In Kenya, healthcare services are provided by public and private healthcare facilities at different hierarchical levels. It is a legal requirement for suppliers to sell drugs strictly to licensed healthcare practitioners and registered premises. Further, depending on the license granted, a facility is required to employ defined cadres of staff [1
]. Facilities licensed to offer pharmaceutical services are required to employ a pharmacist or pharmaceutical technologist, being the two cadres licensed by the local regulator, the Pharmacy and Poisons Board (PPB) to practice pharmacy in Kenya. Pharmaceutical technologists possess an intermediate level diploma qualification while pharmacists undergo degree training. Current regulations, however, do not require that all pharmaceutical services have to be provided by these cadres. First, facilities solely licensed to offer health services do not face this requirement but are nevertheless permitted to keep a limited range of drugs for treatment purposes. Second, facilities with a pharmacy license are not explicitly forbidden to employ other staff to ‘support’ the pharmacist or pharmaceutical technologist. As a consequence, they commonly employ untrained personnel who gradually attain sufficient experience to perform limited functions otherwise reserved for the legitimate professionals [5
]. These personnel members constitute the unofficial cadre of ‘pharmacy assistants’ who routinely work alongside the licensed actors in an act of task-shifting and delegation.
Such task-shifting is subject to growing debate. On the one hand, task-shifting is seen as a key strategy to address the substantial shortage of trained health workers in low-income settings [7
] while on the other hand, this practice may affect the quality of services provided. Research on the extent to which the latter applies to pharmaceutical services is scant. A study on 270 public health facilities in Tanzania suggests that such task-shifting ‘may have negative implications both for availability of medicines and quality of clinical care’, but acknowledges that it ‘does not yield any conclusion on possible causality’ [9
]. Other studies have demonstrated that knowledge of service providers in low-income settlements is often inadequate, and so is appropriateness of treatments, diagnostic accuracy, prescription processing and competitive behaviour [10
]. Much is, however, yet unknown about the impact of task-shifting on service quality, particularly so for private sector providers. Enhancing understanding of this matter is highly needed to guide policy framework and context for stimulating task-shifting.
The current study was designed to fill this gap by focusing on private drug-selling facilities in informal settlements (low-income areas) in Nairobi. Informal settlements account for about 60% of the population in Nairobi County. Their inhabitants are characterised as part of the bottom of the (wealth) pyramid (BOP) ecosystem [11
]. They prefer private to public facilities due to proximity, friendly interactions, shorter queue times, and flexibility of pricing and payment [14
The purpose of the study was to describe private sector pharmaceutical services with respect to licensure of facilities, drug procurement behaviour and various indicators of service quality. A better understanding of pharmaceutical services delivery provides an insight into the quality of care and possible areas of interventions to improve patient outcomes.
2. Materials and Methods
2.1. Study Design, Area, Sites and Sampling
This descriptive cross-sectional survey was nested in a larger descriptive study conducted in Nairobi County, Kenya between September and December 2016 as previously reported in a related paper by Ongarora et al. using similar tools within the study setting described [15
2.2. Recruitment and Informed Consent of Participants
During recruitment, the owners of the healthcare facilities were approached and informed that study interviews were targeted at personnel responsible for drug procurement at the facility. Potential participants were fully informed about the purpose of the study, possible risks and benefits at the time of enrolment. They were however, permitted to make a conscious decision to decline participation or withdraw from the study at any time. Willing healthcare facilities thus provided written informed consent and signed a contract under the terms of research for the duration of the study. The identity of participating facilities was blinded by assigning unique identifiers in the database and reports.
2.3. Pharmaceutical Services
The pharmaceutical services targeted during the survey were derived from the statutory guidelines, policy documents and literature reports on related studies [5
]. Drug supply chain management aspects such as licensing, executing cadres, opinions on supplier services and brand preferences were evaluated. Additionally, the prevailing practices with regard to handling of government subsidized items (such as antimalarials and selected family planning commodities), dispensing of drugs, business competition, stocking and use of rapid diagnostic test kits at the facilities, and disposal of expired drugs were investigated. The specific questions incorporated in the questionnaire are listed in Table 1
2.4. Data Collection and Management
Trained field agents used structured, electronic questionnaires designed to capture pertinent information on the prevailing practice and status of the pharmaceutical services provided by the study facilities. Data collected include: cadre of the respondent, license that was used to order drugs and the person who actually made the drug order. For these variables total responses (denominator) were 45 as only one option is applicable per facility. Other indicators were modes of ordering drugs, and reasons for stocking a given brand and changing a drug supplier; perceived reasons that made patients buy a given brand; stocking and use of rapid diagnostics test (RDT) kits; and modes of disposal of expired drugs. For the latter, it was possible to obtain multiple responses for the same variable, thus yielding a denominator greater than 45.
The respondents were interviewed on four occasions at weeks 1, 3, 6 and 8, whereof data were captured at each time point. All data were stored in data files with password secured access to ensure security and confidentiality.
