Parasite-based malaria diagnosis: Are Health Systems in Uganda equipped enough to implement the policy?

Study design and setting

This cross sectional survey assessed the health system capacity for malaria diagnostics at LLHF; health centre (HC) II and III that have a catchment of up to 10,000 and 20,000 people respectively. The larger LLHF (HC IIIs), led by a clinical officer (a clinical officer receives three years pre-service medical and surgical training), provide basic preventive and curative care, maternity care, initial pre-referral care, supervision to the HC II under their jurisdiction and laboratory diagnostic services including microscopy and RDTs. The smaller LLHF (HC IIs), led by a comprehensive nurse (a comprehensive nurse receives medical and midwifery pre-service training), have provision for RDTs although they do not have facilities for microscopy. This sub-study was conducted within a larger study to assess malaria case management practices at different levels of health care delivery within high malaria transmission (Eastern region) and low-medium malaria transmission (mid-Western region) regions in Uganda [12]. Multi-stage cluster sampling methods were used to select study health facilities. Of the 250 LLHF in the two regions, 125 (50%) were randomly selected for health system capacity assessment. This representative sample was deemed feasible and adequate to document capacity needs at LLHF. At the LLHF, all health workers found at the study facility were consecutively offered an interview based on study tools. Health facility specific questions were addressed to the head of the health facilities. The interviews were conducted by 5 teams (4 medical research assistants per team) working in parallel. The teams were trained for one week prior to the survey to ensure that interview questions were asked appropriately and responses consistently recorded. Most survey instruments were adapted from the WHO hospital care assessment tools (http://whqlibdoc.who.int/publications/2008/9789241596428_eng.pdf); see study tools as Additional file 1.

Study measurements

The health system capacity assessment pyramid was used which considers a four-tier hierarchy of capacity needs; that is: 1) tools 2) skills 3) staff and infrastructure and 4) structure, systems and roles (see Figure 1). Using a pre-tested, pre-coded questionnaire, all health workers were interviewed about the available capacity and their perceived capacity building needs. Health facility assessment check lists were used by the research assistants to collect observed data.

The tools assessed included among others: availability of diagnostic kits, equipment, drugs and medical consumables as part of performance capacity for PMD. Routine availability and use of microscopy at LLHF were presented as proportions. However, this survey did not assess the quality of the tests done. The inventory system was assessed by the reported frequency and duration of stock-out period of anti-malarial drugs and other medical supplies. LLHF that reported at least one three-months-long stock out period were taken to have an inadequate inventory and stock management system.

Skills capacity was assessed by evaluation of the staffs’ individual knowledge, technical skills as well as training and motivation to perform their duties including malaria diagnostic tests. For knowledge assessment, health workers were asked about identification and management of severe malaria. Knowledge was graded inadequate if health workers mentioned only 1 out of 10 types of severe malaria, <4 out of 12 danger signs, only one cause of non-malaria febrile illness, could not identify parenteral quinine as the first-line treatment for severe malaria and did not mention two life saving practices for severe malaria. Previous training in integrated management of childhood illnesses was used as an indicator of skills training in management of febrile illnesses.

Staff and infrastructure were assessed by determining proportion of vacancies occupied by qualified health workers (as recommended by the Ministry of Health staffing norms as shown in Additional file 2). The proportion of filled positions was the number of employed qualified health staff over the expected number of staff. Given the 125 LLHF, (64 HCII and 61 HC III) the optimal professional health worker numbers expected is 730 (excludes nursing assistants). The health workers were asked about presence of job descriptions and use of standard operating procedures. Supervisory capacity was assessed by documenting frequency and source of health workers’ supervision. The methods and frequency of supervision were used to assess the regular reporting, monitoring and evaluation systems available. Facility capacity was assessed through asking about physical space for diagnostic activities, storage and mechanisms of safe disposal of infectious waste.

