M.Sc.(Med.), Faculty of Health Sciences, University of the Witwatersrand, 2009 / Introduction: Despite the steady growth of the elderly population in developing countries, this group, remains neglected in health related policies in developing countries, largely due to lack of empirical data on the health problems of elderly. There is need for research and development of convenient and cost effective ways of generating information on the health status of the elderly. Self reports of health are becoming common in health surveys of elderly throughout the world. Despite the considerable use of self reports in developed countries, in developing countries such research is only beginning. Therefore there is need for systematic documentation of factors affecting self reported health status in developing country settings for effective usage of self reports in surveys. Material and methods: The Adult Health and Aging Survey undertaken by Navrongo Health Research Centre, Ghana, as part of WHO SAGE (Study on Global Aging) aimed at generating longitudinal data on health and wellbeing of the elderly in Kassena-Nankana district of Ghana. This survey provides an opportunity to assess consistency between various dimensions of self reported health by comparing measures in an effort to establish the validity of information obtained by self reports. Analysis: Statistical analysis of self reported overall health (SRH), experiences of difficulty encountered in work and day to day activities (Overall Difficulty) and component experiences of health over various domains was carried out using ordered logistic regression and kappa analysis in order to understand what type of relationship exists between different types of measures of health. Overall self reported status of health (SRH) was the main outcome variable and three sets of variables were used as explanatory variables. The first set of variables captured functionality, the second captured psychosocial aspects of health, while the third involved demographic characteristics as possible confounders. Results: An analysis involving 4483 elderly individuals showed that functionality was associated with overall self reported health status in both summary and component forms. Addition of psychosocial domains to the model improved the model when summary functionality was used. However, addition of possible confounders did not improve the model. Conclusion and recommendations: The findings indicate that sex, marital status and ethnic background are important factors to be taken into account while interpreting the responses of self reported health in the Kassena-Nankana district of Ghana. For the current analysis both outcome and explanatory variables were self reported. The findings of the study would get validated with further research into associations between self reported measures and performance based measures and qualitative inquiries on meanings of overall and component health experiences in the same population.

ABSTRACT INTRODUCTION : Malaria remains a serious health burden among children in sub- Saharan Africa. The Home Management of Malaria (HMM) programme was adopted by African heads of states in 2000 as a strategy to achieve high coverage of prompt and effective anti malarial treatment within 24 hours of the onset of symptoms by home caregivers in areas with poor access to facility based health care. Strategic components of the programme include communication for behavioural change, training of community based public-private health service providers and making antimalarials available in communities[1]. AIM: To determine the impact of HMM strategy in the home treatment of uncomplicated malaria in children in Kassena-Nankana District (KND) of upper-east region of Ghana. Specific objectives described the knowledge, attitude and practice and tested the association between knowledge, attitude and other factors and accurate HMM. METHOD: Secondary data from a survey on the role of health information recipients in access and utilization of treatment for malaria management in children under 5 years (U5s) conducted among 818 women in KND from 2005 to 2006 were analyzed using a cross sectional study design. A total of 708 Home caregivers (HCGs) aged 15-49 years who responded to knowledge of the treatment of uncomplicated malaria was obtained after data cleaning. Knowledge of the treatment of uncomplicated malaria was used as a proxy for accurate HMM (correct dosage and correct duration of antimalarial) in U5s because the questionnaire did not contain information on the actual treatment given by the HCGs. Data analysis was done in STATA 10 using Chi squared test for categorical variables. Logistic regression models were used to quantify the associations and adjust for potential confounders and effect modification. RESULTS: The study found that 59% of the women had good knowledge of the symptoms of uncomplicated malaria and 25% knew that only mosquitoes transmit malaria. On treatment seeking attitude (advice and autonomy), the majority (91%) of the home caregivers received various forms of advice from the older women. Such advice included: using herbs (77%), buying drugs (41%), visiting Health Clinic (24%), and visiting the Community Health Officers (19%). On receiving advice, only 15% would utilize the services of the Community Health Officers (CHOs) who are the main source of treatment information for the communities. Thirty percent (30%) of the HCGs had autonomy of health care decision-making in the households. Accurate HMM in children was 28%. Knowledge of malaria and treatment seeking attitude were not significantly associated with accurate HMM (p>0.05). In the multivariate model, the HCGs were more likely to do accurate HMM in children if they had secondary education (OR = 2.54, 95% CI 1.18 ; 5.60), were of Nankani ethnicity (OR = 3.00, 95% CI 2.08 ; 4.35) and belonged to the very poor socio-economic status (OR = 2.31, 95% CI 1.25 ; 4.30). A Chi squared analysis to further identify the differences between the women who gave drugs and those who did not showed that the women differed significantly in their ethnicity (p<0.001), occupation (p<0.001) and relationship as the biological mothers to the children (p=0.008). The major limitation of this study was that knowledge of the treatment of uncomplicated malaria was used as a proxy for accurate HMM hence the finding is not a true reflection of the actual malaria treatment practice HCGs give to U5s. Another limitation is that the study could not measure the promptness of initiating malaria treatment within 24 hours of the onset of symptoms in children because of the absence of such variable in the data. CONCLUSION: Although HCGs had good knowledge of the symptoms of uncomplicated malaria, it did not translate to accurate HMM. The study identified poor dosage of treatment with Chloroquine (the first line antimalarial at the time of the study) was responsible for inaccurate HMM. Therefore, HCGs need to receive adequate information on the dosage of the current first line Artemesinin Combination Therapy drugs which have replaced Chloroquine in the treatment of malaria. Home caregivers need to be encouraged to utilize the services of the CHOs as the main source of malaria related information in the HMM programme. Specific groups to be targeted include the older women and the HCGs at risk of inaccurate HMM. Further research on the actual treatment given to children is recommended with particular emphasis on qualitative technique to unpack culturally related ethnic beliefs associated with HMM in children.