainees and underscore the degree to which both country-level and Ministry-level staff continue to integrate the material into their positions.
Data collection and measures
We conducted a self-administered survey of all participants completing the course at the end of each session to measure change in key indicators as the transfer process occurred (cohort 1 was taught 100% by Yale, cohort 2 was taught 50% by Yale and 50% by Mother Patern College, and cohort 3 was taught 100% by Mother Patern College). Indicators included self-rated management skills (before and after the course), course evaluations, and faculty members' responsiveness to participant needs and ability to teach and manage the training. Participants rated each indicator on a scale, with possible responses tailored to the survey item. For respondents' self-rated management skills, response categories were: very strong and confident, strong, moderate, weak, and very weak. For course evaluations, the response categories were: extremely well, well, a little, not at all, no opinion. For faculty members' ability to teach and manage the course effectively, response categories were: yes, definitely; yes, somewhat; no; not at all; and no opinion. Evaluations were designed by the research team and took approximately 30 minutes to complete (see Additional File 1
We utilized data from surveys administered in the final session for each cohort. We used standard descriptive statistics to characterize item responses, stratified by cohort. We evaluated differences between cohorts for the items of interest using Fisher's exact tests. We calculated 95% confidence intervals around proportions using Wilson's method, including a continuity correction [9
]. Analyses were conducted using SAS version 9.1 (SAS Institute, SAS. 2003: Cary, NC.).