Structures, systems and roles capacity were assessed by determining availability of support services such as; supply chain management, coverage of microscopy quality control/assurance systems and patient referral systems. Health workers were interviewed about their drug ordering and stock management systems including questions about drug stock-out periods. In addition, we assessed the referral systems through which patients at LLHF can access care at larger hospitals. Review of clinical notes and patient/care-giver interviews were conducted on all patients with suspected malaria that were admitted at the district hospitals through referrals from LLHF to document the prevalent pre-referral and referral systems and practices. Data were collected on reasons for referral, use of referral notes, pre-referral medication and transport support. Patients were considered to have had adequate referral if they received all of; a) pre-referral medicine, b) a referral note and c) transport support [12] based on the Uganda Ministry of Health National Treatment guidelines, 2^nd Edition, of 2003. The Health Management Information System (HMIS) was assessed by availability of patient registers, completeness of registers, storage and utilization of records to inform drug stock management. Administrative and organizational capacity was assessed by documentation of available decision-making forums as evidenced by schedules and minutes from staff meetings. Management capacity was determined by cadres in leadership of the LLHF. LLHF leadership was deemed adequate if a HCII was headed by a comprehensive nurse and a HCIII by a clinical officer.

Data analysis

Data were double entered in EPI-info software program version 6 and analyzed using STATA version 10.0 (StataCorp LP, College Station, TX, USA). Summary measures were presented as frequencies and proportions.

Ethical approval

The study was conducted according to the principles of the declaration of Helsinki and the international guidelines of biomedical research involving human subjects (http://www.cioms.ch/frame_guidelines_nov_2002.htm). Ethical approval was granted by the Uganda National Council for Science and Technology (UNCST) which also acts as an ethics committee. Support for the study was acquired from LLHF leader and each participant provided verbal informed consent to participate since this was considered appropriate for a routine audit of the health services.

Health facility characteristics

Tools

Skills

Staff and infrastructure

Structures, systems and roles

Tools for diagnosis and case management

The tools necessary to support clinical practice were limited. There was limited use of microscopy and limited availability of RDTs. Provision of testing tools like RDTs and microscopes is only a starting point of the PMD intervention. Subsequently, routine use of diagnosis then implies health workers’ ability to adopt PMD as a shift from presumptive treatment of malaria. This initial survey considered coverage rather than quality of microscopy. We recommend further studies to evaluate the quality of microscopy given the limited availability of laboratory technologists at the LLHF. In addition to limited diagnostic tools, the quality of malaria case management depends largely on availability of anti-malarial drugs. This survey found that over half of the LLHF faced frequent and prolonged stock-outs of essential medical supplies such as: anti-malarial drugs, syringes and intravenous fluids. Although there is no documented minimum anti-malarial drug stock-out period compatible with adequate malaria case management in an endemic region, the authors felt that drug stock out periods of over three months represent an inadequate stock management system, which is a major drawback to malaria diagnosis and treatment. These levels of stock-outs are similar to those reported in a recent study in two districts in Uganda [13] highlighting the limited health system capacity to manage and maintain a supply chain for routine commodities. This is likely to limit effective implementation of diagnostics such as RDTs that will be provided through the current delivery mechanism for medicines, consumables and reagents. There is need for a well coordinated inventory system at the LLHF in order to prevent stock-outs of anti-malarial drugs and other medical supplies. Given that malaria accounts for 30-50% of out-patient hospital visits and 8% of under-five mortality [14], provision of medical supplies at LLHF is a critical step in reducing infant mortality. Similarly, availability of diagnostic tools is critical to appropriate management of malaria and non-malaria febrile illnesses given that the latter contributes a higher mortality rate among children < five years. Therefore, investment in diagnostic tools should be equally matched with intensive efforts to provide essential medicines and health supplies in order to ensure improved quality of health care delivery. In addition, the procurement capacity should be strengthened to ensure timely delivery and equitable distribution of essential medical supplies at LLHF.

Skills and knowledge capacity

There was limited health worker capacity in terms of numbers, level of training and technical skills required for quality health care delivery. Only 18% of the recommended staff positions at the LLHF were occupied by qualified personnel. Majority (66%) of the health workers were nursing assistants; a cadre with limited pre-service training. This survey found that the nursing assistants had limited technical skills to enable effective malaria case management. Moreover there was limited support supervision from senior health workers which would otherwise improve the performance of nursing assistants.

Since RDTs require less skill than microscopy [15–19], and provide accurate diagnosis [20]; they can be used to scale up PMD especially at LLHF where microscopy is not feasible due to personnel and facility limitations. This is a potential area for utilization of task-shifting to delegate RDTs to less specialized health workers [21, 22]. It is therefore possible to deploy RDTs after brief training of care providers using job aides [9]. Whereas health worker training has been shown to be critical for improving case management, other systems capacity needs must be addressed alongside implementation of RDTs. In addition to guidelines that emphasize PMD before therapy, it is critical to train health workers on the diagnosis and management of alternative causes of fever [23] in order to maximize impact of diagnostics [24]. Given the limited technical skills among the health care providers at LLHF, support supervision and on-job training is required to promote prescription behavior change and maintain high quality health care delivery.

