Prescription practices for malaria in Mozambique: poor adherence to the national protocols for malaria treatment in 22 public health facilities

Accurate and timely diagnosis and treatment of malaria are critical for proper management of the disease and other treatable febrile illness. Presumptive diagnosis and treatment of malaria, with no reliance on laboratory confirmation, was for a long time the common practice in sub-Saharan Africa, including Mozambique [1‐3]. While ineffective, this leads to over-diagnosis of malaria, over-treatment with anti-malarials, rapid emergence of drug-resistant strains and increased mortality among febrile patients because the true aetiology remains untreated [4‐6].

In 2010, the World Health Organization (WHO) changed the recommendation for the management of uncomplicated malaria from presumptive diagnosis to the ‘test and treat’ strategy. As per this guidance, prescription of artemisinin-based combination treatment (ACT) should be entirely based on laboratory confirmation, either by microscopy or rapid diagnostic test (RDT), where this is available [7]. Most African countries, including Mozambique, adopted this recommendation and are rapidly scaling up RDT to areas where microscopy is not available [8]. Implementation of this recommendation, will not only avoid unnecessary prescription of ACT to non-malaria cases, but will also improve the management of non-malaria febrile illnesses and will delay the occurrence of drug-resistance strains [9]. Despite the rapid expansion of microscopy and widespread availability of malaria RDT, adherence to the guidelines for malaria treatment is still a challenge in sub-Saharan Africa, where administration of ACT to malaria negative patients is still high [10‐12]. For instance, a recent study conducted in Tanzania, found that up to 90 % of anti-malarial drugs were prescribed to patients with negative malaria results [13]. Zurovac et al. [14] investigated the financial aspects of malaria management and found that adherence to the guidelines for malaria treatment might save up to 50 % of the costs of malaria programmes in sub-Saharan Africa. Poor adherence to the guidelines for malaria treatment in sub-Saharan Africa has been attributed to insufficient training, lack of supervision, socio-cultural aspects and lack of trust in the laboratory systems [15, 16]. To data, no study has been conducted to investigate the magnitude of this problem in Mozambique.

Although health care workers’ adherence to the new malaria guidelines has been extensively investigated in sub-Saharan Africa, data from different countries has not shown a consistent pattern. While significant improvement in the management of malaria was achieved in several countries [9, 17, 18], poor adherence to the guidelines for malaria was observed in others [10, 19].

This debate has been renewed in the light of the recent decline in malaria transmission in several parts of sub-Saharan Africa [20‐23]. It is conceivable that reduction of malaria attributable fever, may increases over-prescription of ACT. With regards to Mozambique, previous reports demonstrate that malaria prevalence is in decline, given national efforts to control the disease [20, 24, 25], and also because of the implementation of the Lubombo Spatial Development Initiative and Roll Back Malaria activities [25‐28].

Despite these gains, no study has yet been conducted in Mozambique to investigate the clinician’s adherence to the new malaria management protocols. This study was conducted to assess the prescription practices among health care workers in 22 health facilities in Mozambique. Results of this study will guide the National Malaria Control Programme (NMCP) in Mozambique to improve malaria case management in the country.

In 2011 Mozambique adopted the new WHO guidelines for the management of malaria and stipulated that prescription of ACT for the treatment of uncomplicated malaria should be guided and based on laboratory confirmation using either microscopy or RDT as opposed to presumptive diagnosis and treatment. Health care workers’ adherence to this guideline has never been assessed in Mozambique and data from other sub-Saharan countries are conflicting [9, 10, 18, 19], suggesting that prescription practices for ACT is country specific.

