This study shows that both providers and community members attach importance to the concept of testing especially to distinguish malaria from other illnesses and an important step towards appropriate treatment, however, what providers and community actually do with a test result is different. For instance some of the providers were of the opinion that test result does not affect the treatment they give. This finding is consistent with what has been reported elsewhere where despite the availability of diagnostics, patients with negative test results continued to receive anti-malarials
[6, 8]. This could be explained by the perception of providers that tests results are not always accurate due to their seeming lack of trust in the competence of laboratory personnel. Perhaps a more worrying aspect on the part of community members is the reported inconsistencies in test results leading them to question the need for tests. Such beliefs if not properly addressed could result in a lack of demand for tests even if they are made available, reinforce the current practice of presumptive treatment for febrile illness and in turn, lead to the over use of anti-malarials.
In this study, the reasons for not testing included the unavailability of testing equipment in health facilities. Despite the expressed desire to carry out tests, lack of testing equipment compels them to still use symptom-based approach for diagnosis; this could limit the ability of health providers to look for other causes of fever in non malaria febrile illnesses as all febrile cases are treated as malaria
[8, 18, 19]. Thus it has been argued that providers should be equipped with testing facilities to enable treatment that is in line with a test result which will also safeguard the efficacy of ACT
. As reported in other setting
, cost was mentioned as a barrier to testing and this is not surprising since individuals bear most of the cost of their treatment in this region and an additional cost of test may increase the burden of payment on patients and could result in their refusal to go for test. Though studies have identified that even when traditional barriers to testing such as availability and cost are minimized, people’s perceptions and practice behaviours emerge as continued barriers to use of test results
An issue of concern emanating from this study is the notion of recommending tests to certain categories of people. In both the public and private facilities a common category was pregnant women, those who have previously had antimalarials but not recovered and patients with confusing symptoms. This could be because health messages have often portrayed certain categories of people like pregnant women as more vulnerable to malaria. While it suffices to give appropriate treatment to vulnerable groups, this impression could pose a problem if it becomes normal practice because even when these tests are made available, not everyone who presents with fever may get a chance to be tested as is currently being advocated. It will thus benefit National Malaria Control Programmes if health messages and treatment algorithms emphasize the importance of testing every febrile patient. Addressing this issue will help to improve access to proper treatment to everyone and to achieve the Roll Back Malaria (RBM) target of universal access to malaria diagnostic testing.
Both providers and community members expressed reservation and a lack of trust for a negative malaria test result and deem it necessary to repeat the same test for malaria elsewhere rather than explore other causes of the illness. This lack of trust in test result has been reported in other studies for microscopy as well as for RDTs
[21, 23, 24] and have often resulted in continued prescription of antimalarials
[8, 24, 25]. Other interpretations of negative test results were also found in this study, for instance there were views that in the event of a negative result, the illness could then be due to the activities of witchcraft or poison, or could be due to supernatural forces. This is similar to what was reported by Muela et al where 62% of respondents shared the view that witches could interfere with normal malaria by hiding the parasites and making them invisible and thus undetectable. The authors attributed the interpretation of malaria in terms of witchcraft as being possible when biomedical treatment does not provide the expected outcome hence people give their own interpretation of the illness when the outcome does not conform to their preconception
. This belief could lead to further delays in seeking treatment and in treatment being sought from unorthodox health care providers.
One practical limitation of this study was that it was carried out at a time that RDTs had not been widely used by the cadre of health providers involved in the study, thus many people had not seen or experienced them. Their perceptions may thus differ when they have utilized these tests. Finally, this study was done in one state in the country and may not be generalisable to the whole country especially with existing diverse ethnic and religious groups.
In conclusion, the study shows that perceived importance of malaria tests by health providers and community members does not translate to trust in the results and thus test results may be of limited use in patient management. Consequently, introducing testing equipments may overcome the barrier of unavailability, but not some of the perceptions about accuracy and usefulness of tests. If tests are to be incorporated into treatment-seeking and provision, there should be behavioural change interventions for both the providers and community. In the first instance testing facilities should be made available in health facilities and health providers properly trained on how to perform tests. This should be backed up by supportive visits and supervision to ensure that providers adhere to testing guidelines and treat based on test results. Proper training has been shown to improve health provider’s ability to accurately perform tests, improve their confidence in the results and foster their acceptability in the community
[14, 27]. The use of supportive supervision was also reported to be effective elsewhere
. Health worker training on malaria case management should also emphasize better ways of communicating with patients on the need for tests. Treatment algorithms that clearly indicate to test everyone suspected of malaria should be provided, these algorithms should also indicate the course of action for a negative test result.
Though it is not evident on what mix of interventions will work in this setting, getting evidence about the effectiveness of alternative intervention strategies is important. A cluster randomized trial following on from this study is ongoing to assess the impact of a combination of approaches on treatment-seeking and provider prescribing in public and private facilities (clinicaltrials.gov NCT01350752).