Testing new methods to improve providers' performance of IMCI in rural Kenya

The Integrated Management of Childhood Illness (IMCI) is a strategy developed by UNICEF and WHO to improve the diagnosis and treatment of children in primary health facilities in developing countries. Since 1995, IMCI has been introduced into 82 countries. Initially, the algorithm lacked an HIV/AIDS component. Because of growing recognition that earlier diagnosis of HIV positive children can result in more effective treatment of opportunistic infections and a timelier start of anti-retroviral treatment, WHO developed an HIV/AIDS assessment for IMCI in 2001. More recently, WHO encouraged countries also to add a “care for development” component to assess caretakers' communication and play with the child. The challenge for developing countries is to introduce these new components to health providers without incurring costly re-training. Another concern is that adding elements to IMCI can make it more time-consuming and erode care quality.

The purpose of this study was to test whether low-cost job aids for providers could be developed which would introduce the new components, be easy to disseminate and use, maintain or enhance the quality of IMCI performance, and be liked by providers. The team developed two job aids: (1) a child outpatient booklet, based on a Zambian model that used a standard exercise book, which gave providers step-by-step guidance to the revised algorithm; and (2) a laminated tablemat, which could be placed on the providers' desk and contained all IMCI-recommended treatments. The job aids were introduced to 14 providers in 6 randomly-selected primary care facilities in Vihiga District, Kenya, in March 2007. Six other facilities served as controls. Providers were observed caring for six sick children before and 2-3 months after the job aid introduction, and also queried on their usefulness.

Results of the study indicate that the job aids alone, without refresher training, led 58% of the providers to perform the HIV assessment, and 45% to perform the care for development assessment. For the other components of IMCI, the job aids appeared to significantly improve providers' assessment and treatment, and somewhat improved their classification. Most providers reported that the job aids were time-saving and user-friendly. Implications of this study for IMCI performance elsewhere will be discussed.