This is the first study investigating health care workers’ adherence to these guideline in Mozambique. Results of this study showed that ACT was prescribed to 72.0 % (25,448/35,361) of patients with negative malaria results. This suggests that despite the availability of laboratory testing using either RDT or microscopy in Mozambique, health care workers’ adherence to the guideline for malaria treatment is poor and the disease is largely treated presumptively. Similar results were found in other sub-Saharan Africa countries where malaria is endemic [10, 13, 32]. A study conducted in Zambia demonstrated that 84 % of patients who received ACT were malaria negative [16] and another study conducted in Malawi found that ACT was prescribed to 58 % of febrile patients with negative RDT results [10]. As previously mentioned, over-treatment with ACT is expensive, and leads to unnecessary wastage of drugs, increased morbi-mortality associated with misdiagnosis of treatable non-malaria febrile illnesses and rapid acquisition of drug resistance [6]. Adherence to the new protocol for malaria management is considered a cost effective intervention, as a study conducted in Kenya demonstrated that correct management of malaria can save up to 60 % of costs associated with malaria treatment [14].

Concern on the over treatment of ACT has recently increased, because the burden of malaria is declining in many sub-Saharan Africa countries, including in Mozambique [20, 21, 25], and as a consequence, the likelihood of treating malaria negative with ACT may increase.

While the results of this study suggest that RDT results are underutilized in Mozambique, the underlying reasons for poor uptake and utilization of malaria RDTs results were not addressed in this study and merit further investigation. However, socio-behavioural studies from other countries demonstrate that patients’ expectations to receive anti-malarial treatment along with a lack of skills to deal with patients’ expectations in the presence of negative result places a strong psychological pressure on clinicians [33‐35]. This suggests that community education on management of febrile illness is also critical. Other reasons for the overtreatment of ACT described in previous studies conducted in other sub-Saharan Africa countries are: (i) clinicians’ resistance to adhere to the new guidelines, because presumptive diagnosis was a common and accepted practice for many years, (ii) lack of trust in the national laboratory systems, (iii) fear of false negative test results [16, 36], and (iv) lack of epidemiological information on the alternative diagnosis of acute fever [6]. Because of this, there is an urgent need of studies to investigate the aetiology of fever of unknown origin in sub-Saharan Africa.

It’s well documented that comprehensive training, periodic refresher courses and regular supervision visits leads to improvement of clinicians’ adherence to the new guidelines for malaria management. In fact, recent studies conducted in Tanzania demonstrated that adherence to new malaria guidelines was enhanced by comprehensive training, education and supervision [9, 37]. Another study conducted in Ghana found similar results [38]. Implementation of quality control measures is also important, and previous studies demonstrated that the expansion of external quality control programmes for RDT and microscopy improved clinicians’ confidence of results [9, 39, 40].

Data from this study showed that stock-out of RDT was a common finding and resulted in increase in the frequency of presumptive ACT prescription. These findings are similar to those from a study conducted in Benin in 2010, which found that stock-out of RDT occurred in up to 38.5 % of health facilities and influenced prescription of ACT [41]. This suggests that improvement in the supply chain for ACT is critical to minimize over-prescription of ACT.

In this study, adherence to the new protocol for malaria management was lower in the central part of the country, suggesting that health care workers’ prescription practices varies across the country. Over-prescription was also slightly higher in rural areas as compared to urban areas. This might be due to higher literacy and better access to information among clinicians working in urban settings, but additional research is needed to investigate this.

Of note, the quality of medical records is a limiting factor when conducting retrospective studies. To minimize some of these problems, the followings strategies were implemented during this study, (i) comprehensive training of all staff who were involved in data collection, (ii) random selection of questionnaires for quality checks and (iii) only recent data, corresponding to the period between July and December 2011 were collected, since the risk of loosing medical records increases as the storage time increases. Despite these efforts, common errors may still have occurred. The Ministry of Health recently launched its strategy to improve the quality of routine data.

Overall, the findings of this study showed that clinicians’ adherence to the new WHO guidelines for malaria management in Mozambique are poor and varies by regions of the country. Stock out of RDT is a strong contributing factor for the presumptive treatment with ACT. This study suggests that the roll out of RDT should be accompanied by continuous training, frequent refresh courses, regular supervision, community education, implementation of quality assurance schemes and improvement of supply chain for RDT.