Effective supervision requires qualified health workers to mentor junior staff in a systematic manner as a form of ‘in-service’ training. The job descriptions of junior staff should include clear lines of accountability to their supervisors. However this survey found that support supervision was hampered by lack of qualified staff and logistics. In addition, one third of the health workers did not have appointment letters or job descriptions to determine how they would be supervised. There is evidence that supervision and simple guidelines after deploying RDTs positively impact on prescription behavior [10, 25–27] and limited supervision causes sub-optimal utilization of tests kits [28–31]. It is therefore important to re-emphasize the roles of qualified supervisors at LLHF as part of malaria diagnosis intervention and general improvement of the quality of health care.

Staff and infrastructure capacity

Introducing a new intervention into routine practice should consider its impact on workload or patients’ waiting time at the facility. For example, using RDTs to test all febrile patients prior to treatment may increase health workers’ workload. Therefore health workers require motivation to sustain performance.

We found a shortage of qualified health workforce in all the LLHF. Health workers were already overworked since only 1 out of 5 (20%) of the required positions were filled with qualified health workers. This level of staffing is way below the 80% national Ministry of Health recruitment target. It is likely that this shortage of staff affected the quality of leadership since only 15% of the larger LLHF had appropriately trained leaders. At several LLHF, nurses and nursing assistants provided a broad range of services under minimal supervision due to shortage of qualified health workers. This causes an ethical challenge because nursing assistants are neither recognized nor regulated by any medical professional body in Uganda. Therefore they offer unregulated task-shifting services and it is difficult to monitor their competences, training needs as well as professional discipline. There is need for an institutional and legal framework to protect both the service providers and the patients served by this category of health care providers.

One third of the LLHF had microscopy although they lacked a proper quality control and quality assurance system. There were no defined procedures for monitoring analytical performance, consistent documentation and resolution of quality control (QC) issues. The laboratories were not participating in any external quality assurance scheme. These systems are vital to the maintenance of credible RDT results because inaccurate results have been shown to demoralize health workers, and damage the credibility of PMD [28]. Quality assurance schemes need to be put in place to support RDTs deployment at the peripheral levels. These schemes will ensure that periodic blood smears are prepared and sent to a reference laboratory to validate RDT results using microscopy or other gold standards. In collaboration with WHO/TDR and other agenies, Foundation for Innovative New Diagnostics (FIND) has developed some prototype positive control wells (PCWs) containing antigens that may be used to control the quality of RDTs at the peripheral level. Other centrally planned lot and batch testing will not require local capacity building at LLHF but rather coordination and communication [32, 33].

Structure, systems and roles capacity

Rational referral is part of effective resource utilization. Only one-in-ten of the patients received adequate referral. The concept of a rational and effective referral system was non functional at all LLHF. There were no clear linkages between the various tiers of the health care delivery systems. For effective malaria case management, key elements such as blood transfusion services and oxygen and admission beds were lacking at LLHF. Some of the reasons for referral reported in this study are preventable if the supply chain management is operational without frequent stock-outs of essential supplies like antimalarial drugs and intravenous fluids. Poor response to treatment was one of the four leading causes of referral but this was clearly in absence of confirmatory test or documentation of the effective treatment [34]. Therefore, improving health facilities’ pharmaceutical management capacity is required to ensure non-interrupted supply of pre-referral medications like rectal artesunate as a strategy to improve outcomes of referred patients. It is therefore critical to include inventory management as part of the training package for malaria diagnostics.

Most of the LLHF reported health data to the district through a paper based system. However since 2012, an SMS-based reporting (mTRAC) system is being rolled out to improve local utilization of this data and to report stock-outs of medicines and health supplies. This was not yet in use at the LLHF surveyed. The LLHF leaders need skills to utilize consumption data to make appropriate orders and prevent stock-outs or drug expiry at the facilities. There is need to develop specific capacity to manage stock data at both the LLHF and district level including utilization of simple technology to prevent stock-outs and wastage of resources.

Limitations

This study did not evaluate governance capacity at the district level which is important for effective service delivery at LLHF. For example, the authority and responsibility to make the decisions essential to effective performance, regarding schedules, budgets, and staff appointments were vested at higher level facilities within the prevalent decentralized health system.