2.5. Data Analysis
Data analysis was carried out using Microsoft Excel spreadsheets. Descriptive analysis was undertaken for pharmaceutical services indicators. All the indicators were summarized by their frequency (presented using frequency tables) of occurrence and also in percentages. No attempt of analysing associations was done.
2.6. Ethical Approval
The University of Nairobi-Kenyatta National Hospital, Ethics and Research Committee (UoN-KNH ERC) granted approval for the study (approval number KNH-ERC/A/371), as per legal requirements in Kenya.
The pharmaceutical services recorded during the study indicate a deviation from recommended practice for the aspects considered. Several reports on the practices of retail pharmacies have shown propensity to non-adherence to regulations [7
]. This is the first study combining all the types of facilities with respect to pharmaceutical practices.
Comparison of premises names and the functional licenses of the facilities showed 20 mismatches, which might be misleading to clients. Correlation between the names and the services offered is important because clients are likely to choose a facility by virtue of its name, size and reputation rather than licensing [17
]. Furthermore, use of misleading words is prohibited under existing laws [2
The number of pharmaceutical personnel acting as respondents (11) could not be rationalised with that against whose licenses were used to order drugs (21) or who participated in the ordering process (22). This discrepancy indicates that some facilities may be ‘renting’ licenses from the pharmacy professionals to facilitate operations [18
Task-shifting and delegation was observed with half of the healthcare facilities placing drugs orders through non-pharmaceutical cadres. This may be an expedient way of controlling costs for business survival [7
]. This is not surprising given the lack of role clarity and blurred professional boundaries as previously reported [18
]. It is disturbing that ‘pharmacy assistants’, a cadre that is not recognized by the laws of Kenya, were offering pharmaceutical services in some facilities in consonance with previous studies [11
As is typical for Kenyan facilities, drugs orders were placed through telephone calls to or from the suppliers [20
]. However, e-mail and other automated or digitized methods were not commonly used within the study population contrary to developed countries, where integrated software is used to manage inventory and prescriptions with improved efficiency and accountability [21
]. Furthermore, the slum study setting may not readily embrace ICT platforms for operations due to cost and capacity building constraints.
Studies have demonstrated that stocking and prescribing patterns are significantly determined by supplier aggressiveness rather than logical professional judgment [18
]. Brand preference based on quality (Table 2
) was most likely perceptual as corroborated by studies conducted in Pakistan and Saudi Arabia [23
], since the facilities have no suitable mechanisms of determining the product quality. For patients, brand preference is influenced by price since they mostly pay out-of-pocket for healthcare [15
Prescription of specific brands by physicians has been recognized as a major factor in determining the products dispensed at retail outlets. Individual prescribers may prefer specific innovator brands due to previous experience, brand loyalty and perceptual poor performance by generics. In some cases, direct financial inducements by drug companies are employed to influence prescribers to promote branded drug products which undermines generic prescribing and increases costs of treatment to patients. Counter incentives by the health plan financiers to promote generic prescribing have also been used in the United States with similar ethical dilemmas [25
]. The study brought out different factors such as prescriber and patient needs, price and efficacy. The stocking behaviour of pharmacies is likely to be modelled on prescriber actions since they are obliged to satisfy prescription needs of their clients [26
]. Patients on the other hand tend to consider the cost of the treatment, perceived quality and efficacy. This outcome is in tandem with the observations of Guttier et al. [28
] that linked patient preference of generic products to perceived safety, efficacy, previous experience and financial ability. Some patients however, may prefer branded products due to previous experience or influence from the attending physician [29
Prescription substitution as encountered during the study is most likely a shielding mechanism to keep customers from competitors. This practice has been a point of conflict between prescribers, dispensing personnel and the patient which highlights the ethical issues surrounding the practice [25
The artemisinin-combination-therapy (ACT) antimalarials and contraceptive pills are supplied on government subsidy for improved availability and affordability [33
]. It is disturbing that half of the facilities that stocked these drugs were making 100% profits for these items, thus impeding affordability.
Stiff competition from local players inspired by the desire for client retention and business survival was a major factor influencing ethics of practice. This underscores the need for the regulators to rationalize catchment areas using established models to curb unhealthy competition, promote professional colleagueship and ethical adherence [35
The practice environment in resource limited settings, compels the facilities to use rapid diagnostic test (RDT) kits since they present a relatively inexpensive and convenient method to carry out diagnosis and generate appropriate prescriptions. The use of these devices as a basis for prescription in the study facilities is a right step towards rational practices. Indeed, RDT kits are actively promoted for malaria, HIV and pregnancy diagnosis [36
With regard to the disposal of expired drugs, it is worrying that 44% of facilities used unorthodox methods such as throwing them away. Drugs disposed through general garbage may cause environmental pollution. Furthermore, they eventually reach the dumpsites where they are liable to misuse by street families. Twelve facilities passed on near-expiry drugs to other facilities for use on unsuspecting clients. Some of the responses provided about drugs disposal were doubtful under the prevailing practice environment in Kenya. For instance, private facilities cannot submit expired drugs to a government facility or the PPB for disposal as claimed. It appears, the drug disposal mechanisms prescribed by the PPB are ineffective as the majority of the facilities did not employ them and therefore may require review [16